Neurological impairment and disability in children in rural Kenya.

The surveys used parent recall for pregnancy, birth, and neonatal history, so these risk factors may be inaccurate. Some impairment groups, especially visual impairment, had very few cases, so differences in those domains are hard to detect. Hospital admission data may miss births and deaths outside hospital, so neonatal risk trends may be undercounted. The surveys compared different children at different times, not the same individuals, so changes cannot be attributed to one cause. Attrition between screening and assessment was higher in the second survey.

Funded by the Wellcome Trust. No other conflicts of interest are stated in the supplied text.

Do not read the decline as proof that specific health interventions caused it. The study is observational, it compares different cohorts, and the authors say admission increases may reflect better service use or reporting.