Table 3.

Relationship between calendar period, exposure to protease inhibitor (PI) and nonnucleoside reverse-transcriptase inhibitor (NNRTI) therapy, and the incidence of myocardial infarction, before and after adjustment for changes in risk factors for cardiovascular disease.

	Adjustment
	Baseline risk factors only		Changes in non-lipid risk factors		Changes in non-lipid risk factors and lipid levels
Variable	Relative rate (95% CI)	P	Relative rate (95% CI)	P	Relative rate (95%) CI)	P
Relationship with combination antiretroviral therapy
Cumulative exposure to PIs (per year)	1.16 (1.11–1.21)	<.001	1.15 (1.11–1.20)	<.001	1.10 (1.05–1.16)	<.001
Cumulative exposure to NNRTIs (per year)	1.05 (0.99–1.11)	.09	1.04 (0.98–1.10)	.19	1.03 (0.96–1.10)	.39
Relationship with calendar period
1999–2000	1		1		1
2001	1.20 (0.84–1.71)	.31	1.15 (0.81–1.64)	.45	1.60 (0.97–2.65)	.06
2002	0.84 (0.58–1.21)	.34	0.75 (0.51–1.08)	.12	1.11 (0.66–1.86)	.69
2003	0.90 (0.62–1.29)	.56	0.73 (0.50–1.05)	.09	1.06 (0.63–1.77)	.82
2004	0.83 (0.57–1.20)	.32	0.64 (0.44–0.94)	.02	1.02 (0.61–1.71)	.95
2005–2006	0.50 (0.33–0.76)	.001	0.36 (0.24–0.56)	<.001	0.63 (0.36–1.09)	.10

NOTE. Cumulative exposure to PIs and NNRTIs were both treated as time-updated covariates in all analyses. Fixed risk factors were male sex, HIV exposure group, ethnicity, and cohort. Time-updated non-lipid covariates were age, body mass index (calculated as the weight in kilograms divided by the square of the height in meters), family history of cardiovascular disease, smoking status, and a previous cardiovascular event. Time-updated lipid covariates were cholesterol, high-density lipoprotein cholesterol, and triglyceride levels.