Introduction

Police-community relations in New York City have long been a topic of contentious debate (see Fagan et al, 2010 for a review), with investigative stops, alternatively known as “stop-and-frisk,” “Terry stops,” or “stop question and frisk” activity, a practice of particular prominence in public discourse and litigation since the 1990s [1–4]. Implemented by the New York City Police Department (NYPD) as a method of Broken Windows policing [5], stop-and-frisk was promoted as a tactic for enforcement of laws aimed against low-level social disorder with a parallel focus on the proactive detection and seizure of illegal guns [6]. However, despite officers recording more than 4 million stops over a ten-year period from 2004–2013 [7], the vast majority of these stops did not lead to arrest or the detection of illegal activity. Stop-and-frisk encounters are a source of stress and disempowerment [8–10] and carry the possibility of escalating to violence, raising the risk of physical injury and mortality [9, 10]. A single encounter with the police may be an acute stressor; moreover, individuals who are stopped repeatedly or live in high-stop neighborhoods, may experience this police presence as a chronic stressor.

The NYPD’s stop-and-frisk program was found to violate 4^th Amendment protections against unlawful searches and seizures, and these practices were further found to violate 14^th Amendment protections [2, 11]; nearly 90% of individuals who were stopped were Black and Hispanic/Latinx, a disparity that exceeded what might be predicted either by race differences in either population or neighborhood crime patterns [12]. Black men and women face unacceptable brutality and death at the hands of law enforcement [13]. In response to legal action taken against the NYPD stop-and-frisk program, reported stop activity has declined by more than 90% since its 2011 peak [7]. The decline reflects both a change in policing practices and recording practices [14]. However, the independent police monitor overseeing police stops has cited underreporting of stops as a serious problem [14], and it is widely recognized that New Yorkers face far more invasive policing than official statistics indicate [14]. The magnitude of the underreporting is unclear given a dearth of research on the magnitude of stop-and-frisk in the city.

While racial disparities in policing are well documented, we know little about other dimensions of inequity in policing. There is evidence that sexual minority groups face disproportionate risk of policing; in a 2013 study conducted among approximately 2,000 LGBTQ persons, almost three-quarters had had contact with the police in the past five years [15]. Within that sample, approximately 20% reported that the police had a hostile attitude, with about 10% being physically searched and 14% being verbally assaulted; those who identified as a person of color were about two times more likely to report these experiences [15]. There has been scant research on the burden of police contact faced by sexual minority men. Given evidence of the disproportionate burden of police exposure among sexual minority groups, a study measuring levels of stop-and-frisk experience that also assesses risk of police exposure among those at the intersection of both sexual and racial/ethnic minority status is warranted.

The purpose of the current study was to describe levels of and factors associated with self-reported stop-and-frisk experience in young sexual minority men (YSMM) in New York City. We used data collected during the P18 Cohort Study (Version 2), a study conducted to measure determinants of inequity in STI/HIV risk in a racially and ethnically diverse cohort of young sexual minority men (YSMM) in New York City. To address the gap in the literature on burden of stop-and-frisk in YSMM, we aimed to describe levels and trends of stop-and-frisk experience over the cohort follow-up period (2014 to 2019). We hypothesized we would observe a substantial proportion of YSMM have a history of stop-and-frisk experience, with a disproportionate burden in racial/ethnic YSMM. We also hypothesized we would observe decreases in stop-and-frisk over time consistent with the decreases in reported stop-and-frisk activity by NYPD yet that across-time racial/ethnic disparities in stop-and-frisk would persist. To address concerns that official counts of stop-and-frisk in NYC based on NYPD records do not reflect the experience of stop-and-frisk burden in the population we aimed to compare rates of self-reported stop-and-frisk experience in the P18 cohort with stop-and-frisk rates among young men based on NYPD official reports, hypothesizing rates of police contact would be higher in the P18 sample. To better understand racial/ethnic disparities in stop-and-frisk of YSMM and where in NYC racial/ethnic and sexual minority men faced the greatest inequity in police exposure, we sought to describe the geographic distribution of police exposure and disparities therein. We hypothesized that respondents of all racial/ethnic backgrounds living in higher-stop areas would have had high risk of prior contact with the police [12], while racial/ethnic disparities in stop-and-frisk would be pronounced in neighborhoods where stop-and-frisk activity was less prevalent overall. Finally, to best understand the social and health needs of YSMM who come into contact with the police, we also assessed socio-demographic, psychosocial vulnerability, and behavioral correlates of stop-and-frisk. We hypothesized YSMM with the greatest social vulnerability (poverty, mental health symptoms) would have had greatest risk of contact with the police. We also hypothesized drug use would be a strong correlate of stop-and-frisk among White YSMM but not among minority YSMM given disproportionate stop-and-frisk risk in minority YSMM including among those who do not endorse drug use.

Materials and methods

Results

Levels of stop-and-frisk by race/ethnicity among YSMM

Over the study period, 43% of P18 participants reported a history of stop-and-frisk experience in the prior year, with higher levels among Black (47%) and Hispanic/Latinx (45%) than White (38%) participants.

As shown in Fig 1a, at baseline, the prevalence of any self-reported stop-and-frisk experience for the 12-months preceding study interview was higher among Black (35%) and Hispanc/Latinx (34%) YSMM compared to their White (31%) counterparts; in the sample overall 33% experienced stop-and-frisk in the past year. While there were declines in the prevalence of stop-and-frisk encounters for all groups between the baseline and 12-month follow-up visit, the racial disparities in stop-and-frisk evident at the baseline visit remained at the 12-month follow-up visit and persisted over the remaining study period despite declines in all three groups. Specifically, for White and Hispanic/Latinx YSMM, there was almost a 65% (31% to 11% and 34% to 12%, respectively) decrease in stop-and-frisk over the study period. However, among Black YSMM, the decline in stop-and-frisk encounters was only 54% (35% to 16%). Most importantly, at every visit during the study period, prevalence of any stop-and-frisk reports were higher among Black YSMM than among Hispanic/Latinx and White YSMM, with Black-White differences consistently statistically significant. There was substantial heterogeneity in the levels of past year stop-and-frisk among those who endorsed stop-and-frisk over time. The maximum number of stops per person was markedly higher among Black and Hispanic/Latinx vs White respondents (Black: 20, Hispanic: 23, White: 12 stops) and remained disparate over follow-up through the 36-month follow-up visit (Black: 66, Hispanic: 5, White: 2 stops). Of note, the Black participant with 66 stops was an outlier who also reported high scores on measures of personal stigma, public stigma, internalized homophobia, and PTSD symptoms. We discuss the sensitivity of our findings to this participant’s inclusion below.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1.

a. Race differences in past year stop-and-frisk prevalence of past year stop-and-frisk* among P18 participants over cohort follow-up (2014–2019; N = 591 at baseline). *P18 questionnaire item: “How many times have you been stopped, questioned, and/or frisked in the past year?” We coded a dichotomous indicator of any prior stop-and-frisk in the past year. In the sample overall, 32.3% reported being stopped-and-frisk at baseline and 21.5% reported being stopped-and-frisked at least once over the course of the follow-up. b. Race differences in past year stop-and-frisk rate* among P18 participants over cohort follow-up (above bar chart) and unadjusted rate ratios (RRs) and 95% confidence intervals (CIs) for associations between race and stop-and-frisk rate (below table) (2014–2019; N = 591 at baseline). *P18 questionnaire item: “How many times have you been stopped, questioned, and/or frisked in the past year?” P18 rates were calculated as the race-specific number of stop-and-frisk events reported by P18 participants/race-specific P18 study population times 1,000. Note: Rate among Black P18 participants at the 36-month visit was driven by a high stop rate among one participant (66 stops in the past year); when this participant was excluded from the analysis the stop rate was reduced to 0.5 from 0.9 among the remaining participants.

https://doi.org/10.1371/journal.pone.0256201.g001

Next, we examined per-participant stop rates (i.e., the reported number of encounters per participant) across racial/ethnic groups. As shown in Fig 1b, participant experiences of stop-and-frisk decreased between the baseline visit and the 36-month follow-up visit. However, these decreases in rates of stop-and-frisk were not consistent across race/ethnicity. Specifically, among White YSMM, the rate of stop-and-frisk decreased by 83% from 0.6 to 0.1 from baseline to the end of the study period. A similar trend was observed among Hispanic/Latinx YSMM with stop-and-frisk rates decreasing by 80% from 1 to 0.2 over the study period. However, among Black YSMM, rates of stop-and-frisk encounters did not decline consistently over the study period. Black participants experienced a comparable decline in per-participant stop rate (75% decline; from 1.2 to 0.3 encounters per person) through the 24-month follow-up, but a considerable uptick at the 36-month interview (66% increase; from 0.3 to 0.9 encounters per person). This increase was driven in part by a small number of respondents reporting very high stop rates (e.g., the aforementioned participant reporting 66 times stopped in the prior year at the 36-month follow-up) but suggests that the decline was not sustained for Black participants to the extent it was for those who were White or Hispanic/Latinx. Black participants had twice the stop-and-frisk rate as White participants at the baseline visit (RR: 2.09, 95% CI: 1.68, 2.63) and 12-month visit (RR: 2.13, 95% CI: 1.46, 3.18). By the 36-month follow-up the disparity grew to five times the rate (RR: 5.89, 95% CI: 3.63, 10.28); yet when we removed from the analysis one Black participant who reported 66 stop-and-frisk experiences the stark disparity remained (RR: 3.15, 95%: 1.89, 5.59).

Levels of self-reported stop-and-frisk among YSMM versus official reports in the general population

In Fig 2 we compare rates of per capita stop-and-frisk experience among study participants to that of young men in New York City at large. Notably, the overall decline in stop-and-frisk in our study sample parallel the population-level decline in stop-and-frisk among males ages 20–29 years old during this same period. Racial/ethnic disparities are similarly comparable to those at the population level, though the magnitudes of these disparities are less extreme. However, the per capita rate of stops reported by P18 participants of all races is significantly higher than that recorded citywide, in some cases by orders of magnitude.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Stop-and-frisk rate per 1,000 population* reported by P18 participants vs recorded by the New York City Police Department (NYPD), by race.

*P18 rates were calculated as the race-specific number of stop-and-frisk events reported by P18 participants/race-specific P18 study population. These were compared to rates in the general population, calculated as the race-specific number of stop-and-frisk events recorded by the New York City Police Department/race-specific American Community Survey New York City population estimates. Note: Rate among Black P18 participants at the 36-month visit was driven by a high stop rate among one participant (66 stops in the past year); when this participant was excluded from the analysis the stop rate was reduced to 58.6 per 1,000.

https://doi.org/10.1371/journal.pone.0256201.g002

Spatial distribution of stop-and-frisk in New York City among YSMM, by race/ethnicity

Fig 3 provides the geographic distribution of P18 respondents based on their reported ZIP codes of residence and indicates that at baseline, P18 participants residing throughout all five New York City boroughs including Manhattan and the four outer boroughs (Brooklyn, Queens, the Bronx, and Staten Island) reported stop-and-frisk. Observed racial disparities in self-reported stop-and-frisk experiences may have been driven in neighborhood exposure. White P18 respondents were more likely to live in Manhattan than their Black and Hispanic/Latinx counterparts, who were more likely to live in high-stop areas of the Brooklyn and the Bronx, which are neighborhoods that have historically experienced high rates of aggressive policing [12]. Stop experience among Black and Hispanic/Latinx participants was not limited to those living in high-stop neighborhoods; those living outside of New York City also reported having been stopped.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Geographic distribution of past year stop-and-frisk among P18 participants at cohort baseline, by race: P18 participants are mapped by participant ZIP code of residence.

Those with a history of stop-and-frisk are indicated by red circles (White race), green circles (Hispanic/Latinx race/ethnicity), blue circles (Black race) with increasing circle size indicating increasing number of stop-and-frisk events, while those with no stop-and-frisk history are indicated by Black circles.

https://doi.org/10.1371/journal.pone.0256201.g003

Racial/ethnic disparities in stop-and-frisk, by neighborhood-level stop activity

Given the likely role of neighborhood policing in participants’ exposure to police encounters, we examined racial/ethnic differences in their reported rates of stop-and-frisk (i.e., reported baseline stops per participant) among study participant by tertiles of neighborhood-level rates (Fig 4). Within the lowest stop New York City neighborhoods (i.e., the lowest tertile), White (0.45) and Hispanic/Latinx (0.42) YSMM had substantially lower stop-and-frisk rates than Black YSMM (1.2). In neighborhoods with the second-highest tertile of neighborhood-level stop-and-frisk rates, White YSMM had the lowest rate of stop-and-frisk encounters (0.51) compared to their Hispanic/Latinx (0.78) and Black (1.33). However, in neighborhoods comprising the highest tertile of neighborhood-level stop-and-frisk rates, rates of stop-and-frisk encounters among White YSMM (0.92) were only slightly lower than that among Black YSMM (1.11). Moreover, rates of stop-and-frisk encounters among Hispanic/Latinx (2.38) YSMM were more than two times higher than among their White and Black YSMM. In neighborhoods with low and moderate stop-and-frisk rates, Black participants had over 2.5 times the stop-and-frisk rates (lowest tertile neighborhood stop-and-frisk: RR 2.66, 95% CI: 1.68, 4.44; middle tertile neighborhood stop-and-frisk: RR: 2.63, 95% CI: 1.87, 3.75) while in the highest stop-and-frisk tertile Black and White P18 participants had comparable rates (RR: 1.15, 95% CI: 0.75, 1.82).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Race differences in past year stop-and-frisk rate among P18 participants at cohort baseline in 2014 by neighborhood-level stop-and-frisk rate (counts/1000)*.

*P18 participants were categorized according to their neighborhood-level stop-and-frisk rate, which were calculated as followed. Participant’s ZIP code was translated to latitude and longitude, which was then used to identify the New York City Police Department (NYPD) precinct in which the respondent lived (total of 77 possible precincts). NYPD precinct-level stop-and-frisk rates were calculated as the 2014 NYPD recorded number of stop-and-frisk events divided by the precinct population of male residents aged 20–29 years old; restricted given the younger age range of the P18 sample. The population estimates for New York City were calculated from 2014 estimates from the 2010–2014 5-year American Community Survey; population estimates for each census tract of New York City were calculated after restricting the sample to males aged 20–29 years. Each of the 77 precincts were categorized into group groups based on the precinct-level stop-and-frisk tertile. Within each tertile indicating neighborhood-level stop-and-frisk, stop-and-frisk rates were calculated at the race-specific number of stop-and-frisk events reported by P18 participants/race-specific P18 study population.

https://doi.org/10.1371/journal.pone.0256201.g004

Individual-level factors associated with stop-and-frisk

Finally, in analyses assessing associations between stop-and-frisk and socio-demographic, psychosocial and substance use factors (Table 1), the odds of multiple stop-and-frisk encounters among Black and Hispanic/Latinx YSMM were almost 2 times higher than for White YSMM (OR = 2.35, 95% CI 1.39, 4.08 and OR = 1.82, 95% CI 1.02, 3.32, respectively). Next, while 2% of participants reported they did not identify with a binary gender, the odds of any stop-and-frisk encounter as well as multiple encounters were 2 to almost 5 times higher in this group (OR = 3.51, 95% CI 1.15, 11.7 and OR = 5.78, 95% CI 1.88, 18.28, respectively). Among YSMM who reported being more likely to socialize or hang out in the Bronx, the odds of both any and multiple stop-and-frisk encounters where higher (OR = 2.90, 95% CI 1.57, 5.43 and OR = 2.76, 95% CI 1.42, 5.25, respectively) than among YSMM reporting socializing in other New York City boroughs.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Odds ratios (ORs) and 95% confidence intervals (CIs) for associations between participant socio-demographic, psychosocial, and behavioral factors and past year stop-and-frisk at cohort baseline in 2014.

https://doi.org/10.1371/journal.pone.0256201.t001

In terms of psychosocial factors, YSMM reporting higher levels of internalized homophobia had increased odds of reporting both any and multiple stop-and-frisk encounters (OR = 1.58, 95% CI 1.05, 2.37 and OR = 1.64, 95% CI 1.01, 2.63, respectively). With respect to mental health burdens, YSMM who reported any and multiple stop-and-frisk encounters had consistently higher odds of severe levels of depressive symptoms, severe levels of anxiety, and experiences of childhood victimization.

With regard to recent substance use, YSMM who reported using marijuana in the 30 days preceding study visits had higher odds of any and multiple stop-and-frisk encounters (OR = 1.97, 95% CI 1.41, 2.79 and OR = 1.79, 955 CI 1.18, 2.75). Similar relationships between any and multiple stop-and-frisk encounters were observed among individuals who reported hard drug use. Finally, the odds of multiple stop-and-frisk encounters were higher among YSMM who reported recent cocaine use (OR = 1.92, 95% CI 1.06, 3.34). While among White respondents levels of past year stop-and-frisk were markedly higher among those who reported past 30 day marijuana use (41%) than those reporting no use (17%), among Black and Hispanic/Latinx respondents, levels were comparable regardless of past 30 day marijuana use (marijuana use: 38%, no marijuana use: 31%) (race differences in associations between drug use and stop-and-frisk not presented in Table 1).

Discussion

This study is among the first to measure the prevalence of recent and repeat stop-and-frisk experience in a population-based sample [8], and the only study to our knowledge to measure stop-and-frisk in YSMM. We observed that being stop-and-frisked by the police is a normative life event among YSMM in New York City with one in three YSMM having experienced the event at baseline, primarily administered in 2014–2015, with comparable levels of the event reported among Black, non-Black Hispanic/Latinx, and White participants. While levels decreased over time, consistent with documented city-wide trends in stop-and-frisk [7], at the final follow-up a significant minority of P18 participants continued to report past year stop-and-frisk, including one in nine White and non-Black participants and one in seven Black participants. The vast differences in the levels of self-reported stop-and-frisk experience in the P18 compared with per capita rates based on the NYPD records are consistent with documented underreporting in NYPD records [14], but also suggest that sexual minority men across race/ethnicity may experience much higher rates of police contact than non-sexual minority populations. We observed Black YSMM who lie at the intersection of racial minority and sexual minority status face the greatest vulnerability and that YSMM with the greatest economic and social vulnerability were most likely to report stop-and-frisk experience. Those with very high numbers of stop-and-frisk events demonstrated substantial psychosocial vulnerability highlighting how syndemics of stigma, internalized homophobia, and adverse mental health, in conjunction with criminal legal system contact, negatively affect some Black YSMM. This increased exposure was loosely associated with Black participants’ residing in neighborhoods with histories of being heavily policed. Stop-and-frisk adds to the burden of structural and systematic discrimination that is known to degrade health, well-being and sense of safety in this already vulnerable group [26, 27]. Together, results indicating the pervasive experience of stop-and-frisk among YSMM, particularly Black YSMM, highlight the need for reduction of this stressful and deleterious life event among YSMM. The results underscore the call for an alternative to policing, and a community response that addresses the economic, social, and heath care needs of the population.

While YSMM of all races and ethnicities face high rates of police contact, racial inequality persists within this vulnerable group. Black YSMM faced a significantly higher burden of repeat stop-and-frisk than their White counterparts. At baseline, nearly one-quarter of Black YSMM reported being stopped-and-frisked two or more times in the past year compared with one in five non-Black Hispanic/Latinx and approximately 10% of White participants. Further, the cumulative number of stop-and-frisk events in the past year per Black participants, was substantially higher than the stop-and-frisk rate among non-Black participants across follow-up. Race disparities appeared to increase over time as the experience became less normative overall. We also observed that the Black vs non-Black disparities were particularly pronounced in neighborhoods with overall lower levels of stop-and-frisk experience while in the highest stop neighborhoods non-Black Hispanic/Latinxs had a disproportionate rate of being stopped. Disproportionate rates of policing were observed among Black and Hispanic/Latinx participants who did not endorse drug use, which is a primary reason motivating police stops.

The inequitable policing of racial/ethnic minorities observed in this study is consistent with the vast scientific literature and current media reports highlighting the inequitable policing including aggressive policing of Black individuals and communities [13, 28, 29]. Although the threat of police violence faced by Black Americans has recently gained prominence in public discourse, a long list of police-related deaths over the past 40 years highlights the long history of police related violence that has persisted into the modern era. Moreover, incidents which lead to death represent only a small fraction of violent police encounters; nonlethal physical force is far more common, if less consistently documented, and has been associated with a variety of adverse health and social outcomes [30].

We observed levels of stop-and-frisk based on self-report of the experience were disproportionately high in this sample compared to the official NYPD stop-and-frisk counts. The substantial magnitude of difference in reporting by participants compared to official records is consistent with concerns that official reporting by the NYPD does not fully reflect the burden of police practices on local populations. Moreover, our findings are suggestive of a heavy burden on YSMM, as has been suggested in qualitative studies and advocacy reports [31, 32]. There is agreement that levels of stop-and-frisk as a practice has become less common since the Floyd v. City of New York decision in 2013. However, the contrast between general population rates and the reports of P18 participants provide empirical evidence that the experience is likely far more common than official reports suggest.

In addition to a lack of information about the nature of the stop-and-frisk experience, which makes comparison to the NYPD official case counts of stop-and-frisk as well as unerstanding of the severity of the policing exposure challenging, additional limitations of the study include selection and measurement errors. First, attrition over cohort follow-up may have impacted measured levels of stop-and-frisk and race differences in stop-and-frisk levels if those who did not return for a study visit had different stop-and-frisk experience as those who presented for study visits. In addition, due to relatively modest sample size of the cohort, we collapsed multiple racial/ethnic categories and further omitted some (e.g., Asians) who did not identify as one of the three racial/ethnic groups most prominent in discussions of American policing. Although not explicitly limitations, we had no non-sexual minority group against which to compare levels of stop-and-frisk and additionally our analyses of correlates of stop-and-frisk were by design unadjusted to describe the social, mental health, and behavioral health needs of those who come in contact with the police. Our findings highlight the need for future studies in larger samples to include questions of stop-and-frisk experience to enable comparisons of YSMM to non-sexual minority youth. New York City does not report stop-and-frisk rates by sexual minority status, making these comprisons somewhat limited. In addition, future studies should be powered to consider racial/ethnic differences in the burden of this exposure considering a heterogeneity of racial/ethnic groups compared YSMM and non-sexual minority individuals.

In conclusion, the current study highlights the substantial burden of stop-and-frisk police experience among YSMM in New York City, a group whose experience has remained understudied and underdocumented, and a population that continues to face a heigtened burden of health disaprities including but not limted to HIV, mental helath burden, substance use, and violence [33, 34]. We observed that while rates of stop-and-frisk have declined over time, consistent with official NYPD reporting of the experience, levels remain high and are disproportionately high among Black YSMM. Findings that the most socially vulnerable YSMM are those most likely to experience stop-and-frisk policing highlight the need for community based solutions among those at risk of contact with the police as well as the need for institutions that do not explicitly focus on public safety (e.g., primary care physicians, schools) to raise awareness of police contact as a threat to health and wellbeing [35]. Doing so will reduce a key driver of adverse health in YSMM and particularly those who possess intersectional identities as both sexual minority and racial/ethnic minority individuals.

Acknowledgments

We would like to acknowledge P18 participants for their participation in the Project 18 Cohort study and Natalia Irvine for her editorial contributions.