Using health-facility data to assess subnational coverage of maternal and child health indicators, Kenya.

OBJECTIVE: To develop a systematic approach to obtain the best possible national and subnational statistics for maternal and child health coverage indicators from routine health-facility data. METHODS: Our approach aimed to obtain improved numerators and denominators for calculating coverage at the subnational level from health-facility data. This involved assessing data quality and determining adjustment factors for incomplete reporting by facilities, then estimating local target populations based on interventions with near-universal coverage (first antenatal visit and first dose of pentavalent vaccine). We applied the method to Kenya at the county level, where routine electronic reporting by facilities is in place via the district health information software system. FINDINGS: Reporting completeness for facility data were well above 80% in all 47 counties and the consistency of data over time was good. Coverage of the first dose of pentavalent vaccine, adjusted for facility reporting completeness, was used to obtain estimates of the county target populations for maternal and child health indicators. The country and national statistics for the four-year period 2012/13 to 2015/16 showed good consistency with results of the 2014 Kenya demographic and health survey. Our results indicated a stagnation of immunization coverage in almost all counties, a rapid increase of facility-based deliveries and caesarean sections and limited progress in antenatal care coverage. CONCLUSION: While surveys will continue to be necessary to provide population-based data, web-based information systems for health facility reporting provide an opportunity for more frequent, local monitoring of progress, in maternal and child health.

Completeness of malaria indicator data reporting via the District Health Information Software 2 in Kenya, 2011-2015.

BACKGROUND: Health facility-based data reported through routine health information systems form the primary data source for programmatic monitoring and evaluation in most developing countries. The adoption of District Health Information Software (DHIS2) has contributed to improved availability of routine health facility-based data in many low-income countries. An assessment of malaria indicators data reported by health facilities in Kenya during the first 5 years of implementation of DHIS2, from January 2011 to December 2015, was conducted. METHODS: Data on 19 malaria indicators reported monthly by health facilities were extracted from the online Kenya DHIS2 database. Completeness of reporting was analysed for each of the 19 malaria indicators and expressed as the percentage of data values actually reported over the expected number; all health facilities were expected to report data for each indicator for all 12 months in a year. RESULTS: Malaria indicators data were analysed for 6235 public and 3143 private health facilities. Between 2011 and 2015, completeness of reporting in the public sector increased significantly for confirmed malaria cases across all age categories (26.5-41.9%, p < 0.0001, in children aged <5 years; 30.6-51.4%, p < 0.0001, in persons aged ≥5 years). Completeness of reporting of new antenatal care (ANC) clients increased from 53.7 to 70.5%, p < 0.0001). Completeness of reporting of intermittent preventive treatment in pregnancy (IPTp) decreased from 64.8 to 53.7%, p < 0.0001 for dose 1 and from 64.6 to 53.4%, p < 0.0001 for dose 2. Data on malaria tests performed and test results were not available in DHIS2 from 2011 to 2014. In 2015, sparse data on microscopy (11.5% for children aged <5 years; 11.8% for persons aged ≥5 years) and malaria rapid diagnostic tests (RDTs) (8.1% for all ages) were reported. In the private sector, completeness of reporting increased significantly for confirmed malaria cases across all age categories (16.7-23.1%, p < 0.0001, in children aged <5 years; 19.4-28.6%, p < 0.0001, in persons aged ≥5 years). Completeness of reporting also improved for new ANC clients (16.2-23.6%, p < 0.0001), and for IPTp doses 1 and 2 (16.6-20.2%, p < 0.0001 and 15.5-20.5%, p < 0.0001, respectively). In 2015, less than 3% of data values for malaria tests performed were reported in DHIS2 from the private sector. CONCLUSIONS: There have been sustained improvements in the completeness of data reported for most key malaria indicators since the adoption of DHIS2 in Kenya in 2011. However, major data gaps were identified for the malaria-test indicator and overall low reporting across all indicators from private health facilities. A package of proven DHIS2 implementation interventions and performance-based incentives should be considered to improve private-sector data reporting.

Feasibility of an innovative electronic mobile system to assist health workers to collect accurate, complete and timely data in a malaria control programme in a remote setting in Kenya.

BACKGROUND: The cornerstone of decision making aimed at improving health services is accurate and timely health information. The Ministry of Public Health and Sanitation in Kenya decided to pilot feasibility of Fionet, an innovation that integrates diagnostics, data capture and cloud services, in its malaria control programme to demonstrate usability and feasibility by primary level workers in a remote setting in Kenya. METHODS: Eleven sites comprising one sub-district hospital, ten health centres and dispensaries were selected in three districts of Kisumu County to participate. Two health workers per site were selected, trained over a two-day period in the use of the Deki Reader™ to undertake rapid diagnostic testing (RDT) for malaria and data capture of patients' records. Health managers in the three districts were trained in the use of Fionet™ portal (web portal to cloud based information) to access the data uploaded by the Deki Readers. Field Support was provided by the Fio Corporation representative in Kenya. RESULTS: A total of 5812 malaria RDTs were run and uploaded to the cloud database during this implementation research study. Uploaded data were automatically aggregated into predetermined reports for use by service managers and supervisors. The Deki Reader enhanced the performance of the health workers by not only guiding them through processing of a malaria RDT test, but also by doing the automated analysis of the RDT, capturing the image, determining whether the RDT was processed according to guidelines, and capturing full patient data for each patient encounter. Supervisors were able to perform remote Quality assurance/Quality control (QA/QC) activities almost in real time. CONCLUSION: Quality, complete and timely data collection by health workers in a remote setting in Kenya is feasible. This paperless innovation brought unprecedented quality control and quality assurance in diagnosis, care and data capture, all in the hands of the health worker at point of care in an integrated way.

Patient adherence to prescribed artemisinin-based combination therapy in Garissa County, Kenya, after three years of health care in a conflict setting.

BACKGROUND: Current day malaria cases and deaths are indicative of a lack of access to both methods of prevention, diagnosis, and treatment; an important determinant of treatment efficacy is adherence. This study is a follow up to the baseline study of adherence to artemether-lumefantrine (AL) carried out in Garissa District in 2010. The study presented evaluates any changes in adherence levels which may have occurred in the area during this period and after nearly three years of sustained use of ACT across the public health sector. METHODS: The study was carried out in Garissa County in the North Eastern Province of Kenya and included patients fitting the suspected malaria case definition and having been prescribed AL, regardless of confirmatory diagnosis. A questionnaire assessed the intake of AL via both self-reporting by the participant and observation of blister packs by the interviewer. On separate occasions exit interviews with patients and observations of prescribers were also carried out. RESULTS: Of the 218 participants enrolled, 195 were successfully followed up. 60% of participants were found to be adherent to the three-day AL regimen, this is 4.7% lower than the proportion of participants adherent in 2010; the result of a two-sided z-test was not significant (p = 0.23). The odds of the patient being adherent to AL increased by 65% with each additional correct statement regarding how to take AL that a patient could recall (between zero and four statements), this was the only variable significantly associated with patient adherence (p = 0.01). CONCLUSION: Sustaining the ACT adherence rates at the 2010 levels, through 2.5 years of insecurity in the study area is an achievement and suggests that if security can be improved barriers to improving health service quality and patient adherence to AL would be removed. This study, by looking specifically at anti-malarial adherence over a prolonged period and in a setting of severe conflict, provides a valuable and rare insight in to the challenges and barriers to ACT adherence in such settings.

Perceptions of malaria and acceptance of rapid diagnostic tests and related treatment practises among community members and health care providers in Greater Garissa, North Eastern Province, Kenya.

BACKGROUND: Conventional diagnosis of malaria has relied upon either clinical diagnosis or microscopic examination of peripheral blood smears. These methods, if not carried out exactly, easily result in the over- or under-diagnosis of malaria. The reliability and accuracy of malaria RDTs, even in extremely challenging health care settings, have made them a staple in malaria control programmes. Using the setting of a pilot introduction of malaria RDTs in Greater Garissa, North Eastern Province, Kenya, this study aims to identify and understand perceptions regarding malaria diagnosis, with a particular focus on RDTs, and treatment among community members and health care workers (HCWs). METHODS: The study was conducted in five districts of Garissa County. Focus group discussions (FGD) were performed with community members that were recruited from health facilities (HFs) supported by the MENTOR Initiative. In-depth interviews (IDIs) and FGDs with HCWs were also carried out. Interview transcripts were then coded and analysed for major themes. Two researchers reviewed all codes, first separately and then together, discussed the specific categories, and finally characterized, described, and agreed upon major important themes. RESULTS: Thirty-four FGDs were carried out with a range of two to eight participants (median of four). Of 157 community members, 103 (65.6%) were women. The majority of participants were illiterate and the highest level of education was secondary school. Some 76% of participants were of Somali ethnicity. Whilst community members and HCWs demonstrated knowledge of aspects of malaria transmission, prevention, diagnosis, and treatment, gaps and misconceptions were identified. Poor adherence to negative RDT results, unfamiliarity and distrust of RDTs, and an inconsistent RDT supply were the main challenges to become apparent in FGDs and IDIs. CONCLUSION: Gaps in knowledge or incorrect beliefs exist in Greater Garissa and have the potential to act as barriers to complete and correct malaria case management. Addressing these knowledge gaps requires comprehensive education campaigns and a reliable and constant RDT supply. The results of this study highlight education and supply chain as key factors to be addressed in order to make large scale roll out of RDTs as successful and effective as possible.

Health disparities across the counties of Kenya and implications for policy makers, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016.

BACKGROUND: The Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2016 provided comprehensive estimates of health loss globally. Decision makers in Kenya can use GBD subnational data to target health interventions and address county-level variation in the burden of disease. METHODS: We used GBD 2016 estimates of life expectancy at birth, healthy life expectancy, all-cause and cause-specific mortality, years of life lost, years lived with disability, disability-adjusted life-years, and risk factors to analyse health by age and sex at the national and county levels in Kenya from 1990 to 2016. FINDINGS: The national all-cause mortality rate decreased from 850·3 (95% uncertainty interval [UI] 829·8-871·1) deaths per 100 000 in 1990 to 579·0 (562·1-596·0) deaths per 100 000 in 2016. Under-5 mortality declined from 95·4 (95% UI 90·1-101·3) deaths per 1000 livebirths in 1990 to 43·4 (36·9-51·2) deaths per 1000 livebirths in 2016, and maternal mortality fell from 315·7 (242·9-399·4) deaths per 100 000 in 1990 to 257·6 (195·1-335·3) deaths per 100 000 in 2016, with steeper declines after 2006 and heterogeneously across counties. Life expectancy at birth increased by 5·4 (95% UI 3·7-7·2) years, with higher gains in females than males in all but ten counties. Unsafe water, sanitation, and handwashing, unsafe sex, and malnutrition were the leading national risk factors in 2016. INTERPRETATION: Health outcomes have improved in Kenya since 2006. The burden of communicable diseases decreased but continues to predominate the total disease burden in 2016, whereas the non-communicable disease burden increased. Health gains varied strikingly across counties, indicating targeted approaches for health policy are necessary. FUNDING: Bill & Melinda Gates Foundation.

Pyrethroid susceptibility of malaria vectors in four Districts of western Kenya.

BACKGROUND: Increasing pyrethroid resistance in malaria vectors has been reported in western Kenya where long lasting insecticidal nets (LLINs) and indoor residual spraying (IRS) are the mainstays of vector control. To ensure the sustainability of insecticide-based malaria vector control, monitoring programs need to be implemented. This study was designed to investigate the extent and distribution of pyrethroid resistance in 4 Districts of western Kenya (Nyando, Rachuonyo, Bondo and Teso). All four Districts have received LLINs while Nyando and Rachuonyo Districts have had IRS campaigns for 3-5 years using pyrethroids. This study is part of a programme aimed at determining the impact of insecticide resistance on malaria epidemiology. METHODS: Three day old adult mosquitoes from larval samples collected in the field, were used for bioassays using the WHO tube bioassay, and mortality recorded 24 hours post exposure. Resistance level was assigned based on the 2013 WHO guidelines where populations with <90% mortality were considered resistant. Once exposed, samples were identified to species using PCR. RESULTS: An. arabiensis comprised at least 94% of all An. gambiae s.l. in Bondo, Rachuonyo and Nyando. Teso was a marked contrast case with 77% of all samples being An. gambiae s.s. Mortality to insecticides varied widely between clusters even in one District with mortality to deltamethrin ranging from 45-100%, while to permethrin the range was 30-100%. Mortality to deltamethrin in Teso District was < 90% in 4 of 6 clusters tested in An arabiensis and <90% in An. gambiae s.s in 5 of 6 clusters tested. To permethrin, mortality ranged between 5.9-95%, with <90% mortality in 9 of 13 and 8 of 13 in An. arabiensis and An. gambiae s.s. respectively. Cluster specific mortality of An. arabiensis between permethin and deltamethrin were not correlated (Z = 2.9505, P = 0.2483). CONCLUSION: High levels of pyrethroid resistance were observed in western Kenya. This resistance does not seem to be associated with either species or location. Insecticide resistance can vary within small geographical areas and such heterogeneity may make it possible to evaluate the impact of resistance on malaria and mosquito parameters within similar eco-epidemiological zones.