Development of novel composite data quality scores to evaluate facility-level data quality in electronic data in Kenya: a nationwide retrospective cohort study.

BACKGROUND: In this evaluation, we aim to strengthen Routine Health Information Systems (RHIS) through the digitization of data quality assessment (DQA) processes. We leverage electronic data from the Kenya Health Information System (KHIS) which is based on the District Health Information System version 2 (DHIS2) to perform DQAs at scale. We provide a systematic guide to developing composite data quality scores and use these scores to assess data quality in Kenya. METHODS: We evaluated 187 HIV care facilities with electronic medical records across Kenya. Using quarterly, longitudinal KHIS data from January 2011 to June 2018 (total N = 30 quarters), we extracted indicators encompassing general HIV services including services to prevent mother-to-child transmission (PMTCT). We assessed the accuracy (the extent to which data were correct and free of error) of these data using three data-driven composite scores: 1) completeness score; 2) consistency score; and 3) discrepancy score. Completeness refers to the presence of the appropriate amount of data. Consistency refers to uniformity of data across multiple indicators. Discrepancy (measured on a Z-scale) refers to the degree of alignment (or lack thereof) of data with rules that defined the possible valid values for the data. RESULTS: A total of 5,610 unique facility-quarters were extracted from KHIS. The mean completeness score was 61.1% [standard deviation (SD) = 27%]. The mean consistency score was 80% (SD = 16.4%). The mean discrepancy score was 0.07 (SD = 0.22). A strong and positive correlation was identified between the consistency score and discrepancy score (correlation coefficient = 0.77), whereas the correlation of either score with the completeness score was low with a correlation coefficient of -0.12 (with consistency score) and -0.36 (with discrepancy score). General HIV indicators were more complete, but less consistent, and less plausible than PMTCT indicators. CONCLUSION: We observed a lack of correlation between the completeness score and the other two scores. As such, for a holistic DQA, completeness assessment should be paired with the measurement of either consistency or discrepancy to reflect distinct dimensions of data quality. Given the complexity of the discrepancy score, we recommend the simpler consistency score, since they were highly correlated. Routine use of composite scores on KHIS data could enhance efficiencies in DQA at scale as digitization of health information expands and could be applied to other health sectors beyondHIV clinics.

The association between health workforce availability and HIV-program outcomes in Côte d'Ivoire.

OBJECTIVE: The purpose of this study was to assess the distribution of HIV-program staff and the extent to which their availability influences HIV programmatic and patient outcomes. METHODS: The study was a facility level cross-sectional survey. Data from October 2018 to September 2019 were abstracted from HIV program reports conducted in 18 districts of Côte d'Ivoire. The distribution of staff in clinical, laboratory, pharmacy, management, lay, and support cadres were described across high and low antiretroviral therapy (ART) volume facilities. Non-parametric regression was used to estimate the effects of cadre categories on the number of new HIV cases identified, the number of cases initiated on ART, and the proportion of patients achieving viral load suppression. RESULTS: Data from 49,871 patients treated at 216 health facilities were included. Low ART volume facilities had a median of 8.1 staff-per-100 ART patients, significantly higher than the 4.4 staff-per-100 ART patients at high-ART volume facilities. One additional laboratory staff member was associated with 4.30 (IQR: 2.00-7.48, p < 0.001) more HIV cases identified and 3.81 (interquartile range [IQR]: 1.44-6.94, p < 0.001) additional cases initiated on ART. Similarly, one additional lay worker was associated with 2.33 (IQR: 1.00-3.43, p < 0.001) new cases identified and 2.24 (IQR: 1.00-3.31, p < 0.001) new cases initiated on ART. No cadres were associated with viral suppression. CONCLUSIONS: HCWs in the laboratory and lay cadre categories were associated with an increase in HIV-positive case identification and initiation on ART. Our findings suggest that allocation of HCWs across health facilities should take into consideration the ART patient volume. Overall, increasing investment in health workforce is critical to achieve national HIV goals and reaching HIV epidemic control.

Strengthening health systems in poor countries: a code of conduct for nongovernmental organizations.

The challenges facing efforts in Africa to increase access to antiretroviral HIV treatment underscore the urgent need to strengthen national health systems across the continent. However, donor aid to developing countries continues to be disproportionately channeled to international nongovernmental organizations (NGOs) rather than to ministries of health. The rapid proliferation of NGOs has provoked "brain drain" from the public sector by luring workers away with higher salaries, fragmentation of services, and increased management burdens for local authorities in many countries. Projects by NGOs sometimes can undermine the strengthening of public primary health care systems. We argue for a return to a public focus for donor aid, and for NGOs to adopt a code of conduct that establishes standards and best practices for NGO relationships with public sector health systems.

Measuring adherence to antipsychotic medications for schizophrenia: Concordance and validity among a community sample in rural China.

BACKGROUND: Despite the abundance of measures to assess medication adherence by persons suffering schizophrenia, few studies have evaluated their concordance and validity against a reference standard in resource-poor community settings. We explored the concordance and validity of several measures to assess antipsychotic medication adherence in a resource-poor community. METHOD: Based on a random sample of 278 villagers diagnosed with schizophrenia from Liuyang, Hunan Province, China, we used a concordance correlation coefficient (rc) and Kappa statistic to assess agreement among pill counts, refill records, clinician rating, Drug Attitude Inventory (DAI), and the Brief Adherence Rating Scale (BARS). The validity of various measures was evaluated by their concordance and sensitivity/specificity to home-based unannounced pill count (UPC) as the reference standard. RESULTS: The estimated proportion of adherent patients according to all measures (41% ~ 88%) was substantially higher than identified by UPC (35%). Concordance between any two measures was poor (rc/Kappa mostly <0.30). Validity of various measures also was poor against the UPC (rc < 0.20; Kappa <0.16), although refill records and the structured instruments (BARS) performed better than office-based pill counts and clinician impression. BARS, DAI and clinician rating were not sensitive to changes in adherence and would likely underestimate any program effect. CONCLUSION: In resource-poor community settings, most measures assessed in this study should not be used alone as they overestimated adherence, underestimated program effect, and had poor validity. A combination of UPC and several other measures may provide more insight into clinical trials and programmatic management.

Antiretroviral-associated toxicity among HIV-1-seropositive pregnant women in Mozambique receiving nevirapine-based regimens.

OBJECTIVE: To assess toxicities associated with highly active antiretroviral therapy (HAART) among HIV-1-infected pregnant women treated with nevirapine-based regimens according to Mozambican national guidelines. STUDY DESIGN: Prospective cohort study. METHODS: HIV-1-infected antiretroviral-naive pregnant women with CD4 counts < or =350 cells/microL were initiated on nevirapine, lamivudine, and stavudine or zidovudine and followed monthly. Severe hepatotoxicity was defined as aspartate aminotransferase (AST) or alanine aminotransferase (ALT) levels > or =5-fold the upper limit of normal. Analyses were stratified by baseline CD4 count (<250 vs. 250-350 cells/microL). RESULTS: Among 146 pregnant women, 75 (52%) began nevirapine, lamivudine, and zidovudine and 71 (48%) began nevirapine, lamivudine, and stavudine. Overall, 79 (54%) women had CD4 counts <250 cells/microL, 7 (5%) had grade II hepatotoxicity, and 4 (3%) had severe (grade III or IV) hepatotoxicity. All 4 women with severe hepatotoxicity had baseline CD4 counts > or =250 cells/microL (P = 0.02). Rates of skin toxicity, anemia, and peripheral neuropathy did not differ by CD4 cell count group. Overall, 12 (8%) women changed or discontinued HAART as a result of drug toxicity. CONCLUSIONS: Severe hepatotoxicity from nevirapine-containing HAART in this cohort of pregnant women was more common at higher CD4 counts (6% vs. 0% among women with CD4 counts > or =250 cells/microL and CD4 counts <250 cells/microL, respectively), suggesting that laboratory monitoring is necessary when administering nevirapine-containing regimens to pregnant women with CD4 counts > or =250 cells/microL.