Table 2.
Analysis methods stratified by unit of analysis
| Cluster-level analysis | Individual-level analysis | |
|---|---|---|
| Methods of analysis | 6 (100) | 117 (100) |
| Generalized estimating equations | - | 27 (23.1) |
| Mixed effects models | - | 55 (47.0) |
| Repeated measures analysis of variance | - | 5 (4.3) |
| Generalized linear model with sandwich variance | - | 16 (13.7) |
| Chi square accounting for clustering | - | 1 (0.8) |
| Survival analysis accounting for clustering | - | 1 (0.8) |
| Other methods ignoring clustering a | - | 12 (10.3) |
| Weighted regression b | 1 (16.7) | - |
| Other methods without weighting a | 5 (83.3) | - |
| Methods of analysis when non-adherence was addressed | 1 (100) | 18 (100) |
| Generalized estimating equations | - | 4 (22.2) |
| Mixed effects models | - | 9 (50.0) |
| Generalized linear model with sandwich variance | - | 4 (22.2) |
| Poisson regression ignoring clustering c | - | 1 (5.6) |
| Unweighted t-test d | 1 (100) | - |
The numbers in brackets are the column percentages.
a
Generalized linear model, analysis of variance, analysis of covariance, T-test, Mann-Whitney U test, Chi square test.
b
Number of events (cluster size) used as weights (Buttha et al [36]). The use of weights is applicable when cluster-level summaries analysis is performed while accounting for clustering may be required for individual-level analysis.
c
t-test with multiple testing adjustment but ignoring clustering was applied to perform a per protocol analysis at individual-level (Neuzil et al. [23]).
d
Per protocol analysis with unweighted t-test comparing rates at cluster level (Tagbor et al. [28]).