The reliability of a newborn foot length measurement tool used by community volunteers to identify low birth weight or premature babies born at home in southern Tanzania

In 2012, the identification of low birth weight infants (<2500 g) within 48 hours of birth among those born at home was ranked by scientists and public health practitioners as the number one research priority to reduce global mortality from preterm birth and low birth weight [1]. Low birth weight infants need additional care: each year in sub-Saharan Africa an estimated 1.2 million newborns die during the first month of life [2]. Being born too small – because of low birth weight, prematurity, or intra-uterine growth retardation – is a very strong predictor of newborn mortality [3, 4]. Life saving interventions [5] implemented in the community and in health facilities could save as many as 85% of African maternal, newborn and child deaths if optimal coverage was achieved, [6] but despite some improvements, intervention coverage still remains low for the majority of families across Africa [7].

One immediate obstacle to realising the full potential to save lives is that, despite a global increase in skilled attendance at birth in the last decade, around half of all births in Africa continue to occur at home, [8, 9] and most families have no contact with formal health services during the first week of a newborn’s life. In such settings, community health workers or volunteers have considerable potential to improve newborn survival by promoting skin-to-skin contact, thermal care, immediate and frequent breastfeeding, and active care seeking [10]. A meta-analysis of studies to evaluate the impact of home visits to newborns in programmatic settings estimated a 12% reduction in newborn mortality, equivalent to a five percent reduction in overall child mortality [11].

One persistent limitation of home visits is that community members, including volunteers, usually lack the tools and measurement skills with which to identify newborns as small and at-risk. We previously reported a hospital based study showing that short foot length had the potential to be used as a low-cost tool to identify low birth weight or preterm newborns born at home in the event that weighing scales and clinical assessment were not available [12].

Here we report on a community based study to estimate the reliability of a simple foot length measurement tool used by community volunteers to identify low birth weight or premature babies born at home in the first days of life.

In this community based study we have shown that a newborn foot length measurement tool, printed on a counselling card, can be used at the community level with moderate reliability by community volunteers who have received 5 days of training in total, which included approximately 5 hours on foot size measurement. This training was supplemented by quarterly review meetings, lasting a few hours each time, at which volunteers were encouraged to discuss their experiences from the field.

Qualitative and quantitative evidence from the study area [15, 18] informed the healthy newborn behaviours to be promoted by INSIST volunteers during the main INSIST trial. The key behaviours were hygiene during childbirth, immediate and exclusive breastfeeding, and extra care, including skin to skin care, for low birth weight or premature babies. In order to identify low birth weight or premature newborns, volunteers required a simple and easily sustainable identification tool. Key stakeholders at national level advised that weighing scales would not be sustainable, and many primary health care facilities do not have scales, thus the foot length tool was developed [19]. Our findings show that community volunteers tended to under estimate newborn foot length. The implication of this may have been that a large number of newborns were said to be in need of extra care, something that could have led volunteers and parents to lose confidence in the identification tool. Nonetheless, by putting focus on examining the newborn, the tool may have provided a starting point for discussion with parents about whether or not they thought that their baby needed extra care – beyond that recommended for every newborn - in the first days of life. The issue of early care seeking from formal health providers urgently needs to be addressed. A meta-analysis of care seeking for neonatal illness in low and middle income countries [20] highlighted that there is a paucity of high quality data on the subject, especially from Africa, and estimated that just 20% of caregivers of sick newborns reported seeking care from a health care provider outside the home. In southern Tanzania, previous studies have suggested that care seeking is particularly low for babies born at home [14] – precisely the group of newborns that were targeted by the foot length tool.

The goal of health care providers is to have contact with every newborn at birth (through facility delivery) or immediately after (through post-natal care) and take a high quality measure of birth weight to inform decision making about any care needed. But sub-optimal coverage of institutional deliveries and post-natal care means that around half the newborns of sub-Saharan Africa do not have this opportunity and alternative methods to identify small babies should be encouraged if they can be linked to identification of newborn danger signs. In an analysis of Demographic and Health Survey data from three countries, [21] mother’s perception of a baby’s size at birth was found to be closely linked to birth weight at national levels, although within communities there was considerable variation in size perceptions. The authors concluded that care was needed when interpreting individual maternal perceptions because of the influence of other variables such as gender of the child, place of birth, and the size of other babies in the local community.

Measuring newborn foot length has been found to be acceptable, [12] but of the anthropometric proxies for low birth weight it is not the most accurate. Our earlier hospital based study estimated the sensitivity and specificity of foot length <8 cm to predict low birth weight (<2500 g) to be 0.87 and 0.60 respectively [12]. A meta-analysis of evidence from hospital studies concluded that chest circumference had the highest accuracy when predicting low birth weight, compared to other anthropometric measures, [22] but its main limitation for application in a community setting was that it can be difficult to measure accurately without medical training, and can be disruptive to the newborn since it requires undressing [23, 24]. Further to this, integrating the foot length tool as a printed image on a counselling card may not have been the most accurate method of determining foot length, and some discordant categorisation was observed between the counselling card and the ruler. But having just one simple, cheap, locally produced tool for volunteers to carry and explain, and being able to leave a copy of this tool with families, may have outweighed this limitation by removing important barriers to its use.

A simple newborn foot length measurement tool can be applied by community volunteers with moderate reliability. However, using this tool volunteers over estimated the number of small newborns needing extra care by thirteen percent compared to estimates by the project researcher. Newborn foot length is simple to understand, easy to construct and can be implemented at home in the first days of life when the risk of newborn death is highest, but monitoring its use and providing supportive supervision to users of the tool is important.