The analyses were carried out using the fifth round of National Family Health Survey (NFHS-5) conducted during 2019–2021 in 28 states and 8 union territories of India. NFHS is a large-scale multi-stage cross sectional survey conducted in representative sample of households throughout India. The protocol for the NFHS-5 survey with the content of all the survey questionnaires, was approved by the Institutional Ethical Review Board of International Institute for Population Sciences, the ICF Institutional Ethical Review Board and was reviewed by the U.S. Centers for Disease Control and Prevention (CDC). NFHS-5 had collected cross-sectional data for wide range of socioeconomic, demographic and health entities from 707 districts of India, which included 30,198 primary sampling units. A primary sampling unit in an urban area is census enumeration block, and a village in rural area. Prior consent was procured from all the women before participating in the survey. NFHS-5 collected month wise data explicitly identifying the key events in the reproductive history of the women within the period of 60 months prior to the interview date. Chronological sequencing of the data provided an opportunity to identify the event of terminating the pregnancy and the following contraceptive use in each calendar month. The event of the recent termination of pregnancy was recorded in the month when the pregnancy was terminated using the letter ‘M’ for miscarriage, ‘A’ for abortion and ‘S’ for stillbirth. Among the 724,115 women in the reproductive age group 2.9% pregnancies were terminated [15]. The final sample consisted of 6,862 women for the analysis who underwent abortion within the period of 60 months prior to the survey date.

Main outcome variable for the study was adoption of any contraceptive method after the event of abortion. The utilization of any contraceptive method was determined by analyzing data from a calendar matrix, which displayed month wise data for various contraceptive options. No exposure period was considered as there is evidence that earliest ovulation can occur as early as on tenth postabortal day suggesting that they should be supplied with contraceptive method within ten days after the abortion [24,25,26].

The explanatory variables were main reason for abortion (unplanned pregnancy, contraceptive failure, any health related reasons and other reasons like sex of the child, other family member’s opinion, last child too young, economic reasons and foetus had congenital abnormality), method of abortion (medical method of abortion, manual vacuum aspiration, surgical methods and other unlisted methods too which may include unsafe methods of abortion), person who performed the abortion (trained healthcare personal, Ayush/Dai (AYUSH-Ayurveda, Yoga & Naturopathy, Unani, Siddha & Homeopathy and Dai are traditional birth attendant), untrained person like a relative or a friend and the woman herself) and the timing of abortion (first trimester, second trimester and third trimester). Additional sociodemographic and socioeconomic explanatory variable considered were wealth index of the household, age (estimated age at the time of abortion from the current reported age), caste (Schedule caste (SC), Schedule tribe (ST), Other backward caste (OBC is a collective term used by the Government of India to identify the educationally or socially disadvantaged castes) and other categories of caste), religion (Hindu, Muslim, Christian and Other), educational attainment (No education, primary education, Secondary education and higher), place of residence (Urban and rural) and exposure to mass media. Stata 14.2 statistical package was used to carry out all the analyses. Sampling weights available with the dataset were applied for all the analyses to account for the survey design.

The multinomial regression model was used to identify the correlates of post-abortion contraceptive adoption among the women seeking abortion services. The contraceptive methods were grouped into four categories; permanent methods, which consist of male and female sterilization cases; reversible method considering the method of intrauterine contraceptive device (IUCDs); short term modern method which included pills, condoms, foam & jelly, injectable and diaphragm and traditional method including rhythm, withdrawal and periodic abstinence. Those who did not use any method were categorized under the heading of ‘no method’. For the analysis purpose permanent methods and IUCDs have been combined as these two are long term methods as compared to the listed methods.