Table 2. Results of meta-regression from 231 surveys of the prevalence of wasting among children aged 6–59 months in crisis-affected areas within Ethiopia, 2000–2013.
| Category | Posterior OR (95% CrI) |
|---|---|
| Conflict events affecting survey areaa | |
| No | Ref. |
| Yes | 1.02 (0.82–1.26) |
| Drought episodes affecting survey areab | |
| None | Ref. |
| Mild | 1.04 (0.91–1.19) |
| Moderate | 1.34 (1.05–1.72) |
| Severe | 0.96 (0.68–1.35) |
| Survey periodc | 0.96 (0.94–0.98) |
| Administrative region of survey | |
| Afar | 1.51 (1.15–2.00) |
| Amhara | 1.29 (1.08–1.54) |
| Benishangul-Gumz | 1.61 (0.80–3.26) |
| Dire Dawa | 1.53 (0.75–3.13) |
| Oromia | Ref. |
| Somali | 2.21 (1.74–2.81) |
| Southern Nations, Nationalities, and People’s | 0.82 (0.69–0.97) |
OR: odds ratio; CrI: credible interval; Ref.: reference group.
a Conflict events were occurrence of a conflict in a given location within six months of the survey starting date, irrespective of magnitude or impact.
b Drought episodes in a given location were based on the three-month standardized precipitation evapotranspiration index (SPEI): no drought (SPEI > 0), mild drought (−1 < SPEI ≤ 0), moderate drought (−1.5 < SPEI ≤ −1) and severe to extreme drought (SPEI ≤ −1.5).
c For a one-year increase in the survey period.
Notes: Data were estimated from a pooled sample of 175 607 children. See Fig. 2 for the number of surveys, samples sizes, numbers of children affected and prevalence of wasting by category. Wasting was defined as weight-for-height z-scores below −2.