Health Management Information System (HMIS)

hospital management information system

HMIS is the cornerstone of monitoring and evaluation of health programmes. However, the data from HMIS need to be interpreted with caution. HMIS are based on routine data collection to monitor the performance of health services and therefore are bound to be affected by data quality problems. I have analyzed the data coming from many types of HMIS by taking into account reliability issues and by validating the data through chart audits.

I have more than 10-year working experience in the monitoring and evaluation of Tuberculosis control programmes. Between 2001 and 2008 I provided technical assistance in epidemiology to the Department of Health (DOH) of KwaZulu-Natal (KZN), South Africa. This included the analysis of the complex architecture of the electronic TB register and the critical interpretation of treatment outcome indicators.

Examples include:

(a) The validation of the TB electronic register of KZN. Interpreting the TB treatment outcome indicators is fraught with problems because data are taken as valid without considering the many potential sources of unreliability. The TB register is particularly complex because it is based on a cohort system to monitor the different phases of diagnosis, follow-up visits, transfers and final outcomes. I managed a chart audit of the electronic TB register in the district of Umzinyathi, KwaZulu-Natal, South Africa. One of the issues that I evaluated was the influence of double counting due to the absence of a unique patients' ID in the TB register. This identified one source of unreliability in estimating treatment outcome rates, especially in districts that were characterized by high frequency of transfers.

(b) Analysis of the Management Information System of the Antiretroviral Treatment (ART) Programme in KZN. Each ART clinic was sending to the DOH weekly aggregates on the number of HIV/AIDS patients who were counseled, tested for HIV and CD4, began treatment, had side effects, died and defaulted. I assisted the DOH in critically reviewing the reliability of the ART management indicators.

(c) I provided technical assistance to the KZN DOH to evaluate its hospital management information system. This included the analysis of the hospital indicators to identify reliability problems and to correctly interpret the indicators, the pilot test of a methodology to validate the hospital management information system in a limited sample of hospitals, and the full fledge validation. This was carried out by visiting all the 67 hospitals of KZN to collect primary data and compare them with the reported data.

(d) I assisted the Policy and Planning Unit of the Federal Ministry of Health in Ethiopia to critically review the hospital performance. This included a critical analysis of the data produced by the hospital information system and field visits to identify factors influencing hospital performance.