Umbilical cord-care practices in low- and middle-income countries: a systematic review

Background Neonatal sepsis is the third leading cause of deaths for infants in their first month of life. The newly cut umbilical cord can be a pathway for bacteria that can cause newborn sepsis and death. Optimal umbilical cord care practices for newborns and during the first week of life, especially in settings with poor hygiene, has the potential to avoid these preventable neonatal deaths. The purpose of this review of cord care practices is to assist in the development of behavior-change strategies to support introduction of novel cord-care regimens, particularly 7.1% chlorhexidine digluconate for umbilical cord care. Methods We searched domestic and international databases for articles that were published in English between January 1, 2000, and August 24, 2016. We found 321 articles and reviewed 65 full-text articles using standardized inclusion criteria. The primary criteria for inclusion was a description of substances applied to the umbilical cord stump in the days following birth. Results We included 46 articles in this review of umbilical cord-care practices. Articles included data from 15 low- and middle-income countries in sub-Saharan Africa (8 countries), Asia (5 countries), North Africa (1 country), and Latin America and the Caribbean (1 country). Findings from this review suggest that documentation of cord-care practices is not consistent throughout low- and middle-income countries, yet existing literature depicts a firm tradition of umbilical cord care in every culture. Cord-care practices vary by country and by regions or cultural groups within a country and employ a wide range of substances. The desire to promote healing and hasten cord separation are the underlying beliefs related to application of substances to the umbilical cord. The frequency of application of the substance (either the number of days or the number of times per day the substance was applied), and source and cost of products used is not well-characterized. Conclusions This desire to actively care for the umbilical cord of a newborn—as noted in the variety of cord care practices and beliefs identified in this review—points toward the need to contextualize any behavior change approach to align with the local culture. Electronic supplementary material The online version of this article (doi:10.1186/s12884-017-1250-7) contains supplementary material, which is available to authorized users.


Background
Neonatal sepsis is responsible for more than 15% of neonatal deaths worldwide [1] and is the third leading cause of deaths for infants in their first month of life. A newly cut umbilical cord can be a pathway for bacteria that can cause newborn sepsis and death.. Optimal umbilical cord care practices for newborns and during the first week of life, especially in settings with poor hygiene, has the potential to avoid these preventable neonatal deaths.
Harmful traditional cord-care practices are often cited as an important public health concern [2,3]. A clear understanding of behavioral intention underlying traditional cord care practices in low-and middle-income countries can be helpful in addressing high rates of neonatal sepsis. Although systematic evidence reviews of cord-cleansing practices have been conducted previously [4,5], the qualitative nature of cord-care practices has not been summarized to-date. This review fills a gap in the literature by systematically reviewing available evidence related to traditional cord-care practices and assessing the likely impact of product categories on infection risk.

Methods
Our initial search focused on studies that described traditional umbilical cord care practices globally. For the purposes of this article, traditional practices were those that focused on the cultural beliefs and customs that guided how the umbilical cord was cared for, including the length of the cord stump, substances applied, and the decisions regarding disposal of the cord stump. We developed systematic searches for PubMed and Google Scholar using controlled vocabulary (Additional file 1: Search Terminology). Initial criteria for eligibility were determined by topic, time period, and language of the publication. We included articles that were published between January 1, 2000, and January 30, 2016. A second search was performed in August 2016 to account for any publications during the intervening months. The language of publication was limited to English. References in the identified articles were reviewed to determine if other sources would be pertinent and additional articles were abstracted if relevant.
The original search yielded 321 articles, from which 107 duplicates were excluded. A reviewer then screened titles and abstracts of the remaining 214 articles to determine suitability for inclusion. Articles that did not meet the criteria included those unrelated to the application of substances to the umbilical cord, articles focused on clinical trials comparing a variety of antiseptic applications to the umbilical cord, articles wherein the authors related only secondary data sources regarding umbilical cord-care practices, and articles that were unrelated to umbilical cord care, but had appeared in the search due to a common term, such as spinal cord. A total of 65 full-text articles were then reviewed using standardized inclusion criteria. The primary criterion for inclusion was a description of substances applied to the umbilical cord stump in the days following birth. Based on these criteria, a total of 46 of the 65 articles were included in this review. Secondary data about beliefs in relation to umbilical cord care and other cord-care practices were also recorded, if available. Data regarding cord-care practices were extracted from the articles using a standardized tracking form in Excel. Data items included: ○ What was used to cut the umbilical cord? ○ What was used to tie the umbilical cord? ○ Applications of a substance to the umbilical cord stump.
○ What substance was applied? ○ How often it was applied? ○ How many days it was applied? ○ Why was it applied (belief )? ○ Who applied the substance? ○ Source of product supply. ○ Cost of product applied. ○ Other newborn skin-care practices, such as infant massage, that could contribute to the development of neonatal sepsis or tetanus were also tracked.
We synthesized data relating to cord-care practices and substance used on the cord by country. Because most of the studies were qualitative or observational in design, we were unable to draw any statistical comparisons. This review reporting followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (known as PRISMA) reporting guidelines [6], as warranted.

Results
A total of 46 articles were included in this review of umbilical cord-care practices. Figure 1 presents the flow diagram of the review process.
The 46 articles included data from 15 distinct lowand middle-income countries in sub-Saharan Africa (8 countries), Asia (5 countries), North Africa (1 country), and Latin America and the Caribbean (1 country). In sub-Saharan Africa, the majority of articles came from Uganda (6), followed by Tanzania (4), Ethiopia and Nigeria (3 each), Ghana and Zambia (2 each), and Benin and Sierra Leone (1 each). In Asia, the majority of articles came from Pakistan (7), followed by India and Nepal (5 each), Bangladesh (3), and Turkey (2). In North Africa, one article came from Egypt. In the Latin America/Caribbean region, one article came from Haiti. Table 1 provides general details about the articles included in this review. While country income classification was not a predetermined criteria for inclusion or exclusion, umbilical cord care articles from high-income countries were solely focused on the comparisons and uses of antiseptics and; therefore, none are included in this review based on the exclusion criteria.

Beliefs
Beliefs related to the application of substances to the umbilical cord varies by country and by regions or cultural groups within a country. The intention behind applying a substance to the umbilical cord is to promote healing [7][8][9][10][11] and hasten the separation of the cord [7,[12][13][14] either by keeping the cord stump moist [10,14,15] or by drying it out [8,[16][17][18][19][20] to prevent pain/infection/bleeding [10,12,16,17,21], or to keep the "wind" (evil spirits) or cold/air [10,16,22] out of the infant. Table 2 provides an overview of cord-care practices described in each article included in this review. Table 3 illustrates the types of substances applied to the cord by country. The substance applied may depend upon the perceived nature of the cord. A moisturizing substance is applied if the cord is too brittle and a drying substance is applied if the cord takes too long to separate. For example, in southern Zambia, petroleum jelly or mabono (wild fruit) oil might be used if the cord is cracking or bleeding, while charcoal dust, baby powder, or burnt pumpkin stem may be used if the cord takes too long to separate [8]. In areas of Tanzania, Uganda, and Zambia [7,8,18,20,23], the infant cannot leave the home until the cord separates and/or the mother cannot return to her chores until that time. In the Tonkolili district of Sierra Leone, a traditional birth attendant noted the purpose of applying pounded cassava to the cord: "It will help the umbilical recover easily and the child will walk fast" [11].
The substances applied may also vary by the infant's perceived gestational age or state of health. For example, in the Choma District in Zambia, the substance applied to a newborn's umbilical cord differs by the newborn's perceived gestational age, as substances used for fullterm babies are considered too strong for preterm babies. A black powder made from the burnt stem of the pumpkin plant is applied to the umbilical cord of fullterm infants while a green powder made from the dried roots of the mweeye plant is applied to the cord of preterm/small infants as it is considered to be gentler than the black powder [18]. Also in southern Zambia, substances such as petroleum jelly, mabono (wild fruit) oil, cooking/motor oil, charcoal, dried cow dung or chicken droppings, burnt pumpkin stem, and crushed loma (wasps' nest) are applied to the cord of the healthy infant. However, a separate set of substances that are considered to be medicinal are applied if the cord is red or if pus appears. The substances that are considered to be medicinal include: python oil, breast milk, alcohol, banana, cow dung, mukunku (bark of a tree), traditional herbs, or dirt from a pounding stick [8].
The length of the cord may have specific importance. In southern Zambia, it is believed that if the cord is too long, it will take too long to heal, and if it is too short, "the air will go in and this will make the baby die" [8].        Substances applied to the stump vary by whether child is classified (at birth) as "sick" or "healthy". If there are blood clots in the umbilical cord, a sign of abnormality and future illness, the child is taken to a healer who treats the stump with herbs or the child is taken to a hospital. Until the cord separates, mother is expected to apply a wrap around the infant's waist so that the cord does not touch the genitals (will make them infertile as an adult) and the cord is not supposed to fall on to the floor. Once the cord separates, the baby is bathed in cold water to make them strong.
Breast milk dripped onto cord. Full-term healthy babies: black powder (made from burn stem of pumpkin). Pre-term babies: green powder (ground dried roots of mweeye plant). This is considered gentler for preterm babies.
If the appropriate herbs cannot be found then brick ash is used. Less common: after the cord falls off, fresh dried chicken dung is mashed and put onto the wound (rooster for male babies, hen for female babies). Participants explained that this tradition originated in Zimbabwe.
Shamba [19] J Health Popul Nutr. 2013 Tanzania A study to explore childbirth-related hygiene and newborn care practices in community deliveries in southern Tanzania. New blades were used in almost all instance, but new cord ties were rarely used. Substance was applied to cord by many mothers.
Sharkey [11] Health Policy Plan. 2016 Sierra Leone A study to understand maternal and neonatal care practices in four rural districts of Sierra Leone.
Sharma [28] BMC Pregnancy Childbirth. 2016 Nepal A study to understand beliefs around childbirth and postnatal card in rural Nepal, particularly focusing on beliefs related to women being polluted following childbirth. A hasiya (scythe), a sickle, or a razor blade may be used to cut the umbilical cord and then a substance may be applied.
Antiseptic, cooking oil, ghee, plain water, toothpaste, and ash. The length of the cord is also believed to indicate the length of the genitals in both men and women [8]. Further, in southern Zambia and on the island of Pemba in the Zanzibar archipelago, Tanzania, the umbilical cord may be wrapped or bound with a strip of cloth to prevent it from touching the groin area [8,18,24], which is sometimes believed to cause infertility as an adult [18]. In areas of Zambia and Tanzania, where it is important to protect the cord from being taken by someone who wishes the infant or the family ill, the cord, and often the placenta, are disposed of through burial in a sacred place or by burning, or are sometimes put in a pit latrine/toilet to prevent them from being unearthed [8,23]. In southeastern Turkey, the cord may be buried in a special place to help define the child's path in life, such as burying the cord at a mosque or a school to help the child grow up to be a religious or educated person, respectively [25].

Frequency of application
Few articles reported on the frequency of application of the substance (either the number of days or the number of times per day the substance was applied). In Ethiopia, the substance is applied one to three times per day up to the seventh day of life. However, the applications may not begin until the newborn is two or three days old [14]. In the Brong Ahafo region of Ghana, a substance is applied anywhere from every 30 min to 3 times per day [15]. In urban Uganda, mothers reported cleaning the cord with a substance at least twice per day [9]. On Pemba Island in the Zanzibar archipelago, Tanzania, multiple substances were applied to the umbilical area beginning on the sixth day after birth. In this study of more than 1000 infants born at home, only 10% (n = 109) had a substance applied to the umbilical cord and less than 11% of those applications were made in the first 48 h of life [26]. A study in Sylhet District in Bangladesh, involving 39 in-depth interviews of mothers, fathers, grandmothers, and traditional birth attendants and data from more than 6000 household surveys of mothers, found that substances were applied until the cord separated and that either turmeric or ginger are applied at birth and then a combination of mustard oil and garlic are applied twice daily until the separation of the cord [27]. Few studies identified who usually applied the substance to the cord. In the 19 studies where it was reported, either the mother or grandmother of the infant [9, 12-19, 21, 25, 27-32] or a senior woman [15,19,33] in the household applied the substance to the cord. In only a very few cases did a traditional birth attendant or health worker [11,14,19] apply a substance to the cord.

Cost and source of substance
Few articles investigated the source of the substances applied to the umbilical cord and none reported on the cost of the substances applied. Shea butter or cooking oil were purchased from the market [15,18]. One study reported that the cooking oil applied to the cord is generally a recycled product bought at the local market that had been previously used [18].

Discussion
This study presents a view of traditional cord-care practices as reported during the last fifteen years in low-and middle-income countries. The desire to care for the umbilical cord of an infant appears to be universal in all cultures. Of interest is the description of the range of products applied to the newly cut cord and that substances are applied to the umbilical cord to infants born at home as well as in facilities. Participants in studies from Petit-Gôave, Haiti, and Karachi, Pakistan, reported that if an infant was born in a health facility, a substance would be applied to the umbilical cord upon returning home [22,34]. The possible harm of these substances has not been fully quantified. In some cases, such as with kohl and surma used in Egypt and Pakistan, the lead and antimony included in the product is most likely harmful. Anecdotal literature often refers to the use of dung as a harmful traditional practice. In contrast, we found use of chicken/lizard/cow dung reported in only three countries (Haiti, Uganda, and Zambia), thereby suggesting that this practice is not as widespread as depicted or that it, as reported by a traditional birth attendant in northern Ghana, was practiced in the past and has since ceased [21]. It is also possible that application of preparations using dung is underreported in published literature. For example, unpublished formative research from the Kenieba and Koutiala Districts of Mali report use of bassa bo (lizard excrement powder mixed with shea butter) and bagani dji (insect powder mixed with shea Table 3 Substances used, by category (Continued) Antiseptics (unspecified) [26,28,31,57,58,62,67,68] Egypt, India, Nepal, Tanzania Baby powder/talcum powder [8,17,19,20,24,26,32,55,65]  Other product categories (i.e., oils, herbs/spices/plants, mineral/powder, water, bodily fluids, food, personal care/ medical products) and processes such as heat treatment may or may not be harmful and could warrant further investigation. Neonates and young infants are more prone to infection than older children and adults and evidence suggests that their immune systems are still developing rather than being fully formed at the time of birth. Further, newborns appear to be particularly vulnerable to the types of intracellular pathogens that commonly cause neonatal sepsis [35], such as some strains of streptococcus, Escherichia coli, and Listeria monocytogenes.
Concern has been expressed about transmission of HIV from mother to child through the application of breast milk to the umbilical cord [18]. Also, the use of numbati (water that has been used to wash an adult woman's genitals) in Tanzania [23], could potentially pose a risk for transmission of HIV or other diseases. The application of unhygienic substances to or around the umbilical cord stump has been linked to tetanus in infants [2,13,25,[36][37][38][39]. Clostridium tetani bacteria, found in soil, dust, saliva, animal dung, and other sources, are the cause of tetanus infection [40].
Some substances, which are known to be harmful in other contexts, pose an unclear risk when applied to the umbilical cord. For example, motor/machine oil contains high levels of chemical additives and, once used for its intended purpose, can contain high levels of heavy metals and other minerals [41,42]. While used motor/ machine oil is well known to be harmful to the environment and causes dermatitis through long-term exposure, the health risks due to short-term exposure through an open wound is not well defined. Boric acid powder is used as a pesticide and has caused seizures and death in infants when ingested [43], but also has known antifungal properties and has been used as a treatment for conditions such yeast infections [44]. However, sources differ on whether it is safe to use boric acid on open wounds. Pyriproxyfen powder, a pesticide, poses minimal risk to humans in small quantities [45]. Other substances, such as oils, herbs, plants, food products, and heat treatments applied to the cord may be harmful depending on whether they have been contaminated in some way, such as through unhygienic preparation or storage.
This critical information regarding traditional cordcare practices can serve as the stepping stone to behavior change. We focused our review on a comprehensive description of reported cord-care practices with the intention of employing the knowledge to develop behavior-change strategies to support introduction of novel cord-care regimens, particularly 7.1% chlorhexidine digluconate for umbilical cord care. Randomized controlled trials investigating the use of 7.1% chlorhexidine digluconate for umbilical cord care have been conducted in Nepal [46], Bangladesh [47], and Pakistan [48]. A meta-analysis of the three studies demonstrated that application of chlorhexidine to the umbilical cord of the newborn led to a 23% reduction in all-cause neonatal mortality and a reduction in omphalitis ranging from 27 to 56% compared to control group depending on severity of infection [49]. Further, the World Health Organization recommends the application of 7.1% chlorhexidine digluconate (gel or solution) to the umbilical cord of neonates who are born at home in settings with high neonatal mortality or to replace the use of a harmful, traditional substances [50]. Despite previously reported substantial reductions in South Asia, results from recent trials in Zambia [51] and Tanzania [52] show that application of 7.1% chlorhexidine to the umbilical cord did not significantly reduce neonatal mortality rates in the study sites. This suggests that programmatic context and level of risk in the population as well as cord care practices must be considered in any behavior change initiative.
Findings from this review suggest that documentation of cord-care practices is not consistent throughout lowand middle-income countries. Given the heterogeneity of practices described in the literature, it is not clear if data from one country in a region also pertains to other surrounding countries and/or nearby ethnic groups. Overall, however, existing literature depicts a firm tradition of umbilical cord care in every culture. The desire to take some kind of action to address the newly cut umbilical cord seems to be a strong human desire. Participants in several studies noted that adhering to dry cord care was very difficult. For example, in Ghana, there was a belief that applying nothing to the cord would delay separation, cause discomfort, and potentially cause death for the infant by preventing the sore from healing and causing a sickness in the stomach [15]. In Uganda, the practice of dry cord care was noted as difficult to follow as it delays the cord separation and the return of the mother to chores [7]. This desire to actively care for a newborn could be utilized as a trigger for behavior change. For example, as applied in the Health Belief Model [53], knowledge of effective cord care (i.e., product, application procedure, number of days to use, number of times/day to apply) could function as a "cue to action" to allow caretakers to act in a positive manner.
This review has limitations, as noted. Reporting is not global in nature as the review was limited to sources published in English in peer-reviewed journals. Few of the studies conducted have been indepth, qualitative assessments of umbilical cord care practices; therefore, much of the detailed information about beliefs and practices surrounding umbilical cord care comes from a few sources or countries, such as Tanzania and Zambia. This could lead to the assumption that some countries place greater importance on umbilical cord care practices than other countries where the data collection focused on a variety of newborn care practices. The paucity of data from Latin America and the Caribbean reflected in this review could stem from our decision to include only English-language material. Also, it is likely that additional anecdotal information is available in gray literature.
Additional research around cost and source of products used in cord care practices could assist programs that are targeting newborn care behavior change. Likewise, deeper ethnographic and/or qualitative investigation into the underlying meaning and significance of traditional cord-care practices could assist in formulating key messages being used to generate demand for novel prophylactic products.

Conclusions
Findings from this review suggest that documentation of cord-care practices is not consistent throughout low-and middle-income countries, yet existing literature depicts a firm tradition of umbilical cord care in every culture studied. Cord-care practices vary by country and by regions or cultural groups within a country and employ a wide range of substances. The desire to promote healing and hasten cord separation are the underlying beliefs related to application of substances to the umbilical cord. The frequency of application of the substance (either the number of days or the number of times per day the substance was applied) and source and cost of products used are not well characterized. This desire to actively care for the umbilical cord of a newborn could be utilized to promote positive behavior change such as the introduction of 7.1% chlorhexidine digluconate for umbilical cord care. The variety in cord care practices and beliefs noted in this review however points toward the need to contextualize any behavior change approach to align with the local culture.

Additional file
Additional file 1: Search Terminology. Exact narrative sequences used for literature search. (DOCX 12 kb)