Preterm birth and subsequent pregnancies in Matlab, Bangladesh: a recurrent survival analysis

Preterm birth and subsequent pregnancies in Matlab, Bangladesh: a recurrent survival analysis | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 26 March 2026 V1 Latest version Share on Preterm birth and subsequent pregnancies in Matlab, Bangladesh: a recurrent survival analysis Authors : Wnurinham Silva 0000-0001-8129-890X [email protected] , Monjur Rahman 0000-0002-7877-858X , Markku Nurhonen , Syed Manzoor Ahmed Hanifi , Jesmin Pervin , Suvi Alenius , Johanna Metsälä , Sylvain Sebert , Anisur Rahman , and Eero Kajantie Authors Info & Affiliations https://doi.org/10.22541/au.177452316.68521391/v1 141 views 79 downloads Contents Abstract Supplementary Material Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Objective: To test the association between preterm-birth and time to occurrence of a subsequent pregnancy, compared with term-birth. Design: Prospective cohort study using health and demographic surveillance system data. Setting: Matlab area, Bangladesh (1990-2020). Population or Sample: 52,502 pregnancies from 24,559 women, excluding pregnancies with multiple fetuses and women whose first pregnancy occurred after 45 years of age. Methods: Main exposure was preterm-birth. Previous miscarriages and stillbirths served as secondary exposures. Associations were assessed using recurrent survival analysis. Impact of infant mortality was assessed by including death of the previous child under one year of age as time-dependent variable. Hazard ratio (HR) of a subsequent pregnancy was calculated with 95% confidence intervals (CIs). Main Outcome Measures: time to occurrence of a subsequent pregnancy counted as the date of the previous event until the date of the next last menstrual period. Results: Women with a previous extremely-very and moderately preterm-birth had an increased likelihood for subsequent pregnancy, compared to women with a term-birth (Adj.HR: 1.22 (95% CI 1.08-1.39; 1.13 (1.04-1.23), respectively). After adjusting for death of the previous child HRs reduced to 0.97 (0.85-1.09) and 1.03 (0.95-1.12), suggesting that the increased HRs observed was explained by infant death. Previous stillbirth and miscarriage were associated with approximately three-fold likelihood of a subsequent pregnancy. Conclusions: In Matlab, women with a previous preterm-birth were more likely to get pregnant than women with a previous term-birth. However, after adjusting for infant mortality there was no difference in pregnancy rates between previous preterm or term-birth. INTRODUCTION Approximately 13.4 million babies, one in ten, are born preterm each year globally 1 . Children and adults born preterm are at an increased risk of a range of poor short- and long-term outcomes 2–5 . However, the effects of preterm birth may also extend to the parents, impacting several areas of their lives. One of such areas is later parental reproductive outcomes. Previous studies have shown that parents are less likely to have a subsequent child following a preterm birth (< 37 weeks of gestation) 6–13 . To what extent this is due to altered fertility intentions following a preterm born offspring or to biological fertility problems, it is not known. Regardless of the reason, the association is relevant because of the fundamental effect of childbirth on family life. Moreover, previous studies have suggested that only-children are at an increased risk of poor behavioural outcomes compared to children with siblings 14–16 . However, to our knowledge, all studies investigating the association between preterm birth and subsequent children have emerged from high income countries 6–8,11 . Using cohort data from Matlab subdistrict, a rural low-income setting in Bangladesh, we investigated the association between a previous preterm birth and time to occurrence of a subsequent pregnancy. We hypothesised that in Matlab, the time to a subsequent pregnancy would be longer among women with a previous preterm birth, compared to previous term birth. Setting and population This study was based on prospectively collected data in the Matlab area, a subdistrict of Chandpur, Bangladesh, where a health and demographic surveillance system (HDSS) has been running since 1966, covering a population of 220,000 individuals 17–19 . HDSS is divided into two areas: the International Centre for Diarrhoeal Disease Research, Bangladesh (icddr,b) and the government service areas. The present study used data collected from the icddr,b service area. We used data available for the period of January 1990 to December 2020, including a population of 43,277 women and 91,052 pregnancies. Women with twins and triplets (n=758), and whose first pregnancy occurred after the age of 45 years were excluded (n=2). We included women of reproductive age who had null parity at time of entry in the study and whose first pregnancy occurred between ages of 12 and 45 years (n=24,604). Following inclusion and exclusion criteria, a total of 24,559 women and 52,502 pregnancies were included in the main analyses (Figure 1). <> Outcome The primary outcome of interest was defined as time to occurrence of a subsequent pregnancy assessed using the Prentice-Williams-Peterson gap time (PWP-GT) model, an extension of the Cox Proportional Hazards regression, as described in the statistical analysis section. The underlying time scale was counted as the date of the previous event (i.e. livebirth, stillbirth, miscarriage), until the date of the next last menstrual period. Exposure The main exposure was whether the previous birth (birth preceding the current one) was a preterm birth (livebirth occurring between 23 and 36 completed weeks of gestation) or a term birth (livebirths occurring between 37 to 44 weeks of gestation, reference category). In additional analysis, we used preterm birth categorized into three subcategories: extremely-very preterm 23 to 31 completed weeks’ gestation; moderately preterm 32 to 33 weeks; and late preterm 34 to 36 weeks). There were only 23 cases of extremely preterm births (23 to 27 weeks), which had been recorded only from 2009. To reduce risk of bias and to increase statistical validity, the extremely preterm and very preterm subcategories were combined. Stillbirths (late fetal deaths ≥28 weeks) 20 miscarriages (fetal deaths before 28 weeks of gestation) 21 served as secondary exposures. Gestational age was calculated using the reported date of the women’s last menstrual period, subtracted by the delivery date, expressed in weeks. Last menstrual period data was collected by community health research workers visiting the women in the area regularly to identify pregnancies 18 . Visits were carried out fortnightly until 2001, monthly from 2002 to 2006, and bi-monthly from 2007. Pregnancy was after that confirmed at subsequent visits and followed up for type and date of pregnancy outcomes 18 . Previous and current pregnancies Previous pregnancy was defined as the pregnancy preceding the current pregnancy (including miscarriages, stillbirths and livebirths). Current pregnancy was the pregnancy for which we were interested in calculating the hazard ratio (HR). Covariates A list of confounders was selected a priori based on the available evidence on their association with preterm birth and subsequent pregnancy 6,22 . Maternal age at the end of the previous pregnancy was used as a continuous variable and expressed in years. Maternal and paternal educational level, defined as the number of years completed at school (0 years= No education; 1-5 years= Primary education; 6-10 years= Secondary education; 11+ years= Higher education). Asset scores (continuous variable from 1 to 5, with category one representing the poorest and category five representing the richest) were used as a proxy of wealth status expressed in quintiles, calculated using factor analysis and consists of a collection of household items owned by the family 19 . In addition, for the sensitivity analysis we included a time-dependent variable on the death of a previous child under one year of age, defined as whether a previous liveborn offspring had died before one year of age in the two years preceding a new pregnancy (Yes/No). Statistical analyses The most traditional way of analysing survival data is the Cox model, which analyses the time to first event. However, that model ignores recurrent events, which leads to substantial loss of information and to reduced hazard ratios 23 . Extensions of the Cox model allow to include the full information of each individual factoring that they may experience reoccurrence of an event over time, while accounting for the intra-subject correlation among events by using robust variance estimates 24–26 . Hence, we have used the PWP-GT, which calculates the hazard function by stratification 24 . For every woman the number of events determines the number of strata. All women in the same risk set have had the same number of prior events and survived the same length of follow-up since the most recent event. Recurrent events analysis was used to determine hazard ratios and 95% confidence intervals (CIs) for association between previous preterm birth and time to occurrence of subsequent pregnancy. In the PWP-GT model, the HR is interpreted as an average increase in the effect of the previous pregnancy on the subsequent pregnancy from one exposure group relative to the comparison group across the duration of the study. Time at risk started from the date of occurrence of the previous pregnancy event. Follow up ended at outmigration (N = 2,023, 8%), maternal death (N = 67, 0,3%) or end of follow-up (31 st of December 2020), whichever occurred first. We conducted unadjusted and adjusted analyses, using two different models in the main analyses. Model 1 was unadjusted. Model 2 was adjusted for maternal age at previous pregnancy, asset, maternal education and paternal education. In sensitivity analysis Model 3 was adjusted as in model 2 plus for death of the previous child under one year of age. Firstly, we assessed the association between preterm birth and time to a subsequent pregnancy using preterm birth as one category (23-36 weeks of gestation). Secondly, we tested the association between preterm birth and time to a subsequent pregnancy using preterm birth subcategories (extremely-very preterm/moderately preterm/late preterm). Thirdly, we conducted sensitivity analyses by analysing the effect of death of the previous child when under one year of age and while death occurred in less than 2 years ago, using death of the previous child as a time-dependent variable. Additionally, we tested the association between previous pregnancy outcomes (miscarriages, stillbirths and livebirths), stratified by sex, and time to occurrence of subsequent pregnancy. To independently test the effect of death, we also tested the association between death of the previous liveborn child (Yes/No) and time to occurrence of subsequent pregnancy. Lastly, we calculated estimated survival curves by previous pregnancy outcome, sex of the previous child, and by parity (one, two and three), adjusting for the time‑dependent effect of the death of a previous child. The rate of missing data for the maternal and paternal education was (n=73), <1% and (n=2,108) 9%, respectively. To address the issue of missing data, we added an additional category “Unknown”. All analyses were conducted using R Studio software version 4.3.0. Ethical approval The present study used routine prospectively collected Data by the HDSS. HDSS periodically receives consent from the household heads for routine household visits and data collection and receives approval from the Institutional Review Board (IRB) of the icddr,b. The IRB of icddr,b also approved the use HDSS data to understand preterm births and related morbidity in the HDSS area (PR- 14069). RESULTS Population characteristics The characteristics of the population are described in Table 1, separately for women with at least one preterm birth, women without preterm births, and the whole population. The median age of the women at time of the first pregnancy was 20 years in both groups. The median age at the current pregnancy was 24 years for the group of women with at least one preterm birth and 23 years for women without preterm births. Children of the group of women with at least one preterm birth had an increased mortality during their first year of life (5% vs 2%), the women were more likely to have a higher number of children, lower levels of parental education and marginally lower socio-economic status, indicated by asset quintiles, and lower rates of birth by caesarean section compared to the group of women without preterm births. Women with only one pregnancy throughout the study period had approximately the same percentage of preterm births as women with two to four pregnancies (12%, 13%, respectively) (Table S1). However, the percentage of preterm births was higher among women with more than four pregnancies (16%). Women with one pregnancy during the study period had higher educational level compared to women with more than one pregnancy (Table S1). <<>> Recurrent event analyses Adjusted and unadjusted estimates for association between previous preterm birth and time to occurrence of subsequent pregnancy using preterm birth as a one homogenous category (23-36 weeks of gestation) are shown in Table S2. In the unadjusted analysis we observed an HR of 1.13 (CI=1.09-1.17) for time of occurrence of subsequent pregnancy (Table S2). After adjusting the HR weakened to 1.06 (CI=1.02-1.09). The adjusted analysis including preterm birth subcategories (extremely-very preterm/moderately preterm/late preterm), showed an increased HR for time to occurrence the previous extremely-very preterm and moderately preterm subcategories (Adj.HR=1.22;1.08-1.39 and Adj.HR=1.13;1.04-1.23, respectively) (Table 2). <<>> In all analyses, stillbirth and miscarriage were consistently associated with approximately three times higher HR for time to occurrence of subsequent pregnancy (Tables 2 and S2). Sensitivity Analyses Decreasing gestational age is associated with increased infant mortality. Previous studies have shown that infant death, in turn, could prompt more rapid reproduction 7 . We accounted for the death of the previous child under one year of age in a time-dependent manner (Tables 3 and S3). After adjusting for death of the previous child, the HRs decreased to 1 suggesting that HRs observed in the main analyses were explained by increased mortality of the previous child (Tables 3, S3 and S5). The estimates indicated that women whose previous child died under one year of age had approximately seven times higher hazard of subsequent pregnancy, compared to women whose child survived (Tables 3 and S3). <<>> Additional analyses Analysis showing sex specific effects of the previous birth indicated that when the previous child was a girl, women had a higher hazard of a subsequent pregnancy than when the previous child was a boy. This was similar for those born preterm and at term (Adj.HR=1.15;1.11-1.18 and Adj.HR=1.12;1.06-1.17, respectively) (Table S4). In analysis using death of the previous liveborn child as exposure, the death of the child was strongly associated with shorter time to occurrence of subsequent pregnancy (Table S5). Estimated survival curves Estimated probability of remaining without a subsequent pregnancy by previous pregnancy outcome, sex of the previous liveborn child, and by parity, adjusting for the time‑dependent effect of the death of a previous child is shown in Figure S6. The curves correspond to the mean age of women in gravidities one, two, three and four (21, 25, 28 and 31 years, respectively). The plot curves illustrate the higher rates of subsequent children among women with previous stillbirth and previous miscarriage. Sex data for stillbirths and miscarriages was not available, therefore only the overall categories are shown in the plot. DISCUSSION Main findings In this cohort study using prospectively collected data of 24,559 women and 52,502 pregnancies in the Matlab area, we found evidence supporting that infant mortality was a predictor of subsequent pregnancies among women in this rural area of Bangladesh. Results of the main analyses revealed increased HRs for time to subsequent pregnancy among women with a previous extremely-very and moderately preterm births. However, the effect disappeared after adjusting for infant death, suggesting that the association observed in the initial analyses was explained by death of the previous child. In our study, approximately 26%, 12% and 5% of the extremely-very preterm, moderately preterm and late preterm babies, respectively, died before one year of age, compared to 2% of those born at term. In addition, our results showed that previous stillbirth and miscarriage were also associated with subsequent pregnancies. Women with a previous stillbirth or miscarriage had 3-fold increased estimates for time to occurrence of subsequent pregnancy. Approximately 9% and 2% of the pregnancies in this study ended in miscarriage and stillbirth, respectively. Interpretation Contrary to our hypothesis, we found no evidence to support an association between preterm birth and time to occurrence of subsequent pregnancy. The finding that infant death plays a key role is consistent with those of a Finnish study involving 223,615 children, which showed that the most extreme category of prematurity was associated with higher HRs for having a subsequent sibling, but only if the child did not survive infancy 7 . Another study conducted in Norway showed that women who had experienced preterm birth had a relative risk (RR) of 1.35 (1.33-1.38) for having only one pregnancy, however the RR reduced to 0.64 (0.60-0.69) if the women had lost the child 6 . Together with other studies showing an association between loss of a child and additional pregnancies 6,7,27–30 , our findings suggest that among women in the Matlab area, reproductive behaviour was associated with infant mortality. One proposed mechanism to explain the trends observed in this study is selective fertility or replacement child, which have been used to describe the tendency to replace reproductive losses to obtain a desired number of children 6,27 . Similarly, theories of planned behaviour argue that fertility outcomes are purposive, based on fertility intentions and modified when unexpected developments occur 31,32 . Results of the sex stratified analysis further validated the lack of association between preterm birth and the likelihood of a subsequent pregnancy. This analysis showed higher hazards for subsequent pregnancy following the birth of a female infant, irrespective of whether the previous birth was preterm or term. An important factor when interpreting our findings are the demographics of our study population. In this study, the median total number of children was three and the median interpregnancy interval among women in the Matlab district was four years. These are consistent with figures reported in previous studies indicating a 3.2 fertility rate in the Matlab area in 1995 and an interpregnancy interval of approximately three years in 1996 33 . However, in this study, among the group of women who lost a child under one year of age the median interpregnancy interval was two years. The median age of the women included in this study at time of their first pregnancy was 20 years. This is lower than the age of women included in previous works assessing the existence of subsequent children following preterm birth 7,11,13 . Strengths and limitations Major strengths of this study include the use of a large population of women and children. We have assessed the association between preterm birth and subsequent pregnancies using recurrent events analysis which allowed us to use all pregnancy events data, rather than just the first pregnancies. However, this study carries important limitations. Our findings should be interpreted within the unique setting that is Matlab, a rural area which has benefitted from health and reproductive interventions since the 1960s and, consequently, holds better than average health indicators than the ones observed in the remaining Bangladesh. In addition, we were not able to test for contraception use, religious affiliation, or more qualitative measures such as desire or intention to get pregnant,. That could have provided further insight into the mechanisms behind preterm birth and its association with subsequent pregnancies in Matlab. Other limitations include inability to adjust for the condition of the child and maternal gestational disorders in previous pregnancies. CONCLUSIONS Contrary to our hypothesis, women with a previous preterm birth were more likely to get pregnant than women with a previous term birth. This was, however, wholly explained by increased mortality of preterm infants: when infant death was accounted for, there was no difference in the likelihood of getting pregnant. This contrasts findings in high-resource settings and calls for studies on the effect of preterm birth on parental reproduction in different settings. Author’s contribution Wnurinham Silva was responsible for article administration tasks including management and coordination of the research activities, conceptualized and designed the study, curated the data, conducted the data analyses, drafted the initial manuscript, revised the manuscript and incorporated all co-authors’ comments and suggestions for improvements. Monjur Rahman provided the data and critically reviewed and revised the manuscript. Markku Nurhonen provided statistical support and critically reviewed and revised the manuscript. Syed Manzoor Ahmed Hanifi, Jesmin Pervin, Suvi Alenius and Johanna Metsälä critically reviewed and revised the manuscript. Anisur Rahman conceptualized and designed the study, provided the data and critically reviewed and revised the manuscript. Sylvain Sebert and Eero Kajantie supervised the work, conceptualized and designed the study, provided the data and critically reviewed and revised the manuscript. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work. Acknowledgements: Authors would like to acknowledge Maryam Nasserinejad, who provided statistical support in the early stages of this study and Iiro Nerg, who helped checking the validity of the analyses conducted. Funding: Wnurinham Silva was supported by University of Oulu Graduate School and Research Council of Finland, Novo Nordisk Foundation and Sigrid Jusélius Foundation. Eero Kajantie was supported by Research Council of Finland (358384, 374278), European Commission (101156325 IMPROVE PRETERM), Finska Läkaresällskapet, Foundation for Pediatric Research, Novo Nordisk Foundation, Signe and Ane Gyllenberg Foundation, Sigrid Jusélius Fundation. Conflicts of Interests: The authors have no conflicts of interest relevant to this article to disclose. Data availability statement: References 1. Ohuma EO, Moller AB, Bradley E, Chakwera S, Hussain-Alkhateeb L, Lewin A, et al. National, regional, and global estimates of preterm birth in 2020, with trends from 2010: a systematic analysis. The Lancet. 2023 Oct 7;402(10409):1261–71. doi:10.1016/S0140-6736(23)00878-4 PubMed PMID: 37805217. 2. Hollanders JJ, Schaëfer N, Van Der Pal SM, Oosterlaan J, Rotteveel J, Finken MJJ. Long-Term Neurodevelopmental and Functional Outcomes of Infants Born Very Preterm and/or with a Very Low Birth Weight. Neonatology. 2019 Jun 1;115(4):310–9. doi:10.1159/000495133 PubMed PMID: 30836372. 3. Kajantie E, Johnson S, Heinonen K, Anderson PJ, Wolke D, Evensen KAI, et al. Common Core Assessments in follow-up studies of adults born preterm-Recommendation of the Adults Born Preterm International Collaboration. Paediatr Perinat Epidemiol. 2021 May 1;35(3):371–87. doi:10.1111/PPE.12694 PubMed PMID: 32990377. 4. Crump C. An overview of adult health outcomes after preterm birth. Early Hum Dev. 2020 Nov 1;150. doi:10.1016/j.earlhumdev.2020.105187 PubMed PMID: 32948365. 5. Smyrni N, Koutsaki M, Petra M, Nikaina E, Gontika M, Strataki H, et al. Moderately and Late Preterm Infants: Short- and Long-Term Outcomes From a Registry-Based Cohort. Front Neurol. 2021 Feb 12;12. doi:10.3389/FNEUR.2021.628066 PubMed PMID: 33643206. 6. Pirnat A, DeRoo LA, Skjærven R, Morken NH. Risk of having one lifetime pregnancy and modification by outcome of pregnancy and perinatal loss. Acta Obstet Gynecol Scand. 2019 Jun 1;98(6):753–60. doi:10.1111/AOGS.13534 PubMed PMID: 30648732. 7. Alenius S, Kajantie E, Sund R, Näsänen-Gilmore P, Vääräsmäki M, Gissler M, et al. The Missing Siblings of Infants Born Preterm. Pediatrics. 2018 Jan 1;141(1). doi:10.1542/PEDS.2017-1354 PubMed PMID: 29247118. 8. Congdon JL, Trope LA, Bruce JS, Chung PJ, Dehlendorf C, Chamberlain LJ. Meeting the Needs of Postpartum Women With and Without a Recent Preterm Birth: Perceptions of Maternal Family Planning in Pediatrics. Matern Child Health J. 2020 Mar 1;24(3):378–88. doi:10.1007/S10995-019-02829-X PubMed PMID: 31875305. 9. Rivers A, Caron B, Hack M. Experience of Families with Very Low Birthweight Children with Neurologic Sequelae. Clin Pediatr (Phila). 1987;26(5):223–30. doi:10.1177/000992288702600502 PubMed PMID: 3568527. 10. Singer LT, Salvator A, Guo S, Collin M, Lilien L, Baley J. Maternal psychological distress and parenting stress after the birth of a very low-birth-weight infant. JAMA. 1999 Mar 3;281(9):799–805. doi:10.1001/jama.281.9.799 PubMed PMID: 10071000. 11. Saigal S, Burrows E, Stoskopf BL, Rosenbaum PL, Streiner D. Impact of extreme prematurity on families of adolescent children. Journal of Pediatrics. 2000;137(5):701–6. doi:10.1067/mpd.2000.109001 PubMed PMID: 11060538. 12. Cronin CMG, Shapiro CR, Casiro OG, Cheang MS. The impact of very low-birth-weight infants on the family is long lasting. A matched control study. Arch Pediatr Adolesc Med. 1995;149(2):151–8. doi:10.1001/ARCHPEDI.1995.02170140033005 PubMed PMID: 7849876. 13. Trause M, Kramer lloyd I. The effects of premature birth on parents and their relationship. Dev Med Child Neurol. 1983 Aug;25(4):459–65. doi:10.1111/J.1469-8749.1983.TB13790.X PubMed PMID: 6618024. 14. Sheng L, Yang B, Story M, Wu W, Xi X, Zhou Y, et al. Emotional and behavioral changes and related factors of firstborn school-aged compared to same age only children. Front Public Health. 2022 Mar 3;10:822761. doi:10.3389/FPUBH.2022.822761/FULL PubMed PMID: 35309215. 15. Downey D, Condron DJ. Playing well with others in kindergarten: The benefit of siblings at home. Journal of Marriage and Family. 2004 May;66(2):333–50. doi:10.1111/J.1741-3737.2004.00024.X 16. Silva W, Avraam D, Cederkvist L, Nader JL, Popovic M, Marroun H El, et al. Internalising and externalising behaviour in siblings of children born preterm. PLOS Mental Health. 2025 Jun 11;2(6):e0000334. doi:10.1371/JOURNAL.PMEN.0000334 17. Alam NN, Ali TT, Razzaque AA. Health and Demographic Surveillance System (HDSS) in Matlab, Bangladesh. [Internet]. 2017. doi:10.5555/20173380629 18. Rahman A, Rahman M, Pervin J, Razzaque A, Aktar S, Ahmed JU, et al. Time trends and sociodemographic determinants of preterm births in pregnancy cohorts in Matlab, Bangladesh, 1990-2014. BMJ Glob Health. 2019 Aug 1;4(4). doi:10.1136/BMJGH-2019-001462 PubMed PMID: 31423346. 19. Pervin J, Rahman SM, Rahman M, Aktar S, Rahman A. Association between antenatal care visit and preterm birth: a cohort study in rural Bangladesh. BMJ Open. 2020 Jul 23;10(7). doi:10.1136/bmjopen-2019-036699 PubMed PMID: 32709651. 20. World Health Organization. Making Every Baby Count: audit and review of stillbirths and neonatal deaths. WHO Library Cataloguing-in-Publication Data. 2016;144. 21. World Health Organization. Why we need to talk about losing a baby [Internet]. 2025 [cited 2025 Nov 10]. Available from: https://www.who.int/news-room/spotlight/why-we-need-to-talk-about-losing-a-baby 22. Vogel JP, Chawanpaiboon S, Moller AB, Watananirun K, Bonet M, Lumbiganon P. The global epidemiology of preterm birth. Best Pract Res Clin Obstet Gynaecol. 2018 Oct 1;52:3–12. doi:10.1016/j.bpobgyn.2018.04.003 PubMed PMID: 29779863. 23. Ip EH, Efendi A, Molenberghs G, Bertoni AG. Comparison of risks of cardiovascular events in the elderly using standard survival analysis and multiple-events and recurrent-events methods. BMC Med Res Methodol. 2015 Mar 8;15(1). doi:10.1186/S12874-015-0004-3 PubMed PMID: 25887387. 24. Prentice R, Williams B, AV Peterson. On the regression analysis of multivariate failure time data. Biometrika [Internet]. 1981 Aug [cited 2025 Nov 10];68. Available from: https://academic.oup.com/biomet/article-abstract/68/2/373/260402 25. Andersen P, Gill RD. Cox’s regression model for counting processes: a large sample study. The annals of statistics. 1982;10:1100–20. 26. Wei LJ, Lin DY, Weissfeld L. Regression analysis of multivariate incomplete failure time data by modeling marginal distributions. J Am Stat Assoc. 1989;84(408):1065–73. doi:10.1080/01621459.1989.10478873 27. Skjærven R, Melve KK. Selective fertility – the examples of perinatal death and preeclampsia. Norsk Epidemiologi. 2007 Oct;17(2):175–80. doi:10.5324/NJE.V17I2.159 28. Akram R, Sarker AR, Sheikh N, Ali N, Mozumder MGN, Sultana M. Factors associated with unmet fertility desire and perceptions of ideal family size among women in Bangladesh: Insights from a nationwide Demographic and Health Survey. PLoS One. 2020 May 1;15(5). doi:10.1371/JOURNAL.PONE.0233634 PubMed PMID: 32442230. 29. Plana-Ripoll O, Basso O, László KD, Olsen J, Parner E, Cnattingius S, et al. Reproduction after the loss of a child: A population-based matched cohort study. Human Reproduction. 2018 Aug 1;33(8):1557–65. doi:10.1093/humrep/dey233 PubMed PMID: 30010921. 30. Finnäs F, Rostila M, Saarela J. Divorce and parity progression following the death of a child: A register-based study from Finland. Popul Stud (NY). 2018 Jan 2;72(1):41–51. doi:10.1080/00324728.2017.1337918 PubMed PMID: 28789590. 31. Mencarini L, Vignoli D, Gottard A. Fertility intentions and outcomes. Implementing the Theory of Planned Behavior with graphical models. Adv Life Course Res. 2015;23:14–28. doi:10.1016/j.alcr.2014.12.004 PubMed PMID: 26047838. 32. Schoen R, Astone N, Kim Y, A.Nathanson C, N.Fields J. Do fertility intentions affect fertility behavior? J Marriage Fam [Internet]. 1999 Aug [cited 2025 Nov 10];61. Available from: https://www.jstor.org/stable/353578 33. Joshi S, Schultz P. Family planning and women’s and children’s health: Long-term consequences of an outreach program in Matlab, Bangladesh. Demography [Internet]. 2013 [cited 2025 Nov 10];50:149–80. Available from: https://read.dukeupress.edu/demography/article-abstract/50/1/149/169652 Supplementary Material File (figure 1.docx) Download 44.43 KB File (table 1. population characteristics.docx) Download 30.46 KB File (table 2. categorical analysis.docx) Download 17.90 KB File (table 3. categorical death1yoldtimedependent.docx) Download 16.79 KB Information & Authors Information Version history V1 Version 1 26 March 2026 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords epidemiology preterm labour: clinical research Authors Affiliations Wnurinham Silva 0000-0001-8129-890X [email protected] Oulun yliopisto View all articles by this author Monjur Rahman 0000-0002-7877-858X International Centre for Diarrhoeal Disease Research Bangladesh View all articles by this author Markku Nurhonen Oulun yliopisto View all articles by this author Syed Manzoor Ahmed Hanifi International Centre for Diarrhoeal Disease Research Bangladesh View all articles by this author Jesmin Pervin International Centre for Diarrhoeal Disease Research Bangladesh View all articles by this author Suvi Alenius Oulun yliopisto View all articles by this author Johanna Metsälä Oulun yliopisto View all articles by this author Sylvain Sebert Oulun yliopisto View all articles by this author Anisur Rahman International Centre for Diarrhoeal Disease Research Bangladesh View all articles by this author Eero Kajantie Oulun yliopisto View all articles by this author Metrics & Citations Metrics Article Usage 141 views 79 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Wnurinham Silva, Monjur Rahman, Markku Nurhonen, et al. Preterm birth and subsequent pregnancies in Matlab, Bangladesh: a recurrent survival analysis. Authorea . 26 March 2026. DOI: https://doi.org/10.22541/au.177452316.68521391/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.177452316.68521391/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fe1a24c4abead07',t:'MTc3OTE3ODMxOA=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();