How likely is unmeasured confounding to explain meta-analysis-derived associations between alcohol, other substances, and mood-related conditions with HIV risk behaviors?

Table 2 Assessment of unmeasured confounding among meta-analyses assessed by Point Estimate sensitivity analysis only

Exposure	Outcome	Pooled RR (95% CI)¹	E-value² (95% CI LL)	Bias factor³	Bias-adjusted RR (95% CI)
Alcohol	Non-adherence	1.40 (1.30–1.50)	2.15 (1.93)	1.11	1.26 (1.17–1.35)
Tobacco	Non-adherence	1.28 (1.20–1.38)	1.89 (1.68)	1.11	1.16 (1.08–1.24)
Stimulants	Non-adherence	1.59 (1.44–1.75)	2.55 (2.23)	1.11	1.43 (1.29–1.58)
Depression	Non-adherence	1.52 (1.35–1.72)	2.42 (2.04)	1.01	1.51 (1.34–1.71)
Anxiety	Non-adherence	1.44 (1.24–1.68)	2.24 (1.78)	1.01	1.43 (1.23–1.67)
Pain	Non-adherence	1.20 (1.08–1.34)	1.70 (1.38)	1.01	1.20 (1.08–1.33)
Pain	Unprotected sex	1.32 (1.08–1.62)	1.97 (1.37)	1.01	1.31 (1.07–1.61)
Alcohol	Transactional sex	1.54 (1.4–1.68)	2.44 (2.16)	1.17	1.31 (1.20–1.44)
Stimulants	Transactional sex	1.73 (1.58–1.89)	2.85 (2.54)	1.17	1.48 (1.35–1.62)
Depression	Transactional sex	1.32 (1.22–1.43)	1.98 (1.74)	1.01	1.31 (1.21–1.42)
Anxiety	Transactional sex	1.22 (1.08–1.38)	1.73 (1.37)	1.01	1.21 (1.07–1.37)
Tobacco	Multiple sexual partners	1.53 (1.42–1.64)	2.43 (2.20)	1.17	1.31 (1.22–1.40)
Opioids	Multiple sexual partners	1.14 (0.93–1.41)	1.55 (1.00)	1.17	0.98 (0.79–1.21)
Anxiety	Multiple sexual partners	0.97 (0.81–1.16)	1.21 (1.00)	1.01	0.98 (0.82–1.17)
Pain⁴	Multiple sexual partners	1.30 (1.10–1.55)	1.93 (1.42)	1.01	1.29 (1.09–1.54)

¹ Converted from pooled ORs of Fig. 3
² E-value: quantifies how strongly an unmeasured confounder would need to be associated with the exposure (minimum risk ratio RR_XU) and outcome (minimum risk ratio RR_UY) to fully explain the observed association (RR_XY)
³ Bias factor: quantifies the maximum amount by which an unmeasured confounder could alter RR^T_XY
⁴ Single study

ISSN: 1471-2288