This study employed a descriptive cross-sectional study design. The study was conducted at Mbarara Regional Referral Hospital (MRRH) located along Kabale road, about 270km from Kampala, the capital city of Uganda. MRRH serves about 4 million people in the entire catchment area of south western Uganda and handles about 21 deliveries per day. We recruited newborns from labour suit and maternity ward of the Obstetrics and Gynaecology Department in the months of July and September 2017.
Characteristics of study participants
In this study, systematic random sampling at an interval of 2 newborns, on a daily basis, using patient register as sampling frame was used to recruit 638 eligible newborns within their first 24hours after birth. . We excluded 85 newborns who were not traced on maternal ward for CHWs to take their anthropometric measurements. Therefore, data from 553 newborns were analyzed. Purposive sampling was used to select two midwives to participate in the study. Similarly, two community health workers (CHWs) with at least ordinary secondary education and working experience of 6 months as CHWs in Mbarara municipality were recruited to take anthropometric measurements. These CHWs were introduced to MRRH setting for familiarization and acclimatization. Midwives recruited to collect the data enrolled newborns into the study, took birth weight and other anthropometric measurements. Using unique identifiers, the CHWs repeated measurements of other anthropometric parameters on the same newborns.
Using Kish Leslie formula  and a design effect of 2.0 used in childhood anthropometrics , a sample size of 638 newborns was determined. From a total of 1,200 newborns born during the study period, 638 newborns were recruited and assessed for anthropometrics. Of the 638 newborns recruited, 85 newborns were not assessed for anthropometrics by midwives CHWs because they could not be traced on the martenity ward (Figure 1). Thus only data from 553 newborns were analysed and used in reporting the findings.
Two midwives working in the maternity wing of the hospital and two CHWs from Mbarara municipality were recruited and trained for two days, under one roof, on anthropometric techniques. This was to ensure harmonization with the use of anthropometric techniques prior to data collection. Training was done using tools already validated for use in Uganda .
Our main outcome variable was birth weight of newborns. Birth weight was determined by midwives using a weighing scale (Salter model 180). Weighing scale calibration was done on a daily basis using a 1 kg stone throughout the process of data collection. The naked newborn would be put lying supine on the leveled pan scale of the weighing scale and the midwife read and record weight in grams nearest to 2 decimal places.
Anthropometric parameters including birth weight, circumferences of head, mid upper arm, chest, thigh, and calf, foot and body lengths of newborn were measured . Using non extendable measuring tapes, with a width of 1.0 cm and subdivisions of 0.1cm, midwives measured circumferences of newborns’ head, mid upper arm, chest, thigh, and calf. Head circumference was measured in a clockwise direction starting from the glabella through the occipital prominence back to the starting point on the glabella. Mid upper arm circumference was measured from midpoint between tip of shoulder and elbow by rolling the tape around the arm to the starting point. Chest circumference was measured by fixing the starting point of a tape measure at the tip of xiphoid process and rolling it around the back of the newborn to the starting point at the tip of the xiphoid process. While keeping the newborn lying supine, the measurement was recorded only at full inspiration.
The thigh circumference was measured from midpoint between anterior superior iliac spine and knee joint by rolling the tape around the thigh and back to the starting point. The calf circumference was measured from midpoint between knee and ankle joints by rolling the tape around the calf and back to the starting point. The length of the newborn was measured using calibrated length measuring board. The newborn was made to lie supine on the calibrated measuring board. The heel of the newborn was fixed on zero point, then the body length was measured from the heel to the crown and recorded in centimeters. Foot length was measured using a transparent ruler starting from the heel to the tip of the big toe of the right foot. All the measurements, except birth weight, were independently repeated on each newborn by the CHWs. For each anthropometric parameter, two measurements were made and its average calculated and recorded. The midwives recruited newborns into the study, took anthropometric parameters before admitting them into the maternity ward. Unique identifiers were then assigned to each newborn recruited. Unique identifiers were used by CHWs access the newborns, previously recruited by midwives, and take their anthropometric measurements.
Data was entered in Microsoft excel version 2010 from where it was edited; checked for completeness and consistency. Data were then exported into SPSS version 20 for cleaning and analysis. Categorical characteristics of participants were analyzed and summarized using frequencies. Continuous data were summarized and recorded as means (standard deviation).
Pearson correlation analysis of linearity between birth weight and all other anthropometric parameters understudy was conducted, and Correlation coefficient (r) and confidence intervals reported. Non-parametric receiver operating characteristic (ROC) curve analysis was carried out to determine the sensitivity, specificity and area under the curve (AUC). Finally the cut offs for various anthropometric measurements were calculated to predict LBW. We used an independent paired t-test to find out if there was a statistical significant difference between the anthropometric measurements taken by midwives and those taken by CHWs. Mean, standard deviation, mean and p values at 95% confidence interval were determined and reported.