Skip to main content

Table 3 Multivariable Longitudinal Regression Results of the Cumulative Number Nights Homeless and Medical Service Utilization (n= 1045 observations).

From: Episodic homelessness and health care utilization in a prospective cohort of HIV-infected persons with alcohol problems

 

IRR†(95% CI)

Number nights homeless

Ambulatory

Emergency Room

Hospitalization

0

1

1

1

1–7

1.24 (0.90, 1.71)

1.31 (0.80, 2.15)

1.07 (0.64, 1.80)

8–30

1.15 (0.84, 1.57)

1.49 (1.02, 2.19)*

1.83 (1.01, 3.32)*

31–120

1.27 (0.97, 1.67)

2.17 (1.54, 3.07) *

1.85 (1.17, 2.94)**

121–180

0.79 (0.61, 1.03)

2.65 (1.94, 3.61) ***

2.88 (1.95, 4.25) ***

  1. † IRR: incidence rate ratio of medical service utilization for 6-month observations periods Separate multivariate regression models were used for each utilization outcome. All models include age, gender, race/ethnicity (2 df), CD4 cell, HIV RNA viral load log10, antiretroviral medication use, adherence intervention participation, presence of depressive symptoms (Center for Epidemiologic Studies Depression Scale), and addiction severity (Addiction Severity Index, alcohol and drug composite scores).
  2. ‡ Number nights homeless was defined by the survey question, "In the last six months, how many nights have you spent in an overnight shelter, on the street, without shelter?" Homelessness was categorized based upon the median nights homeless for the sample (i.e., 30) and interquartile range (i.e., 7, 120).