Like food, water, rest, and shelter, safety is one of our basic needs in life. Safety is the condition being safe from undergoing or causing hurt, injury, or loss. Safety is important in all environments — home, school, work, and public spaces. Low perceptions of safety can result from witnessing, experiencing, or fearing numerous types of events, including: violence, property crime or other forms of social disorder, emotional abuse, threats to financial security, or other threats to an individual’s sense of control over their life. In this analysis, we measure safety primarily by tracking the prevalence of reported crimes or violent events that necessitate medical care.
Crime and violence are rarely caused by a single risk factor, but rather by the presence of multiple risk factors and the absence of protective (or resiliency) factors. Risk factors for crime and violence include: poverty and economic disparity, illiteracy and school failure, alcohol and other drugs, firearms, negative family dynamics, mental illness, incarceration, community deterioration, discrimination and oppression, power and control, exposure to media violence, experiencing and witnessing violence, and gender socialization.[1] Safety can be promoted through environmental interventions, including ensuring access to quality housing, reliable transportation to services and jobs, and opportunities for positive recreation (e.g. sports and libraries), while decreasing access to harmful influences such as tobacco, firearms, and alcohol. [1, 2]
Injury, death, and emotional trauma are adverse and often direct health outcomes resulting from physical assaults, homicides, and sexual assaults. In many marginalized communities, homicides account for the largest number of avoidable years of life lost, because of their disproportionate occurrence among young men. Witnessing and experiencing violence disrupts early brain development and causes longer term behavioral, physical, and emotional problems in children and youth, including perpetrating or being a victim of violence, depression, suicide ideation and attempts, smoking, obesity, high-risk sexual behaviors, school absenteeism, unintended pregnancy, eating disorders, and alcohol and drug abuse. [3, 4, 5, 6]
Community violence decreases the real and perceived safety of a neighborhood, inhibiting social interactions and adversely affecting social cohesion. [7, 8] This can create a vicious circle, as social cohesion can be a valuable tool in decreasing crime. [9] Studies have found a negative relationship between neighborhood residents’ levels of mutual trust/willingness to take action and levels of violent crime.[8, 10] In addition, the level of safety perceived by residents of a neighborhood may differ from objective measures of the level of safety (e.g. crime rates), and may be influenced by the residents’ feelings of integration into the social fabric of the neighborhood, or by other aspects of social cohesion. [11, 12]
Residents’ worries about safety in their neighborhoods can be a cause of chronic stress and can also be a disincentive to engage in physical activity outdoors, particularly among women, children, and older persons. [10, 13, 14, 15, 16, 17] A study in Baltimore, Maryland ranked 65 neighborhoods on the Neighborhood Psychosocial Hazards Scale, a combined measure of social disorganization, public safety, physical disorder, and economic deprivation. The researchers then linked the neighborhood measures with health data for a sample of residents. Regardless of age, gender, race, education, smoking or medical history (e.g. hypertension, diabetes), residents were more likely to have had a heart attack if they lived in the most hazardous neighborhoods compared to the least hazardous neighborhoods. [18] In a separate study using the same data, researchers found that living in the most hazardous neighborhoods increased the odds of being obese compared to living in the least hazardous neighborhoods of Baltimore. [18] More importantly, this relationship could not be explained away by differences in resident demographics, wealth, education, alcohol consumption, tobacco use, diet, or physical activity.
Crime rates and perceptions of safety: Between 2013 and 2017, all crime rates increased, with the exception of drug crime (Figure 1). Increases primarily occurred between 2013-2015, with rates plateauing or declining between 2015 and 2017. Among all crime types, property crime has the highest rate and has experienced the most significant increase. By neighborhood, South of Market, Tenderloin, Financial District, Mission, and Bayview have the highest violent crime rates (Figure 2). Property crime rates are mostly concentrated in neighborhoods with heavy tourist activity, including South of Market (this includes Westfield shopping center), Financial District/South Beach, North Beach, and Japantown. Drug crime rates are heavily concentrated in Tenderloin and South of Market. All crime rate types are higher in parts of the city defined as Areas of Vulnerability compared to the rest of the city (Figure 3).
Perceptions of safety are measured through the San Francisco City Survey. Between 2001 and 2017, there do not appear to have been any significant changes in the percent of residents that feel safe walking alone in their neighborhood during the day or at night (Figure 4). The percentage of residents that feel safe at night is consistently lower than the percent during the day. Perceptions of safety are lower in the eastern neighborhoods compared to the western side of the city (Figure 5). The zip codes with the lowest perceptions of safety during the day or night include 94102, 94103, 94124, and 94134 – roughly covering Tenderloin, South of Market, Bayview, Visitacion Valley, and Portola neighborhoods. Asian, Black, and Latino/a residents have significantly lower perceptions of safety during the day and night compared to White residents (Figure 6).
Youth violence and crime:
Child maltreatment – Over the past decade, substantiated cases of child maltreatment has decreased from 9.8 to 4 incidents per 1,000 children citywide (Figure 7). Despite this citywide decrease, a large disparity in the rate of substantiated child maltreatment cases for Black/African American children continues to exist with a rate 17 times that in Asian/Pacific Islander children or White children. In fact, the disparity in the rates has increased in recent years as the rates for White and API children decreased by over 55% while rates decreased by only 38% for Black children. The rate for Latino/a children has remained about 4 times as high as the rate for White children between 2007-2017.
Bullying: Nearly 30% of all middle school students reported being bullied on school property, while 15% report being bullied electronically (Figure 8). Those rates drop in high school to 13% and 11% respectively. By gender, significantly more middle school girls report being electronically bullied compared to boys (18% vs. 12%). By ethnicity, a greater percentage of White middle students have experienced bullying on school property compared to Chinese students. Comparisons between other ethnic groups were not significant.
In middle school, a greater percentage of students that identify as bisexual report being bullied on school property compared to heterosexual identified students (53% vs. 28%). In high school, a higher percent of all students that do notidentify as heterosexual experience bullying, both on school property and electronically (at school: bi-20%, GL-29%, unsure-21%, het-12%) (electronic: bi-19%, GL-29%, unsure-17%, het-10%).
Youth dating violence: Between survey periods 2009-2011 and 2015-2017, the percent of middle school youth reporting that they had experienced physical violence from someone they were dating dropped from 7% to 3% (Figure 9). There also appears to have been a drop among high school youth, but this is not significant. By gender, more male middle school students reported experiencing dating violence from 2009-2013 compared to female students (8% vs. 6%). Gay, lesbian, and bisexual identified middle and high school students also experienced at least twice the rate of dating violence than their heterosexual peers – around 15-20% in middle school 10-30% in high school. Chinese students have the lowest rate of dating violence at around 1-3% in middle school and 4-5% in high school – significantly lower than the rate for Latino/a and Black students. Comparisons between other ethnic groups were not significant.
Youth Juvenile Hall bookings: Between 2006 and 2015 there was a 71% drop in the number of juvenile hall bookings in San Francisco (Figure 10). While this drop is impressive, there are ethnic and gender disparities in the youth population that gets booked. Figure 11 demonstrates that Black youth make up over 57% of the youth booked even though they make up only 6% of the youth population. Similarly, Samoan youth make up 3% of the bookings, but only account for less than 1% of the youth population. Boys are similarly over represented – 76% percent of bookings in 2017. The majority of youth booked at Juvenile Hall reside in the southeastern part of the city, the Mission, South of Market, or Western Addition neighborhoods (Figure 12). Zip code 94124, which roughly covers the Bayview neighborhood, was home to nearly 22% of all of the youth booked at Juvenile Hall in 2017 (Figure 12).
Assault: Between 2005-2016, there appears to have been a slight increase in the age-adjusted rate of emergency room visits for assault, from 40.3 during 2005-2009 to 47.5 in 2012-2016 (Figure 13). This increase is more pronounced for adults alone, while youth experienced a decrease in their ER visit rate for assault. The visit rate for males is about twice the rate for females. By ethnicity, Black San Franciscan’s have by far the highest visit rate for assault – 255 per 10,000 compared to 13 for Asian residents and 36 for White residents. Pacific Islanders have the second highest rate at 131. The disparity between males and females is also lower for Black and Pacific Islander residents. While the visit rates for Asian, Latino, and White males are about twice as high as the rates for females, the rate for Black men is only 1.29 times higher and for Pacific Islander the rates are nearly the same.
When the visit rate is calculated for specific age groups, transitional aged youth (18-24 years) have the highest rate among both males and females (Figure 14). When examined by race, 25-34 year-old Black adults have the highest rate for their ethnicity (447), while 45-54 year old Pacific Islanders have the highest rate within their ethnic group (270). The zip codes with the highest rates of residents being admitted to the ER for assault are 94102 (Tenderloin), 94103 (SOMA), and 94124 (Bayview) (Figure 15). When examined just for youth, 94124 have by far the highest rate.
In 2014 SFUSD completed a Memorandum of Understanding limiting police presence on school campuses {35} and the board of education adopted a resolution to end suspensions for “willful defiance” which accounted for more than 80% of suspension of Black and Latinx students. [36].
SFPD San Francisco Police Department. https://data.sfgov.org/Public-Safety/Police-Department-Incident-Reports-Historical-2003/tmnf-yvry
SFC San Francisco Controller’s Office, “San Francisco City Survey,” http://sfcitysurvey.weebly.com/
UCB University of California at Berkeley, “California Child Welfare Indicators Project.” http://cssr.berkeley.edu/ucb_childwelfare/allegations.aspx
YRBS “Youth Risk Behavior Surveillance survey,” http://www.cdc.gov/healthyyouth/data/yrbs/index.htm
SFJPD San Francisco Juvenile Probation Department, “2017 Statistical Report,” https://sfgov.org/juvprobation/sites/default/files/2017AnnualReport_Statistics.pdf
OSHPD Office of Statewide Health Planning and Development. http://www.oshpd.ca.gov/
Violent crime and drug crime: Crime rates are calculated based only on crimes that are reported to the San Francisco Police Department. Violent crime includes 1) forcible sexual offenses, 2) robbery, and 3) assault. Homicide data is excluded, because it was not publicly available. Property crimes included 4) burglary, 5) larceny/theft, 6) vehicle theft, and 7) arson. Drug crimes include incidents coded as 8) drug/narcotic. All other crimes fall in the “other” category. Because crime incidents may include multiple crime categories, incidents were coded with the most severe crime category listed (1-8 listed previously).
Perceived safety: San Francisco City Survey respondents were asked to categorize their level of safety as very safe, safe, neither safe nor unsafe, unsafe, or very unsafe when walking alone in their neighborhood during the day or night. In 2015, the survey methodology changed from mail to phone in order to reach a more representative sample of San Francisco residents. This included changes to ethnic groupings – in 2015 Pacific Islanders (previously included in Asian) and Middle Eastern became distinct groups. Thus, prior to 2015, Asian should be interpreted as Asian/Pacific Islander. Because of these methodological changes, data before and after 2015 should be compared with caution.
Substantiated child abuse: Rates of substantiated child maltreatment includes cases of physical, sexual, and emotional abuse, as well as child neglect, exploitation, caretaker absence/incapacity, at-risk siblings, and allegations of substantial risk. Rates were calculated per 1,000 children using CA Department of Finance population estimates.
Bullying: Students were asked the following questions: 1) During the past 12 months, have you ever been bullied on school property? 2) During the past 12 months, have you ever been electronically bullied?
Dating violence: Students were asked the following question: during the past 12 months, how many times did someone you were dating or going out with physically hurt you on purpose? The denominator for dating violence rates is the number of youth that dated someone in the past 12 months.
Juvenile Hall Bookings: Juvenile hall bookings indicate that a youth (<18 years) was detained for a criminal offense. Both duplicated and unduplicated juvenile hall bookings are presented. Duplicated bookings count each booking instance, even if it is for the same youth. Unduplicated counts each youth only once, even if they are booked multiple times.
Emergency room visits due to assault: Emergency room visits among San Francisco Residents due to assault were identified with the following list of primary injury code (E-Codes): E960, E961, E962, E963, E964, E965, E966, E967, E968.
Hospitalizations and Emergency Room Visits:
Hospitalization and ER rates measure the number of admissions or visits, not the number of residents who are hospitalized. Admissions records may include multiple admissions by the same person.
In October 2015, the diagnosis coding standard for Hospitalizations and Emergency Room visits was changed from ICD-9 to ICD-10. Caution should be used in comparing data using the two different standards.
ICD-9 and ICD-10 codes for assault were obtained from the CDPH Safe and Active Communities Branch [19]. Records with ICD-9 code E967 or ICD-10 codes Y060, Y061, or Y07 indicated domestic abuse.
Estimates for assault as the primary cause were obtained by searching the primary diagnosis field only while estimate for assault as the primary, co-morbid, or coexisting cause was obtained by searching all available diagnosis fields. Similarly, A primary diagnosis of domestic abuse was defined as an assiciated code in the primary diagnosis field or, where assault was the primary diagnosis, in any diagnosis field. A domestic abuse code in any diagnosis filed, regardless of assault as the primary cause, indicated domestic abuse as primary, co-morbid or co-existing.
Population estimates for rates:
Standard Population for age adjustment:
[1] Prevention Institute. A lifetime commitment to violence prevention: The alameda county blueprint. Technical report, 2005.
[2] JD Bell, J Bell, R Colmenar, R Flournoy, M McGehee, V Rubin, M Thompson, J Thompson, and VB Vasquez. Reducing health disparities through a focus on communities. Oakland, CA: PolicyLink, 2002.
[3] A M Perez-Smith, K E Albus, and M D Weist. Exposure to violence and neighborhood affiliation among inner-city youth. Journal of clinical child psychology, 30:464–472, December 2001.
[4] Emily J Ozer and Kristen L McDonald. Exposure to violence and mental health among chinese american urban adolescents. The Journal of adolescent health : official publication of the Society for Adolescent Medicine, 39:73–79, July 2006.
[5] Diann M Ackard and Dianne Neumark-Sztainer. Date violence and date rape among adolescents: associations with disordered eating behaviors and psychological health. Child abuse & neglect, 26:455–473, May 2002.
[6] Donna E Howard and Min Qi Wang. Psychosocial correlates of u.s. adolescents who report a history of forced sexual intercourse. The Journal of adolescent health : official publication of the Society for Adolescent Medicine, 36:372–379, May 2005.
[7] Mindy Thompson Fullilove, Veronique Heon, Walkiria Jimenez, Caroline Parsons, Lesley L Green, and Robert E Fullilove. Injury and anomie: effects of violence on an inner-city community. American Journal of Public Health, 88(6):924–927, 1998.
[8] R J Sampson, S W Raudenbush, and F Earls. Neighborhoods and violent crime: a multilevel study of collective efficacy. Science (New York, N.Y.), 277:918–924, August 1997.
[9] Putnam R. Bowling Along: The Collapse and Revival of the American Community. New York,. NY: Simon & Schuster, 2000.
[10] B P Kennedy, I Kawachi, D Prothrow-Stith, K Lochner, and V Gupta. Social capital, income inequality, and firearm violent crime. Social science & medicine (1982), 47:7–17, July 1998.
[11] Robert J Sampson and Stephen W Raudenbush. Neighborhood stigma and the perception of disorder. Focus, 24(1):7–11, 2005.
[12] Pamela Wilcox Rountree and Kenneth C Land. Perceived risk versus fear of crime: Empirical evidence of conceptually distinct reactions in survey data. Social forces, pages 1353–1376, 1996.
[13] Andrea Altschuler, Carol P Somkin, and Nancy E Adler. Local services and amenities, neighborhood social capital, and health. Social science & medicine, 59(6):1219–1229, 2004.
[14] Mahasin S Mujahid, Ana V Diez Roux, Jeffrey D Morenoff, Trivellore E Raghunathan, Richard S Cooper, Hanyu Ni, and Steven Shea. Neighborhood characteristics and hypertension. Epidemiology, 19(4):590–598, 2008.
[15] Sarah Foster and Billie Giles-Corti. The built environment, neighborhood crime and constrained physical activity: an exploration of inconsistent findings. Preventive medicine, 47(3):241–251, 2008.
[16] Kristina Sundquist, Holger Theobald, Min Yang, Xinjun Li, Sven-Erik Johansson, and Jan Sundquist. Neighborhood violent crime and unemployment increase the risk of coronary heart disease: a multilevel study in an urban setting. Social science & medicine, 62(8):2061–2071, 2006.
[17] Jacqueline Kerr, Dori Rosenberg, James F Sallis, Brian E Saelens, Lawrence D Frank, Terry L Conway, et al. Active commuting to school: associations with environment and parental concerns. Medicine and science in sports and exercise, 38(4):787, 2006.
[18] Toms Augustin, Thomas A Glass, Bryan D James, and Brian S Schwartz. Neighborhood psychosocial hazards and cardiovascular disease: the baltimore memory study. American Journal of Public Health, 98(9):1664–1670, 2008.
[19] California Department of Public Health, Safe and Active Communities Branch, Cause of Injury Definition Codes
2017. https://www.cdph.ca.gov/Programs/CCDPHP/DCDIC/SACB/Pages/EpiCenter/OverviewofICDE9and10codes.aspx
