Health system lessons from the global fund-supported procurement and supply chain investments in Zimbabwe: a mixed methods study.

BACKGROUND: The Global Fund partnered with the Zimbabwean government to provide end-to-end support to strengthen the procurement and supply chain within the health system. This was accomplished through a series of strategic investments that included infrastructure and fleet improvement, training of personnel, modern equipment acquisition and warehouse optimisation. This assessment sought to determine the effects of the project on the health system. METHODS: This study employed a mixed methods design combining quantitative and qualitative research methods. The quantitative part entailed a descriptive analysis of procurement and supply chain data from the Zimbabwe healthcare system covering 2018 - 2021. The qualitative part comprised key informant interviews using a structured interview guide. Informants included health system stakeholders privy to the Global Fund-supported initiatives in Zimbabwe. The data collected through the interviews were transcribed in full and subjected to thematic content analysis. RESULTS: Approximately 90% of public health facilities were covered by the procurement and distribution system. Timeliness of order fulfillment (within 90 days) at the facility level improved from an average of 42% to over 90% within the 4-year implementation period. Stockout rates for HIV drugs and test kits declined by 14% and 49% respectively. Population coverage for HIV treatment for both adults and children remained consistently high despite the increasing prevalence of people living with HIV. The value of expired commodities was reduced by 93% over the 4-year period. Majority of the system stakeholders interviewed agreed that support from Global Fund was instrumental in improving the country's procurement and supply chain capacity. Key areas include improved infrastructure and equipment, data and information systems, health workforce and financing. Many of the participants also cited the Global Fund-supported warehouse optimization as critical to improving inventory management practices. CONCLUSION: It is imperative for governments and donors keen to strengthen health systems to pay close attention to the procurement and distribution of medicines and health commodities. There is need to collaborate through joint planning and implementation to optimize the available resources. Organizational autonomy and sharing of best practices in management while strengthening accountability systems are fundamentally important in the efforts to build institutional capacity.

Progress and Challenges in Scaling Up Laboratory Monitoring of HIV Treatment.

In a perspective on Habiyambere and colleagues, Peter Kilmarx and Raiva Simbi discuss the disconnect between HIV testing instrument capacity and utilization.

Diagnostic Performance of STANDARD™ M10 Multidrug-resistant Tuberculosis Assay for Detection of Mycobacterium tuberculosis and Rifampicin and Isoniazid Resistance in Zimbabwe.

BACKGROUND: Although Zimbabwe has transitioned out of the 30 high-burden countries, it still remained in the 30 high multidrug-resistant (MDR)/rifampicin-resistant tuberculosis (TB) burden. Rapid detection of rifampicin (RIF) and isoniazid (INH) is essential for the diagnosis of MDR-TB. The World Health Organization has recommended the use of molecular WHO-recommended rapid diagnostic (mWRD) for TB and DR-TB. STANDARD™ M10 MDR-TB assay is a new molecular rapid diagnostic assay developed by SD Biosensor for the detection of Mycobacterium tuberculosis (MTB) and RIF and INF resistance. This study aims to determine the diagnostic accuracy of STANDARD™ M10 MDR-TB assay. METHODS: The study was conducted on 214 samples with different MTB and RIF and INH resistance status. The STANDARD™ M10 MDR-TB assay was performed according to the manufacturer's instructions. Xpert MTB/RIF Ultra, MGIT culture, and phenotypic drug susceptibility testing are used as comparative methods. RESULTS: The sensitivity and specificity of STANDARD™ M10 MDR-TB assay for the detection of MTB are 99% and 97.9%, respectively. The sensitivity and specificity of the assay for detection of MDR-TB were 97.8% and 100%, respectively. CONCLUSION: The STANDARD™ M10 MDR-TB assay demonstrated high diagnostic accuracy in the detection of MTB and RIF and INH resistance. This molecular assay can also be used as an alternative to other mWRD assays.

Laboratory capacity strengthening in Zimbabwe as part of the COVID-19 response: what has worked? What still needs to be done?

The COVID-19 pandemic was declared a Public Health Emergency of International Concern on January 30, 2020. The government of Zimbabwe through the Ministry of Health and Child Care set up the COVID-19 national preparedness and response plan in which the laboratory was a key pillar. The implementation of PCR testing, genomic sequencing, and the establishment of quality management systems during the COVID-19 response strengthened the capacity of the public health laboratory system in responding to the pandemic. Here we present the different strategies taken by the government that strengthened laboratory capacity, the lessons learned during the COVID-19 response, and recommendations on how the capacity can be sustained and leveraged for outbreak response in the future.

Zimbabwe's emergency response to COVID-19: Enhancing access and accelerating COVID-19 testing as the first line of defense against the COVID-19 pandemic.

The Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) spreads rapidly, causing outbreaks that grow exponentially within a short period before interventions are sought and effectively implemented. Testing is part of the first line of defense against Corona Virus Disease of 2019 (COVID-19), playing a critical role in the early identification and isolation of cases to slow transmission, provision of targeted care to those affected, and protection of health system operations. Laboratory tests for COVID-19 based on nucleic acid amplification techniques were rapidly developed in the early days of the pandemic, but such tests typically require sophisticated laboratory infrastructure and skilled staff. In March 2020, Zimbabwe confirmed its first case of COVID-19; this was followed by an increase in infection rates as the pandemic spread across the country, thus increasing the demand for testing. One national laboratory was set to test all the country's COVID-19 suspect cases, building pressure on human and financial resources. Staff burnout and longer turnaround times of more than 48 h were experienced, and results were released late for clinical relevance. Leveraging on existing PCR testing platforms, including GeneXpert machines, eased the pressure for a short period before facing the stockout of SARs-CoV-2 cartridges for a long time, leading to work overload at a few testing sites contributing to long turnaround times. On September 11, WHO released the interim guidance to use antigen rapid diagnostic test as a diagnostic tool. The Zimbabwe laboratory pillar quickly adopted it and made plans for its implementation. The National Microbiology Reference Laboratory verified the two emergency-listed kits, the Panbio Abbott and the Standard Q, Biosensor, and they met the WHO minimum performance of ≥97% specificity and ≥80% sensitivity. Decentralizing diagnostic testing leveraging existing human resources became a game-changer in improving COVID-19 containment measures. Task shifting through training on Antigen rapid diagnostic tests (Ag-RDT) commenced, and testing was decentralized to all the ten provinces, from 1 central testing laboratory to more than 1,000 testing centers. WhatsApp platforms made it easier for data to be reported from remote areas. Result turnaround times were improved to the same day, and accessibility to testing was enhanced.

Near-point-of-care viral load testing during pregnancy and viremia at delivery.

OBJECTIVES: Assess whether near-point-of-care (POC) viral load testing at the first antenatal care visit (ANC1) increased the proportion of women taking antiretroviral therapy who were virally suppressed at delivery through expedited clinical action. DESIGN: Difference-in-difference analysis. METHODS: At 20 public sector facilities in Zimbabwe, 10 implemented near-POC viral load testing at ANC1 (August 2019 to November 2020) and 10 used centralized viral load testing at ANC1. Study endpoints included time to result received, clinical action, and unsuppressed viral load (UVL; >1000 copies/ml) at delivery. RESULTS: Of 1782 women, only 46% came for ANC1 before their third trimester. Preimplementation, 28% of women received viral load testing at ANC1, increasing to 86% during implementation. In the near-POC viral load arm, women were more likely to receive their result within 30 days of ANC1 sample collection compared with the centralized laboratory arm [54 versus 14%, relative risk (RR): 4.17, 95% confidence interval (CI) 1.82-9.55], as well as receive clinical action among those with UVL (63 versus 8%, RR 7.88; 95% CI 1.53-40.47). However, we did not observe significant changes in risk of UVL at delivery with near-POC viral load (RR 1.02, 95% CI 0.95-1.10). CONCLUSION: ANC1 viral load coverage was initially low. Near-POC viral load testing at ANC1 dramatically improved the timeliness of result receipt by patients and clinical action for those with an UVL. Although we did not observe a significant impact of provision of near-POC viral load at ANC1 on re-suppression at delivery, potentially because of late presentation for ANC1, continued near-POC viral load testing during pregnancy and delivery may reduce UVL and mother-to-child transmission risk.

Feasibility and impact of near-point-of-care integrated tuberculosis/HIV testing in Malawi and Zimbabwe.

OBJECTIVES: Near-point-of-care (POC) testing for early infant diagnosis (EID) and viral load expedites clinical action and improves outcomes but requires capital investment. We assessed whether excess capacity on existing near-POC devices used for TB diagnosis could be leveraged to increase near-POC HIV molecular testing, termed integrated testing, without compromising TB services. DESIGN: Preimplementation/postimplementation studies in 10 health facilities in Malawi and 8 in Zimbabwe. METHODS: Timeliness of EID and viral load test results and clinical action were compared between centralized and near-POC testing using Somers' D tests (continuous indicators) and risk ratios (RR, binary indicators); TB testing/treatment rates and timeliness were analyzed preintegration/postintegration. RESULTS: With integration, average device utilization increased but did not exceed 55%. Despite the addition of HIV testing, TB test volumes, timeliness, and treatment initiations were maintained. Although few HIV-positive infants were identified, near-POC EID testing improved treatment initiation within 1 month by 57% compared with centralized EID [Malawi RR: 1.57, 95% confidence interval (CI) 0.98-2.52], and near-POC viral load testing significantly increased the proportion of patients with elevated viral load receiving clinical action within 1 month (Zimbabwe RR: 5.26, 95% CI 3.38-8.20; Malawi RR: 3.90, 95% CI 2.58-5.91). CONCLUSION: Integrating TB/HIV testing using existing multidisease platforms is feasible and enables increased access to rapid diagnostics without disrupting existing TB services. Our results serve as an example of a novel, efficient implementation model that can increase access to critical testing services across disease silos and should be considered for additional clinical applications.

Evaluation of near point-of-care viral load implementation in public health facilities across seven countries in sub-Saharan Africa.

INTRODUCTION: In many low- and middle-income countries, HIV viral load (VL) testing occurs at centralized laboratories and time-to-result-delivery is lengthy, preventing timely monitoring of HIV treatment adherence. Near point-of-care (POC) devices, which are placed within health facility laboratories rather than clinics themselves (i.e. "true" POC), can offer VL in conjunction with centralized laboratories to expedite clinical decision making and improve outcomes, especially for patients at high risk of treatment failure. We assessed impacts of near-POC VL testing on result receipt and clinical action in public sector programmes in Cameroon, Democratic Republic of Congo, Kenya, Malawi, Senegal, Tanzania and Zimbabwe. METHODS: Routine health data were collected retrospectively after introducing near-POC VL testing at 57 public sector health facilities (2017 to 2019, country-dependent). Where possible, key indicators were compared to data from patients receiving centralized laboratory testing using hazard ratios and the Somers' D test. RESULTS: Data were collected from 6795 tests conducted on near-POC and 17614 tests on centralized laboratory-based platforms. Thirty-one percent (2062/6694) of near-POC tests were conducted for high-risk populations: pregnant and breastfeeding women, children and those with suspected failure. Compared to conventional testing, near-POC improved the median time from sample collection to return of results to patient [six vs. sixty-eight days, effect size: -32.2%; 95% CI: -41.0% to -23.4%] and to clinical action for individuals with an elevated HIV VL [three vs. fourty-nine days, effect size: -35.4%; 95% CI: -46.0% to -24.8%]. Near-POC VL results were two times more likely to be returned to the patient within 90 days compared to centralized tests [50% (1781/3594) vs. 27% (4172/15271); aHR: 2.22, 95% CI: 2.05 to 2.39]. Thirty-seven percent (340/925) of patients with an elevated near-POC HIV VL result had documented clinical follow-up actions within 30 days compared to 7% (167/2276) for centralized testing. CONCLUSIONS: Near-POC VL testing enabled rapid test result delivery for high-risk populations and led to significant improvements in the timeliness of patient result receipt compared to centralized testing. While there was some improvement in time-to-clinical action with near-POC VL testing, major gaps remained. Strengthening of systems supporting the utilization of results for patient management are needed to truly capitalize on the benefits of decentralized testing.

Prevalence, risk factors and treatment outcomes of isoniazid resistant TB in Bulawayo city, Zimbabwe: A cohort study.

INTRODUCTION: The isoniazid-resistant TB poses a threat to TB control efforts. Zimbabwe, one of the high TB burden countries, has not explored the burden of isoniazid resistant TB. Hence among all bacteriologically-confirmed TB patients diagnosed in Bulawayo City during March 2017 and December 2018, we aimed to assess the proportion with isoniazid resistant TB and associated factors. Also, we aimed to describe the TB treatment outcomes. METHODOLOGY: A cohort study involving routinely collected data by the National TB Reference Laboratory (NTBRL) in Bulawayo City and National TB programme of Zimbabwe. The percentage with 95% confidence interval (CI) was used to express the proportion with isoniazid-resistant TB. The modified Poisson regression was used to assess the association of demographic and clinical characteristics with isoniazid mono-resistant TB. RESULTS: Of 2160 bacteriologically-confirmed TB patients, 1612 (74.6%) had their sputum received at the NTBRL and 743 (46.1%) had culture growth. Among those with culture growth, 34 (4.6%, 95% CI: 3.5-6.7) had isoniazid mono-resistant TB, 25 (3.3%, 95% CI: 2.2-4.9) had MDR-TB. Thus, 59 (7.9%, 95% CI: 6.1-10.1) had isoniazid-resistant TB. Children < 15 years had a higher prevalence of isoniazid mono-resistant TB (aPR= 3.93; 95% CI: 1.24-12.45). Among those with rifampicin sensitive TB, patients with isoniazid-sensitive TB had higher favourable treatment outcomes compared to those with isoniazid-resistant TB (86.3% versus 75.5%, p = 0.039). CONCLUSIONS: The prevalence of isoniazid-resistant TB was low compared to neighbouring countries with high burden of TB-HIV. However, Zimbabwe should consider reviewing treatment guidelines for isoniazid mono-resistant TB due to the observed poor treatment outcomes.

Data connectivity: A critical tool for external quality assessment.

Point-of-care (POC) tests have been useful in increasing access to testing and treatment monitoring for HIV. Decentralising testing from laboratories to hundreds of sites around a country presents tremendous challenges in training and quality assurance. In order to address these concerns, companies are now either embedding connectivity in their new POC diagnostic instruments or providing some form of channel for electronic result exchange. These will allow automated key performance and operational metrics from devices in the field to a central database. Setting up connectivity between these POC devices and a central database at the Ministries of Health will allow automated data transmission, creating an opportunity for real-time information on diagnostic instrument performance as well as the competency of the operator through external quality assessment. A pilot programme in Zimbabwe shows that connectivity has significantly improve the turn-around time of external quality assessment result submissions and allow corrective actions to be provided in a timely manner. Furthermore, by linking the data to existing supply chain management software, stock-outs can be minimised. As countries are looking forward to achieving the 90-90-90 targets for HIV, such innovative technologies can automate disease surveillance, improve the quality of testing and strengthen the efficiency of health systems.

Maximising mentorship: Variations in laboratory mentorship models implemented in Zimbabwe.

BACKGROUND: Laboratory mentorship has proven to be an effective tool in building capacity and assisting laboratories in establishing quality management systems. The Zimbabwean Ministry of Health and Child Welfare implemented four mentorship models in 19 laboratories in conjunction with the Strengthening Laboratory Management Toward Accreditation (SLMTA) programme. OBJECTIVES: This study outlines how the different models were implemented, cost involved per model and results achieved. METHODS: Eleven of the laboratories had been trained previously in SLMTA (Cohort I). They were assigned to one of three mentorship models based on programmatic considerations: Laboratory Manager Mentorship (Model 1, four laboratories); One Week per Month Mentorship (Model 2, four laboratories); and Cyclical Embedded Mentorship (Model 3, three laboratories). The remaining eight laboratories (Cohort II) were enrolled in Cyclical Embedded Mentorship incorporated with SLMTA training (Model 4). Progress was evaluated using a standardised audit checklist. RESULTS: At SLMTA baseline, Model 1-3 laboratories had a median score of 30%. After SLMTA, at mentorship baseline, they had a median score of 54%. At the post-mentorship audit they reached a median score of 75%. Each of the three mentorship models for Cohort I had similar median improvements from pre- to post-mentorship (17 percentage points for Model 1, 23 for Model 2 and 25 for Model 3; p > 0.10 for each comparison). The eight Model 4 laboratories had a median baseline score of 24%; after mentorship, their median score increased to 63%. Median improvements from pre-SLMTA to post-mentorship were similar for all four models. CONCLUSION: Several mentorship models can be considered by countries depending on the available resources for their accreditation implementation plan.

Weighing the costs: Implementing the SLMTA programme in Zimbabwe using internal versus external facilitators.

BACKGROUND: In 2010, the Zimbabwe Ministry of Health and Child Welfare (MoHCW) adopted the Strengthening Laboratory Management Toward Accreditation (SLMTA) programme as a tool for laboratory quality systems strengthening. OBJECTIVES: To evaluate the financial costs of SLMTA implementation using two models (external facilitators; and internal local or MoHCW facilitators) from the perspective of the implementing partner and to estimate resources needed to scale up the programme nationally in all 10 provinces. METHODS: The average expenditure per laboratory was calculated based on accounting records; calculations included implementing partner expenses but excluded in-kind contributions and salaries of local facilitators and trainees. We also estimated theoretical financial costs, keeping all contextual variables constant across the two models. Resource needs for future national expansion were estimated based on a two-phase implementation plan, in which 12 laboratories in each of five provinces would implement SLMTA per phase; for the internal facilitator model, 20 facilitators would be trained at the beginning of each phase. RESULTS: The average expenditure to implement SLMTA in 11 laboratories using external facilitators was approximately US$5800 per laboratory; expenditure in 19 laboratories using internal facilitators was approximately $6000 per laboratory. The theoretical financial cost of implementing a 12-laboratory SLMTA cohort keeping all contextual variables constant would be approximately $58 000 using external facilitators; or $15 000 using internal facilitators, plus $86 000 to train 20 facilitators. The financial cost for subsequent SLMTA cohorts using the previously-trained internal facilitators would be approximately $15 000, yielding a break-even point of 2 cohorts, at $116 000 for either model. Estimated resources required for national implementation in 120 laboratories would therefore be $580 000 using external facilitators ($58 000 per province) and $322 000 using internal facilitators ($86 000 for facilitator training in each of two phases plus $15 000 for SLMTA implementation in each province). CONCLUSION: Investing in training of internal facilitators will result in substantial savings over the scale-up of the programme. Our study provides information to assist policy makers to develop strategic plans for investing in laboratory strengthening.