Predictors of congenital anomalies among newborns in Arsi zone public hospitals, Southeast Ethiopia: a case-control study

The multi-center institutional-based case-control study design was conducted among newborns in Arsi zone public hospitals from December 01, 2020 to May 30, 2020. The study was conducted in Arsi Zone Public Hospitals. Arsi Zone is located in Oromia Regional State Southeastern part of Ethiopia. The capital city of Arsi zone is Asella, which is 174 km away from Addis Ababa, the capital city of Ethiopia. The hospitals are Asella Referral Hospital, Bekoji, Abomsa, Robe, Gobessa, Kersa, Sude, and Bale District Hospitals.

The source population is all newborns delivered in Arsi Zone Public Hospitals. The study population is all newborns delivered in Arsi Zone Public Hospitals and fulfill the inclusion criteria during the study period.

The sample size was calculated by using Fleiss with continuity correction factor formula with a case-control ratio of 1:3; power of 80%, a significance level of 5%, the proportion of controls exposed (alcohol drinking) 55% [16], assuming the minimum odds ratio to be detected was 2. Using this value the expected percent of exposure among cases was 70%. The computed sample size was 418 (105 cases, and 313 controls).

All public hospitals found in the Arsi zone were included and an equal proportion allocation of samples was performed to include participants from each hospital. All mothers who delivered infants with congenital anomalies were invited to participate in the study. Mothers who delivered infants with no congenital anomalies were also included in the study using a systematic random sampling procedure. In the present study, all newborns with visible congenital anomalies and the next three newborns (delivered after cases) without congenital anomalies were included.

In our study, congenital anomalies are considered as the dependent variable. On the other hands, Socio-demographic characteristics (age, educational status, marital status, religion, ethnicity, income, occupation, husband’s education, and occupation), obstetric history (folic acid during and prior to pregnancy, history of antenatal care (ANC), contraceptive use, parity, gravidity, gestational age, family history of congenital anomalies, previous history of a child with congenital anomalies), medical condition-related factors (unidentified medication use, medical illness), drug and radiation (smoking and alcoholic history, khat chewing, use of herbal medicines, radiation, exposure to chemicals like pesticides), and others like sex of the newborn, consanguinity are considered as independent variables.

Cases:- are mothers who gave birth to babies with externally visible defects to any system of the body identified by clinical examinations by experienced medical doctors at the time of delivery.

Controls:- are mothers who gave birth babies without externally visible defects to any system of the body after clinical examination by experienced medical doctors at the time of delivery.

Consanguineous marriage:- couples related by blood at least to the second cousin.

Radiation:- Exposure to x-ray.

Data were collected using a pre-tested structured interviewer-administered questionnaire, which includes questions of socio-demographic, obstetric, medical, alcohol intake and smoking history, use of herbal medicine, exposure to chemicals, radiation, and folic acid supplementation (5 mg of folic acid was given throughout first trimesters from conception to 12th week of pregnancy). The questionnaire was prepared in English then translated into Amharic and Afan Oromo and then back to English by a third party who is native in Amharic and Afan Oromo respectively and good in English to ensure consistency of translation. Newborns were carefully examined and assessed for congenital anomalies by experienced medical doctors. Following a proper medical examination, the data were gathered from mothers regarding the information explained above using a structured questionnaire.

Six data collectors and six supervisors who are fluent in Amharic and Afan Oromo were involved in data collection. Accordingly, the supervisors were supervised respective data collectors during the data collection time and receive the collected and crosschecked questionnaires from them. Finally, the supervisors reported and discussed and crosschecked data with the investigator on a daily basis throughout the data collection period.

To maintain data quality, data collectors who are fluent in Amharic and Afan Oromo were selected based on their educational level and experience in data collection. Detail orientation and training about data collection and the purpose of the study were given to supervisors and data collectors by the principal investigator. Based on previous related peer-reviewed published studies, the questionnaire were developed by the investigators. A pretest study was conducted at Halila health centers on fifty individuals to see for validity and reliability of tools. The collected data were reviewed and checked for mistakes, completeness and consistency by the investigators and supervisors on a daily basis during data collection.

The collected data were coded, checked, and entered using Epi-Info version 7.0 and exported to IBM SPSS Version 21 for cleaning, checking for missing values, and analysis of data. Descriptive statistics and frequency tables were used to describe study populations.

A total of 418 (105 cases and 313 controls) mothers were successfully interviewed at the labor ward after delivery of a child with a response rate of 100%. About 84.7% of controls and more than half of cases (66.7%) were in the age group of 20–34. Nearly above half of the women among cases (54.3%) and most of the women among controls (77.6%) were married as presented in Table 1. The majority of women (54.3% of cases and 46% of controls) were Muslims in religion. Around 44.8% of cases and 42.2% of controls were illiterate. Almost half of the women.

(50.5% of cases and 52.7% of controls were urban dwellers. Concerning the ethnicity of mothers, the majority of cases (67.6%) and controls (59.1%) were Oromo.

As it is presented in Table 2, the majority of women (81% of cases and 75.1% of controls) were multigravida. Almost half of the controls (52.1%) and 41.9% of cases were born at second order. Fifty-six percent of cases and 74% of controls were born at 37 weeks and above and a small proportion of both cases (2.9%) and controls (2.9) were born prior to 28 weeks. Nearly equal proportion of cases (5.7%) and controls (5.4) had a previous abortion history. Regarding family history and previous history of children with congenital anomalies, less than 20% of both cases and controls reported a family history and previous history of a child with congenital anomalies. This study revealed that the number of females with congenital anomalies (60%) were higher than those without congenital anomalies (38.7%). About 41% of mothers of children with congenital anomalies and 55% of mothers of children without congenital anomalies were taken folic acid supplementation in the early period of pregnancy. Similarly, contraception use around conception was 59% in mothers who had children with congenital anomalies and 44.7% of mothers who had children without congenital anomalies.

Among 105 newborns presented with congenital anomalies, neural tube defects (70.5%) were the most manifested anomalies, succeeded by an orofacial cleft (10.5%) and musculoskeletal anomalies (8.6%). Out of the diagnosed anomalies, genitourinary and gastrointestinal anomalies were 4.8% (Fig. 1). Furthermore, the Percentage distributions of selected congenital anomalies by type of anomaly is presented in Table 3.

In the present study, consanguineous marriage was reported in 58.1% of case mothers and 20% of control mothers (Table 4). About 64% of case mothers and 18% of control mothers have been drinking alcohol during their pregnancy period. Out of the study participants, maternal medical illness during pregnancy in mothers of newborns with and without congenital anomalies was 39 and 20.4%, respectively. About 67% of case mothers and 21% of control mothers were exposed to chemicals like pesticides at the time of pregnancy. Chewing khat and using other herbal medicines were 65.7 and 46.7% of mothers who had infants with congenital anomalies and 19.5 and 19.2% in mothers who had infants without congenital anomalies, respectively.

The binary logistic regression analyses (using crude odds ratio) displayed that consanguinity, alcohol drinking during pregnancy, smoking during pregnancy, maternal illness, history of drug use during the current pregnancy, khat chewing during pregnancy, exposure to chemicals during pregnancy, exposure to herbal during pregnancy, lack of folic acid supplementation during early pregnancy, and use of contraception around pregnancy were significantly associated with the occurrence of congenital anomalies. However, maternal age, educational level of the mother, parental occupation, residence, gravidity, birth order, family history of CAs, previous history of a child with CAs, ANC follow-up, stillbirth, and abortion were not statistically significant (Table 4).

Those statistically significant variables in the binary logistic regression analyses were entered into a multivariable logistic regression to identify the predictor variables associated with congenital anomalies. As a result, women who have been drinking alcohol during the current pregnancy were 3.48 times more prone to give birth to newborns with congenital anomalies than their counterparts (AOR = 3.48; 95% CI: 1.38, 8.74). Likewise, the likelihood of having newborns with congenital anomalies was six and four times higher for women who had a maternal illness during pregnancy (AOR = 6.10; 95% CI: 2.39, 15.57) and chewing khat during pregnancy (AOR = 4; 95% CI: 1.49, 10.65), respectively. Women who have been directly exposed to chemicals in work fields like pesticides during pregnancy were 4.76 times more likely to have newborns with congenital anomalies as compared to those who did not expose (AOR = 4.76; 95% CI: 1.57, 14.47). Besides, women who had not taken folic acid supplementation during early pregnancy were three times more likely (AOR = 3.25; 95% CI: 1.6, 6.61) to have newborns with congenital anomalies as compared to those who had taken folic acid supplementation (Table 4).

Although diversified study participants were included in the present study,hospital-based unmatched case-control study makes the result of our study less generalizable to the community. The present study did not take the terminated pregnancy/cases of CAs into account rather it considered only externally visible CAs which are identified by clinical examination at birth.