Introducing an Ethics Framework for health priority-setting in South Africa on the path to universal health coverage.

BACKGROUND: South Africa (SA) has embarked on a process to implement universal health coverage (UHC) funded by National Health Insurance (NHI). The 2019 NHI Bill proposes creation of a health technology assessment (HTA) body to inform decisions about which interventions NHI funds will cover under UHC. In practice, HTA often relies mainly on economic evaluations of cost-effectiveness and budget impact, with less attention to the systematic, specific consideration of important social, organisational and ethical impacts of the health technology in question. In this context, the South African Values and Ethics for Universal Health Coverage (SAVE-UHC) research project recognised an opportunity to help shape the health priority-setting process by providing a way to take account of multiple, ethically relevant considerations that reflect SA values. The SAVE-UHC Research Team developed and tested an SA-specific Ethics Framework for HTA assessment and analysis. OBJECTIVES: To develop and test an Ethics Framework for use in the SA context for health priority-setting. METHODS: The Framework was developed iteratively by the authors and a multidisciplinary panel (18 participants) over a period of 18 months, using the principles outlined in the 2015 NHI White Paper as a starting point. The provisional Ethics Framework was then tested with multi-stakeholder simulated appraisal committees (SACs) in three provinces. The membership of each SAC roughly reflected the composition of a potential SA HTA committee. The deliberations and dedicated focus group discussions after each SAC meeting were recorded, analysed and used to refine the Framework, which was presented to the Working Group for review, comment and final approval. RESULTS: This article describes the 12 domains of the Framework. The first four (Burden of the Health Condition, Expected Health Benefits and Harms, Cost-Effectiveness Analysis, and Budget Impact) are commonly used in HTA assessments, and a further eight cover the other ethical domains. These are Equity, Respect and Dignity, Impacts on Personal Financial Situation, Forming and Maintaining Important Personal Relationships, Ease of Suffering, Impact on Safety and Security, Solidarity and Social Cohesion, and Systems Factors and Constraints. In each domain are questions and prompts to enable use of the Framework by both analysts and assessors. Issues that arose, such as weighting of the domains and the availability of SA evidence, were discussed by the SACs. CONCLUSIONS: The Ethics Framework is intended for use in priority-setting within an HTA process. The Framework was well accepted by a diverse group of stakeholders. The final version will be a useful tool not only for HTA and other priority-setting processes in SA, but also for future efforts to create HTA methods in SA and elsewhere.

A prospective observational study of developmental outcomes in survivors of neonatal hypoxic ischaemic encephalopathy in South Africa.

BACKGROUND: Neonatal hypoxic ischaemic encephalopathy (NHIE) is an important cause of long-term handicap in survivors. There is limited information on the burden of handicap from NHIE in sub-Saharan Africa. OBJECTIVES: To determine the developmental outcomes in survivors of NHIE in South Africa (SA). METHODS: In this prospective observational study, the developmental outcomes in 84 infants who had survived hypoxic ischaemic encephalopathy (the NHIE group) were compared with those in 64 unaffected infants (the control group). The Bayley Scales of Infant Development version III were used for assessment of developmental outcomes. RESULTS: Significant differences were found between the developmental outcomes of the two groups, with a significantly lower composite language score and higher proportions with language, motor and cognitive developmental delays in the NHIE group than in the control group. Cerebral palsy (CP) was present in 13 of the infants with NHIE (15.5%) and none in the control group (p<0.001). CP was associated with developmental delay, and also with the severity of NHIE. Therapeutic hypothermia (TH) was administered in 58.3% of the study group, but although it was associated with lower rates of CP and developmental delay than in the group without TH, the only significant difference was for delay on the language subscale. CONCLUSIONS: Survivors of NHIE in SA are at risk of poor developmental outcomes.

Development of the FAST-M maternal sepsis bundle for use in low-resource settings: a modified Delphi process.

OBJECTIVE: To develop a sepsis care bundle for the initial management of maternal sepsis in low resource settings. DESIGN: Modified Delphi process. SETTING: Participants from 34 countries. POPULATION: Healthcare practitioners working in low resource settings (n = 143; 34 countries), members of an expert panel (n = 11) and consultation with the World Health Organization Global Maternal and Neonatal Sepsis Initiative technical working group. METHODS: We reviewed the literature to identify all potential interventions and practices around the initial management of sepsis that could be bundled together. A modified Delphi process, using an online questionnaire and in-person meetings, was then undertaken to gain consensus on bundle items. Participants ranked potential bundle items in terms of perceived importance and feasibility, considering their use in both hospitals and health centres. Findings from the healthcare practitioners were then triangulated with those of the experts. MAIN OUTCOME MEASURE: Consensus on bundle items. RESULTS: Consensus was reached after three consultation rounds, with the same items deemed most important and feasible by both the healthcare practitioners and expert panel. Final bundle items selected were: (1) Fluids, (2) Antibiotics, (3) Source identification and control, (4) Transfer (to appropriate higher-level care) and (5) Monitoring (of both mother and neonate as appropriate). The bundle was given the acronym 'FAST-M'. CONCLUSION: A clinically relevant maternal sepsis bundle for low resource settings has been developed by international consensus. TWEETABLE ABSTRACT: A maternal sepsis bundle for low resource settings has been developed by international consensus.

A review of antenatal corticosteroid use in premature neonates in a middle-income country.

BACKGROUND: Antenatal corticosteroid (ANS) use in premature neonates has become a standard of practice. However, there is low ANS coverage in low- to middle-income countries (LMICs). Recent studies have questioned the efficacy of ANSs in such countries. OBJECTIVE: To review the use of ANSs in preterm neonates at Charlotte Maxeke Johannesburg Academic Hospital (CMJAH), South Africa. METHODS: This was a retrospective observational study of all neonates with a birth weight of 500 - 1 800 g born at CMJAH between 1 January 2013 and 30 June 2016. Neonatal and maternal characteristics of neonates exposed to ANSs were compared with those of neonates who were not exposed. RESULTS: The ANS coverage of the final sample was 930/2 109 (44.1%). The mean (standard deviation (SD)) birth weight was 1 292.4 (323.2) g and the mean gestational age 30.2 (2.9) weeks. Attending antenatal care and maternal hypertension were associated with increased use of ANSs, whereas vaginal delivery was associated with decreased use. In neonates weighing <1 500 g, the use of ANSs was associated with decreased mortality, decreased intraventricular haemorrhage and decreased patent ductus arteriosus. There was no association between ANSs and respiratory distress syndrome, necrotising enterocolitis, sepsis or need for respiratory support in all premature neonates, and no association with improved outcomes in those weighing ≥1 500 g. CONCLUSION: The benefits of ANSs in terms of neonatal morbidity in this study were not as marked as those published in high-income countries. A randomised controlled trial may be indicated in LMICs.

Correlation between rivaroxaban (Xarelto) plasma activity, patient clinical variables and outcomes in a South African centre.

BACKGROUND: Low-molecular-weight heparin and vitamin K antagonists such as warfarin are the gold standard for prohylaxis and treatment of venous thromboembolic disease (VTED). Direct oral anticoagulants (DOACs) result in predictable anticoagulation with significantly reduced inter- and intra-patient variability. DOAC absorption is rapid, with a short half-life and relatively few drug interactions. DOACs are effective and safe at fixed doses without activity monitoring. However, specific situations may require assessment of accurate drug activity. Rivaroxaban, a DOAC targeting activated coagulation factor X (FXa), is registered for the prevention and treatment of VTED in South Africa. OBJECTIVES: To establish a prophylactic rivaroxaban activity level range and determine any associations with clinical complications, viz. haemorrhage and/or thrombosis. METHODS: Samples from 115 orthopaedic patients were tested 3 hours after a prophylactic oral dose of 10 mg rivaroxaban with STAGO rivaroxaban anti-FXa reagent on an automated coagulation analyser. Patient demographics and clinical outcomes were documented. RESULTS: The mean rivaroxaban anti-FXa level was 105.7 ng/mL. Two patients developed adverse events on therapy. One patient had minor bleeding (menorrhagia) (drug activity level 288.7 ng/mL) and another a deep-vein thrombosis (drug activity level 34.7 ng/mL). Statistical analysis demonstrated an association between drug activity and advancing age (p=0.008), most apparent among those aged ≥65 years. CONCLUSIONS: Measuring rivaroxaban activity levels reduces uncertainty if treatment failure and complications occur. Patients aged ≥65 years should be closely monitored. A local rivaroxaban activity level for patients on rivaroxaban prophylaxis has been established.

Hepatitis C genotype distribution in patient and blood donor samples in South Africa for the period 2008-2012.

There are limited molecular epidemiological studies of hepatitis C at a national level in South Africa. The introduction of newer treatment modalities for hepatitis C requires knowledge of the genotypes as these may have different prognostic and therapeutic implications. This retrospective study describes genotype distributions of patients attending specialist clinics and a blood donor group studied during the period 2008-2012 in South Africa. Residual samples from diagnostic viral load testing from specialist clinics in South Africa (n=941) and from the South African National Blood Service (n=294) were analysed quantitatively by real-time PCR and genotyped using the Versant line probe assay or sequencing. Genotype 1 was predominant in blood donors (34%), whilst genotype 5a was prevalent in patients (36%). In the blood donor group, genotype 4 was detected for the first time. Genotype 2 was rare in the patient group and not detected in blood donors. Genotype 1 was the predominant genotype in the younger age groups (less than 30 years), whereas genotype 5a was found at higher proportions in the older age groups for both the patient and blood donor groups, comprising more than 60% of genotypes in those older than 50 years. Genotypes 1 and 5 were at highest proportions across all provinces compared to other genotypes. In blood donors, genotype 1 was predominant among Caucasians (43%) and genotype 5a among Blacks (54%). Such information is required for planning the impact on the health sector with regard to newly emerging therapies for hepatitis C and burden of disease.

Glycaemic, blood pressure and cholesterol control in 25 629 diabetics.

OBJECTIVE: To examine and compare the extent to which people with type 2 diabetes (T2DM) are achieving haemoglobin A1c (HbA1c), blood pressure (BP) and LDL cholesterol (LDL-C) treatment targets. METHODS: A review of databases (MEDLINE Ovid, Pubmed and Sabinet) was performed and limited to the following terms: type 2 diabetes mellitus AND guideline AND goal achievement for the years 2009 to 2014 (five years). RESULTS: A total of 14 studies (25 629 patients) were selected across 19 different countries. An HbA1c level of 7.0% (or less) was achieved by 44.5% of subjects (range 19.2-70.5%), while 35.2% (range 7.4-66.3%) achieved BP of 130/80 mmHg (or less), and 51.4% (range 20.0-82.9%) had an LDL-C level of either 2.5 or 2.6 mmol/l (100 mg/dl or less). CONCLUSION: Despite guideline recommendations that lowering of HbA1c, BP and lipids to target levels in T2DM will lead to a reduction in morbidity and mortality rates, we found that control of these risk factors remains suboptimal, even across different settings.

Identifying high-risk areas for sporadic measles outbreaks: lessons from South Africa.

OBJECTIVE: To develop a model for identifying areas at high risk for sporadic measles outbreaks based on an analysis of factors associated with a national outbreak in South Africa between 2009 and 2011. METHODS: Data on cases occurring before and during the national outbreak were obtained from the South African measles surveillance programme, and data on measles immunization and population size, from the District Health Information System. A Bayesian hierarchical Poisson model was used to investigate the association between the risk of measles in infants in a district and first-dose vaccination coverage, population density, background prevalence of human immunodeficiency virus (HIV) infection and expected failure of seroconversion. Model projections were used to identify emerging high-risk areas in 2012. FINDINGS: A clear spatial pattern of high-risk areas was noted, with many interconnected (i.e. neighbouring) areas. An increased risk of measles outbreak was significantly associated with both the preceding build-up of a susceptible population and population density. The risk was also elevated when more than 20% of infants in a populous area had missed a first vaccine dose. The model was able to identify areas at high risk of experiencing a measles outbreak in 2012 and where additional preventive measures could be undertaken. CONCLUSION: The South African measles outbreak was associated with the build-up of a susceptible population (owing to poor vaccine coverage), high prevalence of HIV infection and high population density. The predictive model developed could be applied to other settings susceptible to sporadic outbreaks of measles and other vaccine-preventable diseases.

Adult deaths and the future: a cause-specific analysis of adult deaths from a longitudinal study in rural Tanzania 2003-2007.

OBJECTIVES: To determine patterns and risk factors for cause-specific adult mortality in rural southern Tanzania. METHODS: The study was a longitudinal open cohort and focused on adults aged 15-59 years between 2003 and 2007. Causes of deaths were ascertained by verbal autopsy (VA). Cox proportion hazards regression model was used to determine factors associated with cause-specific mortality over the 5-year period. RESULTS: Thousand three hundred and fifty-two of 65 548 adults died, representing a crude adult mortality rate (AMR) of 7.3 per 1000 person years of observation (PYO). VA was performed for 1132 (84%) deaths. HIV/AIDS [231 (20.4%)] was the leading cause of death followed by malaria [150 (13.2%)]. AMR for communicable disease (CD) causes was 2.49 per 1000 PYO, 1.21 per 1000 PYO for non-communicable diseases (NCD) and 0.53 per 1000 PYO for accidents/injury causes. NCD deaths increased from 16% in 2003 to 24% in 2007. High level of education was associated with a reduction in the risk of dying from NCDs. Those with primary education (HR = 0.67, 95% CI: 0.49, 0.92) and with education beyond primary school (HR = 0.11, 95% CI: 0.02, 0.40) had lower mortality than those who had no formal education. Compared with local residents, in-migrants were 1.7 (95% CI: 1.37, 2.11) times more likely to die from communicable disease causes. CONCLUSION: NCDs are increasing as a result of demographic and epidemiological transitions taking place in most African countries including Tanzania and require attention to prevent increased triple disease burden of CD, NCD and accident/injuries.

Bacterial vaginosis in female facility workers in north-western Tanzania: prevalence and risk factors.

OBJECTIVES: To determine prevalence of, and risk factors for, bacterial vaginosis (BV) among herpes simplex virus (HSV) 2 seropositive Tanzanian women at enrollment into a randomised, placebo-controlled trial of HSV suppressive treatment. METHODS: 1305 HSV-2 seropositive women aged 16-35 years working in bars, guesthouses and similar facilities were interviewed, examined and tested for HIV, syphilis, Neisseria gonorrhoeae, Chlamydia trachomatis, BV, candidiasis and trichomoniasis. Factors associated with BV were analysed using logistic regression to estimate odds ratios and 95% confidence intervals. RESULTS: BV prevalence was 62.9%; prevalence of Nugent score 9-10 was 16.1%. Independent risk factors for BV were work facility type, fewer dependents, increasing alcohol consumption, sex in the last week (adjusted OR 2.03; 95% CI 1.57 to 2.62), using cloths or cotton wool for menstrual hygiene, HIV (adjusted OR 1.41; 95% CI 1.09 to 1.83) and Trichomonas vaginalis infection. There was no association between BV and the frequency or method of vaginal cleansing. However, BV was less prevalent among women who reported inserting substances to dry the vagina for sex (adjusted OR 0.44; 95% CI 0.25 to 0.75). CONCLUSION: BV was extremely prevalent among our study population of HSV-2 positive female facility workers in North-western Tanzania. Although recent sex was associated with increased BV prevalence, vaginal drying was associated with lower BV prevalence. Further studies of the effects of specific practices on vaginal flora are warranted.

Performance of selected nonparametric tests for discrete longitudinal data under different patterns of missing data.

Comparison of changes over time of a continuous response variable between treatment groups is often of main interest in clinical trials. When the distributional properties of the continuous response variable are not regular enough, or when the response is discrete, nonparametric techniques have been used. The relative performances of selected repeated measures nonparametric two-sample tests proposed by Wei and Lachin, Koziol, Wei and Johnson, and the adapted Wilcoxon Rank-Sum test are compared through simulations based on quality of life data. The Wilcoxon Rank-Sum test is the most powerful and is not significantly affected by the different patterns of missing data.