Estimating the impact of alternative programmatic cotrimoxazole strategies on mortality among children born to mothers with HIV: A modelling study

Shrey Mathur; Melanie Smuk; Ceri Evans; Catherine J Wedderburn; Diana M Gibb; Martina Penazzato; Andrew J Prendergast

doi:10.1371/journal.pmed.1004334

. 2024 Feb 20;21(2):e1004334. doi: 10.1371/journal.pmed.1004334

Estimating the impact of alternative programmatic cotrimoxazole strategies on mortality among children born to mothers with HIV: A modelling study

Shrey Mathur ^1,^*, Melanie Smuk ¹, Ceri Evans ^1,^2,³, Catherine J Wedderburn ^4,⁵, Diana M Gibb ⁴, Martina Penazzato ⁶, Andrew J Prendergast ^1,²

Editor: Marie-Louise Newell⁷

PMCID: PMC10914273 PMID: 38377150

Abstract

Background

World Health Organization (WHO) guidelines recommend cotrimoxazole prophylaxis for children who are HIV-exposed until infection is excluded and vertical transmission risk has ended. While cotrimoxazole has benefits for children with HIV, there is no mortality benefit for children who are HIV-exposed but uninfected, prompting a review of global guidelines. Here, we model the potential impact of alternative cotrimoxazole strategies on mortality in children who are HIV-exposed.

Methods and findings

Using a deterministic compartmental model, we estimated mortality in children who are HIV-exposed from 6 weeks to 2 years of age in 4 high-burden countries: Côte d’Ivoire, Mozambique, Uganda, and Zimbabwe. Vertical transmission rates, testing rates, and antiretroviral therapy (ART) uptake were derived from UNAIDS data, trial evidence, and meta-analyses. We explored 6 programmatic strategies: maintaining current recommendations; shorter cotrimoxazole provision for 3, 6, 9, or 12 months; and starting cotrimoxazole only for children diagnosed with HIV.

Modelled alternatives to the current strategy increased mortality to varying degrees; countries with high vertical transmission had the greatest mortality. Compared to current recommendations, starting cotrimoxazole only after a positive HIV test had the greatest predicted increase in mortality: Mozambique (961 excess annual deaths; excess mortality 339 per 100,000 HIV-exposed children; risk ratio (RR) 1.06), Uganda (491; 221; RR 1.04), Zimbabwe (352; 260; RR 1.05), and Côte d’Ivoire (125; 322; RR 1.06). Similar effects were observed for 3-, 6-, 9-, and 12-month strategies. Increased mortality persisted but was attenuated when modelling lower cotrimoxazole uptake, smaller mortality benefits, higher testing coverage, and lower vertical transmission rates. The study is limited by uncertain estimates of cotrimoxazole coverage in programmatic settings; an inability to model increases in mortality arising from antimicrobial resistance due to limited surveillance data in sub-Saharan Africa; and lack of a formal health economic analysis.

Conclusions

Changing current guidelines from universal cotrimoxazole provision for children who are HIV-exposed increased predicted mortality across the 4 modelled high-burden countries, depending on test-to-treat cascade coverage and vertical transmission rates. These findings can help inform policymaker deliberations on cotrimoxazole strategies, recognising that the risks and benefits differ across settings.

Mathur and colleagues estimate the potential impact of alternative cotrimoxazole strategies on mortality in children born to mothers with HIV in Côte d’Ivoire, Mozambique, Uganda, and Zimbabwe.

Author summary

Why was this study done?

Cotrimoxazole prophylaxis is recommended in World Health Organization (WHO) guidelines for all children born to mothers with HIV until HIV infection has been excluded by an age-appropriate HIV test to establish the final diagnosis after complete cessation of breastfeeding.
Though there is a proven mortality benefit for children who acquire HIV, recent trial evidence has shown that cotrimoxazole does not reduce mortality for majority of children who are HIV-exposed uninfected (HEU), which has led to countries considering changing their guidelines.
In many resource-limited settings, however, it is difficult to reliably distinguish children with HIV from children who are HEU, due to incomplete coverage of early infant diagnosis (EID) of HIV.
There is a need to model to what extent alternative cotrimoxazole strategies, which either do not provide universal cotrimoxazole for all infants who are HIV-exposed, or provide it for a shorter duration, would be predicted to increase mortality in different settings among infants who acquire HIV but are undiagnosed.

What did the researchers do and find?

This study uses mathematical modelling based on epidemiological data from 4 high-burden settings (Côte d’Ivoire, Mozambique, Uganda, and Zimbabwe) to estimate the effect on mortality of alternative programmatic cotrimoxazole strategies.
The model incorporates the HIV status of the infant, perinatal and postnatal transmission rates, testing rates, and mortality benefits from trial evidence for cotrimoxazole and antiretroviral therapy (ART) across 6 different programmatic strategies: maintaining current recommendations; reducing the duration of cotrimoxazole provision to 3, 6, 9, or 12 months; or starting cotrimoxazole only once a child tests positive for HIV.
We demonstrate that changing the current strategy is predicted to increase mortality in all 4 settings, with the greatest increase in mortality in countries with the highest vertical transmission rates.
Increased predicted mortality persisted in sensitivity analyses considering conservative model estimates, although cotrimoxazole had fewer predicted benefits when vertical transmission rates were lowered, testing coverage improved or uptake of cotrimoxazole was reduced.

What do these findings mean?

Changing the current strategy of cotrimoxazole provision for all children born to mothers with HIV is estimated to increase mortality in these 4 high-burden settings to varying degrees as countries continue to scale up prevention of mother-to-child transmission (PMTCT) of HIV and EID coverage.
Cotrimoxazole continues to provide important protection to children who acquire HIV and are missed by gaps in the test-to-treatment cascade, but does not replace the importance of timely testing and treatment.
Our study is limited by lack of cost-effectiveness analysis, lack of data on cotrimoxazole uptake, and limited antimicrobial resistance surveillance data in sub-Saharan Africa.
Policymakers need to weigh the risks and benefits of cotrimoxazole prophylaxis through any change to current recommendations, noting that these differ across settings: where lower vertical transmission rates and improved testing and treatment uptake occurs, the estimated mortality benefits of cotrimoxazole are attenuated.

Introduction

Vertical transmission of HIV occurs during pregnancy, labour, delivery, and breastfeeding. Despite substantial reductions in vertical transmission globally due to increased uptake of antiretroviral therapy (ART) among pregnant and breastfeeding women, transmission rates remain at 9% for Eastern and Southern Africa and 21% for Western and Central Africa [1]. Poor uptake of prevention of mother-to-child transmission (PMTCT) interventions arises due to late antenatal booking, low rates of HIV testing and ART initiation, and disengagement with services during pregnancy or breastfeeding [2]. Infants acquiring HIV have rapid disease progression, with over 50% mortality by 2 years of age without infant ART [3]. Given ongoing high transmission rates, limited PMTCT uptake, and rapid progression in undiagnosed children, programmatic strategies are needed to reduce morbidity and mortality among children born to mothers with HIV, until elimination of vertical transmission is achieved.

Cotrimoxazole is an inexpensive, well-tolerated, broad-spectrum antibiotic, which is recommended in World Health Organization (WHO) guidelines for all infants born to mothers with HIV from age 4 to 6 weeks until the end of HIV-exposure—typically defined as the end of breastfeeding, 18 months of age, or conclusive determination of being HIV-free [4,5]. There is strong evidence of mortality benefit from receiving cotrimoxazole among children with HIV. The CHAP trial demostrated a 43% reduction in mortality in ART-naïve children, and the ARROW trial showed that benefits of cotrimoxazole persist despite immune reconstitution on long-term ART [6,7]. Conversely, there is no evidence of mortality benefit from cotrimoxazole among children who are HIV-exposed but uninfected (HEU) [8,9]. Furthermore, there are concerns about antibiotic resistance and microbiome dysbiosis with cotrimoxazole use [10]. Given the distinct cotrimoxazole strategy required in each group, there is a critical need to identify and distinguish children with HIV from children who are HEU across different periods of transmission.

HIV testing for children has improved substantially over the past 2 decades. However, only 62% of infants globally currently receive the recommended virological test within 2 months of birth [1]. Overall, 40% of children with HIV therefore remain undiagnosed [11]. Even where infants are tested, challenges remain across the testing-to-treatment cascade. These include lengthy turnaround times for results, lack of integration with child health services, loss to follow-up, and delayed ART initiation [12,13]. Vertical transmission through prolonged breastfeeding now accounts for over half of transmissions in some settings and necessitates longitudinal testing [2]. WHO guidelines encourage testing of HIV-exposed children at 9 months of age and after complete cessation of breastfeeding [4].

A recent systematic review highlighted the lack of mortality benefit from cotrimoxazole for children who are HEU [14]. However, in settings with insufficient PMTCT and inadequate early infant diagnosis (EID) programmes, there is a substantial population of undiagnosed infants with HIV, who may benefit from cotrimoxazole prophylaxis to reduce mortality. In other settings, where PMTCT uptake and EID coverage are high and the majority of infants are HEU, the risk-benefit balance of cotrimoxazole may differ. Therefore, given these competing concerns, the optimal duration of cotrimoxazole prophylaxis needs to be determined. Here, we model the mortality impact of alternative cotrimoxazole strategies in 4 high-burden settings.

Methods

Model structure

We constructed a deterministic compartmental model with a decision tree to compare alternative cotrimoxazole prophylaxis strategies for children who are HIV-exposed. The model aimed to mimic epidemics in 4 high-burden countries: Zimbabwe (ZWE), Côte d’Ivoire (CIV), Mozambique (MOZ), and Uganda (UGA). These countries represent differing sub-Saharan African contexts and have sufficient information to enable estimates for the model parameters. The model structure and all estimates or assumptions applied are shown in Fig 1 and Table A in S1 Text. The study did not have a prospective protocol. This study is reported as per the transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD) guideline (S1 Checklist).

Fig 1 — **Panel A (Scenarios):** Scenarios considered were (1) children with HIV (diagnosed), (2) children with HIV (undiagnosed), or (3) HEU. Superscripts denote applied mortality estimates: (a) age-specific baseline mortality estimates for infants with HIV and no ART, based on Newell and colleagues [3]. (b) Age-specific baseline mortality estimates for HEU infants based on Arikawa and colleagues [15], (c) 77% uptake of ART following positive HIV test based on Luo and colleagues [17], (d) 76% mortality reduction from ART based on CHER [18], (e) 43% mortality reduction from CTX in a child with HIV based on CHAP [7], (f) 0% mortality reduction from CTX in an HEU child based on Botswana and South Africa trials [8,9]. **Panel B (Strategies)**: Illustration of considered programmatic strategies. Pink represents period of programmatic CTX provision. Blue represents period of risk of HIV transmission due to breastfeeding (18 months). Strategies illustrated are (1) current WHO strategy (2) initiating CTX at 6 weeks and stopping at 12 months, (3) initiating CTX at 6 weeks and stopping at 9 months, (4) initiating CTX at 6 weeks and stopping at 6 months, (5) initiating CTX at 6 weeks and stopping at 3 months, and (6) initiating CTX prophylaxis only once a positive HIV test result is confirmed and the child is started on ART. CTX, cotrimoxazole; ART, antiretroviral therapy; EID, early infant diagnosis at 6 weeks; 9mo, 9-month HIV test; EoB, end of breastfeeding; HEU, HIV-exposed uninfected.

The model estimates mortality from 6 weeks of age, when cotrimoxazole initiation is recommended, to 24 months of age. Testing time points were assigned at 6 weeks (defined as EID), at 9 months, and at 18 months (defined as end of breastfeeding, EoB). At 6 weeks, infants are categorised according to whether they have acquired perinatal HIV and whether they undergo EID. Children who are HIV-free at 6 weeks may acquire HIV through breastfeeding and may or may not receive additional HIV testing at 9 months and end of breastfeeding. Cotrimoxazole was provided based on programmatic strategy (Fig 1B). If the child was diagnosed with HIV, cotrimoxazole was started in conjunction with ART. ART was initiated only after a positive HIV test and was not initiated after negative or missed tests (i.e., for undiagnosed cases). Disease progression was incorporated through survival probabilities for children with HIV and children who are HEU, extracted from trial data and natural history studies [3,15]. Mortality occurs at each stage of the model and depends on current age, HIV status, ART use, and cotrimoxazole use. We defined model parameters for HIV transmission, coverage of EID and ART uptake using epidemiological data, as outlined below, while the effects of cotrimoxazole and ART on mortality were derived from trials and meta-analyses (Table A in S1 Text) [2,3,15–18].

Compartments were created to compare strategies of receiving cotrimoxazole: (1) from age 6 weeks to end of breastfeeding (current WHO strategy; base case); (2) from 6 weeks to 12 months; (3) from 6 weeks to 9 months; (4) from 6 weeks to 6 months; (5) from 6 weeks to 3 months; and (6) only once a positive HIV test result is confirmed. In all strategies, we assume that infants testing positive for HIV will start cotrimoxazole.

The primary outcome of deaths among children who are HIV-exposed from age 6 weeks to 2 years is expressed as (1) percentage mortality; (2) risk ratio (RR) compared to the current strategy; (3) excess deaths per year relative to the current strategy; and (4) excess annual deaths per 100,000 compared to the current strategy. The number of infants who are HIV-exposed in each country was calculated based on the number of live births and proportion of mothers living with HIV [19].

Data sources and parameters

We included parameters to reflect child HIV status based on UNAIDS perinatal and postnatal transmission rates. The probability of an HIV test yielding a positive result was based on age-specific transmission rates, obtained from country-specific UNAIDS estimates from the most recent year with available data for all countries, including six-week transmission rates (MOZ 6%, ZWE 5%, CIV 4%, and UGA 3%) [2]. For postnatal transmission, we calculated the cumulative probability of acquiring HIV by age 9 months (MOZ 3.4%, ZWE 1.7%, CIV 1.7%, UGA 1.3%), and by the end of breastfeeding (MOZ 4.1%, ZWE 2.0%, CIV 2.1%, UGA 1.6%), based on derived weekly transmission rates from country-specific final vertical transmission rates and breastfeeding duration, making the assumption that postnatal transmission rates were consistent over time [16].

EID rates at 6 weeks were derived from country-specific UNAIDS estimates (MOZ 82.9%, ZWE 75.9%, UGA 66.2%, CIV 60.8%) [16]. Since data on subsequent testing are not available, we assumed that 80% of children undergoing EID would have a 9-month test and 50% would have an EoB test, as these children would likely be engaged in the test-to-treatment cascade. For children not undergoing EID at 6 weeks, we assumed the probability of subsequent testing was much lower (10% for 9-month test, 30% for EoB), as these children were likely less engaged with services. We derived the proportion of children starting ART after a positive HIV test result based on point-of-care and standard-of-care testing coverage and corresponding ART uptake from a recent meta-analysis [17].

Scenarios

To estimate mortality, we considered several scenarios: (1) children with HIV diagnosed by EID; (2) children with HIV not yet diagnosed (due to missed 6-week, 9-month, and/or EoB tests) or with postnatal acquisition after an earlier negative test; and (3) children who are HIV-exposed and uninfected.

In the first scenario, children with known positive HIV status after EID started cotrimoxazole and 77% received ART based on a recent meta-analysis [17]. We applied a 43% mortality reduction following cotrimoxazole initiation based on results of the CHAP trial [7], and a 76% mortality reduction following ART initiation based on the CHER trial, to underlying mortality estimates [3,18].

In the second scenario, children had undiagnosed HIV and did not therefore receive ART. We based the survival probability on a pooled analysis of natural history studies conducted prior to the availability of ART [3] and applied a 43% mortality reduction due to cotrimoxazole alone. With the same rationale as in the first scenario, for 77% of children diagnosed with HIV (by 9 month or EoB testing), we applied mortality reduction of 76% from ART and 43% from cotrimoxazole.

In the third scenario, we derived mortality estimates for children who are HEU from an individual pooled analysis [15] and applied no mortality reduction due to cotrimoxazole, based on trial data [8,9].

Sensitivity analyses

To explore uncertainties in the model assumptions, we conducted sensitivity analyses by adjusting estimates of cotrimoxazole uptake, mortality reduction from cotrimoxazole, vertical transmission rates (perinatal and postnatal), and EID testing rates. For cotrimoxazole, we considered the effect of lower uptake (40%, 60%, and 80% of HIV-exposed children starting cotrimoxazole at age 6 weeks) compared to 100% uptake in the base model. For mortality effects of cotrimoxazole, we altered the base case assumption of 43% mortality reduction, derived from the CHAP trial, which compared cotrimoxazole and placebo among hospitalised children with HIV in Zambia. Since CHAP included very few infants, meaning mortality reductions are uncertain in children <12 months old, we used conservative estimates of 15%, 20%, 25%, 30%, 35%, 40% mortality reduction from cotrimoxazole to test a range of assumptions. To consider future progress in the prevention of MTCT, we reduced perinatal MTCT in 1% decrements to 0% and, separately, reduced postnatal MTCT rates in 0.5% decrements to 0%. Finally, we increased EID coverage in 10% increments, from current country rates up to 100%, to estimate the mortality benefits of cotrimoxazole with improved testing coverage over time. In a final sensitivity analysis, we combined the variables above to create a conservative estimate of effects, assuming low levels of cotrimoxazole uptake (40%), low ascribed mortality reduction from cotrimoxazole (15%), low perinatal (1%) and postnatal (1%) vertical transmission, and optimal EID coverage (100%).

Data access and ethics

The model was constructed in Microsoft Excel, is freely available at https://osf.io/8kjgp/, and can be adapted for other settings. No ethical approval was sought for this modelling study.

Results

Modelling the current WHO strategy, predicted mortality was 5.58% for Zimbabwe, 5.59% for Côte d’Ivoire, 5.78% for Mozambique, and 5.39% for Uganda. Limiting cotrimoxazole exposure from 6 weeks to 3 months of age, resulted in higher predicted mortality in each setting compared to the current strategy: Zimbabwe (increase in predicted mortality from 5.58% to 5.83%, RR 1.04, predicted excess mortality rate 246 deaths per 100,000 infants HIV-exposed, or 334 excess annual deaths), Côte d’Ivoire (from 5.59% to 5.90%, RR 1.05, 304 deaths per 100,000, 118 excess deaths), Mozambique (from 5.78% to 6.11%, RR 1.06, 328 deaths per 100,000, 928 excess deaths), and Uganda (from 5.39% to 5.60%, RR 1.04, 209 deaths per 100,000, 466 excess deaths); Table 1 and Fig 2. Similar effects were observed for 6-, 9-, and 12-month strategies: In each case, predicted infant mortality increased compared to the current policy, with the number of excess deaths lower with longer durations of cotrimoxazole prophylaxis (Table 1 and Fig 2).

Table 1. Comparison of predicted mortality, between ages 6 weeks to 2 years, with different programmatic cotrimoxazole strategies for children born to mothers with HIV.

For the current WHO strategy (base case) of providing cotrimoxazole (CTX) to all HIV-exposed children, mortality rate is expressed in percent (%). Alternative strategies explored were (1) initiating cotrimoxazole at 6 weeks and stopping at 12 months; (2) initiating cotrimoxazole at 6 weeks and stopping at 9 months; (3) initiating cotrimoxazole at 6 weeks and stopping at 6 months; (4) initiating cotrimoxazole at 6 weeks and stopping at 3 months; and (5) initiating cotrimoxazole prophylaxis only once a positive HIV test result is confirmed. Mortality for alternative strategies are reported in comparison to the base case (current strategy): risk ratio, excess mortality rate per 100,000 HIV-exposed infants, and excess annual deaths. CTX = cotrimoxazole.

		Programmatic strategy
		Current	Alternate
		Current	CTX from 6 weeks to 12 months	CTX from 6 weeks to 9 months	CTX from 6 weeks to 6 months	CTX from 6 weeks to 3 months	Start CTX only after positive HIV test result
Mozambique	Mortality rate (%)	5.78	5.91	5.95	6.07	6.11	6.12
	Risk ratio	-	1.02	1.03	1.05	1.06	1.06
	Excess mortality rate (per 100,000)	-	131	174	295	328	339
	Excess deaths	-	371	492	835	928	961
Cote d’Ivoire	Mortality rate (%)	5.59	5.72	5.76	5.85	5.90	5.91
	Risk ratio	-	1.02	1.03	1.05	1.06	1.06
	Excess mortality rate (per 100,000)	-	127	172	254	304	322
	Excess deaths	-	49	67	98	118	125
Zimbabwe	Mortality rate (%)	5.58	5.68	5.71	5.79	5.83	5.84
	Risk ratio	-	1.02	1.02	1.04	1.04	1.05
	Excess mortality rate (per 100,000)	-	99	134	208	246	260
	Excess deaths	-	135	181	281	334	352
Uganda	Mortality rate (%)	5.39	5.48	5.51	5.57	5.60	5.61
	Risk ratio	-	1.02	1.02	1.03	1.04	1.04
	Excess mortality rate (per 100,000)	-	87	117	177	209	221
	Excess deaths	-	193	261	394	466	491

Open in a new tab

Fig 2 — Columns represent additional deaths from each alternate strategy in comparison to the current WHO programmatic strategy of providing cotrimoxazole to all HIV-exposed infants. 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months.

A programmatic strategy of no routine prophylaxis, and instead initiating cotrimoxazole only once HIV is diagnosed, led to the greatest increase in predicted mortality across settings: Zimbabwe (increase in predicted mortality from 5.58% to 5.84%, RR 1.05, predicted excess mortality rate 260 deaths per 100,000 infants HIV-exposed, 352 excess annual deaths), Côte d’Ivoire (from 5.59% to 5.91%, RR 1.06, 322 deaths per 100,000, or 125 excess deaths), Mozambique (from 5.78% to 6.12%, RR 1.06, 339 deaths per 100,000, or 961 excess deaths), and Uganda (from 5.39% to 5.61%, RR 1.04, 221 deaths per 100,000, 491 excess deaths); Table 1.

Sensitivity analyses

Sensitivity analyses for Mozambique are shown in Fig 3. Reducing cotrimoxazole uptake to 40%, predicted excess mortality was lower overall, but the fewest predicted deaths still occurred under the current strategy of universal cotrimoxazole provision. Lowering the efficacy of cotrimoxazole in 5% intervals, to as little as 15% predicted mortality reduction (compared to 43% in the CHAP trial), the current strategy continued to have the lowest number of predicted deaths. When EID coverage was increased to 100%, the current strategy still had the lowest predicted mortality, compared to alternative approaches of shorter durations of cotrimoxazole, although predicted excess mortality reduced overall. To explore the effect of reaching and exceeding MTCT elimination targets, we reduced rates for both perinatal and postnatal vertical transmission to zero. Finally, a combined conservative estimate of low cotrimoxazole coverage (40%) and efficacy (15%), low vertical transmission (2%) and high EID (100%), designed to model a plausible future scenario in each country, showed the current strategy still led to fewer predicted deaths, though effects were attenuated.

Predicted mortality effects remained consistent across countries in all our sensitivity analyses. Mozambique has the highest vertical transmission rate of the 4 countries (13.5%) and high EID coverage (82.9%). If uptake was as low as 40% in Mozambique, starting cotrimoxazole only after a positive HIV test is estimated to result in 369 excess annual deaths compared to the current strategy (Fig 3). Figures for sensitivity analysis in Cote d’Ivoire (Fig G in S1 Text), Zimbabwe (Fig E in S1 Text), and Uganda (Fig J in S1 Text) are available in S1 Text. Uganda has the lowest vertical transmission rate (5.9%) and EID coverage of 66.2%. Further reducing perinatal and postnatal MTCT without changing EID is estimated to result in 311 and 251 excess annual deaths, respectively, if a strategy of starting cotrimoxazole only after a positive HIV test was deployed. Cote d’Ivoire has the lowest EID coverage of the 4 countries (60.8%) and vertical transmission of 7.8%. In the most optimistic consideration, if EID coverage was increased to 100%, 3-month provision of cotrimoxazole is predicted to result in 51 excess annual deaths, in part due to continued vertical transmission through breastfeeding. In Zimbabwe, where the vertical transmission rate (8.7%) is similar to the regional estimate for Eastern and Southern Africa, a conservative ascribed mortality reduction to cotrimoxazole of 15% would still result in 352 excess annual deaths when cotrimoxazole is only started after a positive HIV test result.

Discussion

Our study suggests that changing the current cotrimoxazole policy for children born to mothers with HIV is predicted to result in more deaths across the 4 modelled high-burden settings representing different epidemic contexts. The highest increase in mortality is predicted to occur with no routine cotrimoxazole provision (ranging from predicted excess mortality rate of 221 per 100,000 for Uganda to 339 per 100,000 for Mozambique). These predicted effects persisted, but were attenuated, when modelling lower cotrimoxazole uptake, smaller mortality benefits from prophylaxis, higher EID coverage, and lower perinatal and postnatal vertical transmission. These effects are driven entirely by the benefits of cotrimoxazole for infants who acquire HIV. Even as the epidemic context matures and most infants now remain HIV-free, many countries still have slow progress towards elimination targets. This study demonstrates the important “safety net” that cotrimoxazole may provide for children who are missed by the test-to-treat cascade in these settings.

Recent studies have questioned the ongoing value of cotrimoxazole in settings where most children born to women with HIV now remain HIV-free [8–10]. Our study shows that across these 4 modelled settings, predicted mortality would increase by limiting provision of cotrimoxazole. When comparing modelled strategies, the highest predicted mortality occurred under an alternative strategy of providing no universal prophylaxis and instead starting cotrimoxazole only after a positive infant HIV test result. This is driven by high predicted mortality among infants with HIV who are not picked up by testing, and therefore not started on ART or cotrimoxazole [6,7]. Predicted mortality increased to different degrees when cotrimoxazole was given universally but for a shorter duration than currently. This is due to inadequate EID coverage, vertical transmission through breastfeeding after an initial negative test, and shortening cotrimoxazole duration during a period when disease progression among children with HIV remains rapid [3]. While the benefits of retaining the current strategy persisted even with suboptimal cotrimoxazole coverage, reflecting the reality in many settings, there were fewer benefits with low coverage, emphasising the continued effort needed to retain infants in care [20]. When comparing countries, Mozambique had the highest predicted excess mortality and greatest number of excess annual deaths under alternative cotrimoxazole strategies. This likely reflects high perinatal and postnatal transmission in Mozambique compared to the other modelled countries. Conversely, Uganda has the lowest MTCT rates, having made significant progress towards the elimination of paediatric HIV, and lowest predicted increase in mortality with alternative cotrimoxazole strategies. Mozambique has the highest EID testing rate but also had the greatest estimated increase in predicted mortality in our models [16]. Taken together, this suggests that while increased testing coverage is advantageous, continued focus should be placed on interventions to reduce vertical transmission. Where vertical transmission does occur, timely testing and initiation of antiretroviral treatment remains a cornerstone of care. Countries with ongoing high rates of transmission would be most affected by any change in cotrimoxazole strategy.

There are several strengths and limitations to the analysis. Our model was developed with up-to-date, publicly available epidemiological estimates. These were supplemented by data extracted from the highest available levels of evidence. The model incorporates multiple aspects of transmission, testing and treatment uptake and mortality, and has been developed in an accessible format which can be adapted to other settings. Assumptions in the model were interrogated in sensitivity analyses. Although our effects of cotrimoxazole were derived from the CHAP trial—which enrolled children after infancy in the pre-ART era—fewer predicted deaths occurred under the current strategy even when the estimated mortality reduction from cotrimoxazole was as low as 15%. There are limited randomised controlled trial data for cotrimoxazole among breastfed infants with HIV, but it is unlikely that mortality benefits are substantially lower than in the original CHAP trial, given the more rapid disease progression at this age, lack of prognostic markers, and higher risk of Pneumocystis jirovecii pneumonia [9]. Our sensitivity analysis shows that, even if this assumption is untrue and the benefits are much lower, the current policy leads to fewer predicted deaths than a revised policy of shorter (or no) prophylaxis. There are also limitations to this analysis. There are several parameters for which data were not available (e.g., the probability of having a 9-month test if early infant diagnosis has been missed). We assumed equal uptake of cotrimoxazole regardless of final child HIV status, but it is likely that children most likely to acquire HIV are the least engaged in care and are therefore less likely to receive cotrimoxazole. Countries achieving 95-95-95 testing and treatment targets and Path to Elimination of MTCT goals were not modelled and may no longer require routine cotrimoxazole given that few infants acquire HIV and those who do are most likely to be missed by testing [21].

We only modelled mortality increases following reductions in cotrimoxazole provision, given the known benefits of cotrimoxazole for children with HIV; however, our goal was to quantify the magnitude of mortality increase for each strategy across epidemic contexts to provide policymakers with population estimates when considering guideline revisions. We recognise, however, that our model did not incorporate the potential risks of cotrimoxazole, such as antimicrobial resistance, as mortality effects are difficult to quantify given limited quality data from sub-Saharan Africa [22]. There is evidence that cotrimoxazole use in infants who are HEU leads to antimicrobial resistance, including 1 small study showing cross-class resistance to amoxicillin, but no evidence of major microbiome dysbiosis from available data [23]. Although there are estimates of deaths attributable to resistance for specific antibiotics in several global regions [24], it is difficult to model what proportion of these deaths would be averted in each of our scenarios, particularly as it remains unclear whether cotrimoxazole resistance is reversible to a clinically relevant level with decline in use [25]. Estimates of pathogen-specific deaths attributable to AMR from cotrimoxazole in sub-Saharan Africa in the post-neonatal period range from 0.0 (0.0 to 0.1) per 100,000 for Proteus spp. to 15.07 (2.3 to 31.8) per 100,000 for Streptococcus pneumoniae, but even collectively these estimates are much lower than the excess deaths modelled in our programmatic scenarios from restricted cotrimoxazole use. Responsive country-level surveillance is required to evaluate local resistance data in the context of evolving HIV epidemics. Lastly, this study did not consider the economic implications of the proposed strategies. Future research should investigate the potential cost savings and the financial impacts of implementing each strategy on national budgets [26].

Countries need to continue to strengthen the test-to-treat cascade, to ensure timely testing and ART initiation, in order to reduce morbidity and mortality among infants who are HIV-exposed. However, our models show that cotrimoxazole still remains an important part of the package of care. Decisions around appropriate provision of cotrimoxazole to infants who are HIV-exposed require a pragmatic, public health approach with difficult decisions that weigh risks and benefits that differ across settings. Results from this model may be considered as part of the breadth of evidence which help inform policymaker decisions. Though there has been progress in prevention of paediatric HIV, the “last mile” to elimination is particularly challenging. Furthermore, in many high-burden settings, testing coverage is not universal, making it difficult to reliably distinguish infants who are HIV-infected or HEU. Individualised algorithms dependent on testing contingencies may be unrealistic given resource and service limitations at most low-level health facilities. The current programmatic cotrimoxazole strategy provides important benefits for the small number of high-risk infants who still acquire HIV through vertical transmission and are missed due to gaps in the test-to-treat cascade. As demonstrated in the sensitivity analysis for each strategy, lower vertical transmission rates and improved testing and treatment uptake attenuate the predicted increase in mortality. This demonstrates the importance of considering how vertical transmission should inform cotrimoxazole policy: in the context of high ongoing transmission, the current universal cotrimoxazole prophylaxis approach is predicted to lead to substantially fewer deaths than alternative strategies, while settings with low transmission rates may derive less benefit from universal cotrimoxazole prophylaxis. Therefore, hybrid cotrimoxazole policies may be a consideration noting we did not model countries achieving HIV elimination targets or with strong test-to-treat cascades. Further research is required to better understand cotrimoxazole coverage for children who are HIV-exposed; to determine the proportion who are subsequently found to be living with HIV; to better understand community values and preferences; and to identify barriers to engagement with the test-to-treat cascade. Finally, cost-effectiveness analysis informed by national programme cost data is required to better understand the most optimal health investment and cost per death averted and economic impact of each strategy.

In conclusion, despite impressive progress in prevention of vertical HIV transmission, elimination is not yet in sight in most high-burden settings. Priorities include maintaining optimal care and ART for mothers; providing postnatal HIV prophylaxis for infants; and ensuring timely testing and linkage to care. However, cotrimoxazole continues to provide a vital safety net for infants who acquire HIV but are not tested or started on ART. Our data suggest that changing the current strategy of universal cotrimoxazole provision for all infants who are HIV-exposed would increase predicted mortality to varying degrees across the 4 modelled countries, depending on vertical transmission rates and test-to-treatment cascade coverage. Policymakers need to balance the possible negative effects of cotrimoxazole with potential excess deaths resulting from changes in programmatic use depending on their specific context.

Supporting information

S1 TRIPOD Checklist. TRIPOD Checklist: Prediction model development.

(DOCX)

pmed.1004334.s001.docx^{(88.6KB, docx)}

S1 Text

Table A in S1 Text. Model assumptions and data sources. CTX = cotrimoxazole, ART = antiretroviral therapy, EID = early infant diagnosis at 6 weeks, ZIM = Zimbabwe, CIV = Cote d’Ivoire, MOZ = Mozambique, UGA = Uganda. *The authorship brings multidisciplinary expertise in paediatric HIV (AJP, CE, DMG, CW, MP), epidemiology (MS, DMG), clinical trials (DMG, AJP, MP, MS), and policy (SM, MP). Fig A in S1 Text. Predicted mortality percentage per year by country under alternative cotrimoxazole strategies. Columns represent additional deaths from each alternate strategy in comparison to the current WHO programmatic strategy of providing cotrimoxazole to all HIV-exposed infants. 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig B in S1 Text. Predicted excess mortality rate (per 100,000) per year by country under alternative cotrimoxazole strategies. Columns represent additional deaths from each alternate strategy in comparison to the current WHO programmatic strategy of providing cotrimoxazole to all HIV-exposed infants. 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig C in S1 Text. Predicted risk ratio per year by country under alternative cotrimoxazole strategies. Columns represent additional deaths from each alternate strategy in comparison to the current WHO programmatic strategy of providing cotrimoxazole to all HIV-exposed infants. 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig D in S1 Text. Sensitivity Analysis for Zimbabwe (risk ratios). Sensitivity analysis, for Zimbabwe, exploring the effect of varying assumptions on the risk ratio for deaths (6 weeks to 2 years) compared to the current WHO strategy. CTX = cotrimoxazole, ART = antiretroviral therapy, EID = early infant diagnosis at 6 weeks, MTCT = mother-to-child transmission, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig E in S1 Text. Sensitivity Analysis for Zimbabwe (excess deaths). Sensitivity analysis, for Zimbabwe, exploring the effect of varying assumptions on the risk ratio for deaths (6 weeks to 2 years) compared to the current WHO strategy. CTX = cotrimoxazole, ART = antiretroviral therapy, EID = early infant diagnosis at 6 weeks, MTCT = mother-to-child transmission, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig F in S1 Text. Sensitivity Analysis for Cote d’Ivoire (risk ratios). Sensitivity analysis, for Cote d’Ivoire, exploring the effect of varying assumptions on the risk ratio for deaths (6 weeks to 2 years) compared to the current WHO strategy. CTX = cotrimoxazole, ART = antiretroviral therapy, EID = early infant diagnosis at 6 weeks, MTCT = mother-to-child transmission, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig G in S1 Text. Sensitivity Analysis for Cote d’Ivoire (excess deaths). Sensitivity analysis, for Cote d’Ivoire, exploring the effect of varying assumptions on the risk ratio for deaths (6 weeks to 2 years) compared to the current WHO strategy. CTX = cotrimoxazole, ART = antiretroviral therapy, EID = early infant diagnosis at 6 weeks, MTCT = mother-to-child transmission, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig H in S1 Text. Sensitivity Analysis for Mozambique (risk ratios). Sensitivity analysis, for Mozambique, exploring the effect of varying assumptions on the risk ratio for deaths (6 weeks to 2 years) compared to the current WHO strategy. CTX = cotrimoxazole, ART = antiretroviral therapy, EID = early infant diagnosis at 6 weeks, MTCT = mother-to-child transmission, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig I in S1 Text. Sensitivity Analysis for Uganda (risk ratios). Sensitivity analysis, for Uganda, exploring the effect of varying assumptions on the risk ratio for deaths (6 weeks to 2 years) compared to the current WHO strategy. CTX = cotrimoxazole, ART = antiretroviral therapy, EID = early infant diagnosis at 6 weeks, MTCT = mother-to-child transmission, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig J in S1 Text. Sensitivity Analysis for Uganda (excess deaths). Sensitivity analysis, for Uganda, exploring the effect of varying assumptions on the risk ratio for deaths (6 weeks to 2 years) compared to the current WHO strategy. CTX = cotrimoxazole, ART = antiretroviral therapy, EID = early infant diagnosis at 6 weeks, MTCT = mother-to-child transmission, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig K in S1 Text. Sensitivity Variable—Cotrimoxazole uptake, Zimbabwe. Risk ratio of mortality for varying risk reduction from cotrimoxazole uptake for Zimbabwe, (risk ratio). CTX = cotrimoxazole, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig L in S1 Text. Sensitivity Variable–Risk reduction from CTX, Zimbabwe. Risk ratio of mortality for varying risk reduction from cotrimoxazole while infant with HIV is taking antiretroviral therapy for Zimbabwe, (risk ratio). Risk reduction from 15% to 43%. CTX = cotrimoxazole, ART = antiretroviral therapy, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig M in S1 Text. Sensitivity Variable—Risk reduction from CTX, Zimbabwe. Risk ratio of mortality for varying risk reduction from cotrimoxazole while infant with HIV is taking antiretroviral therapy for Zimbabwe, (risk ratio). Risk reduction from 25% to 60%. CTX = cotrimoxazole, ART = antiretroviral therapy, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig N in S1 Text. Sensitivity Variable—EID testing, Zimbabwe. Risk ratio of mortality for varying probability of HIV-exposed infants undergoing early infant diagnosis (EID) for Zimbabwe, (risk ratio). EID = early infant diagnosis at 6 weeks, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig O in S1 Text. Sensitivity Variable—Perinatal MTCT, Zimbabwe. Risk ratio of mortality for varying probability of perinatal mother-to-child transmission (MTCT) for Zimbabwe, (risk ratio). MTCT = mother-to-child transmission, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig P in S1 Text. Sensitivity Variable—Postnatal MTCT, Zimbabwe. Risk ratio of mortality for varying probability of postnatal mother-to-child transmission (MTCT) for Zimbabwe, (risk ratio). MTCT = mother-to-child transmission, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig Q in S1 Text. Sensitivity Variable—Cotrimoxazole uptake, Cote d’Ivoire. Risk ratio of mortality for varying risk reduction from cotrimoxazole uptake for Cote d’Ivoire, (risk ratio). CTX = cotrimoxazole, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig R in S1 Text. Sensitivity Variable—Risk reduction from CTX, Cote d’Ivoire. Risk ratio of mortality for varying risk reduction from cotrimoxazole while infant with HIV is taking antiretroviral therapy for Cote d’Ivoire, (risk ratio). Risk reduction from 15% to 43%. CTX = cotrimoxazole, ART = antiretroviral therapy, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig S in S1 Text. Sensitivity Variable—Risk reduction from CTX, Cote d’Ivoire. Risk ratio of mortality for varying risk reduction from cotrimoxazole while infant with HIV is taking antiretroviral therapy for Cote d’Ivoire, (risk ratio). Risk reduction from 15% to 60%. CTX = cotrimoxazole, ART = antiretroviral therapy, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig T in S1 Text. Sensitivity Variable—EID testing, Cote d’Ivoire. Risk ratio of mortality for varying probability of HIV-exposed infants undergoing early infant diagnosis (EID) for Cote d’Ivoire, (risk ratio). EID = early infant diagnosis at 6 weeks, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig U in S1 Text. Sensitivity Variable—Perinatal MTCT, Cote d’Ivoire. Risk ratio of mortality for varying probability of perinatal mother-to-child transmission (MTCT) for Cote d’Ivoire, (risk ratio). MTCT = mother-to-child transmission, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig V in S1 Text. Sensitivity Variable—Postnatal MTCT, Cote d’Ivoire. Risk ratio of mortality for varying probability of postnatal mother-to-child transmission (MTCT) for Cote d’Ivoire, (risk ratio). MTCT = mother-to-child transmission, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig W in S1 Text. Sensitivity Variable—Cotrimoxazole uptake, Mozambique. Risk ratio of mortality for varying risk reduction from cotrimoxazole uptake for Mozambique, (risk ratio). CTX = cotrimoxazole, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig X in S1 Text. Sensitivity Variable—Risk reduction from CTX, Mozambique. Risk ratio of mortality for varying risk reduction from cotrimoxazole while infant with HIV is taking antiretroviral therapy for Mozambique, (risk ratio). Risk reduction from 15% to 43%. CTX = cotrimoxazole, ART = antiretroviral therapy, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig Y in S1 Text. Sensitivity Variable—Risk reduction from CTX, Mozambique. Risk ratio of mortality for varying risk reduction from cotrimoxazole while infant with HIV is taking antiretroviral therapy for Mozambique, (risk ratio). Risk reduction from 15% to 60%. CTX = cotrimoxazole, ART = antiretroviral therapy, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig Z in S1 Text. Sensitivity Variable—EID testing, Mozambique. Risk ratio of mortality for varying probability of HIV-exposed infants undergoing early infant diagnosis (EID) for Mozambique, (risk ratio). EID = early infant diagnosis at 6 weeks, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig AA in S1 Text. Sensitivity Variable—Perinatal MTCT, Mozambique. Risk ratio of mortality for varying probability of perinatal mother-to-child transmission (MTCT) for Mozambique, (risk ratio). MTCT = mother-to-child transmission, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig AB in S1 Text. Sensitivity Variable—Postnatal MTCT, Mozambique. Risk ratio of mortality for varying probability of postnatal mother-to-child transmission (MTCT) for Mozambique, (risk ratio). MTCT = mother-to-child transmission, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig AC in S1 Text. Sensitivity Variable—Cotrimoxazole uptake, Uganda. Risk ratio of mortality for varying risk reduction from cotrimoxazole uptake for Uganda, (risk ratio). CTX = cotrimoxazole, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig AD in S1 Text. Sensitivity Variable—Risk reduction from CTX, Uganda. Risk ratio of mortality for varying risk reduction from cotrimoxazole while infant with HIV is taking antiretroviral therapy for Uganda, (risk ratio). Risk reduction from 15% to 43%. CTX = cotrimoxazole, ART = antiretroviral therapy, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig AE in S1 Text. Sensitivity Variable—Risk reduction from CTX, Uganda. Risk ratio of mortality for varying risk reduction from cotrimoxazole while infant with HIV is taking antiretroviral therapy for Uganda, (risk ratio). Risk reduction from 15% to 60%. CTX = cotrimoxazole, ART = antiretroviral therapy, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig AF in S1 Text. Sensitivity Variable—EID testing, Uganda. Risk ratio of mortality for varying probability of HIV-exposed infants undergoing early infant diagnosis (EID) for Uganda, (risk ratio). EID = early infant diagnosis at 6 weeks, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig AG in S1 Text. Sensitivity Variable—Perinatal MTCT, Uganda. Risk ratio of mortality for varying probability of perinatal mother-to-child transmission (MTCT) for Uganda, (risk ratio). MTCT = mother-to-child transmission, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months. Fig AH in S1 Text. Sensitivity Variable—Postnatal MTCT, Uganda. Risk ratio of mortality for varying probability of postnatal mother-to-child transmission (MTCT) for Uganda, (risk ratio). MTCT = mother-to-child transmission, 12m = 12 months, 9m = 9 months, 6m = 6 months, 3m = 3 months.

(DOCX)

pmed.1004334.s002.docx^{(7MB, docx)}

Acknowledgments

Disclaimer: The content of this publication is solely the responsibility of the authors and does not necessarily represent the official views of the World Health Organization.

Abbreviations

ART: antiretroviral therapy
EID: early infant diagnosis
HEU: HIV-exposed uninfected
PMTCT: prevention of mother-to-child transmission
RR: risk ratio
WHO: World Health Organization

Data Availability

The data underlying the results presented in the study are available from https://osf.io/8kjgp/.

Funding Statement

SM is supported by a National Institute for Health and Care Research (NIHR) Academic Clinical Fellowship (https://www.nihr.ac.uk/, award ref: ACF-2020-19-502). AJP is funded by the Wellcome Trust (https://wellcome.org/, grant number: 108065/Z/15/Z). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Joint United Nations Programme on HIV/AIDS. UNAIDS data 2022. Geneva: UNAIDS; 2022. [PubMed]
2.Joint United Nations Programme on HIV/AIDS. Start Free, Stay Free, AIDS Free Final report on 2020 targets. 2021.
3.Newell ML, Coovadia H, Cortina-Borja M, Rollins N, Gaillard P, Dabis F, et al. Mortality of infected and uninfected infants born to HIV-infected mothers in Africa: a pooled analysis. Lancet. 2004;364(9441):1236–1243. doi: 10.1016/S0140-6736(04)17140-7 . [DOI] [PubMed] [Google Scholar]
4.World Health Organization. Guidelines on post-exposure prophylaxis for HIV and the use of co-trimoxazole prophylaxis for HIV-related infections among adults, adolescents and children: recommendations for a public health approach: December 2014 supplement to the 2013 consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection. World Health Organization; 2014. [PubMed]
5.Church JA, Fitzgerald F, Walker AS, Gibb DM, Prendergast AJ. The expanding role of co-trimoxazole in developing countries. Lancet Infect Dis. 2015;15(3):327–39. Epub 20150121. doi: 10.1016/S1473-3099(14)71011-4 . [DOI] [PubMed] [Google Scholar]
6.Bwakura-Dangarembizi M, Kendall L, Bakeera-Kitaka S, Nahirya-Ntege P, Keishanyu R, Nathoo K, et al. A randomized trial of prolonged co-trimoxazole in HIV-infected children in Africa. N Engl J Med. 2014;370(1):41–53. doi: 10.1056/NEJMoa1214901 . [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Chintu C, Bhat GJ, Walker AS, Mulenga V, Sinyinza F, Lishimpi K, et al. Co-trimoxazole as prophylaxis against opportunistic infections in HIV-infected Zambian children (CHAP): a double-blind randomised placebo-controlled trial. Lancet. 2004;364(9448):1865–1871. doi: 10.1016/S0140-6736(04)17442-4 . [DOI] [PubMed] [Google Scholar]
8.Daniels B, Coutsoudis A, Moodley-Govender E, Mulol H, Spooner E, Kiepiela P, et al. Effect of co-trimoxazole prophylaxis on morbidity and mortality of HIV-exposed, HIV-uninfected infants in South Africa: a randomised controlled, non-inferiority trial. Lancet Glob Health. 2019;7(12):e1717–e1727. doi: 10.1016/S2214-109X(19)30422-X . [DOI] [PubMed] [Google Scholar]
9.Lockman S, Hughes M, Powis K, Ajibola G, Bennett K, Moyo S, et al. Effect of co-trimoxazole on mortality in HIV-exposed but uninfected children in Botswana (the Mpepu Study): a double-blind, randomised, placebo-controlled trial. Lancet Glob Health. 2017;5(5):e491–e500. doi: 10.1016/S2214-109X(17)30143-2 . [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Daniels B, Kuhn L, Spooner E, Mulol H, Goga A, Feucht U, et al. Cotrimoxazole guidelines for infants who are HIV-exposed but uninfected: a call for a public health and ethics approach to the evidence. Lancet Glob Health. 2022;10(8):e1198–e1203. doi: 10.1016/S2214-109X(22)00120-6 [DOI] [PubMed] [Google Scholar]
11.Teasdale CA, Zimba R, Abrams EJ, Sachathep K, Ndagije F, Nuwagaba-Biribonwoha H, et al. Estimates of the prevalence of undiagnosed HIV among children living with HIV in Eswatini, Lesotho, Malawi, Namibia, Tanzania, Zambia, and Zimbabwe from 2015 to 2017: an analysis of data from the cross-sectional Population-based HIV Impact Assessment surveys. Lancet HIV. 2022;9(2):e91–e101. doi: 10.1016/S2352-3018(21)00291-5 . [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Ciaranello AL, Park JE, Ramirez-Avila L, Freedberg KA, Walensky RP, Leroy V. Early infant HIV-1 diagnosis programs in resource-limited settings: opportunities for improved outcomes and more cost-effective interventions. BMC Med. 2011;9:59. Epub 20110520. doi: 10.1186/1741-7015-9-59 . [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Chatterjee A, Tripathi S, Gass R, Hamunime N, Panha S, Kiyaga C, et al. Implementing services for Early Infant Diagnosis (EID) of HIV: a comparative descriptive analysis of national programs in four countries. BMC Public Health. 2011;11:553. Epub 20110713. doi: 10.1186/1471-2458-11-553 . [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Wedderburn CJ, Evans C, Slogrove AL, Rehman AM, Gibb DM, Prendergast AJ, et al. Co-trimoxazole prophylaxis for children who are HIV-exposed and uninfected: a systematic review. J Int AIDS Soc. 2023;26(6):e26079. doi: 10.1002/jia2.26079 . [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Arikawa S, Rollins N, Jourdain G, Humphrey J, Kourtis AP, Hoffman I, et al. Contribution of Maternal Antiretroviral Therapy and Breastfeeding to 24-Month Survival in Human Immunodeficiency Virus-Exposed Uninfected Children: An Individual Pooled Analysis of African and Asian Studies. Clin Infect Dis. 2018;66(11):1668–1677. doi: 10.1093/cid/cix1102 . [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Joint United Nations Programme on HIV/AIDS. UNAIDS data 2021. Geneva: UNAIDS; 2021. [PubMed]
17.Luo R, Fong Y, Boeras D, Jani I, Vojnov L. The clinical effect of point-of-care HIV diagnosis in infants: a systematic review and meta-analysis. Lancet. 2022;400(10356):887–895. doi: 10.1016/S0140-6736(22)01492-1 . [DOI] [PubMed] [Google Scholar]
18.Violari A, Cotton MF, Gibb DM, Babiker AG, Steyn J, Madhi SA, et al. Early antiretroviral therapy and mortality among HIV-infected infants. N Engl J Med. 2008;359(21):2233–2244. doi: 10.1056/NEJMoa0800971 . [DOI] [PMC free article] [PubMed] [Google Scholar]
19.World Bank. World development indicators 2022. The World Bank; 2022. https://data.worldbank.org/. Accessed 2022 Dec 2.
20.Mofenson LM, Cohn J, Sacks E. Challenges in the early infant HIV diagnosis and treatment cascade. J Acquir Immune Defic Syndr. 2020;84:S1–S4. doi: 10.1097/QAI.0000000000002366 [DOI] [PubMed] [Google Scholar]
21.Frescura L, Godfrey-Faussett P, Feizzadeh AA, El-Sadr W, Syarif O, Ghys PD, et al. Achieving the 95 95 95 targets for all: A pathway to ending AIDS. PLoS ONE. 2022;17(8):e0272405. Epub 20220804. doi: 10.1371/journal.pone.0272405 . [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Tadesse BT, Ashley EA, Ongarello S, Havumaki J, Wijegoonewardena M, Gonzalez IJ, et al. Antimicrobial resistance in Africa: a systematic review. BMC Infect Dis. 2017;17(1):616. Epub 20170911. doi: 10.1186/s12879-017-2713-1 . [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Powis KM, Souda S, Lockman S, Ajibola G, Bennett K, Leidner J, et al. Cotrimoxazole prophylaxis was associated with enteric commensal bacterial resistance among HIV-exposed infants in a randomized controlled trial, Botswana. J Int AIDS Soc. 2017;20(3). doi: 10.1002/jia2.25021 . [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Antimicrobial Resistance C. Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. Lancet. 2022;399(10325):629–55. Epub 20220119. doi: 10.1016/S0140-6736(21)02724-0 . [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Pouwels KB, Batra R, Patel A, Edgeworth JD, Robotham JV, Smieszek T. Will co-trimoxazole resistance rates ever go down? Resistance rates remain high despite decades of reduced co-trimoxazole consumption. J Glob Antimicrob Resist. 2017;11:71–4. Epub 20170731. doi: 10.1016/j.jgar.2017.07.013 . [DOI] [PubMed] [Google Scholar]
26.Ryan M, Griffin S, Chitah B, Walker AS, Mulenga V, Kalolo D, et al. The cost-effectiveness of cotrimoxazole prophylaxis in HIV-infected children in Zambia. AIDS. 2008;22(6):749–757. doi: 10.1097/QAD.0b013e3282f43519 . [DOI] [PubMed] [Google Scholar]

PLoS Med. doi: 10.1371/journal.pmed.1004334.r001

Decision Letter 0

Alexandra Schaefer

13 Jun 2023

Dear Dr Mathur,

Thank you for submitting your manuscript entitled "The Impact of Alternative Programmatic Cotrimoxazole Strategies on Mortality Among Children Born to Mothers with HIV: A Modelling Study" for consideration by PLOS Medicine.

Your manuscript has now been evaluated by the PLOS Medicine editorial staff and I am writing to let you know that we would like to send your submission out for external peer review.

However, before we can send your manuscript to reviewers, we need you to complete your submission by providing the metadata that is required for full assessment. To this end, please login to Editorial Manager where you will find the paper in the 'Submissions Needing Revisions' folder on your homepage. Please click 'Revise Submission' from the Action Links and complete all additional questions in the submission questionnaire.

Please re-submit your manuscript within two working days, i.e. by Jun 15 2023 11:59PM.

Once your full submission is complete, your paper will undergo a series of checks in preparation for peer review. Once your manuscript has passed all checks it will be sent out for review.

Feel free to email us at plosmedicine@plos.org if you have any queries relating to your submission.

Kind regards,

Alexandra Schaefer, PhD

Associate Editor

PLOS Medicine

PLoS Med. doi: 10.1371/journal.pmed.1004334.r002

Decision Letter 1

Louise Gaynor-Brook

23 Aug 2023

Dear Dr. Mathur,

Thank you very much for submitting your manuscript "The Impact of Alternative Programmatic Cotrimoxazole Strategies on Mortality Among Children Born to Mothers with HIV: A Modelling Study" (PMEDICINE-D-23-01623R1) for consideration at PLOS Medicine.

Your paper was evaluated by four independent reviewers, including a statistical reviewer, and discussed among all the editors here and with an academic editor with relevant expertise. The reviews are appended at the bottom of this email and any accompanying reviewer attachments can be seen via the link below:

[LINK]

In light of these reviews, I am afraid that we will not be able to accept the manuscript for publication in the journal in its current form, but we would like to consider a revised version that addresses the reviewers' and editors' comments. Obviously we cannot make any decision about publication until we have seen the revised manuscript and your response, and we plan to seek re-review by one or more of the reviewers.

In revising the manuscript for further consideration, your revisions should address the specific points made by each reviewer and the editors. Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments, the changes you have made in the manuscript, and include either an excerpt of the revised text or the location (eg: page and line number) where each change can be found. Please submit a clean version of the paper as the main article file; a version with changes marked should be uploaded as a marked up manuscript.

In addition, we request that you upload any figures associated with your paper as individual TIF or EPS files with 300dpi resolution at resubmission; please read our figure guidelines for more information on our requirements: http://journals.plos.org/plosmedicine/s/figures. While revising your submission, please upload your figure files to the PACE digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at PLOSMedicine@plos.org.

We expect to receive your revised manuscript by Sep 13 2023 11:59PM. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

We ask every co-author listed on the manuscript to fill in a contributing author statement, making sure to declare all competing interests. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. If new competing interests are declared later in the revision process, this may also hold up the submission. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT. You can see our competing interests policy here: http://journals.plos.org/plosmedicine/s/competing-interests.

Please use the following link to submit the revised manuscript:

https://www.editorialmanager.com/pmedicine/

Your article can be found in the "Submissions Needing Revision" folder.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

We look forward to receiving your revised manuscript.

Sincerely,

Louise Gaynor-Brook, MBBS PhD

Senior Editor, PLOS Medicine

plosmedicine.org

-----------------------------------------------------------

Requests from the editors:

General comments:

Throughout the paper, please adapt reference call-outs to the following style: "... HIV-free [4,5]." (noting the absence of spaces within the square brackets).

Please revise ‘impact’ to ‘potential impact’ or similar when referring to your own results

Throughout the main text, please refer to ‘predicted mortality’ when referring to your own results

Data availability:

PLOS Medicine requires that the de-identified data underlying the specific results in a published article be made available, without restrictions on access, in a public repository or as Supporting Information at the time of article publication, provided it is legal and ethical to do so. If the data are freely or publicly available, please note this and state the location of the data (include the DOI or accession number); please note that a study author cannot be responsible for permitting access to the data or be the contact person for the data (I note that Andrew Prendergast approved my access request on OSF).

Title: Please revise your title according to PLOS Medicine's style. We suggest “Estimating the Impact of Alternative Programmatic Cotrimoxazole Strategies on Mortality Among Children Born to Mothers with HIV: A Modelling Study” or similar

Abstract:

Line 48 - please revise to ‘potential impact’

In the last sentence of the Abstract Methods and Findings section, please describe 2-3 of the main limitations of the study's methodology.

Abstract Conclusions: Please address the implications of your study, being careful to avoid assertions of primacy.

Author Summary:

At this stage, we ask that you include a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract. Please see our author guidelines for more information: https://journals.plos.org/plosmedicine/s/revising-your-manuscript#loc-author-summary

In the final bullet point of ‘What Do These Findings Mean?’, please describe the main limitations of the study in non-technical language.

Introduction:

If there has been a systematic review of the evidence related to your study (or you have conducted one), please refer to and reference that review and indicate whether it supports the need for your study.

Methods:

Did your study have a prospective protocol or analysis plan? If so, please include the relevant prospectively written document with your revised manuscript as a Supporting Information file to be published alongside your study, and cite it in the Methods section. Changes in the analysis-- including those made in response to peer review comments-- should be identified as such in the Methods section of the paper, with rationale.

We suggest that TRIPOD may be an appropriate checklist for your study design; please include a completed checklist as a supplementary file. Please add the following statement, or similar, to the Methods: "This study is reported as per the Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD) guideline (S1 Checklist)." The TRIPOD guideline can be found here: https://www.equator-network.org/reporting-guidelines/tripod-statement/ When completing the checklist, please use section and paragraph numbers, rather than page numbers which will likely no longer correspond to the appropriate sections after copy-editing.

Using guidance from Geoffrey P Garnett, Simon Cousens, Timothy B Hallett, Richard Steketee, Neff Walker. Mathematical models in the evaluation of health programmes. (2011) Lancet DOI:10.1016/S0140-6736(10)61505-X;

Please provide a diagram that shows the model structure, including how the disease natural history is represented and how the putative intervention (cotrimoxazole) could affect the system.

Please provide a complete list of model parameters and important caveats about the use of these values noted.

Please provide a clear statement about how the model was fitted to the data (may include goodness-of-fit measure, the numerical algorithm used, which parameter varied, constraints imposed on parameter values, and starting conditions)

For uncertainty analyses, please state the sources of uncertainties quantified and not quantified [can include parameter, data, and model structure].

Please discuss the scientific rationale for this choice of model structure and identify points where this choice could influence conclusions drawn. Please also describe the strength of the scientific basis underlying the key model assumptions.

Results:

Line 236 - please clarify that sensitivity analyses for Mozambique are shown in Fig 3

Throughout the results section, please refer to ‘predicted mortality’

Please provide tables/figures for all of the sensitivity analyses presented in the main text (these can be included in the supplementary information)

Line 252 - please refer to specific supplementary figures for each of these countries. NB) these supplementary figures have not been included in this submission.

Discussion:

Please present and organize the Discussion as follows: a short, clear summary of the article's findings; what the study adds to existing research and where and why the results may differ from previous research; strengths and limitations of the study; implications and next steps for research, clinical practice, and/or public policy; one-paragraph conclusion.

Throughout the discussion, please refer to ‘predicted mortality’ when referring to your own results

Figures:

Please consider avoiding the use of red and green in order to make your figure more accessible to those with colour blindness.

Please define all abbreviations used in the figure legend of each figure.

Tables:

Please define all abbreviations used in the table legend of each table (including supplementary tables).

References:

Please ensure that journal name abbreviations match those found in the National Center for Biotechnology Information (NCBI) databases (http://www.ncbi.nlm.nih.gov/nlmcatalog/journals), and are appropriately formatted and capitalised.

Please also see https://journals.plos.org/plosmedicine/s/submission-guidelines#loc-references for further details on reference formatting.

Where website addresses are cited, please specify the date of access.

Comments from the reviewers:

Reviewer #1: Mathur and colleagues describe in this manuscript deterministic compartmental models developed by the authors relying upon up-to-date, publicly available estimates including of HIV early infant HIV testing (EIT), prevalence of infant HIV acquisition, and antiretroviral treatment (ART) uptake to evaluate mortality between 6- and 24-months of life among HIV-exposed children comparing a strategy of universal cotrimoxazole use for all children to other strategies with lower coverage, as cotrimoxazole has not bene shown to have any survival benefit when used in children who are HIV-exposed uninfected, but has been shown to improve survival in children living with HIV not yet on treatment. Models were applied to four high burden HIV-countries yet to achieve 95-95-95 HIV program targets or Eliminating Mother-to-Child-Transmission goals, specifically Cote d'Ivoire, Mozambique, Uganda, and Zimbabwe. Country-specific data was supplemented with data extracted from peer-reviewed publications of clinical trials, observational studies, and other epidemiological data in eras both before and after universal ART was recommended at time of HIV diagnoses. The paper is well-written with excellent tables and figures. Considerable work has been devoted to testing multiple strategies. The authors conclude changing the current recommendations of universal cotrimoxazole provision for all HIV-exposed infants would increase mortality to varying degrees depending on vertical transmission rates and test-to-treatment cascade strategies would increase mortality to varying degrees. While this is an important manuscript that will be of interest to PLOS Medicine readers, there are two issues raised by the final recommendation of the modeling exercise that need to be addressed by the authors. Major and minor revision recommendations are presented below.

Major Revision Recommendations:

The authors' final conclusions advocate for universal cotrimoxazole use by HIV-exposed infants until infection is excluded and vertical transmission risk has ended. Conflictingly, the authors make mention in the discussion that among countries achieving 95-95-95 testing and treatment targets and Elimination of Mother-to-Child goals, cotrimoxazole prophylaxis may not be required. Interestingly, in at least one of the two studies cited for the lack of survival benefit of cotrimoxazole use among infants and children HIV-exposed uninfected, the Mpepu randomized controlled trial was conducted in a setting where the 95-95-95 targets have been achieved and infant HIV acquisition is currently estimated to be approximately 1%. This presents an argument for a hybrid policy, as the absence of using a country such as Botswana in the modeling exercise may be influencing an incorrect policy recommendation about universal use of cotrimoxazole.

The model assumes that 100% of infants who are HIV-exposed and for whom HIV infection status has not been finalized or vertical transmission risk continues to exist will be prescribed cotrimoxazole and parents or caregivers will have 100% adherence to dosing. While the authors acknowledged that they lacked a data source to estimate the number of children most likely to acquire HIV who are not engaged in care, this is a major flaw in the model that leads to questionable justification for continued universal cotrimoxazole use. If there is an issue with timely HIV testing of infants at risk for HIV, or timely ART initiation for those who found to be infected, monthly or quarterly encounters with the health care system for refills of cotrimoxazole are highly unlikely. If they are occurring, they could be better spent employing HIV testing or initiation of treatment, as applicable. It would be important in the Discussion section to call for accurate accounting of cotrimoxazole use in HIV-exposed infants to determine the proportion who actually receive this treatment and are subsequently found to be living with HIV, with a further understanding of the hindrances that preclude timely diagnoses and treatment initiation. The elephant in the room is engagement in the care cascade. To justify a policy that, on paper appears to save lives, but exchanges one medication regimen, namely ART, for another, cotrimoxazole, promulgates continued weaknesses in the treatment and care cascade and may only save lives on paper.

Minor Revision Recommendations:

Please use person first language throughout the manuscript. For example, at line 45 reference is made for "cotrimoxazole prophylaxis for HIV-exposed infants". Please revised this to read "cotrimoxazole prophylaxis for infants who are HIV-exposed".

Please review references, as some have "abbreviated" authorship. (eg references 1, 2, 15 and 20).

At lines 349-350, reference is made to "EMTCT Path to Elimination goals". However, since the "E" in EMTCT stand for Elimination, this does not make sense. It might be clearer to take out the word "Elimination" in this phrase.

Reviewer #2: This paper presents the mortality effects of providing cotrimoxazole for periods shorter than the WHO guidelines. The conclusion is obvious, shorter periods of cotrim results in more deaths, but the paper is useful for showing the magnitude of the effect.

1. In my opinion the paper is not really an evaluation of the WHO guidelines because it doe not consider any costs or negative effects. It would be relatively easy to show the additional costs of longer periods of cotrim and the cost per death averted. Since cotrim is inexpensive I assume it is very cost-effective. The authors do not include increased antimicrobial resistance because they are unsure of the effect size. But they did extensive sensitivity analysis of the model parameters. Why not include sensitivity around assumptions about resistance? It would make the paper much stronger.

2. The paper states that the model is freely available and provide a website. I tried to download the model but have been unsuccessful so far. I went to the OSF website using the link provided in the paper, but had to first create an account with OSF. Once I did that I received a message that I had to request permission to review the file. I requested permission. The next day I was informed that permission had been granted. When I went to the site I just saw a folder that had been set up for collaboration. There was no Excel file in that folder. So I still have not seen the model. Other readers may be frustrated with this process. Since I had to request permission there was no hope of a confidential review.

Reviewer #3: The study appears to be highly relevant and valuable in informing the ongoing debate regarding the revision of the current WHO policy on administering cotrimoxazole to all HIV-exposed infants during breastfeeding. I believe that the research findings will serve as a credible and significant piece of evidence for understanding the disease and its treatment. Here are several suggested comments that the authors could address to enhance the manuscript:

1. The manuscript requires revisions to enhance clarity and readability. The authors should clarify certain sections to ensure their intended meaning is clear to the reader. Additionally, there are ambiguous statements that need to be revised for better comprehension. Coherence should be improved to ensure a logical flow of ideas between paragraphs. By addressing these specific areas for improvement, the authors can significantly enhance the overall quality of the manuscript, making it more accessible to readers. I have provided specific examples below:

a) Until the elimination of vertical transmission is achieved, therefore, children born to mothers with HIV require programmatic strategies to reduce morbidity and mortality.

b) There is therefore a critical need to identify and distinguish children with HIV from children who are HEU across different periods of transmission, given the distinct cotrimoxazole strategy required in each group.

c) Even where infants are tested, challenges remain across the testing-to-treatment cascade, with delay in the return of results, lack of integration with child health services, loss to follow-up, and delayed ART initiation.

d) The model starts at age 6 weeks when early infant diagnosis and cotrimoxazole initiation are recommended and estimates mortality to 24 months of age

e) Sensitivity analysis (results) paragraph one: "Reducing perinatal and postnatal vertical transmission rates towards elimination targets and beyond decreased the effect of cotrimoxazole on mortality until it reached zero effect at 0% transmission."

f) Once HIV was diagnosed (by 9 months or EoB), 76% mortality reduction from ART plus 43% mortality reduction from cotrimoxazole was applied, for the 77% of children with HIV who start ART

2. The authors referenced "expert opinion" as a data source in Supplementary Table 1. To improve clarity, it is recommended to provide additional details in the method section regarding the nature of the experts used and how "expert opinion" was utilised to extract data. Specifically, explain the qualifications or expertise of the experts involved and describe the process by which their opinions were obtained and incorporated into the analysis. Additionally, within the manuscript, it would be beneficial to clarify the assumptions ("Since data on subsequent testing are not available, we assumed that 80% of children undergoing EID would have a 9-month test, and 50% would have an EoB test, as these children would likely be engaged in the test-to-treatment cascade. For children not undergoing 162 EID at 6 weeks, we assumed the probability of subsequent testing was much lower (10% for 9-month 163 test, 30% for EoB), as these children were likely less engaged with services.") that were made for testing if they are based on expert opinion. This will help readers understand the methodology and the basis for the analysis, ensuring transparency and reproducibility.

3. The authors state that in all strategies, they assume that infants testing positive for HIV will initiate cotrimoxazole treatment. However, it would be beneficial for the manuscript to explicitly mention the probability of an HIV test yielding a positive result and provide clarification regarding the specific time points at which testing was assigned in the model. This additional information will enhance the transparency of the study and enable readers to better understand the assumptions made regarding HIV test outcomes and the timing of testing within the modeling.

4. If the following details mentioned inside the manuscript are linked with comment 3 please revise the manuscript accordingly. "To determine the probability of a child testing HIV-positive by age 6 weeks, perinatal transmission rates were obtained from country-specific UNAIDS estimates from the most recent year with available data for all countries, including six-week transmission rates (MOZ 6%, ZWE 152 5%, CIV 4%, and UGA 3%) (2). For postnatal transmission, we calculated the cumulative probability of acquiring HIV by age 9 months (MOZ 3.4%, ZWE 1.7%, CIV 1.7%, UGA 1.3%), and by the end of breastfeeding (MOZ 4.1%, ZWE 2.0%, CIV 2.1%, UGA 1.6%), based on derived weekly transmission rates from country-specific final vertical transmission rates and breastfeeding duration, making the assumption that postnatal transmission rates were consistent over time."

5. The authors mentioned in the manuscript that globally, 62% of infants currently receive the recommended virological test within two months of birth, and as a result, 40% of children with HIV remain undiagnosed. It would be helpful if the authors could clarify whether these specific values were utilized within the model being discussed in the study. Providing this clarification will enable readers to better understand the extent to which these statistics were incorporated into the modeling process.

6. Please clarify within the manuscript whether the model explicitly incorporates the following strategies: providing antiretroviral therapy (ART) for undiagnosed cases and cotrimoxazole for cases testing positive for HIV, as well as whether it considers the progression of the disease among children with HIV. This clarification is important for understanding the interventions considered in the study and their potential impact on disease progression and treatment outcomes.

7. Please clarify within the manuscript whether the model explicitly incorporates the following strategies among children with HIV:

a) Providing antiretroviral therapy (ART) for undiagnosed cases and cotrimoxazole for cases testing positive for HIV

b) Disease progression among children with HIV into the mode

8. Could the author kindly include the confidence interval (CI), if possible, for the risk ratio of strategies in comparison to the current strategy?

9. Could the authors please provide a summary of the main findings? This summary should be provided as a paragraph at the beginning of the discussion section.

10. Could the authors please provide a clear and detailed discussion of the limitations of the research? Additionally, it would be valuable to explore the need for further research, such as conducting a cost-effectiveness analysis and examining the potential impact on antibiotic resistance for the proposed strategies.

Reviewer #4: The manuscript reports the impact of different cotrimoxazole strategies on mortality among children born to HIV-positive mothers. This work may help inform WHO recommendations on cotrimoxazole prophylaxis given some concerns on its necessity among HIV-exposed children given concerns on antibiotic resistance.

The manuscript is well written and informative: I only have minor comments regarding the methods:

Model structure

In my opinion, the author should clearly state what they consider as "early initiation diagnosis (EID)" before detailing the model in order to clarify it.

Two model structures are provided: Figure 1 and Supplementary table 1. In my opinion, providing two model structures may be misleading and authors may consider clearly explaining the differences between both. Regarding the model structure in Supplementary Material, the document states that infants are categorized according to whether they have acquired perinatal HIV and whether they undergo early infant diagnosis. If I understand well, perinatal HIV is defined as acquiring HIV by 6 weeks, but the model classifies the children as born with HIV or not, which in my opinion is different.

Data sources and parameters

Data on subsequent testing after EID and engagement of children with HIV services: in my opinion it would be important to support the assumptions with literature. For Mozambique, I suggest to read the "Inquérito de Indicadores de Imunização, Malária e HIV/SIDA" (https://www.misau.gov.mz/index.php/inqueritos-de-saude?download=566:relatorio-final-imasida). Despite it dates from 2015, I think it might provide useful information to inform the assumptions.

Any attachments provided with reviews can be seen via the following link:

[LINK]

Attachment

Submitted filename: Review Plos Medicine.docx

pmed.1004334.s003.docx^{(12.9KB, docx)}

PLoS Med. 2024 Feb 20;21(2):e1004334. doi: 10.1371/journal.pmed.1004334.r003

Author response to Decision Letter 1

3 Oct 2023

Attachment

Submitted filename: CTX Modelling PLoS Medicine REVISION RESPONSE.docx

pmed.1004334.s004.docx^{(40KB, docx)}

PLoS Med. doi: 10.1371/journal.pmed.1004334.r004

Decision Letter 2

Louise Gaynor-Brook

24 Oct 2023

Dear Dr. Mathur,

Thank you very much for submitting your manuscript "Estimating the Impact of Alternative Programmatic Cotrimoxazole Strategies on Mortality Among Children Born to Mothers with HIV: A Modelling Study" (PMEDICINE-D-23-01623R2) for consideration at PLOS Medicine.

Your paper was re-evaluated by three independent reviewers, including a statistical reviewer, and discussed among all the editors here. The reviews are appended at the bottom of this email and any accompanying reviewer attachments can be seen via the link below:

[LINK]

In light of these reviews, I am afraid that we will not be able to accept the manuscript for publication in the journal in its current form, but we would like to consider a revised version that fully addresses Reviewer 1's comments. Obviously we cannot make any decision about publication until we have seen the revised manuscript and your response, and we plan to seek re-review by one or more of the reviewers.

We expect to receive your revised manuscript by Nov 14 2023 11:59PM. Please email me (lgaynor@plos.org) if you have any questions or concerns.

Please use the following link to submit the revised manuscript:

https://www.editorialmanager.com/pmedicine/

Your article can be found in the "Submissions Needing Revision" folder.

We look forward to receiving your revised manuscript.

Sincerely,

Louise Gaynor-Brook, MBBS PhD

Senior Editor, PLOS Medicine

plosmedicine.org

lgaynor@plos.org

-----------------------------------------------------------

Requests from the editors:

In your revised submission, please ensure that you fully address the comments from Reviewer 1, relating to strengthening the model to take into account cotrimoxazole usage among infants not engaged in care, and consideration of the Mpepu RCT as suggested.

Comments from the reviewers:

Reviewer #1: Mathur and colleagues have significantly improved their manuscript entitled "The Impact of Alternative Programmatic Cotrimoxazole Strategies on Mortality Among Children Born to Mothers with HIV: A Modelling Study" by addressing opportunities raised by the editorial staff of PLOS Medicine, as well as that of the reviewers. Again noted is the considerable effort in further customizing the multiple strategies to quantify predicted mortality. There remain opportunities to further strengthen the manuscript, including:

The premise of this modeling exercise is that infants and children living with HIV are not diagnosed and treated proximal to the incident HIV infection. While cotrimoxazole does have mortality benefit among children infected with HIV, all models presume there will be no change in pediatric HIV testing and rapid antiretroviral treatment initiation. In the limitations paragraph of the discussion section it should be pointed out that investment in ensuring that all children HIV-exposed uninfected receive cotrimoxazole until risk of HIV infection has been eliminated may not represent the optimal health investment compared to investing in timely testing and treatment, an alternative known to substantially reduce morbidity and mortality, preserve immune function, and limit seeding of the HIV viral reservoir, that latter two of which cotrimoxazole cannot affect.

At lines 222 through 224, while it is appropriate for the purposes of modeling to assume that children who do not test at 6 weeks would have a lower prevalence of testing at 9 months and following cessation of breastfeeding, as the authors appropriately note this group is less likely to be engaged in care, it is imperative that this population also have a much lower use of cotrimoxazole in model. If they are not engaged in care, they will not be prescribed cotrimoxazole. Reducing the proportion of infants in this group who actually take cotrimoxazole will likely substantially reduce the overall mortality benefit of a policy of universal use of cotrimoxazole in high burden HIV settings.

At lines 258 and 259, rationale is used for selection of conservative estimates of mortality benefit with cotrimoxazole use, since the CHAP study included very few infants. Why wouldn't the authors use the Mpepu study, a more current study were infants HIV-exposed uninfected were randomized to cotrimoxazole or placebo through 15 months of life? This is particularly important as at lines 372 through 374 it is stated that there is no randomized controlled trial data for cotrimoxazole among breastfed infants. The Mpepu study in Botswana was randomized controlled trial that included breastfed infants.

At lines 361 through 362, in addition to recognizing to the need to focus on reducing vertical transmission, it would be equally important to highlight, where vertical transmission does occur, timely testing and initiation of antiretroviral treatment.

Under "Why was this study done?", under the first bullet, how would one ascertain the period of HIV infection risk through breastfeeding has ended without performing testing? It might be clearer to the reader to state "until the child is no longer at risk for HIV acquisition and documentation of a negative HIV test appropriately timed after the risk has concluded is available".

For the statement between lines 91-94, it would be important to use the term "model" in place of "determine" and/or to state "would be predicted to increase mortality" rather than "would increase mortality".

At line 100, can you clarify how the model "incorporates the HIV status of the infant"? All infants/children in the model start off HIV-exposed uninfected and that same bullet point goes on to explain that perinatal and postnatal transmission rates are employed in the model.

Please consider changing the text at lines 121 and 122 to read "Policymakers need to weigh the risks and benefits of guidelines for cotrimoxazole prophylaxis among infants HIV-exposed uninfected, recognizing that vertical transmission rates and timely test and treat programming significantly influences predicted mortality outcomes".

The text between lines 164 and 166 indicate that modeling of mortality impact of alternative cotrimoxazole strategies in four high-burden settings was performed to inform potential cotrimoxazole recommendations in different contexts. However, modeling in four high-burden settings really does not afford the ability to inform recommendations in different contexts. It is more accurate to simply state "Here, we model the mortality impact of alternative cotrimoxazole strategies in four high-burden settings".

The manuscript has many places where first person language is not presented, including at line 92 (HIV-exposed infants should be infants HIV-exposed), line 173, line 204, line 246, and line 392.

PMTCT is used for the first time at line 115. Please spell it out.

Lines 180 and 181 do not need "Early Infant Diagnosis" as the abbreviation has already been used previously.

At line 87, the "now" is not necessary.

Reviewer #2: The authors have done a good job of replying to all my questions and concerns. I have no further concerns.

Reviewer #3: The authors have diligently addressed all comments, and I now find the manuscript suitable for publication in PLOS Medicine. The only minor revision I identified is related to the "What Did the Researchers Do and Find?" section. Specifically, there is duplication of the phrase "limited by" that should be removed. The sentence should read as follows: "Our study is limited by the lack of a cost-effectiveness analysis, data on cotrimoxazole uptake, and comprehensive antimicrobial resistance surveillance data in sub-Saharan Africa."

Any attachments provided with reviews can be seen via the following link:

[LINK]

PLoS Med. 2024 Feb 20;21(2):e1004334. doi: 10.1371/journal.pmed.1004334.r005

Author response to Decision Letter 2

13 Nov 2023

Attachment

Submitted filename: CTX Modelling PLoS Medicine REVISION2 RESPONSE.docx

pmed.1004334.s005.docx^{(21.8KB, docx)}

PLoS Med. doi: 10.1371/journal.pmed.1004334.r006

Decision Letter 3

Louise Gaynor-Brook

11 Dec 2023

Dear Dr. Mathur,

Thank you very much for re-submitting your manuscript "Estimating the Impact of Alternative Programmatic Cotrimoxazole Strategies on Mortality Among Children Born to Mothers with HIV: A Modelling Study" (PMEDICINE-D-23-01623R3) for review by PLOS Medicine.

I have discussed the paper with my colleagues and the academic editor and it was also seen again by two reviewers. I am pleased to say that provided the remaining editorial and production issues are dealt with we are planning to accept the paper for publication in the journal.

The remaining issues that need to be addressed are listed at the end of this email. Any accompanying reviewer attachments can be seen via the link below. Please take these into account before resubmitting your manuscript:

[LINK]

In revising the manuscript for further consideration here, please ensure you address the specific points made by each reviewer and the editors. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments and the changes you have made in the manuscript. Please submit a clean version of the paper as the main article file. A version with changes marked must also be uploaded as a marked up manuscript file.

Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. If you haven't already, we ask that you provide a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract.

We expect to receive your revised manuscript within 1 week. Please email me (lgaynor@plos.org) if you have any questions or concerns.

We ask every co-author listed on the manuscript to fill in a contributing author statement. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT.

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript.

Please note, when your manuscript is accepted, an uncorrected proof of your manuscript will be published online ahead of the final version, unless you've already opted out via the online submission form. If, for any reason, you do not want an earlier version of your manuscript published online or are unsure if you have already indicated as such, please let the journal staff know immediately at plosmedicine@plos.org.

If you have any questions in the meantime, please contact me or the journal staff on plosmedicine@plos.org.

We look forward to receiving the revised manuscript by Dec 18 2023 11:59PM.

Sincerely,

Louise Gaynor-Brook, MBBS PhD

Senior Editor, PLOS Medicine

lgaynor@plos.org

plosmedicine.org

------------------------------------------------------------

Thank you for your responses to the comments from the editors and reviewers in your revised submission. A few minor issues detailed below require further attention before we are able to accept your manuscript for publication.

Comments from the Academic Editor:

In some places in the discussion, the authors could change the order of the relevant sentence to start with insisting that HIV diagnosis and immediate treatment of infants found to be infected is priority, and that provision of cotrimoxazole should not be seen as a 'way out' and reducing the costs of the HIV prevention and care programmes, before stating that co-trimoxazole should be provided to HEU.

RE: availability of trials in breastfed populations - the authors should amend their sentence re no trials being available to note that only limited data is available re use of CTX in HEU in breastfeeding populations, and reference the Mpepu trial, where 20% of infants were breastfed. In many high HIV prevalence settings breastfeeding does happen but often not for that long - for a variety of reasons - which likely underlies the finding in Arikiwa that breastfeeding was not clearly associated with mortality in HEU infants.

There is no need to compare mortality between Mpepu and CHAPS trials.

Requests from the Editors:

To help us extend the reach of your research, please provide any Twitter handle(s) that would be appropriate to tag, including your own, your coauthors’, your institution, funder, etc.

Author Summary:

Line 111 - please revise to ‘Increased predicted morality…’ and ‘although it was predicted that cotrimoxazole would have fewer benefits’ or similar

Methods:

Please state early in the Methods section that your study did not have a prospective protocol.

Please add the following statement, or similar, to the Methods: "This study is reported as per the Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD) guideline (S1 Checklist)."

Results:

Lines 285, 295, 304, 309 - please revise to ‘predicted excess mortality’ (or similar)

Line 320 - please specify which supplementary files are relevant to each setting.

Figures:

Please ensure that all abbreviations are defined in the figure legend of each figure, including those in the supplementary information.

Comments from Reviewers:

Reviewer #1: Mathur and colleagues have made further changes to their manuscript entitled "The Impact of Alternative Programmatic Cotrimoxazole Strategies on Mortality Among Children Born to Mothers with HIV: A Modelling Study" in an effort to be responsive to reviewer suggestions. Two key points which were not adequately addressed in this latest revision should be addressed prior to publication.

First, it is again requested that in the limitations paragraph of the discussion section it be pointed out that investment in ensuring that all children HIV-exposed uninfected receive cotrimoxazole until risk of HIV infection has been eliminated may not represent the optimal health investment compared to investing in timely testing and treatment, an alternative known to substantially reduce morbidity and mortality, preserve immune function, and limit seeding of the HIV viral reservoir, the latter two of which cotrimoxazole cannot affect. While the response from the authors to this previous request characterize prescribing of cotrimoxazole as a "safety net" when gaps in the test-to-treat cascade exist, the "safety net" is the health care system. The modeling exercise does not sufficiently account for lack of engagement in the health care "safety net". It is again requested that the wording be clear that funding and programming of cotrimoxazole may not represent the best investment compared to investment in the test-to-treat cascade. It is recommended that this appear in the limitations paragraph of the discussion section.

Secondly, although the authors do reference the Botswana-based Mpepu study and the South African study, lines 379 through 383 continue to state:

"There are no randomized controlled trial data for cotrimoxazole among breastfed infants, but it is unlikely that mortality benefits are substantially lower than in the original CHAP trial, given the more rapid disease progression at this age, lack of prognostic markers, and higher risk of Pneumocystis jirovecii."

As previously pointed out, the Mpepu study does represent a randomized controlled trial where 20% of the infants in the study were breastfeeding and these infants were randomized to cotrimoxazole or placebo. Please change the statement in the discussion to reflect this and consider comparing Mpepu mortality data to that of the original CHAP trial. If the fact that mortality is only reported through 18 months is the reason Mpepu mortality data was not used, then the text should reflect that rather than indicating that no randomized clinical trial has occurred.

Reviewer #3: To enhance the clarity and precision of the manuscript, I propose a revision to lines 410 and 411 in the Discussion section as detailed below:

Lastly, this study did not consider the economic implications of the proposed strategies. Future research should investigate the potential cost savings and the financial impacts of implementing each strategy on national budgets.

Any attachments provided with reviews can be seen via the following link:

[LINK]

PLoS Med. 2024 Feb 20;21(2):e1004334. doi: 10.1371/journal.pmed.1004334.r007

Author response to Decision Letter 3

19 Dec 2023

Attachment

Submitted filename: CTX Modelling PLoS Medicine REVISION3 RESPONSE.docx

pmed.1004334.s006.docx^{(19.7KB, docx)}

PLoS Med. doi: 10.1371/journal.pmed.1004334.r008

Decision Letter 4

Louise Gaynor-Brook

10 Jan 2024

Dear Dr Mathur,

On behalf of my colleagues and the Academic Editor, Prof. Marie-Louise Newell, I am pleased to inform you that we have agreed to publish your manuscript "Estimating the Impact of Alternative Programmatic Cotrimoxazole Strategies on Mortality Among Children Born to Mothers with HIV: A Modelling Study" (PMEDICINE-D-23-01623R4) in PLOS Medicine.

Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow up email. Please be aware that it may take several days for you to receive this email; during this time no action is required by you. Once you have received these formatting requests, please note that your manuscript will not be scheduled for publication until you have made the required changes.

In the meantime, please log into Editorial Manager at http://www.editorialmanager.com/pmedicine/, click the "Update My Information" link at the top of the page, and update your user information to ensure an efficient production process.

PRESS

We frequently collaborate with press offices. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximise its impact. If the press office is planning to promote your findings, we would be grateful if they could coordinate with medicinepress@plos.org. If you have not yet opted out of the early version process, we ask that you notify us immediately of any press plans so that we may do so on your behalf.

We also ask that you take this opportunity to read our Embargo Policy regarding the discussion, promotion and media coverage of work that is yet to be published by PLOS. As your manuscript is not yet published, it is bound by the conditions of our Embargo Policy. Please be aware that this policy is in place both to ensure that any press coverage of your article is fully substantiated and to provide a direct link between such coverage and the published work. For full details of our Embargo Policy, please visit http://www.plos.org/about/media-inquiries/embargo-policy/.

Thank you again for submitting to PLOS Medicine. We look forward to publishing your paper.

Sincerely,

Louise Gaynor-Brook, MBBS PhD

Senior Editor

PLOS Medicine

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 TRIPOD Checklist. TRIPOD Checklist: Prediction model development.

(DOCX)

pmed.1004334.s001.docx^{(88.6KB, docx)}

S1 Text

(DOCX)

pmed.1004334.s002.docx^{(7MB, docx)}

Attachment

Submitted filename: Review Plos Medicine.docx

pmed.1004334.s003.docx^{(12.9KB, docx)}

Attachment

Submitted filename: CTX Modelling PLoS Medicine REVISION RESPONSE.docx

pmed.1004334.s004.docx^{(40KB, docx)}

Attachment

Submitted filename: CTX Modelling PLoS Medicine REVISION2 RESPONSE.docx

pmed.1004334.s005.docx^{(21.8KB, docx)}

Attachment

Submitted filename: CTX Modelling PLoS Medicine REVISION3 RESPONSE.docx

pmed.1004334.s006.docx^{(19.7KB, docx)}

Data Availability Statement

The data underlying the results presented in the study are available from https://osf.io/8kjgp/.

[pmed.1004334.ref001] 1.Joint United Nations Programme on HIV/AIDS. UNAIDS data 2022. Geneva: UNAIDS; 2022. [PubMed]

[pmed.1004334.ref002] 2.Joint United Nations Programme on HIV/AIDS. Start Free, Stay Free, AIDS Free Final report on 2020 targets. 2021.

[pmed.1004334.ref003] 3.Newell ML, Coovadia H, Cortina-Borja M, Rollins N, Gaillard P, Dabis F, et al. Mortality of infected and uninfected infants born to HIV-infected mothers in Africa: a pooled analysis. Lancet. 2004;364(9441):1236–1243. doi: 10.1016/S0140-6736(04)17140-7 . [DOI] [PubMed] [Google Scholar]

[pmed.1004334.ref004] 4.World Health Organization. Guidelines on post-exposure prophylaxis for HIV and the use of co-trimoxazole prophylaxis for HIV-related infections among adults, adolescents and children: recommendations for a public health approach: December 2014 supplement to the 2013 consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection. World Health Organization; 2014. [PubMed]

[pmed.1004334.ref005] 5.Church JA, Fitzgerald F, Walker AS, Gibb DM, Prendergast AJ. The expanding role of co-trimoxazole in developing countries. Lancet Infect Dis. 2015;15(3):327–39. Epub 20150121. doi: 10.1016/S1473-3099(14)71011-4 . [DOI] [PubMed] [Google Scholar]

[pmed.1004334.ref006] 6.Bwakura-Dangarembizi M, Kendall L, Bakeera-Kitaka S, Nahirya-Ntege P, Keishanyu R, Nathoo K, et al. A randomized trial of prolonged co-trimoxazole in HIV-infected children in Africa. N Engl J Med. 2014;370(1):41–53. doi: 10.1056/NEJMoa1214901 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004334.ref007] 7.Chintu C, Bhat GJ, Walker AS, Mulenga V, Sinyinza F, Lishimpi K, et al. Co-trimoxazole as prophylaxis against opportunistic infections in HIV-infected Zambian children (CHAP): a double-blind randomised placebo-controlled trial. Lancet. 2004;364(9448):1865–1871. doi: 10.1016/S0140-6736(04)17442-4 . [DOI] [PubMed] [Google Scholar]

[pmed.1004334.ref008] 8.Daniels B, Coutsoudis A, Moodley-Govender E, Mulol H, Spooner E, Kiepiela P, et al. Effect of co-trimoxazole prophylaxis on morbidity and mortality of HIV-exposed, HIV-uninfected infants in South Africa: a randomised controlled, non-inferiority trial. Lancet Glob Health. 2019;7(12):e1717–e1727. doi: 10.1016/S2214-109X(19)30422-X . [DOI] [PubMed] [Google Scholar]

[pmed.1004334.ref009] 9.Lockman S, Hughes M, Powis K, Ajibola G, Bennett K, Moyo S, et al. Effect of co-trimoxazole on mortality in HIV-exposed but uninfected children in Botswana (the Mpepu Study): a double-blind, randomised, placebo-controlled trial. Lancet Glob Health. 2017;5(5):e491–e500. doi: 10.1016/S2214-109X(17)30143-2 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004334.ref010] 10.Daniels B, Kuhn L, Spooner E, Mulol H, Goga A, Feucht U, et al. Cotrimoxazole guidelines for infants who are HIV-exposed but uninfected: a call for a public health and ethics approach to the evidence. Lancet Glob Health. 2022;10(8):e1198–e1203. doi: 10.1016/S2214-109X(22)00120-6 [DOI] [PubMed] [Google Scholar]

[pmed.1004334.ref011] 11.Teasdale CA, Zimba R, Abrams EJ, Sachathep K, Ndagije F, Nuwagaba-Biribonwoha H, et al. Estimates of the prevalence of undiagnosed HIV among children living with HIV in Eswatini, Lesotho, Malawi, Namibia, Tanzania, Zambia, and Zimbabwe from 2015 to 2017: an analysis of data from the cross-sectional Population-based HIV Impact Assessment surveys. Lancet HIV. 2022;9(2):e91–e101. doi: 10.1016/S2352-3018(21)00291-5 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004334.ref012] 12.Ciaranello AL, Park JE, Ramirez-Avila L, Freedberg KA, Walensky RP, Leroy V. Early infant HIV-1 diagnosis programs in resource-limited settings: opportunities for improved outcomes and more cost-effective interventions. BMC Med. 2011;9:59. Epub 20110520. doi: 10.1186/1741-7015-9-59 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004334.ref013] 13.Chatterjee A, Tripathi S, Gass R, Hamunime N, Panha S, Kiyaga C, et al. Implementing services for Early Infant Diagnosis (EID) of HIV: a comparative descriptive analysis of national programs in four countries. BMC Public Health. 2011;11:553. Epub 20110713. doi: 10.1186/1471-2458-11-553 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004334.ref014] 14.Wedderburn CJ, Evans C, Slogrove AL, Rehman AM, Gibb DM, Prendergast AJ, et al. Co-trimoxazole prophylaxis for children who are HIV-exposed and uninfected: a systematic review. J Int AIDS Soc. 2023;26(6):e26079. doi: 10.1002/jia2.26079 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004334.ref015] 15.Arikawa S, Rollins N, Jourdain G, Humphrey J, Kourtis AP, Hoffman I, et al. Contribution of Maternal Antiretroviral Therapy and Breastfeeding to 24-Month Survival in Human Immunodeficiency Virus-Exposed Uninfected Children: An Individual Pooled Analysis of African and Asian Studies. Clin Infect Dis. 2018;66(11):1668–1677. doi: 10.1093/cid/cix1102 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004334.ref016] 16.Joint United Nations Programme on HIV/AIDS. UNAIDS data 2021. Geneva: UNAIDS; 2021. [PubMed]

[pmed.1004334.ref017] 17.Luo R, Fong Y, Boeras D, Jani I, Vojnov L. The clinical effect of point-of-care HIV diagnosis in infants: a systematic review and meta-analysis. Lancet. 2022;400(10356):887–895. doi: 10.1016/S0140-6736(22)01492-1 . [DOI] [PubMed] [Google Scholar]

[pmed.1004334.ref018] 18.Violari A, Cotton MF, Gibb DM, Babiker AG, Steyn J, Madhi SA, et al. Early antiretroviral therapy and mortality among HIV-infected infants. N Engl J Med. 2008;359(21):2233–2244. doi: 10.1056/NEJMoa0800971 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004334.ref019] 19.World Bank. World development indicators 2022. The World Bank; 2022. https://data.worldbank.org/. Accessed 2022 Dec 2.

[pmed.1004334.ref020] 20.Mofenson LM, Cohn J, Sacks E. Challenges in the early infant HIV diagnosis and treatment cascade. J Acquir Immune Defic Syndr. 2020;84:S1–S4. doi: 10.1097/QAI.0000000000002366 [DOI] [PubMed] [Google Scholar]

[pmed.1004334.ref021] 21.Frescura L, Godfrey-Faussett P, Feizzadeh AA, El-Sadr W, Syarif O, Ghys PD, et al. Achieving the 95 95 95 targets for all: A pathway to ending AIDS. PLoS ONE. 2022;17(8):e0272405. Epub 20220804. doi: 10.1371/journal.pone.0272405 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004334.ref022] 22.Tadesse BT, Ashley EA, Ongarello S, Havumaki J, Wijegoonewardena M, Gonzalez IJ, et al. Antimicrobial resistance in Africa: a systematic review. BMC Infect Dis. 2017;17(1):616. Epub 20170911. doi: 10.1186/s12879-017-2713-1 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004334.ref023] 23.Powis KM, Souda S, Lockman S, Ajibola G, Bennett K, Leidner J, et al. Cotrimoxazole prophylaxis was associated with enteric commensal bacterial resistance among HIV-exposed infants in a randomized controlled trial, Botswana. J Int AIDS Soc. 2017;20(3). doi: 10.1002/jia2.25021 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004334.ref024] 24.Antimicrobial Resistance C. Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. Lancet. 2022;399(10325):629–55. Epub 20220119. doi: 10.1016/S0140-6736(21)02724-0 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004334.ref025] 25.Pouwels KB, Batra R, Patel A, Edgeworth JD, Robotham JV, Smieszek T. Will co-trimoxazole resistance rates ever go down? Resistance rates remain high despite decades of reduced co-trimoxazole consumption. J Glob Antimicrob Resist. 2017;11:71–4. Epub 20170731. doi: 10.1016/j.jgar.2017.07.013 . [DOI] [PubMed] [Google Scholar]

[pmed.1004334.ref026] 26.Ryan M, Griffin S, Chitah B, Walker AS, Mulenga V, Kalolo D, et al. The cost-effectiveness of cotrimoxazole prophylaxis in HIV-infected children in Zambia. AIDS. 2008;22(6):749–757. doi: 10.1097/QAD.0b013e3282f43519 . [DOI] [PubMed] [Google Scholar]

PERMALINK

Estimating the impact of alternative programmatic cotrimoxazole strategies on mortality among children born to mothers with HIV: A modelling study

Shrey Mathur

Melanie Smuk

Ceri Evans

Catherine J Wedderburn

Diana M Gibb

Martina Penazzato

Andrew J Prendergast

Roles

Abstract

Background

Methods and findings

Conclusions

Author summary

Why was this study done?

What did the researchers do and find?

What do these findings mean?

Introduction

Methods

Model structure

Fig 1. Schematic illustration of model structure (scenarios and strategies).

Data sources and parameters

Scenarios

Sensitivity analyses

Data access and ethics

Results

Table 1. Comparison of predicted mortality, between ages 6 weeks to 2 years, with different programmatic cotrimoxazole strategies for children born to mothers with HIV.

Fig 2. Excess deaths per year by country under alternative cotrimoxazole strategies.

Sensitivity analyses

Fig 3. Sensitivity analysis (Mozambique)—Conservative scenarios.

Discussion

Supporting information

Acknowledgments

Abbreviations

Data Availability

Funding Statement

References

Decision Letter 0

Alexandra Schaefer

Roles

Decision Letter 1

Louise Gaynor-Brook

Roles

Author response to Decision Letter 1

Decision Letter 2

Louise Gaynor-Brook

Roles

Author response to Decision Letter 2

Decision Letter 3

Louise Gaynor-Brook

Roles

Author response to Decision Letter 3

Decision Letter 4

Louise Gaynor-Brook

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases