Chlorhexidine for Umbilical Cord Care

Published: April 2021

What is the problem?

Neonatal mortality is high in low- and middle-income countries, and infection is an important contributing factor to mortality in the neonatal period.

UNICEF estimates that in 2019, 2.4 million infants died in the first 28 days of life.1 Neonatal mortality, or the death of a newborn in the first four weeks of life, is a problem largely confined to low and middle-income countries, according to WHO.2 One paper estimates that Sub-Saharan Africa and South Asia comprise 79 percent of global neonatal mortality.3 A study across 194 countries found that higher neonatal mortality rates overall tended to correlate with a higher proportion of neonatal deaths occurring due to infection.4

More specifically, infection of the umbilical cord (omphalitis) is considered to be a significant cause of neonatal mortality in areas where neonatal mortality is high.5

We have not vetted these estimates, and we are unsure whether there is substantial regional variation in the portion of neonatal mortality attributable to infection — and, in high-mortality areas, infection of the umbilical cord in particular.

What is the program?

Chlorhexidine (CHX) is a topical antiseptic.6 When applied to the newborn umbilical stump at birth and/or daily in the weeks after birth, CHX may prevent bacterial colonization of the umbilical cord,7 resulting in fewer infections and deaths.8 In this review, we focus on the delivery of CHX by community health workers. Studies described here included participants with both home and facility-based deliveries.9

Current WHO recommendations for umbilical cord care support the application of CHX for home-based deliveries in high neonatal mortality settings (i.e. settings with 30 or more neonatal deaths per 1,000 live births).10

Does the program have strong evidence of effectiveness?

Overall, we believe there is strong evidence that topical CHX applied to the umbilical cord stump reduces neonatal mortality in the community setting. We are highly uncertain about how these effects will generalize, however, and expect there to be meaningful differences by region and/or baseline neonatal mortality rate.

We conducted a meta-analysis of five cluster-randomized community-based trials assessing the impact of cord cleansing with CHX on all-cause neonatal mortality.11 The analysis included three studies from South Asia and two from Sub-Saharan Africa. The South Asian studies exhibited higher neonatal mortality rates in the control group.12 The studies were all of high quality, with low risk of bias.13 The duration of CHX application ranged from 1 to 10 days, or up to three days after cord stump separation.14

We found that CHX application is associated with a 15% reduction in neonatal mortality across studies (95% confidence interval -2% to 29%),15 but there were meaningful differences in the size of the effect across studies that are unlikely to be due to chance.16

Studies with higher neonatal mortality in the control group tended to have a greater effect size,17 and observed effects were also different by region, with larger effects observed in South Asia than in Africa.18 Because neonatal mortality rates were also higher in the control groups in the South Asian trials, it is unclear whether region, baseline neonatal mortality rates, or other factors are contributing to the differences across these studies. Only one included study met the threshold for neonatal mortality (30 deaths per 1,000 live births) required to apply CHX per the current WHO criteria,19 and that study also exhibited the largest effect size (27% reduction in neonatal mortality).20 Other factors that may be contributing to differences in observed effects include:

• Differences in the level of neonatal mortality due to infection
• Differences in infection rates driven by the proportion of home vs. hospital deliveries in the studies
• Differences in baseline cord care practices
• Study-specific differences, such as the interventions offered to control groups, health worker and/or maternal adherence to CHX protocols, or the number of CHX administrations. Note that we briefly investigated whether the number of days of CHX application was associated with a differential effect on neonatal mortality, but we did not find evidence of such an association.21

A facility-based trial currently being conducted in an area of Uganda with a higher neonatal mortality rate will provide additional data that may help to disentangle the confounded influences of region and neonatal mortality rate on outcomes.22

Internal validity

Included studies exhibited a low risk of bias, as described in the study summary table. Although we don't have specific, serious concerns about internal validity, our expectation is that published studies are more likely to overstate an intervention's efficacy than understate it.

External validity

We have a high level of uncertainty about the extent to which findings from trials will generalize to future implementation, though we haven’t explored this issue in depth yet. Context or study-specific factors may matter in terms of the expected effect, as described above. We will take contextual factors into account if we are considering a specific funding opportunity, and at that time may revise our external validity estimate.

In general, we expect that a downward adjustment to effectiveness, relative to the level measured in published trials, is likely when an intervention is delivered at scale. On the other hand, in all five trials included in this analysis, participants in the control group were provided instructions on proper dry cord care practices and, in some cases, additional interventions like the provision of a clean delivery kit.23 We think that in future settings where topical CHX is applied to the umbilical cord stump, the standard-of-care would not include these additional interventions, which would suggest the effects of topical CHX would be larger in these future settings. We have assumed for the sake of this preliminary analysis that these competing considerations cancel each other out and thus have not included an external validity adjustment to the model; however, this is not based on any specific evidence and is highly uncertain.

Additional benefits

This intervention was found to reduce the incidence of omphalitis,24 but our cost-effectiveness model only incorporates benefits due to reductions in neonatal mortality. There is evidence of a long-term neuro-developmental cost associated with certain neonatal infections.25 Due to a lack of evidence regarding the incidence of these severe infections among cases of neonatal omphalitis and uncertainty around the omphalitis findings,26 we did not incorporate this into our model. By not taking these benefits into account, the model may underestimate the overall benefit of the program. We may investigate this further.

Are there any potential negative impacts of the program?

We believe that there are unlikely to be substantial adverse effects associated with this intervention.
In the only trial reporting mild adverse effects, nine out of 18,510 participants reported adverse events, eight of which were mild local skin irritation.27 None of the other trials reported any serious adverse effects.28 CHX is broadly regarded as safe to apply to the skin of neonates at standard concentrations (safety at higher concentrations is not known).29

Adverse effects have been noted with respect to the misuse of CHX: specifically, 45 cases of eye injuries in nine African countries resulting from improper application of CHX to the eyes have been reported to WHO as of February 2019.30 It is unclear what proportion of these injuries were in newborns.
CHX may carry some risk of promoting bacterial resistance. Evidence of bacterial resistance to CHX has been reported, but the extent and clinical significance of this is unclear.31

How cost-effective is it?

We conducted a preliminary cost-effectiveness analysis and found that CHX may be within the range of cost-effectiveness of the opportunities that we expect to direct marginal donations to as of 2021 (about 10x cash or higher). Cost-effectiveness is driven by baseline neonatal mortality, estimated impacts on neonatal mortality, and low estimated costs. Specifically:

• Baseline neonatal mortality: We use an estimate of neonatal mortality based on the World Bank’s reported average neonatal mortality of 27 per 1,000 in low-income country settings.32
• Effect size: CHX may reduce the probability of neonatal death by 15% in some circumstances.33
• Cost of the program: We are currently estimating that the program costs $5 per newborn, but we are highly uncertain (see "Program costs" below).

Note that our cost-effectiveness analyses are simplified models with significant limitations. We believe that cost-effectiveness estimates such as these should not be taken literally, due to the significant uncertainty around them. We provide these estimates (a) for comparative purposes and (b) because working on them helps us ensure that we are thinking through as many of the relevant issues as possible.

The model described above has a number of major uncertainties:

• The cost-effectiveness of technical assistance. Our best guess is that this intervention would be delivered via technical assistance within existing government service infrastructure like health facilities or community health worker programs. We believe that it is unlikely that the model presented here reflects the true costs and benefits of CHX technical assistance at scale. If we identify a giving opportunity, we will adapt this model to account for the likely costs and benefits of those technical assistance activities.
• Number of treatments. Interventions varied in the number of CHX treatments applied to each child. We have included 10 days of treatments in the cost-effectiveness model.
• Program costs. We are currently estimating that the program costs $5 per newborn, but we are highly uncertain. One paper indicated that the cost of a single-use tube of CHX in Nepal was $0.23 per dose, but we have not vetted this estimate.34 As noted above, we have assumed 10 doses of CHX used per child, which would result in a total of $2.30 per child treated at $0.23 per dose. To account for implementation costs, we have added $2.70 per child treated, such that delivering CHX costs a total of $5.00 per child. These cost estimates are highly likely to change, and changes in the cost could meaningfully change our view on the cost-effectiveness of the intervention. In our experience, as we learn more about a program, the cost per person reached is often significantly higher than it appeared in our preliminary estimates.
• Benefits of reduced infection. We have not included benefits of averting non-fatal infections in our model. By not taking these benefits into account, the model may underestimate the overall benefit of the program.
• Baseline neonatal mortality rate. We’ve used the average baseline neonatal mortality rate for low-income countries from the UN Inter-agency Group for Child Mortality Estimation.35 Actual neonatal mortality in countries of implementation will likely vary, which would impact the bottom line cost-effectiveness estimate.
• Deaths beyond day 28. Because deaths from infections tend to occur later in the neonatal period,36 it is likely that the intervention prevented deaths beyond week four. Because all studies examined deaths through day 28 only, the full scope of mortality reduction was likely not included in these estimates. We have not included a model for mortality reduction after day 28 in our cost-effectiveness model, and for this reason we may be underestimating the overall benefit of the program.
• External validity. In all five trials, participants in the control group were provided, at a minimum, instructions on proper cord care practices. In two trials, control group participants were offered sterile instruments for cord cutting – in another trial, control group participants were provided materials for soap and water washing of the umbilical stump. For this reason, we believe that population-based implementation of CHX for umbilical cord care might have a greater impact than that which was measured in the included studies. On the other hand, we are highly uncertain of the extent to which favorable results would translate when the intervention is scaled up. In our model we have assumed that these competing considerations cancel each other out and thus have not included an external validity adjustment to the model; however, this is not based on any specific evidence.
• Negative impacts. We have not accounted for any potential negative or offsetting impacts, including increased bacterial resistance or harms stemming from inappropriate use. We do not expect these factors to have a meaningful effect on the bottom line.

Does the program have room for more funding?

We have not yet carefully quantified the degree to which there is room for more funding in this area. In a recent conversation, PATH, an international global health nonprofit that is spearheading a CHX working group, indicated that there was room for more funding for its work in West Africa.37 We have not vetted this statement. We plan to continue to investigate room for more funding with other charities.

Key questions for further investigation

Below, we list areas we may research as we deepen our investigation of this program:

• What are the most likely underlying reasons for the heterogeneity observed across studies?
• How is chlorhexidine typically administered (i.e., by a healthcare provider or by the parent)? Does the typical method of administration match the administration method in the included studies? Does the effect size vary with type of administration?
• Does the prevalence of deaths due to umbilical cord infection vary meaningfully in contexts with high neonatal mortality rates, or is it relatively stable? Meaningful variation in deaths due to infection even in the context of high neonatal mortality could compromise the expected impact of this intervention.
• What is the benefit of averting a non-fatal umbilical cord infection?
• What giving opportunities are there to increase the use of chlorhexidine?
• Is technical assistance to governments for CHX scale-up cost-effective?
• What is the expected number of deaths averted after the 28-day mark?
• How does caregiver willingness to apply CHX vary in different contexts?

Our process

We reviewed the studies summarized in the most recent meta-analysis we could find. The meta-analysis has since been retracted, so we re-analyzed the data reported in the five articles summarized therein (see this spreadsheet, Table 1).

We also conducted a light literature review to determine whether any additional community-based trials should be included in this analysis and did not find any.

Sources

Document	Source
Arifeen et al. 2012	Source
Banerjee 2019	Source (archive)
Deeks et al. 2021	Source (archive)
GiveWell, Chlorhexidine CEA draft	Source
GiveWell, Chlorhexidine meta analysis	Source
GiveWell, Chlorhexidine studies summary	Source
GiveWell's non-verbatim summary of a conversation with Dr. Patricia Coffey, December 13, 2019	Source
Hodgins et al. 2013	Source (archive)
Hug et al. 2019	Source (archive)
Imdad et al. 2013	Source
Kohli-Lynch et al. 2017	Source (archive)
Mayo Clinic: Chlorhexidine, Topical Application Route	Source (archive)
Mullany et al. 2006	Source
Mullany, Darmstadt, and Tielsch 2006	Source (archive)
Nankabirwa et al. 2017	Source (archive)
Oza et al. 2015	Source (archive)
Roba et al. 2019	Source (archive)
Roba et al. 2019, retraction note	Source (archive)
Romoser 2016	Source (archive)
Sankar et al. 2016	Source (archive)
Sazawal et al. 2016	Source (archive)
Semrau et al. 2016	Source (archive)
Smales 1988	Source (archive)
Soofi et al. 2012	Source
UNICEF, "Neonatal mortality," 2020	Source (archive)
Weinstein et al. 2008	Source
WHO recommendations on postnatal care of the mother and newborn 2013	Source (archive)
WHO, "Newborn death and illness," 2011	Source (archive)
World Bank, Mortality rate, neonatal	Source (archive)

• 1

  "Globally, 2.4 million children died in the first month of life in 2019– approximately 6,700 neonatal deaths every day." UNICEF, "Neonatal mortality," 2020

• 2

  "Ninety-nine percent of newborn deaths occur in low and middle-income countries." WHO, "Newborn death and illness," 2011

• 3

  "In 2017, the annual NMR [neonatal mortality rate] was highest in west and central Africa, at 30.2 deaths per 1000 livebirths (90% uncertainty interval 25.7–37.2), and in south Asia, at 26.9 deaths per 1000 livebirths (24.1–30.3; figure 1; table). The annual NMR in these regions was more than 9 times higher than the average NMR in high-income countries, which was 3.0 deaths per 1000 livebirths (3.0–3.5). Together, south Asia and sub-Saharan Africa accounted for 79% of the total burden of neonatal deaths." Hug et al. 2019, p. e713

• 4

  Oza et al. 2015 finds that 11% of deaths in countries with <5 neonatal deaths per thousand are due to infection, compared to 56% of deaths in countries with >30 neonatal deaths per thousand. See p. 21, Appendix M, "Overall neonatal period" rows, "Median proportion (with IQR) by NMR level" columns. Deaths due to infection include deaths due to sepsis, tetanus, diarrhea and pneumonia. Sepsis accounts for the majority of infection-related deaths.

• 5

  "Infection of the umbilical cord stump (omphalitis), caused by skin bacteria, is a significant cause of illness and death in newborn babies in developing countries… In populations with high neonatal mortality rates, infections account for approximately half of all newborn deaths." Imdad et al. 2013, p. 2-4

• 6

  "Chlorhexidine belongs to a group of medicines called antiseptic antibacterial agents. It is used to clean the skin after an injury, before surgery, or before an injection." Mayo Clinic: Chlorhexidine, Topical Application Route

• 7

  "Of the different antiseptics available, chlorhexidine is the most studied agent in newborn infants. Research has shown a significant reduction in the rates of bacterial colonization of the umbilical cord after chlorhexidine application(s)." Sankar et al. 2016, p. S12

• 8

  "Infection of the umbilical cord stump (omphalitis), caused by skin bacteria, is a significant cause of illness and death in newborn babies in developing countries… In populations with high neonatal mortality rates, infections account for approximately half of all newborn deaths." Imdad et al. 2013, p. 2-4

• 9

  See the study summary spreadsheet for more information on delivery setting.

• 10

  "Daily chlorhexidine (7.1% chlorhexidine digluconate aqueous solution or gel, delivering 4% chlorhexidine) application to the umbilical cord stump during the first week of life is recommended for newborns who are born at home in settings with high neonatal mortality (30 or more neonatal deaths per 1000 live births). Clean, dry cord care is recommended for newborns born in health facilities and at home in low neonatal mortality settings. Use of chlorhexidine in these situations may be considered only to replace application of a harmful traditional substance, such as cow dung, to the cord stump." WHO recommendations on postnatal care of the mother and newborn 2013, p. 3

• 11

  We were unable to replicate the inputs to the 2019 meta-analysis conducted by Roba et al. 2019 (neonatal mortality inputs and findings summarized in Figure 2, p. 5), which has since been retracted, so we re-analyzed the data reported in the five articles summarized therein (see this spreadsheet, Table 1). We also conducted a brief literature review to determine whether any additional community-based trials should be included in this analysis and did not find any. One facility-based trial is currently (as of February 2021) underway in Uganda (see Nankabirwa et al. 2017).

• 12

  All South Asian trials had higher neonatal mortality rates in the control groups than the African trials:

  • South Asia: 28.3 per 1,000 (Bangladesh, Arifeen et al. 2012, table 4, p. 1026), 19.3 per 1,000 (Nepal, Mullany et al. 2006, table 5, p. 915), and 36.1 per 1,000 (Pakistan, Soofi et al. 2012, table 2, p. 1033)
  • Africa: 11.7 per 1,000 (Tanzania, Sazawal et al. 2016, table 3, p. e843), and 13.6 per 1,000 (Zambia, Semrau et al. 2016, table 4, p. 7)

• 13

  For more details on study quality, see the study summary spreadsheet.

• 14

  Details of intervention in each study:

  • “We divided the study area into 133 clusters, which were randomly assigned to one of the two chlorhexidine cleansing regimens (single cleansing as soon as possible after birth; daily cleansing for 7 days after birth) or promotion of dry cord care.” (Arifeen et al. 2012, Design, p. 1022)
  • “In intervention clusters, the newborn cord was cleansed in the home on days 1−4, 6, 8, and 10.” (Mullany et al. 2006, Methods, p. 910)
  • “One intervention comprised birth kits containing 4% CHX solution for application to the cord at birth by TBAs and once daily by family members for up to 14 days along with soap and educational messages promoting handwashing. One intervention was CHX solution only and another was handwashing only. Standard dry cord care was promoted in the control group.” (Soofi et al. 2012, Methods, p. 1029)
  • “For babies allocated to the chlorhexidine group, mothers or caretakers were advised to apply the solution to the cord every day until 3 days after the cord had dropped off.” (Sazawal et al. 2016, Methods, p. e837)
  • “Newborn babies received clean dry cord care (control) or topical application of 10 mL of a 4% chlorhexidine solution once per day until 3 days after cord drop (intervention), according to cluster assignment.” (Semrau et al. 2016, Methods, p. e827)

• 15

  We calculated a random effects model using Comprehensive Meta-Analysis Pro, version 3. A summary of all included studies, meta-analysis inputs, and pooled results are available here. We found an average effect of 14.6% reduction in neonatal mortality. The result was marginally statistically significant (p = 0.084).

• 16

  I² = 75.74, defined as substantial heterogeneity per Deeks et al. 2021.

• 17

  We conducted a meta-regression using a random effects model (method of moments) with the log risk ratio as the dependent variable and the control group neonatal mortality rate as an independent variable. We found a significant (p=0.05) positive association between higher baseline mortality and intervention effect size, with 1 additional neonatal death per thousand at baseline carrying a 1.5% increase in the percentage reduction in neonatal mortality due to the intervention. Heterogeneity was reduced from I² = 75.74 to 59.88 percent. The full meta-regression output is available here (see Table 4).

• 18

  Random effects risk ratio for South Asian trials was 0.760 (p=0.007) (see this spreadsheet, cells F19:G19), and for African trials was 1.006 (p=0.960) (see this spreadsheet, cells H19:I19).

• 19

  "Daily chlorhexidine (7.1% chlorhexidine digluconate aqueous solution or gel, delivering 4% chlorhexidine) application to the umbilical cord stump during the first week of life is recommended for newborns who are born at home in settings with high neonatal mortality (30 or more neonatal deaths per 1000 live births)." WHO recommendations on postnatal care of the mother and newborn 2013, p. 3

• 20

  See this spreadsheet, Table 1.

• 21

  We conducted a meta-regression using a random effects model (method of moments) with the log risk ratio as the dependent variable and days of CHX application and control group neonatal mortality rate as an independent variable (days of application was not available for one trial). The number of days of CHX application was not statistically significantly associated with the observed effect. The full meta-regression output is available here (see Table 5).

• 22

  "The CHX study is a facility-based, individually randomized controlled trial that will be conducted among 4760 newborns in Uganda. The primary outcomes are severe illness and omphalitis during the neonatal period…Uganda has one of the highest neonatal death rates in the world. Nearly 29 out of 1000 live-born children die within the first month of life." Nankabirwa et al. 2017, pp. 1-8

• 23

  Details of control group intervention in each study:

  • "CHWs visited participating pregnant women twice during pregnancy and provided birth and neonatal care preparedness counselling, clean birthing kits, and referral for antenatal care." (Arifeen et al. 2012, p. 1024)
  • "All women received iron and folic acid supplementation (90 days), deworming (albendazole 400 mg), vitamin A supplementation every week (7000 retinol equivalents), and a locally manufactured clean delivery kit, which included soap, a sterile blade and tie, a small plastic disc, and a plastic sheet." (Mullany et al. 2006, p. 911)
  • "Families in group D (control cluster) received standard birth kits (without any CHX solution or soap). They were advised to practise dry cord care." (Soofi et al. 2012, p. 1030)
  • "Participants were randomly assigned to either 4% free chlorhexidine for cord care or to dry cord care using a computer-generated random sequence." (Sazawal et al. 2016, Methods, p. e837)
  • "The antenatal visit was completed within 2 weeks of enrolment, during which the field monitor confirmed the home location for follow-up, provided a standard clean delivery kit (which included soap, sterile razor blade, sterile gloves, two cord clamps, candle and matches, and a plastic mat) to all study participants irrespective of study group, reviewed study procedures with the mother, and screened for pregnancy danger signs." (Semrau et al. 2016, p. e829)

• 24

  Six trials measured the effect of the intervention on omphalitis. Based on the evidence presented in these trials, we are relatively confident that CHX cleansing can reduce the incidence of omphalitis. However, due to inconsistencies in the measure of omphalitis used and likely under-reporting of omphalitis in some cases, we do not believe that these data are of sufficient quality to provide an estimate of the effect of the intervention on the prevalence of omphalitis. Among trials reporting mild to severe omphalitis results, the prevalence of mild to severe omphalitis in the control group ranged from 6.1 (Semrau et al. 2016, p. e834, Table 6, "Dry cord care" group; "The observed incidence of omphalitis was lower than expected; it is possible that mild or moderate cases were under-reported as diagnosis was based on purulent discharge or redness at the umbilical stump." P. e834) to 155.7 (Arifeen et al. 2012, p. 1026, Table 3, "Dry cord care" group).

• 25
  • "Survivors of infant group B streptococcal (GBS) disease are at risk of neurodevelopmental impairment (NDI)… We identified 6127 studies, of which 18 met eligibility criteria, all from middle- or high-income contexts. All 18 studies followed up survivors of GBS meningitis; only 5 of these studies also followed up survivors of GBS sepsis and were too few to pool in a meta-analysis. Of meningitis survivors, 32% (95% CI, 25%–38%) had NDI at 18 months of follow-up, including 18% (95% CI, 13%–22%) with moderate to severe NDI." Kohli-Lynch et al. 2017, p. S190
  • Chlorhexidine kills GBS: "Chlorhexidine treatment of the cord was associated with an overall reduction in bacterial colonisation of the cord. This was most marked for coagulase positive staphylococci and was not associated with an increase in gram negative organisms." Smales 1988, abstract

• 26

  We suspect that there were differences across studies in the level of monitoring for infection, as well as the evidentiary bar for reporting suspected infection, but have not investigated this in detail.

• 27

  "Nine adverse events associated with chlorhexidine were reported; one case was of ocular exposure (grade 2 and did not require hospital admission) and eight episodes were of local skin irritation (grade 1). No cases of accidental ingestion, contact dermatitis, or anaphylaxis were reported." Semrau et al. 2016, p. e833

• 28

  From each of the included trials:

  • South Asia: Bangladesh – “We recorded no adverse events during this study.” Arifeen et al. 2012, p. 1026
  • South Asia: Nepal – No mention of adverse events in Mullany et al. 2006.
  • South Asia: Pakistan – “We recorded no serious adverse events." Soofi et al. 2012, p. 1029
  • Africa: Zanzibar – No mention of any adverse events in Sazawal et al. 2016.

• 29

  "Cleansing of the cord with 4.0% chlorhexidine has been implemented routinely in many clinical settings throughout the world for the past 4 decades and is considered safe…after topical applications of chlorhexidine, some percutaneous absorption occurs, particularly in preterm newborns, but only at trace levels. The data on safety, however, are incomplete. For example, the concentration of chlorhexidine used for multiple different interventions spans a wide range. Although the concentrations reported thus far appear to be safe, the upper level of chlorhexidine that can be considered safe is not known." Mullany, Darmstadt, and Tielsch 2006, p. 4

• 30

  "WHO/HQ has been made aware of over forty-five (45) reports of eye injuries, including cases of blindness, associated with the improper use of Chlorhexidine diclucanate 7.1% (CHX) in nine African countries —Nigeria (2015), Senegal (2015), DRC (2015), Liberia (2015), Niger (2017), Mali (2018), Kenya (2018), Chad (2018) and Cameroon (2019). The reported injuries are associated with both the liquid and the gel CHX formulations." Banerjee 2019, p. 1

• 31

  "Although decreased susceptibility to chlorhexidine has been reported, it has not been convincingly shown to be associated with repeated exposure to chlorhexidine. Proposed mechanisms include drug inactivation, efflux, and decreased uptake. Vancomycin-resistant enterococci (VRE) and vancomycin-susceptible enterococci, for example, have equivalent susceptibilities to chlorhexidine. Some Pseudomonas species and other nonfermenting Gram-negative organisms have high-level resistance to chlorhexidine. Plasmid-mediated resistance mechanisms to antiseptics are documented for staphylococci, as are increased MICs [minimum inhibitory concentrations]; however, the clinical significance of these increased MICs has not been determined, because they are usually far less than the commonly used chlorhexidine concentrations. Decreased susceptibility and the potential for emergence of resistance exist. Several laboratory studies have raised concerns that emergence of biocide nonsusceptibility may result in cross-resistance to antibiotics, but data are limited. As antiseptics become used more broadly, surveillance of clinical isolates should be considered to identify an epidemiologically significant trend toward decreasing antiseptic and antibiotic susceptibility." Weinstein et al. 2008, p. 274-276

• 32

  See World Bank, Mortality rate, neonatal, "Low income countries" row (as of 2019 data).

• 33

  As described above, our meta-analysis found an average effect of 14.6% reduction in neonatal mortality. The result was marginally statistically significant (p = 0.084) (see this spreadsheet, cells D19:E19).

• 34

  "In Nepal, the bulk procurement cost of single-use chlorhexidine tubes is about US$0.23/newborn" Hodgins et al. 2013, p. 6

• 35

  See World Bank, Mortality rate, neonatal.

• 36

  “Preterm birth (40.8%) and intrapartum complications (27.0%) accounted for most early neonatal deaths while infections caused nearly half of late neonatal deaths. Preterm birth complications were the leading cause of death in all regions of the world.” Oza et al. 2015, p. 19

• 37

  "PATH would likely allocate any additional funding it received for work on chlorhexidine to its activities in Francophone West Africa. For example, additional funding would significantly bolster its efforts to produce multi-language materials promoting community-level awareness of chlorhexidine." GiveWell's non-verbatim summary of a conversation with Dr. Patricia Coffey, December 13, 2019, p. 4