How does hard-to-reach status affect antiretroviral therapy adherence in the HIV-infected population? Results from a meta-analysis of observational studies

Table 2 Odds ratios describing the association between HTR status and optimal ART adherence, categorized by subgroups

Subgroup	Number of measures of correlation	OR	95% CI		Tests for heterogeneity		Tests for interaction
Subgroup	Number of measures of correlation	OR	95% CI		P Value (Q Statistic)	I² (%)	OR	95% CI		P value
Study design
Cohort	18	0.54	0.45	0.65	0.00	58.3	0.98	0.77	1.25	0.88
Cross-sectional	14	0.55	0.47	0.64	0.00	31.4	0.98	0.77	1.25	0.88
Adherence threshold/cut-off point (%)
100	13	0.56	0.46	0.67	0.00	52.1	–
≥ 95	12	0.52	0.42	0.66	0.00	54.1
< 95	7	0.55	0.43	0.71	0.00	47.9
Adherence measure
Self-reporting	22	0.56	0.49	0.64	0.00	42.2	–
Pharmacy refill	9	0.47	0.35	0.64	0.00	64.1
Self-reporting & pill count	1	0.72	0.53	0.97	0.03	–
Region/country
USA	16	0.58	0.50	0.68	0.00	38.5	–
Canada	9	0.44	0.34	0.58	0.00	51.7
Others	7	0.60	0.47	0.78	0.00	58.5
Observational period
≥ 6 months	11	0.44	0.35	0.56	0.00	44.3	0.72	0.55	0.94	0.02
< 6 months	21	0.61	0.53	0.69	0.00	41.5	0.72	0.55	0.94	0.02
Quality assessment results
High	18	0.57	0.48	0.67	0.00	53.2	1.1	0.85	1.42	0.48
Moderate	14	0.52	0.44	0.65	0.00	41.3	1.1	0.85	1.42	0.48

ISSN: 1471-2458