Bill & Melinda Gates Foundation

Combination vaccines combine several components in a single dose administration. They offer programmatic and public health advantages, particularly as vaccine schedules become increasingly crowded. They are often more expensive to develop and produce, which discourages manufacturer investment without clear market signals. Hence their benefits need to be captured with existing health economic evaluation reference cases used by decision-makers to guide vaccine investments. We propose that the value of combination vaccines can be captured through at least four domains: (1) reductions in tangible and intangible costs to caregivers, (2) operational efficiencies to the health system, (3) opportunity costs of vaccine schedule slots, and (4) more streamlined vaccine schedules. We demonstrate the practicality of our framework by comparing the value of introducing a hypothetical vaccine to a crowded schedule as a standalone formulation, a replacement for a vaccine already in the schedule, or a combination product. The framework could also be applied to estimate the value of reducing the number of separate administrations needed for a standalone vaccine. Applying it in real-world situations could be facilitated by further data collection, particularly on collating results on the value of existing vaccines in the schedule and estimating willingness-to-pay for fewer vaccine administrations.

Randomized, placebo-controlled trials are widely accepted as the scientific reference standard to determine a causal association between an intervention and an outcome. They are pivotal for product licensure. Although such trials establish safety and efficacy with the assumption that bias has been eliminated, limitations remain. For vaccines, establishing the duration of protection and understanding how they perform in diverse populations or in terms of less common outcomes are important to inform use in clinical practice. Furthermore, randomized, controlled trials may be difficult or unethical when a vaccine is available and recommended. Thus, alternative methods are important adjuncts to traditional randomized, .

Invasive non-typhoidal Salmonella (iNTS) disease is a global health concern, particularly for sub-Saharan Africa. Despite high case fatality risks, there is no vaccine available against the disease. An obstacle to vaccine development is a lack of data on the economic burden of iNTS disease in many parts of the world. The main aim of the current study is to estimate the economic burden of iNTS disease in 123 countries. Several multivariate regression models were parameterised with the data obtained from an existing systematic literature review on the economic burden of iNTS disease and all forms of NTS disease. Various model diagnostics were performed to validate the statistical significance of each model outcome, and the most suitable model was selected to predict costs in countries where no data points were available. A generalised linear model with gamma distribution with log link was chosen based on model diagnostics. While the average economic burden per iNTS disease episode ranged from US$341 in Africa to US$2194 in Europe, the total economic burden of iNTS disease was the highest in Africa due to the high burden of the disease in the region. The current study indicates that the economic burden of iNTS disease is substantial. Given the scarcity of field-based iNTS disease economic burden estimates, cost extrapolation through an econometric framework can be helpful for understanding the associated cost implications in a resource-limited setting and for informing the cost-effectiveness of public health interventions including future vaccination strategies.

Background:Despite efforts to increase global immunization, vaccination seroconversion in low and middle income countries (LMICs) is often lower than in high income countries (HICs). The reasons for this disparity are not fully understood. Given the role of the gut microbiome in immune development, we investigated the relationship between the gut microbiome and polio vaccine seroconversion in children in the Democratic Republic of the Congo (DRC).Methods:This cross-sectional analysis included children ages 6-24 months old (n=90) enrolled in the DRC. Vaccine history was obtained from health records and/or self-report and poliovirus serostatus surveyed using dried blood spots. Nutritional status was evaluated using anthropomorphic measures. Microbiome profiling (16S rRNA gene) was performed and associations with poliovirus serostatus and malnutrition were examined.Results:The average age of the study population was 13.6 months (SD=5.6) with 58% female. Poliovirus seropositivity was 65.5% and 22% of the children were malnourished. We found that presence of Campylobacter and Veillonella, especially at an early age, was associated with low poliovirus vaccination seroconversion. These bacterial taxa differed from those associated with malnutrition.Conclusions:The presence of enteropathogens such as Campylobacter at a young age could be an important factor contributing to low vaccination seroconversion in children in LMICs.

IntroductionDeveloping and implementing trauma clinical guidance is integral to providing quality care to all trauma patients while maintaining a minimum standard of treatment. A mixed-methods novel consensus-building approach was used to identify the current barriers to developing and implementing trauma clinical guidance and highlight the priority areas for change to better support end users.MethodsAs part of year 1 of the Design for Implementation: The Future of Trauma Clinical Guidance and Research Conference Series, preconference participant surveys and hybrid, professionally facilitated, structured dialogue were used to define the ideal future state of trauma clinical guidance development and dissemination. Novel to this context, in-person and virtual “user stories”, a form of structured focus group, were generated, and a “minimum viable product” (MVP), a form of brokered dialogue, was developed. Descriptive statistics and thematic analysis were used to evaluate preconference survey and “user story” results.Results72 in-person and up to 35 virtual attendees participated. The majority (92%) of in-person attendees and nearly half (48%) of virtual attendees completed the preconference survey. Participants identified barriers along the continuum of clinical guidance development, dissemination, and adoption. Areas for improvement centered around the creation, storage, and use of guidance. Across the survey and user stories, participants expressed the need for clinical guidance that is comprehensive, evidence-based, coordinated, and easily accessible by all clinicians both domestically and abroad. The MVP targeted the risks and objectives to improved guidance. A prominent theme throughout this consensus-building assessment was the imperative for collaboration between professional societies for clinical guidance development and dissemination.DiscussionTrauma clinical guidance must be current, consolidated, and coordinated with patient-centered outcomes prioritized. Next steps include turning the MVP produced into a prototype and refining it to inform a national redesign of trauma clinical guidance.Level of evidenceLevel III.

Introduction Cervical cancer remains the fourth most common cancer among women globally, despite being preventable with the human papillomavirus (HPV) vaccine. However, HPV vaccine uptake remains a challenge in low- and middle-income countries (LMICs), where cervical cancer elimination faces significant delays. This study aims to identify the social determinants impacting HPV vaccine uptake in LMICs. Methods This systematic review and meta-analysis included studies published between 2010 and 2025, identified through PubMed, Google Scholar, and ScienceDirect. Eligible studies reported HPV vaccine uptake (initiation, completion, or both) among adolescent girls aged 9–19 and examined at least one individual- or household-level social determinant. Data were thematically synthesized, and a meta-analysis was conducted using the random-effects model, with results expressed as odds ratios (ORs), with 95% confidence intervals (CIs). Results Eight studies, conducted in Ethiopia, Tanzania, and Uganda, were included. Key determinants assessed included age, religion, residence, parental education, occupation, wealth index, marital status, and household factors. Meta-analyses revealed wealth index (OR = 1.34; 95% CI: 1.05–1.70; P = 0.02) and parental marital status (OR = 0.86; 95% CI: 0.78–0.95; P < 0.01) as significant predictors of HPV vaccine uptake among adolescent girls in LMICs. Other factors, such as age, residence, parental education, etc., showed inconsistent effects or no significant association, with high heterogeneity across studies limiting the generalizability of some findings. Conclusions This review highlights the complex, context-specific individual and household factors influencing HPV vaccine uptake among adolescent girls in LMICs. While wealth index and parental marital status showed consistent associations, other factors varied across studies. Community-based, culturally sensitive, tailored interventions are critical to improve the vaccine uptake. Continued research with standardized mixed-methods is vital to address multilevel factors and ensure equitable HPV vaccine uptake in LMICs.

Abstract Efforts to improve maternal health in the Democratic Republic of Congo (DRC) have often focused on expanding access and improving structural readiness. These are critical first steps, but this analysis reveals a critical missing link: women’s perceptions of care—including how they are treated, how much they trust providers, and how affordable care seems—which are more predictive of health-seeking behavior than objective facility infrastructure or actual cost. Drawing on two rounds of household and facility surveys, we find that care seeking is strongly impacted by perceptions. Women’s perceptions of quality was defined by a composite index that captures interpersonal experiences (e.g., respectful treatment, clear communication, and provider competence), alongside perceptions of facility conditions like cleanliness and drug availability. In regression models, we found that perceived quality was more predictive of care-seeking than structural readiness. Women who perceived care as good were significantly more likely to initiate antenatal care, return for postnatal visits, and seek help for complications—even in under-resourced facilities. These findings highlight the individual-dependent nature of health system interaction: perceived quality modulates behavior even within the same infrastructure. Similarly, perceptions of affordability better predicted care utilization than objective financial hardship, operating through household coping strategies like asset sales. The type of asset sold mattered: selling food reserves was linked to earlier care but reduced continuity, while selling productive assets enabled sustained engagement. These patterns highlight women’s agency and that affordability is shaped by expectations, vulnerability, and risk assessment. Health financing models also shaped these dynamics. Performance-based financing (PBF) drove improvements in centrally-incentivized services such as vaccination and laboratory testing, while direct facility financing (DFF) enabled more responsive delivery for high-burden local priorities like malaria care. Crucially, perceptions had a stronger influence on utilization in DFF settings, where facility responsiveness varied more. This suggests that in more adaptive systems, interpersonal trust and perceived quality play a greater role in guiding behavior, highlighting how systems-dependent factors mediate how care is delivered and how patients respond to it. Improving maternal health in fragile settings requires re-centering perceptions as core indicators of system performance. Respect, clarity, cleanliness, and competence must not only be measured but actively addressed. Financing reforms should empower frontline providers to respond to patient feedback, not just meet centrally imposed targets. And affordability interventions must reflect behavioral realities: what feels affordable is shaped by hardship, expectations, and perceived benefit. Measuring household coping strategies like selling assets can offer insight into the how to mitigate the sacrifices households make to access care and suggest interventions. Contrary to traditional models of primary healthcare investment, far more attention must be placed on the drivers of perception—friendliness, affordability, trustworthiness—if we aim to expand coverage. Perception data are not mere satisfaction scores; they are proxies for trust, predictors of use, and a missing link in health systems that aspire to be people centered.

SummaryBackgroundStudies have consistently demonstrated beneficial effects of intermittent preventive treatment during pregnancy (IPTp) with sulfadoxine-pyrimethamine (SP) on reducing malaria infection and improving birth outcomes among pregnant women in endemic areas. However, data on its impact on maternal gestational weight gain (GWG) are very limited. We aimed to conduct a two-stage meta-analysis of individual participant data to examine the effect of IPT with SP on GWG compared to other antimalarial regimens.MethodsIn this systematic review and individual participant data meta-analysis, we conducted electronic literature searches of PubMed, Embase, Web of Science, and the Cochrane Library to identify eligible RCTs among pregnant women. We did not apply any language or publication date restrictions in the search. The initial search was conducted on August 4th, 2021, and updated on February 15th, 2025. The study-level inclusion criteria were as follow: 1) the studies must be randomised controlled trials (RCTs), which could be individually randomised, cluster randomised, or a combination of both; 2) study participants were pregnant at enrollment or enrolled before pregnancy and followed up in pregnancy; 3) studies were conducted in a low-income, lower-middle-income, or upper-middle-income economy defined by the World Bank country classification for the 2021 fiscal year; 4) antimalaria and/or antibiotic interventions were provided during pregnancy; and 5) the intervention was provided alone or in combination with a co-intervention that was similar across arms. Since we focused on the intervention's effect on GWG in generally healthy pregnant women, we applied the following study-level exclusion criteria: 1) studies without any measures of maternal weight during pregnancy; and 2) studies conducted exclusively among women with pre-existing health conditions, such as anemia, human immunodeficiency virus (HIV) infection, or diabetes. Within each eligible trial, we further applied individual-level criteria to identify eligible individual participants, including 1) singleton pregnancies, 2) at least one weight measurement in the second or third trimesters, 3) known gestational ages at the time of weight measurements, and 4) availability of maternal height measure. Risk of bias for each trial was assessed using the Cochrane risk-of-bias tool, version 2 (RoB 2). GWG percent adequacy (%) and total weight gain (gram) at delivery were calculated according to the Institute of Medicine 2009 guidelines. Linear regression models were used to estimate mean difference (MD) and 95% confidence intervals (CIs) in GWG percent adequacy and total weight gain across intervention arms. Results from individual trials were pooled using fixed-effects inverse-variance meta-analysis models. This study is registered with PROSPERO, CRD42023428794.FindingsA total of 97 trials were identified in the search and sough for IPD, of them eight trials including 8550 pregnant women were included in the current analysis. Women who received IPTp with only 2 doses of SP had a greater GWG percent adequacy (MD: 5.61%; 95% CI: 2.61%, 8.60%; P = 0.0002; I2 = 84.26%), and total GWG in grams at delivery (MD: 702; 95% CI: 321, 1083; P = 0.0003; I2 = 83.78%) than those who received weekly chloroquine as prophylaxis. No significant differences in GWG percent adequacy (MD: −0.53%; 95% CI: −2.89%%, 1.83%; P = 0.66; I2 = 0.00%) or GWG grams (MD: −80; 95% CI: −380, 221; P = 0.60; I2 = 0.00%) were found between IPTp with 2-dose SP and monthly IPTp-SP (3-dose or more). Compared to women who received monthly IPTp-SP, those who received monthly IPTp with dihydroartemisinin-piperaquine (IPTp-DHA + PPQ) had a lower GWG percent adequacy (MD: −5.56%; 95% CI: −8.22%, −2.90%; P < 0.0001; I2 = 13.47%) and total GWG in grams (MD: −723; 95% CI: −1037, −410; P < 0.0001; I2 = 46.29%). Adding azithromycin to an antimalarial regimen was associated with a greater GWG percent adequacy (MD: 2.75%; 95% CI: 0.46%, 5.05%; P = 0.19; I2 = 0.00%) and total GWG in gram at delivery (MD: 485; 95% CI: 210, 760; P = 0.0005; I2 = 75.66%).InterpretationOur findings suggest that monthly IPTp-SP has superior effect on GWG compared to weekly chloroquine or IPTp-DHA + PPQ in malaria-endemic areas. The result provides further evidence indicating that IPTp-SP improves maternal weight gain, an important determinant of fetal growth beyond its antimalarial effects. Due to the limited number of trials with weight and height measures available for the IPD meta-analysis we were likely underpowered to detect any significant difference between 2-dose SP and monthly IPTp-SP. More efforts are warranted to examine the potential beneficial effect of adding azithromycin or DHA + PPQ to the standard antimalarial regimens.FundingGates Foundation.

Large Language Models (LLMs) have shown remarkable performance across various tasks, yet significant disparities remain for non-English languages, and especially native African languages. This paper addresses these disparities by creating approximately 1 million human-translated words of new benchmark data in 8 low-resource African languages, covering a population of over 160 million speakers of: Amharic, Bambara, Igbo, Sepedi (Northern Sotho), Shona, Sesotho (Southern Sotho), Setswana, and Tsonga. Our benchmarks are translations of Winogrande and three sections of MMLU: college medicine, clinical knowledge, and virology. Using the translated benchmarks, we report previously unknown performance gaps between state-of-the-art (SOTA) LLMs in English and African languages. Finally, using results from over 400 fine-tuned models, we explore several methods to reduce the LLM performance gap, including high-quality dataset fine-tuning (using an LLM-as-an-Annotator), cross-lingual transfer, and cultural appropriateness adjustments. Key findings include average mono-lingual improvements of 5.6% with fine-tuning (with 5.4% average mono-lingual improvements when using high-quality data over low-quality data), 2.9% average gains from cross-lingual transfer, and a 3.0% out-of-the-box performance boost on culturally appropriate questions. The publicly available benchmarks, translations, and code from this study support further research and development aimed at creating more inclusive and effective language technologies.