Skip to main content
. Author manuscript; available in PMC: 2020 Jul 9.
Published in final edited form as: N Engl J Med. 2020 Jan 9;382(2):152–162. doi: 10.1056/NEJMsa1906848

Table 1:

Summary Statistics

Overall Treatment Control
Study Population (N=800):
 Index Admission Match Rate (%) 97.8 98.5 97.0

Analysis Sample (N=782):
 Observations 782 393 389
 Male (%) 50.0 52.4 47.6
Age at index admission (%)
 <44 17.1 16.0 18.3
 45–64 55.4 55.0 55.8
 >65 27.5 29.0 26.0
Race (%)
 African American Non-Hispanic 54.9 57.8 51.9
 Hispanic 29.5 26.7 32.4
 White Non-Hispanic 15.1 14.8 15.4
 Asian / Multiracial / Other 0.5 0.8 0.3
Number of inpatient admissions prior to the index admission
 0–6 months prior 1.75 1.72 1.78
 7–12 months prior 0.74 0.74 0.75
Primary payer (%)
 Medicaid 44.6 43.0 46.3
 Medicare 48.2 47.6 48.8
 Other 7.0 9.2 4.9
Employment status (%)
 Currently Employed 5.5 4.8 6.2
 No response 0.5 0.3 0.8
Mental health diagnoses at index admission (%)
 Depression 30.2 32.3 28.0
 Substance Abuse 44.0 41.2 46.8

Notes: “Study Population” and “Analysis Sample” are defined in Figure 1. Within “Analysis Sample,” data on gender, age, number of admissions prior to index, and mental health diagnoses are from the hospital discharge data. Race, primary payer, and employment status come from the baseline survey. The analysis sample excluded 18 participants with missing outcome data because they could not be matched to the discharge record for their index admission.