Understanding patterns of unplanned pregnancies and abortions in India during the COVID-19 pandemic using Google Trends data

February  4, 2021

Arnab Dey1, Nabamallika Dehingia1, Anita Raj1
1 Center on Gender Equity and Health, University of California San Diego, USA

The COVID-19 pandemic has affected almost every aspect of our lives in profound ways. In addition to the catastrophic loss of life and livelihoods 1,2, the pandemic has severely impacted the physical and mental health of people around the world. 3,4 In forcing us to change the ways we eat, sleep, work and socialize, the pandemic has altered our health behaviors significantly. One of its major effects on our health and wellbeing has been on our Sexual and Reproductive Health (SRH) behaviors. A recent study conducted in Turkey found that female sexual desires and frequency of sexual intercourse increased during the pandemic, while contraceptive use and continuation declined during the period.5 Another study conducted in Italy indicated that unplanned pregnancies and pregnancy terminations might have increased during the pandemic. 6 These problems are further accentuated in countries like India that already have low rates of contraceptive use,7 and high rates of unintended pregnancies and abortions. 8,9

While studies indicate that unplanned pregnancies and abortions might have increased during the pandemic, lack of robust data in India, which is understandable during the pandemic, limits our ability to study and understand these phenomena. The limitations of availability of primary data can be somewhat addressed by using publicly available data sources such Google Trends to analyze patterns in health behaviors based on Google searches in a particular geography, over a specific period. Given the popularity and hegemony of Google as the primary search engine in India 10 and its rapidly increasing share of internet users in both rural and urban areas,11 Google searches can be used to gain a reasonable understanding of these patterns.

In this brief, we analyze trends of searches related to unplanned pregnancies and abortions in India between January 01, 2019 and December 31, 2020. We study the patterns in these search terms and try to assess if there was a meaningful increase in search terms related to unplanned pregnancies and abortions in India during the COVID-19 pandemic. Findings from this analysis can help us understand how the COVID-19 pandemic affected reproductive health behaviors in India and can inform programs and policy on the need to strengthen Sexual and Reproductive Health  services during the ongoing COVID-19 crisis.

Our Approach
Selecting search terms: The first step in identifying search terms for the analysis involved creating a list of ‘base terms’ related to unplanned pregnancies and abortions in India. 12 Two researchers, with contextual understanding of SRH in India, brainstormed related search terms that might be used to search for unplanned pregnancies and abortions. The researchers initially identified 17 search terms and narrowed it down to 13 base terms by removing terms that were not specifically associated unplanned pregnancies or abortions. The 13 base terms identified are as follows:

  • abortion
  • abortion medicine
  • abortion pill
  • abortion tablet
  • emergency contraceptive pills
  • i-pill
  • medical termination of pregnancy
  • morning after pill
  • pregnancy termination
  • unintended pregnancy
  • unplanned pregnancy
  • unwanted pregnancy
  • unwanted 72

Subsequently, the researchers used Google’s autocomplete feature to identify nuanced search terms used by people related to these base-terms.12 In order to identify search terms specific to India, the researchers changed the ‘Region Setting’ in Google to India. 13 The researchers then switched to incognito mode to avoid their personal search patterns from affecting the autocomplete results.12,14 For each of the base-terms, the researchers first identified the search terms suggested by the autocomplete feature after the ‘base-term’ (Figure 1) and then also identified search terms suggested by the autocomplete feature before the ‘base-term by adding a * in front of the base-term as shown in Figure 2.

This process led to the identification of a total of 243 search terms, including the base-terms. Two researchers then independently identified search terms that were not related to the thematic areas of unplanned pregnancy or abortion e.g. one of the search terms suggested by the autocomplete feature for the base-term ‘unplanned pregnancies’ was ‘unplanned pregnancy romance novels’. The researchers identified 63 such search terms and excluded them from analysis. The next step involved identifying and removing duplicate search terms or terms that were very similar to each other. A total of 21 such terms were identified, and the remaining 159 terms were used for subsequent analysis. The next step involved identifying search terms that did not have enough data to produce results on Google Trends (https://trends.google.com/trends/?geo=IN) for our study period of January 01, 2019 to December 31, 2020 with the region specified as India. An example of such a term excluded from the analysis is shown in Figure 3.

Removal of search terms that yielded no results on Google Trends excluded 88 terms and 71 search terms remained for further analysis. The researchers then categorized these terms into the following four thematic categories: 12

  • Theme-1: Emergency Contraceptive Pills (27 terms)
  • Theme-2: Unplanned Pregnancy (7 terms)
  • Theme-3: Abortion Procedure (16 terms)
  • Theme-4: Abortion pills (21 terms)

The researchers then created a single ‘search-string’ for each of the categories by combining the search terms in a way that captured the thematic category sufficiently and succinctly, while ensuring that the combined search string the terms yielded results on Google Trends for the study period and geography. The following search-strings were finally used for the analysis for each of the thematic categories:

  • Theme-1: Emergency Contraceptive Pills – “emergency contraceptive pills” + “morning after pill” + “unwanted 72” + “i pill”
  • Theme-2: Unplanned Pregnancy – “unwanted pregnancy” + “unintended pregnancy” + “unplanned pregnancy”
  • Theme-4: Abortion Pills – “abortion tablet” + “abortion pill price” + “abortion pill” + “abortion pill side effects”
  • Theme-3: Abortion Procedure – “abortion” + “types of abortion” + “abortion process” + “surgical abortion”

Data Analysis: We used the search-strings described above to download ‘Interest over time’ data from Google Trends, for our study period and geography for each of the five thematic categories. Google Trends generated one data point on interest over time for each week in our study period. To identify periods where there was higher or lower interest in our thematic categories, we used changepoint analysis. 12 Changepoint analysis is a statistical tool that helps in detecting meaningful changes in time-series data by identifying periods that are qualitatively different from their neighboring points. 12,15 We used the ‘changepoint’ package 16 in R (version 4.0.3) and used the pruned exact linear time (PELT) algorithm as the method for detecting changepoints. This allowed us to break our study period into discrete regions where the interest over time for a specific thematic category was statistically different from their neighboring periods. Finally, we plotted graphs for each of the thematic categories with the changepoint periods laid atop the trend for interest over time data obtained from Google Trends.

Results: Our analysis reveals clear patterns of increased interest over time around all four thematic categories during the onset of the COVID-19 pandemic. Search trends for terms related to Emergency Contraceptive pills indicate an increased volume of searches around emergency contraceptive pills between December 2019 and March 2020 (Figure 5). The search volume then decreased till June 2020 and then seem to be increasing gradually again since that time. Search patterns related to unplanned pregnancies also show a similar trend, but they peaked a little late, relative to the emergency contraceptive pills, between March and April 2020 (Figure 6).

Search terms related to abortion pills and abortion procedure had peak periods between February to May 2020 and January to May 2020 (Figures 6 and 7) respectively. The peak periods for these searches were wider as compared to those for emergency contraceptive pills and unplanned pregnancies.

Interestingly, the search volume for emergency contraceptive pills, unplanned pregnancy, and abortion pill (figures 4, 5 and 6) decreased after their peak periods for a few months and seem to be increasing in terms of interest over time towards the end of 2020. The trend for searches around abortion procedure, however, does not follow this pattern and seems to be stable towards the end of the year.

The sharp increase in search terms around these topics can be considered as a reflection of the SRH needs of women in India during the COVID-19 pandemic. These needs were at their peak during the onset of the pandemic in 2020 and seemed to decline from May 2020 onwards.  However, recent trends around search terms related to emergency contraceptive pills, unplanned pregnancies, and abortion pills seem to suggest that these needs might be on an incline again. While our analyses of these search terms are some reflection of SRH needs, they do not tell us  about how these needs were addressed by the health system, especially during the country wide lock-downs that were in effect during the same months where we see a peak in these trends. 17 This calls for further research into the access and utilization of SRH services in India during the COVID-19 pandemic, especially given the premonition of an increase in interest around these topics as indicated in this analysis. Having a deeper understanding of the SRH needs of women and the challenges in ensuring access and utilization of SRH services can help the system to be better prepared to provide timely and effective services and can contribute significantly in improving the health outcomes of women in the country.

This work was supported by a grant to UC San Diego from the Bill and Melinda Gates Foundation (INV-018007; PI: Raj)


  1. WHO. Impact of COVID-19 on people’s livelihoods, their health and our food systems. 2020. https://www.who.int/news/item/13-10-2020-impact-of-covid-19-on-people’s-livelihoods-their-health-and-our-food-systems.
  2. UNDP. Socio-economic impact of COVID-19. 2020. https://www.undp.org/content/undp/en/home/coronavirus/socio-economic-impact-of-covid-19.html.
  3. Cullen W, Gulati G, Kelly B. Mental health in the Covid-19 pandemic. QJM: An International Journal of Medicine 2020; 113(5): 311-2.
  4. Woods J, Hutchinson NT, Powers SK, et al. The COVID-19 pandemic and physical activity. Elsevier; 2020.
  5. Yuksel B, Ozgor F, Obstetrics. Effect of the COVID‐19 pandemic on female sexual behavior. International Journal of Gynecology 2020.
  6. Caruso S, Rapisarda AMC, Minona P, Care RH. Sexual activity and contraceptive use during social distancing and self-isolation in the COVID-19 pandemic. The European Journal of Contraception 2020; 25(6): 445-8.
  7. International Institute for Population Sciences (IIPS) and MoHFW. National Family Health Survey (NFHS-4), 2015-16.
  8. Dixit P, Ram F, Dwivedi LK, childbirth. Determinants of unwanted pregnancies in India using matched case-control designs. BMC pregnancy 2012; 12(1): 84.
  9. Sebastian MP, Khan M, Sebastian D. Unintended pregnancy and abortion in India: country profile report. 2014.
  10. Statista. Share of desktop search traffic originating from Google in selected countries as of October 2020. 2020. https://www.statista.com/statistics/220534/googles-share-of-search-market-in-selected-countries/.
  11. Mishra D, Chanchani M. For the first time, India has more rural net users than urban 2020. https://timesofindia.indiatimes.com/business/india-business/for-the-first-time-india-has-more-rural-net-users-than-urban/articleshow/75566025.cms.
  12. Matsa KE, Mitchell A, Stocking G. Methodology. 2017. https://www.journalism.org/2017/04/27/google-flint-methodology/.
  13. wikiHow. How to Change Your Location in Google Chrome. 2018. https://www.wikihow.com/Change-Your-Location-in-Google-Chrome.
  14. Google Chrome Help. Browse in private. https://support.google.com/chrome/answer/95464?co=GENIE.Platform%3DDesktop&hl=en.
  15. Taylor WA. Change-point analysis: a powerful new tool for detecting changes. 2000.
  16. Killick R, Eckley I. changepoint: An R package for changepoint analysis. Journal of statistical software 2014; 58(3): 1-19.
  17. Wikipedia. COVID-19 lockdown in India. 2020. https://en.wikipedia.org/wiki/COVID-19_lockdown_in_India.