Table 3. Effect sizes from logistic regressions: Free and convenient distribution compared to other arms.
| Free and Convenient Versus Control | Free and Convenient Versus Convenient Only (role of price) | Free and Convenient Versus Free Only (role of convenience) | ||||
|---|---|---|---|---|---|---|
| Unadjusted | Adjusted | Unadjusted | Adjusted | Unadjusted | Adjusted | |
| Primary Outcome | ||||||
| Used ORS (primary outcome) | 0.21 | 0.19 | 0.12 | 0.12 | 0.03 | 0.02 |
| (0.13 to 0.28) | (0.13 to 0.26) | (0.05 to 0.19) | (0.06 to 0.18) | (−0.04 to 0.10) | (−0.04 to 0.08) | |
| Secondary Outcomes | ||||||
| Used ORS on same day as diarrhea began | 0.19 | 0.18 | 0.19 | 0.18 | 0.07 | 0.07 |
| (0.11 to 0.27) | (0.11 to 0.25) | (0.11 to 0.27) | (0.10 to 0.25) | (−0.01 to 0.16) | (−0.01 to 0.15) | |
| Used ORS + zinc | 0.33 | 0.31 | 0.18 | 0.17 | 0.03 | 0.03 |
| (0.22 to 0.44) | (0.21 to 0.41) | (0.09 to 0.28) | (0.08 to 0.26) | (−0.06 to 0.13) | (−0.06 to 0.12) | |
| Used antibiotics | −0.07 | −0.08 | −0.05 | −0.05 | 0.04 | 0.03 |
| (−0.14 to 0.01) | (−0.14 to −0.01) | (−0.12 to 0.03) | (−0.12 to 0.02) | (−0.03 to 0.12) | (−0.04 to 0.10) | |
| Ex-Post Outcome | ||||||
| Used zinc | 0.29 | 0.27 | 0.15 | 0.15 | 0.02 | 0.02 |
| (0.18–0.41) | (0.17 to 0.37) | (0.06–0.24) | (0.06 to 0.23) | (−0.08 to 0.12) | (−0.07 to 0.12) | |
95% CIs from logistic regression in parentheses. Differences estimated using recycled predictions from logistics regressions (see S1 Appendix Section S2.1). Standard errors clustered at the village level and estimated using the Delta method [28]. Adjusted models control for prespecified control variables including branch fixed-effects, caretaker characteristics (age, education, number of children), child characteristics (age, diarrhea frequency, blood in stool, concurrent fever), household characteristics (water source, latrine type, main source of income), and baseline village characteristics (percentage of households using respective treatment, percentage of households visited by CHW in past month, percentage of households aware of free ORS in village, percentage of households with ORS stored in their home). See S1 Appendix Table A for full regression results. Secondary outcomes were prespecified, and P values were adjusted for multiple hypotheses (see S1 Appendix Section S6). Ex-post outcome was not prespecified and was requested ex-post by an anonymous reviewer.
Abbreviations: CHW, community health worker; ORS, oral rehydration salts.