Table 2.
Compliance with 7 Key Epidemiological Principles Among Published Cluster-Randomized Trials (CRT) in Infection Control and Hospital Epidemiology (N=44)
| Epidemiological Principle | Characteristics | No. (%) |
|---|---|---|
| 1 | Type of design used | |
| CRT with crossover | 15 (34.1) | |
| Parallel CRT | 11 (25.0) | |
| Stratified CRT | 7 (15.9) | |
| CRT with stepped wedge | 4 (9.1) | |
| Matched CRT | 3 (6.8) | |
| CRT with crossover and multiple periods | 2 (4.5) | |
| Stratified CRT with crossover | 2 (4.5) | |
| Factorial CRT | 0 | |
| Fractional factorial CRT design | 0 | |
| Justified use of CRT | 22 (50.0) | |
| 2 | Sample size estimates | |
| Accounted for clustering when estimating sample size | 20/33 (60.6) | |
| Reported design effect (ICC or CV) | 15/33 (45.5) | |
| 3 | Consent | |
| Obtained waived consent | 15 (34.1) | |
| Did not report how they dealt with consent | 14 (31.8) | |
| Obtained consent from individuals | 8 (18.2) | |
| Obtained consent at the cluster level | 7 (15.9) | |
| 4 | Level of inference | |
| Individual level | 32 (72.7) | |
| Cluster level | 11 (25.0) | |
| Both individual and cluster level | 1 (2.3) | |
| 5 | Matching and/or stratification | |
| Employed matching or stratification at time of randomization | 17 (38.6) | |
| 6 | Bias and/or contamination | |
| Reported some effort to reduce bias/contamination | 34 (77.3) | |
| 7 | Analysis | |
| Accounted for clustering in the analysis | 29/33 (87.9) |