Effect of antenatal care on low birth weight prevention in Lao PDR : A case control study

Low Birth Weight (LBW) is a worldwide public health problem, Background: which subsequently may affect the health status of the child. Lao PDR has high incidence of LBW. Antenatal care (ANC) is provided to improve maternal and child health outcomes. The aim of this study was to identify the effect ANC on LBW prevention in Lao PDR. This case control study was conducted in tertiary hospitals of Lao Methods: PDR. The ratio of case: control was 1:3, of which there were 52 cases and 156 controls that passed the inclusion criteria included in the study. In our analysis information on pregnancy and ANC including height of mother, maternal weight gain during pregnancy, maternal gestational age at delivery, type of delivery, supplementary vitamins, and other covariates including age, marital status, educational attainment, occupation, family income, health insurance, family size and living condition were described and determine their association with LBW using multiple logistic regression analysis. There were only 32.69 % of complete ANC among cases and 57.69% Results: in control. Incomplete ANC (<4 times) were significant increased the odds of having LBW (adj. OR=2.97; 95%CI: 1.48 to 5.93; p-value =0.002). Other covariates which also influenced LBW were having maternal weight gain during pregnancy less than 10 kg. (adj.OR=2.28; 95%CI: 1.16 to 4.49; p-value = 0.017), maternal gestation age at delivery less than 40 weeks (adj. OR=3.33; 95%CI: 1.52 to 7.32; p-value =0.003). Complete ANC could help both mother and child in term of weight Conclusion: gain and full term delivery which may effect on LBW reduction.


Introduction
Low birth weight (LBW) refers to a baby who has a birth weight of le<2,500 grams 1 . LBW may cause birth asphyxia, amniotic fluid aspiration, hypoglycemia and hyponatremia. An infant weight of 1,500-2,500 grams has been shown to have 5-10 times higher mortality rate than normal infants 2 . During 2005-2010, LBW incidences in countries of the Association of Southeast Asian nations ASEAN were 21.0% in Philippines, 11.0% in Malaysia, Cambodia and Lao PDR, 9.0% in Indonesia, 7.0% in Thailand and 5.0% in Vietnam 3 . The incidence of LBW reflects a country's socio-economic development 4 . Mortality of LBW babies is as high as 1% when compared with 0.2% among normal children 5 . At the 2005 World Summit a plan was announced to improve quality of life called Millennium Development Goals (MDGs)", covered goals to be achieved worldwide by the year 2015 5 .
Maternal and child health problems have been considered as indicators of the health service performances. In Southeast Asia, 28% of all deaths of 1 month-old infants were from infection, and 20% from preterm birth and LBW 6 . LBW is commonly used as an indicator of health status and is important for national health policy development 7 . Important factors associated with LBW are maternal factors, such as socioeconomic status, food consumption behaviors, calorie intake, urinary tract infection and prenatal care, smoking, genital infections, maternal health and stress 8,9 . Antenatal care (ANC) is care provided to pregnant women by health personal. Care includes risk identification, prevention and management of pregnancy-related or concurrent disease, and health education and health promotion 10 . The World Health Organization's ANC model, also known as focused or basic ANC is a goal-orientated approach to delivering evidence-based interventions at four critical times for ANC during pregnancy 11 ; therefore completed ANC in this study is having ANC for at least 4 times.
High incidence of LBW and high mortality of both mothers and children in Lao PDR has been hypothesized as being caused by various influencing factors, such as low socio economic status and limited access to qualified health services related to pregnancy 12 . However, there are limited studies identifying the role of ANC in reducing LBW incidence in Lao PDR. Therefore this study aimed to determine whether having ANC at least four times during pregnancy could help reduce LBW when controlling for other covariates related to socioeconomic and pregnancy factors. The results could be used to develop appropriate measures for prevention of LBW and obtaining better maternal and child health statuses.

Participants
Postpartum mothers who had delivered babies and came for regular checkup after delivery in four tertiary hospitals in Vientiane, Lao PDR between July and December 2016 were included in this study. The four hospitals were Mahosot Hospital, Sethathirat Hospital, Mothers and Child Center Hospital, and Mittraphab Hospital.
The sample size was calculated using the formula for the analysis of a relationship in a case-control study 13 . The formula indicated that the sample size of the study group (cases) should be 52 participants. The control group was 3 times the size of the study group (case:control ratio of 1:3). Therefore, the control group included 156 participants, which made a final total of 208 participants. Cases and controls were not matched for demographics.
Inclusion criteria were mothers who had delivered babies in the four hospitals during the study period, who were 18 -49 years of age. This age range was chosen as the reproductive age of women is 15-49 years old. However, those aged below 18 are considered as a vulnerable group, therefore we selected 18-49 year-olds. The exclusion criteria were mothers who delivered twins, did not live in the study area and who were not willing to participate. The samples in this study were divided into two groups: 1) Case group: mothers of babies whose birth weight was <2,500 grams (LBW); 2) Control group: mothers of babies whose births weight was ≥2,500 grams who were born during the same period as the cases.

Data collection
The tool for data collection was a structured questionnaire interview that consisted of seven parts (Supplementary File 1): Part I, general information; Part II, sociodemographic characteristics; Part III, knowledge of health care during pregnancy; Part IV, maternal factors and pregnancy status; Part V, environment factors and support for ANC; Part VI, prenatal distress (as assessed using the Edinburgh Prenatal Depression Scale); Part VII, obstetric information at delivery (gathered from the mother's medical records).
The questionnaire was content validity tested by five experts in terms of theory and understanding. Unclear questions were edited and some information that was missing was added. Reliability was tested among 30 mothers (from Xaythany and Sisattanak Hospitals), indicating the high reliability with the Cronbach's alpha coefficient of 0.84.
Data was collected by four physicians who were trained in using the questionnaire. These interviewers were blinded for the infant status of normal or LBW. The interview of the participants took place in a postpartum patient room or nursing room within 5 days of agreement to participate.
Statistical analysis STATA version 10.0 was used to analyze data 14 . Descriptive statistics frequency, percentage, means, standard deviation, minimum, maximum were used to present data on the following characteristics: characteristics of mothers, knowledge on health care during pregnancy of the mother, pregnancy status, environment factors and support for ANC, prenatal distress and obstetric information at delivery.
Simple logistic regression was used to determine factors associated with LBW. The factors which had association with LBW (p-value <0.25) were analyzed using multiple logistic regression. Multiple logistic regression was applied to identify the association of ANC and LBW when controlling for other covariates, presenting adjusted OR, 95% confidence interval (95% CI) with the levels of significance at 0.05 15 .

Ethical statement
We submitted the proposal and questionnaire to the Ethical Committee of Khon Kaen University, Thailand (reference No. HE 592087) and the Ethical Committee of University of Health Sciences, Lao PDR (reference No.13/16) for approval. Both committees granted approval of this study, including the reliability test of the questionnaire. Written informed consent was obtained from all participants.

Results
The study involved a total of 208 mothers, of which 52 were cases and 156 were controls. There were a higher proportion of younger mothers (<20 years) among cases than controls (17.31% and 11.54%, respectively). Controls had higher educational attainment of upper secondary (43.59%) compared to cases (28.85%). More controls lived in urban areas than cases (60.26% and 51.92%, respectively), and 21.15% of cases had average monthly family income ≤1,000,000 (Kips) which was 9.62% among controls (see Table 1).
Only 32.69 % of cases completed ANC (≥ 4 times) check-ups while 57.69% controls completed these. The bivariable analysis of ANC, socioeconomic and maternal factors with LBW showed that going to ANC check-ups <4 times had higher odds of having LBW than those who went to at least 4 times (OR=2.80; 95%CI: 1.44 to 5.43; p-value = 0.002).
In the bivariate analysis, which considers the association of one independent variable with the outcome (LBW), we selected factors with a p-value <0.25 to proceed to the multivariable analysis. These factors were height of mother (cm), taking any supplementary vitamins, prenatal depression, maternal weight gain during pregnancy (kg), maternal gestational age during delivery (weeks), and type of delivery (see Table 2).  Table 3).  29 , and some studies reported that the older the mother, the higher the risk of having LBW infants 30 . LBW babies among older mothers, whose age is 35 years and above, was 23.89%. It was significantly higher than the percentage of LBW babies for mothers in other age categories (p = 0.004) 16,31 . For maternal weight gain during pregnancy it was found that weight gain <10 kg were risk factors for LBW in the present study, which is similar to the result of a study in Indonesia among others 26,32-34 . Low maternal weight gain reflects poor child growth, which puts both mother and child at risk for morbidity and mortality 35,36 .
Maternal gestational age at delivery of <40 weeks was another associated factor for LBW in the present study, which is similar to a result found in Malaysia 37 . Some studies showed that the risk factor of LBW infants was a gestational ages of <37 weeks 24,38 . When the mother delivers a baby before the baby is at full term, the baby is not fully grown. Therefore the babies are more likely to be small (LBW) and have higher risk for mortality since some organs such as lung is not fully functioning.
One limitation in this case-control study could be data collection bias due to interviewer prejudices. However, we minimized this by blinding the interviewers; therefore the manner in which they asked the questions were the same in both case and control groups.

Conclusion
This hospital-based case-control study was conducted in Vientiane, Lao PDR and indicated that ANC checkups at least 4 times could help reduce LBW of babies. Consequently, policy should improve coverage and quality of ANC of at least 4 times for all pregnant women in this population.    ○ Abstract: suggest changing the word "problem" to concern. "Background: Low Birth Weight (LBW) is a worldwide public health problem.." ○ Abstract: suggest changing this sentence from: Antenatal care (ANC) is provided to improve maternal and child health outcomes." to "Women are offered antenatal care (ANC) to improve maternal, pregnancy and newborn outcomes". ○ Introduction: second line "le" should be "less than" ○ care (ANC > 4 times) and the low birth weight (LBW) in Lao PDR. It demonstrated the significance of ANC on the prevention of neonatal complication however there are a number of issues with the methods and analysis section that need to be clarified with some of the conclusion are not correspond with the results. The detail of comments by section is below: Abstract: The conclusion that complete ANC help mother and child in term of weight gain and full term delivery is not from the analysis of the results. The multiple logistic regression from this study only showed that antenatal care and maternal weight gain and GA at delivery had effect to the LBW. Total birth, total LBW, total preterm birth and total cesarean deliveries in the study period should be added. ○ Table 1-Maternal age should be 18-20 instead of <20 (<18 was excluded from this study), Famer should be changed to farmer. Only mean or median should be selected depend on the distribution of data. If median was selected, the min-max should be changed to interquartile range. Maternal medical and obstetrics complications should be added ○ because it was an important factor for LBW. Table 2-The gestational age normally divided to <37, 37-<42, > 42 instead of 40. The BMI should be used instead of maternal height and why 10 kg was used instead of the recommended weight gain during pregnancy which depends on maternal BMI. Very high cesarean section in both groups was shown, please add the reasons or indications.

○
The continuous data such as age, income should be analyzed in continuous form, especially income which was statistical difference between two groups if analyzed in continuous form and should be analyzed in multiple logistic regression analysis. The downgrade of continuous data to ordinal data should be avoided.

○
Normally, the premature birth is the birth before 37 weeks GA and is the most important cause of LBW. This should be used instead of 40 weeks. ○ Conclusion: As mentioned above, the new WHO recommendation in 2016 recommend at least 8 times of antenatal care to reduce perinatal mortality.

If applicable, is the statistical analysis and its interpretation appropriate? Partly
Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? Partly