Predictors of HIV testing rates among eligible patients: meta-regression of results from studies identified
| Covariate | N studies | OR (95% CI) | p Value |
|---|---|---|---|
| Patient group | |||
| Patients presenting with indicator disease conditions | 10 | 0.8 (0.2 to 2.6) | 0.67 |
| Persons attending screening settings | 20 | 1 (ref) | |
| Location of study | |||
| London | 14 | 1 (ref) | |
| Non-London | 16 | 0.5 (0.2 to 1.6) | 0.26 |
| Type of HIV test | |||
| Laboratory | 7 | 1 (ref) | |
| Point-of-care | 6 | 0.8 (0.2 to 4) | 0.75 |
| Service model* | |||
| Usual practice | 18 | 1 (ref) | |
| Added staff training | 9 | 1.3 (0.4 to 4.8) | 0.62 |
| GUM specialist testing | 3 | 3.5 (0.5 to 24) | 0.2 |
| Testing strategy | |||
| Opt-out | 12 | 1 (ref) | |
| Opt-in | 9 | 1 (0.3 to 3.2) | 0.99 |
| Study type | |||
| Retrospective | 14 | 1 (ref) | |
| Prospective | 16 | 2.6 (0.9 to 7.7) | 0.08 |
*With only three studies in one of the categories, this result should be interpreted with caution due to lack of power.
GUM, genitourinary medicine.