Figure 1
An educational poster used by program facilities. Text is written in Chichewa and describes how to use P&G water purification packets, buckets, and sieves to purify water.
Figure 2
Geographical depiction of Safe Water Program facilities in Lilongwe, Malawi. Facility locations are depicted as points; light blue circles indicate non‑program facilities (n = 29); dark blue triangles indicate program facilities (n = 33). The dark gray line indicates the Malawi‑Mozambique border.
Figure 3
Program distribution and enrollment characteristics from January 2018 to December 2022. Panel (A) shows monthly kit distribution counts at program facilities, stratified by participant status. Months where no data were reported are due to commodity stock outs or fuel shortages impairing distribution. Non‑program facilities that distributed materials in 2018 in response to cholera and flooding are not represented on the plot. Panel (B) shows counts of all participants who participated in the program; data were unavailable for 2018.
Table 1
Characteristics of Safe Water Program facilities and non‑program facilities.
| VARIABLE | PROGRAM FACILITIES | NON‑PROGRAM FACILITIES | P‑VALUE | |||
|---|---|---|---|---|---|---|
| N | % | N | % | |||
| Total number | 33 | 29 | ||||
| Facility type* | Health center | 24 | 72.7 | 24 | 82.8 | 0.2 |
| Hospital | 6 | 18.2 | 5 | 17.2 | ||
| Urbanicity | Rural | 28 | 84.8 | 14 | 48.3 | <0.001 |
| Urban | 2 | 6.1 | 15 | 51.7 | ||
| Ownership | Government | 25 | 75.8 | 29 | 100 | 0.02 |
| Private | 5 | 15.2 | 0 | 0 | ||
| N | % | N | % | |||
| Water access | 25 | 75.8 | – | – | ||
| Electricity access | 25 | 75.8 | – | – | ||
| MEAN | SE | MEAN | SE | |||
| Catchment population, children under 5 years† | 15,936 | 20,932 | 10,029 | 11,825 | 0.2 | |
| Facility staff | Doctors | 0.7 | 1.8 | – | – | |
| Clinical officers | 3.1 | 4.8 | – | – | ||
| Medical assistants | 2.8 | 1.7 | – | – | ||
| Nurses | 14.6 | 18.1 | – | – | ||
| Community midwives | 3.9 | 10.3 | – | – | ||
[i] Note:missing data ‑ three program facilities are missing data on facility type, urbanicity, ownership, and catchment population. Eight program facilities are missing data on water access, electricity access, and facility staff.
[ii] *Hospitals are larger than health centers and have inpatient facilities. Health centers are primarily outpatient facilities although most have inpatient maternity services.
[iii] †The catchment population of children under 5 years for each facility was averaged over the years 2018–2022 if facilities reported populations yearly.
Figure 4
Annual case incidence of (A) non‑bloody diarrhea and (B) bloody diarrhea among children under 5 years from 2018 to 2022, stratified by program and other non‑program facilities (n = 300). Panel (C) represents the annual coverage of the Safe Water Program, the proportion of women benefitting from the program out of all women enrolled in PVTH at program facilities; data were unavailable for 2018. Ten observations fall outside plot area A with non‑bloody diarrhea incidence in children > 0.4 and 3 observations fall outside plot area B with bloody diarrhea incidence in children > 80 per 1,000. Thirty‑one observations (facilities in a given year) are missing non‑bloody diarrhea data, 32 observations are missing bloody diarrhea data, and 50 observations are missing coverage data.
Figure 5
Antiretroviral therapy (ART) retention plotted quarterly from 2018 to 2022. Results are stratified by program facilities (n = 33) and non‑program facilities (n = 29). Lines represent the mean quarterly ART retention values for program facilities, non‑program facilities, and a national estimate (n = 874).
Table 2
Bivariate and multivariate associations between facility participation in the Safe Water Program and diarrheal incidence among children under 5 years, 2018–2022, and antiretroviral therapy (ART) retention from 2018 to 2022.
| OUTCOME | MODEL | ESTIMATE (INCIDENCE RATE RATIO) | 95% CONFIDENCE INTERVAL | P‑VALUE |
|---|---|---|---|---|
| Non‑bloody diarrhea incidence | Unadjusted | 0.99 | 0.54–1.83 | 1.0 |
| Adjusted* | 1.20 | 0.62–2.34 | 0.6 | |
| Bloody diarrhea incidence | Unadjusted | 0.61 | 0.32–1.16 | 0.1 |
| Adjusted* | 0.81 | 0.41–1.63 | 0.6 | |
| ART retention | Unadjusted | 1.06 | 1.01–1.10 | 0.008 |
| Adjusted* | 1.05 | 0.99–1.12 | 0.09 |
[i] *Adjusted for urbanicity and facility type.