Page 74 - IJPS-7-1

Risk factors for cesarean section

get a total of 83 CS deliveries. We tried approaching all 83 women who delivered through CS, out of which we could reach
70 mothers. Out of the 83 women who had CS, three of them moved out of the service area and 10 were not reachable
even through mobile. Thus, finally, we finally arrived at 70 cases, and we went on choosing eligible age- and year-matched
controls in the ratio of 1:1. Thus, the study progressed with 70 matched pairs (cases and controls). The controls were then
age matched for the cases by selecting them with a within an age difference range of ±2 years. The controls were also
month-matched by choosing them of the same month of delivery as that of the cases.
After obtaining the list of cases and eligible controls, four trained doctors were chosen to do data collection. The data
collectors were enrolled after familiarizing them regarding the questionnaire and objectives of the study. Data collection
was done through regular house to house visits on a daily basis, data collection process was supervised and reviewed by
the principal investigator posted in health center to ensure completion of data and to address the issues faced by the data
collectors. If the households remained closed even after making two home visits, we tried contacting them through phone
and if they were: Not traceable through phone, moved out of service area, and houses remaining locked after two visits
were excluded from the study. The same excluding criteria were applicable for the controls too.
The importance of the study was explained to the women before administration of the questionnaire. They were
enrolled only after obtaining informed consent and assured confidentiality. Interview had three sections: Sociodemographic
characteristics; indications for cesarean; and last section had delivery details. Since delivery being a significant life event,
the recall period of 3 years was chosen as the chances of remembering the details are highly possible. The indications for
CS were asked from the mother and cross-verified with the discharge slip for all cases. The delivery details were obtained
from the delivery card.

2.4. Data analysis
Data were entered into Microsoft Excel 2013 and analyzed using STATA 14 (Stata Corp., 2012). Continuous variables
were summarized as mean (SD) or median (IQR) depending on their distribution. The sociodemographic risk factors were
summarized as proportions. Association between CS and various risk factors and sociodemographic factors was analyzed
using Chi-square test with 95% confidence intervals. P < 0.05 was taken to be statistically significant. Bivariable logistic
regression was carried out for each variable and those variables whose odds ratios (OR) were significant at P < 0.2 were
taken into the model for calculating adjusted odds ratio.

2.5. Ethical statement
The study was done as a component of MCH care services delivered in the urban health center toward obstetric care. Any
queries arising from the participants were clarified after obtaining written consent. The study was reviewed and approved
by the department review committee.

3. Results
There were totally 140 mothers included into the study (70 case and 70 age and period matched controls). Table 1 shows
the sociodemographic characteristics of the study participants. The mean age ± standard deviation (SD) of the cases and
controls were 29.4 ± 4.9 and 29.2 ± 4.5 years, respectively. Among 140 mothers interviewed, almost 50% of the cases and
controls had education up to 10 standard. Majority 61 (87.1%) and 62 (88.5%) of the cases and controls were Hindu by
th
religion. About 56 (80%) of the cases and controls 60 (85%) were housewives by occupation. More than 2/3 of the study
rd
population in each group belonged to middle or upper class according to Modified BG Prasad classification (cases 78%
and controls 66%) (Majhi & Bhatnagar, 2021). Both the groups had almost equal proportion of low birth weight (cases
48% and controls 52%). A higher proportion of the cases (31.4%; n = 22) had delivered in a private institution compared
with controls (18.5%; n = 13).
Among sociodemographic factors, only socioeconomic status (P = 0.04) emerged as a significant risk factor for CS.
Other factors such as education, occupation, and religion and birth weight of the baby of the baby did not show significant
association.
The association of cases and controls with the various antenatal complications assessed in the study is represented
in Table 2. In unadjusted analysis [Table 2], history of gestational diabetes (OR = 3.7; CI: 1.1 – 14.2), previous LSCS
(OR = 29.5; CI: 3.8 – 227.4), and malpresentation during delivery (OR = 5; CI: 10 – 24.1) emerged as the three significant
risk factors for CS in this study.
Table 3 shows the final multivariable analysis which includes variables whose OR had a significance level P < 0.2 in
unadjusted bivariable analysis. In multivariable analysis, previous LSCS (aOR of 45.4, CI: 4.3 – 483.6), malpresentation

68 International Journal of Population Studies | 2021, Volume 7, Issue 1