Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Pregnancy trends and associated factors among Kenyan adolescent girls and young women pre- and post-COVID-19 lockdown

Pregnancy trends and associated factors among Kenyan adolescent girls and young women pre- and... Congo O, et al. 2022. Pregnancy trends and associated factors among Kenyan adolescent girls and young women pre- and post- COVID-19 lockdown. Adv Glob Health, 1: 1. DOI: https://doi.org/ 10.1525/agh.2022.1811306 RESEARCH ARTICLE Pregnancy trends and associated factors among Kenyan adolescent girls and young women pre- and post-COVID-19 lockdown 1, 1 1 2 Ouma Congo *, George Otieno , Imeldah Wakhungu , Elizabeth K. Harrington , 3 4 5 4,5 Syovata Kimanthi , Charlene Biwott , Hannah Leingang , Nelly Mugo , 6 1,5 1 Ruanne V. Barnabas , Elizabeth A. Bukusi , and Maricianah Onono Globally, COVID-19 has had a negative impact on health systems and health outcomes, with evidence of differential gender impacts emerging. The COVID-19 timeline of events spanning from closures and restrictions to phased reopenings is well-documented in Kenya. This unique COVID-19 situation offered us the opportunity to study a natural experiment on pregnancy trends and outcomes in a cohort of Kenyan adolescent girls and young women (AGYW), enrolled in the KENya Single-dose HPV-vaccine Efficacy (KEN SHE) Study. The KEN SHE Study enrolled sexually active AGYW aged 15–20 years from central and western Kenya. Pregnancy testing was performed at enrollment and every 3 months. We determined pregnancy incidence trends pre- and post-COVID-19 lockdown, pregnancy outcomes (delivery, spontaneous, or induced abortion), and postabortion and postpartum contraceptive uptake. Kaplan–Meier survival estimates of incidence rates were used to estimate the cumulative probability of pregnancy during the study period. Cox regression was used to investigate factors associated with pregnancy incidence. Of the 2,223 AGYW included in the analysis, median age was 18.6 IQR (17.6–20.3), >90% had at least secondary school education, 95% were single at the time of enrollment, and 82% had a steady/primary sexual partner. Pregnancy incidence peaked at 2.27 (95% CI [1.84, 2.81])/100 women-years of observation at the end of the first quarter of 2020, a period coinciding with the government-imposed lockdown. AGYW had 60% increased risk of being pregnant during the lockdown when compared to prelockdown period (HR ¼ 1.60, 95% CI [1.25, 2.05]). Among the 514 pregnancies reported, 127 (25%) ended in abortion, of which 66 (52%) were induced abortions. Our findings demonstrate the adverse sexual and reproductive health (SRH) consequences of the COVID-19 pandemic and the lockdown measures among AGYW. As services continue to be disrupted by the pandemic, there is an urgent need to strengthen and prioritize AGYW-centered SRH services, including contraception and safe abortion. Keywords: Adolescent, Pregnancy, COVID-19, Lockdown, AGYW Introduction increasing evidence of differential gender implications The COVID-19 pandemic has had a negative impact on [1, 2]. Government and policy organization projections health systems and health outcomes globally, with and media reports suggest COVID-19-related disruptions to health systems and restrictions in the movement are leading to worsening health outcomes for women and Center for Microbiology Research, Kenya Medical Research Institute, Kisumu, Kenya girls globally. Reports indicate increased violence against Department of Obstetrics and Gynaecology, University of women, increased maternal morbidity and mortality, and Washington, Seattle, WA, USA a lack of access to essential sexual and reproductive health Centre for Clinical Research, Partners in Health and Research (SRH) services [3]. In resource-limited settings, the combi- Development, Kenya Medical Research Institute, Thika, Kenya nation of school closures and interruptions in access to Centre for Clinical Research, Kenya Medical Research SRH services has uniquely impacted adolescent girls and Institute, Nairobi, Kenya 5 young women (AGYW) [4]. Department of Global Health, University of Washington, Pregnancy among Kenyan adolescent girls is a complex Seattle, WA, USA and “multidimensional” social and public health issue [5]. Division of Infectious Disease, Massachusetts General Hospital, Boston, MA, USA Data suggest approximately 1 in every 5 girls between 15 and 19 years is either pregnant or already a mother [6]; this * Corresponding author: Email: dismasplanetcongo@gmail.com age-group accounts for 14% of all births in Kenya [7]. Art. 1(1) page 2 of 9 Congo et al: Pregnancy trends and associated factors among Kenyan adolescent girls and young women Adolescent pregnancy is a key risk factor for school drop- components or latex, hysterectomy, or history of immu- out, which is associated with lifelong social, economic, and nosuppressive conditions [17]. health consequences, including higher maternal and infant KEN SHE Study locations health risks and HIV acquisition [8]. Furthermore, over 60% The study was conducted at 3 Kenya Medical Research of adolescent pregnancies are estimated to be unintended, Institute (KEMRI) clinical sites in Thika, Nairobi, and of which 35% are estimated to end in abortion. Research Kisumu. All participants, and their parents/guardians in the suggests pregnant adolescents may face social stigma, iso- case of minors, provided informed consent/assent, which lation, worsening poverty, and associated poor mental included counseling about randomization, risks and bene- health outcomes; they are also at risk of complications of fits of participation, study procedures, and their rights as unsafe abortion practices in settings with restricted abor- research participants. Thika is a rapidly growing industrial tion access, such as in much of sub-Saharan Africa [9, 10]. city of 280,000 people, in Kiambu County, lying 40 km Media reports at the height of COVID-19 pandemic northeast of Nairobi. The Nairobi study site is located near called attention to a spike in the number of teenage preg- to the Kibera slum area; Nairobi is a multiethnic city with nancies in Kenya and attributed them to government- a population over 4 million. Kisumu is the third-largest city mandated COVID-19 containment measures such as school in Kenya with a population of 610,000 people and is situ- closures [11]. Despite significant concern about rising teen ated on the shores of Lake Victoria in Western Kenya. pregnancies during the COVID-19 pandemic, there are few sources of high-quality data [12, 13]. From January to June KEN SHE Study procedures 2020, Nairobi County recorded a higher number of adoles- Participants in this study are offered SRH services (contra- cent pregnancies compared to the prior year, though the ception, sexually transmitted infection diagnosis and analysis was limited to those presenting to government treatment, and HIV pre-exposure prophylaxis) at enroll- health centers [14]. A recent analysis from western Kenya ment and every visit. Questionnaires are conducted using concluded that compared with historical controls, female electronic case report forms (eCRFs; DF/Net Research, secondary school students who were out of school for at Inc.© , Seattle, WA, USA). least 6 months had twice the pregnancy risk and were 3 Participants are followed up at month 1, and thereafter, times more likely to drop out of school [15]. every 3 months, during which they have comprehensive In this study, we evaluated incident pregnancy trends assessment including genital swab collection. Providers and outcomes in a large cohort of Kenyan AGYW, pre- and administer clinical questionnaires at each visit. Pregnancy post-government lockdown during the COVID-19 pan- testing was done at enrollment, every 3 months, and when demic. We aim to contribute to the body of literature clinically indicated. Pregnant participants were followed- demonstrating the gendered impacts of the COVID-19 up and monitored to the end of pregnancy when the pandemic, which has further limited progress toward real- outcome was established. All these data were available for izing the Sustainable Development Goal 5 on gender the analysis presented in this article. equality [16]. COVID-19 lockdown and pregnancy estimation Methods Kenya reported its first case of confirmed COVID-19 on Study description and population March 12, 2020, and the government closed all learning The COVID-19 timeline of events spanning from closures institutions from March 13, 2020, to as one of the mea- and restrictions to phased reopenings is well-documented sures to curb viral spread. An estimated date of fertiliza- in Kenya. This unique COVID-19 situation offered us the tion (EDF) was computed as the first day of last menstrual opportunity to study a natural experiment on pregnancy period (LMP) plus 14 days to determine whether the preg- trends and outcomes in a cohort of Kenyan AGYW, nancy occurred before or after initiation of lockdown mea- enrolled in the KENya Single-dose HPV-vaccine Efficacy sures (pre- or post-COVID lockdown). Participants whose (KEN SHE) Study. EDF fell after March 13, 2020, were considered to have The KEN SHE Study is an individual randomized con- a pregnancy that occurred post-COVID lockdown; this is trolled trial to evaluate whether a single dose of the biva- equivalent to LMP on or after March 1, 2020. See the lent or nonavalent HPV vaccine prevents persistent HPV definition of independent variables in the Supplementary infection, a surrogate marker for precancerous lesions, Tables 1 and 2. and cervical cancer [17]. Briefly, between December 2018 and November 2019, the KEN SHE Study enrolled Measures and statistical analysis girls and young women aged 15–20 years in a double- Person-time started on December 20, 2018, and observa- blind randomized clinical trial testing the efficacy of a sin- tions were censored on March 31, 2021, or the last gle-dose bivalent and nonvalent HPV vaccination as time the person participated in the study. Person-time a catch-up strategy for cervical cancer prevention. Girls ended at date of pregnancy, lost to follow-up, or end of and young women from Thika, Nairobi, and Kisumu were the study period (March 31, 2021), whichever occurred eligible to participate if they were HIV-negative, reported first. If individuals came back to the study, they reentered 1 to 5 sexual partners in their lifetime, were not pregnant, the cohort (open cohort). Cumulative probabilities of and had a uterus and cervix. Study exclusion criteria pregnancy were estimated using Kaplan–Meier (K–M) included history of HPV vaccination, allergies to vaccine method. Pregnancy trends were determined among this Congo et al: Pregnancy trends and associated factors among Kenyan adolescent girls and young women Art. 1(1) page 3 of 9 cohort, pre- and post-COVID-19 lockdown (March 13, Table 1. Baseline sociodemographic characteristics of 2020). Pregnancy incidence rates were computed using the study population number of pregnancies as the numerator and person- Variables N % time in follow-up as the denominator. Pregnancy rates per 100 person-years observation were determined by pre-/ Study area post-COVID-19 lockdown, study site, and quarterly. Kisumu 1,168 52.5 Supplementary Figure 1 shows goodness-of-fit and model diagnostics. Thika 768 34.5 We assessed (1) pregnancy outcome (term delivery Nairobi 287 12.9 defined as delivery 37 weeks of gestational age, sponta- Do you have a main or steady sex partner? neous or induced abortion, ongoing or undetermined) and (2) postabortion and postpartum contraceptive No 402 18.4 uptake. We used a Cox proportional hazards model to Yes 1,781 81.6 investigate factors associated with pregnancy incidence Age-group in years and binary logistic regression to determine factors associ- ated with postabortion contraceptive initiation during the <18 years 998 44.9 pandemic period. 18 years 1,225 55.1 Variables significant at <0.2 level in the univariable Education level analysis or set a priori underwent further examination using a multivariable regression. Potential confounding None/Primary incomplete 162 7.3 effect of each covariate and a 2-way interaction were Secondary 1,614 72.6 addressed, and we obtained final variable selection by Tertiary/University 447 20.1 applying backward elimination method. Analysis and comparisons were done at 5% level of significance. The Marital status analyses were completed using STATA version 16.1 (STATA Single 2,102 94.7 Corporation, College Station, TX, USA). Married 96 4.3 Results Divorced/separated/widowed 22 1 Our analysis was restricted to 2,223 participants with com- Number of sexual partners plete information on pregnancy. Overall, median age of One 1,567 95.1 the participants was 18.6 years, IQR (17.6–20.3); girls and young women aged 18 years were slightly more than Two or more 80 4.9 a half (55%). Majority of the participants were single Condom use last sex (95%) at the time of enrollment, with majority of them No 876 53.2 reporting secondary school level as the highest at the time of enrollment (73%). Approximately 81.6% of girls Yes 771 46.8 reported a main or steady sexual partner, while approxi- Earned income in the last 12 months mately 3-quarters of the participants (75%) reported own- No 1,928 86.7 ing a mobile phone. Only 13% of the girls and young women earned their own income in the last 12 months. Yes 295 13.3 A total of 519 incident pregnancies were recorded over Own a phone 2 years of follow-up. Baseline characteristics by pregnancy No 418 24.6 status are presented in Table 1. The overall pregnancy incidence rate was 21 pregnancies per 100 woman-years Yes 1,283 75.4 of observation (95% CI [20, 23]). Nairobi and Thika had Do you smoke? a higher risk of pregnancy in comparison to Kisumu as No 2,212 99.5 shown by the K–M curves in Figure 1. There was a steady increase in pregnancy incidence rates toward the end of Yes 11 0.5 the last quarter of 2019, this trend peaked at the end Take alcohol of first quarter of 2020, corresponding to declaration of No 2,102 94.6 COVID-19 lockdown measures in Kenya. A steady, though not statistically significant, decline in pregnancy rate is Yes 121 5.4 shown between the third quarter of 2020 and first quarter N ¼ 2,223. of 2021. Pregnancy rates in Kisumu were significantly lower post-COVID-19 lockdown (Figures 2 and 3), whereas in Thika, there was a significant increase in rates post COVID-19 lockdown. Nairobi showed a slight reduc- abortion (6%). In Thika, 60% had term live birth and tion, which was not statistically significant. 51% were spontaneous abortions (Figure 4). Nairobi had Kisumu cohort pregnancy outcomes were (67%) term 60% term live births, 21% elective abortions, and 11% live birth, induced abortion (24%), and spontaneous spontaneous abortions. Art. 1(1) page 4 of 9 Congo et al: Pregnancy trends and associated factors among Kenyan adolescent girls and young women Kaplan-Meier pregnancy incidence survival estimates by the study sites Log-rank test p < 0.001 0 5 10 15 20 25 Analysis time Study site = Kisumu Study site = Thika Study site = Nairobi Figure 1. Kaplan–Meier curves for the cumulative risk of pregnancy by the study site. O V ERALL PRE G NANCY RATE PER 100 P YOS Test for trend overme p -value <0.001 2019Q1 2019Q2 2019Q3 2019Q4 2 020Q1 2 020Q2 2020Q3 2 020Q4 2 021Q1 Figure 2. Pregnancy rate for all sites combined. In adjusted analysis (Table 2), pre-/post-COVID-19 secondary school were nearly 3 times more likely to get lockdown, study site, age, education level, and condom pregnant during the COVID-19 lockdown period com- use independently predicted pregnancy incidence among pared to those in primary school or those with no educa- girls. In the post-COVID-19 lockdown period, girls and tion (HR ¼ 2.49, 95% CI [1.37, 4.53], P < 0.003), while young women from Nairobi and Thika had higher risk of those in postsecondary education were twice as likely to being pregnant, compared with pre-COVID-19 lockdown get pregnant compared to those with primary or no edu- cation (HR ¼ 2.06, 95% CI [1.11, 3.81], P < 0.021). (Uni- (HR ¼ 2.29, 95% CI [1.17, 4.48], P ¼ 0.016 and HR ¼ 3.10, variate analysis of factors associated with pregnancy 95% CI [1.76, 5.62], P ¼ 0.016, respectively). Condom use incidence are contained in Supplementary Table 3.) at last sex was associated with 71% reduced hazard of pregnancy (HR ¼ 0.29, 95% CI [0.18, 0.46], P < 0.001); however, participants aged 18 years who reported hav- Discussion ing used a condom during last sexual intercourse were We demonstrated that the Government of Kenya- more likely to be pregnant compared to minors (HR ¼ mandated COVID-19 lockdown, which included move- 2.00, 95% CI [1.14, 3.50], P < 0.015). Participants in ment restriction and closure of schools starting in March RATE PER 100 PYOS 0.00 0.25 0.50 0.75 1.00 Congo et al: Pregnancy trends and associated factors among Kenyan adolescent girls and young women Art. 1(1) page 5 of 9 Pregnancy rate pre and post COVID-19 lockdown by the study site 60.00 37.52 50.00 35.30 30.85 40.00 30.00 20.95 17.76 15.66 20.00 10.00 0.00 Pre covid-19 lockdown Post covid-19 lockdown Kisumu Thika Nairobi Figure 3. Overall pregnancy incidence rate pre- and post-COVID-19 lockdown. Pregnancy outcome by the study site (N= 514) Other/unknown 0% 1% 0% Spontaneous fetal death and/or stillbirth > 20 weeks 2% 3% 3% Therapeuc/elecve aboron 24% 6% 21% Spontaneous aboron (second trimester and < 20 weeks) 0% 4% 1% Spontaneous aboron (first trimester) 6% 25% 11% Ectopic pregnancy 0% 0% 1% Preterm live birth (<37 weeks) 0% 1% 4% Full-term live birth 37 weeks 67% 60% 60% Kisumu Thika Nairobi Figure 4. Overall pregnancy outcomes. 2020, was temporally associated with a rise in the hazard treatment for sexually transmitted infections, and obstet- of pregnancy among AGYW in central and western Kenya. ric care [20–22]. These barriers to access and quality care These findings correlate with the spike in teenage preg- expose adolescents to negative health and life-altering nancy that was publicized by media and programmatic socioeconomic consequences: consequences that are even organizations in the wake of the COVID-19 lockdown more pronounced in times of crisis such as disease out- [11]. The pregnancy incidence rate differed by site, which breaks or war, when data suggest they are disproportion- we hypothesize is related to regional secular adolescent ately affected. Furthermore, prior studies suggest that pregnancy trends as well as geographically specific school closures, lack of comprehensive sexuality educa- COVID-19 movement restrictions, as Thika and Nairobi tion, and increased risk of violence against girls and young were put under more stringent lockdown measures. This women put them at higher risk of unintended pregnancy. is consistent with experience from previous public health A recently published study from rural western Kenya com- crises, such as recent Ebola outbreak in Liberia and Sierra pared school drop-out and pregnancy rates pre- and post- Leone, which demonstrated reduced access to essential COVID-19-related school disruptions and found a 3-fold SRH and other health services [18, 19]. increase in school drop-outs and a 2-fold increase in preg- Our study demonstrates a rise in pregnancy incidence nancy among secondary school students whose school was during the COVID-19 lockdown period when compared to disrupted, which is similar to our study findings [15]. the pre-COVID-19 period. AGYW constitutes a special The study population had 1 abortion for every 4 preg- population who already face social, developmental, and nancies, with 19% being induced abortions. The conse- health systems barriers to accessing quality SRH care, quences of an unwanted pregnancy may be devastating including contraception, abortion, screening and for many adolescents, which may lead them to seek Pregnancy rate per 100 PYOs Art. 1(1) page 6 of 9 Congo et al: Pregnancy trends and associated factors among Kenyan adolescent girls and young women Table 2. Factors associated with pregnancy incidence among girls aged 15-20 years Unadjusted HR Adjusted HR Factor HR [95% CI] P Value HR [95% CI] P Value COVID-19 lockdown Pre-COVID-19 lockdown Ref Post-COVID-19 lockdown 1.32 [1.10, 1.57] 0.003 0.72 [0.46, 1.12] 0.147 Interaction (pre–post#Site) Post Covid#Thika 2.82 [1.59, 4.99] <0.001 Post Covid#Nairobi 2.09 [1.07, 4.07] 0.029 Study area Kisumu Ref Thika 1.46 [1.21, 1.78] <0.001 0.84 [0.59, 1.21] 0.366 Nairobi 2.4 [1.90, 3.03] <0.001 1.56 [0.95, 2.54] 0.074 Age-group in years 15–17 Ref 18–20 1.26 [1.05, 1.50] 0.011 0.76 [0.58, 1.09] 0.157 Marital status Single Ref Married 0.83 [0.54, 1.27] 0.392 Divorced/separated/widowed 1.05 [0.52, 2.10] 0.894 Highest education level None/primary incomplete Ref Secondary 1.43 [1.00, 2.05] 0.052 2.45 [1.35, 4.45] 0.003 Tertiary/university 1.68 [1.14, 2.48] 0.009 2.04 [1.11, 3.77] 0.023 Number of sexual partners One Ref Two or more 0.85 [0.52, 1.37] 0.500 Do you have a main or steady sex partner? No Ref Yes 1.08 [0.84, 1.38] 0.549 Own a phone No Ref Yes 1.38 [1.07, 1.76] 0.011 1.16 [0.82, 1.63] 0.382 Condom use last sex No Ref Yes 0.41 [0.33, 0.52] <0.001 0.28 [0.18, 0.45] <0.001 Interaction (condom use last sex# age-group) >18 years#Yes 2.06 [1.18, 3.59] 0.010 Do you smoke? No Ref Yes 0.48 [0.07, 3.47] 0.467 Take alcohol No Ref Yes 1.04 [0.72, 1.51] 0.842 Congo et al: Pregnancy trends and associated factors among Kenyan adolescent girls and young women Art. 1(1) page 7 of 9 abortion in unsafe circumstances, particularly in countries Table 3. Univariate analysis of factors associated with where abortion is highly restricted [23, 24]. Based on pregnancy incidence. extensive clinical experience in the region and confirmed Figure 1. Goodness-of-fit and model diagnostics. reports of induced abortion in this cohort, our team sus- pects a far higher proportion of abortions were induced, Acknowledgments often without a skilled provider or using practices known We thank the young women who participated in this study to be harmful [25]. Our data underscore the known high for their motivation and dedication and the communities unmet need for contraception among girls and young that supported this work. women in sub-Saharan Africa [26] and the need to protect these services in times of crisis. Funding Our study has a number of strengths, including its The KEN SHE Study was funded by the Bill & Melinda innovative leverage of a natural experiment and use of Gates Foundation (OPP1188693) and the University of a multisite design representing urban and rural settings Washington King K. Holmes Endowed Professorship in in Kenya with very low loss to follow-up. We had labora- STDs and AIDS. The content is solely the responsibility tory ascertainment of pregnancy status with routine preg- of the authors and does not necessarily represent the nancy tests done quarterly. While the present study had views, decisions, or policies of the institutions with which a number of advantages because of the availability of they are affiliated or the KEN SHE Study funders. The a well-studied cohort, this feature was also a limitation funders had no separate role in the study design; data of the study because the girls and young women in the collection, analysis, and interpretation; writing of the cohort had continued, though disrupted, access to SRH report; or in the decision to submit for publication. care including contraceptive counseling and provision of The corresponding author had full access to all the data STI treatment throughout the COVID-19 lockdown due to in the study and had final responsibility for the decision to study participation. It is likely, therefore, that age-matched submit for publication. girls and young women who were not members of the cohort had a higher pregnancy risk than we report, and Competing interests thus, the pregnancy incidence rates may not be general- The authors have no competing interests to declare. izable to nonstudy participants’ risk. We did not collect repeated measures of pregnancy intentions and the con- Ethics approval and consent to participate sistency of condom use, which limit our understanding of The KEMRI Scientific and Ethics Review Unit (SERU), in these factors as influences on pregnancy risk. Finally, our Kenya, and the University of Washington Institutional study design cannot completely account for age as a time- Review Board, Seattle, WA, approved this study. The study varying confounder of pregnancy incidence. was registered at ClinicalTrials.gov, number NCT03675256. In conclusion, our findings contribute to the evidence All participants provided written informed consent, empirically demonstrating the social and health conse- which included counseling about randomization, the vac- quences of the COVID-19 pandemic, which widen cine used in each study group, strategies for cervical can- already-existing SRH inequities confronting AGYW. We cer prevention, and their rights as research participants. demonstrate that despite frequent contact and compre- hensive contraceptive care offered through the study References clinic, girls and young women in this cohort had a similar 1. Pulse survey on continuity of essential health services pregnancy rate to the general aged-matched population in during the COVID-19 pandemic: interim report, 27 Kenya, many of whom did not have adequate access to August 2020 [Internet]. 2020 Aug 27 [cited 2021 SRH services during COVID-19 containment measures. Oct 18]. Available from: https://www.who.int/ Given the many barriers to contraceptive use among this publications/i/item/WHO-2019-nCoV-EHS_continuity- age-group, ongoing investigation into their unique contra- survey-2020.1. ceptive needs and fertility desires, power in sexual rela- 2. Mooi-Reci I, Risman BJ. The gendered impacts of tionships, and agency in reproductive decision-making is COVID-19: lessons and reflections. Gend Soc [Inter- critical to improving adolescent SRH outcomes. net]. 2021 [cited 2022 Jan 23];35(2):161-7. Available from: https://journals.sagepub.com/doi/10.1177/ Data accessibility statement 3. How COVID-19 has affected global progress [Internet]. The main study that provided data for this paper is ongo- 2020 Sep [cited 2022 Jan 30]. Available from: https:// ing; data can be accessed by writing to RVB. www.gatesfoundation.org/goalkeepers/report/2020- report/progress-indicators/maternal-mortality/. Supplemental files 4. de Paz C, Muller M, Munoz Boudet AM, Gaddis I. The supplemental files for this article can be found as Gender dimensions of the COVID-19 pandemic [Inter- follows: net]. 2020 Apr 16 [cited 2022 Jan 27]. Available from: Table 1. Independent variables (demographic https://openknowledge.worldbank.org/handle/ characteristics). 10986/33622. Table 2. Independent variables (sexual behavior 5. Onono Id MA, Rutherford GW, Bukusi EA, Whiteid JS, practices). Goosby E, Brindis CD. Political prioritization and the Art. 1(1) page 8 of 9 Congo et al: Pregnancy trends and associated factors among Kenyan adolescent girls and young women competing definitions of adolescent pregnancy in 16. United Nations: gender equality and women’s empow- Kenya: an application of the Public Arenas Model. erment [Internet]. 2022 [cited 2022 Mar 3]. Available PLoS One. 2020 [cited 2022 Mar 8]. Available from: from: https://www.un.org/sustainabledevelopment/ https://doi.org/10.1371/journal.pone.0238136. gender-equality/. 17. Barnabas RV, Brown ER, Onono M, Bukusi EA, Njoroge 6. National Bureau of Statistics Nairobi K. Republic of B, Winer RL, et al. Single-dose HPV vaccination efficacy Kenya—Kenya Demographic and Health Survey 2014. among adolescent girls and young women in Kenya 2015 [cited 2022 Jan 28]. Available from: www. (the KEN SHE Study): study protocol for a randomized DHSprogram.com. controlled trial. Trials [Internet]. 2021 [cited 2022 Mar 7. Darroch J, Audam S, Biddlecom A, Kopplin G, Riley T, 7];22(1):1-19. Available from: https://trialsjournal. Singh S. Adding it up: investing in contraception and biomedcentral.com/articles/10.1186/s13063-021- maternal and newborn health. New York (NY): The 05608-8. Guttmacher Institute; 2017. 18. The socio-economic impacts of Ebola in Liberia: 8. Christofides NJ, Jewkes RK, Dunkle KL, Nduna M, Shai results from a high frequency cell phone survey, NJ, Sterk C. Early adolescent pregnancy increases risk Round 5 [Internet]. [cited 2022 Mar 7]. Available of incident HIV infection in the Eastern Cape, South from: https://openknowledge.worldbank.org/ Africa: a longitudinal study. J Int AIDS Soc [Internet]. handle/10986/21893. 2014 [cited 2022 Mar 8];17(1). Available from: 19. Elston JWT, Cartwright C, Ndumbi P, Wright J. The https://www.ncbi.nlm.nih.gov/pmc/articles/PMC39 health impact of the 2014–15 Ebola outbreak. Public 62027/. Health. 2017;143:60-70. 9. Osok J, Kigamwa P, Stoep AV, Huang K-Y, Kumar M. 20. Durowade KA, Omokanye LO, Elegbede OE, Adeto- Depression and its psychosocial risk factors in preg- kunbo S, Olomofe CO, Ajiboye AD, et al. Barriers to nant Kenyan adolescents: a cross-sectional study in contraceptive uptake among women of reproductive a community health centre of Nairobi. BMC Psychiatry age in a semi-urban community of Ekiti State, South- [Internet]. 2018 [cited 2022 Mar 8];18(1):136. Available west Nigeria. Ethiop J Health Sci [Internet]. 2017 [cited from: https://bmcpsychiatry.biomedcentral.com/ 2022 Mar 4];27(2):121-8. Available from: https://www. articles/10.1186/s12888-018-1706-y. ajol.info/index.php/ejhs/article/view/153144. 10. The Case of Korogocho Slums Final Report. Nairobi: 21. Ababor AA, Tesso DW, Cheme MC. Addressing the KHRC and RHRA; 2010. deprived: need and access of sexual reproductive 11. COVID-19: lockdown linked to high number of unin- health services to street adolescents in Ethiopia. The tended teen pregnancies in Kenya. Kenya | ReliefWeb case of Nekemte town: mixed methods study. BMC Res [Internet]. [cited 2022 Jan 30]. Available from: Notes [Internet]. 2019 [cited 2022 Mar 3];12(1):1-6. https://reliefweb.int/report/kenya/covid-19- Available from: https://bmcresnotes.biomedcentral. lockdown-linked-high-number-unintended-teen- com/articles/10.1186/s13104-019-4850-7. pregnancies-kenya. 22. Machiyama K, Casterline JB, Mumah JN, Huda FA, 12. Covid-19 pandemic: 13,000 girls got pregnant in Obare F, Odwe G, et al. Reasons for unmet need for Narok in 2021. The Standard [Internet]. [cited 2022 family planning, with attention to the measurement Mar 8]. Available from: https://www.standardmedia. of fertility preferences: protocol for a multi-site cohort co.ke/the-standard-insider/article/2001436813/ study. Reprod Health [Internet]. 2017 [cited 2022 Mar covid-19-pandemic-13000-girls-got-pregnant-in- 4];14(1):1-11. Available from: https://reproductive- narok-in-2021. health-journal.biomedcentral.com/articles/10.1186/ 13. Kilifi girls: Teen pregnancies in the region a pandemic. s12978-016-0268-z. Kenya [Internet]. [cited 2022 Mar 8]. Available from: 23. WHO/UNICEF. School closures and teenage preg- https://nation.africa/kenya/news/gender/kilifi-girls- nancy. Bull World Health Organ [Internet]. 2021 [cited teen-pregnancies-in-the-region-a-pandemic-3644438? 2022 Jan 30];99(1):6. Available from: https://www. view¼htmlamp. ncbi.nlm.nih.gov/pmc/articles/PMC7924898/. 14. Teen pregnancy in Kenya: verifying the data and the 24. Preventing Unintended Pregnancies in the Context of facts. African Institute for Development Policy—AFI- COVID-19. Wilson Center [Internet]. [cited 2022 Feb DEP [Internet]. 2020 [cited 2022 Jan 30]. Available 25]. Available from: https://www.wilsoncenter.org/ from: https://www.afidep.org/news-release-teen- event/preventing-unintended-pregnancies-context- pregnancy-in-kenya-verifying-the-data-and-the-facts/. covid-19. 15. Zulaika G, Bulbarelli M, Nyothach E, Van Eijk A, Mason 25. WHO. Preventing unsafeabortion. Geneva: WHO; L, Fwaya E, et al. Impact of COVID-19 lockdowns on adolescent pregnancy and school dropout among sec- 26. Dickson KS, Adde KS, Ahinkorah BO. Socio-economic ondary schoolgirls in Kenya. BMJ Glob Heal [Internet]. determinants of abortion among women in Mozam- 2022 [cited 2022 Mar 2];7(1):e007666. Available bique and Ghana: evidence from demographic and from: https://gh.bmj.com/content/7/1/e007666. health survey. Arch Public Heal. 2018;76(1):1-10. Congo et al: Pregnancy trends and associated factors among Kenyan adolescent girls and young women Art. 1(1) page 9 of 9 How to cite this article: Congo O, Otieno G, Wakhungu I, Harrington EK, Kimanthi S, Biwott C, et al. Pregnancy trends and associated factors among Kenyan adolescent girls and young women pre- and post-COVID-19 lockdown. Adv Glob Health. 2022;1(1). https://doi.org/10.1525/agh.2022.1811306 Editor-in-Chief: Craig R. Cohen, University of California, San Francisco, CA, USA Senior Editor: Sarah Ssali, Makerere University, Kampala, Uganda Section: Achieving Gender Equality Published: December 8, 2022 Accepted: October 29, 2022 Submitted: March 18, 2022 Copyright: © 2022 The Author(s). This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (CC-BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. See http://creativecommons.org/licenses/by/4.0/. Elem Sci Anth is a peer-reviewed open access journal published by University of California Press. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Advances in Global Health University of California Press

Pregnancy trends and associated factors among Kenyan adolescent girls and young women pre- and post-COVID-19 lockdown

Download PDF
9 pages

Loading next page...
 
/lp/university-of-california-press/pregnancy-trends-and-associated-factors-among-kenyan-adolescent-girls-piLY2yC6NP

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
University of California Press
Copyright
© 2022 The Author(s)
eISSN
2689-9558
DOI
10.1525/agh.2022.1811306
Publisher site
See Article on Publisher Site

Abstract

Congo O, et al. 2022. Pregnancy trends and associated factors among Kenyan adolescent girls and young women pre- and post- COVID-19 lockdown. Adv Glob Health, 1: 1. DOI: https://doi.org/ 10.1525/agh.2022.1811306 RESEARCH ARTICLE Pregnancy trends and associated factors among Kenyan adolescent girls and young women pre- and post-COVID-19 lockdown 1, 1 1 2 Ouma Congo *, George Otieno , Imeldah Wakhungu , Elizabeth K. Harrington , 3 4 5 4,5 Syovata Kimanthi , Charlene Biwott , Hannah Leingang , Nelly Mugo , 6 1,5 1 Ruanne V. Barnabas , Elizabeth A. Bukusi , and Maricianah Onono Globally, COVID-19 has had a negative impact on health systems and health outcomes, with evidence of differential gender impacts emerging. The COVID-19 timeline of events spanning from closures and restrictions to phased reopenings is well-documented in Kenya. This unique COVID-19 situation offered us the opportunity to study a natural experiment on pregnancy trends and outcomes in a cohort of Kenyan adolescent girls and young women (AGYW), enrolled in the KENya Single-dose HPV-vaccine Efficacy (KEN SHE) Study. The KEN SHE Study enrolled sexually active AGYW aged 15–20 years from central and western Kenya. Pregnancy testing was performed at enrollment and every 3 months. We determined pregnancy incidence trends pre- and post-COVID-19 lockdown, pregnancy outcomes (delivery, spontaneous, or induced abortion), and postabortion and postpartum contraceptive uptake. Kaplan–Meier survival estimates of incidence rates were used to estimate the cumulative probability of pregnancy during the study period. Cox regression was used to investigate factors associated with pregnancy incidence. Of the 2,223 AGYW included in the analysis, median age was 18.6 IQR (17.6–20.3), >90% had at least secondary school education, 95% were single at the time of enrollment, and 82% had a steady/primary sexual partner. Pregnancy incidence peaked at 2.27 (95% CI [1.84, 2.81])/100 women-years of observation at the end of the first quarter of 2020, a period coinciding with the government-imposed lockdown. AGYW had 60% increased risk of being pregnant during the lockdown when compared to prelockdown period (HR ¼ 1.60, 95% CI [1.25, 2.05]). Among the 514 pregnancies reported, 127 (25%) ended in abortion, of which 66 (52%) were induced abortions. Our findings demonstrate the adverse sexual and reproductive health (SRH) consequences of the COVID-19 pandemic and the lockdown measures among AGYW. As services continue to be disrupted by the pandemic, there is an urgent need to strengthen and prioritize AGYW-centered SRH services, including contraception and safe abortion. Keywords: Adolescent, Pregnancy, COVID-19, Lockdown, AGYW Introduction increasing evidence of differential gender implications The COVID-19 pandemic has had a negative impact on [1, 2]. Government and policy organization projections health systems and health outcomes globally, with and media reports suggest COVID-19-related disruptions to health systems and restrictions in the movement are leading to worsening health outcomes for women and Center for Microbiology Research, Kenya Medical Research Institute, Kisumu, Kenya girls globally. Reports indicate increased violence against Department of Obstetrics and Gynaecology, University of women, increased maternal morbidity and mortality, and Washington, Seattle, WA, USA a lack of access to essential sexual and reproductive health Centre for Clinical Research, Partners in Health and Research (SRH) services [3]. In resource-limited settings, the combi- Development, Kenya Medical Research Institute, Thika, Kenya nation of school closures and interruptions in access to Centre for Clinical Research, Kenya Medical Research SRH services has uniquely impacted adolescent girls and Institute, Nairobi, Kenya 5 young women (AGYW) [4]. Department of Global Health, University of Washington, Pregnancy among Kenyan adolescent girls is a complex Seattle, WA, USA and “multidimensional” social and public health issue [5]. Division of Infectious Disease, Massachusetts General Hospital, Boston, MA, USA Data suggest approximately 1 in every 5 girls between 15 and 19 years is either pregnant or already a mother [6]; this * Corresponding author: Email: dismasplanetcongo@gmail.com age-group accounts for 14% of all births in Kenya [7]. Art. 1(1) page 2 of 9 Congo et al: Pregnancy trends and associated factors among Kenyan adolescent girls and young women Adolescent pregnancy is a key risk factor for school drop- components or latex, hysterectomy, or history of immu- out, which is associated with lifelong social, economic, and nosuppressive conditions [17]. health consequences, including higher maternal and infant KEN SHE Study locations health risks and HIV acquisition [8]. Furthermore, over 60% The study was conducted at 3 Kenya Medical Research of adolescent pregnancies are estimated to be unintended, Institute (KEMRI) clinical sites in Thika, Nairobi, and of which 35% are estimated to end in abortion. Research Kisumu. All participants, and their parents/guardians in the suggests pregnant adolescents may face social stigma, iso- case of minors, provided informed consent/assent, which lation, worsening poverty, and associated poor mental included counseling about randomization, risks and bene- health outcomes; they are also at risk of complications of fits of participation, study procedures, and their rights as unsafe abortion practices in settings with restricted abor- research participants. Thika is a rapidly growing industrial tion access, such as in much of sub-Saharan Africa [9, 10]. city of 280,000 people, in Kiambu County, lying 40 km Media reports at the height of COVID-19 pandemic northeast of Nairobi. The Nairobi study site is located near called attention to a spike in the number of teenage preg- to the Kibera slum area; Nairobi is a multiethnic city with nancies in Kenya and attributed them to government- a population over 4 million. Kisumu is the third-largest city mandated COVID-19 containment measures such as school in Kenya with a population of 610,000 people and is situ- closures [11]. Despite significant concern about rising teen ated on the shores of Lake Victoria in Western Kenya. pregnancies during the COVID-19 pandemic, there are few sources of high-quality data [12, 13]. From January to June KEN SHE Study procedures 2020, Nairobi County recorded a higher number of adoles- Participants in this study are offered SRH services (contra- cent pregnancies compared to the prior year, though the ception, sexually transmitted infection diagnosis and analysis was limited to those presenting to government treatment, and HIV pre-exposure prophylaxis) at enroll- health centers [14]. A recent analysis from western Kenya ment and every visit. Questionnaires are conducted using concluded that compared with historical controls, female electronic case report forms (eCRFs; DF/Net Research, secondary school students who were out of school for at Inc.© , Seattle, WA, USA). least 6 months had twice the pregnancy risk and were 3 Participants are followed up at month 1, and thereafter, times more likely to drop out of school [15]. every 3 months, during which they have comprehensive In this study, we evaluated incident pregnancy trends assessment including genital swab collection. Providers and outcomes in a large cohort of Kenyan AGYW, pre- and administer clinical questionnaires at each visit. Pregnancy post-government lockdown during the COVID-19 pan- testing was done at enrollment, every 3 months, and when demic. We aim to contribute to the body of literature clinically indicated. Pregnant participants were followed- demonstrating the gendered impacts of the COVID-19 up and monitored to the end of pregnancy when the pandemic, which has further limited progress toward real- outcome was established. All these data were available for izing the Sustainable Development Goal 5 on gender the analysis presented in this article. equality [16]. COVID-19 lockdown and pregnancy estimation Methods Kenya reported its first case of confirmed COVID-19 on Study description and population March 12, 2020, and the government closed all learning The COVID-19 timeline of events spanning from closures institutions from March 13, 2020, to as one of the mea- and restrictions to phased reopenings is well-documented sures to curb viral spread. An estimated date of fertiliza- in Kenya. This unique COVID-19 situation offered us the tion (EDF) was computed as the first day of last menstrual opportunity to study a natural experiment on pregnancy period (LMP) plus 14 days to determine whether the preg- trends and outcomes in a cohort of Kenyan AGYW, nancy occurred before or after initiation of lockdown mea- enrolled in the KENya Single-dose HPV-vaccine Efficacy sures (pre- or post-COVID lockdown). Participants whose (KEN SHE) Study. EDF fell after March 13, 2020, were considered to have The KEN SHE Study is an individual randomized con- a pregnancy that occurred post-COVID lockdown; this is trolled trial to evaluate whether a single dose of the biva- equivalent to LMP on or after March 1, 2020. See the lent or nonavalent HPV vaccine prevents persistent HPV definition of independent variables in the Supplementary infection, a surrogate marker for precancerous lesions, Tables 1 and 2. and cervical cancer [17]. Briefly, between December 2018 and November 2019, the KEN SHE Study enrolled Measures and statistical analysis girls and young women aged 15–20 years in a double- Person-time started on December 20, 2018, and observa- blind randomized clinical trial testing the efficacy of a sin- tions were censored on March 31, 2021, or the last gle-dose bivalent and nonvalent HPV vaccination as time the person participated in the study. Person-time a catch-up strategy for cervical cancer prevention. Girls ended at date of pregnancy, lost to follow-up, or end of and young women from Thika, Nairobi, and Kisumu were the study period (March 31, 2021), whichever occurred eligible to participate if they were HIV-negative, reported first. If individuals came back to the study, they reentered 1 to 5 sexual partners in their lifetime, were not pregnant, the cohort (open cohort). Cumulative probabilities of and had a uterus and cervix. Study exclusion criteria pregnancy were estimated using Kaplan–Meier (K–M) included history of HPV vaccination, allergies to vaccine method. Pregnancy trends were determined among this Congo et al: Pregnancy trends and associated factors among Kenyan adolescent girls and young women Art. 1(1) page 3 of 9 cohort, pre- and post-COVID-19 lockdown (March 13, Table 1. Baseline sociodemographic characteristics of 2020). Pregnancy incidence rates were computed using the study population number of pregnancies as the numerator and person- Variables N % time in follow-up as the denominator. Pregnancy rates per 100 person-years observation were determined by pre-/ Study area post-COVID-19 lockdown, study site, and quarterly. Kisumu 1,168 52.5 Supplementary Figure 1 shows goodness-of-fit and model diagnostics. Thika 768 34.5 We assessed (1) pregnancy outcome (term delivery Nairobi 287 12.9 defined as delivery 37 weeks of gestational age, sponta- Do you have a main or steady sex partner? neous or induced abortion, ongoing or undetermined) and (2) postabortion and postpartum contraceptive No 402 18.4 uptake. We used a Cox proportional hazards model to Yes 1,781 81.6 investigate factors associated with pregnancy incidence Age-group in years and binary logistic regression to determine factors associ- ated with postabortion contraceptive initiation during the <18 years 998 44.9 pandemic period. 18 years 1,225 55.1 Variables significant at <0.2 level in the univariable Education level analysis or set a priori underwent further examination using a multivariable regression. Potential confounding None/Primary incomplete 162 7.3 effect of each covariate and a 2-way interaction were Secondary 1,614 72.6 addressed, and we obtained final variable selection by Tertiary/University 447 20.1 applying backward elimination method. Analysis and comparisons were done at 5% level of significance. The Marital status analyses were completed using STATA version 16.1 (STATA Single 2,102 94.7 Corporation, College Station, TX, USA). Married 96 4.3 Results Divorced/separated/widowed 22 1 Our analysis was restricted to 2,223 participants with com- Number of sexual partners plete information on pregnancy. Overall, median age of One 1,567 95.1 the participants was 18.6 years, IQR (17.6–20.3); girls and young women aged 18 years were slightly more than Two or more 80 4.9 a half (55%). Majority of the participants were single Condom use last sex (95%) at the time of enrollment, with majority of them No 876 53.2 reporting secondary school level as the highest at the time of enrollment (73%). Approximately 81.6% of girls Yes 771 46.8 reported a main or steady sexual partner, while approxi- Earned income in the last 12 months mately 3-quarters of the participants (75%) reported own- No 1,928 86.7 ing a mobile phone. Only 13% of the girls and young women earned their own income in the last 12 months. Yes 295 13.3 A total of 519 incident pregnancies were recorded over Own a phone 2 years of follow-up. Baseline characteristics by pregnancy No 418 24.6 status are presented in Table 1. The overall pregnancy incidence rate was 21 pregnancies per 100 woman-years Yes 1,283 75.4 of observation (95% CI [20, 23]). Nairobi and Thika had Do you smoke? a higher risk of pregnancy in comparison to Kisumu as No 2,212 99.5 shown by the K–M curves in Figure 1. There was a steady increase in pregnancy incidence rates toward the end of Yes 11 0.5 the last quarter of 2019, this trend peaked at the end Take alcohol of first quarter of 2020, corresponding to declaration of No 2,102 94.6 COVID-19 lockdown measures in Kenya. A steady, though not statistically significant, decline in pregnancy rate is Yes 121 5.4 shown between the third quarter of 2020 and first quarter N ¼ 2,223. of 2021. Pregnancy rates in Kisumu were significantly lower post-COVID-19 lockdown (Figures 2 and 3), whereas in Thika, there was a significant increase in rates post COVID-19 lockdown. Nairobi showed a slight reduc- abortion (6%). In Thika, 60% had term live birth and tion, which was not statistically significant. 51% were spontaneous abortions (Figure 4). Nairobi had Kisumu cohort pregnancy outcomes were (67%) term 60% term live births, 21% elective abortions, and 11% live birth, induced abortion (24%), and spontaneous spontaneous abortions. Art. 1(1) page 4 of 9 Congo et al: Pregnancy trends and associated factors among Kenyan adolescent girls and young women Kaplan-Meier pregnancy incidence survival estimates by the study sites Log-rank test p < 0.001 0 5 10 15 20 25 Analysis time Study site = Kisumu Study site = Thika Study site = Nairobi Figure 1. Kaplan–Meier curves for the cumulative risk of pregnancy by the study site. O V ERALL PRE G NANCY RATE PER 100 P YOS Test for trend overme p -value <0.001 2019Q1 2019Q2 2019Q3 2019Q4 2 020Q1 2 020Q2 2020Q3 2 020Q4 2 021Q1 Figure 2. Pregnancy rate for all sites combined. In adjusted analysis (Table 2), pre-/post-COVID-19 secondary school were nearly 3 times more likely to get lockdown, study site, age, education level, and condom pregnant during the COVID-19 lockdown period com- use independently predicted pregnancy incidence among pared to those in primary school or those with no educa- girls. In the post-COVID-19 lockdown period, girls and tion (HR ¼ 2.49, 95% CI [1.37, 4.53], P < 0.003), while young women from Nairobi and Thika had higher risk of those in postsecondary education were twice as likely to being pregnant, compared with pre-COVID-19 lockdown get pregnant compared to those with primary or no edu- cation (HR ¼ 2.06, 95% CI [1.11, 3.81], P < 0.021). (Uni- (HR ¼ 2.29, 95% CI [1.17, 4.48], P ¼ 0.016 and HR ¼ 3.10, variate analysis of factors associated with pregnancy 95% CI [1.76, 5.62], P ¼ 0.016, respectively). Condom use incidence are contained in Supplementary Table 3.) at last sex was associated with 71% reduced hazard of pregnancy (HR ¼ 0.29, 95% CI [0.18, 0.46], P < 0.001); however, participants aged 18 years who reported hav- Discussion ing used a condom during last sexual intercourse were We demonstrated that the Government of Kenya- more likely to be pregnant compared to minors (HR ¼ mandated COVID-19 lockdown, which included move- 2.00, 95% CI [1.14, 3.50], P < 0.015). Participants in ment restriction and closure of schools starting in March RATE PER 100 PYOS 0.00 0.25 0.50 0.75 1.00 Congo et al: Pregnancy trends and associated factors among Kenyan adolescent girls and young women Art. 1(1) page 5 of 9 Pregnancy rate pre and post COVID-19 lockdown by the study site 60.00 37.52 50.00 35.30 30.85 40.00 30.00 20.95 17.76 15.66 20.00 10.00 0.00 Pre covid-19 lockdown Post covid-19 lockdown Kisumu Thika Nairobi Figure 3. Overall pregnancy incidence rate pre- and post-COVID-19 lockdown. Pregnancy outcome by the study site (N= 514) Other/unknown 0% 1% 0% Spontaneous fetal death and/or stillbirth > 20 weeks 2% 3% 3% Therapeuc/elecve aboron 24% 6% 21% Spontaneous aboron (second trimester and < 20 weeks) 0% 4% 1% Spontaneous aboron (first trimester) 6% 25% 11% Ectopic pregnancy 0% 0% 1% Preterm live birth (<37 weeks) 0% 1% 4% Full-term live birth 37 weeks 67% 60% 60% Kisumu Thika Nairobi Figure 4. Overall pregnancy outcomes. 2020, was temporally associated with a rise in the hazard treatment for sexually transmitted infections, and obstet- of pregnancy among AGYW in central and western Kenya. ric care [20–22]. These barriers to access and quality care These findings correlate with the spike in teenage preg- expose adolescents to negative health and life-altering nancy that was publicized by media and programmatic socioeconomic consequences: consequences that are even organizations in the wake of the COVID-19 lockdown more pronounced in times of crisis such as disease out- [11]. The pregnancy incidence rate differed by site, which breaks or war, when data suggest they are disproportion- we hypothesize is related to regional secular adolescent ately affected. Furthermore, prior studies suggest that pregnancy trends as well as geographically specific school closures, lack of comprehensive sexuality educa- COVID-19 movement restrictions, as Thika and Nairobi tion, and increased risk of violence against girls and young were put under more stringent lockdown measures. This women put them at higher risk of unintended pregnancy. is consistent with experience from previous public health A recently published study from rural western Kenya com- crises, such as recent Ebola outbreak in Liberia and Sierra pared school drop-out and pregnancy rates pre- and post- Leone, which demonstrated reduced access to essential COVID-19-related school disruptions and found a 3-fold SRH and other health services [18, 19]. increase in school drop-outs and a 2-fold increase in preg- Our study demonstrates a rise in pregnancy incidence nancy among secondary school students whose school was during the COVID-19 lockdown period when compared to disrupted, which is similar to our study findings [15]. the pre-COVID-19 period. AGYW constitutes a special The study population had 1 abortion for every 4 preg- population who already face social, developmental, and nancies, with 19% being induced abortions. The conse- health systems barriers to accessing quality SRH care, quences of an unwanted pregnancy may be devastating including contraception, abortion, screening and for many adolescents, which may lead them to seek Pregnancy rate per 100 PYOs Art. 1(1) page 6 of 9 Congo et al: Pregnancy trends and associated factors among Kenyan adolescent girls and young women Table 2. Factors associated with pregnancy incidence among girls aged 15-20 years Unadjusted HR Adjusted HR Factor HR [95% CI] P Value HR [95% CI] P Value COVID-19 lockdown Pre-COVID-19 lockdown Ref Post-COVID-19 lockdown 1.32 [1.10, 1.57] 0.003 0.72 [0.46, 1.12] 0.147 Interaction (pre–post#Site) Post Covid#Thika 2.82 [1.59, 4.99] <0.001 Post Covid#Nairobi 2.09 [1.07, 4.07] 0.029 Study area Kisumu Ref Thika 1.46 [1.21, 1.78] <0.001 0.84 [0.59, 1.21] 0.366 Nairobi 2.4 [1.90, 3.03] <0.001 1.56 [0.95, 2.54] 0.074 Age-group in years 15–17 Ref 18–20 1.26 [1.05, 1.50] 0.011 0.76 [0.58, 1.09] 0.157 Marital status Single Ref Married 0.83 [0.54, 1.27] 0.392 Divorced/separated/widowed 1.05 [0.52, 2.10] 0.894 Highest education level None/primary incomplete Ref Secondary 1.43 [1.00, 2.05] 0.052 2.45 [1.35, 4.45] 0.003 Tertiary/university 1.68 [1.14, 2.48] 0.009 2.04 [1.11, 3.77] 0.023 Number of sexual partners One Ref Two or more 0.85 [0.52, 1.37] 0.500 Do you have a main or steady sex partner? No Ref Yes 1.08 [0.84, 1.38] 0.549 Own a phone No Ref Yes 1.38 [1.07, 1.76] 0.011 1.16 [0.82, 1.63] 0.382 Condom use last sex No Ref Yes 0.41 [0.33, 0.52] <0.001 0.28 [0.18, 0.45] <0.001 Interaction (condom use last sex# age-group) >18 years#Yes 2.06 [1.18, 3.59] 0.010 Do you smoke? No Ref Yes 0.48 [0.07, 3.47] 0.467 Take alcohol No Ref Yes 1.04 [0.72, 1.51] 0.842 Congo et al: Pregnancy trends and associated factors among Kenyan adolescent girls and young women Art. 1(1) page 7 of 9 abortion in unsafe circumstances, particularly in countries Table 3. Univariate analysis of factors associated with where abortion is highly restricted [23, 24]. Based on pregnancy incidence. extensive clinical experience in the region and confirmed Figure 1. Goodness-of-fit and model diagnostics. reports of induced abortion in this cohort, our team sus- pects a far higher proportion of abortions were induced, Acknowledgments often without a skilled provider or using practices known We thank the young women who participated in this study to be harmful [25]. Our data underscore the known high for their motivation and dedication and the communities unmet need for contraception among girls and young that supported this work. women in sub-Saharan Africa [26] and the need to protect these services in times of crisis. Funding Our study has a number of strengths, including its The KEN SHE Study was funded by the Bill & Melinda innovative leverage of a natural experiment and use of Gates Foundation (OPP1188693) and the University of a multisite design representing urban and rural settings Washington King K. Holmes Endowed Professorship in in Kenya with very low loss to follow-up. We had labora- STDs and AIDS. The content is solely the responsibility tory ascertainment of pregnancy status with routine preg- of the authors and does not necessarily represent the nancy tests done quarterly. While the present study had views, decisions, or policies of the institutions with which a number of advantages because of the availability of they are affiliated or the KEN SHE Study funders. The a well-studied cohort, this feature was also a limitation funders had no separate role in the study design; data of the study because the girls and young women in the collection, analysis, and interpretation; writing of the cohort had continued, though disrupted, access to SRH report; or in the decision to submit for publication. care including contraceptive counseling and provision of The corresponding author had full access to all the data STI treatment throughout the COVID-19 lockdown due to in the study and had final responsibility for the decision to study participation. It is likely, therefore, that age-matched submit for publication. girls and young women who were not members of the cohort had a higher pregnancy risk than we report, and Competing interests thus, the pregnancy incidence rates may not be general- The authors have no competing interests to declare. izable to nonstudy participants’ risk. We did not collect repeated measures of pregnancy intentions and the con- Ethics approval and consent to participate sistency of condom use, which limit our understanding of The KEMRI Scientific and Ethics Review Unit (SERU), in these factors as influences on pregnancy risk. Finally, our Kenya, and the University of Washington Institutional study design cannot completely account for age as a time- Review Board, Seattle, WA, approved this study. The study varying confounder of pregnancy incidence. was registered at ClinicalTrials.gov, number NCT03675256. In conclusion, our findings contribute to the evidence All participants provided written informed consent, empirically demonstrating the social and health conse- which included counseling about randomization, the vac- quences of the COVID-19 pandemic, which widen cine used in each study group, strategies for cervical can- already-existing SRH inequities confronting AGYW. We cer prevention, and their rights as research participants. demonstrate that despite frequent contact and compre- hensive contraceptive care offered through the study References clinic, girls and young women in this cohort had a similar 1. Pulse survey on continuity of essential health services pregnancy rate to the general aged-matched population in during the COVID-19 pandemic: interim report, 27 Kenya, many of whom did not have adequate access to August 2020 [Internet]. 2020 Aug 27 [cited 2021 SRH services during COVID-19 containment measures. Oct 18]. Available from: https://www.who.int/ Given the many barriers to contraceptive use among this publications/i/item/WHO-2019-nCoV-EHS_continuity- age-group, ongoing investigation into their unique contra- survey-2020.1. ceptive needs and fertility desires, power in sexual rela- 2. Mooi-Reci I, Risman BJ. The gendered impacts of tionships, and agency in reproductive decision-making is COVID-19: lessons and reflections. Gend Soc [Inter- critical to improving adolescent SRH outcomes. net]. 2021 [cited 2022 Jan 23];35(2):161-7. Available from: https://journals.sagepub.com/doi/10.1177/ Data accessibility statement 3. How COVID-19 has affected global progress [Internet]. The main study that provided data for this paper is ongo- 2020 Sep [cited 2022 Jan 30]. Available from: https:// ing; data can be accessed by writing to RVB. www.gatesfoundation.org/goalkeepers/report/2020- report/progress-indicators/maternal-mortality/. Supplemental files 4. de Paz C, Muller M, Munoz Boudet AM, Gaddis I. The supplemental files for this article can be found as Gender dimensions of the COVID-19 pandemic [Inter- follows: net]. 2020 Apr 16 [cited 2022 Jan 27]. Available from: Table 1. Independent variables (demographic https://openknowledge.worldbank.org/handle/ characteristics). 10986/33622. Table 2. Independent variables (sexual behavior 5. Onono Id MA, Rutherford GW, Bukusi EA, Whiteid JS, practices). Goosby E, Brindis CD. Political prioritization and the Art. 1(1) page 8 of 9 Congo et al: Pregnancy trends and associated factors among Kenyan adolescent girls and young women competing definitions of adolescent pregnancy in 16. United Nations: gender equality and women’s empow- Kenya: an application of the Public Arenas Model. erment [Internet]. 2022 [cited 2022 Mar 3]. Available PLoS One. 2020 [cited 2022 Mar 8]. Available from: from: https://www.un.org/sustainabledevelopment/ https://doi.org/10.1371/journal.pone.0238136. gender-equality/. 17. Barnabas RV, Brown ER, Onono M, Bukusi EA, Njoroge 6. National Bureau of Statistics Nairobi K. Republic of B, Winer RL, et al. Single-dose HPV vaccination efficacy Kenya—Kenya Demographic and Health Survey 2014. among adolescent girls and young women in Kenya 2015 [cited 2022 Jan 28]. Available from: www. (the KEN SHE Study): study protocol for a randomized DHSprogram.com. controlled trial. Trials [Internet]. 2021 [cited 2022 Mar 7. Darroch J, Audam S, Biddlecom A, Kopplin G, Riley T, 7];22(1):1-19. Available from: https://trialsjournal. Singh S. Adding it up: investing in contraception and biomedcentral.com/articles/10.1186/s13063-021- maternal and newborn health. New York (NY): The 05608-8. Guttmacher Institute; 2017. 18. The socio-economic impacts of Ebola in Liberia: 8. Christofides NJ, Jewkes RK, Dunkle KL, Nduna M, Shai results from a high frequency cell phone survey, NJ, Sterk C. Early adolescent pregnancy increases risk Round 5 [Internet]. [cited 2022 Mar 7]. Available of incident HIV infection in the Eastern Cape, South from: https://openknowledge.worldbank.org/ Africa: a longitudinal study. J Int AIDS Soc [Internet]. handle/10986/21893. 2014 [cited 2022 Mar 8];17(1). Available from: 19. Elston JWT, Cartwright C, Ndumbi P, Wright J. The https://www.ncbi.nlm.nih.gov/pmc/articles/PMC39 health impact of the 2014–15 Ebola outbreak. Public 62027/. Health. 2017;143:60-70. 9. Osok J, Kigamwa P, Stoep AV, Huang K-Y, Kumar M. 20. Durowade KA, Omokanye LO, Elegbede OE, Adeto- Depression and its psychosocial risk factors in preg- kunbo S, Olomofe CO, Ajiboye AD, et al. Barriers to nant Kenyan adolescents: a cross-sectional study in contraceptive uptake among women of reproductive a community health centre of Nairobi. BMC Psychiatry age in a semi-urban community of Ekiti State, South- [Internet]. 2018 [cited 2022 Mar 8];18(1):136. Available west Nigeria. Ethiop J Health Sci [Internet]. 2017 [cited from: https://bmcpsychiatry.biomedcentral.com/ 2022 Mar 4];27(2):121-8. Available from: https://www. articles/10.1186/s12888-018-1706-y. ajol.info/index.php/ejhs/article/view/153144. 10. The Case of Korogocho Slums Final Report. Nairobi: 21. Ababor AA, Tesso DW, Cheme MC. Addressing the KHRC and RHRA; 2010. deprived: need and access of sexual reproductive 11. COVID-19: lockdown linked to high number of unin- health services to street adolescents in Ethiopia. The tended teen pregnancies in Kenya. Kenya | ReliefWeb case of Nekemte town: mixed methods study. BMC Res [Internet]. [cited 2022 Jan 30]. Available from: Notes [Internet]. 2019 [cited 2022 Mar 3];12(1):1-6. https://reliefweb.int/report/kenya/covid-19- Available from: https://bmcresnotes.biomedcentral. lockdown-linked-high-number-unintended-teen- com/articles/10.1186/s13104-019-4850-7. pregnancies-kenya. 22. Machiyama K, Casterline JB, Mumah JN, Huda FA, 12. Covid-19 pandemic: 13,000 girls got pregnant in Obare F, Odwe G, et al. Reasons for unmet need for Narok in 2021. The Standard [Internet]. [cited 2022 family planning, with attention to the measurement Mar 8]. Available from: https://www.standardmedia. of fertility preferences: protocol for a multi-site cohort co.ke/the-standard-insider/article/2001436813/ study. Reprod Health [Internet]. 2017 [cited 2022 Mar covid-19-pandemic-13000-girls-got-pregnant-in- 4];14(1):1-11. Available from: https://reproductive- narok-in-2021. health-journal.biomedcentral.com/articles/10.1186/ 13. Kilifi girls: Teen pregnancies in the region a pandemic. s12978-016-0268-z. Kenya [Internet]. [cited 2022 Mar 8]. Available from: 23. WHO/UNICEF. School closures and teenage preg- https://nation.africa/kenya/news/gender/kilifi-girls- nancy. Bull World Health Organ [Internet]. 2021 [cited teen-pregnancies-in-the-region-a-pandemic-3644438? 2022 Jan 30];99(1):6. Available from: https://www. view¼htmlamp. ncbi.nlm.nih.gov/pmc/articles/PMC7924898/. 14. Teen pregnancy in Kenya: verifying the data and the 24. Preventing Unintended Pregnancies in the Context of facts. African Institute for Development Policy—AFI- COVID-19. Wilson Center [Internet]. [cited 2022 Feb DEP [Internet]. 2020 [cited 2022 Jan 30]. Available 25]. Available from: https://www.wilsoncenter.org/ from: https://www.afidep.org/news-release-teen- event/preventing-unintended-pregnancies-context- pregnancy-in-kenya-verifying-the-data-and-the-facts/. covid-19. 15. Zulaika G, Bulbarelli M, Nyothach E, Van Eijk A, Mason 25. WHO. Preventing unsafeabortion. Geneva: WHO; L, Fwaya E, et al. Impact of COVID-19 lockdowns on adolescent pregnancy and school dropout among sec- 26. Dickson KS, Adde KS, Ahinkorah BO. Socio-economic ondary schoolgirls in Kenya. BMJ Glob Heal [Internet]. determinants of abortion among women in Mozam- 2022 [cited 2022 Mar 2];7(1):e007666. Available bique and Ghana: evidence from demographic and from: https://gh.bmj.com/content/7/1/e007666. health survey. Arch Public Heal. 2018;76(1):1-10. Congo et al: Pregnancy trends and associated factors among Kenyan adolescent girls and young women Art. 1(1) page 9 of 9 How to cite this article: Congo O, Otieno G, Wakhungu I, Harrington EK, Kimanthi S, Biwott C, et al. Pregnancy trends and associated factors among Kenyan adolescent girls and young women pre- and post-COVID-19 lockdown. Adv Glob Health. 2022;1(1). https://doi.org/10.1525/agh.2022.1811306 Editor-in-Chief: Craig R. Cohen, University of California, San Francisco, CA, USA Senior Editor: Sarah Ssali, Makerere University, Kampala, Uganda Section: Achieving Gender Equality Published: December 8, 2022 Accepted: October 29, 2022 Submitted: March 18, 2022 Copyright: © 2022 The Author(s). This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (CC-BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. See http://creativecommons.org/licenses/by/4.0/. Elem Sci Anth is a peer-reviewed open access journal published by University of California Press.

Journal

Advances in Global Health – University of California Press

Published: Dec 8, 2022

Keywords: Adolescent; Pregnancy; COVID-19; Lockdown; AGYW

References

  • The gendered impacts of COVID-19: lessons and reflections
    Mooi-Reci, I; Risman, BJ
  • Gender dimensions of the COVID-19 pandemic [Internet]
    de Paz, C; Muller, M; Munoz Boudet, AM; Gaddis, I
  • Political prioritization and the competing definitions of adolescent pregnancy in Kenya: an application of the Public Arenas Model
    Onono Id, MA; Rutherford, GW; Bukusi, EA; Whiteid, JS; Goosby, E; Brindis, CD
  • Republic of Kenya—Kenya Demographic and Health Survey 2014
  • Early adolescent pregnancy increases risk of incident HIV infection in the Eastern Cape, South Africa: a longitudinal study
    Christofides, NJ; Jewkes, RK; Dunkle, KL; Nduna, M; Shai, NJ; Sterk, C.
  • Depression and its psychosocial risk factors in pregnant Kenyan adolescents: a cross-sectional study in a community health centre of Nairobi
    Osok, J; Kigamwa, P; Stoep, AV; Huang, K-Y; Kumar, M
  • Kenya [Internet]
  • African Institute for Development Policy—AFIDEP [Internet]
  • Impact of COVID-19 lockdowns on adolescent pregnancy and school dropout among secondary schoolgirls in Kenya
    Zulaika, G; Bulbarelli, M; Nyothach, E; Van Eijk, A; Mason, L; Fwaya, E
  • Single-dose HPV vaccination efficacy among adolescent girls and young women in Kenya (the KEN SHE Study): study protocol for a randomized controlled trial
    Barnabas, RV; Brown, ER; Onono, M; Bukusi, EA; Njoroge, B; Winer, RL
  • The health impact of the 2014–15 Ebola outbreak
    Elston, JWT; Cartwright, C; Ndumbi, P; Wright, J
  • Barriers to contraceptive uptake among women of reproductive age in a semi-urban community of Ekiti State, Southwest Nigeria
    Durowade, KA; Omokanye, LO; Elegbede, OE; Adetokunbo, S; Olomofe, CO; Ajiboye, AD
  • Addressing the deprived: need and access of sexual reproductive health services to street adolescents in Ethiopia. The case of Nekemte town: mixed methods study
    Ababor, AA; Tesso, DW; Cheme, MC
  • Reasons for unmet need for family planning, with attention to the measurement of fertility preferences: protocol for a multi-site cohort study
    Machiyama, K; Casterline, JB; Mumah, JN; Huda, FA; Obare, F; Odwe, G
  • School closures and teenage pregnancy
  • Socio-economic determinants of abortion among women in Mozambique and Ghana: evidence from demographic and health survey
    Dickson, KS; Adde, KS; Ahinkorah, BO