Reports

Signals of Distress: High Utilization of Criminal Legal and Urgent and Emergent Health Services in San Francisco

Map of San Francisco is overlaid with five icons, including a justice gavel, a heart with a beat going through it, a home, a person thinking, and a hand holding smaller silhouetted figures.

PRESS RELEASE: New Report Shows Patterns of Health and Legal System Utilization for Small Number of Extremely Vulnerable San Francisco Residents

People with multiple, complex health and housing needs frequently receive fragmented care because the providing systems operate independently. Typically, individuals who come into frequent contact with the emergency medical system (e.g., emergency departments; emergency medical services) also interact with other health services and public systems such as psychiatric facilities, substance use treatment centers, shelters, and jails. Cross-sector care coordination is limited, in part, because data systems are not linked across physical health, behavioral health (mental health and substance use), housing, and criminal legal systems.

To help San Francisco better serve this high need population, the California Policy Lab at UC Berkeley and the UCSF Benioff Homelessness and Housing Initiative worked with our partners in San Francisco’s public health and criminal legal systems to link together ten years of data from the physical health, behavioral health, housing, and criminal legal sectors. Using these linked data, we identify individuals with high utilization of the criminal legal system and the medical and behavioral health systems in a single year. High criminal legal utilization is defined as at least three jail bookings in a year, while high healthcare utilization is seven or more urgent/emergent healthcare contacts in a year.

To understand trends before and after a year of high utilization, we analyze two cohorts. The 2011 cohort includes 211 people with high utilization of both systems in fiscal year 2011, while the 2020 cohort includes 161 individuals with high utilization of both systems in fiscal year 2020.

Key Findings

1. Twenty-five percent of the individuals in the 2011 cohort of high, dual-system utilization continued their dual-system high frequency contact the following year. Overall, the 2011 cohort had a startling death rate: by the end of the ten-year period, at least one quarter of the cohort was deceased (26%).
Bar chart showing the composition of the FY10-11 cohort by year starting in fiscal year 2010-11. All individuals are dual high utilizers in the first year, but that decreases to 24% the second year and is at 3% the final year of the data, fiscal year 2019-20. The figure shows an extremely high death rate of 23% by the start of fiscal year 2019-20. A large portion do not show up in the data in many years, peaking with 36% in 2019-20.

2. Forty-two percent of individuals in the 2020 cohort of high, dual-system utilization were in San Francisco and received urgent/emergent healthcare services or were booked into jail 10 years prior to their year of high utilization.Bar chart showing the composition of the FY19-20 cohort by year starting in fiscal year 2010-11. High contact with both systems gradually increases from 4% in 2010-11 to 100% in 2019-20. The other categories stay relatively consistent before '19-20, with not being in the dataset decreasing rapidly and high contact with CCMS increasing to 25% from 3% before '19-20.

3. The share of individuals in the 2011 cohort who had frequent criminal justice system contact declined over time, however many continued to be arrested and booked into jail after FY 2011.

Line graph with one line showing year by year the percentage of FY 10-11 cohort individuals who are booked at least once in that year. The percentage declines dramatically, going from 100% in fiscal year 2010-11 to 19% in fiscal year 2019-20.

4. The most frequent services accessed by the 2020 cohort in the decade before their high utilization were acute services for physical health and behavioral health needs. Over time, the frequency of contact of routine medical visits and transitional health and behavioral health services remained steady, while contacts with acute services increased. Acute physical health contacts include emergency department visits and medical hospital stays, and acute behavioral health services include psychiatric emergency services visits, detoxification stays, and mental health hospitalizations. In particular, acute medical service contacts grew from fewer than 2 per year on average in FY 2011, to more than 12 per year in FY 2020. The growth in acute behavioral health contacts was driven by an increase in acute services for mental health needs leading up to FY 2020, while the average contacts for acute substance use services remained stable over time.

Line graph with 5 different lines showing the year to year average usage of different health services for the FY 19-20 cohort. All services start at similar levels, but acute physical health services fast outstrips the others, at the end getting to 3 times the next highest service.

Select Media Coverage

SF ExaminerThree ways to improve San Francisco’s fragmented homelessness system (10-18-22)
SF ChronicleWill a new lawsuit finally convince San Francisco its homeless sweeps are futile? (10-1-22)
SF ChronicleStudy: San Francisco’s fragmented city services are harming—and killing—the most vulnerable (9-27-22)