Use of data quality assessment to evaluate the level of HIV data completeness in Ministry of Health facilities: a case study of Eastern province-Kenya

Human Immunodeficiency Virus (HIV) and Acquired Immune Deficiency Syndrome data feedback and decision making is enhanced by data quality assessments which are done using well designed tools and methods. HIV data quality is useful in HIV prevention, care and treatment. The main objective of the study was to evaluate the quality of data collected in Ministry of Health facilities offering HIV prevention, care and treatment services in Eastern Province of Kenya. Assessment of factors that contribute to HIV data incompleteness in Ministry of health facilities was done by interviewing 60 health care providers and patients records reviewed in 42 health facilities to determine whether any difference exists in level of HIV data completeness in Eastern Province-Kenya. Data on perceptions behind missing HIV data entries in patients‟ records and competence in HIV data quality assessment for completeness was obtained by administering a semi-structured questionnaire to health care providers. Logistic regression model was used to create a relationship between training of health care providers in health management information systems (HMIS), experience in filling patient records, gender and respective competence on how to conduct HIV data quality assessment for completeness. Second, patient records (Blue cards) were evaluated for complete entries and the level of data completeness was determined. This was done by evaluating the patients‟ records for complete entries in the fields of weight, new opportunistic infections (OI), pregnancy status and Tuberculosis (TB) screening. The level of HIV data completeness varied in different HIV data elements. Weight had the least difference in level of data completeness at 6.89% while screening of new opportunistic infection and pregnancy status for patients had the biggest difference at 10.86% and 10.77% respectively. It was evident that training health care providers in health management information systems (P=0.007) and having had an experience in the comprehensive care clinic (P<0.000) contributed significantly to competence on how to conduct a HIV DQA for completeness. It was noted that many HIV data collection tools for documentation 26 (43%) and few record officers 24 (40%), were perceived as the major causes of data incompleteness in patient records. Documentation of patients‟ weight, pregnancy status, screening of new opportunistic infections and tuberculosis in HIV records was incomplete. Many HIV data collection records for documentation and few record officers assigned to work in the comprehensive care clinics were identified as the major causes of missing entries in HIV records. Training health care providers in health management and information systems and having had worked in the HIV clinic for at least ten months contributed positively in competence of conducting a structured HIV DQA for completeness.