Effect of perinatal sterile gloves on neonatal mortality: a post hoc analysis of a randomised clinical trial in Zambia

Introduction

Neonatal infection is a primary cause of neonatal mortality worldwide, especially in low- and middle-income countries (LMICs) (1). In an international, multi-site, prospective observational study from Bangladesh, Ethiopia, Pakistan, India, Rwanda, Nigeria, and South Africa, the incidence of clinically suspected sepsis was 166.0 per 1,000 live births, and that of laboratory-confirmed sepsis was 46.9 per 1,000 live births (2). People affected by conflict are suffering in many parts of the world. The military actions in Sudan, Ukraine, and Gaza in recent years have been deeply shocking. Basic necessities such as clean water, food, safe shelter, and sanitation are in short supply, and health systems are consistently compromised (3). Effective and efficient prophylaxes for neonatal infections are needed to target interventions to reduce their incidence and subsequent mortality (4).

The World Health Organization (WHO) promotes hand hygiene in healthcare settings through its annual global campaign, “SAVE LIVES: Clean Your Hands” on 5 May each year (5). To help monitor and implement a safe health care environment and clean birth practices in health care facilities, the WHO has identified ‘the six cleans’ (6), which are guidelines for reducing infections in newborns and mothers and are important prerequisites for hygienic childbirth. Individual financial constraints, difficulty accessing clean delivery kits (CDKs) and religious or cultural beliefs have limited the widespread use of these potentially life-saving interventions (7). Currently, conflicts around the world continue to occur with water shortages and deteriorating sanitary conditions (8), limiting the application of CDKs and certain components. In LMICs, more than half of all deliveries take place at home, and neonatal infections are 3–20 times more common than in newborns born in health facilities in high-income countries (7). CDK use was associated with reductions in early newborn mortality at both facility deliveries and home birth (9). One review (10) revealed the positive impact of CDKs on reducing the incidence of neonatal morbidity and mortality. However, other studies have reached different conclusions (4,7). The constituent elements of CDKs were substantially similar in contemporary studies. However, there was still some variation in the specific components in different implementation settings. This variation highlights the need for rigorous research to establish the clinical significance of each component. Relatively little research has been conducted to assess the impact of the individual components of CDKs.

Given that sterile gloves are among the most commonly used components of CDKs and that it is easier to obtain a single glove as opposed to a CDK in LMICs, we hypothesize that the use of sterile gloves would be associated with neonatal mortality. We plan to conduct this cohort study utilizing data from the Zambia Chlorhexidine Application Trial (ZamCAT), a cluster-randomised controlled trial (cRCT) that compares the effects of 4% chlorhexidine with dry cord care on neonatal mortality rates (11) with the aim of evaluating the associations between perinatal sterile glove use and neonatal mortality in home and facility deliveries. We present this article in accordance with the STROBE reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2025-88/rc).

Methods

Study design

This was a post hoc secondary analysis of a cRCT comparing daily cord cleansing with 4% chlorhexidine to the standard Ministry of Health practice (dry cord care) in Zambia.

Study population

Between 15 February 2011 and 30 January 2013, a cohort of 39,679 pregnant women in their second or third trimester, aged 15 years and older, planning to stay in the catchment area until 28 days postpartum, and willing to provide informed consent and complete cord care according to cluster assignment were enrolled. During a home visit 2 weeks after enrollment, all participants were provided a standard CDK containing sterile gloves, a bar of soap, a sterile razor blade, 2 cord clamps, a plastic sheet, matches, and a candle, and were instructed by field monitors on how to use the CDKs. At each of the 4 postnatal visits, maternal and neonatal health status was assessed and documented. The informed consent was obtained from all participating pregnant women. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Data collection

At the postpartum visit on day 4, the field monitors asked mothers about the use of each component of the CDKs, which was categorized as a binary variable on the basis of whether it was used. The data were collected in tabular paper forms designed in the Teleforms system (HP, Cambridge, UK). The forms, which were checked by supervisors for illegible and missing information, were transported to the central study office in Choma, where they were scanned and entered into a Microsoft Access database by a data management team. The data is available at Boston University’s data repository: https://open.bu.edu/handle/2144/42348. The following data were extracted from the main study: ZamCAT study ID, cluster number, maternal age, marital status, mother’s education level, date of enrollment, gestational age (GA) at enrollment (weeks), baby’s sex, birth weight (BW) (g), place of delivery, use of CDKs components, type of infant death, date of delivery, date of last event, age at death (days), original cause of death, post-autopsy cause of death and date of presence of infectious risk signs at visit (including day 1, day 4, day 10, and day 28).

Outcome

The primary endpoint was all-cause neonatal mortality within 28 days postpartum, categorized as immediate if in the first 24 hours of life, early if at 1–6 days of life, and late if at 7–28 days of age (12). Stillbirth was defined as a neonate who did not cry, breathe, or move at the time of delivery at ≥20 weeks gestation. The cause of death was determined by verbal autopsy, not clinical/physical autopsy. Neonatal deaths were documented via systematic oral interviews with the mothers. For deaths that occurred at home, verbal autopsies were completed, including information on the time of death, birth and surrounding circumstances; trained clinicians then reviewed the verbal autopsy reports and determined the cause of death. For deaths in health facilities, the cause of death depended on the diagnosis of the serious adverse event.

Statistical analysis

The data sets were analyzed with R (http://www.R-project.org) and EmpowerStats software (www.empowerstats.com, X&Y solutions, Inc., Boston, MA, USA). The data were initially preprocessed and checked for outliers. Women with a nonviable pregnancy outcome (stillbirth, abortion, miscarriage, or loss to follow-up after delivery) or missing data on glove use, newborn outcome, or delivery location were excluded. All analyses conducted here are exploratory. Quantitative variables were converted into categorical variables according to tertiles. Bivariate logistic regression was used to assess the effect of gloves on neonatal mortality, stratified by delivery location and time of death. The baseline characteristics selected as important covariates were based on the main study, which focused on known risk factors for neonatal mortality and sepsis, including low BW, prematurity, newborn sex, maternal education, and maternal age (11). In addition, other potential confounders, selected on the basis of the statistical analysis plan for the main trial, including socioeconomic status and delivery location, were considered if they were associated with the outcome variable with a P value <0.2 in the univariate models. GA was not adjusted for in the multivariate model because of collinearity with BW (13). The interaction between potentially confounding variables and gloves was also assessed. Sensitivity analyses were also performed via logistic models between the infection-related neonatal mortality and noninfection-related neonatal mortality groups.

Results

Study population characteristics

Prior to delivery, 1,542 women (3.8%) were lost to follow up, withdrew, experienced false pregnancies or underwent spontaneous or elective abortions, and six women died. A total of 38,131 women gave birth to 38,642 children. We used data from 38,642 births available between 15 February 2011 and 30 January 2013. Births with missing information on glove use, neonatal outcome or place of birth were excluded from the analysis (n=991 births). A total of 648 stillborn, aborted, miscarried and lost to follow-up after delivery were excluded, resulting in 37,003 newborns included in the 28-day analysis (Figure 1). Data on glove use were missing for 1.9% (n=745) of the births. Given that few data points were missing, we did not present differences between infants with missing data on glove use and those with complete data. On average, women were 25.6 years old, 82.5% were married, 39.7% had completed upper primary education (grades 5–7), and 50.0% delivered at a health center. The average number of infants was 28.2 GA, and 50.5% were female. The overall neonatal mortality rate was 14 deaths per 1,000 live births. Gloves were used in 96.9% of the cases. There were statistically significant differences between the use and nonuse of gloves in terms of maternal age, BW, plurality of infants, marital status, place of delivery, and neonatal mortality. There were no significant differences in the remaining baseline factors between the two groups (Table 1).

Figure 1 Flowchart of the study.

Perinatal sterile gloves and risk of neonatal mortality

The results of the univariate logistic regression analysis are shown in Table 2. Covariates such as the male sex of the newborn and doubleton were associated with increased neonatal mortality. Single and separated mothers were more likely to have a neonatal death than were married mothers. Older age (21–27 and 28–56 years), lower educational level and delivery at a health center or at home were associated with higher neonatal mortality. In addition, the lower the BW of the newborn was, the greater the risk of neonatal mortality. Unadjusted logistic regression revealed an association between glove use and lower neonatal mortality [odds ratio (OR) =0.13; 95% confidence interval (CI): 0.10–0.16]. Table 3 shows that perinatal glove use and neonatal mortality varied across the 3 places of delivery. Table 4 shows the associations between perinatal glove use and neonatal mortality according to logistic regression adjusted for other risk factors. Three models were estimated for the outcome: a model adjusting for BW (Model 0), a model adjusting for BW and infant sex (Model I), and a model adjusting for infant sex, BW, clustering, maternal age, marital status, maternal education, and place of birth (Model II). The use of gloves was associated with significantly lower odds of death in the 3 models. In Model II, with sterile gloves, we found a 90% relative reduction in mortality for all deliveries (OR =0.10; 95% CI: 0.07–0.15). We generated an E value to assess the sensitivity to unmeasured confounding in Model II. The primary findings were robust unless there was an unmeasured confounder with an OR greater than 19.5.

Findings from cause-of-death data

The 28-day mortality rates for hospital, health center and nonfacility deliveries were 2.7%, 1.1% and 1.3%, respectively (P<0.001). Cause of death data from post-autopsy were available for 261 of 521 newborn deaths. The most common cause overall was newborn sepsis/meningitis, accounting for 80 (30.6%) of the 261 newborn deaths, with infectious causes, including sepsis/meningitis, maternal infections that can affect the fetus, diarrhea, tetanus, and pneumonia, accounting for 105 (40.2%) of the neonatal deaths.

To assess the plausibility of the effect sizes, we used cause-specific mortality data to examine the associations of glove use with infection-related neonatal death and with death due to other noninfection-related neonatal deaths. The cause of death after autopsy was categorized as ‘infection-related death’ if it was sepsis/meningitis; maternal infections that can affect the fetus; diarrhea; pneumonia and tetanus; or ‘non-infection-related death’ if otherwise. When the analysis was restricted to patients with infection-related death, glove use was associated with a large relative reduction in ‘noninfection-related death’ (OR =0.09; 95% CI: 0.06–0.13) after adjustment for the confounders mentioned above. Risk-adjusted mortality for infection-related deaths was not performed because it did not fit into the regression model because of the small number of events.

Multivariable-adjusted logistic regression revealed decreased odds of death within 24 hours when sterile gloves were used (OR =0.47; 95% CI: 0.24–0.91). A reduction in early neonatal death (1–6 days) was associated with the use of sterile gloves (OR =0.26; 95% CI: 0.16–0.39). In contrast, glove use was not significantly associated with late neonatal death (7–28 days) (P=0.64) (Table 5).

Discussion

This retrospective cohort study revealed that the use of perinatal sterile gloves was associated with a reduction in neonatal mortality after adjusting for covariates in the total study population and in different delivery locations. Gloves were used in 96.9% of the cases. Reductions in early neonatal death (1–6 days) and death within 24 hours were associated with the use of sterile gloves.

Several previous studies have explored the association between sterile gloves and the risk of neonatal mortality. Since the Maternal and Neonatal Tetanus Elimination Initiative, which included CDKs, was launched in 1999, the estimated proportion of neonatal mortality attributed to tetanus decreased 84% (14). Glove use after hand hygiene prior to patient and line contact was associated with fewer gram-positive bloodstream infections and possible central line-associated bloodstream infections in preterm infants (15). CDKs did not significantly reduce cause-specific neonatal mortality [the leading cause of death was infection (44%)] in a cRCT in Pakistan (7). Differences in outcome definitions, methodologies and sample sizes may partly explain the differences found in the role of CDKs. Two randomised controlled trials (RCTs) investigating the effects of gloves on neonatal outcomes are still ongoing and have not yet published their results [NCT03078335 (16) and NCT02208960 (17)]. According to an earlier joint report by the WHO and United Nations Children’s Fund summarizing 20 years of data on water and sanitation, one in 10 health facilities worldwide had no sanitation, and approximately 857 million people had no water in their health facility in 2021 (18). Water scarcity may affect hand hygiene compliance. Furthermore, in the current global context, including the Russian-Ukrainian conflict, inadequate water, sanitation and hygiene infrastructure pose major health risks to women and children. Gloves are undoubtedly a readily available method for maintaining a relatively clean environment. In addition, 73% of healthcare facilities offering neonatal health services did not meet the WHO ‘six cleans’ guidelines (6). We are still a long way from achieving this goal.

The development of antimicrobial gloves, most of which use antimicrobial chemicals and some of which use mechanical approaches, is a recent development in glove research (19). Current guidelines on the use of antimicrobial gloves do not recommend replacing hand hygiene practices with antimicrobial gloves, as antimicrobial gloves are not a solution but rather an additional precautionary measure to combat nosocomial infections. However, we believe that with the rapid development of science and technology, novel, multifunctional, low-cost glove designs will certainly benefit different types of populations in the near future.

Several caveats to this exploratory analysis must be mentioned, including its retrospective nature. First, the main limitation of our study was the use of older data (despite the excellent population-based sample data available to researchers). Second, as with all observational studies, although we controlled for known covariates, residual unmeasured confounders may have affected the validity of the analyses. Third, there may be recall bias in maternal recall of antenatal glove use. Fourth, we are also aware that approximately 50% of the patient death data (260 cases) are not from postmortem examinations. Fifth, reverse causation cannot be ruled out, with high-risk neonatal delivery areas receiving only sporadic or no cleaning attention.

Despite these limitations, we believe that our study provides useful insights. First, interventions that focus on perinatal sterile gloves alone would likely be more appropriate than interventions that include CDKs in low- and middle-income settings. We assessed the role of glove use on neonatal mortality separately from the Park et al.’s study (9). Logistic regression was additionally adjusted for multiple gestation (singleton and multiple) as well as neonatal sex, low BW, prematurity, maternal age and maternal education. Second, we provide valuable guidance on interventions to improve neonatal health in low- and middle-income settings with a relatively large sample size. Third, the stratified analyses by place of delivery are a strength of this study.

Conclusions

The use of perinatal sterile gloves was associated with reduced neonatal mortality. This association persisted in analyses stratified by place of delivery. Our analyses revealed a significant association between glove use and improved neonatal outcomes within 24 hours and 7 days of birth. The provision of gloves to promote newborn health may be feasible in LMICs and conflict settings.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tp.amegroups.com/article/view/10.21037/tp-2025-88/rc

Peer Review File: Available at https://tp.amegroups.com/article/view/10.21037/tp-2025-88/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tp.amegroups.com/article/view/10.21037/tp-2025-88/coif). X.C. is from Empower U, X&Y Solutions Inc., a for-profit organization. However, the company had no role in the design, conduct, analysis, or interpretation of the study, nor in the preparation or decision to submit the manuscript for publication, no financial or non-financial conflicts of interest is related to this work. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Kuti BP, Ogunlesi TA, Oduwole O, et al. Hand hygiene for the prevention of infections in neonates. Cochrane Database Syst Rev 2023;6:CD013326. [Crossref] [PubMed]
2. Milton R, Gillespie D, Dyer C, et al. Neonatal sepsis and mortality in low-income and middle-income countries from a facility-based birth cohort: an international multisite prospective observational study. Lancet Glob Health 2022;10:e661-72. [Crossref] [PubMed]
3. Homer CSE. End humanitarian catastrophe in conflict settings. Lancet 2024;403:24-5. [Crossref] [PubMed]
4. Pell LG, Turab A, Bassani DG, et al. Effect of an integrated neonatal care kit on neonatal health outcomes: a cluster randomised controlled trial in rural Pakistan. BMJ Glob Health 2019;4:e001393. [Crossref] [PubMed]
5. Kilpatrick C, Tartari E, Pittet D, et al. "Accelerate Action Together" the 5 May 2023 World Health Organization SAVE LIVES- Clean Your Hands campaign. Infect Control Hosp Epidemiol 2023;44:1032-3. [Crossref] [PubMed]
6. Cronk R, Guo A, Folz C, et al. Environmental conditions in maternity wards: Evidence from rural healthcare facilities in 14 low- and middle-income countries. Int J Hyg Environ Health 2021;232:113681. [Crossref] [PubMed]
7. Duby J, Pell LG, Ariff S, et al. Effect of an integrated neonatal care kit on cause-specific neonatal mortality in rural Pakistan. Glob Health Action 2020;13:1802952. [Crossref] [PubMed]
8. Rawtani D, Gupta G, Khatri N, et al. Environmental damages due to war in Ukraine: A perspective. Sci Total Environ 2022;850:157932. [Crossref] [PubMed]
9. Park JH, Hamer DH, Mbewe R, et al. Components of clean delivery kits and newborn mortality in the Zambia Chlorhexidine Application Trial (ZamCAT): An observational study. PLoS Med 2021;18:e1003610. [Crossref] [PubMed]
10. Lassi ZS, Fisher Z, Andraweera P, et al. Effectiveness of birthing kits for clean childbirth: a systematic review. Int Health 2020;12:3-10. [Crossref] [PubMed]
11. Semrau KEA, Herlihy J, Grogan C, et al. Effectiveness of 4% chlorhexidine umbilical cord care on neonatal mortality in Southern Province, Zambia (ZamCAT): a cluster-randomised controlled trial. Lancet Glob Health 2016;4:e827-36. [Crossref] [PubMed]
12. Saito A, Kondo M. Association between maternal and child health care and neonatal death in Angola: a secondary analysis of Angola Demographic Health Survey 2015-16. Trop Med Health 2024;52:87. [Crossref] [PubMed]
13. Shah KP, deRegnier RO, Grobman WA, et al. Neonatal Mortality After Interhospital Transfer of Pregnant Women for Imminent Very Preterm Birth in Illinois. JAMA Pediatr 2020;174:358-65. [Crossref] [PubMed]
14. Jones CE, Yusuf N, Ahmed B, et al. Progress Toward Achieving and Sustaining Maternal and Neonatal Tetanus Elimination - Worldwide, 2000-2022. MMWR Morb Mortal Wkly Rep 2024;73:614-21. [Crossref] [PubMed]
15. Kaufman DA, Blackman A, Conaway MR, et al. Nonsterile glove use in addition to hand hygiene to prevent late-onset infection in preterm infants: randomized clinical trial. JAMA Pediatr 2014;168:909-16. [Crossref] [PubMed]
16. Khan S, Tsang KK, Hu ZJ, et al. GloveCare: a pilot study in preparation for a cluster crossover randomized controlled trial of non-sterile glove-based care in preventing late-onset infection in the NICU. Pilot Feasibility Stud 2023;9:50. [Crossref] [PubMed]
17. Pell LG, Bassani DG, Nyaga L, et al. Effect of provision of an integrated neonatal survival kit and early cognitive stimulation package by community health workers on developmental outcomes of infants in Kwale County, Kenya: study protocol for a cluster randomized trial. BMC Pregnancy Childbirth 2016;16:265. [Crossref] [PubMed]
18. Naddaf M. The world faces a water crisis - 4 powerful charts show how. Nature 2023;615:774-5. [Crossref] [PubMed]
19. How SW, Low DYS, Leo BF, et al. A critical review on the current state of antimicrobial glove technologies: advances, challenges and future prospects. J Hosp Infect 2023;137:24-34. [Crossref] [PubMed]