Assessment of Routine Immunization Data Quality in Silti District of Southern Ethiopia

Background: Immunization coverage shows increment worldwide including Ethiopia in the past years but validity of facility reports has been questioned. Studies indicated that discrepancies in coverage and poor monitoring system to ensure data quality continue as challenges.

Objective: To assess routine immunization data quality from facility to district level in Silti district, Southern Ethiopia.

Methodology: It was facility based cross-sectional study using secondary data and quality index questionnaire. It was conducted in Silti district, Southern Ethiopia from August to September, 2010. The WHO data quality self-assessment procedure was used to select three health centers and 24 health posts. Descriptive analysis was done using Excel having special feature and SPSS 15.0 for internal comparison of means.

Results: The study indicated data quality monitoring system was 70%, 54%, and 71% for district, health centers and health posts’ respectively. Health workers trained on immunization service and data handling scored higher mean of quality index (78.2) than those who didn’t (55.9) with mean difference of 22.4 (P < 0.001). Similarly, significant mean difference of 18.8 (P < 0.001) was noted among facilities supervised regularly (79.5) than less frequently supervised (60.7). The study also showed over-reporting in the number of vaccinated children from lower to next higher level for all antigens. The overall weighted accuracy ratio of penta-3 coverage in the audit year from facility to district report is 77% which is out of WHO consistent AR level (85% ≤ AR ≤ 115%). Based on this finding the reported 98% coverage of 2009/10 is reduced by 23%. Therefore, the actual coverage will be 75%.

Conclusions and Recommendations: Accuracy ratio of penta-3 coverage showed overestimation. This over-reporting is highly associated with training and supervision. Therefore, district health office and partners should conduct in-service training and frequent supervision. Moreover, performance evaluation should not depend only on coverage.