Table 3. Association between living in a state not contributing to the annual ADAP budget and ART initiation and virologic suppression, U.S. NA-ACCORD, 2001–2009.

	Outcome: 6-month ART initiation		Outcome: 1-year virologic suppression
	HR	95% CI	HR	95% CI
Overall
No contribution (vs. any contribution)
Crude (N = 8,874)	0.56	0.49–0.63	0.75	0.67–0.83
Regression-adjusted (N = 8,874)	0.80	0.69–0.93	1.02	0.88–1.18
Propensity score-matched (N = 1,082) *	0.73	0.60–0.88	1.13	0.93–1.36
Dose-response effect (Ptrend) (N = 8,874)		<0.001		0.25
No contribution	0.75	0.63–0.88	1.06	0.91–1.24
Contribution <20%	0.90	0.82–0.99	1.07	0.97–1.17
Contribution >20%	1.00	Ref.	1.00	Ref.
Injection drug users only (N = 1,824)
No contribution (vs. any contribution)
Crude	0.40	0.31–0.51	0.78	0.64–0.96
Regression-adjusted	0.67	0.47–0.95	1.14	0.82–1.59
Dose-response effect (Ptrend)		0.005		0.29
No contribution	0.58	0.40–0.86	1.21	0.83–1.74
Contribution <20%	0.81	0.63–1.04	1.10	0.85–1.42
Contribution >20%	1.00	Ref.	1.00	Ref.

ART = antiretroviral therapy, CI = confidence interval, HR = hazard ratio.

All analyses use Cox proportional hazards regression.

^*Hazard ratios obtained after 1∶3 matching (with replacement) 683 “exposed” to 399 “unexposed” individuals based on propensity of living in a state contributing to the ADAP budget.

Both regression-adjusted and propensity-score matched analyses account for the following variables: age; sex; race/ethnicity; transmission risk; CD4+ count and viral load at eligibility; history of alcohol abuse, substance abuse, and mental disorders; year of eligibility; type of cohort; clinic-specific mechanisms to help obtain ART; state-level population density, % population of black race, % population below poverty line, median household income, and per capita Medicaid spending on HIV.