Measuring newborn foot length to identify small babies in need of extra care: a cross sectional hospital based study with community follow-up in Tanzania
BMC Public Health volume 10, Article number: 624 (2010)
Neonatal mortality because of low birth weight or prematurity remains high in many developing country settings. This research aimed to estimate the sensitivity and specificity, and the positive and negative predictive values of newborn foot length to identify babies who are low birth weight or premature and in need of extra care in a rural African setting.
A cross-sectional study of newborn babies in hospital, with community follow-up on the fifth day of life, was carried out between 13 July and 16 October 2009 in southern Tanzania. Foot length, birth weight and gestational age were estimated on the first day and foot length remeasured on the fifth day of life.
In hospital 529 babies were recruited and measured within 24 hours of birth, 183 of whom were also followed-up at home on the fifth day. Day one foot length <7 cm at birth was 75% sensitive (95%CI 36-100) and 99% specific (95%CI 97-99) to identify very small babies (birth weight <1500 grams); foot length <8 cm had sensitivity and specificity of 87% (95%CI 79-94) and 60% (95%CI 55-64) to identify those with low birth weight (<2500 grams), and 93% (95%CI 82-99) and 58% (95%CI 53-62) to identify those born premature (<37 weeks). Mean foot length on the first day was 7.8 cm (standard deviation 0.47); the mean difference between first and fifth day foot lengths was 0.1 cm (standard deviation 0.3): foot length measured on or before the fifth day of life identified more than three-quarters of babies who were born low birth weight.
Measurement of newborn foot length for home births in resource poor settings has the potential to be used by birth attendants, community volunteers or parents as a screening tool to identify low birth weight or premature newborns in order that they can receive targeted interventions for improved survival
Despite important gains in the survival of children under five years in sub-Saharan Africa during the last twenty years, the neonatal mortality rate (NMR) remains virtually unchanged at 41 deaths within 28 days of birth for every 1000 live births. This currently represents an estimated 41% of all under five deaths .
The majority of these deaths occur in settings where the health system is weak and robust data is scarce. However, a recent focus on the topic has lead to some consensus about causes and actions to be taken. The three main causes of neonatal death are intra-partum related, complications due to prematurity or low birth weight, and infection [1, 2]. As many as three-quarters of deaths occur in the first week of life, and up to 90% of all babies who die are born low birth weight (LBW <2500 grams) either because of prematurity or intrauterine growth restriction . Community level interventions including skin-to-skin contact (also known as Kangaroo Mother Care), immediate and frequent breastfeeding, and active care seeking could reduce neonatal mortality rates by as much as 40% in high mortality settings [3–6].
There are a number of reasons why our potential to reduce the burden of neonatal death is not currently realised on a large scale [7–9]. One important road block is identification of babies at risk. Over half of all babies born in sub-Saharan Africa for example are born at home  and the majority of communities and families have no access to scales or other means by which to identify a baby as small, at risk, and in need of extra care.
Simple anthropometric alternatives to measuring birth weight have been investigated in various settings. Six separate research studies from UK, India, Nepal and Taiwan have reported on studies to investigate newborn foot length as a screening tool for small babies, showing consistent foot length cut-offs for identifying small babies across different countries (Figure 1) [11–16]. Moreover, despite having different aims and objectives, five out of six of the studies concluded that for high risk babies born at home, measuring foot length in the community may have advantages over other methods. Unlike measuring chest circumference, for example, there is a relative lack of training required, and lack of disturbance to the infant that undressing might introduce.
Here we report on the first African study of newborn foot length to identify low birth weight and premature babies and give the sensitivity and specificity estimates, and positive and negative predictive values for different foot length cut-offs. The aim was to determine the utility of using foot length as a screening tool in home delivery settings to identify low birth weight or premature babies in need of extra care. Our goal was to define a simple operational foot length cut-off that could be used to screen babies as follows: (1) babies who have an urgent need for care because they are very low birth weight (<1500 grammes); (2) babies who would benefit from additional care in the home because they have birth weight 1500-2499 grams or gestational age <37 weeks; and (3) babies who have birth weight >2499 grams and gestation age >36 weeks, and thus do not need special care because they are small.
We conducted a cross-sectional study of newborn babies born in hospital, with a non random subset followed up at home on the fifth day of life. A clinical officer examined babies and measured their feet on the first day, an experienced field supervisor tracked and measured the feet of babies on the fifth day; a medical doctor supervised the work. All staff were trained together on the maternity ward for one week prior to the start of data collection, each repeating foot length, birth weight and gestational age measures and comparing findings.
Study setting and participants
The research was carried out between 13 July and 16 October 2009 within the framework of an ongoing randomised controlled trial (clinical trials identifier NCT01022788) called "Improving newborn survival in southern Tanzania" (INSIST). It is based in Lindi and Mtwara regions, a predominantly rural and poor area with high neonatal (43 per 1000 live births) and infant (76 per 1000 live births) mortality . Seventy one percent of the population live within 5 kilometres of a health facility but around two-thirds of births take place at home.
During the study period, all babies born at the Mtwara Regional hospital, known as "Ligula hospital", were assessed for inclusion within 24 hours of birth and mothers asked to give informed written consent. Babies showing signs of any one of the following were excluded: severe respiratory distress, birth asphyxia, congenital deformity, birth weight <1000 grams or gestational age <28 weeks.
Anthropometric data collection
On the first day of life, each baby recruited in hospital for whom consent was given had their right foot measured from the heel to the tip of hallux (big) toe using a stiff transparent plastic metric ruler. Birth weight was measured using digital Salter scales which were calibrated at each use with a bottle weighing 1000 grams. Gestational age was estimated by the Eregie method [18, 19] which adapts the Dubowitz score  for African babies; in short, 6 physical characteristics of babies were examined and scored for signs of maturity (skin, eyes/ears, breast size, genitalia, mid-upper arm circumference, and head circumference).
Mothers were asked to give their home address and those estimated to live within 20 kilometres of the hospital were asked for permission for project staff to make a home visit on the fifth day of baby's life to re-measure foot length. This was then compared to the first day newborn foot length, birth weight, and gestational age estimate to explore the association up to five days after birth. Finally, project staff were asked to comment on the ease of implementation of the foot length measurement, and to report any perceptions of acceptability by parents.
Data processing and statistical analysis
Data was entered directly into personal digital assistants at point of measurement on the first and fifth day of life, with internal consistency checks pre-programmed. Data was analysed using STATA version 10.0. The distribution of gender, birth weight, foot length and gestational age estimates for the babies measured on the first day and the fifth day subset were compared using Chi2 test for heterogeneity to look for evidence of difference between the two samples that could introduce bias. We defined three binary variables to describe outcomes: (1)less than 1500 grams (very low birth weight), (2)less than 2500 grams (low birth weight) and (3)less than 37 weeks gestation (premature). Babies with these outcomes are referred to as 'small' in this context.
The absolute difference in foot length measured on the first and the fifth day for individual babies was calculated. Non-parametric receiver operating characteristics (ROC) analysis was conducted on first and fifth day foot lengths for the three outcome variables. To generate a simple operational foot length cut-off for babies with low birth weight or prematurity, a range of foot lengths was then defined as those that approximated 80% sensitivity and 80% specificity across the three outcomes. Sensitivity, specificity, positive and negative predictive values were calculated for these foot lengths.
The study was approved by the Institutional Review Board of Ifakara Health Institute, Tanzania, the Medical Research Coordinating Committee, Tanzania and the London School of Hygiene and Tropical Medicine, UK. Written, informed consent was obtained from the caregiver of each infant in the study.
During the study period, 635 babies were born at Ligula Hospital of whom 529 (83%) were recruited to the foot length project and had their feet measured on the first day of life. According to maternal reports, 257 lived within 20 km of the hospital and of these 183 (71%) had their feet measured again on the fifth day of life (Figure 2). The mean birth weight was 2.9 grams (SD 0.4), and mean gestational age was 39.5 weeks (SD 2.4). The two samples were similar with respect to gender, birth weight, newborn foot length and prematurity (χ2 p > 0.05, table 1). Of the 529 babies, eight (2%) were born very low birth weight (<1500 grams); 78 (15%) were born low birth weight (<2500 grams); 44 (8%) were born premature <37 weeks. Amongst the premature babies, seven (16%) had birth weight <1500 grams and 37 (84%) had birth weight between 1500-2500 grams.
The mean foot length of the 529 babies measured on the first day of life was 7.8 cm (standard deviation 0.4), and the mean foot length of the 182 babies measured on the fifth day was 8.1 cm (SD 0.3). On average, each baby's foot was 0.2 cm (SD 0.3) longer on the fifth day than on the first day.
Sensitivity, specificity, positive and negative predictive values of foot length
Sensitivity and specificity estimates for different foot lengths measured on the first day to identify small babies are shown in figures 3, 4 and 5 in the form of non-parametric receiver operating characteristics (ROC) curves. Assuming an 80% cut-off for both sensitivity and specificity to be desirable, we observe that this is achieved for very low birth weight (<1500 g) at foot lengths <7.2 cm and <7.5 cm respectively (figure 3), for low birth weight (<2500 g) at foot lengths <7.9 cm and <7.6 cm (figure 4), and for prematurity (<37 weeks) at foot lengths <7.7 cm and <7.5 cm (figure 5).
Using this range of foot length estimates as a guide, we show in table 2 the sensitivity and specificity for two simple and potentially operational foot length cut-offs: <7.0 cm to identify very low birth weight babies <1500 g who need urgent care and <8.0 cm to identify other small babies (low birth weight <2500 g or gestational age <37 weeks).
Fourteen newborns (3%) had a foot length <7 cm, and 275 (52%) a foot length of <8 cm on the day of birth. Of the eight babies with birth weight <1500 g, six had foot length <7 cm and all had feet shorter than 8 cm. Of the 77 babies with birth weight <2500 g, 14 had foot length <7 cm and 67 had feet shorter than 8 cm. Of the 44 premature babies, 12 had foot length <7 cm and 41 had feet shorter than 8 cm.
Newborn foot length of <7 cm had sensitivity and specificity of 75% (95% confidence interval 36-100) and 99% (95% confidence interval 97-99) respectively to identify birth weight <1500 grams, foot length of 8 cm had sensitivity and specificity of 87% (95% confidence interval 79-94) and 60% (95% confidence interval 55-64) respectively to identify birth weight <2500 grams, and foot length of 8 cm had sensitivity and specificity of 93% (95% confidence interval 82-99) and 58% (95% confidence interval 53-62) to identify premature babies.
On day five, foot length of 8 cm had sensitivity and specificity of 77% (95% confidence interval 74-86) and 74% (95% confidence interval 51-88) respectively to identify low birth weight babies, and foot length of 8 cm had sensitivity and specificity of 79% (95% confidence interval 73-85) and 72% (95% confidence interval 44-98) to identify premature babies (table 2).
The positive predictive values (PPV) for these foot length cut-offs were low, being highest at 43% for the 7 cm cut-off to predict very low birth weight. The negative predictive values (NPV) were high at 96% for low birth weight and 99% for very low birth weight (table 2).
Project staff reported that using the ruler to measure feet was simple to learn and to explain to others: overall, the approach was easily understood by parents. The exceptions were the four mothers who refused in hospital who reported a fear of association with measuring baby length for coffins, and the eight fathers who refused at the home visit because they had not been present for, or informed about, the participation in hospital on the first day.
Measuring newborn foot length in Tanzania using a simple transparent plastic ruler can identify babies needing extra care because they are low birth weight or premature. A foot length of less than 7 cm at birth was75% sensitive and 99% specific to identify very low birth weight babies (<1500 grams) and foot length less than 8 cm had sensitivity and specificity of 87% and 60% to identify those with low birth weight (<2500 grams), or 93% and 58% to identify those born premature (<37 weeks). The average newborn foot length increased by only 0.2 cm by the fifth day of life, and foot length measured on day five identified more than three-quarters of babies who were born small. The foot length cut-offs of 7 and 8 cm defined in this African setting were very similar to that previously recommended to identify small babies in Europe (7.2 cm) , and in Asia (between 6.3 cm and 7.2 cm) [12–15].
Strengths and limitations of the study
This was a cross sectional hospital based study of babies at birth with a community follow-up on the fifth day of life: findings suggest that measuring newborn foot length even days after birth could be useful to identify babies needing extra care. We defined 'small babies' both by birth weight and by gestational age but had a small sample size for analysis of the relatively rare occurrence of very low birth weight (<1500 g). Although the hospital based recruitment was not representative of the population level, the prevalence of low birth weight (<2500 g, 15%) was consistent with previous findings . Very little data exists at the population level with which to calculate prematurity rates in sub-Saharan Africa. Our estimate of 8% was lower than the 20% estimate reported previously from Malawi  with the implication that we may have underestimated the specificity for prematurity.
Public health implications
The utility of an anthropometric surrogate to identify small babies in the community will largely be determined by the implementing environment . The Lives Saved Tool (LiST)  estimates that in a high mortality setting such as Tanzania, 76% of neonatal deaths could be prevented if mothers and newborns were connected to comprehensive obstetric care, neonatal resuscitation and management of encephalopathy . However, the majority of newborns live in environments characterised by gaps in coverage of essential interventions, in quality of health services, and in equity .
In 2009, WHO and UNICEF released a joint statement promoting community volunteer programmes to make home visits to newborns as a strategy to improve survival, especially where access to facility-based skilled care is limited . In Tanzania, and other African and Asian countries, there is growing political support for such programmes, but some reluctance for volunteers to carry weighing scales, because of fears about cost, maintenance, and sustainability. In these settings, the use of a simple, cheap and reliable foot length tool to screen for low birth weight or premature babies could greatly strengthen the potential health impact of community volunteer programmes.
In figure 6 we consider foot length screening in the context of a hypothetical community volunteer programme for the estimated annual 800,000 home births in Tanzania, of whom 12,000 babies could be expected to be born with very low birth weight and 120,000 with low birth weight. If volunteers used a screening cut-off of 7 cm, 20,820 newborns would be targeted with advice about urgent medical care for very low birth weight, 9,000 of whom would be correctly targeted; 3000 very low birth weight babies would be missed. If a screening cut-off of 8 cm was applied, 415,560 newborns (half the home birth population) would be targeted with advice about additional care for low birth weight babies in the home, 100,920 of whom would be correctly identified. Using these cut-offs in combination, fifteen thousand low birth weight babies across the country would be missed, but all very low birth weight babies would be captured.
Two previous studies have reported high inter- and intra-observer agreement of foot length when measured by trained medical staff [11, 12]. However, the reliability of foot length when measured by birth attendants, community volunteers and the mothers themselves is not yet known. We are unaware of any research or programme using foot length as a proxy for birth weight or prematurity on a large scale. This home screening tool has unknown, but promising, potential for mortality impact when integrated with home-based interventions for moderately low birth weight babies including skin-to-skin care, exclusive breastfeeding, and recognition and referral of danger signs [3–6, 8].
In resource poor settings where neonatal mortality remains disproportionately high, there are many missed opportunities to provide extra care to premature or low birth weight babies born at home. This study has shown that the simple and inexpensive measurement of newborn foot length can be used to screen for low birth weight and prematurity, with great potential for impact on newborn survival.
Black RE, Cousens S, Johnson HL, Lawn JE, Rudan I, Bassani DG, Jha P, Campbell H, Walker CF, Cibulskis R: Global, regional, and national causes of child mortality in 2008: a systematic analysis. Lancet. 2010, 375 (9730): 1969-87. 10.1016/S0140-6736(10)60549-1.
Lawn JE, Cousens S, Zupan J: 4 million neonatal deaths: when? Where? Why?. Lancet. 2005, 365 (9462): 891-900. 10.1016/S0140-6736(05)71048-5.
Lawn JE, Kerber K, Enweronu-Laryea C, Massee Bateman O: Newborn survival in low resource settings--are we delivering?. BJOG. 2009, 116 (Suppl 1): 49-59.
Baqui AH, Ahmed S, El Arifeen S, Darmstadt GL, Rosecrans AM, Mannan I, Rahman SM, Begum N, Mahmud AB, Seraji H: Effect of timing of first postnatal care home visit on neonatal mortality in Bangladesh: a observational cohort study. BMJ. 2009, 339: b2826-10.1136/bmj.b2826.
Darmstadt GL, Bhutta ZA, Cousens S, Adam T, Walker N, de Bernis L: Evidence-based, cost-effective interventions: how many newborn babies can we save?. Lancet. 2005, 365 (9463): 977-988. 10.1016/S0140-6736(05)71088-6.
Lawn JE, Mwansa-Kambafwile J, Horta BL, Barros FC, Cousens S: 'Kangaroo mother care' to prevent neonatal deaths due to preterm birth complications. Int J Epidemiol. 2010, 39 (Suppl 1): i144-154. 10.1093/ije/dyq031.
Bhutta ZA, Darmstadt GL, Haws RA, Yakoob MY, Lawn JE: Delivering interventions to reduce the global burden of stillbirths: improving service supply and community demand. BMC Pregnancy Childbirth. 2009, 9 (Suppl 1): S7-10.1186/1471-2393-9-S1-S7.
Darmstadt GL, Lee AC, Cousens S, Sibley L, Bhutta ZA, Donnay F, Osrin D, Bang A, Kumar V, Wall SN: 60 Million non-facility births: who can deliver in community settings to reduce intrapartum-related deaths?. Int J Gynaecol Obstet. 2009, 107 (Suppl 1): S89-112. 10.1016/j.ijgo.2009.07.010.
Lawn JE, Kinney M, Lee AC, Chopra M, Donnay F, Paul VK, Bhutta ZA, Bateman M, Darmstadt GL: Reducing intrapartum-related deaths and disability: can the health system deliver?. Int J Gynaecol Obstet. 2009, 107 (Suppl 1): S123-140. 10.1016/j.ijgo.2009.07.021. S140-122
UNICEF: State of the World's Children 2010. 2009, New York: United Nations Children's Fund (UNICEF)
James DK, Dryburgh EH, Chiswick ML: Foot length--a new and potentially useful measurement in the neonate. Arch Dis Child. 1979, 54 (3): 226-230. 10.1136/adc.54.3.226.
Daga SR, Daga AS, Patole S, Kadam S, Mukadam Y: Foot length measurement from foot print for identifying a newborn at risk. J Trop Pediatr. 1988, 34 (1): 16-19.
Hirve SS, Ganatra BR: Foot tape measure for identification of low birth weight newborns. Indian Pediatr. 1993, 30 (1): 25-29.
Mullany LC, Darmstadt GL, Khatry SK, Leclerq SC, Tielsch JM: Relationship between the surrogate anthropometric measures, foot length and chest circumference and birth weight among newborns of Sarlahi, Nepal. Eur J Clin Nutr. 2007, 61 (1): 40-46. 10.1038/sj.ejcn.1602504.
Mathur A, Tak SK, Kothari P: 'Foot length'--a newer approach in neonatal anthropometry. J Trop Pediatr. 1984, 30 (6): 333-336.
Ho TY, Ou SF, Huang SH, Lee CN, Ger LP, Hsieh KS, Huang SM, Weng KP: Assessment of growth from foot length in Taiwanese neonates. Pediatr Neonatol. 2009, 50 (6): 287-290. 10.1016/S1875-9572(09)60079-6.
Armstrong Schellenberg JR, Mrisho M, Manzi F, Shirima K, Mbuya C, Mushi AK, Ketende SC, Alonso PL, Mshinda H, Tanner M: Health and survival of young children in southern Tanzania. BMC Public Health. 2008, 8: 194-10.1186/1471-2458-8-194.
Eregie CO: A new method for maturity determination in newborn infants. J Trop Pediatr. 2000, 46 (3): 140-144. 10.1093/tropej/46.3.140.
Sunjoh F, Njamnshi AK, Tietche F, Kago I: Assessment of gestational age in the Cameroonian newborn infant: a comparison of four scoring methods. J Trop Pediatr. 2004, 50 (5): 285-291. 10.1093/tropej/50.5.285.
Dubowitz LM, Dubowitz V, Goldberg C: Clinical assessment of gestational age in the newborn infant. J Pediatr. 1970, 77 (1): 1-10. 10.1016/S0022-3476(70)80038-5.
UNICEF and WHO: Low Birthweight: Country, regional and global estimates. 2004, [http://whqlibdoc.who.int/publications/2004/9280638327.pdf]
van den Broek N, Ntonya C, Kayira E, White S, Neilson JP: Preterm birth in rural Malawi: high incidence in ultrasound-dated population. Hum Reprod. 2005, 20 (11): 3235-3237. 10.1093/humrep/dei208.
Victora CG: LiST: using epidemiology to guide child survival policymaking and programming. Int J Epidemiol. 39 (3): 650-652. 10.1093/ije/dyq097.
Kinney MK, Black KJ, Cohen RE, Nkrumah B, Coovadia F, Nampala H, PM Lawn JE: Sub-Saharan Africa's Mothers, Newborns, and Children: Where and Why Do They Die?. PLOS Medicine. 2010, 7 (6): e1000294-10.1371/journal.pmed.1000294.
WHO and UNICEF: Home visits for the newborn child: a strategy to imporve survival: Joint statement. 2009, [http://whqlibdoc.who.int/hq/2009/WHO_FCH_CAH_09.02_eng.pdf]
The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2458/10/624/prepub
We are thankful to the members of staff at the maternity hospital at Ligula Hospital and to all caregivers of newborns enrolled in this study. Elibariki Mkubo, Mwajuma Chemba, and Rose Sagga collected the foot length data. The work was supported by grants from The Batchworth Trust and the Bill and Melinda Gates Foundation through Save the Children US. The funders had no role in the design and conduct of the study; in the collection, analysis and interpretation of the data; or in the preparation, review, or approval of the manuscript. This paper is published with the permission of the Director-General of NIMR, for whose support we are grateful.
The authors declare that they have no competing interests.
TM and JAS were responsible for the study concept and design. Statistical analysis and interpretation was conducted by TM, JJ, SP, MT and JAS. TM wrote the first draft, TM, JAS, SP, JJ and MT revised the paper and contributed to discussion. All authors read and approved the final manuscript. TM acts as guarantor for the study.
Authors’ original submitted files for images
About this article
Cite this article
Marchant, T., Jaribu, J., Penfold, S. et al. Measuring newborn foot length to identify small babies in need of extra care: a cross sectional hospital based study with community follow-up in Tanzania. BMC Public Health 10, 624 (2010). https://doi.org/10.1186/1471-2458-10-624
- Birth Weight
- Premature Baby
- Receiver Operating Characteristic
- Home Birth
- Extra Care