systematic literature review french

Crime and safety in transit environments: a systematic review of the English and the French literature, 1970–2020

• Original Research
• Open access
• Published: 31 January 2022
• Volume 14 , pages 105–153, ( 2022 )

Cite this article

You have full access to this open access article

• Vania Ceccato   ORCID: orcid.org/0000-0001-5302-1698 1 ,
• Nathan Gaudelet 2 &
• Gabin Graf 2  

15k Accesses

16 Citations

4 Altmetric

Explore all metrics

This article reviews five decades of English and French literature on transit safety in several major databases, with the focus on Scopus and ScienceDirect. The review explores the nature and frequency of transit crime and passengers’ safety perceptions in transport nodes and along the trip using bibliometric analysis and a systematic review of the literature. The number of retrieved documents was 3137, and 245 were selected for in-depth analysis. Transit safety as a research area took off after the mid-1990s and peaked after the 2010s. The body of research is dominated by the English-language literature (mostly large cities), with a focus on the safety of rail-bound environments and examples of interventions to improve actual and perceived safety for public transportation (PT) users. Highlighting the importance of transit environments along the whole trip, the article also helps advocate for more inclusion of passengers’ safety needs and the involvement of multiple stakeholders in implementing PT policies.

Similar content being viewed by others

Practical Challenges and New Research Frontiers for Safety and Security in Transit Environments

Studies on the impact of road freight transport and alternative modes in Australia: a literature study

Theoretical perspectives of safety and security in transit environments.

Avoid common mistakes on your manuscript.

1 Introduction

When we use public transportation (PT), we spend a part of our travel time waiting at transport nodes (train stations or bus stops) or walking on our way to them. If we feel at risk of crime or feel anxious about something unpleasant happening while in transit, there is a risk we may start avoiding PT and start looking for other travel alternatives. Therefore, feeling safe in transit environments is a fundamental need of all travelers and a guarantee for a sustainable city (UN-Habitat 2019 ).

How safe are we while in transit? Our understanding of the conditions that affect safety is growing, but the international literature lacks an updated systematic review of the research evidence. One of the first reviews in this area was carried out by Smith and Clarke ( 2000 ), followed later by Smith and Cornish ( 2006 ), then in the mid-2010s by Newton ( 2014 ), and more recently with a specific focus on gender issues by Ding et al. ( 2020 ). None of these were systematic reviews (Higgins and Green 2011 ), nor did they incorporate studies devoted to safety interventions. They also reflect scholarly material published in English only, and none performed a bibliometric analysis. We argue for the need of a systematic review of the literature on transit crime and safety. Therefore, in this article we collect and systematize scholarly knowledge on the topic covering five decades of studies, written in English primarily but extending also to the French literature as a benchmark. Although this is not the first literature review on transit safety in English or in French (e.g., Bradet and Normandeau 1987 ; Crossonneau 2003 ; Noble 2015a , b ), this article is the first that covers studies dealing with transit crime (various types) as well as safety perceptions of transit environments, providing:

A special focus on studies that deal with the transit environment (stations and bus stops as well as the last-mile mobility and/or safety door-to-door, the so-called “whole journey perspective”). We still can agree with Smith and Clarke ( 2000 ) who suggest there is a lack of understanding of “how the conditions favoring crime on public transport arise and why they persist. We have limited knowledge of the mix of forces and constraints—political, geographical, economic, engineering, and others—which have combined to shape and form modern public transport.”

Extra attention to studies devoted to safety according to “users’ perspectives” and safety needs. The temporal dimension of these studies is also an interesting feature that has been highlighted in this review but was not pointed out in previous reviews.

Lessons from different research traditions, from English and French, with as many perspectives on the topic as possible. The French literature has been chosen because of the long tradition of research on public transportation and also because French is the fifth most common world language, spoken on nearly every continent, providing therefore a good international overview of this field of research.

1.1 Definitions and delimitation

Transit safety is about safety conditions experienced and perceived by public transportation (PT) users along their trip. This may involve PT travelers’ risk of being victimized by crime or feeling unsafe at a particular place (e.g. a station) or in a combination of environments, situations and contexts. For example, Smith and Clarke ( 2000 ) indicate that crime can occur when a PT traveler is walking from home to a bus stop, or it may occur from or between transportation nodes; it is when the traveler is exposed to environments that are more or less criminogenic. Similarly, safety perceptions may vary along and within these environments. When the PT traveler is waiting for transportation or is on the move between different sections of stations (e.g. on the metro station platform or walking from the ticket area to the platform). Third, when the PT traveler is on board a mode of transport (see e.g. Newton 2004 ).

Transit crime includes offenses and/or safety-disturbing behaviors against PT users, personnel, and property. Moreover, crime targets vary and can include the system itself (e.g. vandalism and fare evasion), employees (e.g. assaults on ticket collectors or guards), and PT users (e.g. pickpocketing or assault). These offences and/or safety-disturbing behaviors may not in a strict sense be a “crime” but they affect PT users’ safety perceptions, such as shouting or what Moore ( 2011 ) calls “low-level behavior,” ranging from groups of young people behaving boisterously to people talking loudly on mobile phones on trains or buses.

Safety perceptions is used here as an umbrella term for fear of crime and other anxieties that are expressed by PT users along the trip and can vary over time. How individuals perceive transit environments depends on their individual characteristics (age, gender, previous victimization) and the features and contexts of these transit environments (transport mode, type of transport node—bus stop or station—and quality of the environments they spend time in, from their home to a transport node).

For the purposes of this article, we adopted the term public transportation (PT) to capture what North American readers may think of as “public transit,” “mass transit,” “rapid transit,” or just “public transport” systems. Note that we use the terms “PT users,” “PT travelers,” “PT riders,” and “PT passengers” interchangeably. As suggested by Newton ( 2014 ), there is no clear consensus of a definition of public transport. PT refers here to “a system used by the public, often a means of transporting passengers in mass numbers, generally a for-hire system that occurs across a fixed route or line,” consisting of a range of transport modes, including rail-bound (railroads, light rail, metro/subway/underground railway, high-speed rail, and intercity rail), buses, trolleybuses, and trams. We limit this literature review to PT only, excluding taxis and Uber services that may be connected to PT. We exclude also studies devoted to events of accidents and suicides and focus instead on crime and safety perceptions only. This perspective includes studies focused on safety during the trip on buses and rail-bound modes, at transport nodes, such as bus stops and stations, and/or on the way to them.

1.2 Research questions

The study is based on materials from an open search from several databases (Scopus, ScienceDirect, Google scholar) which were later split into two parts (a bibliometric analysis and an in-depth analysis of materials both in English and French from 1970). We used the software VOSviewer ( https://www.vosviewer.com ) to organize and visualize the vast material spanning five decades. This literature review aims to respond to the following questions:

When and where have most studies on transit safety been published?

Which are the main research domains in transit safety?

What are the most common characteristics of transit research in English and French?

What is the nature and extent of crime in transit environments?

How do temporal, environmental, and other contextual factors influence transit safety?

How do the physical and social characteristics of transit environments affect safety perceptions of PT travelers?

How does safety vary by different types of PT users?

How does technology affect safety in transit environments?

Which are the safety interventions used to tackle safety problems in transit environments?

Which are the recommendations for future research and practice?

This study is composed of six parts. First, we briefly present the methodology, then we report the results by answering the above questions. In the final section, we identify gaps in the literature and suggest a research agenda as well as policy implications of the current knowledge.

2 Methodology

We used two complementary methods: (1) one was the systematic review using Scopus, Science Direct and Google Scholar database (in French) and (2) the other was a list of publication from the Transit crime network.

We conducted a comprehensive search for academic publications (focus was on articles but a bit over 30% were other relevant publications, such as reports and conference papers). We used a variety of bibliographic databases, Scopus and ScienceDirect in particular because they are reliable sources of internationally published research in English language (note that duplets were excluded). In French, because the literature in this area is composed of a mixture of articles and reports distributed in different databases, we used Google Scholar as a search engine to capture multiple sources (Cairn.info, Erudit, Persée, Hal, etc.). A list of keywords used during this process is shown in “ Appendix ” Table 3 .

In addition, the literature review used a pre-selected list publication from transit safety researchers provided by a user list Transit crime network ( https://maillist.sys.kth.se/mailman/listinfo/abe.kth.se_tcr-network ). This sample of references is a convenient sample based on the contributions of experts that actively participate in a forum that discusses transit safety. Although the decision of including the literature in a specialized research forum such as this user list might have created a bias in the sample since many members were experts in the area, we believe that the benefits outweigh the problems. Eventually, we included a few articles by hand during the finalizing process to include newly published articles, resulting in 3137 publications (Fig.  1 ).

The methodological steps to perform the literature search based on five criteria of selection

From these 3137 publications, 245 were selected (Fig.  1 ), first eliminating duplicates and later excluding those that were not relevant. We adopted the systematic review protocol of type PRISMA-P 2015 (Moher et al. 2015 ) to support inclusion based on five criteria of importance:

Studies aiming at explaining the link between the transit environment, the perceived safety of riders, and the actual crime occurrences in PT. By “transit environment” we mean the design (e.g. size and layout of platform, lighting, visibility), the technology (e.g. CCTV, apps, RTI), the users (flow, crowdedness), the personnel (e.g. patrols), and the immediate context of the PT node (safety door-to-door).

Studies should preferably show temporal patterns of crime and users’ safety perceptions (hours, days, and seasons).

Studies that distinguish safety perception and victimization in transit by individual and socio-demographic characteristics, in particular by gender but also by age, ethnic background, disabilities, and socio-economic status. Studies that analyze fear of crime in transit from the user’s perspective.

Studies with a focus on crime prevention initiatives. Studies that show qualitative and quantitative evaluation of the impacts of programs or environmental changes that tackle crime and/or perceived safety and/or crime rates in PT.

Studies that provide perspectives on the topic that are as varied as possible, in different contexts. We cover therefore publications from 1970 to 2020, in English and in French. Because there is less material in French and a large part consists of reports resulting from surveys (often having a looser structure compared to English-language journal articles). We made the criteria 1–4 more flexible. For instance, we accepted reports from the French Ministry of Transport that were relevant to the topic (e.g. Barjonet et al. 2011 ).

2.1 Bibliometric analysis

The bibliometric analysis included all English-language articles obtained through the user list and in the data collection process in *.ris (Scopus, Sciencedirect and the TCR network; the French publications were too few to produce meaningful visualization on their own). We used VOSviewer version 1.6.12 ( https://www.vosviewer.com ) to create bibliometric maps based on the terms cited in the titles of each article and to group the terms in clusters according to their linkages. With these clusters, we were able to visualize several topics. The co-occurrence analysis was performed using the terms in titles adopting the full counting method. For VOSviewer mapping of most frequent terms in titles, a minimum occurrence of two was used as a cutoff point for inclusion of the terms in mapping analysis. From a total of 675 terms, only 106 met the threshold of the minimum number of two occurrences/repetitions. This criterion of a minimum of two repetitions was selected to avoid terms without links or with weak links to the topic and at the same time to ensure the coverage of the terms and representativeness of the articles. The final selection resulted in 62 items out of the 106 keywords. We did not apply any filter based on the software “score relevance.” Instead, we selected the relevant terms ourselves to fit our subject. The map of words was created using a minimum cluster size of 10 to maximize the readability of the map; the rest of the parameters were the default ones. Output files from the database were used to produce informative network maps by topic and a heat map showing the frequency of keywords indicating potential research domains. An example of each is discussed in the section “ Findings .”

2.2 In-depth analysis

The bibliometric analysis supports the selection of “research domains” that are discussed in detail. The in-depth analysis was guided by our research questions, namely our interest in knowing about studies on the nature and extent of crime in transit environments, the temporal, spatial, and other factors influencing these crimes, and the potential impact of physical and social environments on safety perceptions of travelers. We also investigated using the literature review to determine whether and how safety perceptions vary by different types of users as well as how technology affects safety in transit environments. In addition, we collected literature about safety door-to-door as well as the types of strategies and interventions that are used to reduce crimes and improve perceived safety in transit environments. From the initial 245, about two thirds are discussed in detail in a set of tables and appendices referenced in the next section.

3.1 Bibliometric analysis

3.1.1 number and types of retrieved documents.

From 245 eligible publications that constitute the base for the analysis, 70% are articles and the remainder are reports, chapters, and conference papers; 77% are in English, and 23% in French. Figure  2 a shows the annual share of total publications on transit safety (N = 245). The same is shown in Fig.  2 b but divided by language.

Annual share of total publications on transit safety, 1970–2020 in percentage, a Total and b by language. Source: Selection from Scopus, ScienceDirect, Google Scholar, TCR users list (in percentage). Up to June 2020 only

3.1.2 Evolution of the field over time

The earliest publication within the defined time period was written by the National Technical Information Service (SRI - Stanford Research Institute 1970 ). This study discussed the nature and causes of robbery and assault of bus drivers and suggested solutions. One of the most recent articles was published by Jun et al. ( 2020 ) and was devoted to the gap between users’ needs and the practitioners’ prioritization of accessibility features. A focus on users’ perspectives on safety, particularly on women’s safety, became more common in the last decade. From 1970 to 2010, the number of publications grew slowly, but note that 56% of publications (131) came out in the last 10 years, 37% of the total in the last 5 years (86 publications). The maximum number of publications was recorded in 2017. In addition, 24 papers were published in 2017, and a double special issue in “Crime Prevention and Community Safety” on this theme contributed to almost half of these publications.

The literature is dominated by the English literature. While publications in English are characterized by academic articles published in peer-reviewed journals, the French publications are often reports directed at security professionals, organizations, and practitioners. Therefore, it is no surprise that the French documents follow a looser structure than the English publications do. The target of publications in French is mostly the transit system in France (38 publications), while four reports focus on Canada (Montréal only) and two focus on Belgium. The remainder is about different transit systems in cities around the world.

In terms of content, there are differences between the English and French literature. While the English-language literature has focused more on the causalities between transit environments and crime, the French literature has a more sociological/psychological nature, focusing on travelers’ surveys and users’ behaviors and perspectives on safety (e.g. Barjonet et al. 2010 ; Noble 2015a , b ; Wilow 2015 ; Vanier and D'arbois 2018a ). Examples of the French literature include the seminal work of Bradet and Normandeau ( 1987 ) and one of the latest by Noble and Fussy ( 2020 ) on feelings of insecurity in public transit. A few deal with issues of safety and governance, including the interplay between public and private sectors (Malochet 2015 ; Castagnino 2016 ; Hamelin 2010 ; Bonnet 2008 , 2006 ). Reports written in French are mostly about safety perceptions, with a strong focus on sexual crimes and sexual harassment in PT (e.g. d’Arbois de Jubainville and Vanier 2017 ; HCEEFH 2015 ; Debrincat et al. 2016 ; Colard 2018 ; Alessandrin et al. 2016 ).

3.1.3 Most active countries

Most publications are based on examples from the United States (Table 1 ), with nearly twice as many as in France, followed by the United Kingdom, Sweden, and Canada. However, the United Kingdom is in second position if we exclude reports and count the number of published articles only. More recently, the literature has been expanded by contributions from Latin America (Brazil, Colombia, Mexico), Australia, India, Norway, Malaysia, Japan, and Hong Kong, with more than three publications each (e.g. Shibata et al. 2014 ; Javier and Ceccato 2020 ; Otu and Agugua 2020 ; Chowdhury and van Wee 2020 ; Fillone and Mateo-Babiano 2018 ).

3.1.4 Publications by type of geographical area

Public transit systems such as metro, commuting trains, and bus lines in global and capital cities are often the focus of transit safety studies (44% of all publications). Cities such as New York, London, Paris, Tokyo, Los Angeles, Sao Paulo, New Delhi, and Santiago belong to this group. Next, studies that are devoted to an entire country, such as the United Kingdom, United States, France, and Sweden, account for 28% of the publications, followed by transportation systems in cities, such as Liverpool, Lucknow, Manchester, Lyon, and Milan. Thus, 6% of the papers focus on other areas and 10% do not specify the location of their study.

3.1.5 Visualization of author keywords

The visualization map of author keywords from the English literature shows that “crime,” “safety,” “public transport,” “environment,” “fear,” and “transit” were the most frequently encountered keywords (Fig.  3 a). These maps show how keywords in the papers’ titles are linked to each other and their occurrences in our sample. The maps are divided into four clusters, which show families of terms based on their co-occurrences in titles. In the density visualization (Fig.  3 b) the strength of each keyword is shown in different shades. The words “crime” and “public transport” are linked to many other keywords and have many occurrences.

a Network visualization map of words contained in the titles of the English publications, b Density visualization map and c Network visualization map for the English selected publications (N = 188), centered in the words “crime” ( c ) and “public transport” ( d )

Note that “crime,” “public transport,” “environment,” “fear,” “safety,” “crime prevention,” “bus,” and “women/woman” are recurrent terms in these publications; in particular, the word “crime” constitutes the major hotspot. The keywords “women/woman” signal a growing literature on sexual transit crime, a safety problem that is largely underreported (e.g. Ding et al. 2020 ; Solymosi and Newton 2018 ; Newton et al. 2020 ; Priya Uteng et al. 2019 ; Chowdhury and van Wee 2020 ; Jun et al. 2020 ; Whitzman et al. 2020 ; Dunckel-Graglia 2013 , 2016 ; Natarajan 2016 ; Moreira and Ceccato 2020 ). These keywords in the titles are also represented when the network visualization map is created for the selected English publications (N = 188), centered on the words “crime” (Fig.  3 c) and “public transport” (Fig.  3 d). In the next section, we discuss the interconnections between these terms as research domains.

3.1.6 The nature of studies in transit environments

Based on a title search of 245 articles, we found that 81 (33%) publications were about crime, 29 (12%) were about fear and/or safety perceptions, 129 (53%) both crime and safety perceptions, 6 on other topics (2%).

Among those articles that were devoted to the analysis of crime, 70% focused on violence, 18% on property crimes with aggravated violence, such as robberies, 2% on property crimes only, and 10% on vandalism and other types of crime. Vandalism is often associated with the analysis of other crimes (either violent or property crimes), rarely studied by itself. Studies that are devoted to violent crimes, assaults of all types, such as fights and attacks, are common, followed by gendered and sexual crimes and terrorism, which are a minority. With regards to transport mode, as many as 76% (188) of the publications deal with rail-bound PT; of these, about half relate to trains and the other half the metro. About 65% are related to buses, while 8% (22) of these focus on other types of transportation, such as ride-hailing services  “Tourist Vehicle with Driver” (Weber 2019 ) or Vikram transports in India (Tripathi et al. 2017 ). In other words, many publications deal with both buses and rail-bound modes.

3.1.7 The types of methods

Quantitative analyses (descriptive or confirmatory statistical analyses) are used in 63% (155) of all 245 publications, 19% (47) are qualitative publications, and 18% (43) used mixed methods. Among the quantitative analyses, the majority used descriptive statistics, such as frequency analysis of crimes and/or perceived safety, cross-tables, and different types of regression models. These publications are often in English and intend to assess the impact of traveling conditions of one or more aspects of the environment on crime and/or safety. As many as 17% (33) of these quantitative publications use regression models. Among them, half (17) were classical multivariate linear regressions and half (15) were logistic regressions (nine binomial, three multinomial, and three ordinal). Publications using Geographic Information Systems (GIS) and interstitial analyses have appeared at the end of the twentieth century and represent 12% (25 out of 198) of quantitative papers. None of them are from the French literature. These methods are useful to study certain aspects of transit safety and crime, such as socio-economic characteristics (e.g. Buckley 1996 ; Sung-suk 2009 ) and transit surroundings (e.g. Liggett et al. 2001 ; Newton et al. 2014b ). Out of 47 qualitative publications, half were in French. These findings show the dominance of qualitative analyses in the French literature, while the English publications are denominated by more quantitative approaches. Exceptions are publications in French from Canada that follow an Anglo-Saxon tradition of taking a more quantitative approach to analyze transit safety (Grandmaison and Tremblay 1997 ; Ouimet and Tremblay 2001 ; Browne 2010 ). The studies based on mixed methods are varied in nature and often aim at producing results relevant for practice; see for instance Newton et al. ( 2004 ).

3.2 In-depth analysis by research domains

3.2.1 the temporal patterns of crime and perceived safety.

The flow of passengers both at peak and off-peak times create the necessary conditions for crime, namely the presence of a possible motivated offender, a desirable target, and a lack of people ready to intervene if anything happens (Cohen and Felson 1979 ). This temporal dimension has been recognized by the international literature as key to understanding safety conditions in transit environments. Of the total publications, 22% investigate one or several dimensions of the temporal variations of crime and/or safety perceptions in transit environments. Nearly all publications refer to hourly patterns of crime and/or perceived safety, often differences between peak and off-peak hours. These “time windows” are important for transit safety conditions because crowded environments tend to attract certain types of offenses, such as pickpocketing (Zahnow and Corcoran 2019 ; Ceccato et al. 2015 ), while empty stations or bus stops tend to promote crimes that demand anonymity, such as robbery. “ Appendix ” Table 4 summarizes findings from these studies split into different time windows.

Studies show that morning peak hours attract robberies (Block and Davis 1996 ; Newton et al. 2014b ), assaults, thefts, public disorder (Stringer 2007 ; Ceccato 2018 ), and sexual crimes (Ceccato and Paz 2017 ), while Smith ( 1986 ) finds fewer crimes in the early morning hours. Research suggests that people feel unsafe during morning rush hours, especially women (Mitra-Sarkar and Partheeban 2011 ), which is when sexual crimes can take place (Ceccato et al. 2011 , 2017 ). As exemplified by Vanier and D'arbois ( 2018b ), seven articles deal with peak hours in the mornings, a time that is both “anxiety-inducing” and criminogenic. However, almost half of the papers focusing on the nighttime describe it as the time of day when most people feel the least safe, especially women (e.g. Austin 1984 ; d’Arbois de Jubainville and Vanier 2017 ). This feeling is stronger when travelers have to wait for transportation (Chowdhury and van Wee 2020 ; Mahmoud and Currie 2010 ), or when the location is associated with a particular land use, such as being near nightclubs (Gosselin 2012 ). Fear may reflect the risk of different crimes. For instance, violent crimes are said to be prevalent after rush hour, after 6:00 pm (Moreira and Ceccato 2020 ; Newton 2014 ), while robberies seem to occur more often late at night than in the early evening (Block and Davis 1996 ; Chaiken et al. 1974 ; Clarke et al. 1996 ).

Among the publications that deal with the temporal dimension of crime or perceived safety in transit environments, 43% made a distinction between crime occurring on weekdays or on weekends. Seven of these (30%) suggest that more crimes occur on weekends (e.g. Ceccato et al. 2011 , 2017 ). The end of the week (Thursday to Saturday) witnesses increases in robbery (Newton 2018a ). Finally, one article (Smith 1986 ) suggests that maybe there are fewer crimes during weekends in absolute numbers but, after the author compares the number of crimes with the number of travelers, crime rates are actually higher. In addition, despite the fact that Monday is posited as a low-crime day (TCRP 2001 ; Capasso da Silva and Rodrigues da Silva 2020 ), 22% of the publications suggest that most crimes happened on a weekday, especially on Wednesday (e.g. Chaiken et al. 1974 ; Le Grâet and Vanier 2016 ). There are indications that lack of surveillance and fewer patrols affect crime during weekends (Ouimet and Tremblay 2001 ). Also, trains are less frequent on weekends, the density of passengers decreases and waiting times increase, affecting safety perceptions (Yavuz and Welch 2010 ; Mahmoud and Currie 2010 ). Weekends tend to be perceived as unsafe, especially by women (Vanier and d’Arbois 2017 ). “ Appendix ” Table 4 summarizes the remaining time windows.

3.2.2 The influence of environment on transit safety

According to Smith and Clarke ( 2000 ) “crimes cannot be properly explained, nor effectively prevented, without a thorough understanding of the environments in which they occur. Nowhere is this more apparent than in urban public transport.” (page 169). Transit environments can be planned in a way that reduces the possibility of crime occurring, by stimulating surveillance, fostering territoriality, and reducing areas of conflict by controlling access and improving overall perceived safety. These principles are based on what is called Crime Prevention Through Environmental Design (CPTED) and on situational crime prevention theory (Clarke 1983 ) that, among other things, seek to enhance natural surveillance through planning and modifying the environment. CPTED asserts “that the physical environment can encourage or discourage opportunities for crime by its very design and management” (Cozens et al., 2003a , b , c ). Research shows that crime and safety in transit environments depend on the physical and social environmental attributes at the transport node, either bus stops or train/subway stations, the characteristics of the immediate environment and neighborhood, and the relative position of both the station and neighborhood in the city (Loukaitou-Sideris et al. 2001 ; Ceccato 2013 ). Yet, how do the physical and the social characteristics of transit environments affect safety perceptions of PT travelers? Below we discuss the importance of the quality of transit environments, in particular, design, layout, obstacles and hiding spots that affect visibility and surveillance; maintenance and lighting conditions, CCTV cameras; the presence and flow of passengers over time, specially crowdedness, and the quality of immediate areas.

3.2.2.1 Visibility, obstacles and hiding spots

Elements of the station, such as elevators, pillars or facilities should not hinder visibility for PT users. Instead, elevators with large glass side panels should be endorsed (La Vigne 1996 ). Similarly, columns no wider than necessary, glassed internal walls and ticketing in one central location in the station (Felson et al. 1996 ) should be encouraged. In an underground station with long corridors, often with sharp corners and restricted sight lines, a solution is the establishment of corner convex mirrors that increase visibility and natural surveillance (Crime-Concern 2004 ; Felson et al. 1996 ). Moreover, it is important to ensure visibility from any point in a station (Umaña-Barrios and Gil 2017 ; Mohamed and Stanek 2020 ); high-domed and/or white ceilings in the station are encouraged (e.g.: Felson et al. 1996 ; Sham et al. 2013 ) as is avoiding spiral ramps (Atkins 1990 ). Moreover, the design of bus stops is important. Transparent bus shelters are recommended by seven articles (27%) focusing on buses and bus stops. They insist on the avoidance of enclosed brick shelters; indeed, the lack of visibility is an opportunity for offenders to commit crimes (Noble 2015a ; Sham et al. 2013 ). Moreover, curved shelter structures can be a solution that gives the feeling of an open and secure space for users (Diec et al. 2010 ). Finally, bus stops should be located in front of a place that provides natural visibility (Liggett et al. 2001 ). Regarding the interior of the buses, large, transparent windows are encouraged, while dark-tinted windows reduce visibility from the outside during daylight and should be avoided (Levine and Wachs 1985 ). Thus, the same article suggests that seats in the back could be arranged in a circular pattern to allow better visibility between passengers and to cut down on pickpockets. Overall, the internal design of the bus should maximize clear sight lines and visibility for both driver and passengers (Sham et al. 2013 ).

The external and internal design of the stations are essential elements that can promote visibility and opportunities for surveillance. Uittenbogaard ( 2015 ) suggests that visibility as the possibilities (promoted by the environment) a person has for observing others or situations while surveillance , in contrast, relates to the possibilities for others to observe a person, an object or a place. It can also be carried out by CCTV cameras or involving a diverse array of agents in a train station, for example: from police, security guards, safety hosts to drivers, shop owners, passengers and residents. More recently, with the wide use of cameras in mobile phones, ‘surveillance’ (‘eye-in-the-sky’ watching from above) gives room to the process of ‘sousveillance’ which denotes bringing the camera or other means of observation down to a human level (Ceccato 2019 ).

Impaired visibility is clearly associated with fear of crime (e.g.: Cozens et al. 2003b , 2004 ; Ceccato and Paz 2017 ). Having control of where others are, the capacity to see and be seen by others, increases the confidence of travelers (Umaña-Barrios and Gil 2017 ). When passengers feel isolated, they feel themselves becoming an easy target for criminals (Noble 2015b ; Buckley 1996 ). Research shows that underpass platforms are often related to higher rates of disorder and crime (e.g. Ceccato 2013a ; Loukaitou-Sideris et al. 2002 ). However, the literature shows unexpected findings when vandalism is often associated with increased visibility, perhaps because it reflects the size of passenger flows (Ceccato and Uittenbogaard 2013 ). Ten articles (45%) raise the question of visibility inside stations. These focus on the importance of a wide-open design, which straightening sight lines provides, and avoiding obscured areas, corners, hiding spots, and enclosed spaces (e.g. CEMT 2003 ; Ceccato et al. 2011 ). In addition, plants should not become obstacles and impede the field of view (Mohamed and Stanek 2020 ; Crime-Concern 2004 ; Diec et al. 2010 ).

3.2.2.2 Management and maintenance

Mechanisms linking poor management and maintenance of transit environments to crime and poor safety perceptions have long been associated with Wilson and Kelling’s broken window syndrome (Wilson and Kelling 1982 ), which suggests that unrepaired damage to property encourages further vandalism and other types of crime. We argue that poor management and maintenance of transit environments indicate other types of underlying problems that can have a direct effect on safety perceptions and also lead to victimization (see also Eck 2019 ). The effect of poor management and maintenance is well documented in the international literature. It may reflect poor cleaning practices in and around transit nodes (e.g. Liggett et al. 2001 ; Ceccato and Masci 2017 ; Loukaitou-Sideris 2012 ) or poor lighting conditions, as further discussed below.

3.2.2.3 Lighting

The effect of illumination (either artificial or sunlight) is documented in the international literature on transit environments with 71 publications out of 245 publications: 60 in English and 11 in French. From 245 articles, 27 publications deal directly with the effect of illumination of bus stops and/or stations. Out of these publications, 17% deal with the effect of illumination on crime, 76% on perceived safety, and 7% on both. For example, according to La Vigne ( 1996 ) suggested that stations well-lit and well-maintained contribute to the overall feeling of safety among Metro users in Washington, DC. Overall, lighting seems to have a reductive effect on crime: as many as 66% of publications that focus on crime state that good illumination decreases crime, while 11% have found that lighting can increase crime or fear in certain conditions (e.g. a Canadian study shows that good lighting should extend from the bus stops to the adjacent streets so that bus stops avoid the passengers to be exposed to the ‘fishbowl effect’ (Women’s Centre/METRAC 1991 ). As many as 14% have found inconclusive results in the sense that lighting can increase or decrease crime, and 9% found that illumination shows no significant effect on crime. For a summary of these findings on safety perceptions, see Appendix Table 5 .

3.2.2.4 Crowdedness

The literature lacks quantitative reference for what would be a ‘crowded’ station ( Appendix Table 6 ). Evidence shows that nearly empty transit environments (e.g. platforms, stations, bus stops) increase certain types of crimes in western countries, such as exhibitionism and sexual assaults (e.g. Beller et al. 1980 ; Colard 2018 ; Mohamed and Stanek 2020 ), but also robberies (Clarke et al. 1996 ), social disorder (Ceccato et al. 2011 ), and violent assaults (Noble 2015b ). These findings show that the absence of informal surveillance motivates offenders to consider crime as an alternative. People are worried about empty transit places, especially women (Vanier and d’Arbois 2017 ), perhaps because, as suggested by Ball and Wesson ( 2017 ), unwanted sexual behavior is perceived to be more severe in empty transit environments.

However, crowdedness also has an impact on transit safety. Some studies show an increase in fights, insults, and pickpocketing in crowded transit environments (Ceccato et al. 2015b ; Noble 2015b ). Pickpocketing in particular occurs at overcrowded spots at which offenders can take advantage of a high density of people who are close together (Loukaitou-Sideris et al. 2001 ; Ceccato et al. 2015 ). Research shows that these conditions are affected by the transit environment, for instance, Solymosi et al. ( 2015 ) refer to a pedestrian motion analysis that found that in crowded environments safety problems occur due to bottlenecks, which are areas in which there is a significant capacity drop in pedestrian movement, such as a narrow doorway in a corridor, when jamming occurs when the incoming flow exceeds the capacity of the bottleneck. Other studies found a diminution of overall crime under crowded circumstances (e.g. violence, and vandalism) or just inconclusive results (Marteache et al. 2015 ; Ceccato et al. 2011 ).

In very crowded transit environments, sexual harassment and especially groping are recurrent problems since they are observed in 54% (6) of the 11 selected publications. Crowdedness in transit environments facilitates women’s sexual harassment (Beller et al. 1980 ; Alessandrin et al. 2016 ; Colard 2018 ), and unwanted sexual behavior appears to be more ambiguous for men (Ball and Wesson 2017 ). Poor safety conditions in crowded places impact women’s behavior and force them either to avoid crowded transportation (Mitra-Sarkar and Partheeban 2011 ) or to use women-only carriages (Bachok et al. 2014 ; Horii and Burgess 2012 ; Dunckel-Graglia 2013 ) in India, Malaysia, and Japan. In South America, Kash ( 2019 ) found that even if women experience high stress in crowded transports, they have no choice but to commute in them. However, Gaylord and Galljher ( 1991 ) found that crowdedness decreases crime in the Hong Kong transit system.

3.2.3 The technology and safety in transit environments

From 245 articles, 18 (7%) directly deal with some aspect of transport technology and how it has influenced safety. Only six publications are quantitative. Most articles fall within the spectrum of Information Communication Technology (ICT), access control systems, smartphone apps, emergency alarms, crime-resistant materials, ITM machines, RTI, apps, and social media. As a target of crime, mobile phones are said to be the first target by theft offender in transit environments in the United Kingdom and the United States (Newton 2016 , 2018a ; Newton et al. 2014a ; Gentry 2015 ). They may also facilitate certain types of transit crimes, such as cyber-theft, thanks to Wi-Fi or cyber-harassment (Newton 2016 ). On the other hand, technological innovations might decrease transit crime and enhance perceptions of safety. For instance, researchers found that emergency alarms and ICT are efficient in preventing and controlling crime (Debrincat et al. 2017 ; Shen 1997 ). The use of big data analysis in PT is a growing research area; for instance, a recent study shows how passengers’ routine mobility patterns can be analyzed using information from subway travel cards and how pickpocketing suspects can be detected by matching abnormal cardholder trajectories and occurred crime locations (Gu et al. 2019 ).

To prevent vandalism, the use of new resistant materials has been suggested by previous research (La Vigne 1997 ; Shen 1997 ), while some access control systems and exact fare transaction machines may decrease fare evasion to the detriment of perceived safety (Weidner 1996 ; La Vigne 1997 ). More recent technologies such as safety apps (McCarthy et al. 2016 ; Debrincat et al. 2017 ) and real-time information (RTI) (Abenoza et al. 2018 ; Newton 2016 ) increase safety perceptions at transport nodes. However, technologies used to prevent the occurrence of crime may not be efficient everywhere in transit environments (Shen 1997 ; Swain 2015 ) and has to be adapted to different types of users. For instance, Sochor ( 2012a ) presents the use of ICT to enhance mobility in a case of a navigation system for visually impaired users in Sweden [see also (Sochor 2012b )]. Modern analytical tools and visualization software, such as Geographical Information Systems (GIS), free of charge software, and virtual reality (VR) have the potential to improve communication when implementing safety intervention, in particular between practitioners, researchers, and/or PT users and to create a better basis for improving transit safety (Cozens and Hillier 2002 ; Cozens et al. 2003a , c ; Newton 2007 ; Smith et al. 2013 ; Ceccato et al. 2015 ). Finally, big data and social media provide useful data for the analysis of passenger flow and safety (Marteache et al. 2015 ; Casas and Delmelle 2017 ; Zahnow and Corcoran 2019 ). For more in-depth discussion of the effects of technology on transit safety, see the next section.

3.2.3.1 CCTVs in transit environments and BWCs

Closed-circuit television (CCTV) is associated with a modest but significant decrease in crime, particularly property crime, vehicle crime, and drug crime in public places in general (Piza et al. 2019 ).We found 30 publications (12%) focusing on CCTV. Findings show conflicting evidence. CCTVs can reduce the risk of violence in metro stations in Sao Paulo, Brazil (Moreira and Ceccato 2020 ) and car crimes, such as breaking into cars in parking lots in the United Kingdom (Tilley 1993 ; Webb and Laycock 1992 ). They are also depicted as being efficient in increasing perceived safety in several studies (Cozens et al. 2003b ; Debrincat et al. 2017 ). On the other hand, CCTV impact on crime reduction and/ perceived safety was not evident in transit systems in Sweden or in the United States (e.g. Priks 2009 ; Shellow et al. 1974 ). The reason is most likely that, most of the time, CCTV cameras are installed together with other interventions, making it difficult to know which intervention influenced the outcome (Tilley 1993 ). Other studies show no significant impact on safety, such as in Los Angeles (Loukaitou-Sideris and Fink 2008 ), mostly when they are implemented alone, with no other intervention (Ceccato 2013a ; Newton and Bowers 2007 ).

Women are the most skeptical about CCTV (e.g. Vanier and D'arbois 2018a ; Yavuz and Welch 2010 ). Indeed, CCTV may be seen as an intrusion of privacy (Ceccato 2013a ) or even voyeurism (Smith and Clarke 2000 ). Yet, studies claim that when they are visible, CCTV may enhance perceived safety (Ceccato 2013a ; Riley and Dean 1985 ). They claim that the preventive effect of a camera is greater if it is positioned in a visible place (e.g. Mariotte et al. 2004 ; Burrows 1980 ). In addition, publicity and communication around the installation of these devices may help users to feel safer and may have discouraged potential criminals (e.g. Richards 1980 ; Tilley 1993 ). People cannot feel safer if they do not know that the cameras exist (Vanier and D'arbois 2018a ). It is interesting to note that CCTV may have a greater positive impact on older adults’ perceived safety than on other riders’ (Wallace et al. 1999 ).

The way CCTV is implemented is important. Grandmaison and Tremblay ( 1997 ) suggest that to improve the effect of CCTV, the systems must be “credible”, meaning that the products of the system must have good resolution, agents must be trained to use this tool, and specific objectives should be identified (Mariotte 2004 ). If someone is monitoring the live feed, CCTV can be useful to find the criminals or suspects, especially if the camera’s sight field is continuous, in a tight mesh (Malochet and Le Goff 2013 ). In addition, the effectiveness of CCTV depends on the situation (Riley and Dean 1985 ), for instance, on how much potential criminals fear the system and see in it an increased risk of getting caught (Webb and Laycock 1992 ). Scholars have claimed that the effectiveness of CCTV as a deterrent of crime has to do to some extent with its novelty in a station, even if the cameras are “fake” (Burrows 1980 ). Cameras with face-recognition software might be used in the future (Nourani et al. 2020 ) to the detriment of personal privacy.

Another technological development is the use of Body Worn Cameras (BWC). These systems seem to generate fewer use-of-force reports and complaints from citizens, yet there is little research as yet in the use of BWC in transit environments. In one of the few studies in transit systems, Ariel et al. ( 2019 ) indicate that BWC reduce assaults on staff members unequipped, yet working close to equipped colleagues and the cameras also provide train operating company staff with an effective protection that reduces physical aggression toward them.

3.2.4 Location and surrounding areas

A transportation node, such as a train station, is often planned to facilitate the movement of as many passengers as possible. It tends to be within walking distance of a residential area, working place, industrial area, or commercial center. This centrality feature of transportation nodes has criminogenic implications. Different types of land use affect the social interactions at those places and, consequently, their geographies of crime (Ceccato 2018 ). The literature also indicates that transport nodes can become crime attractors, crime generators, crime absorbers, or crime radiators (Brantingham and Brantingham 1995 ; Bowers 2014 ; Sedelmaier 2003 ; Newton 2018 ) partially because they contain or are close to risky facilities. Clarke and Eck ( 2007 ) suggest that certain facilities can be called risky and affect crime occurrence at their locations as well as within their vicinities. Examples of risky facilities are bars, restaurants, stores, shopping malls, ATMs, bus stops, railway stations, parking lots, apartment buildings, mobile home parks, libraries, hospitals, schools, public swimming pools, and marinas. Adams et al. ( 2015 ) suggest that these different land uses interact to form what they call “malignant mixes” because individually or in combination, they engender a greater risk of crime and according to La Vigne ( 2015 ) should be approached by crime prevention initiatives in a different way. These areas are important because they embody the safety needs of travelers along the trip from home to the closest transport node and vice-versa (what is also called the safety in the “last mile.”).What makes these areas unsafe?

The location of transport nodes in cities also has an independent impact on crime levels and geography (Newton et al. 2014a ; Phillips and Sandler 2015 ; Liggett et al. 2003 ). While Newton et al. ( 2014b ) found that terminal stations were at lower risk, Marteache et al. ( 2015 ) showed that central stations and one terminal station are more criminogenic than others. Ceccato et al. ( 2011 ) found more crime happens in inner city areas (where the central station is located) but also in the so-called “end stations.” On the other hand, Ihlanfeldt ( 2003 ) found no evidence that transit in suburbs increases crime. The size of transport nodes may also matter for the safety of passengers. For instance, Marteache ( 2018 ) found that smaller airports experience higher theft rates than larger ones, while Ceccato et al. ( 2011 ) suggested that larger subway stations attract more crime, controlling for station and neighborhood contexts. More interestingly, there seem to be gender differences in victimization by station location. Moreira and Ceccato ( 2020 ) found that women are at higher risk of victimization than men in central metro stations, while men run a higher risk of violence at end stations—both notably during late-night periods. Some suggest that it is the surrounding contexts that affect crime and/or the perceived safety of riders. While Abenoza et al. ( 2018 ) showed that mixed land use tends to increase perceived safety of users at night, two other studies found that liquor stores, bars, and pawnshops near transit increased robberies (Liggett et al. 2001 ; Sung-suk 2009 ). They also found that transport nodes near vacant lots had higher property crime rates. It appears that the direct surroundings of transport nodes have great influence on transit crime (Buckley 1996 ), but also the other way around, namely research suggests that crime concentrates in close proximity to public bus stop locations. Hart and Miethe ( 2014 ) noted that robberies around bus stops occur when specific combinations of risk factors are present, and Gerell ( 2018 ) found that for violent victimization the risk was highest in local centers of disadvantaged neighborhoods having low collective efficacy, but such a risk was dependent on temporal and city contexts.

Out of 245 publications, 15% (36) deal with the safety conditions around transportation nodes. Evidence of transport nodes functioning as crime absorbers is found in the literature (Pearlstein and Wachs 1982 ; Newton 2008 ). Scholars have found crime on bus routes mostly associated with high crime neighborhoods. Similarly, Loukaitou-Sideris et al. ( 2002 ) suggest that crimes at subway stations are related to overall crime in the neighborhood, as a crime absorber. There are also examples of transportation nodes that radiate crime. Several studies showed that transit environments seem to radiate crime to their surroundings (Block and Davis 1996 ; Buckley 1996 ; Kooi 2013 ; Piza and Kennedy 2003 ; Liu et al. 2020 ; Poister 1996 ). In addition, Spicer and Song ( 2017 ) demonstrated how the perception of crime increased and intensified over 10 years: as a public transit hub grew in size and ridership, crime increased and became more concentrated. Transit nodes might have a positive effect on safety. For instance, Billings et al. ( 2011 ) found that the opening of a transit line led to a decrease of property crimes. It seems that the link between transit systems and crime is dependent on multiple factors (Ihlanfeldt 2003 ). For instance, Sedelmaier ( 2003 ) found that a new small transit system had not generated more crime than usual, and this was confirmed by Sedelmaier ( 2014 ). Liu et al. ( 2017 ) found that only the bus stops with high capacities attracted property crimes. More recently, Gómez et al. ( 2020 ) suggested that the development of major highways benefitted mobility in a metropolitan area but compromised the sustainability of particular places and called for a more holistic approach when assessing the effects of transit nodes on urban environments.

3.2.5 Safety perceptions by types of user

Victimization and safety perceptions are often influenced by gender and other individual characteristics (Whitzman 2007 ; Madan and Nalla 2015 ), but it is the intersection of some individual and contextual characteristics that determines an individual’s vulnerability to crime and fear of crime (Crenshaw 1989 ). For instance, scholars have indicated that previous victimization continues to be an important determinant of perceived safety in public places (Otis 2007 ; Hirtenlehner and Farrall 2014 ; Yates and Ceccato 2020 ; Hale 1996 ). The risk of victimization is expected to vary across types of individuals (age, gender, (dis)ability, LGBTQI status, IT literacy, previous victimization, frequency of use of public transportation, length of trip, etc.) (Gekoski et al. 2015 ).

The level of service appears to be linked to people’s perceptions of safety. Indeed, a high frequency of service—especially at late hours- increases perceived safety and consequently, people rely more on public transportation when they feel safe (e.g. Reed et al. 2000 ). The gender dimension is the one that is by far the most studied in transit safety studies. Not only PT passengers but also transit personnel express a variety of safety perceptions (Browne 2010 ); some personnel refrain from coming to work because of previous experience with crime (Bradet and Normandeau 1987 ).

From our analysis, 40% of the publications (100) deal with the issue of perceived safety by type of user. Among them, 72 publications focus on women’s safety perceptions. Women feel less safe than men while transiting (e.g.: Cozens et al. 2003b ; Levine and Wachs 1986 ); some researchers argue that women feel three times less safe than men (Wilow 2015 ) and express greater precautionary behavior than men (Ingalls et al. 1994 ; Newton et al. 2020 ; Ceccato and Nalla 2020 ). Women are far from being a uniform group of passengers. Vanier and d’Arbois ( 2017 ) distinguished four types of female PT users: daily workers (63% of whom feel unsafe), young women (66% of whom feel unsafe), inactive women (who feel the safest) and casual users (the most unsafe, 70% of them do not feel safe). Safety perceptions are affected by whether or not women are alone (Ceccato et al. 2017 ; Crime-Concern 2004 ). Worldwide, women are often “transit captives” (Yu and Smith 2014 ; Ceccato 2017a , b ; Guimarães et al. 2019 ), which means they do not have any other choice of transport than PT. It is also this group who is overrepresented among those who feel unsafe (D'Arbois 2019 ; Vanier and D'arbois 2018a ), compared to non-captives. Gender status is another determinant of perceived transit safety. LGBTI + (and especially ethnic minorities) feel particularly unsafe and fear the reactions of other PT users (Lubitow et al. 2020 ; Nourani et al. 2020 ), as do disabled people, who declare that their transit safety needs are not entirely satisfied (e.g. Currie et al. 2013 ; Barjonet et al. 2011 ; Jun et al. 2020 ; Iudici 2015 ; Iudici et al. 2017 ).

The ages of users also influence the declared transit safety. Young people’s safety seems not to be a priority (but see e.g. Wiebe et al. 2014 ; Mahmoud and Currie 2010 ; Currie et al. 2013 ). Young people, especially women, are often said to feel particularly unsafe while in transit, especially those who fear sexual harassment and violence (e.g. Natarajan et al. 2017 ; Nourani et al. 2020 ; Gekoski et al. 2015 ; Wiebe et al. 2014 ). Interestingly, the elderly feel less safe in transit environments than the rest of the population (e.g.Loukaitou-Sideris 2009 ; Bon et al. 2009 ) even if they have never previously been victimized (Bradet and Normandeau 1987 ), perhaps because, as suggested by Delbosc and Currie ( 2012 ), age and gender have an indirect effect on safety.

Another characteristic that influences users’ transit safety is the neighborhood context in which they are exposed along the trip. The relationship between surrounding land uses and crime incidents at stations tends to be statistically significant, as certain environmental features either attract offenders (i.e. offer good opportunities) or influence criminal activities by concentrating potential offenders and encouraging anti-social behavior (Loukaitou-sideris 1999 ). Poor perceived safety may reflect more than crime. Evidence from the United States shows that ethnic minorities are more likely to declare they feel unsafe when using public transport, in particular Asians, who show more safety concerns than members of other ethnic groups (e.g.Buckley 1996 ; Loukaitou-Sideris and Eck 2007 ). In addition, travelers from poor neighborhoods have higher levels of fear of crime and property damage in public environments (e.g. Loukaitou-Sideris 2006 ; Vilalta 2011 ). Finally, despite these relationships, keep in mind that social groups are not uniform or stereotypical and that every user is different from another (Loukaitou-Sideris and Fink 2008 ). Hong and Chen ( 2014 ) suggest that residents in safe and high-density areas are more likely to walk and that the built environment is significantly related to walking behavior as well as people’s perception of safety.

3.2.6 Safety door-to-door: the whole journey approach

When we refer to transit safety, the journey not only involves the bus ride or the trip by train; it includes safety throughout the whole trip, from door to door (e.g. Ceccato and Newton 2015 ), for instance, safety in the environments to which an individual is exposed when walking to the bus stop, waiting for the bus to come, inside the carriage on the trip, changing transportation modes, arriving at a destination, and returning back home. Empirical studies are now being informed by this “whole journey approach” dealing with victimization and safety perceptions. Kim et al. ( 2007 ) analyzed factors that influence the mode choice for trips between home and light rail stations and found that the risk of crime made women more likely to be picked up and dropped off at the station, especially when it was dark. Recent research shows that the perceived “walk accessibility” of metro stations was found to be highly influenced by micro-aspects of the landscape, such as streetlights, police patrols, and traffic signs and signals (Bivina et al. 2019 , 2020 ). There are a number of studies that deal with barriers that disrupt mobility, such as gated communities (e.g. Smit et al. 2015 ; Landman 2012 ). In a study by Cao and Duncan ( 2019 ), they show that park-and-riders are willing to walk longer if they experience safer intersections, better pedestrian infrastructure, and an attractive building appearance. Other studies that focused on youth safety, in particular women’s safety, showed different degrees of perceived safety along the trip and by transportation mode (see e.g. Ceccato and Loukaitou-Sideris 2020 ; Evans 2009 ; Natarajan et al. 2017 ).

3.2.7 Strategies and interventions in transit safety

Ensuring transit safety is not a task for a single stakeholder. There is a range of actors that are supposed to share the responsibility of dealing with safety problems together with the major actors in public safety. Recent research recognizes that safety depends on the coordinated cooperation of transportation service providers, safety experts (security guards, security hosts), municipalities, the private sector, and NGOs (e.g. Levin 2015 ; Uittenbogaard and Ceccato 2014 ). Despite the rapid proliferation of these constellations, few efforts have been made to determine their effectiveness in reducing either crime or its related fear in transit environments. Out of the 245 publications, 167 publications deal with interventions or make suggestions for better preventing crime and/or improving safety perceptions. Some of these interventions have their effectiveness tested, while the great majority of authors state them as potential solutions for the pressing safety problems dealt with in their studies. These recommendations can be split into five types: design of transport nodes, policy/cooperation, campaigns/education, data/knowledge, and technology (Table 2 ). Note that our initial intention was to split these interventions into two categories, against crime and for safety perceptions, but this was not possible for two reasons. First, many interventions belonged to a program or package with multiple interventions. Second, there was an overlap of the intention to prevent crime and, at the same time, improve safety perceptions of PT users. Table 2 also lists examples of the most important recommendations.

As much as 26% of the publications suggested improving the knowledge base by building up better data and knowledge. These involve better statistics, codes, surveys, and consultations with groups of users (e.g. d’Arbois de Jubainville and Vanier 2017 ; Smith 2008 ; Riley and Dean 1985 ). Because many of these problems go beyond the transport nodes, also involving the surrounding areas, establishing a better cooperation between public and private actors (e.g. multi-agency police cooperation) is pointed out in 26% of these studies (e.g. Church et al. 2000 ; Ceccato et al. 2020 ; Kenney 1986 ). Other policy changes include female police forces, more patrols, increased supply of public transportation/schedule, women-only carriages/spaces, and stop on demand, as exemplified by Chantal ( 2011 ), Gilbert and Schultz ( 1996 ), or Horii and Burgess ( 2012 ).

Recommendations emphasize that safety is a serious problem often linked to the environment of bus stops and train and subway stations as well as their design. As many as 19% of publications mention crime prevention through environmental design (CPTED), which is largely based on the principles of situational crime prevention and seeks to enhance natural surveillance by modifying the transit environment (e.g. Noble 2015a ; La Vigne 1996 ). They suggest improvement of lighting, enhanced visibility and surveillance via design through mirrors and transparent shelters, and also improving maintenance and changes in the internal land use of these transport nodes. Additional ideas include, for instance, adjustable sizes for carriages and platforms to fit passenger density (e.g. Vanier and d’Arbois 2017 ; Shellow et al. 1974 ), and lighting, because 11 articles agree that sufficient lighting increases the perceived safety but that the fishbowl effect must be taken into account (e.g. Nordfjærn et al. 2014 ; Reed et al. 2000 ). Eight articles claim that visibility should be enhanced by promoting transparent materials, mirrors, a minimum of blind spots, and high arched ceilings (e.g. Mohamed and Stanek 2020 ; Cozens et al. 2004 ). Finally, 19 articles suggest improvements in the environment, such as locating bus stops far from alcohol retailers and high crime streets (Levine and Wachs 1986 ; Liggett et al. 2001 ) Re-plan vegetation, because such cover increases the fear of assault and robbery for women (Cozens et al. 2003b ; Rišová and Madajová 2020 ). Other suggestions are focusing on facilities, claiming that if there are no tables, bathrooms, benches, people do not have the opportunity to loiter in the public space, affecting safety positively (e.g. Gaylord and Galljher 1991 ; Turner 2012 ). Authors also suggest putting more emergency buttons and alarms in transit environments (Debrincat et al. 2017 ; Mahmoud and Currie 2010 ) to facilitate reporting. However, there is some controversy about implementing women-only solutions to tackle sexual crime in transit environments (Shibata 2020 ; Dunckel-Graglia 2013 ).

Although the need for campaigns, education, and information was highlighted by 19% of the publications, limited evidence exists to evaluate their effectiveness. Studies often suggest educational programs, information, media, and awareness campaigns to communicate the level of risk and to sensitize users to the important issues in transit environments (e.g. Loukaitou-Sideris 2006 ; Marteache et al. 2015 ), with particular emphasis on sexual violence (Ceccato and Paz 2017 ; HCEEFH 2015 ). They insist that these campaigns should focus on changing people’s perceptions of other users’ attitudes and not the image of the public transport itself (Fyhri and Backer-Grøndahl 2012 ; Currie et al. 2013 ) or the role of bystanders (Orozco-Fontalvo et al. 2019 ; Lubitow et al. 2020 ). Levin ( 2015 ) mentions the need of developing work on gender mainstreaming and gender impact assessment in transport planning, and in particular more interdisciplinary research as a basis for an increased cooperation between different types of professionals. There are suggestions that focus on agents’ training (e.g. CEMT 2003 ; SRI - Stanford Research Institute 1970 ) and on the promotion of safety and security policies, reminders that the station is under surveillance (Richards 1980 ; Mariotte et al. 2004 ). Finally, 10% of recommendations involve security technology solutions, from CCTV and telephone hotlines, to RTI and newer technologies, such as online platforms and smartphone apps, for example, to encourage people to report transit crimes (e.g. Lea et al. 2017 ). As some of these technologies may have the potential to follow riders door-to-door, ICT technologies can be a useful tool to promote safety along the whole journey. Concerns with staff safety have been globally recognized across professions, and transit staff members are no exception. BWC are assumed to have a “calming effect” on personnel and reduce crime in rail-bound transit (Ariel et al. 2019 ).

4 Conclusion

This article reviewed and analyzed a growing literature on transit safety. We carried out this review covering five decades of studies from 1970 to 2020 written in English primarily, but extending also to the French literature as a benchmark.

Although records of research on transit safety are found in the early 1970s and 1980s, this field of knowledge truly took off in the mid-1990s with simultaneous efforts from disciplines such as Criminology/Psychology and Urban planning/Architecture, especially with efforts by scholars in the United States. Transit safety as a research field steadily developed, to peak after the 2010s, covering a variety of subjects. However, a shift in focus has been noticed, from research on crime and victimization at bus stops and train stations, to the perceived safety of PT users (and to a lesser extent of personnel), focusing on gender and the intersectionality of safety and particular groups. More recently, transportation and ICT are also becoming a distinct field of research. Following different theoretical traditions, transit safety is nowadays an interdisciplinary field of research and is executed by a mixture of sociologists, criminologists, safety experts, architects, urban planners, psychologists, economists, geographers, engineers, computer scientists, and others. The body of research is dominated by the English literature (mostly from large cities, often more “quantitative pieces” using situational crime prevention approaches), with a particular focus on rail-bound environments. While the English-language literature is focused more on the relationship between crime and features of transit environments, the French literature has more of a qualitative nature. French studies adopt a more sociological approach, focusing on individuals’ perspectives on safety (mostly on sexual crimes) as well as the governance of transit safety, in particular the relationship of public and private sectors in delivering transportation/safety services.

A research domain that is recurrent is the one devoted to the transit environment at micro-scale, for example, links between bus stop features and crime. Research over the five decades shows definitive links between the type and quality of lighting and crime and/or perceived safety. Physical features, obstacles, and transparency are all elements associated with passengers’ opportunities to exercise guardianship in transit environments. It is important to notice, however, that targeting the microenvironments of crime (e.g. Ding et al. 2020 ; Cozens et al. 2005 ), may not be enough to adequately address issues of transit safety. Findings from research clearly show that the quality of the social environment, both in the transit nodes and surrounding areas (also their location in the urban environment, i.e. center-periphery), have an impact together on levels of crime and fear of crime experienced by PT users (Smith and Cornish 2006 ). The literature review shows many examples of a multi-pronged approach to tackle these safety problems in transit environments.

Many studies are devoted to the understanding of how passenger flow affects temporal variations in crime and/or perceived safety, because transit nodes, such as bus stops and train stations, vary in the amount of people they concentrate. Thus, it is expected that crime and safety perceptions would also vary over time. However, drawing conclusions from the results in temporal patterns of crime is not an easy task, because findings vary by crime type, hour of the day, day of the week, as well as seasonally—and, not least, in relation to their urban and country contexts. It is worth pointing out that many of the current studies lack data to directly capture the flow of passenger numbers, which means they often rely on proxy measures. Despite a relatively large body of literature we lack a strong empirical evidence base that demonstrates the association between more passengers and more crime. Crowdedness at peak hours tends to facilitate violence, sexual crimes, and pickpocketing. Desolated transit spaces may trigger feelings of insecurity and perhaps create the necessary conditions for crimes that demand anonymity, such as robbery and rape.

Overall, more than half of the publications are devoted to the study of both crime and safety perceptions, followed by publications about crime in transit environments. The remaining studies are about safety perceptions or other topics, such as safety interventions. We identify five common research domains, some with overlapping features.

The relationship between transit environments (type, size, design, social environment) and a wide variety of crimes (assault, rape, robbery, pickpocketing, vandalism, terrorism). Issues of context are also considered here, as well as how to tackle safety problems. Environmental criminology theories (e.g. situational crime prevention, CPTED and routine activity) often provide the theoretical bases for these studies.

Safety perceptions of PT users with a recent clear dominance of women’s transit safety though covering a wide group of riders, including “transit captives.” Fear of crime, overall anxieties related to surveillance, and, more recently, the “whole journey approach” characterize studies working with surveys and a more individualized, psychological perspective.

Temporal aspects of crime and perceived safety —These studies attempt to identify the relationship between passenger flow and safety over time, in particular peak and off-peak hours, seasonal variations, and between weekdays and weekends/holidays/special events. Some identify the potential effects of transport nodes in a neighborhood.

Technology and safety —These studies follow the evaluation of a wide range of products and systems, some not necessarily related to safety, from CCTV cameras to smart cards (e.g. electronic ticketing), real-time information (RTI) systems, body-worn cameras (BWC), and safety apps. This domain also includes “technology” as a crime target, such as mobile phones.

Management practices and crime and safety interventions —These studies covered the effects of interventions, roles of stakeholders, types of cooperation, legal systems, campaigns, and overall performance of crime prevention practices. It is difficult to draw conclusions on specific measures, as “what works” is often dependent on the types of safety problems as well as transportation modes and contexts.

Caution should be exercised when conclusions are drawn from findings. Our search of the literature was limited to the set of keywords (see “ Appendix ” Table 3 ) reflecting the concept of crime and disorder but also the ones related to fear and other overall anxieties. There is a chance that publications on terrorism and minor events, such as fare evasion, may not be captured by the search. Other words that turned out to be relevant during the execution of the study were not part of the initial pool of keywords and may have been misrepresented in the sample of publications, an example was ‘personal security’ or ‘personal safety’. In addition, there is a risk of bias using a combination of systematic search of the English literature (from Scopus and Science direct) complemented by a list of studies provided by expert members of a user list (TCR-network) specialized in the topic. However, we believe that the benefits of having a list of studies from this group of expertise outweigh the risk of bias. In addition, the French literature was mostly collected using Google scholar. Moreover, conclusions are drawn here based on hundreds of publications, but it is important to remember that these publications are more likely to show positive “expected” results than results that are uncertain, negative, or “unexpected.” It is therefore a difficult task to estimate how many studies had negative or unexpected results and were not published, which means it is virtually impossible to estimate the “dark figure” of unpublished materials. Finally, the studies used different methods, which makes it difficult to compare findings. Various studies showed, for instance, that the effect of transit environments on safety was dependent on crime types, levels, and contexts, but it is unclear to what extent these differences might be merely an artefact of the types of employed methods.

Despite the limitations, a literature review of this type can be helpful for researchers, as it provides an up-to-date structured review of the relevant research domains in public transit safety. The review itself “adds value” (Wee and Banister 2016 ), because research gaps have been revealed, it is hoped, functioning as guidance for researchers who plan to do research in this area. A systematic international review of topics of this type can inform practitioners about the most common safety transit problems as well as the types of interventions that have been implemented around the world.

5 Recommendations

Our review of the literature points to several areas where more research is needed on the topic of crime and safety in transit environments, so we conclude this article with recommendations for further research and policy in specific areas.

First, we need to recognize that the literature is quite definitive about the complexity of transit environments and their impact on safety. Our findings show that it is not possible to find a single “physical (or social) characteristic of these environments” that, if tackled, will solve a safety problem. There is no doubt that transit safety demands a situational and dynamic perspective that is framed to the different stages of the trip at micro-level—the bus stop, the station, within the station, and so on—and at particular times. Conditions at any time will affect PT riders, environments, and contexts in different ways, against different types of problems—this is essential to the quality of both research and policy in the future. Another important area for future research is the testing of rigorous longitudinal methodologies that can provide assessments over time and be applied in contexts other than those that were initially tested. In addition, a minority of studies dealt with acts of terrorism in transit environments—an area of research that deserves more attention in the future. In particular for the French literature, more research that would inform planners and transit operators about the importance of understanding the relationship between crime and transit environments is desirable. Data permitting, a more quantitative take on this subject by French researchers would allow a generalization of the results in other contexts.

Second, safety conditions in transport nodes are not the same as the ones experienced or perceived on the way to them. Additionally, if policies are to be tailored to attend to the safety needs of those who declare they are most in fear, “fixing” the problem of victimization does not solve issues related to poor perceived safety. This distinction between crime and safety perception remains, yet it is still treated as unproblematic in transit environment research. Despite a growing literature on women’s safety (Ding et al. 2020 ; Gekoski et al. 2015 ; Kunieda and Gauthier 2007 ; Law 1999 ; Kirchhoff et al. 2007 ; Keane 1998 ), there is still a lack of research on the practical implementation of programs tackling the safety needs of vulnerable groups of PT users (e.g. disabled individuals, elderly, LBQTI status) or, if they exist, they disregard an intersectional perspective. In addition, more research is needed in a variety of areas that include the implications of excessive concern for safety expressed by vulnerable groups, the costs of “women’s safety work” in different country contexts, the normalization of inappropriate behaviors while in transit as well as an in-depth characterization of victims’ agency. There is also the need to further investigate the role of bystanders in PT environments and beyond (applying a “whole journey approach” to safety), and precautionary measures and their long-term impact on life opportunities for transit captives, particularly in contexts where the supply of public transportation is limited, for other suggestions see Ding et al. ( 2020 ). For policy, it is essential that personnel recruitment in transit systems aims at reducing safety inequalities among PT rider groups. Mobility and safety are essential qualities of sustainable cities. Yet, women around the world declare being victims of sexual harassment on a daily basis. Therefore, tailored education programs for university students in Architecture, Engineering, Urban Planning, Transport Sciences—to name a few—are essential to sensitize new generations to the omnipresent problem of unequal transit safety and to make them aware of possible ways to combat it.

Third, the quality of the cooperation between actors (transport operators, police, municipalities, crime prevention councils, NGOs, and private citizens) in tackling transit safety is worth investigation. The role of municipalities is fundamental, because they are responsible for day-to-day spatial planning decisions, ranging from designing a new residential area or subway station, to placing and maintaining physical infrastructure (e.g. streets and sidewalks)—all of high relevance for individuals’ safety and accessibility to public transportation. Safety and mobility with a “whole journey approach” require understanding the barriers of governance that lead to poor cooperation between actors within and across sectors and organizational scales. It is often unclear who is in charge of the safety conditions in the immediate vicinities of transportation nodes or in environments that are privately owned but are public, such as a station. The safety of PT travelers is compromised when the assignment of tasks between these stakeholders is unclear, creating a gray zone in which few actors are willing to take charge of problems or share costs beyond their predefined roles. This demands more than quick fixes to the physical environment at transportation nodes. Despite encouraging results with multi-pronged approaches to transit and safety, research is needed to inform “what works and what does not work” in terms of cooperation between public and private sectors and overall governance of PT. At the same time, achieving cooperation should not be the main goal of getting together. Partnership should be a means to support different actors to solve safety problems at transportation nodes—not an end in itself. Safety as an individual right and a public good needs to be further investigated in the context of mobility.

Fourth, we echo Newton ( 2004 ) on the need for studies that investigate if certain paths on the public transport network facilitate crime. In the past, the work by Belanger ( 1999 ) has investigated this subject by examining how far offenders traveled from their place of residence to the place where they committed subway crimes. The use of big data analysis in PT systems is a growing research area. Information from travel cards can provide information about erratic travel patterns that can be associated with criminal behavior or suicides  (Ceccato et al. 2020 ), yet these research areas are in their infancy. Future research can work out how passengers’ routine mobility patterns can be analyzed using information from travel cards and better detect hotspots of crime by time slices, such as minutes—crucial information for crime prevention. By having the total number of passengers, measures of risk can be more precise in comparison with the traditional traveler-flow estimates. These types of measures will certainly allow us to implement a place-oriented public health perspective to transit safety (e.g. Branas et al. 2018 ) in which crime in PT can be treated as any other preventable harm, such as fall incidents, self-harm/suicides, or acts of terrorism.

Finally, transit safety research will open up research questions for a wider spectrum of sciences, from ethics to computer science. With the use of current technologies (e.g. CCTV, information from travel smart cards, apps, MaaS mobility as a service), on top of the new technologies under way (e.g. micromobility vehicles, virtual reality (VR), artificial intelligence (AI), autonomous vehicles (AV)), solutions as well new safety challenges will appear. One area of research is related to the potential impact of AVs on current PT supply, while prioritizing and accommodating the safety needs of travelers. Assessments of the efficiency of autonomous buses in ensuring the safety of PT riders along the trip and in “the last mile” are already opening a new research frontier. When autonomous buses are implemented, issues of trust and safety by passengers, particularly vulnerable users, will be relevant for further investigation. Not least, concerns about trust and data privacy are bound to become more important than they are today. When they do, it is important to evaluate the opportunities and risks for all types of users. Moreover, there is a need to promote in-depth critical discussions with city planners, transit operators, policy makers, real estate developers, architects, and designers about the impact of AVs on future cities’ designs and PT infrastructures. Research that examines the impact of new technologies on PT users is vital to ensuring mobility and safety for all.

Abenoza RF, Ceccato V, Susilo YO, Cats O (2018) Individual, travel, and bus stop characteristics influencing travelers’ safety perceptions. Transp Res Rec J Transp Res Board 2672:19–28

Article   Google Scholar  

Adams W, Herrmann C, Felson M (2015) Crime, transportation and malignant mixes. In: Newton VCAA (ed) Safety and security in transit environments: an interdisciplinary approach. Basinstoke, Palgrave

Google Scholar  

Alessandrin A, César-Franquet L, Dagorn J (2016) Femmes et déplacements. https://entreleslignesentrelesmots.files.wordpress.com/2017/02/femmes-et-deplacements.pdf

Ariel B, Newton M, McEwan L, Ashbridge GA, Weinborn C, Brants HS (2019) Reducing assaults against staff using body-worn cameras (BWCs) in railway stations. Crim Justice Rev 44:76–93

Atkins ST (1990) Personal security as a transport issue: a state-of-the-art review. Transp Rev 10:111–125

Austin TL, Buzawa E (1984) Citizen perceptions on mass transit crime and its deterrence: a case study. Transp Q 38:103–120

Bachok S, Osman MM, Murad M, Ibrahim M (2014) An assessment of commuters’ perceptions of safety and comfort levels of ‘women-only coach’: the case study of KTM Komuter Malaysia. In: The 4th international conference on sustainable future for human security SUSTAIN 2013, vol 20, pp 197–205

Ball KS, Wesson CJ (2017) Perceptions of unwanted sexual behaviour on public transport: exploring transport density and behaviour severity. Crime Prevent Community Saf 19:199–210

Barjonet P, Gezentsvey M, Mores C (2010) Perception des risques et choix du mode de transport. Approche conceptuelle et modélisation prédictive. Flux 81:19–32

Barjonet P, Gezentsvey M, Mores C (2011) Perception du risque dans les transports en France et en Norvege (PETRIS–WP2)—Etude qualitative et quantitative. Ministere de l’Ecologie, de l’Energie, de Developpement durable et de la Mer Commissariat general du developpement durable Direction de la recherche et de l’innovation

Belanger M (1999) Crime mobility and public transport: the case of the New York City subway. Rutgers University, The State University of New Jersey

Beller A, Garelik S, Cooper S (1980) Sex crimes in the subway. Criminology 18:35–52

Billings SB, Leland S, Swindell D (2011) The effects of the announcement and opening of light rail transit stations on neighborhood crime. J Urban Affairs 33:549–566

Bivina GR, Gupta A, Parida M (2019) Influence of microscale environmental factors on perceived walk accessibility to metro stations. Transp Res Part D Transp Environ 67:142–155

Bivina GR, Gupta A, Parida M (2020) Walk accessibility to metro stations: an analysis based on meso- or micro-scale built environment factors. Sustain Cities Soc 55:102047

Block R, Davis S (1996) The environs of rapid transit stations: a focus for street crime or just another risky place? Preventing mass transit crime. Criminal Justice Press, Monsey, pp 237–257

Bon D, Scherer S, Zauberman R, Robert P, Heurtel H (2009) L'insécurité en Île-de-France. Rapport final: enquête: victimation et sentiment d’insécurité

Bonnet F (2008) Les effets pervers du partage de la sécurité. Polices publiques et privées dans une gare et un centre commercial. Sociol Trav 50:505–520

Bonnet F (2006) La production organisée de l'ordre. Contrôler des gares et des centres commerciaux à Lyon et à Milan. Sociologie, Institut d'études politiques de Paris—Sciences Po

Bowers K (2014) Risky facilities: crime radiators or crime absorbers. A comparison of internal and external levels of theft. J Quant Criminol 30:389–414

Bradet C, Normandeau A (1987) La criminalité et les transports publics. Déviance Soc 11:105–122

Branas CC, South E, Kondo MC, Hohl BC, Bourgois P, Wiebe DJ, Macdonald JM (2018) Citywide cluster randomized trial to restore blighted vacant land and its effects on violence, crime, and fear. Proc Natl Acad Sci USA 115:2946–2951

Brantingham P, Brantingham P (1995) Criminality of place—crime generators and crime attractors. Eur J Crim Policy Res 3:5–26

Browne TF (2010) La perception du risque terroriste et de ses conséquences sur la gestion de la sécurité dans le système de transport en commun de Montréal

Buckley JB (1996) Public transit and crime: a routine activities/ecological approach. Simon Fraser University

Burnett J, Pang AY-H (2004) Design and performance of pedestrian subway lighting systems. Tunn Undergr Space Technol 19:619–628

Burrows J (1980) Closed-circuit television and crime on the London Underground. In: Clarke RV, Mayhew P (eds) Designing out crime. HMSO, London, pp 75–83

Castagnino F (2016) Séparer pour mieux surveiller. Spatialité des risques et pratiques de surveillance en Gare du Nord. Flux 103–104:44–56

CEMT (2003) Vandalisme, terrorisme et sûreté dans les transports publics urbains de voyageurs, Éditions. OCDE

CRIME-CONCERN (2004) People’s perceptions of personal security and their about crime on public transport: research findings. Dft, London

Cao J, Duncan M (2019) Associations among distance, quality, and safety when walking from a park-and-ride facility to the transit station in the Twin cities. J Plan Educ Res 39:496–507

Capasso da Silva D, Rodrigues da Silva AN (2020) Sustainable modes and violence: perceived safety and exposure to crimes on trips to and from a Brazilian university campus. J Transp Health 16:100817

Casas I, Delmelle EC (2017) Tweeting about public transit—gleaning public perceptions from a social media microblog. Case Stud Transp Policy 5:634–642

Ceccato V (2013) Moving safely: crime and perceived safety in Stockholm’s subway stations. Lexington, Plymouth

Ceccato V (2017a) Women’s transit safety: Making connections and defining future directions in research and practice. Crime Prevent Community Saf 19:276–287

Ceccato V (2017b) Women’s victimisation and safety in transit environments. Crime Prev Community Saf 19:163–167

Ceccato V (2019) Eyes and apps on the streets: from surveillance to sousveillance using smartphones. Crim Justice Rev 44:25–41

Ceccato V, Cats O, Wang Q (2015) The geography of pickpocketing at bus stops: an analysis of grid cells. In: Ceccato V, Newton A (eds) Safety and security in transit environments. Palgrave Macmillan UK, London

Chapter   Google Scholar  

Ceccato V, Loukaitou-Sideris A (2020) Transit crime and sexual violence in cities: international evidence and prevention. Routledge, London

Book   Google Scholar  

Ceccato V, Masci S (2017) Airport environment and passengers’ satisfaction with safety. J Appl Secur Res 12:356–373

Ceccato V, Nalla MK (2020) Crime and fear in public places: towards safe, inclusive and sustainable cities. Taylor & Francis Group, New York

Ceccato V, Paz Y (2017) Crime in São Paulo’s metro system: sexual crimes against women. Crime Prev Community Saf 19:211–226

Ceccato V, Uittenbogaard AC (2013) Space-time dynamics of crime in transport nodes. Ann Assoc Am Geogr 104:131–150

Ceccato V, Uittenbogaard A, Bamzar R (2011) Security in Stockholm’s underground stations: The importance of environmental attributes and context. Secur J 26:33–59

Ceccato V, Wiebe D, Vrotsou K, Nyberg U, Grundberg A (2020) The situational conditions of suicide in transit environments: an analysis using CCTV footage. J Transp Health. https://doi.org/10.1016/j.jth.2020.100976

Ceccato V, Wiebe DJ, Eshraghi B, Vrotsou K (2017) Women’s mobility and the situational conditions of rape: cases reported to hospitals. J Interpers Violence 35(15–16):2917–2946. https://doi.org/10.1177/0886260517699950

Ceccato V (2018) Crime in transit environments: lessons from Stockholm (Sweden) and São Paulo (Brazil) metro systems: 交通环境犯罪: 从瑞典斯德哥尔摩和巴西圣保罗 地铁系统的经验总结. Landscape architecture

Ceccato V, Newton A (2015) Safety and security in transit environments: an interdisciplinary approach. Palgrave Macmillan, New York

Ceccato V, Näsman P, Langefors L (2020) Sexual violence on the move: an assessment of youth’s victimization in public transportation. Women Crim Justice. https://doi.org/10.1080/08974454.2020.1733732

Chaiken JM, Lawless M, Stevenson KA (1974) The impact of police activity on crime: robberies on the New York city subway system. Santa Monica, CA: RAND Corporation. https://www.rand.org/pubs/reports/R1424.html

Chantal D (2011) Transport et parité des sexes. Forum International des Transports. https://m.centre-hubertine-auclert.fr/sites/default/files/fichiers/transports-et-parite-des-sexes.pdf

Chowdhury S, Van Wee B (2020) Examining women’s perception of safety during waiting times at public transport terminals. Transp Policy 94:102–108

Church A, Frost M, Sullivan K (2000) Transport and social exclusion in London. Transp Policy 7:195–205

Clarke RV (ed) (1983) Situational crime prevention: its theoretical basis and practical scope. University of Chicago Press, Chicago

Clarke RV, Belanger M, Eastman JA (1996) Where angel fears to tread: a test in the New York city subway of the robbery/density hypothesis. Crime prevention studies: preventing mass transit crime. Criminal Justice Press, Monsey

Clarke R, Eck J (2007) Understanding risky facilities [Online]. Center for problem-oriented policing. https://popcenter.asu.edu/content/understanding-risky-facilities . Accessed 2020

Cohen LE, Felson M (1979) Social change and crime rate trends: a routine activity approach. Am Sociol Rev 44:588–608

Colard F (2018) Femmes et transports en commun. Des parcours de combattantes? http://www.femmesprevoyantes.be/wp-content/uploads/2018/12/transports-en-commun-WEB.pdf

Coppola P, Silvestri F (2020) Assessing travelers’ safety and security perception in railway stations. Case Stud Transp Policy 8:1127–1136

Cozens PM, Neale RH, Whitaker J, Hillier D (2002) Investigating perceptions of personal security on the Valley Lines rail network in South Wales (UK). World Transp Policy Pract 8(1):19–29

Cozens P, Neale R, Hillier D, Whitaker J (2004) Tackling crime and fear of crime while waiting at Britain’s Railway stations. J Public Transp 7:23–41

Cozens P, Neale R, Whitaker J, Hillier D (2003a) Investigating personal safety at railway stations using “virtual reality” technology. Facilities 21:188–194

Cozens P, Neale R, Whitaker J, Hillier D (2003b) Managing crime and the fear of crime at railway stations––a case study in South Wales (UK). Int J Transp Manag 1:121–132

Cozens PM, Saville G, Hillier D (2005) Crime prevention through environmental design (CPTED): a review and modern bibliography. Prop Manag 23:328–356

Crenshaw K (1989) Demarginalizing the intersection of race and sex: a black feminist critique of antidiscrimination doctrine, feminist theory and antiracist politics. University of Chicago Legal Forum. 139–167

Crossonneau N (2003) L’insécurité dans les transports publics urbains: bibliographie commentée. Centre d’études sur les réseaux, les transports, l’urbanisme et les constructions publiques (CERTU). https://hal-lara.archives-ouvertes.fr/hal-02150521/document

Currie G, Delbosc A, Mahmoud S (2013) Factors influencing young peoples’ perceptions of personal safety on public transport. J Public Transp 16:23–42

D’arbois H (2019) Sécurité dans les transports en commun: Le cas des usagers captifs. Note Rapide. 33. ONDRP, Paris

d’Arbois de Jubainville H, Vanier C (2017) Women’s avoidance behaviours in public transport in the Ile-de-France region. Crime Prev Community Saf 19:183–198

Debrincat M, Dupart C, Laurent C (2016) Étude sur le harcèlement sexiste et les violences sexuelles faites aux femmes dans les transports publics. FNAUT, Paris

Debrincat M, Dupart S, Moggio C (2017) Harcèlement sexiste dans les transports collectifs routiers et les pôles d’échanges multimodaux: L’analyse de la Fnaut, Paris

Delbosc A, Currie G (2012) Modelling the causes and impacts of personal safety perceptions on public transport ridership. Transp Policy 24:302–309

Diec J, Coxon S, De Bono A (2010) Designing a public train station shelter to minimise anti-social behaviour and crime in Melbourne's metropolitan rail environment. In: ATRF 2010: 33rd Australasian transport research forum, 2010

Ding H, Loukaitou-Sideris A, Agrawal AW (2020) Sexual harassment and assault in transit environments: a review of the English-language literature. J Plan Lit 35(3):267–280

Dunckel-Graglia A (2013) Women-only transportation: How “pink” public transportation changes public perception of women’s mobility. J Public Transp 16:85–105

Dunckel-Graglia A (2016) Finding mobility: women negotiating fear and violence in Mexico City’s public transit system. Gend Place Cult 23:624–640

Eck JE (2019) Place Managers and Crime Places. Oxford University Press

Evans G (2009) Accessibility, urban design and the whole journey environment. Built Environ 1978(35):366–385

Felson M, Belanger ME, Bichler GM, Bruzinski CD, Campbell GS, Fried CL, Grofik KC, Mazur IS, O'regan AB, Sweeney PJ, Ullman AL, Williams LM (1996) Redesigning hell: preventing crime and disorder at the port authority bus terminal. Prev Mass Transit Crime

Fillone AM, Mateo-Babiano I (2018) Do I walk or ride the rickshaw? Examining the factors affecting first- and last-mile trip options in the historic district of Manila (Philippines). J Transp Land Use 11:237–254

Fyhri A, Backer-Grøndahl A (2012) Personality and risk perception in transport. Accid Anal Prev 49:470–475

Gaylord MS, Galliher JF (1991) Riding the underground dragon: crime control and public order on Hong Kong’s mass transit railway. Br J Criminol 31:15–26

Gekoski A, Jacqueline M, Gray MH, Horvath SE, Aliye E, Adler J (2015) ‘What works’ in reducing sexual harassment and sexual offences on public transport nationally and internationally: a rapid evidence assessment. In: Transport BTPADF (ed). London

Gentry K (2015) Apple picking: the rise of electronic device thefts in boston subways. Macmillan, Palgrave

Gerell M (2018) Bus stops and violence, are risky places really risky? Eur J Crim Policy Res 24:351–371

Gilbert S, Schultz D (1996) Women and transit security: a new look at an old issue. In: Gilbert SSD (ed) Women’s travel issues second national conference: Baltimore

Gosselin C (2012) De la conception à la gestion des espaces en gare: La prévention au coeur de la stratégie en sûreté. IAU-ÎdF, Paris

Grandmaison R, Tremblay P (1997) Évaluation des effets de la télé-surveillance sur la criminalité commise dans 13 stations du métro de Montréal. Criminologie 30:93–110

le Grâet A, Vanier C (2016) Les atteintes personnelles dans les transports en commun: Résultats basés sur les déclarations des personnes de 14 ans et plus, interrogées lors des enquêtes «Caedre de vie et sécurité» 2009–2013. ONDRP, Paris

Gu H, Chen P, Guo Y, Yao M, Yang H, Hou J (2019) Detecting pickpocketing offenders by analyzing beijing metro subway data. In: 2019 the 4th IEEE international conference on big data analytics, March. 15–18. Suzhou, China

Guimarães T, Lucas K, Timms P (2019) Understanding how low-income communities gain access to healthcare services: a qualitative study in São Paulo, Brazil. J Transp Health 15:100658

Gómez AJ, Haarhoff E, García E (2020) Highway expansion and crime: challenges on urban development for sustainability. In: Bougdah HVA, Sotoca A, Trapani F, Migliore M, Clark N (eds) Urban and transit planning. Springer, Cham, pp 133–143

HCEEFH (2015) Avis sur le harcèlement sexiste et les violences sexuelles dans les transports en commun: Se mobiliser pour dire stop sur toute la ligne au harcèlement sexiste et aux violences sexuelles dans les transports. HCEFH, Paris

Hale C (1996) Fear of crime: a review of the literature. Int Rev Victimol 4:79–150

Hamelin F (2010) Les polices des transports face aux défis croisés de la mobilité, de la sûreté et de la sécurité. Flux 81:46–56

Hart TC, Miethe TD (2014) Street robbery and public bus stops: a case study of activity nodes and situational risk. Secur J 27:180–193

Higgins JPT, Green S (2011) Cochrane handbook for systematic reviews of interventions. In: Higgins JPTGS (ed) The cochrane collaboration. http://www.mri.gov.lk/assets/Uploads/Research/Cochrane-Hand-booktext.pdf

Hirtenlehner H, Farrall S (2014) Is the ‘shadow of sexual assault’ responsible for women’s higher fear of burglary? Br J Criminol 54(6):1167–1185

Hong J, Chen C (2014) The role of the built environment on perceived safety from crime and walking: examining direct and indirect impacts. Transportation 41:1171–1185

Horii M, Burgess A (2012) Constructing sexual risk: ‘Chikan’, collapsing male authority and the emergence of women-only train carriages in Japan. Health Risk Soc 14:41–55

Hsu H-P (2011) How does fear of sexual harassment affect women’s use of transit? Volume II: technical papers. In: Hsu H-P (ed) Proceedings of the 4th international conference on women’s issues in transportation. Washington, D.C., pp 85–94

Ihlanfeldt KR (2003) Rail transit and neighborhood crime: the case of Atlanta, Georgia. South Econ J 70:273–294

Ingalls GL, Hartgen D, Owens T (1993) Public fear of crime and its role in bus transit use. Transp Res Rec 1433:201–211

Iudici A, Bertoli L, Faccio E (2017) The ‘invisible’ needs of women with disabilities in transportation systems. Crime Prev Community Saf 19:264–275

Iudici A (2015) Sexual harassment against people with mental disabilities in transit environments: implications for services and clinics. In: Ceccato V, Newton A (eds) Safety and security in transit environments: an interdisciplinary approach. Palgrave Macmillan UK, London, pp 328–343

Jun P, Subeh C, Douglas W (2020) Gap between policymakers’ priorities and users’ needs in planning for accessible public transit system. J Transp Eng Part A Syst 146:04020020

Kash G (2019) Always on the defensive: the effects of transit sexual assault on travel behavior and experience in Colombia and Bolivia. J Transp Health 13:234–246

Keane C (1998) Evaluating the influence of fear of crime as an environmental mobility restrictor on women’s routine activities. Environ Behav 30:60–74

Kenney DJ (1986) Crime on the subways: Measuring the effectiveness of the guardian angels. Justice Q 3:481–496

Kim S, Ulfarsson GF, Hennessy JT (2007) Analysis of light rail rider travel behavior: impacts of individual, built environment, and crime characteristics on transit access. Transp Res Part A Policy Pract 41:511–522

Kirchhoff GF, Morosawa H, Barkhuizen J, Bussinger C, Daigaku T, Sutseyo H, Bey JF, Ungul E (2007) The Asian passengers’ safety study of sexual molestation on trains and buses: the Indonesian pilot study. Acta Criminol 20:1–13

Kooi BR (2013) Assessing the correlation between bus stop densities and residential crime typologies. Crime Prev Community Saf 15:81–105

Kunieda M, Gauthier A (2007) Gender and urban transport: smart and affordable. In: Module 7a. Sustainable transport sourcebook for policy makers in developing countries. GTZ, Eschborn

La Vigne N (1996) Safe transport: security by design on the washington metro. In: Preventing Mass Transit Crime, Clarke RV (ed.). US Department of Justice, pp 163–197

La Vigne NG (1997) Visibility and vigilance: metro's situational approach to preventing subway crime. National Institute of Justice-Research in Brief. U.S. Department of Justice

La Vigne N (2015) Crime in and around metro transit stations: Exploring the utility opportunity theories of crime. In: Newton VCA (ed) Safety and security in transit environments: an interdisciplinary approach. Palbrage, Basingstoke

Landman K (2012) Reconsidering crime and urban fortification in South Africa. In: Ceccato V (ed) The urban fabric of crime and fear. Springer, Dordrecht, pp 85–110.

Law R (1999) Beyond ‘women and transport’: towards new geographies of gender and daily mobility. Prog Hum Geogr 23:567–588

Lea SG, D’Silva E, Asok A (2017) Women’s strategies addressing sexual harassment and assault on public buses: an analysis of crowdsourced data. Crime Prev Community Saf 19:227–239

Levin L (2015) Gender equality and safety, a challenge for transport planning: moving away from stereotypes and stereotypical attitudes anda habits. In: Newton VCA (ed) Safety and security in transit environments: an interdisciplinary approach. Palgrave, Basingstoke

Levine NWM (1986) Bus crime in Los Angeles: II—victims and public impact. Transp Res Part A General 20:285–293

Levine N, Wachs M, Shirazi E (1986) Crime at bus stops: a study of environmental factors. J Archit Plan Res 3:339–361

Levine N, Wachs M (1985) Factors affecting the incidence of bus crime in Los Angeles, vol I and II

Levine N, Wachs M (1986) Tracking crime on buses. TR News 18–22

Liggett R, Loukaitou-Sideris A, Iseki H (2001) Bus stop-environment connection: do characteristics of the built environment correlate with bus stop crime? Transp Res Rec J Transp Res Board 1760:20–27

Liggett R, Loukaitou-Sideris A, Iseki H (2003) Journeys to crime: assessing the effects of a light rail line on crime in the neighborhoods. J Public Transp 6:85–115

Liu L, Jiang C, Zhou S, Liu K, Du F (2017) Impact of public bus system on spatial burglary patterns in a Chinese urban context. Appl Geogr 89:142–149

Liu L, Lan M, Eck JE, Kang EL (2020) Assessing the effects of bus stop relocation on street robbery. Comput Environ Urban Syst 80:101455

Loukaitou-Sideris A (1999) Hot spots of bus stop crime: the importance of environmental attributes. J Am Plann Assoc 65:395–411

Loukaitou-Sideris A (2006) Is it safe to walk? Neighborhood safety and security considerations and their effects on walking. J Plann Lit 20:219–232

Loukaitou-Sideris A (2012) Safe on the move: the importance of the built environment. In: Ceccato V (ed) The urban fabric of crime and fear. Springer, Dordrecht

Loukaitou-Sideris A, Eck JE (2007) Crime prevention and active living. Am J Health Promot 21:380–389

Loukaitou-Sideris A, Fink C (2008) Addressing women’s fear of victimization in transportation settings: a survey of U.S. Transit Agencies Urban Affairs Rev 44:554–587

Loukaitou-Sideris A, Liggett R, Iseki H (2002) The geography of transit crime: documentation and evaluation of crime incidence on and around the green line stations in Los Angeles. J Plan Educ Res 22:135–151

Loukaitou-Sideris A, Liggett R, Iseki H, Thurlow W (2001) Measuring the effects of built environment on bus stop crime. Environ Plan B Plan Des 28:255–280

Loukaitou-Sideris A (2008) How safe is the ride? Evaluation of design and policy responses to women’s fear of victimization and crime. Department of Urban Planning, UCLA School of Public Affairs, Los Angeles

Loukaitou-Sideris (2009) How to ease women’s fear of transportation environments: case studies and best practices. Mineta Transportation Institute: Report 09-01, San Jose

Lubitow A, Abelson MJ, Carpenter E (2020) Transforming mobility justice: gendered harassment and violence on transit. J Transp Geogr 82

Lynch G, Atkins S (1988) The influence of personal security fears on women’s travel patterns. Transportation 15:257–277

Madan M, Nalla M (2015) Sexual harassment in public spaces: examining gender differences in perceived seriousness and victimization. Int Crim Justice Rev 26(2):80–97

Mahmoud S, Currie (2010) The relative priority of personal safety concerns for young people on public transport. In: Mahmoud S, Currie G (eds) Australasian transport research forum conference. Canberra, Australia

Malochet V (2015) La gouvernance de la sécurité dans les transports publics franciliens. IAU-ÎdF, Paris

Malochet V, le Goff T (2013) Étude sur la sécurisation des transports publics franciliens. IAU-ÎdF, Paris

Mariotte S (2004) La vidéosurveillance dans la sécurisation des transports collectifs franciliens. IAURIF, Paris

Mariotte S, Heurtel H, le Delliou S (2004) Évaluation de l’impact de la vidéosurveillance sur la sécurisation des transports en commun en région Île-de-France. IAURIF, Paris

Marteache N (2018) “Please refrain from locking your bags”: theft from passengers’ checked-in luggage at US airports. Secur J 31:163–189

Marteache N, Bichler G, Enriquez J (2015) Mind the gap: perceptions of passenger aggression and train car supervision in a commuter rail system. J Public Transp 18:61–73

McCarthy OT, Caulfield B, O’Mahony M (2016) How transport users perceive personal safety apps. Transp Res Part F Traffic Psychol Behav 43:166–182

Mitra-Sarkar S, Partheeban P (2011) Abandon all hope, ye who enter here: understanding the problem of “eve teasing” in Chennai, India. In: Proceedings of the 4th international conference on women’s issues in transportation. Volume II. Technical Papers, Transportation Research Board, Washington, DC, pp 74–84

Mohamed AA, Stanek D (2020) The influence of street network configuration on sexual harassment patterns in Cairo. Cities 98:102583

Moher D, Liberati A, Tetzlaff J, Altman D (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 6:e1000097

Moore S (2011) Understanding and managing anti-social behaviour on public transport through value change: the considerate travel campaign. Transp Policy 18:53–59

Moreira GC, Ceccato V (2020) Gendered mobility and violence in the Sa˜o Paulo metro. Brazil. Urban Studies. 58(1):203–222

Natarajan M (2016) Rapid assessment of “eve teasing” (sexual harassment) of young women during the commute to college in India. Crime Sci 5:1–11

Natarajan M, Schmuhl M, Sudula S, Mandala M (2017) Sexual victimization of college students in public transport environments: a whole journey approach. Crime Prev Community Saf 19:168–182

Newton A (2004) Crime on public transport: ‘Static’ and ‘Non-Static’ (moving) crime events. Western Criminol Rev 5(3):25–42

Newton A (2008) A study of bus route crime risk in urban areas: the changing environs of a bus journey. Built Environ 34:88–103

Newton A (2018) Macro-level generators of crime, including parks, stadiums, and transit stations. The Oxford Handbook of Environmental Criminology, Oxford University Press, Oxford

Newton AD, Johnson SD, Bowers KJ (2004) Crime on bus routes: an evaluation of a safer travel initiative. Polic Int J Police Strat Manag 27:302–319

Newton A, Loukaitou-Sideris A, D’arbois De Jubainville H, Huang D, Norgaard JK, Solymosi R (2020) Precautions and responses. In: Ceccato V, Loukaitou-Sideris A (eds) Transit crime and sexual violence in cities: international evidence and prevention. Routledge, New York, pp. 276–294

Newton AD, Partridge H, Gill A (2014) Above and below: measuring crime risk in and around underground mass transit systems. Crime Sci 3:1–14

Newton AD, Partridge H, Gill A (2014) In and around: Identifying predictors of theft within and near to major mass underground transit systems. Secur J 27:132–146

Newton A (2014) Crime on public transport. In: Bruinsma G, Weisburd D (eds) Encyclopedia of criminology and criminal justice. Springer, New York, NY. https://doi.org/10.1007/978-1-4614-5690-2_301

Newton AD (2007) Routing out the hot-spots: toward using GIS and crime-place principles to examine criminal damage to bus shelters. In: GIS and Evidence-Based Policy Making. Innovations in GIS, 11. CRC Press, London, UK, pp 69–94

Newton A (2016) Transport technology and crime. In: The handbook of technology, crime and justice. Routledge, Routledge, Abingdon 2017, pp 281–294

Newton A (2018) Macro-level generators of crime, including parks, stadiums, and transit stations. In: Bruinsma GJN, Johnson SD (eds) The Oxford handbook of environmental criminology. Oxford University Press, New York, pp 497–517

Newton A, Bowers K (2007) The geography of bus shelter damage: the influence of crime, neighbourhood characteristics and land-use. Internet J Criminol. http://eprints.hud.ac.uk/id/eprint/474/

Noble J (2015a) L’insécurité personnelle dans les transports en commun. Déviance Soc 39:343–364

Noble J (2015b) L’insécurité personnelle des jeunes dans les transports en commun franciliens: une étude dispositionnelle du sentiment d’insécurité. Doctoral Dissertation. Université Paris-Nord

Noble J, Fussy F (2020) Le sentiment d'insécurité dans les transports collectifs franciliens en 2019. 845. Note rapide de l'Institut Paris Région. Institut Paris Région, Paris

Nordfjærn T, Şimşekoğlu Ö, Lind HB, Jørgensen SH, Rundmo T (2014) Transport priorities, risk perception and worry associated with mode use and preferences among Norwegian commuters. Acci Anal Prev 72:391–400

Nourani F, Antonello SL, Govone JS, Ceccato V (2020) Women and LGBTI youth as targets assessing transit safety in Rio Claro, Brazil. In: Ceccato V, Loukaitou-Sideris A (eds) Transit crime and sexual violence in cities international evidence and prevention. Routledge, New York

Orozco-Fontalvo M, Soto J, Arévalo A, Oviedo-TrespalacioS O (2019) Women’s perceived risk of sexual harassment in a Bus Rapid Transit (BRT) system: the case of Barranquilla, Colombia. J Transp Health 14:100598

Otis MD (2007) Perceptions of victimization risk and fear of crime among lesbians and gay men. J Interpers Viol 22:198–217

Otu SE, Agugua A (2020) An analysis of transit safety among college students in Lagos, Nigeria. In: Ceccato V, Nalla M (eds) Crime and fear in public places: towards safe, inclusive and sustainable cities. Routledge, London

Ouimet M, Tremblay P (2001) Trajets urbains et risques de victimisation: les sites de transit et le cas du métro de Montréal. Criminologie 34:157–176

Pearlstein A, Wachs M (1982) Crime in public transit systems: an environmental design perspective. Transportation 11:277–297

Phillips DC, Sandler D (2015) Does public transit spread crime? Evidence from temporary rail station closures. Reg Sci Urban Econ 52:13–26

Piza EL, Welsh BC, Farrington DP, Thomas AL (2019) CCTV surveillance for crime prevention. Criminol Public Policy 18:135–159

Piza EL, Kennedy D (2003) Transit stops, robbery, and routine activities: examining street robbery in the Newark, NJ Subway Environment

Poister TH (1996) Transit-related crime in suburban areas. J Urban Affairs 18:63–76

Priks M (2009) The effect of surveillance cameras on crime: evidence from the stockholm subway. Department of Economics Stockholm University, Stockholm

PriyaUteng T, Singh YJ, Lam T (2019) 12-safety and daily mobilities of urban women—methodolgies to confront the policy of “invisibility.” In: Lucas K, Martens K, Diciommo F, Dupont-Kieffer A (eds) Measuring transport equity. Elsevier, Amterdam

Reed TB, Wallace RR, Rodriguez DA (2000) Transit passenger perceptions of transit-related crime reduction measures. Transp Res Rec J Transp Res Board 1731:130–141

Richards LJI (1980) Passenger value structure model. Final report DOT contract TSC-1314. Dunlap and Associates, Inc., Darien

Riley NE, Dean DL (1985) Bus station security: crime at intercity bus stations. Transp Res Rec 1012:56–64

Rišová K, Madajová MS (2020) Gender differences in a walking environment safety perception: a case study in a small town of Banská Bystrica (Slovakia). J Transp Geogr 85:102723

Romero-Torres J, Ceccato V (2020) Youth safety in public transportation: the case of eastern Mexico City, Mexico. In: Ceccato V, Nalla M (eds) Crime and fear in public places: towards safe, inclusive and sustainable cities. Routledge, London, pp 145–159

Scarborough Women’s Centre/METRAC (1991) Making transit stops safer for women: Scarborough moves forward. Scarborough Women’s Centre/METRAC, Scarborough

Sedelmaier CM (2014) Offender-target redistribution on a new public transport system. Secur J 27:164–179

Sedelmaier CM (2003) Railroaded: the effects of a new public transportation system upon local crime patterns. Rutgers University, The State University of New Jersey.

Sham R, Hussein MZSM, Ismail HN (2013) A dilemma of crime and safety issues among vulnerable travellers in Malaysian urban environment. In: AicE-Bs 2013 London (Asia Pacific International Conference on Environment-Behaviour Studies), 4–6 September 2013, vol 105, pp 498–505

Shellow R, Romualdi JP, Bartel EW (1974) Crime in rapid transit systems: an analysis and a recommended security and surveillance system. Transp Res Rec 487:1-2.

Shen LD, Ospina D, Zhao F, Elbadrawi H (1997) Analyses of technologies and methodologies adopted by US transit agencies to enhance transit security. NUTI report, NUTI96FIU1.2

Shibata S, Hanyu K, Doi Hata T, Yamaoka Y (2014) Expectation and perception of crime and disorder events in railway stations in Tokyo. Secur J 27:210–225

Shibata S (2020) Are women-only cars (WOC) a solution to groping? A survey among college students in Tokyo/Kanagawa, Japan. Int J Comp Appl Crim Justice 44:293–305

Smit T, Landman K, Venter C (2015) The impact of crime and neighbourhood enclosures on travel behaviour and transport patterns in South Africa. In: Newton VCA (ed) Safety and security in transit environments: an interdisciplinary approach. Palgrave, Basingstoke

Smith MJ (2008) Addressing the security needs of women passengers on public transport. Secur J 21:117–133

Smith M, Clarke R (2000) Crime and public transport. Crime Justice 27:169–233

Smith MJ, Cornish DB (2006) Secure and tranquil travel—preventing crime and disorder on public transport. UCL Jill Dando Institute of Crime Science, London

Smith D, Winstanley S, Levine N (2013) Targeting crime on public transport: an example from Greater Manchester, England. In: CrimeStat IV: a spatial statistics program for the analysis of crime incident locations (version 4.0) . Ned Levine & Associates ed, Houston

Smith M (1986) Transit crime study volume II: final report. U.S. Department of Justice National Institute ot Justice, New York

Sochor J (2012a) Mobility-enhancing ICT from an ethical perspective: the case of a navigation system for visually impaired persons. J Transp Res Board

Sochor J (2012b) Impacts of intelligent transportation systems on users’ mobility: a case study analysis. Licentiate Thesis in Transport Science, Stockholm, Sweden

Solymosi R, Borrion H, Fujiyama T (2015) Crowd spatial patterns at bus stops: secuirty implications and effects of warning messages. In: Newton VCA (ed) Safety and security in transit environments: an interdisciplinary approach. Palgrave, Basingstoke, pp 156–178

Solymosi RCK, Newton A (2018) Did they report it to stop it? A realist evaluation of the effect of an advertising campaign on victims’ willingness to report unwanted sexual behaviour. Secur J 31:570–590

Spicer V, Song J (2017) The impact of transit growth on the perception of crime. J Environ Psychol 54:151–159

SRI - Stanford Research Institute (1970) Reduction of robberies and assaults of bus drivers, vol. 1. Summary and Conclusions. National Technical Information Service, Springfield, VA

Stringer MS (2007) Hidden in plain sight: Sexual harassment and assault in the New York City subway system. Office of the Manhattan Borough President, New York

Sung-Suk VY (2009) Bus stops and crime: do bus stops increase crime opportunities in local neighborhoods? The State University of New Jersey, Rutgers

Swain S (2015) Securing the transport system. In: Routledge handbook of law and terrorism. Routledge

TCRP (2001) Guidelines for collecting, analyzing, and reporting transit crime data. Transportation Research Board, National Research Council, Washington, D.C.

Tilley N (1993) Understanding car parks, crime, and CCTV: evaluation lessons from safer cities. Home Office Police Dept., London

Tripathi K, Borrion H, Belur J (2017) Correction to: Sexual harassment of students on public transport: an exploratory study in Lucknow, India. Crime Prev Community Saf 20:223–223

Turner J (2012) Urban mass transit, gender planning protocols and social sustainability—the case of Jakarta. Gend Transp Trans Costs Compet Claims Transp Policy Gaps 34:48–53

UN-Habitat TSCP (2019) Safer cities programme [Online]. UN-Habitat, New York. Accessed 12 Feb 2020

Uittenbogaard A, Ceccato V (2014) Safety in Stockholm’s underground stations: an agenda for action. Eur J Crim Policy Res 20:73–100

Uittenbogaard A (2015) An assessment of guardianship opportunities as provided by the environments of transit stations. In: Newton VCA (ed) Safety and security in transit environments: an interdisciplinary approach. Palgrave, Basingstoke, pp 56-75

Umaña-Barrios N, Gil AS (2017) How can spatial design promote inclusivity, gender equality and overall sustainability in Costa Rica's urban mobility system? In: Urban transitions conference, Shanghai, September 2016, vol 198, pp 1018–1035

Vanier C, D’arbois H (2018b) Le sentiment d’insécurité dans les transports en commun: Situations anxiogènes et stratégies d’évitement. ONDRP, Paris

Vanier C, D’Arbois H (2017) Feeling unsafe in public transportation: a profile analysis of female users in the Parisian region. Crime Prev Community Saf 19:251–263

Vanier C, D’arbois H (2018a) Le sentiment d’insécurité dans les transports en commun: situations anxiogènes et stratégies d’évitement. ONDRP, Paris

Vigne NGL (1996) Safe transport: security by design on the washington metro. In: Crime prevention studies: preventing mass transit crime, vo1 6. Criminal Justice Press, Monsey

Vilalta CJ (2011) Fear of crime in public transport: research in Mexico City. Crime Prev Community Saf 13:171–186

Wallace RR, Rodriguez DA, White C, Levine J (1999) Who noticed, who cares? Passenger reactions to transit safety measures. Transp Res Rec 1666:133–138

Wang B, Zacharias J (2020) Noise, odor and passenger density in perceived crowding in public transport. Transp Res Part A Policy Pract 135:215–223

Webb B, Laycock GK (1992) Reducing crime on the London Underground: an evaluation of three pilot projects. Home Office, London

Weber BS (2019) Uber and urban crime. Transp Res Part A Policy Pract 130:496–506

Wee BV, Banister D (2016) How to write a literature review paper? Transp Rev 36:278–288

Weidner R (1996) Target-hardening at a new york city subway station: decreased fare evasion—at what price? In: Clarke RV (eds) Preventing mass transit crime. Crime Prevention Studies, Vol. 6, Monsey, NY: Criminal Justice Press, pp 5–91

Whitzman C (2007) Stuck at the front door: gender, fear of crime and the challenge of creating safer space. Environ Plan A Econ Sp 39:2715–2732

Whitzman C, Thompson J, Favarin S, Nourani F, Shibata S (2020) Incidence and reporting: making the invisible matter. In: Loukautou-Sideris VCAA (ed) Transit crime and sexual violence in cities: international evidence and prevention. Routledge, London

Wiebe DJ, Richmond TS, Poster J, Guo W, Allison PD, Branas CC (2014) Adolescents’ fears of violence in transit environments during daily activities. Secur J 27:226–241

Wilow M (2015) Déplacements des femmes et sentiment d’insécurité à Bruxelles: perceptions et stratégies. Brussels Studies. Collection générale, n° 87. http://journals.openedition.org/brussels/1274

Wilson JQ, Kelling GL (1982) Broken windows. Atlantic Mon 249:29–38

Yates A, Ceccato V (2020) Individual and spatial dimensions of women’s fear of crime: a Scandinavian study case. Int J Comp Appl Crim Justice 44(4):277–292

Yavuz N, Welch EW (2010) Addressing fear of crime in public space: gender differences in reaction to safety measures in train transit. Urban Stud 47:2491–2515

Yu S-SV, Smith MJ (2014) Commuters using public transit in New York City: using area-level data to identify neighbourhoods with vulnerable riders. Secur J 27:194–209

Zahnow R, Corcoran J (2019) Crime and bus stops: an examination using transit smart card and crime data. Environ Plan B Urban Anal City Sci. https://doi.org/10.1177/2399808319890614

Download references

Acknowledgements

We would like to thank Dr Yuan Pang Wang for sharing his knowledge and experience of systematic literature overviews with the authors. Authors would also like to thank the Swedish Transport Administration for funding this research via the project “The impact of station’s environments on crime and passenger safety”, Grant number TRV 2020/22903.

Open Access funding provided by Royal Institute of Technology. This study was carried out as part of the research project “The impact of station’s environments on crime and passenger safety”, funded by the Swedish Transport Administration, Grant number TRV 2020/22903.

Author information

Authors and affiliations.

Department of Urban Planning and Environment, KTH Royal Institute of Technology, Stockholm, Sweden

Vania Ceccato

National Graduate School of Civil Engineering, University of Lyon, Lyon, France

Nathan Gaudelet & Gabin Graf

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Vania Ceccato .

Ethics declarations

Conflict of interest.

The author(s) declare that they have no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

See Tables 3 , 4 , 5 , and 6 .

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third-party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Ceccato, V., Gaudelet, N. & Graf, G. Crime and safety in transit environments: a systematic review of the English and the French literature, 1970–2020. Public Transp 14 , 105–153 (2022). https://doi.org/10.1007/s12469-021-00265-1

Download citation

Accepted : 18 January 2021

Published : 31 January 2022

Issue Date : March 2022

DOI : https://doi.org/10.1007/s12469-021-00265-1

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Fear of crime
• Transit safety
• Intersectionality
• Whole journey approach
• Bibliometric analysis
• Find a journal
• Publish with us
• Track your research

• Systematic Review
• Open access
• Published: 23 May 2024

Systematic literature review of real-world evidence for treatments in HR+/HER2- second-line LABC/mBC after first-line treatment with CDK4/6i

• Veronique Lambert   ORCID: orcid.org/0000-0002-6984-0038 1 ,
• Sarah Kane   ORCID: orcid.org/0009-0006-9341-4836 2   na1 ,
• Belal Howidi   ORCID: orcid.org/0000-0002-1166-7631 2   na1 ,
• Bao-Ngoc Nguyen   ORCID: orcid.org/0000-0001-6026-2270 2   na1 ,
• David Chandiwana   ORCID: orcid.org/0009-0002-3499-2565 3 ,
• Yan Wu   ORCID: orcid.org/0009-0008-3348-9232 1 ,
• Michelle Edwards   ORCID: orcid.org/0009-0001-4292-3140 3 &
• Imtiaz A. Samjoo   ORCID: orcid.org/0000-0003-1415-8055 2   na1  

BMC Cancer volume  24 , Article number:  631 ( 2024 ) Cite this article

411 Accesses

1 Altmetric

Metrics details

Cyclin-dependent kinase 4 and 6 inhibitors (CDK4/6i) combined with endocrine therapy (ET) are currently recommended by the National Comprehensive Cancer Network (NCCN) guidelines and the European Society for Medical Oncology (ESMO) guidelines as the first-line (1 L) treatment for patients with hormone receptor-positive, human epidermal growth factor receptor 2-negative, locally advanced/metastatic breast cancer (HR+/HER2- LABC/mBC). Although there are many treatment options, there is no clear standard of care for patients following 1 L CDK4/6i. Understanding the real-world effectiveness of subsequent therapies may help to identify an unmet need in this patient population. This systematic literature review qualitatively synthesized effectiveness and safety outcomes for treatments received in the real-world setting after 1 L CDK4/6i therapy in patients with HR+/ HER2- LABC/mBC.

MEDLINE®, Embase, and Cochrane were searched using the Ovid® platform for real-world evidence studies published between 2015 and 2022. Grey literature was searched to identify relevant conference abstracts published from 2019 to 2022. The review was conducted in accordance with PRISMA guidelines (PROSPERO registration: CRD42023383914). Data were qualitatively synthesized and weighted average median real-world progression-free survival (rwPFS) was calculated for NCCN/ESMO-recommended post-1 L CDK4/6i treatment regimens.

Twenty records (9 full-text articles and 11 conference abstracts) encompassing 18 unique studies met the eligibility criteria and reported outcomes for second-line (2 L) treatments after 1 L CDK4/6i; no studies reported disaggregated outcomes in the third-line setting or beyond. Sixteen studies included NCCN/ESMO guideline-recommended treatments with the majority evaluating endocrine-based therapy; five studies on single-agent ET, six studies on mammalian target of rapamycin inhibitors (mTORi) ± ET, and three studies with a mix of ET and/or mTORi. Chemotherapy outcomes were reported in 11 studies. The most assessed outcome was median rwPFS; the weighted average median rwPFS was calculated as 3.9 months (3.3-6.0 months) for single-agent ET, 3.6 months (2.5–4.9 months) for mTORi ± ET, 3.7 months for a mix of ET and/or mTORi (3.0–4.0 months), and 6.1 months (3.7–9.7 months) for chemotherapy. Very few studies reported other effectiveness outcomes and only two studies reported safety outcomes. Most studies had heterogeneity in patient- and disease-related characteristics.

Conclusions

The real-world effectiveness of current 2 L treatments post-1 L CDK4/6i are suboptimal, highlighting an unmet need for this patient population.

Peer Review reports

Introduction

Breast cancer (BC) is the most diagnosed form of cancer in women with an estimated 2.3 million new cases diagnosed worldwide each year [ 1 ]. BC is the second leading cause of cancer death, accounting for 685,000 deaths worldwide per year [ 2 ]. By 2040, the global burden associated with BC is expected to surpass three million new cases and one million deaths annually (due to population growth and aging) [ 3 ]. Numerous factors contribute to global disparities in BC-related mortality rates, including delayed diagnosis, resulting in a high number of BC cases that have progressed to locally advanced BC (LABC) or metastatic BC (mBC) [ 4 , 5 , 6 ]. In the United States (US), the five-year survival rate for patients who progress to mBC is three times lower (31%) than the overall five-year survival rate for all stages (91%) [ 6 , 7 ].

Hormone receptor (HR) positive (i.e., estrogen receptor and/or progesterone receptor positive) coupled with negative human epidermal growth factor 2 (HER2) expression is the most common subtype of BC, accounting for ∼ 60–70% of all BC cases [ 8 , 9 ]. Historically, endocrine therapy (ET) through estrogen receptor modulation and/or estrogen deprivation has been the standard of care for first-line (1 L) treatment of HR-positive/HER2-negative (HR+/HER2-) mBC [ 10 ]. However, with the approval of the cyclin-dependent kinase 4/6 inhibitor (CDK4/6i) palbociclib in combination with the aromatase inhibitor (AI) letrozole in 2015 by the US Food and Drug Administration (FDA), 1 L treatment practice patterns have evolved such that CDK4/6i (either in combination with AIs or with fulvestrant) are currently considered the standard of care [ 11 , 12 , 13 , 14 , 15 , 16 , 17 ]. Other CDK4/6i (ribociclib and abemaciclib) in combination with ET are approved for the treatment of HR+/HER2- LABC/mBC; 1 L use of ribociclib in combination with an AI was granted FDA approval in March 2017 for postmenopausal women (with expanded approval in July 2018 for pre/perimenopausal women and for use in 1 L with fulvestrant for patients with disease progression on ET as well as for postmenopausal women), and abemaciclib in combination with fulvestrant was granted FDA approval in September 2017 for patients with disease progression following ET and as monotherapy in cases where disease progression occurs following ET and prior chemotherapy in mBC (with expanded approval in February 2018 for use in 1 L in combination with an AI for postmenopausal women) [ 18 , 19 , 20 , 21 ].

Clinical trials investigating the addition of CDK4/6i to ET have demonstrated significant improvement in progression-free survival (PFS) and significant (ribociclib) or numerical (palbociclib and abemaciclib) improvement in overall survival (OS) compared to ET alone in patients with HR+/HER2- advanced or mBC, making this combination treatment the recommended option in the 1 L setting [ 22 , 23 , 24 , 25 , 26 , 27 ]. However, disease progression occurs in a significant portion of patients after 1 L CDK4/6i treatment [ 28 ] and the optimal treatment sequence after progression on CDK4/6i remains unclear [ 29 ]. At the time of this review (literature search conducted December 14, 2022), guidelines by the National Comprehensive Cancer Network (NCCN) and the European Society for Medical Oncology (ESMO) recommend various options for the treatment of HR+/HER2- advanced BC in the second-line (2 L) setting, including fulvestrant monotherapy, mammalian target of rapamycin inhibitors (mTORi; e.g., everolimus) ± ET, alpelisib + fulvestrant (if phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha mutation positive [PIK3CA-m+]), poly-ADP ribose polymerase inhibitors (PARPi) including olaparib or talazoparib (if breast cancer gene/partner and localizer of BRCA2 positive [BRCA/PALB2m+]), and chemotherapy (in cases when a visceral crisis is present) [ 15 , 16 ]. CDK4/6i can also be used in 2 L [ 16 , 30 ]; however, limited data are available to support CDK4/6i rechallenge after its use in the 1 L setting [ 15 ]. Depending on treatments used in the 1 L and 2 L settings, treatment in the third-line setting is individualized based on the patient’s response to prior treatments, tumor load, duration of response, and patient preference [ 9 , 15 ]. Understanding subsequent treatments after 1 L CDK4/6i, and their associated effectiveness, is an important focus in BC research.

Treatment options for HR+/HER2- LABC/mBC continue to evolve, with ongoing research in both clinical trials and in the real-world setting. Real-world evidence (RWE) offers important insights into novel therapeutic regimens and the effectiveness of treatments for HR+/HER2- LABC/mBC. The effectiveness of the current treatment options following 1 L CDK4/6i therapy in the real-world setting highlights the unmet need in this patient population and may help to drive further research and drug development. In this study, we conducted a systematic literature review (SLR) to qualitatively summarize the effectiveness and safety of treatment regimens in the real-world setting after 1 L treatment with CDK4/6i in patients with HR+/HER2- LABC/mBC.

Literature search

An SLR was performed in accordance with the Cochrane Handbook for Systematic Reviews of Interventions [ 31 ] and reported in alignment with the Preferred Reporting Items for Systematic Literature Reviews and Meta-Analyses (PRISMA) statement [ 32 ] to identify all RWE studies assessing the effectiveness and safety of treatments used for patients with HR+/HER2- LABC/mBC following 1 L CDK4/6i therapy and received subsequent treatment in 2 L and beyond (2 L+). The Ovid® platform was used to search MEDLINE® (including Epub Ahead of Print and In-Process, In-Data-Review & Other Non-Indexed Citations), Ovid MEDLINE® Daily, Embase, Cochrane Central Register of Controlled Trials, and Cochrane Database of Systematic Reviews by an experienced medical information specialist. The MEDLINE® search strategy was peer-reviewed independently by a senior medical information specialist before execution using the Peer Review of Electronic Search Strategies (PRESS) checklist [ 33 ]. Searches were conducted on December 14, 2022. The review protocol was developed a priori and registered with the International Prospective Register of Systematic Review (PROSPERO; CRD42023383914) which outlined the population, intervention, comparator, outcome, and study design (PICOS) criteria and methodology used to conduct the review (Table  1 ).

Search strategies utilized a combination of controlled vocabulary (e.g., “HER2 Breast Cancer” or “HR Breast Cancer”) and keywords (e.g., “Retrospective studies”). Vocabulary and syntax were adjusted across databases. Published and validated filters were used to select for study design and were supplemented using additional medical subject headings (MeSH) terms and keywords to select for RWE and nonrandomized studies [ 34 ]. No language restrictions were included in the search strategy. Animal-only and opinion pieces were removed from the results. The search was limited to studies published between January 2015 and December 2022 to reflect the time at which FDA approval was granted for the first CDK4/6i agent (palbociclib) in combination with AI for the treatment of LABC/mBC [ 35 ]. Further search details are presented in Supplementary Material 1 .

Grey literature sources were also searched to identify relevant abstracts and posters published from January 2019 to December 2022 for prespecified relevant conferences including ESMO, San Antonio Breast Cancer Symposium (SABCS), American Society of Clinical Oncology (ASCO), the International Society for Pharmacoeconomics and Outcomes Research (ISPOR US), and the American Association for Cancer Research (AACR). A search of ClinicalTrials.gov was conducted to validate the findings from the database and grey literature searches.

Study selection, data extraction & weighted average calculation

Studies were screened for inclusion using DistillerSR Version 2.35 and 2.41 (DistillerSR Inc. 2021, Ottawa, Canada) by two independent reviewers based on the prespecified PICOS criteria (Table  1 ). A third reviewer was consulted to resolve any discrepancies during the screening process. Studies were included if they reported RWE on patients aged ≥ 18 years with HR+/HER2- LABC/mBC who received 1 L CDK4/6i treatment and received subsequent treatment in 2 L+. Studies were excluded if they reported the results of clinical trials (i.e., non-RWE), were published in any language other than English, and/or were published prior to 2015 (or prior to 2019 for conference abstracts and posters). For studies that met the eligibility criteria, data relating to study design and methodology, details of interventions, patient eligibility criteria and baseline characteristics, and outcome measures such as efficacy, safety, tolerability, and patient-reported outcomes (PROs), were extracted (as available) using a Microsoft Excel®-based data extraction form (Microsoft Corporation, WA, USA). Data extraction was performed by a single reviewer and was confirmed by a second reviewer. Multiple publications identified for the same RWE study, patient population, and setting that reported data for the same intervention were linked and extracted as a single publication. Weighted average median real-world progression-free survival (rwPFS) values were calculated by considering the contribution to the median rwPFS of each study proportional to its respective sample size. These weighted values were then used to compute the overall median rwPFS estimate.

Quality assessment

The Newcastle-Ottawa scale (NOS) for nonrandomized (cohort) studies was used to assess the risk of bias for published, full-text studies [ 36 ]. The NOS allocates a maximum of nine points for the least risk of bias across three domains: (1) Formation of study groups (four points), (2) Comparability between study groups (two points), (3) Outcome ascertainment (three points). NOS scores can be categorized in three groups: very high risk of bias (0 to 3 points), high risk of bias (4 to 6), and low risk of bias (7 to 9) [ 37 ]. Risk of bias assessment was performed by one reviewer and validated by a second independent reviewer to verify accuracy. Due to limited methodological data by which to assess study quality, risk of bias assessment was not performed on conference abstracts or posters. An amendment to the PROSPERO record (CRD42023383914) for this study was submitted in relation to the quality assessment method (specifying usage of the NOS).

The database search identified 3,377 records; after removal of duplicates, 2,759 were screened at the title and abstract stage of which 2,553 were excluded. Out of the 206 reports retrieved and assessed for eligibility, an additional 187 records were excluded after full-text review; most of these studies were excluded for having patients with mixed lines of CDK4/6i treatment (i.e., did not receive CDK4/6i exclusively in 1 L) (Fig.  1 and Table S1 ). The grey literature search identified 753 records which were assessed for eligibility; of which 752 were excluded mainly due to the population not meeting the eligibility criteria (Fig.  1 ). In total, the literature searches identified 20 records (9 published full-text articles and 11 conference abstracts/posters) representing 18 unique RWE studies that met the inclusion criteria. The NOS quality scores for the included full-text articles are provided in Table S2 . The scores ranged from four to six points (out of a total score of nine) and the median score was five, indicating that all the studies suffered from a high risk of bias [ 37 ].

Most studies were retrospective analyses of chart reviews or medical registries, and all studies were published between 2017 and 2022 (Table S3 ). Nearly half of the RWE studies (8 out of 18 studies) were conducted in the US [ 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 ], while the remaining studies included sites in Canada, China, Germany, Italy, Japan, and the United Kingdom [ 46 , 47 , 48 , 49 , 50 , 51 , 52 , 53 , 54 ]. Sample sizes ranged from as few as 4 to as many as 839 patients across included studies, with patient age ranging from 26 to 86 years old.

Although treatment characteristics in the 1 L setting were not the focus of the present review, these details are captured in Table S3 . Briefly, several RWE studies reported 1 L CDK4/6i use in combination with ET (8 out of 18 studies) or as monotherapy (2 out of 18 studies) (Table S3 ). Treatments used in combination with 1 L CDK4/6i included letrozole, fulvestrant, exemestane, and anastrozole. Where reported (4 out of 18 studies), palbociclib was the most common 1 L CDK4/6i treatment. Many studies (8 out of 18 studies) did not report which specific CDK4/6i treatment(s) were used in 1 L or if its administration was in combination or monotherapy.

Characteristics of treatments after 1 L CDK4/6i therapy

Across all studies included in this review, effectiveness and safety data were only available for treatments administered in the 2 L setting after 1 L CDK4/6i treatment. No studies were identified that reported outcomes for patients treated in the third-line setting or beyond after 1 L CDK4/6i treatment. All 18 studies reported effectiveness outcomes in 2 L, with only two of these studies also describing 2 L safety outcomes. The distribution of outcomes reported in these studies is provided in Table S4 . Studies varied in their reporting of outcomes for 2 L treatments; some studies reported outcomes for a group of 2 L treatments while others described independent outcomes for specific 2 L treatments (i.e., everolimus, fulvestrant, or chemotherapy agents such as eribulin mesylate) [ 42 , 45 , 50 , 54 , 55 ]. Due to the heterogeneity in treatment classes reported in these studies, this data was categorized (as described below) to align with the guidelines provided by NCCN and ESMO [ 15 , 16 ]. The treatment class categorizations for the purpose of this review are: single-agent ET (patients who exclusively received a single-agent ET after 1 L CDK4/6i treatment), mTORi ± ET (patients who exclusively received an mTORi with or without ET after 1 L CDK4/6i treatment), mix of ET and/or mTORi (patients who may have received only ET, only mTORi, and/or both treatments but the studies in this group lacked sufficient information to categorize these patients in the “single-agent ET” or “mTOR ± ET” categories), and chemotherapy (patients who exclusively received chemotherapy after 1 L CDK4/6i treatment). Despite ESMO and NCCN guidelines indicating that limited evidence exists to support rechallenge with CDK4/6i after 1 L CDK4/6i treatment [ 15 , 16 ], two studies reported outcomes for this treatment approach. Data for such patients were categorized as “ CDK4/6i ± ET ” as it was unclear how many patients receiving CDK4/6i rechallenge received concurrent ET. All other patient groups that lacked sufficient information or did not report outcome/safety data independently (i.e., grouped patients with mixed treatments) to categorize as one of the treatment classes described above were grouped as “ other ”.

The majority of studies reported effectiveness outcomes for endocrine-based therapy after 1 L CDK4/6i treatment; five studies for single-agent ET, six studies for mTORi ± ET, and three studies for a mix of ET and/or mTORi (Fig.  2 ). Eleven studies reported effectiveness outcomes for chemotherapy after 1 L CDK4/6i treatment, and only two studies reported effectiveness outcomes for CDK4/6i rechallenge ± ET. Eight studies that described effectiveness outcomes were grouped into the “other” category. Safety data was only reported in two studies: one study evaluating the chemotherapy agent eribulin mesylate and one evaluating the mTORi everolimus.

Effectiveness outcomes

Real-world progression-free survival

Median rwPFS was described in 13 studies (Tables  2 and Table S5 ). Across the 13 studies, the median rwPFS ranged from 2.5 months [ 49 ] to 17.3 months [ 39 ]. Out of the 13 studies reporting median rwPFS, 10 studies reported median rwPFS for a 2 L treatment recommended by ESMO and NCCN guidelines, which ranged from 2.5 months [ 49 ] to 9.7 months [ 45 ].

Weighted average median rwPFS was calculated for 2 L treatments recommended by both ESMO and NCCN guidelines (Fig.  3 ). The weighted average median rwPFS for single-agent ET was 3.9 months ( n  = 92 total patients) and was derived using data from two studies reporting median rwPFS values of 3.3 months ( n  = 70) [ 38 ] and 6.0 months ( n  = 22) [ 40 ]. For one study ( n  = 7) that reported outcomes for single agent ET, median rwPFS was not reached during the follow-up period; as such, this study was excluded from the weighted average median rwPFS calculation [ 49 ].

The weighted average median rwPFS for mTORi ± ET was 3.6 months ( n  = 128 total patients) and was derived based on data from 3 studies with median rwPFS ranging from 2.5 months ( n  = 4) [ 49 ] to 4.9 months ( n  = 25) [ 54 ] (Fig.  3 ). For patients who received a mix of ET and/or mTORi but could not be classified into the single-agent ET or mTORi ± ET treatment classes, the weighted average median rwPFS was calculated to be 3.7 months ( n  = 17 total patients). This was calculated based on data from two studies reporting median rwPFS values of 3.0 months ( n  = 5) [ 46 ] and 4.0 months ( n  = 12) [ 49 ]. Notably, one study of patients receiving ET and/or everolimus reported a median rwPFS duration of 3.0 months; however, this study was excluded from the weighted average median rwPFS calculation for the ET and/or mTORi class as the sample size was not reported [ 53 ].

The weighted average median rwPFS for chemotherapy was 6.1 months ( n  = 499 total patients), calculated using data from 7 studies reporting median rwPFS values ranging from 3.7 months ( n  = 249) [ 38 ] to 9.7 months ( n  = 121) [ 45 ] (Fig.  3 ). One study with a median rwPFS duration of 5.6 months was not included in the weighted average median rwPFS calculation as the study did not report the sample size [ 53 ]. A second study was excluded from the calculation since the reported median rwPFS was not reached during the study period ( n  = 7) [ 41 ].

Although 2 L CDK4/6i ± ET rechallenge lacks sufficient information to support recommendation by ESMO and NCCN guidelines, the limited data currently available for this treatment have shown promising results. Briefly, two studies reported median rwPFS for CDK4/6i ± ET with values of 8.3 months ( n  = 302) [ 38 ] and 17.3 months ( n  = 165) (Table  2 ) [ 39 ]. The remaining median rwPFS studies reported data for patients classified as “Other” (Table S5 ). The “Other” category included median rwPFS outcomes from seven studies, and included a myriad of treatments (e.g., ET, mTOR + ET, chemotherapy, CDK4/6i + ET, alpelisib + fulvestrant, chidamide + ET) for which disaggregated median rwPFS values were not reported.

Overall survival

Median OS for 2 L treatment was reported in only three studies (Table  2 ) [ 38 , 42 , 43 ]. Across the three studies, the 2 L median OS ranged from 5.2 months ( n  = 3) [ 43 ] to 35.7 months ( n  = 302) [ 38 ]. Due to the lack of OS data in most of the studies, weighted averages could not be calculated. No median OS data was reported for the single-agent ET treatment class whereas two studies reported median OS for the mTORi ± ET treatment class, ranging from 5.2 months ( n  = 3) [ 43 ] to 21.8 months ( n  = 54) [ 42 ]. One study reported 2 L median OS of 24.8 months for a single patient treated with chemotherapy [ 43 ]. The median OS data in the CDK4/6i ± ET rechallenge group was 35.7 months ( n  = 302) [ 38 ].

Patient mortality was reported in three studies [ 43 , 44 , 45 ]. No studies reported mortality for the single-agent ET treatment class and only one study reported this outcome for the mTORi ± ET treatment class, where 100% of patients died ( n  = 3) as a result of rapid disease progression [ 43 ]. For the chemotherapy class, one study reported mortality for one patient receiving 2 L capecitabine [ 43 ]. An additional study reported eight deaths (21.7%) following 1 L CDK4/6i treatment; however, this study did not disclose the 2 L treatments administered to these patients [ 44 ].

Other clinical endpoints

The studies included limited information on additional clinical endpoints; two studies reported on time-to-discontinuation (TTD), two reported on duration of response (DOR), and one each on time-to-next-treatment (TTNT), time-to-progression (TTP), objective response rate (ORR), clinical benefit rate (CBR), and stable disease (Tables  2 and Table S5 ).

Safety, tolerability, and patient-reported outcomes

Safety and tolerability data were reported in two studies [ 40 , 45 ]. One study investigating 2 L administration of the chemotherapy agent eribulin mesylate reported 27 patients (22.3%) with neutropenia, 3 patients (2.5%) with febrile neutropenia, 10 patients (8.3%) with peripheral neuropathy, and 14 patients (11.6%) with diarrhea [ 45 ]. Of these, neutropenia of grade 3–4 severity occurred in 9 patients (33.3%) [ 45 ]. A total of 55 patients (45.5%) discontinued eribulin mesylate treatment; 1 patient (0.83%) discontinued treatment due to adverse events [ 45 ]. Another study reported that 5 out of the 22 patients receiving the mTORi everolimus combined with ET in 2 L (22.7%) discontinued treatment due to toxicity [ 40 ]. PROs were not reported in any of the studies included in the SLR.

The objective of this study was to summarize the existing RWE on the effectiveness and safety of therapies for patients with HR+/HER2- LABC/mBC after 1 L CDK4/6i treatment. We identified 18 unique studies reporting specifically on 2 L treatment regimens after 1 L CDK4/6i treatment. The weighted average median rwPFS for NCCN- and ESMO- guideline recommended 2 L treatments ranged from 3.6 to 3.9 months for ET-based treatments and was 6.1 months when including chemotherapy-based regimens. Treatment selection following 1 L CDK4/6i therapy remains challenging primarily due to the suboptimal effectiveness or significant toxicities (e.g., chemotherapy) associated with currently available options [ 56 ]. These results highlight that currently available 2 L treatments for patients with HR+/HER2- LABC/mBC who have received 1 L CDK4/6i are suboptimal, as evidenced by the brief median rwPFS duration associated with ET-based treatments, or notable side effects and toxicity linked to chemotherapy. This conclusion is aligned with a recent review highlighting the limited effectiveness of treatment options for HR+/HER2- LABC/mBC patients post-CDK4/6i treatment [ 56 , 57 ]. Registrational trials which have also shed light on the short median PFS of 2–3 months achieved by ET (i.e., fulvestrant) after 1 L CDK4/6i therapy emphasize the need to develop improved treatment strategies aimed at prolonging the duration of effective ET-based treatment [ 56 ].

The results of this review reveal a paucity of additional real-world effectiveness and safety evidence after 1 L CDK4/6i treatment in HR+/HER2- LABC/mBC. OS and DOR were only reported in two studies while other clinical endpoints (i.e., TTD, TTNT, TTP, ORR, CBR, and stable disease) were only reported in one study each. Similarly, safety and tolerability data were only reported in two studies each, and PROs were not reported in any study. This hindered our ability to provide a comprehensive assessment of real-world treatment effectiveness and safety following 1 L CDK4/6i treatment. The limited evidence may be due to the relatively short period of time that has elapsed since CDK4/6i first received US FDA approval for 1 L treatment of HR+/HER2- LABC/mBC (2015) [ 35 ]. As such, almost half of our evidence was informed by conference abstracts. Similarly, no real-world studies were identified in our review that reported outcomes for treatments in the third- or later-lines of therapy after 1 L CDK4/6i treatment. The lack of data in this patient population highlights a significant gap which limits our understanding of the effectiveness and safety for patients receiving later lines of therapy. As more patients receive CDK4/6i therapy in the 1 L setting, the number of patients requiring subsequent lines of therapy will continue to grow. Addressing this data gap over time will be critical to improve outcomes for patients with HR+/HER2- LABC/mBC following 1 L CDK4/6i therapy.

There are several strengths of this study, including adherence to the guidelines outlined in the Cochrane Handbook to ensure a standardized and reliable approach to the SLR [ 58 ] and reporting of the SLR following PRISMA guidelines to ensure transparency and reproducibility [ 59 ]. Furthermore, the inclusion of only RWE studies allowed us to assess the effectiveness of current standard of care treatments outside of a controlled environment and enabled us to identify an unmet need in this patient population.

This study had some notable limitations, including the lack of safety and additional effectiveness outcomes reported. In addition, the dearth of studies reporting PROs is a limitation, as PROs provide valuable insight into the patient experience and are an important aspect of assessing the impact of 2 L treatments on patients’ quality of life. The studies included in this review also lacked consistent reporting of clinical characteristics (e.g., menopausal status, sites of metastasis, prior surgery) making it challenging to draw comprehensive conclusions or comparisons based on these factors across the studies. Taken together, there exists an important gap in our understanding of the long-term management of patients with HR+/HER2- LABC/mBC. Additionally, the effectiveness results reported in our evidence base were informed by small sample sizes; many of the included studies reported median rwPFS based on less than 30 patients [ 39 , 40 , 41 , 46 , 49 , 51 , 60 ], with two studies not reporting the sample size at all [ 47 , 53 ]. This may impact the generalizability and robustness of the results. Relatedly, the SLR database search was conducted in December 2022; as such, novel agents (e.g., elacestrant and capivasertib + fulvestrant) that have since received FDA approval for the treatment of HR+/HER2- LABC/mBC may impact current 2 L rwPFS outcomes [ 61 , 62 ]. Finally, relative to the number of peer-reviewed full-text articles, this SLR identified eight abstracts and one poster presentation, comprising half (50%) of the included unique studies. As conference abstracts are inherently limited by how much content that can be described due to word limit constraints, this likely had implications on the present synthesis whereby we identified a dearth of real-world effectiveness outcomes in patients with HR+/HER2- LABC/mBC treated with 1 L CDK4/6i therapy.

Future research in this area should aim to address the limitations of the current literature and provide a more comprehensive understanding of optimal sequencing of effective and safe treatment for patients following 1 L CDK4/6i therapy. Specifically, future studies should strive to report robust data related to effectiveness, safety, and PROs for patients receiving 2 L treatment after 1 L CDK4/6i therapy. Future studies should also aim to understand the mechanism underlying CDK4/6i resistance. Addressing these gaps in knowledge may improve the long-term real-world management of patients with HR+/HER2- LABC/mBC. A future update of this synthesis may serve to capture a wider breadth of full-text, peer-reviewed articles to gain a more robust understanding of the safety, effectiveness, and real-world treatment patterns for patients with HR+/HER2- LABC/mBC. This SLR underscores the necessity for ongoing investigation and the development of innovative therapeutic approaches to address these gaps and improve patient outcomes.

This SLR qualitatively summarized the existing real-world effectiveness data for patients with HR+/HER2- LABC/mBC after 1 L CDK4/6i treatment. Results of this study highlight the limited available data and the suboptimal effectiveness of treatments employed in the 2 L setting and underscore the unmet need in this patient population. Additional studies reporting effectiveness and safety outcomes, in addition to PROs, for this patient population are necessary and should be the focus of future research.

PRISMA flow diagram. *Two included conference abstracts reported the same information as already included full-text reports, hence both conference abstracts were not identified as unique. Abbreviations: 1 L = first-line; AACR = American Association of Cancer Research; ASCO = American Society of Clinical Oncology; CDK4/6i = cyclin-dependent kinase 4/6 inhibitor; ESMO = European Society for Medical Oncology; ISPOR = Professional Society for Health Economics and Outcomes Research; n = number of studies; NMA = network meta-analysis; pts = participants; SABCS = San Antonio Breast Cancer Symposium; SLR = systematic literature review.

Number of studies reporting effectiveness outcomes exclusively for each treatment class. *Studies that lack sufficient information on effectiveness outcomes to classify based on the treatment classes outlined in the legend above. Abbreviations: CDK4/6i = cyclin-dependent kinase 4/6 inhibitor; ET = endocrine therapy; mTORi = mammalian target of rapamycin inhibitor.

Weighted average median rwPFS for 2 L treatments (recommended in ESMO/NCCN guidelines) after 1 L CDK4/6i treatment. Circular dot represents weighted average median across studies. Horizontal bars represent the range of values reported in these studies. Abbreviations: CDK4/6i = cyclin-dependent kinase 4/6 inhibitor; ESMO = European Society for Medical Oncology; ET = endocrine therapy, mTORi = mammalian target of rapamycin inhibitor; n = number of patients; NCCN = National Comprehensive Cancer Network; rwPFS = real-world progression-free survival.

Data availability

All data generated or analyzed during this study are included in this published article [and its supplementary information files]. This study is registered with PROSPERO (CRD42023383914).

Abbreviations

Second-line

Second-line treatment setting and beyond

American Association of Cancer Research

Aromatase inhibitor

American Society of Clinical Oncology

• Breast cancer

breast cancer gene/partner and localizer of BRCA2 positive

Clinical benefit rate

Cyclin-dependent kinase 4/6 inhibitor

Complete response

Duration of response

European Society for Medical Oncology

Food and Drug Administration

Human epidermal growth factor receptor 2

Human epidermal growth factor receptor 2 negative

Hormone receptor

Hormone receptor positive

Professional Society for Health Economics and Outcomes Research

Locally advanced breast cancer

Metastatic breast cancer

Medical Literature Analysis and Retrieval System Online

Medical subject headings

Mammalian target of rapamycin inhibitor

National Comprehensive Cancer Network

Newcastle Ottawa Scale

Objective response rate

Poly-ADP ribose polymerase inhibitor

Progression-free survival

Population, Intervention, Comparator, Outcome, Study Design

Partial response

Preferred Reporting Items for Systematic Literature Reviews and Meta-Analyses

Patient-reported outcomes

• Real-world evidence

San Antonio Breast Cancer Symposium

• Systematic literature review

Time-to-discontinuation

Time-to-next-treatment

Time-to-progression

United States

Łukasiewicz S, Czeczelewski M, Forma A, Baj J, Sitarz R, Stanisławek A, Breast, Cancer—Epidemiology. Risk factors, classification, prognostic markers, and current treatment Strategies—An. Updated Rev Cancers. 2021;13(17):4287.

Google Scholar  

World Health Organization (WHO). Breast Cancer Facts Sheet [updated July 12 2023. https://www.who.int/news-room/fact-sheets/detail/breast-cancer .

Arnold M, Morgan E, Rumgay H, Mafra A, Singh D, Laversanne M, et al. Current and future burden of breast cancer: global statistics for 2020 and 2040. Breast. 2022;66:15–23.

Article   PubMed   PubMed Central   Google Scholar  

Wilkinson L, Gathani T. Understanding breast cancer as a global health concern. Br J Radiol. 2022;95(1130):20211033.

Article   PubMed   Google Scholar  

Giaquinto AN, Sung H, Miller KD, Kramer JL, Newman LA, Minihan A et al. Breast Cancer Statistics, 2022. CA: A Cancer Journal for Clinicians. 2022;72(6):524– 41.

National Cancer Institute (NIH). Cancer Stat Facts: Female Breast Cancer [updated 2020. https://seer.cancer.gov/statfacts/html/breast.html .

American Cancer Society. Key Statistics for Breast Cancer [ https://www.cancer.org/cancer/types/breast-cancer/about/how-common-is-breast-cancer.html .

Zagami P, Carey LA. Triple negative breast cancer: pitfalls and progress. npj Breast Cancer. 2022;8(1):95.

Article   CAS   PubMed   PubMed Central   Google Scholar  

Matutino A, Joy AA, Brezden-Masley C, Chia S, Verma S. Hormone receptor-positive, HER2-negative metastatic breast cancer: redrawing the lines. Curr Oncol. 2018;25(Suppl 1):S131–41.

Lloyd MR, Wander SA, Hamilton E, Razavi P, Bardia A. Next-generation selective estrogen receptor degraders and other novel endocrine therapies for management of metastatic hormone receptor-positive breast cancer: current and emerging role. Ther Adv Med Oncol. 2022;14:17588359221113694.

Cardoso F, Senkus E, Costa A, Papadopoulos E, Aapro M, André F, et al. 4th ESO-ESMO International Consensus guidelines for advanced breast Cancer (ABC 4)†. Ann Oncol. 2018;29(8):1634–57.

Article   CAS   PubMed   Google Scholar  

US Food Drug Administration. Palbociclib (Ibrance) 2017 [updated March 31, 2017. https://www.fda.gov/drugs/resources-information-approved-drugs/palbociclib-ibrance .

US Food Drug Administration. FDA expands ribociclib indication in HR-positive, HER2-negative advanced or metastatic breast cancer 2018 [updated July 18. 2018. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-expands-ribociclib-indication-hr-positive-her2-negative-advanced-or-metastatic-breast-cancer .

US Food Drug Administration. FDA approves abemaciclib for HR positive, HER2-negative breast cancer 2017 [updated Sept 28. 2017. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-abemaciclib-hr-positive-her2-negative-breast-cancer .

NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®). Breast Cancer 2022 [ https://www.nccn.org/professionals/physician_gls/pdf/breast.pdf .

Gennari A, André F, Barrios CH, Cortés J, de Azambuja E, DeMichele A, et al. ESMO Clinical Practice Guideline for the diagnosis, staging and treatment of patients with metastatic breast cancer. Ann Oncol. 2021;32(12):1475–95.

Beaver JA, Amiri-Kordestani L, Charlab R, Chen W, Palmby T, Tilley A, et al. FDA approval: Palbociclib for the Treatment of Postmenopausal Patients with estrogen Receptor-Positive, HER2-Negative metastatic breast Cancer. Clin Cancer Res. 2015;21(21):4760–6.

US Food Drug Administration. Ribociclib (Kisqali) [ https://www.fda.gov/drugs/resources-information-approved-drugs/ribociclib-kisqali#:~:text=On%20March%2013%2C%202017%2C%20the,hormone%20receptor%20(HR)%2Dpositive%2C .

US Food Drug Administration. FDA approves new treatment for certain advanced or metastatic breast cancers [ https://www.fda.gov/news-events/press-announcements/fda-approves-new-treatment-certain-advanced-or-metastatic-breast-cancers .

US Food Drug Administration. FDA expands ribociclib indication in HR-positive, HER2-negative advanced or metastatic breast cancer. 2018 [ https://www.fda.gov/drugs/resources-information-approved-drugs/fda-expands-ribociclib-indication-hr-positive-her2-negative-advanced-or-metastatic-breast-cancer .

US Food Drug Administration. FDA approves abemaciclib as initial therapy for HR-positive, HER2-negative metastatic breast cancer [ https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-abemaciclib-initial-therapy-hr-positive-her2-negative-metastatic-breast-cancer .

Turner NC, Slamon DJ, Ro J, Bondarenko I, Im S-A, Masuda N, et al. Overall survival with Palbociclib and fulvestrant in advanced breast Cancer. N Engl J Med. 2018;379(20):1926–36.

Slamon DJ, Neven P, Chia S, Fasching PA, De Laurentiis M, Im SA, et al. Phase III randomized study of Ribociclib and Fulvestrant in hormone Receptor-Positive, human epidermal growth factor receptor 2-Negative advanced breast Cancer: MONALEESA-3. J Clin Oncol. 2018;36(24):2465–72.

Goetz MP, Toi M, Campone M, Sohn J, Paluch-Shimon S, Huober J, et al. MONARCH 3: Abemaciclib as initial therapy for advanced breast Cancer. J Clin Oncol. 2017;35(32):3638–46.

Gopalan PK, Villegas AG, Cao C, Pinder-Schenck M, Chiappori A, Hou W, et al. CDK4/6 inhibition stabilizes disease in patients with p16-null non-small cell lung cancer and is synergistic with mTOR inhibition. Oncotarget. 2018;9(100):37352–66.

Watt AC, Goel S. Cellular mechanisms underlying response and resistance to CDK4/6 inhibitors in the treatment of hormone receptor-positive breast cancer. Breast Cancer Res. 2022;24(1):17.

Goetz M. MONARCH 3: final overall survival results of abemaciclib plus a nonsteroidal aromatase inhibitor as first-line therapy for HR+, HER2- advanced breast cancer. SABCS; 2023.

Munzone E, Pagan E, Bagnardi V, Montagna E, Cancello G, Dellapasqua S, et al. Systematic review and meta-analysis of post-progression outcomes in ER+/HER2– metastatic breast cancer after CDK4/6 inhibitors within randomized clinical trials. ESMO Open. 2021;6(6):100332.

Gennari A, André F, Barrios CH, Cortés J, de Azambuja E, DeMichele A, et al. ESMO Clinical Practice Guideline for the diagnosis, staging and treatment of patients with metastatic breast cancer. Annals of Oncology. 2021;32(12):1475-95.

European Society for Medical Oncology (ESMO). ESMO ​Metastatic Breast Cancer Living Guideline: ER-positive HER2-negative​ Breast Cancer​ [updated May 2023. https://www.esmo.org/living-guidelines/esmo-metastatic-breast-cancer-living-guideline/er-positive-her2-negative-breast-cancer .

Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Welch PM VA, editors. Cochrane Handbook for Systematic Reviews of Interventions version 6.2 (updated February 2021). www.training.cochrane.org/handbook : Cochrane; 2021.

Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. PLoS Med. 2021;18(3):e1003583.

McGowan J, Sampson M, Salzwedel DM, Cogo E, Foerster V, Lefebvre C. PRESS peer review of electronic search strategies: 2015 Guideline Statement. J Clin Epidemiol. 2016;75:40–6.

Fraser C, Murray A, Burr J. Identifying observational studies of surgical interventions in MEDLINE and EMBASE. BMC Med Res Methodol. 2006;6(1):41.

US Food Drug Administration. Palbociclib (Ibrance). Silver Spring, MD: US Food and Drug Administration; 2017.

Book   Google Scholar  

GA Wells BS, D O’Connell J, Peterson V, Welch M, Losos PT. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses [ https://www.ohri.ca/programs/clinical_epidemiology/oxford.asp .

Lo CK-L, Mertz D, Loeb M. Newcastle-Ottawa Scale: comparing reviewers’ to authors’ assessments. BMC Med Res Methodol. 2014;14(1):45.

Martin JM, Handorf EA, Montero AJ, Goldstein LJ. Systemic therapies following progression on first-line CDK4/6-inhibitor treatment: analysis of real-world data. Oncologist. 2022;27(6):441–6.

Kalinsky KM, Kruse M, Smyth EN, Guimaraes CM, Gautam S, Nisbett AR et al. Abstract P1-18-37: Treatment patterns and outcomes associated with sequential and non-sequential use of CDK4 and 6i for HR+, HER2- MBC in the real world. Cancer Research. 2022;82(4_Supplement):P1-18-37-P1-18-37.

Choong GM, Liddell S, Ferre RAL, O’Sullivan CC, Ruddy KJ, Haddad TC, et al. Clinical management of metastatic hormone receptor-positive, HER2-negative breast cancer (MBC) after CDK 4/6 inhibitors: a retrospective single-institution study. Breast Cancer Res Treat. 2022;196(1):229–37.

Xi J, Oza A, Thomas S, Ademuyiwa F, Weilbaecher K, Suresh R, et al. Retrospective Analysis of Treatment Patterns and effectiveness of Palbociclib and subsequent regimens in metastatic breast Cancer. J Natl Compr Canc Netw. 2019;17(2):141–7.

Rozenblit M, Mun S, Soulos P, Adelson K, Pusztai L, Mougalian S. Patterns of treatment with everolimus exemestane in hormone receptor-positive HER2-negative metastatic breast cancer in the era of targeted therapy. Breast Cancer Res. 2021;23(1):14.

Bashour SI, Doostan I, Keyomarsi K, Valero V, Ueno NT, Brown PH, et al. Rapid breast Cancer Disease Progression following cyclin dependent kinase 4 and 6 inhibitor discontinuation. J Cancer. 2017;8(11):2004–9.

Giridhar KV, Choong GM, Leon-Ferre R, O’Sullivan CC, Ruddy K, Haddad T, et al. Abstract P6-18-09: clinical management of metastatic breast cancer (MBC) after CDK 4/6 inhibitors: a retrospective single-institution study. Cancer Res. 2019;79:P6–18.

Article   Google Scholar  

Mougalian SS, Feinberg BA, Wang E, Alexis K, Chatterjee D, Knoth RL, et al. Observational study of clinical outcomes of eribulin mesylate in metastatic breast cancer after cyclin-dependent kinase 4/6 inhibitor therapy. Future Oncol. 2019;15(34):3935–44.

Moscetti LML, Riggi L, Sperduti I, Piacentini FOC, Toss A, Barbieri E, Cortesi L, Canino FMA, Zoppoli G, Frassoldati A, Schirone A, Dominici MECF. SEQUENCE OF TREATMENTS AFTER CDK4/6 THERAPY IN ADVANCED BREAST CANCER (ABC), A GOIRC MULTICENTER RETRO/ PROSPECTIVE STUDY. PRELIMINARY RESULTS IN THE RETROSPECTIVE SERIES OF 116 PATIENTS. Tumori. 2022;108(4S):80.

Menichetti AZE, Giorgi CA, Bottosso M, Leporati R, Giarratano T, Barbieri C, Ligorio F, Mioranza E, Miglietta F, Lobefaro R, Faggioni G, Falci C, Vernaci G, Di Liso E, Girardi F, Griguolo G, Vernieri C, Guarneri V, Dieci MV. CDK 4/6 INHIBITORS FOR METASTATIC BREAST CANCER: A MULTICENTER REALWORLD STUDY. Tumori. 2022;108(4S):70.

Marschner NW, Harbeck N, Thill M, Stickeler E, Zaiss M, Nusch A, et al. 232P Second-line therapies of patients with early progression under CDK4/6-inhibitor in first-line– data from the registry platform OPAL. Annals of Oncology. 2022;33:S643-S4

Gousis C, Lowe KMH, Kapiris M. V. Angelis. Beyond First Line CDK4/6 Inhibitors (CDK4/6i) and Aromatase Inhibitors (AI) in Patients with Oestrogen Receptor Positive Metastatic Breast Cancer (ERD MBC): The Guy’s Cancer Centre Experience. Clinical Oncology2022. p. e178.

Endo Y, Yoshimura A, Sawaki M, Hattori M, Kotani H, Kataoka A, et al. Time to chemotherapy for patients with estrogen receptor-positive breast Cancer and cyclin-dependent kinase 4 and 6 inhibitor use. J Breast Cancer. 2022;25(4):296–306.

Li Y, Li W, Gong C, Zheng Y, Ouyang Q, Xie N, et al. A multicenter analysis of treatment patterns and clinical outcomes of subsequent therapies after progression on palbociclib in HR+/HER2- metastatic breast cancer. Ther Adv Med Oncol. 2021;13:17588359211022890.

Amaro CP, Batra A, Lupichuk S. First-line treatment with a cyclin-dependent kinase 4/6 inhibitor plus an aromatase inhibitor for metastatic breast Cancer in Alberta. Curr Oncol. 2021;28(3):2270–80.

Crocetti SPM, Tassone L, Marcantognini G, Bastianelli L, Della Mora A, Merloni F, Cantini L, Scortichini L, Agostinelli V, Ballatore Z, Savini A, Maccaroni E. Berardi R. What is the best therapeutic sequence for ER-Positive/HER2- Negative metastatic breast cancer in the era of CDK4/6 inhibitors? A single center experience. Tumori. 2020;106(2S).

Nichetti F, Marra A, Giorgi CA, Randon G, Scagnoli S, De Angelis C, et al. 337P Efficacy of everolimus plus exemestane in CDK 4/6 inhibitors-pretreated or naïve HR-positive/HER2-negative breast cancer patients: A secondary analysis of the EVERMET study. Annals of Oncology. 2020;31:S382

Luhn P, O’Hear C, Ton T, Sanglier T, Hsieh A, Oliveri D, et al. Abstract P4-13-08: time to treatment discontinuation of second-line fulvestrant monotherapy for HR+/HER2– metastatic breast cancer in the real-world setting. Cancer Res. 2019;79(4Supplement):P4–13.

Mittal A, Molto Valiente C, Tamimi F, Schlam I, Sammons S, Tolaney SM et al. Filling the gap after CDK4/6 inhibitors: Novel Endocrine and Biologic Treatment options for metastatic hormone receptor positive breast Cancer. Cancers (Basel). 2023;15(7).

Ashai N, Swain SM. Post-CDK 4/6 inhibitor therapy: current agents and novel targets. Cancers (Basel). 2023;15(6).

Higgins JPTTJ, Chandler J, Cumpston M, Li T, Page MJ, Welch VA, editors. Cochrane Handbook for Systematic Reviews of Interventions version 6.3 (updated February 2022). www.training.cochrane.org/handbook : Cochrane; 2022.

Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71.

Serdar CC, Cihan M, Yücel D, Serdar MA. Sample size, power and effect size revisited: simplified and practical approaches in pre-clinical, clinical and laboratory studies. Biochem Med (Zagreb). 2021;31(1):010502.

US Food Drug Administration. FDA approves elacestrant for ER-positive, HER2-negative, ESR1-mutated advanced or metastatic breast cancer [updated January 27 2023. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-elacestrant-er-positive-her2-negative-esr1-mutated-advanced-or-metastatic-breast-cancer .

US Food Drug Administration. FDA approves capivasertib with fulvestrant for breast cancer [updated November 16 2023. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-capivasertib-fulvestrant-breast-cancer .

Download references

Acknowledgements

The authors would like to acknowledge Joanna Bielecki who developed, conducted, and documented the database searches.

This study was funded by Pfizer Inc. (New York, NY, USA) and Arvinas (New Haven, CT, USA).

Author information

Sarah Kane, Belal Howidi, Bao-Ngoc Nguyen and Imtiaz A. Samjoo contributed equally to this work.

Authors and Affiliations

Pfizer, 10017, New York, NY, USA

Veronique Lambert & Yan Wu

EVERSANA, Burlington, ON, Canada

Sarah Kane, Belal Howidi, Bao-Ngoc Nguyen & Imtiaz A. Samjoo

Arvinas, 06511, New Haven, CT, USA

David Chandiwana & Michelle Edwards

You can also search for this author in PubMed   Google Scholar

Contributions

VL, IAS, SK, BH, BN, DC, YW, and ME participated in the conception and design of the study. IAS, SK, BH and BN contributed to the literature review, data collection, analysis, and interpretation of the data. VL, IAS, SK, BH, BN, DC, YW, and ME contributed to the interpretation of the data and critically reviewed for the importance of intellectual content for the work. VL, IAS, SK, BH, BN, DC, YW, and ME were responsible for drafting or reviewing the manuscript and for providing final approval. VL, IAS, SK, BH, BN, DC, YW, and ME meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of the work, and have given their approval for this version to be published.

Corresponding author

Correspondence to Imtiaz A. Samjoo .

Ethics declarations

Ethics approval and consent to participate.

Not applicable.

Consent for publication

Competing interests.

The authors of this manuscript declare that the research presented was funded by Pfizer Inc. and Arvinas. While the support from Pfizer Inc. and Arvinas was instrumental in facilitating this research, the authors affirm that their interpretation of the data and the content of this manuscript were conducted independently and without bias to maintain the transparency and integrity of the research. IAS, SK, BH, and BN are employees of EVERSANA, Canada, which was a paid consultant to Pfizer in connection with the development of this manuscript.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary material 2, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Lambert, V., Kane, S., Howidi, B. et al. Systematic literature review of real-world evidence for treatments in HR+/HER2- second-line LABC/mBC after first-line treatment with CDK4/6i. BMC Cancer 24 , 631 (2024). https://doi.org/10.1186/s12885-024-12269-8

Download citation

Received : 26 January 2024

Accepted : 16 April 2024

Published : 23 May 2024

DOI : https://doi.org/10.1186/s12885-024-12269-8

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• First-line CDK4/6i

ISSN: 1471-2407

• Submission enquiries: [email protected]
• General enquiries: [email protected]

Help | Advanced Search

Computer Science > Computation and Language

Title: text generation: a systematic literature review of tasks, evaluation, and challenges.

Abstract: Text generation has become more accessible than ever, and the increasing interest in these systems, especially those using large language models, has spurred an increasing number of related publications. We provide a systematic literature review comprising 244 selected papers between 2017 and 2024. This review categorizes works in text generation into five main tasks: open-ended text generation, summarization, translation, paraphrasing, and question answering. For each task, we review their relevant characteristics, sub-tasks, and specific challenges (e.g., missing datasets for multi-document summarization, coherence in story generation, and complex reasoning for question answering). Additionally, we assess current approaches for evaluating text generation systems and ascertain problems with current metrics. Our investigation shows nine prominent challenges common to all tasks and sub-tasks in recent text generation publications: bias, reasoning, hallucinations, misuse, privacy, interpretability, transparency, datasets, and computing. We provide a detailed analysis of these challenges, their potential solutions, and which gaps still require further engagement from the community. This systematic literature review targets two main audiences: early career researchers in natural language processing looking for an overview of the field and promising research directions, as well as experienced researchers seeking a detailed view of tasks, evaluation methodologies, open challenges, and recent mitigation strategies.

Submission history

Access paper:.

• HTML (experimental)
• Other Formats

References & Citations

• Google Scholar
• Semantic Scholar

BibTeX formatted citation

Bibliographic and Citation Tools

Code, data and media associated with this article, recommenders and search tools.

• Institution

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Int J Behav Nutr Phys Act

The Impact of Regular Self-weighing on Weight Management: A Systematic Literature Review

Jeffrey j vanwormer.

1 Department of Education, Minneapolis Heart Institute Foundation, 920 East 28th St, Suite 100, Minneapolis, MN 55407, USA

Simone A French

2 Division of Epidemiology & Community Health, University of Minnesota, Room 300 WBOB, 1300 South 2nd Street, Minneapolis, MN 55454, USA

Mark A Pereira

Ericka m welsh.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( http://creativecommons.org/licenses/by/2.0 ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Regular self-weighing has been a focus of attention recently in the obesity literature. It has received conflicting endorsement in that some researchers and practitioners recommend it as a key behavioral strategy for weight management, while others caution against its use due to its potential to cause negative psychological consequences associated with weight management failure. The evidence on frequent self-weighing, however, has not yet been synthesized. The purpose of this paper is to evaluate the evidence regarding the use of regular self-weighing for both weight loss and weight maintenance.

A systematic literature review was conducted using the MEDLINE, CINAHL, and PsycINFO online databases. Reviewed studies were broken down by sample characteristics, predictors/conditions, dependent measures, findings, and evidence grade.

Twelve studies met the inclusion/exclusion criteria, but nearly half received low evidence grades in terms of methodological quality. Findings from 11 of the 12 reviewed studies indicated that more frequent self-weighing was associated with greater weight loss or weight gain prevention. Specifically, individuals who reported self-weighing weekly or daily, typically over a period of several months, held a 1 to 3 kg/m 2 (current) advantage over individuals who did not self-weigh frequently. The effects of self-weighing in experimental studies, especially those where self-weighing behaviors could be isolated, were less clear.

Based on the consistency of the evidence reviewed, frequent self-weighing, at the very least, seems to be a good predictor of moderate weight loss, less weight regain, or the avoidance of initial weight gain in adults. More targeted research is needed in this area to determine the causal role of frequent self-weighing in weight loss/weight gain prevention programs. Other open questions to be pursued include the optimal dose of self-weighing, as well as the risks posed for negative psychological consequences.

Excess body weight is a leading cause of death in the U.S. [ 1 , 2 ], contributing to the development or complication of many chronic diseases including heart disease, diabetes, and cancer [ 3 , 4 ] Fortunately, even a modest amount of weight loss has been shown to reduce the incidence of chronic diseases and improve obesity-related health conditions [ 5 ]. Many people who have experienced weight issues have learned to manage their weight over the long-term via sustained, moderate caloric restriction and regular physical activity [ 6 ]. Poor adherence to these behaviors, however, is the norm for individuals attempting to lose weight or maintain weight loss [ 7 ] due to substantial physiological [ 8 ], environmental [ 9 - 11 ], and motivational barriers [ 12 ].

Because of the inherent challenges associated with losing weight or preventing weight regain, many people turn to some form of external assistance (e.g., clinical counseling program, community support group, self-help book) to help them initiate or maintain the behavior changes required to lose weight. A major component of these forms of assistance involves instruction in behavioral self-management skills like goal-setting or stimulus control. In particular, regular self-monitoring of weight has been recommended as a key component of behavioral self-regulation of body weight [ 5 ].

Regular self-weighing seems to be a common strategy for individuals who have been successful at losing weight and keeping it off. Klem and colleagues [ 13 ] found that 75 percent of a cohort of weight loss maintainers report self-weighing at least weekly. Weekly self-weighing also seems to be more common among individuals who lost weight on their own versus using an organized weight management program [ 14 ]. In contrast, the prevalence of self-weighing in the general population of healthy weight individuals is not well studied, but one study estimated that about 39 percent self-weigh weekly [ 15 ].

Frequent self-weighing is conceptualized to work via behavioral self-regulation [ 16 ]. Specifically, an individual who self-weighs often is believed to stay focused on and sensitive to changes in their weight. This creates more opportunities for self-reinforcement of even small weight loss (or weight maintenance) progress. Also, the individual is empowered to quickly identify lapses in their progress and adjust their behavior accordingly to head off substantial weight gain [ 17 ].

Despite a plausible rationale, there remains considerable debate on the utility of self-weighing in the context of weight management. Some researchers and practitioners urge caution in the use of frequent self-weighing, at least with some individuals, because it is believed to produce negative psychological conditions such as depression, anxiety, or otherwise unhealthy preoccupations and stress associated with weight [ 18 , 19 ]. Furthermore, the downstream effect of these psychological conditions produced by frequent self-weighing is believed to undermine the effectiveness of weight management interventions by negatively influencing body image and increasing program attrition. Others, however, have noted that most investigations that have observed negative psychological harms secondary to frequent self-weighing have done so only in non-overweight samples [ 20 ] and several studies have shown a strong positive association between self-weighing frequency and magnitude of weight loss.

The findings on the utility of regular self-weighing for weight management have yet to be critically reviewed or synthesized. Therefore, the purpose of this paper was to conduct a systematic review of the literature from observational and experimental studies on self-weighing in order to gauge the effectiveness of regular self-weighing on weight loss and weight maintenance (including primary weight gain prevention) in adults. The central research questions examined are: (1) Do the benefits of frequent self-weighing, in terms of body weight, outweigh the disadvantages, and (2) Do the conclusions in this regard differ by the subgroups of individuals who are interested in weight loss or weight maintenance? We hypothesize that the preponderance of evidence would support the use of frequent self-weighing behaviors to promote both weight loss and prevention of weight (re)gain. Implications in the context of weight loss research and clinical practice are also discussed.

A systematic review of the literature was conducted. MEDLINE, CINAHL, and PsycINFO online databases were searched via the University of Minnesota's Ovid interface http://www.biomed.lib.umn.edu/ovidweb/ovidweb.cgi to produce relevant articles on self-weighing and weight management. The reference sections of all included studies were also manually searched.

Inclusion and Exclusion Criteria

Inclusion criteria were: English language, adult participants, assessment of body weight, assessment of self-weighing frequency (i.e., used as a treatment component, predictor, or outcome variable), quantitative analysis of the relationship between self-weighing and body weight, and published before January 1, 2008. Since the research question was relatively broad and the body of evidence was expected to be small, no restrictions were placed on sample size, setting, research design, body weight of participants, or length of measurement follow-up. The MEDLINE search included one Medical Subject Heading, Body Weight, along with two text words; self-weighing or self-monitoring. Limiters included English language, humans, and adult population. Keywords used in the PsycINFO search included obesity, body weight, weight loss, self-weighing, and self-monitoring (limiters included English language, humans, and adult population).

Data Extraction

The outcome of interest was body weight or change in body weight. For each study, the most conservative approach to data extraction was taken by reporting only findings from the final follow-up visit, and, where possible, only those that were statistically adjusted for potential disturbance variables. To maintain the focus on the scope of the research question of interest, psychological outcomes believed to be related to self-weighing (e.g., depression, obsessive-compulsive disorder, binge eating, body image disorder) were not described. Several investigations on such negative psychological consequences secondary to self-weighing have appeared recently in the scientific literature, therefore the authors agreed this question would be best served by a separate review paper focused on that topic.

Data Synthesis

Studies were broken down by sample characteristics, predictors/conditions, dependent measures, findings, and evidence grade. A previously used adaptation of the American Diabetes Association's (ADA) evidence grading system was used (see Table ​ Table1) 1 ) [ 21 , 22 ]. Studies were assigned an evidence grade of A, B, or C along with a strength grade of 1, 2, or 3 depending on methodological quality, supporting evidence, and estimated benefits to the population at-risk.

ADA-adapted system for grading reviewed studies.

Study Characteristics

As outlined in Figure ​ Figure1, 1 , 249 articles were returned from the initial searches of the online databases. Twelve studies [ 23 - 34 ] were included in the review per the inclusion/exclusion criteria and the reasons for exclusion are also given in Figure ​ Figure1. 1 . With the exception of one study that exclusively recruited males [ 31 ], all other study samples primarily included middle-aged females. Four studies recruited females exclusively [ 24 , 25 , 27 , 33 ] and two studies were conducted in Japan [ 33 , 34 ]. Median baseline body mass index (BMI) was about 30 kg/m 2 across all samples (see Table ​ Table2 2 ).

Synopsis of reviewed studies on self-weighing and weight management.

Flow diagram of the study identification, selection, and exclusion process.

The reviewed studies were almost evenly split in their focus on either weight loss or weight maintenance. Body weight was self-reported in all three cross-sectional studies and one of the prospective cohort studies, while all other studies used objective assessments of body weight. Self-weighing, when assessed independently as a predictor variable (versus implicitly as part of a treatment package), was done so exclusively by self-report.

Research Quality

Using the ADA-adapted evidence grading system [ 21 , 22 ], only one study provided A-level evidence in terms of methodological quality [ 30 ]. All other studies provided B- or C-level evidence, primarily due to weaker research designs, high non-response or loss to follow-up, underpowered samples, and/or incomplete statistical analyses. With strength grades of 1, the studies by Linde et al. [ 26 ] and Levitsky et al. [ 28 ] provided the strongest (positive) associations between self-weighing and weight management. Six studies received strength grades of 2, while the remaining four studies received strength grades of 3.

Self-weighing and Weight Maintenance

All three cross-sectional studies indicated a significant association between the frequency of self-weighing and body weight. Specifically, nationally representative samples revealed that, relative to respondents who did not maintain their weight loss, about 60 [ 25 ] and 80 [ 29 ] percent more respondents who were successful at keeping their weight off reported weekly or daily self-weighing, respectively. The study by Linde et al. [ 27 ] indicated that women who reported daily self-weighing weighed nearly 2 BMI units less than women who reported never self-weighing.

Similar to the results observed in her cross-sectional analysis [ 27 ], Linde et al. [ 26 ] also found that participants in the Pound of Prevention [ 15 ] cohort analysis who reported daily self-weighing at the two-year follow-up weighed nearly 2 BMI units less than participants who reported never self-weighing. Butryn et al. [ 23 ] found that participants who increased their frequency of self-weighing (unspecified magnitude of increase) over one year gained 2.5 kg less weight relative to participants who decreased their frequency of self-weighing over this same time period.

Two randomized-controlled trials focused on weight maintenance. Secondary analyses by Wing and colleagues [ 30 ] found that, in both study treatment conditions, 41 to 55 percent fewer participants who self-weighed daily regained ≥ 2.3 kg relative to participants who did not self-weigh daily. In two separate, short-term experiments conducted by Levitsky et al. [ 28 ] a 3 kg weight advantage was noted for college females who received a daily self-weighing and feedback intervention relative to participants who received information-only or assessment-only treatments.

Self-weighing and Weight Loss

In terms of weight loss, a prospective cohort analyses by Linde et al. [ 26 ] found that participants in the Weigh To Be trial [ 35 ] who self-weighed daily lost about 1 BMI unit more than participants who self-weighed weekly and nearly 3 BMI units more than participants who did not self-weigh at all. Jeffery and colleagues [ 31 ] found that, after two years, participants in a 15-week behavioral weight loss program who reported daily self-weighing lost about 15 pounds more than participants who reported self-weighing less than daily. Qi et al. [ 24 ] found that, compared to participants who did not lost 5 kg or more, participants who lost 5 kg or more in a behavioral weight loss program significantly increased their daily self-weighing frequency between baseline and 6-months follow-up. An uncontrolled cohort study by Tanaka and colleagues [ 33 ] found that participants who completed a nutrition-focused weight loss program whereby they self-weighed 4 times per day lost a significant amount of weight (~4 kg) over 16 weeks.

Two randomized-controlled trials isolated the effects of frequent self-weighing on weight loss by comparing it to an identical intervention that contained everything but the self-weighing component. These two trials found conflicting results. Heckerman and colleagues [ 32 ] found no advantage of frequent self-weighing above and beyond a standard 10-week behavioral weight loss program. Fujimoto and colleagues [ 34 ], however, found that a group that self-weighed very frequently (4 times daily) lost twice as much weight over a two year period relative to a group that received behavioral weight loss therapy alone.

In all but one of the twelve reviewed studies, frequent self-weighing (defined as self-weighing weekly or daily), or treatment groups that utilized frequent self-weighing, was associated with significantly greater weight loss, weight maintenance, or less body weight in general relative to infrequent self-weighing. Generally speaking, weekly and daily self-weighers held approximately a 1 and 2 BMI unit advantage, respectively, over never self-weighers [ 26 , 27 ]. In regard to weight loss, weekly and daily self-weighers lost about 2 to 3 BMI units (~12-18 pounds) more than participants who did not weigh as frequently [ 26 , 31 ]. Based largely on the consistency of the evidence reviewed, frequent self-weighing, at the very least, seems to be a good predictor of moderate weight loss and weight maintenance, both for individuals who have lost weight and are attempting to keep it off and for individuals who are attempting to avoid weight gain in the first place.

Only three studies [ 28 , 32 , 34 ] directly tested a self-weighing intervention that was not part of an extensive treatment package, relative to control groups that did not receive self-weighing advice or support. These trials were small, showed conflicting results, and raised concerns over internal and external validity. Levitsky and colleagues [ 28 ] utilized a brief self-weighing intervention for weight maintenance among female freshman college students (generally non-overweight) and found about 1 BMI unit less weight gain for intervention participants relative to controls. Heckerman, et al. [ 32 ] found no significant weight loss advantage for participants enrolled in a 10-week behavioral weight loss program that included weekly weigh-ins relative to a group that received the same treatment program without weekly weigh-ins. Attrition was extremely high in this small sample, however, and few conclusions could be drawn from the results. In a somewhat similar approach, Fujimoto et al. [ 34 ] also tested a behavioral weight loss program with and without frequent self-weighing. The self-weighing in this study included a recommendation to chart one's weight four times per day. The findings indicated a strong effect in that the self-weighing group lost nearly twice as much weight as compared to the group that did not self-weigh. Analyses from this study, however, were difficult to interpret due to the exclusion of a large part of the randomized sample. As such, selection bias may have been present.

An ideal objective of this review would be to draw conclusions on the optimal dose of self-weighing. At this time, however, the evidence base does not support endorsement of a precise self-weighing frequency and duration that has the most benefit for the most people. In terms of a threshold, weekly self-weighing over several months stands out as what may be the minimum point at which meaningful weight benefits begin to accrue. This assertion is primarily based on the Linde et al. studies [ 26 , 27 ], which were the only ones with enough power to retain a precise assessment of self-weighing frequency across several levels (e.g., daily, weekly, monthly, rarely) versus a dichotomized characterization (e.g., daily, less than daily). These studies found statistically significant benefit, in terms of weight loss, weight maintenance, and weight in general, beginning at weekly self-weighing. It was not clear from any of the reviewed studies if more than daily self-weighing confers added weight benefits.

Perhaps the most significant methodological limitation of the reviewed studies involved the potential for measurement bias. Self-weighing was assessed exclusively by self-report. In order to prevent recall bias, questionnaire items can not be practically designed to examine self-weighing in a timeframe that extends far beyond the point at which the question is asked. As such, the characterization of self-weighing reported may not accurately reflect self-weighing over the time periods they are deemed to represent. In other words, reported self-weighing frequency at the end of a study may not truly represent the degree to which self-weighing actually occurred over the course of the entire study (or in the months since the last follow-up visit). More objective means of assessing self-weighing frequency, such as scales that record time/date of weigh-ins at home, are needed to validate self-reported measures. Also, the overall demographic profile of study samples was somewhat narrow, primarily involving middle-aged American female volunteers. This seems to be the group most likely to present for weight management services, but it limits generalizations on the effects of self-weighing across the general population.

Although frequent self-weighing was included as part of a treatment package in one large, well-conducted, randomized-controlled trial [ 30 ], only three studies were able to experimentally isolate or disaggregate the effects of frequent self-weighing. These studies were small and contained several methodological flaws, however, and therefore strong conclusions could not be drawn. Results from the cross-sectional and prospective cohort studies are also insufficient to make causal claims due to temporality and selection bias issues. Given the controversial endorsement of frequent self-weighing in the scientific community [ 18 , 19 ], it seems timely to experimentally investigate different frequencies and durations of self-weighing in a diverse sample using a large randomized-controlled trial. Also, based on the differential associations of self-weighing across different treatment intensities employed by Wing and colleagues [ 30 ], the interaction between frequent self-weighing and other weight management program components would be useful for practitioners to better understand. Furthermore, more sensitive analyses are needed to identify the subgroups of people who benefit most from frequent self-weighing. For example, many of the weight maintenance studies reviewed had combined samples of individuals who had lost weight previously and were seeking to prevent weight regain alongside individuals who were normal weight and were seeking to prevent weight gain in the first place. There may be subtle differences in such subgroups that could help practitioners and program designers offer the most appropriate advice.

In balancing the strengths and weaknesses of the evidence reviewed, frequent self-weighing seems to be a helpful strategy for adults who have been successful at losing weight, maintaining weight loss, or preventing weight gain. Furthermore, frequent self-weighing may serve as a useful component of standard weight loss treatment packages. At this time, weekly self-weighing seems to be a reasonable strategy to endorse for adults, but more research is needed to firmly establish the independent causal effect, as well as the optimal dose, both in terms of frequency and duration, of self-weighing. Also, more research needs to be done to determine if self-weighing is more or less effective in specific population subgroups and to identify the potential for psychological risks associated with very frequent self-weighing.

Competing interests

The authors declare that they have no competing interests.

Authors' contributions

JJV conceived of the review design and drafted the initial manuscript. SAF participated in the review design and helped draft manuscript revisions. MAP participated in the review design and helped draft manuscript revisions. EMW helped draft manuscript revisions. All authors read and approved the final manuscript.

Acknowledgements

• Ogden CL, Yanovski SZ, Carroll MD, Flegal KM. The epidemiology of obesity. Gastroenterology. 2007; 132 :2087–2102. doi: 10.1053/j.gastro.2007.03.052. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Flegal KM, Graubard BI, Williamson DF, Gail MH. Excess deaths associated with underweight, overweight, and obesity. JAMA. 2005; 293 :1861–1867. doi: 10.1001/jama.293.15.1861. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Bray GA. Medical consequences of obesity. J Clin Endocrinol Metab. 2004; 89 :2583–2589. doi: 10.1210/jc.2004-0535. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Pi-Sunyer FX. Comorbidities of overweight and obesity: Current evidence and research issues. Med Sci Sports Exerc. 1999; 31 :S602–S608. doi: 10.1097/00005768-199911001-00019. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• National Institute of Health. National Heart, Lung, and Blood Institute Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults. The evidence report. Obes Res Suppl. 1998; 6 :51S–209S. [ PubMed ] [ Google Scholar ]
• Wing RR, Hill JO. Successful weight loss maintenance. Annu Rev Nutr. 2001; 21 :323–341. doi: 10.1146/annurev.nutr.21.1.323. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Linde JA, Erickson DJ, Jeffery RW, Pronk NP, Boyle RG. The relationship between prevalence and duration of weight loss strategies and weight loss among overweight managed care organization members enrolled in a weight loss trial. Int J Behav Nutr Phys Act. 2006; 3 :3. doi: 10.1186/1479-5868-3-3. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Perri MG, Corsica JA. Improving maintenance of weight lost in behavioral treatment of obesity. In: Wadden TA, Stunkard AJ, editor. Handbook of Obesity Treatment. 3. New York, Guilford Press; 2002. pp. 357–379. [ Google Scholar ]
• Popkin BM, Duffey K, Gordon-Larsen P. Environmental influences on food choice, physical activity and energy balance. Physiol Behav. 2005; 86 :603–613. doi: 10.1016/j.physbeh.2005.08.051. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Booth SL, Sallis JF, Ritenbaugh C, Hill JO, Birch LL, Frank LD, Glanz K, Himmelgreen DA, Mudd M, Popkin BM, Rickard KA, St Jeor S, Hays NP. Environmental and societal factors affect food choice and physical activity: Rationale, influences, and leverage points. Nutr Rev Suppl. 2001; 59 :S21–S39. [ PubMed ] [ Google Scholar ]
• French SA, Story M, Jeffery RW. Environmental influences on eating and physical activity. Annu Rev Public Health. 2001; 22 :309–335. doi: 10.1146/annurev.publhealth.22.1.309. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Jeffery RW, Kelly KM, Rothman AJ, Sherwood NE, Boutelle KN. The weight loss experience: A descriptive analysis. Ann Behav Med. 2004; 27 :100–106. doi: 10.1207/s15324796abm2702_4. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Klem ML, Wing RR, McGuire MT, Seagle HM, Hill JO. A descriptive study of individuals successful at long-term maintenance of substantial weight loss. Am J Clin Nutr. 1997; 66 :239–246. [ PubMed ] [ Google Scholar ]
• McGuire MT, Wing RR, Klem ML, Seagle HM, Hill JO. Long-term maintenance of weight loss: Do people who lose weight through various weight loss methods use different behaviors to maintain their weight? Int J Obes Relat Metab Disord. 1998; 22 :572–577. doi: 10.1038/sj.ijo.0800627. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Jeffery RW, French SA. Preventing weight gain in adults: The pound of prevention study. Am J Public Health. 1999; 89 :747–751. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Kanfer FH, Karoly P. Self-control. A behavioristic excursion into the lion's den. Behav Ther. 1972; 3 :398–416. doi: 10.1016/S0005-7894(72)80140-0. [ CrossRef ] [ Google Scholar ]
• Boutelle K. Weighing the evidence: Benefits of regular weight monitoring for weight control. J Nutr Educ Behav. 2006; 38 :131. doi: 10.1016/j.jneb.2006.04.002. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Dionne MM, Yeudall F. Monitoring of weight in weight loss programs: A double-edged sword? J Nutr Educ Behav. 2005; 37 :315–318. doi: 10.1016/S1499-4046(06)60162-0. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Neumark-Sztainer D, Berg P van den, Hannan PJ, Story M. Self-weighing in adolescents: Helpful or harmful? Longitudinal associations with body weight changes and disordered eating. J Adolesc Health. 2006; 39 :811–818. doi: 10.1016/j.jadohealth.2006.07.002. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• O'Neil PM, Brown JD. Weighing the evidence: Benefits of regular weight monitoring for weight control. J Nutr Educ Behav. 2005; 37 :319–322. doi: 10.1016/S1499-4046(06)60163-2. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Davidson MB, (editor) American Diabetes Association: Clinical practice recommendations 2003. Diabetes Care. 2003; 26 :S1–S156. Theme issue. [ Google Scholar ]
• VanWormer JJ, Boucher JL, Pronk NP. Telephone-based counseling improves dietary fat, fruit, and vegetable consumption: A best-evidence synthesis. J Am Diet Assoc. 2006; 106 :1434–1444. doi: 10.1016/j.jada.2006.06.008. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Butryn ML, Phelan S, Hill JO, Wing RR. Consistent self-monitoring of weight: A key component of successful weight loss maintenance. Obesity. 2007; 15 :3091–3096. doi: 10.1038/oby.2007.368. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Qi BB, Dennis KE. The adoption of eating behaviors conducive to weight loss. Eat Behav. 2000; 1 :23–31. doi: 10.1016/S1471-0153(00)00003-9. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• McGuire MT, Wing RR, Klem ML, Hill JO. Behavioral strategies of individuals who have maintained long-term weight losses. Obes Res. 1999; 7 :334–341. [ PubMed ] [ Google Scholar ]
• Linde JA, Jeffery RW, French SA, Pronk NP, Boyle RG. Self-weighing in weight gain prevention and weight loss trials. Ann Behav Med. 2005; 30 :210–216. doi: 10.1207/s15324796abm3003_5. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Linde JA, Jeffery RW, Finch EA, Simon GE, Ludman EJ, Operskalski BH, Ichikawa L, Rohde P. Relation of body mass index to depression and weighing frequency in overweight women. Prev Med. 2007; 45 :75–79. doi: 10.1016/j.ypmed.2007.03.008. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Levitsky DA, Garay J, Nausbaum M, Neighbors L, Dellavalle DM. Monitoring weight daily blocks the freshman weight gain: a model for combating the epidemic of obesity. Int J Obes. 2006; 30 :1003–1010. doi: 10.1038/sj.ijo.0803221. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Kruger J, Blanck HM, Gillespie C. Dietary and physical activity behaviors among adults successful at weight loss maintenance. Int J Behav Nutr Phys Act. 2006; 3 :17. doi: 10.1186/1479-5868-3-17. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Wing RR, Tate DF, Gorin AA, Raynor HA, Fava JL. A self-regulation program for maintenance of weight loss. N Engl J Med. 2006; 355 :1563–1571. doi: 10.1056/NEJMoa061883. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Jeffery RW, Bjornson-Benson WM, Rosenthal BS, Lindquist RA, Kurth CL, Johnson SL. Correlates of weight loss and its maintenance over two years of follow-up among middle-aged men. Prev Med. 1984; 13 :155–168. doi: 10.1016/0091-7435(84)90048-3. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Heckerman CL, Brownell KD, Westlake RJ. Self and external monitoring of weight. Psychol Rep. 1978; 43 :375–378. [ PubMed ] [ Google Scholar ]
• Tanaka M, Itoh K, Abe S, Imai K, Masuda T, Koga R, Itoh H, Konomi Y, Kinukawa N, Sakata T. Irregular patterns in the daily weight chart at night predict body weight regain. Exp Biol Med. 2004; 229 :940–945. [ PubMed ] [ Google Scholar ]
• Fujimoto K, Sakata T, Etou H, Fukagawa K, Ookuma K, Terada K, Kurata K. Charting of daily weight pattern reinforces maintenance of weight reduction in moderately obese patients. Am J Med Sci. 1992; 303 :145–150. doi: 10.1097/00000441-199203000-00002. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Jeffery RW, Sherwood NE, Brelje K, Pronk NP, Boyle R, Boucher JL, Hase K. Mail and phone interventions for weight loss in a managed-care setting: Weigh-To-Be one-year outcomes. Int J Obes Relat Metab Disord. 2003; 27 :1584–1592. doi: 10.1038/sj.ijo.0802473. [ PubMed ] [ CrossRef ] [ Google Scholar ]

The impact of regular self-weighing on weight management: a systematic literature review

Affiliation.

• 1 Department of Education, Minneapolis Heart Institute Foundation, 920 East 28th St, Suite 100, Minneapolis, MN 55407, USA. [email protected]
• PMID: 18983667
• PMCID: PMC2588640
• DOI: 10.1186/1479-5868-5-54

Background: Regular self-weighing has been a focus of attention recently in the obesity literature. It has received conflicting endorsement in that some researchers and practitioners recommend it as a key behavioral strategy for weight management, while others caution against its use due to its potential to cause negative psychological consequences associated with weight management failure. The evidence on frequent self-weighing, however, has not yet been synthesized. The purpose of this paper is to evaluate the evidence regarding the use of regular self-weighing for both weight loss and weight maintenance.

Methods: A systematic literature review was conducted using the MEDLINE, CINAHL, and PsycINFO online databases. Reviewed studies were broken down by sample characteristics, predictors/conditions, dependent measures, findings, and evidence grade.

Results: Twelve studies met the inclusion/exclusion criteria, but nearly half received low evidence grades in terms of methodological quality. Findings from 11 of the 12 reviewed studies indicated that more frequent self-weighing was associated with greater weight loss or weight gain prevention. Specifically, individuals who reported self-weighing weekly or daily, typically over a period of several months, held a 1 to 3 kg/m2 (current) advantage over individuals who did not self-weigh frequently. The effects of self-weighing in experimental studies, especially those where self-weighing behaviors could be isolated, were less clear.

Conclusion: Based on the consistency of the evidence reviewed, frequent self-weighing, at the very least, seems to be a good predictor of moderate weight loss, less weight regain, or the avoidance of initial weight gain in adults. More targeted research is needed in this area to determine the causal role of frequent self-weighing in weight loss/weight gain prevention programs. Other open questions to be pursued include the optimal dose of self-weighing, as well as the risks posed for negative psychological consequences.