Essential drugs supply and usage as a reflection of outpatient morbidity in Kirinyaga District, Kenya.

This study was carried out in all the twenty eight Government rural health facilities (RHF's) in Kirinyaga District of Kenya. It aimed at determining whether the quantities and types of drugs supplied to the RHF's matched the morbidity patterns, and also whether the disease patterns continued to be used as a basis for requesting stocks of essential drugs. It also investigated the constraints associated with the programme with an ultimate aim of improving service delivery. The information was obtained through records review of the Bin card books, Out-patients Department (OPD) register, morbidity monthly returns, staff and patient interviews. A highly significant (p < 0.01) variation in morbidity (workload) by season of the year and between the ecological divisions of the district was observed. Significant differences were also observed between the quantities of drugs supplied and the actual monthly requirements in all the RHF's, implying that the disease patterns were not necessarily being used for requesting stocks of essential drugs. A distribution of essential drugs based on two broad ecological divisions of the district was recommended. One of the major constraints found to be associated with the Essential Drugs Programme was the delay in procurement/distribution of essential drugs to the RHF's. This was mainly due to the tendering procedures of Kit 2 by the Ministry of Health (MOH) and also occasional lack of transport at the district level.

Verbal autopsy: a tool for determining cause of death in a community.

Verbal autopsy was used to determine causes of death in 239 children under the age of 5 years. The diagnosis derived from verbal autopsy was corroborated with hospital diagnosis in 39 cases. There was concurrence of diagnosis in 72% of the cases. Using the diagnosis of bronchopneumonia to validate the method, verbal autopsy was found to have a sensitivity of 71%, specificity of 92%, a positive predictive value of 85%. Reliability index of agreement was 0.654. Recall period up to 29 months after death was found to be reliable.