Table 3 Assumptions for sample size required to measure the primary outcome “incidence of births within 8 months of the end of the intervention period”
Parameter | Assumed level | Comment |
|---|---|---|
Incidence of births in control group | 0.06 | Eight months after the end of the intervention period the girls will be on average 17.1 years and we assume that 15% of them will have given birth. This corresponds to an average incidence rate of (15%-3%)/2 = 6% over this 2-year period. |
Effectiveness of combined intervention vs control | −40% | i.e. incidence in combined intervention arm assumed to be 0.036 |
Cluster size | 28 | See assumptions for the outcome “incidence of births before girls’ 18th birthday” |
Person years per cluster | 56 | If 28 participants are followed up for 2 years, there will be 56 person years |
k | 0.20 | The ICC was 0.00737 for “ever pregnant” after the intervention period in the cash transfer trial in Malawi. This corresponds to k = 0.20 when the total proportion who have given birth by this time is 0.15 |
Z1 (acceptable alpha error level) | 1.96 | |
Power for comparison of combined intervention vs control | 90% | We need 36 combined clusters vs 36 control clusters to have 90% power detect the assumed difference. The PASS function for proportions indicates that with 63 combined clusters, we will have > 90% power with 30 control clusters. |