Incorporating respondent-driven sampling into web-based discrete choice experiments: preferences for COVID-19 mitigation measures

Health Services and Outcomes Research Methodology - To slow the spread of COVID-19, most countries implemented stay-at-home orders, social distancing, and other nonpharmaceutical mitigation...     

Antenatal care and pregnancy outcomes in a safe motherhood health voucher system in rural Kenya, 2007–2013

Setting: A rural private health facility, Ruby Medical Centre (RMC), participating in a safe motherhood health voucher system for poor women in Kiambu County, Kenya.Objectives: Between 2007 and 2013, to determine 1) the number of women who delivered at the RMC, their characteristics and pregnancy-related outcomes, and 2) the number of women who received an incomplete antenatal care (ANC) package and associated factors.Design: Retrospective cross-sectional study using routine programme data.Results: During the study period, 2635 women delivered at the RMC: 50% were aged 16–24 years, 60% transferred in from other facilities and 59% started ANC in the third trimester of pregnancy. Of the 2635 women, 1793 (68%) received an incomplete ANC package: 347 (13%) missed essential blood tests, 312 (12%) missed the tetanus toxoid immunisation and 1672 (65%) had fewer than four visits. Presenting late and starting ANC elsewhere were associated with an incomplete package. One pregnancy-related mortality occurred; the stillbirth rate was 10 per 1000 births.Conclusion: This first assessment of the health voucher system in rural Kenya showed problems in ANC quality. Despite favourable pregnancy-related outcomes, increased efforts should be made to ensure earlier presentation of pregnant women, comprehensive ANC, and more consistent and accurate monitoring of reproductive indicators and interventions.     

Classification using ensemble learning under weighted misclassification loss

Binary classification rules based on covariates typically depend on simple loss functions such as zero-one misclassification. Some cases may require more complex loss functions. For example, individual-level monitoring of HIV-infected individuals on antiretroviral therapy requires periodic assessment of treatment failure, defined as having a viral load (VL) value above a certain threshold. In some resource limited settings, VL tests may be limited by cost or technology, and diagnoses are based on other clinical markers. Depending on scenario, higher premium may be placed on avoiding false-positives, which brings greater cost and reduced treatment options. Here, the optimal rule is determined by minimizing a weighted misclassification loss/risk. We propose a method for finding and cross-validating optimal binary classification rules under weighted misclassification loss. We focus on rules comprising a prediction score and an associated threshold, where the score is derived using an ensemble learner. Simulations and examples show that our method, which derives the score and threshold jointly, more accurately estimates overall risk and has better operating characteristics compared with methods that derive the score first and the cutoff conditionally on the score especially for finite samples.     

Effect of phenobarbital on seizure frequency and mortality in childhood cerebral malaria: a randomised, controlled intervention study

This randomized, placebo-controlled study assesses whether a single intramuscular dose of phenobarbital (20 mg/kg) given on admission to Kenyan children with cerebral malaria could lower the frequency of seizures, which complicate cerebral malaria by increasing the risk of death and neurological sequelae. A total of 340 children with cerebral malaria were admitted to the hospital; 170 received phenobarbital and 170 received placebo. The drug was adequately absorbed and well tolerated. Findings revealed a significantly lower frequency of seizures in the phenobarbital group than in the placebo group (18 [11%] vs. 46 [27%] children had 3 or more seizures of any duration; odds ratio (OR), 0.32; 95% confidence interval (CI), 0.18-0.58). However, mortality was doubled (30 [18%] vs. 14 [8%] deaths; OR, 2.39; 95% CI, 1.28-4.64) in the phenobarbital group. In addition, the frequency of respiratory arrest was higher in the phenobarbital group than in the placebo group, and mortality was greatly increased in children who received phenobarbital plus 3 or more doses of diazepam. In conclusion, although the phenobarbital dose of 20 mg/kg given to children with cerebral malaria provides highly effective seizure prophylaxis, an unacceptable increase in mortality is noted; hence, use of this dosage is not recommended.     

Clinical algorithms for malaria diagnosis lack utility among people of different age groups

We conducted a study to determine whether clinical algorithms would be useful in malaria diagnosis among people living in an area of moderate malaria transmission within Kilifi District in Kenya. A total of 1602 people of all age groups participated. We took smears and recorded clinical signs and symptoms (prompted or spontaneous) of all those presenting to the study clinic with a history of fever. A malaria case was defined as a person presenting to the clinic with a history of fever and concurrent parasitaemia. A set of clinical signs and symptoms (algorithms) with the highest sensitivity and specificity for diagnosing a malaria case was selected for the age groups /=15 years. These age-optimized derived algorithms were able to identify about 66% of the cases among those <15 years of age but only 23% of cases among adults. Were these algorithms to be used as a basis for a decision on treatment among those presenting to the clinic, 16% of children /=5000 parasites/microl of blood would be sent home without treatment. Clinical algorithms therefore appear to have little utility in malaria diagnosis, performing even worse in the older age groups, where avoiding unnecessary use of anti-malarials would make more drugs available to the really needy population of children under 5 years of age.