[Skip to Content]
[Skip to Content Landing]
Figure 1.  Absolute Risk of Pregnancy and Birth Outcomes According to Interpregnancy Interval, British Columbia, 2004-2014 in 148 544 Pregnancies
Absolute Risk of Pregnancy and Birth Outcomes According to Interpregnancy Interval, British Columbia, 2004-2014 in 148 544 Pregnancies

Unadjusted and adjusted predicted risks (95% CIs) at each interpregnancy interval length from 3 to 24 months for maternal mortality or severe morbidity (A), adverse fetal and infant composite (B), small-for-gestational age (<10th percentile) (C), spontaneous preterm delivery (D), and indicated preterm delivery (E).

Figure 2.  Absolute Risk of Pregnancy and Birth Outcomes According to Interpregnancy Interval Stratified by Maternal Age at Index Birth, British Columbia, 2004-2014 in 141 360 Pregnancies
Absolute Risk of Pregnancy and Birth Outcomes According to Interpregnancy Interval Stratified by Maternal Age at Index Birth, British Columbia, 2004-2014 in 141 360 Pregnancies

Adjusted predicted absolute risks (95% CIs) of adverse maternal and infant outcome at each interpregnancy interval length from 3 to 24 months according to maternal age category for maternal mortality or severe morbidity (A), adverse fetal and infant composite (B), small-for-gestational age (<10th percentile) (C), spontaneous preterm delivery (D), and indicated preterm delivery (E).

Table 1.  Sociodemographic and Clinical Characteristics of the Study Population According to Interpregnancy Interval Category in 148 544 Pregnancies
Sociodemographic and Clinical Characteristics of the Study Population According to Interpregnancy Interval Category in 148 544 Pregnancies
Table 2.  Incidence of Outcomes According to Categories of Interpregnancy Interval, Stratified by Maternal Age at Index Birth in 148 544 Pregnancies
Incidence of Outcomes According to Categories of Interpregnancy Interval, Stratified by Maternal Age at Index Birth in 148 544 Pregnancies
Table 3.  Adjusted Risk Ratios (aRRs) Comparing Estimated Risks of Each Outcome at 3-, 6-, 9-, and 12-Month Interpregnancy Intervals With Estimated Risks Following 18-Month Intervals in 148 544 Pregnanciesa
Adjusted Risk Ratios (aRRs) Comparing Estimated Risks of Each Outcome at 3-, 6-, 9-, and 12-Month Interpregnancy Intervals With Estimated Risks Following 18-Month Intervals in 148 544 Pregnanciesa
Supplement.

eTable 1. Detailed Variable Definitions

eAppendix 1. Methods for Sensitivity Analyses for Unmeasured or Residual Confounding

eFigure 1. Distribution of Interpregnancy Interval Length by Categories of Maternal Age at Index Birth

eTable 2. Sociodemographic and Clinical Characteristics of the Study Population According to Post-interval Pregnancy Resolution Type

eTable 3. Incidence of Component Outcomes of the Adverse Fetal-Infant Composite Outcome According to Categories of Interpregnancy Interval, Stratified by Maternal Age and Index Birth

eFigure 2. Absolute Risk of Component Outcomes of Adverse Fetal-Infant Composite According to Interpregnancy Interval

eFigure 3. Absolute Risk of Component Outcomes of Adverse Fetal-Infant Composite According to Interpregnancy Interval, Stratified by Maternal Age at Index Birth

eFigure 4. Absolute Risk of Pregnancy and Birth Outcomes According to Interpregnancy Interval Stratified by Maternal Age at Index Birth (20-30 and ≥30), British Columbia, 2004-2014 (n = 138 005).

eAppendix 2. Sensitivity Analyses for Unmeasured and Residual Confounding

eFigure 5. Absolute Risk of Pregnancy and Birth Outcomes According to Interpregnancy Interval Among Women With Infertility Diagnosis or Treatment (n=16,336).

eFigure 6. Absolute Risk of Pregnancy and Birth Outcomes According to Interpregnancy Interval Among Women With Infertility Diagnosis Or Treatment Stratified By Maternal Age At Index Birth (n=16,207

eTable 4. E-Values to Quantify the Minimum Strength of Association That Unmeasured Confounders Would Need to Have With Both Interpregnancy Interval and Each Outcome in Order to Fully Explain the Observed Association Between Short Interpregnancy Interval and Adverse Outcomes

eFigure 7. Absolute Risk of Pregnancy and Birth Outcomes in the Pre-Interval (Index) Birth According to Interpregnancy Interval: a Negative Control Analysis

eFigure 8. Absolute Risk of Pregnancy and Birth Outcomes in the Pre-Interval (Index) Birth According to Interpregnancy Interval, Stratified by Maternal Age at Index Birth: a Negative Control Analysis

1.
Conde-Agudelo  A, Rosas-Bermúdez  A, Kafury-Goeta  AC.  Birth spacing and risk of adverse perinatal outcomes: a meta-analysis.  JAMA. 2006;295(15):1809-1823. doi:10.1001/jama.295.15.1809PubMedGoogle ScholarCrossref
2.
Merklinger-Gruchala  A, Jasienska  G, Kapiszewska  M.  Short interpregnancy interval and low birth weight: a role of parity.  Am J Hum Biol. 2015;27(5):660-666. doi:10.1002/ajhb.22708PubMedGoogle ScholarCrossref
3.
McKinney  D, House  M, Chen  A, Muglia  L, DeFranco  E.  The influence of interpregnancy interval on infant mortality.  Am J Obstet Gynecol. 2017;216(3):316.e1-316.e9. doi:10.1016/j.ajog.2016.12.018PubMedGoogle ScholarCrossref
4.
Smith  GCS, Pell  JP, Dobbie  R.  Interpregnancy interval and risk of preterm birth and neonatal death: retrospective cohort study.  BMJ. 2003;327(7410):313. doi:10.1136/bmj.327.7410.313PubMedGoogle ScholarCrossref
5.
Wendt  A, Gibbs  CM, Peters  S, Hogue  CJ.  Impact of increasing inter-pregnancy interval on maternal and infant health.  Paediatr Perinat Epidemiol. 2012;26(1)(suppl 1):239-258. doi:10.1111/j.1365-3016.2012.01285.xPubMedGoogle ScholarCrossref
6.
Zhang  L, Shen  S, He  J,  et al.  Effect of interpregnancy interval on adverse perinatal outcomes in southern China: a retrospective cohort study, 2000-2015.  Paediatr Perinat Epidemiol. 2018;32(2):131-140. doi:10.1111/ppe.12432PubMedGoogle ScholarCrossref
7.
American College of Obstetricians and Gynecologists.  ACOG Committee opinion 736: optimizing postpartum care.  Obstet Gynecol. 2018;131(5):e140-e150. doi:10.1097/AOG.0000000000002633PubMedGoogle ScholarCrossref
8.
Royal College of Obstetricians and Gynaecologists. Best practice in postpartum family planning. Vol. Best Practice Paper No 1. London, UK: Royal College of Obstetricians and Gynaecologists. https://www.rcog.org.uk/globalassets/documents/guidelines/best-practice-papers/best-practice-paper-1---postpartum-family-planning.pdf. Published June 2015. Accessed March 12, 2018.
9.
Office of Disease Prevention and Health Promotion. Healthy People.gov. US Healthy people 2020 family planning objective 5. Healthy People 2020 Web site. https://www.healthypeople.gov/2020/topics-objectives/topic/family-planning/objectives. Updated September 7, 2018. Accessed March 3, 2018.
10.
World Health Organization. Report of a WHO technical consultation on birth spacing. Geneva, Switzerland, 13-15, 2005. http://apps.who.int/iris/bitstream/10665/69855/1/WHO_RHR_07.1_eng.pdf. Published 2007. Accessed March 3, 2018.
11.
Gemmill  A, Lindberg  LD.  Short interpregnancy intervals in the United States.  Obstet Gynecol. 2013;122(1):64-71. doi:10.1097/AOG.0b013e3182955e58PubMedGoogle ScholarCrossref
12.
Cheslack Postava  K, Winter  AS.  Short and long interpregnancy intervals: correlates and variations by pregnancy timing among US women.  Perspect Sex Reprod Health. 2015;47(1):19-26. doi:10.1363/47e2615PubMedGoogle ScholarCrossref
13.
Lean  SC, Derricott  H, Jones  RL, Heazell  AEP.  Advanced maternal age and adverse pregnancy outcomes: a systematic review and meta-analysis.  PLoS One. 2017;12(10):e0186287. doi:10.1371/journal.pone.0186287PubMedGoogle ScholarCrossref
14.
Zhu  BP, Rolfs  RT, Nangle  BE, Horan  JM.  Effect of the interval between pregnancies on perinatal outcomes.  N Engl J Med. 1999;340(8):589-594. doi:10.1056/NEJM199902253400801PubMedGoogle ScholarCrossref
15.
Finer  LB, Zolna  MR.  Declines in unintended pregnancy in the United States, 2008-2011.  N Engl J Med. 2016;374(9):843-852. doi:10.1056/NEJMsa1506575PubMedGoogle ScholarCrossref
16.
de Weger  FJ, Hukkelhoven  CW, Serroyen  J, te Velde  ER, Smits  LJ.  Advanced maternal age, short interpregnancy interval, and perinatal outcome.  Am J Obstet Gynecol. 2011;204(5):421.e1-421.e9. doi:10.1016/j.ajog.2010.12.008PubMedGoogle ScholarCrossref
17.
Perinatal Services BC. Perinatal Data Registry; 2017. Available at: http://www.perinatalservicesbc.ca/health-professionals/data-surveillance/perinatal-data-registry. Updated 2018. Accessed March 12, 2018.
18.
Population Data BC. Medical Services Plan (MSP) payment information file. V2; 2016. Data extract. Ministry of Health (2016). https://www.popdata.bc.ca/data/health/msp. Updated May 9, 2018. Accessed June 10, 2018.
19.
Population Data BC. PharmaNet. V2. BC Ministry of Health. Data extract. Data Stewardship Committee (2017). https://www.popdata.bc.ca/data/health/PharmaNet. Updated May 9, 2018. Accessed June 10, 2018.
20.
Population Data BC. Canadian Institute for Health Information. Discharge abstract database (hospital separations). V2. Data extract. 2015. Ministry of Health (2015). https://www.popdata.bc.ca/data/health/dad. Updated May 9, 2018. Accessed July 12, 2018.
21.
Population Data BC. BC Ministry of Health (2015). Consolidation file. V2. Data Extract. Ministry of Health (2016). https://www.popdata.bc.ca/data/population/consolidationfile. Updated May 9, 2018. Accessed May 12, 2018.
22.
Population Data BC. Statistics Canada. Small Area and Administrative Data Division (2009). Income band data; catalogue number 13C0016. Data extract (2017). https://www.popdata.bc.ca/data/demographic/incomeband. Updated May 9, 2018. Accessed June 12, 2018.
23.
Population Data BC. BC Vital Statistics Agency (2015). Vital statistics deaths. V2. Data extract. BC Vital Statistics Agency (2016). https://www.popdata.bc.ca/data/population/vsdeaths. Updated May 9, 2018. Accessed June 17, 2018.
24.
Love  ER, Bhattacharya  S, Smith  NC, Bhattacharya  S.  Effect of interpregnancy interval on outcomes of pregnancy after miscarriage: retrospective analysis of hospital episode statistics in Scotland.  BMJ. 2010;341:c3967. doi:10.1136/bmj.c3967PubMedGoogle ScholarCrossref
25.
Geller  SE, Rosenberg  D, Cox  S, Brown  M, Simonson  L, Kilpatrick  S.  A scoring system identified near-miss maternal morbidity during pregnancy.  J Clin Epidemiol. 2004;57(7):716-720. doi:10.1016/j.jclinepi.2004.01.003PubMedGoogle ScholarCrossref
26.
Kramer  MS, Platt  RW, Wen  SW,  et al; Fetal/Infant Health Study Group of the Canadian Perinatal Surveillance System.  A new and improved population-based Canadian reference for birth weight for gestational age.  Pediatrics. 2001;108(2):E35. doi:10.1542/peds.108.2.e35PubMedGoogle ScholarCrossref
27.
Orsini  N, Greenland  S.  A procedure to tabulate and plot results after flexible modeling of a quantitative covariate.  Stata J. 2011;11(1):1-29.Google Scholar
28.
Harrell  FE.  Regression Modeling Strategies with Applications to Linear Models, Logistic Regression, and Survival Analysis. New York, NY: Springer; 2001.
29.
Altman  DG, Bland  JM.  Interaction revisited: the difference between two estimates.  BMJ. 2003;326(7382):219. doi:10.1136/bmj.326.7382.219PubMedGoogle ScholarCrossref
30.
Carlin  JB, Galati  J, Royston  P.  A new framework for managing and analyzing multiply imputed data in Stata.  Stata J. 2008;8(1):49-67.Google Scholar
31.
Corp  S. Stata statistical software: Release 14. College Station, TX: StataCorp, LP; 2015.
32.
Hall  JA, Benton  L, Copas  A, Stephenson  J.  Pregnancy intention and pregnancy outcome: systematic review and meta-analysis.  Matern Child Health J. 2017;21(3):670-704. doi:10.1007/s10995-016-2237-0PubMedGoogle ScholarCrossref
33.
Maxwell  L, Nandi  A, Benedetti  A, Devries  K, Wagman  J, García-Moreno  C.  Intimate partner violence and pregnancy spacing: results from a meta-analysis of individual participant time-to-event data from 29 low-and-middle-income countries.  BMJ Glob Health. 2018;3(1):e000304. doi:10.1136/bmjgh-2017-000304PubMedGoogle ScholarCrossref
34.
Donovan  BM, Spracklen  CN, Schweizer  ML, Ryckman  KK, Saftlas  AF.  Intimate partner violence during pregnancy and the risk for adverse infant outcomes: a systematic review and meta-analysis.  BJOG. 2016;123(8):1289-1299. doi:10.1111/1471-0528.13928PubMedGoogle ScholarCrossref
35.
VanderWeele  TJ, Ding  P.  Sensitivity analysis in observational research: introducing the E-value.  Ann Intern Med. 2017;167(4):268-274. doi:10.7326/M16-2607PubMedGoogle ScholarCrossref
36.
Lipsitch  M, Tchetgen  E, Cohen  T.  Negative controls: a tool for detecting confounding and bias in observational studies.  Epidemiology. 2010;21(3):383-388. doi:10.1097/EDE.0b013e3181d61eebPubMedGoogle ScholarCrossref
37.
Stamilio  DM, DeFranco  E, Paré  E,  et al.  Short interpregnancy interval: risk of uterine rupture and complications of vaginal birth after cesarean delivery.  Obstet Gynecol. 2007;110(5):1075-1082. doi:10.1097/01.AOG.0000286759.49895.46PubMedGoogle ScholarCrossref
38.
Ananth  CV, Skjaerven  R, Klunssoyr  K.  Change in paternity, risk of placental abruption and confounding by birth interval: a population-based prospective cohort study in Norway, 1967-2009.  BMJ Open. 2015;5(2):e007023. doi:10.1136/bmjopen-2014-007023PubMedGoogle ScholarCrossref
39.
Conde-Agudelo  A, Belizán  JM.  Maternal morbidity and mortality associated with interpregnancy interval: cross sectional study.  BMJ. 2000;321(7271):1255-1259. doi:10.1136/bmj.321.7271.1255PubMedGoogle ScholarCrossref
40.
Class  QA, Rickert  ME, Oberg  AS,  et al.  Within-family analysis of interpregnancy interval and adverse birth outcomes.  Obstet Gynecol. 2017;130(6):1304-1311. doi:10.1097/AOG.0000000000002358PubMedGoogle ScholarCrossref
41.
Ball  SJ, Pereira  G, Jacoby  P, de Klerk  N, Stanley  FJ.  Re-evaluation of link between interpregnancy interval and adverse birth outcomes: retrospective cohort study matching two intervals per mother.  BMJ. 2014;349:g4333. doi:10.1136/bmj.g4333PubMedGoogle ScholarCrossref
42.
Hanley  GE, Hutcheon  JA, Kinniburgh  BA, Lee  L.  Interpregnancy interval and adverse pregnancy outcomes: an analysis of successive pregnancies.  Obstet Gynecol. 2017;129(3):408-415. doi:10.1097/AOG.0000000000001891PubMedGoogle ScholarCrossref
43.
Frosst  G, Hutcheon  J, Joseph  KS, Kinniburgh  B, Johnson  C, Lee  L.  Validating the British Columbia Perinatal Data Registry: a chart re-abstraction study.  BMC Pregnancy Childbirth. 2015;15:123. doi:10.1186/s12884-015-0563-7PubMedGoogle ScholarCrossref
44.
Conzuelo-Rodriguez  G, Naimi  AI.  The impact of computing interpregnancy intervals without accounting for intervening pregnancy events.  Paediatr Perinat Epidemiol. 2018;32(2):141-148. doi:10.1111/ppe.12458PubMedGoogle ScholarCrossref
45.
Naimi  AI.  Studying interpregnancy interval effects using observational data: some cautionary remarks.  BJOG. 2016;123(8):1319. doi:10.1111/1471-0528.13743PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    1 Comment for this article
    Postpartum Re-establishment of DHEA and Testosterone Reduces Negative Pregnancy Outcomes
    James Howard, BS | Biologist
    I suggest the findings of Schummers, et al., may be explained by reductions in dehydroepiandrosterone (DHEA) and testosterone. Both of these androgens are reduced postpartum. These reductions may be the bases of postpartum negative outcomes. Sufficient time to re-establish these androgens, especially DHEA, would decrease the probability of negative outcomes.

    It is my hypothesis that human evolution is driven by increases in testosterone levels because it increases intracellular availability of DHEA. ("Androgens in Human Evolution," Rivista di Biologia / Biology Forum 2001; 94: 345-362. This article can be found at: http://anthropogeny.com/Androgens%20in%20Human%20Evolution.htm .)
    CONFLICT OF INTEREST: None Reported
    Original Investigation
    December 2018

    Association of Short Interpregnancy Interval With Pregnancy Outcomes According to Maternal Age

    Author Affiliations
    • 1Department of Epidemiology, Harvard T. H. Chan School of Public Health, Boston, Massachusetts
    • 2Department of Family Practice, University of British Columbia, Vancouver, British Columbia, Canada
    • 3Department of Obstetrics and Gynaecology, University of British Columbia, Vancouver, British Columbia, Canada
    • 4Department of Biostatistics, Harvard T. H. Chan School of Public Health, Boston, Massachusetts
    • 5Department of Obstetrics and Gynecology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts
    JAMA Intern Med. 2018;178(12):1661-1670. doi:10.1001/jamainternmed.2018.4696
    Key Points

    Question  Is the association of short interpregnancy interval with pregnancy outcomes modified by maternal age?

    Findings  In this cohort study of 148 544 pregnancies, maternal mortality or severe morbidity risks were increased at short interpregnancy intervals among women 35 years or older but not for women aged 20 to 34 years; in contrast, increased risks of adverse fetal and infant outcomes and spontaneous preterm delivery were more pronounced for women aged 20 to 34 years than for those 35 years and older. Modest increases in risks of small-for-gestational-age birth and indicated preterm delivery at short intervals were not meaningfully different across maternal age groups.

    Meaning  Short interpregnancy intervals appear to be associated with increased risks for adverse pregnancy outcomes for women of all ages; maternal risks at short intervals may be greater for older women, whereas fetal and infant risks may be greater for younger women.

    Abstract

    Importance  Interpregnancy intervals shorter than 18 months are associated with higher risks of adverse pregnancy outcomes. It is currently unknown whether short intervals are associated with increased risks among older women to the same extent as among younger women.

    Objective  To evaluate whether the association between short interpregnancy (delivery to conception) interval and adverse pregnancy outcomes is modified by maternal age.

    Design, Setting, and Participants  A population-based cohort study conducted in British Columbia, Canada, evaluated women with 2 or more singleton pregnancies from 2004 to 2014 with the first (index) pregnancy resulting in a live birth. Data analysis was performed from January 1 to July 20, 2018.

    Main Outcomes and Measures  Risks of maternal mortality or severe morbidity (eg, mechanical ventilation, blood transfusion >3 U, intensive care unit admission, organ failure, death), small-for-gestational age (<10th birthweight percentile for gestational age and sex), fetal and infant composite outcome (stillbirth, infant death, <third birthweight percentile for gestational age and sex, delivery <28 weeks), and spontaneous and indicated preterm delivery. Risks of each outcome for 3- to 24-month interpregnancy intervals were estimated, according to maternal age at index birth (20-34 and ≥35 years). Adjusted risk ratios (aRRs) comparing predicted risks at 3-, 6-, 9-, and 12-month intervals with risks at 18-month intervals for each age group were calculated. The potential role of other factors explaining any differences (unmeasured confounding) was examined in several sensitivity analyses.

    Results  Among 148 544 pregnancies, maternal mortality or severe morbidity risks were increased at 6-month compared with 18-month interpregnancy intervals for women aged 35 years or older (0.62% at 6 months vs 0.26% at 18 months; aRR, 2.39; 95% CI, 2.03-2.80), but not for women aged 20 to 34 years (0.23% at 6 months vs 0.25% at 18 months; aRR, 0.92; 95% CI, 0.83-1.02). Increased adverse fetal and infant outcome risks were more pronounced for women aged 20 to 34 years (2.0% at 6 months vs 1.4% at 18 months; aRR, 1.42; 95% CI, 1.36-1.47) than women 35 years or older (2.1% at 6 months vs 1.8% at 18 months; aRR, 1.15; 95% CI, 1.01-1.31). Risks of spontaneous preterm delivery at 6-month interpregnancy intervals were increased for women 20 to 34 years old (5.3% at 6 months vs 3.2% at 18 months; aRR, 1.65; 95% CI, 1.62-1.68) and to a lesser extent for women 35 years or older (5.0% at 6 months vs 3.6% at 18 months; aRR, 1.40; 95% CI, 1.31-1.49). Modest increases in risks of small-for-gestational age and indicated preterm delivery at short intervals did not vary meaningfully by maternal age. Sensitivity analyses suggested that observed associations were not fully explained by unmeasured confounding.

    Conclusions and Relevance  The findings of this study suggest that short interpregnancy intervals are associated with increased risks for adverse pregnancy outcomes for women of all ages.

    Introduction

    Interpregnancy intervals (intervals between delivery and conception of the subsequent pregnancy) shorter than 18 months appear to be associated with increased risks of adverse outcomes, including preterm delivery, small-for-gestational-age (SGA) birth, and infant mortality.1-6 Based on these associations, clinical postpartum practice7,8 and public health9,10 guidelines recommend interpregnancy intervals of at least 18 to 24 months.

    However, it is currently unknown whether short intervals are associated with increased risks among older women to the same extent as among younger women. In the United States, women with a first birth at age 30 years or older have shorter interpregnancy intervals compared with women who began childbearing at younger ages.11,12 Older maternal age is associated with increased reproductive birth risks, including infertility, chromosomal anomalies, and adverse birth outcomes.13 Women who have delayed childbearing must balance the risks associated with a shorter interpregnancy interval against the risks of older maternal age at conception if they delay subsequent pregnancies to achieve the recommended interpregnancy interval.

    Pregnancy and maternal characteristics associated with short interpregnancy interval may vary by maternal age (eg, unintended pregnancy, lower socioeconomic position, perinatal loss in the first pregnancy).14,15 Therefore, the association between short interpregnancy interval and adverse outcomes may vary by maternal age such that short intervals might not be associated with increased risks in older women. With limited previous work on this question,16 the extent to which the associations between short interpregnancy interval and adverse pregnancy outcomes vary by maternal age is poorly understood. The objective of this study was to evaluate whether the associations between short interpregnancy interval and pregnancy outcomes vary by maternal age.

    Methods
    Study Population

    This population-based cohort was drawn from all women with at least 2 consecutive singleton pregnancies in British Columbia, Canada, from April 1, 2004, to March 31, 2014, using population-based health databases. Data analysis was performed from January 1 to July 20, 2018. Our primary database was the British Columbia Perinatal Data Registry,17 maintained by Perinatal Services British Columbia, which contains demographic and clinical data abstracted from obstetrical and newborn medical records for nearly 100% of live or stillbirth births in British Columbia of 20 or more completed weeks’ gestation or birthweight of 500 g or more, including maternal postpartum readmissions up to 42 days after delivery and baby transfers and readmissions up to 28 days post birth. Population Data BC, a multi-university data platform that provides researchers with deidentified linkages of British Columbia’s administrative data, linked the British Columbia Perinatal Data Registry with additional population-level databases, including physician billing,18 outpatient prescription,19 and hospital discharge records,20 census and community size data,21,22 and vital statistics records.23 The University of British Columbia/Children’s and Women’s Health Centre of British Columbia Research Ethics Board approved this study with waiver of informed consent.

    Live births and stillbirths of 20 or more completed weeks’ gestation were identified using the British Columbia Perinatal Data Registry. Pregnancies ending before 20 weeks’ gestation were identified using physician billing, hospital discharge, or outpatient prescription codes for spontaneous or induced abortion (codes in eTable 1 in the Supplement). We excluded postinterval (subsequent) pregnancies if the preinterval pregnancy ended in a spontaneous or induced abortion at less than 20 weeks’ gestation, stillbirth, or neonatal death, because the association between interpregnancy interval and outcomes in the subsequent pregnancy may be different following a live birth than following a perinatal or neonatal loss,24 and guidelines differ for pregnancy spacing following a loss.10 Women with more than 2 eligible interpregnancy intervals in the study period could contribute more than 1 observation for the analyses.

    Exposure and Outcome Definitions

    We defined interpregnancy interval as the time in months between the first eligible birth in the study period (the index birth) and the conception of the subsequent pregnancy. To calculate this interval, we subtracted the index birth date from the subsequent birth date minus the gestational age at birth of the subsequent pregnancy (using an algorithm-based estimate of gestational age based on last menstrual period, first ultrasonography [<20 weeks], clinical estimate from newborn examination, and documentation from maternal record). For subsequent pregnancies that ended in spontaneous or induced abortion at less than 20 weeks, we assumed the diagnosis, treatment, or hospital admission date to be the pregnancy loss date, and imputed gestational age at 12 weeks (97% were missing a gestational age estimate).

    Outcomes included (1) maternal mortality or severe morbidity (mechanical ventilation, intensive care unit admission, organ failure, blood transfusion >3 U, unanticipated postpartum surgical procedure, or death) through pregnancy up to 42 days postpartum,25 (2) SGA birth (<10th percentile for sex and gestational age),26 (3) fetal and infant composite (stillbirth, infant death within 1 year after birth, extreme SGA birth [<3rd percentile], delivery <28 weeks), (4) spontaneous preterm delivery at less than 37 weeks, and (5) indicated preterm delivery at less than 37 weeks. Outcomes were defined using International Classification of Disease, Ninth Edition or International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Canadian Classification of Interventions, Version 9, and British Columbia Medical Service Plan billing codes. Detailed variable definitions are provided in eTable 1 in the Supplement.

    Statistical Analysis

    We examined the association between interpregnancy interval and each outcome in the overall population and stratified by categories of maternal age at index birth: younger than 20 years, 20 to 34 years, and 35 years or older. First treating interpregnancy interval as a categorical variable, we tabulated the incidence of each outcome by interpregnancy interval in 6-month categories (<6, 6-11, 12-17, 18-23, and ≥24 months). Next, we used logistic regression and postestimation calculations27 to estimate predicted odds of each outcome in 1-month increments of interpregnancy interval from 3 to 24 months, which we then transformed to predicted absolute risks. To allow for curvilinear shapes, we modeled interpregnancy interval as a continuous variable using restricted cubic splines, with 4 knots at the 5th, 35th, 65th, and 95th percentiles of interpregnancy interval in the study population (5.7, 15.3, 24.8, and 55.3 months).28 For each outcome, we fit unadjusted models including only interpregnancy interval spline terms and adjusted models also including baseline (ie, measured at or before the index birth) covariates identified as potential confounders of the association between short interpregnancy interval and adverse maternal and infant outcomes: nulliparity, smoking before or during index pregnancy, low neighborhood income, inadequate prenatal care, and outcome of index pregnancy (eg, preterm delivery in index pregnancy in model for preterm delivery). In unstratified models using the overall population, we also adjusted for maternal age. Adjusted risks were standardized to population average covariate values based on interpregnancy intervals of 18 months. We examined variance inflation factors to assess multicollinearity between variables included in adjusted models and considered values of 3 or higher as potentially concerning.

    We plotted absolute predicted risks with 95% CIs in the overall population and separately for each age category to display the associations visually. We then used predicted absolute risks to calculate risk ratios (RRs) with 95% CIs comparing predicted risks at 3-, 6-, 9-, and 12-month interpregnancy intervals with predicted risks at the 18-month interval reference length, at population mean covariate values based on interpregnancy intervals of 18 months. In age-stratified analyses, we estimated RRs separately for each category of maternal age and performed a test for interaction for each outcome at each interval length.29 Findings were considered significant at P < .05 (2-sided).

    We used robust variance estimation to account for nonindependence of 2 or more interpregnancy intervals in the same woman. Analyses for all outcomes other than maternal mortality or severe morbidity were restricted to births at 20 weeks’ or more gestation, as pregnancies ending earlier than 20 weeks were not at risk of birth outcomes. Analyses for SGA birth were restricted to those 22 to 43 weeks, owing to availability of reference charts. We excluded 616 records of pregnancies of 20 weeks or more owing to missing gestational age. We imputed missing values for number of antenatal visits (10 382 [7.0%]), using multiple imputation with 30 imputations based on all covariates and outcomes combined according to the Rubin formula.30 For individuals with missing neighborhood income values for the index birth year (7857 [5.3%]), we assigned values from proximal years if available (7654 [5.2%]) and imputed missing values for the remaining 203 (0.1%). All analyses were conducted using Stata, version 14.2 (StataCorp).31

    Sensitivity Analyses

    To assess whether our results were sensitive to age categorization thresholds, we repeated analyses for women younger than 30 years and 30 years or older and for women younger than 40 years and 40 years or older. We conducted several analyses to examine the potential role of unmeasured and residual confounding. We (1) adjusted for additional measured baseline clinical and sociodemographic covariates that could be potential confounding variables (eg, rural residence, obesity, infertility treatment); (2) assessed potential confounding by pregnancy intention by restricting to pregnancies following an infertility diagnosis on the assumption that these pregnancies were more likely to be intended, (3) calculated E values, which represent the minimum strength of association that unmeasured confounding variables (eg, pregnancy intention,32 intimate partner violence33,34) would need to have with both interpregnancy interval and each outcome to fully explain the observed association35; and (4) used the outcome of the index pregnancy as a negative control, on the assumption that no association between the outcome of the index pregnancy and interpregnancy interval could provide evidence against unmeasured confounding.36 These approaches are detailed further in eAppendix 1 in the Supplement. To ensure that our findings were not sensitive to our methodologic approach of imputing missing data, we compared estimates using multiple imputation with those applying simple imputation (mean or missing indicator) and a complete case approach.

    Results

    Our study population included 123 122 women and 148 544 pregnancies subsequent to a live index birth (some women contributed more than 1 interpregnancy interval). Of these, 121 242 pregnancies (81.6%) continued to 20 weeks’ or more gestation and the remaining 27 302 (18.4%) ended in spontaneous or induced abortion at less than 20 weeks. The majority of pregnancies (123 821 [83.4%]) were in women between ages 20 and 34 years at the index (preinterval) birth, while 7184 (4.8%) were in women younger than 20 years and 17 539 (11.8%) were in women 35 years or older at the index birth.

    The distribution of interpregnancy interval differed somewhat by maternal age at index birth (eFigure 1 in the Supplement). Interpregnancy intervals shorter than 6 months were less common among women 35 years or older at the index birth than women aged 20 to 34 years (4.4% vs 5.5%). Women 35 years or older more often had interpregnancy intervals 6 to 11 and 12 to 17 months long than women aged 20 to 34 years (6-11 months, 17.7% vs 16.6% and 12-17 months, 25.2% vs 22.5%) but had fewer intervals 24 months or longer (32.8% vs 37.6%).

    Sociodemographic and clinical characteristics of the study population at the time of index birth according to interpregnancy interval categories are reported in Table 1. Low neighborhood income, inadequate prenatal care, and smoking in pregnancy were most common among women with interpregnancy intervals shorter than 6 months. Women with interpregnancy intervals shorter than 6 months were also more likely to have a history of stillbirth, neonatal death, preterm delivery, or spontaneous abortion. In addition, pregnancies following intervals shorter than 6 months were much more likely to end in spontaneous or indicated abortion at less than 20 weeks. As reported in eTable 2 in the Supplement, maternal age younger than 20 years, smoking, inadequate prenatal care, and low neighborhood income also were associated with spontaneous or indicated abortion at less than 20 weeks.

    The incidence of each outcome according to 6-month categories of interpregnancy interval for all women, and separately for those younger than 20 years, 20 to 34 years, and 35 years or older at index birth, is reported in Table 2. Figure 1 displays the unadjusted and adjusted predicted risk curves for each outcome according to 1-month increments of interpregnancy interval from 3 to 24 months. Risk curves were U- or L-shaped for all outcomes, with nadirs near 18 months, and similarly low risks for most outcomes from 12- to 24-month intervals. Adjusted RRs (aRRs) comparing predicted risks at 3-, 6-, 9-, and 12-month intervals with predicted risks at the referent 18-month interval length are tabulated in Table 3.

    In the overall population, interpregnancy intervals of less than 12 months were associated with modest increases in risk of maternal mortality or severe morbidity (predicted risk, 0.30% at 6 months and 0.25% at 18 months; aRR comparing 6-month with 18-month intervals: 1.18; 95% CI, 1.09-1.27) and larger increases in risk of adverse fetal and infant outcomes (predicted risk, 2.0% at 6 months compared with 1.5% at 18 months: aRR, 1.36; 95% CI, 1.31-1.40), and spontaneous preterm delivery (predicted risk, 5.3% at 6 months compared with 3.3% at 18 months: aRR, 1.59; 95% CI, 1.56-1.2). Risk of SGA birth and indicated preterm delivery were only slightly higher at 6-month intervals (aRRs, 1.11; 95% CI, 1.08-1.14 and 1.08; 95% CI, 1.04-1.13, respectively).

    Figure 2 presents adjusted risk curves after stratification by maternal age category (20-34 years and 35 years or older at index birth). Due to the small number of index pregnancies in women younger than 20 years, predicted risks for this group were unstable and are not presented. Risks for maternal mortality or severe morbidity differed meaningfully by age group. We found increased risks of maternal mortality or severe morbidity at 6-month compared with 18-month interpregnancy intervals for women 35 years or older (predicted risk of 0.62% at 6 months compared with 0.26% at 18 months: aRR, 2.39; 95% CI, 2.03-2.80), but not for women aged 20 to 34 years (predicted risk of 0.23% at 6 months compared with 0.25% at 18 months: aRR, 0.92; 95% CI, 0.83-1.02). Conversely, the increased risk of spontaneous preterm delivery at short interpregnancy intervals was more pronounced for younger women (predicted risk of 5.3% at 6 months compared with 3.2% at 18 months: aRR, 1.65; 95% CI, 1.62-1.68 for women 20-34 years at index birth) than older women (predicted risk of 5.0% at 6 months compared with 3.6% at 18 months: aRR, 1.40; 95% CI, 1.31-1.49 for women 35 years or older at index birth). Similarly, increased risk of adverse fetal and infant outcomes was more pronounced for women 20 to 34 years of age (predicted risk of 2.0% at 6 months vs 1.4% at 18 months: aRR, 1.42; 95% CI, 1.36-1.47) than women 35 years or older (predicted risk of 2.1% at 6 months vs 1.8% at 18 months: aRR, 1.15; 95% CI, 1.01-1.31, respectively). The shapes of the curves were similar across age groups for the other outcomes, generally with the lowest risk at 18-month intervals, although not substantially higher at 12- or 24-month intervals. Women younger than 20 years at index birth had the highest risks of spontaneous preterm delivery across the interpregnancy interval spectrum and similar risks of other outcomes. Incidences and curves for component outcomes of the fetal and infant composite were similar to those of the composite and are presented in eTable 3, eFigures 2 and 3 in the Supplement.

    Adjustment did not meaningfully change estimates for maternal mortality or severe morbidity. Increases in adverse fetal and infant composite, SGA birth, and spontaneous preterm delivery risks at short intervals were attenuated after adjustment. Aside from expected multicollinearity between cubic spline terms,28 no variance inflation factors exceeded 2, indicating no problematic multicollinearity.

    Sensitivity Analyses

    Curves were similar when older maternal age was defined as 30 years or older (eFigure 4 in the Supplement). Because of the small number of women 40 years or older at index birth (n = 896), findings using this threshold were challenging to interpret owing to wide 95% CIs and are not presented. Sensitivity analyses examining the role of unmeasured or residual confounding indicated that confounding was unlikely to entirely explain our observed results (eAppendix 2 in the Supplement). Curves were similar after restricting to pregnancies following an infertility diagnosis (eFigures 5 and 6 in the Supplement), substantial confounding would be required to fully explain the observed association (eTable 4 in the Supplement), and negative control analyses produced different curve shapes, with no relationship or attenuated relationships with short interpregnancy interval (eFigures 7 and 8 in the Supplement).

    Discussion

    Our study found that increased risks of maternal mortality or severe morbidity, adverse fetal and infant outcomes, and spontaneous preterm delivery following short (<12-month) interpregnancy intervals persisted after stratifying by maternal age. Women 35 years or older at index birth had increased risks of maternal mortality or severe morbidity at short interpregnancy intervals, while women aged 20 to 34 years did not. Although short intervals were associated with elevated risks of spontaneous preterm delivery and adverse fetal and infant outcomes for women of all ages, these risks were highest among younger women. For other outcomes, elevation in risks at short interpregnancy intervals did not vary by maternal age categories. Overall, contrary to our hypothesis, risks following short interpregnancy intervals were not attenuated for older women.

    Our results support a large body of literature examining increased risks of some fetal and infant outcomes, including preterm delivery, following short interpregnancy intervals.1-6 While increased preterm delivery risk following short interpregnancy intervals has been studied extensively,1,5,11 to our knowledge, few previous studies4 have examined both spontaneous and indicated preterm delivery separately. We demonstrated that spontaneous preterm delivery was more strongly associated with short interpregnancy interval than was indicated preterm delivery. Furthermore, although the association between short interpregnancy interval and specific maternal outcomes have been studied in high-resource settings (eg, uterine rupture or blood transfusion,37 abnormal placentation38), previous studies of maternal mortality or severe morbidity were restricted to low-resource settings.39 Because previous findings may not be generalizable to high-resource settings owing to differences in maternal age, parity, pregnancy intention, or nutritional status, our findings regarding maternal mortality or severe morbidity add new insight to the literature. The small number of maternal mortality events in our study population (n = 5; 3.4 per 100 000) precluded examination of maternal mortality as a separate outcome, although this may be of interest for future research.

    Current clinical and public health recommendations suggest a minimum interpregnancy interval of 18 months,7,9 and some suggest a range from 18 to 60 months.10 However, our findings suggest that the optimal interpregnancy interval is closer to 18 months, with a range of 12 to 24 months having risks that generally are comparable to the nadir at 18 months. In fact, for some outcomes (adverse fetal and infant outcomes for women ≥35 years old, SGA birth for all women), risks are similar to the nadir even at 9-month intervals.

    More than 20% of interpregnancy intervals among women 35 years or older were shorter than 12 months, when risks of some outcomes remain elevated for older women, and nearly half were shorter than 18 months. Because pregnancies following short interpregnancy intervals among women 35 years or older at index birth are usually planned,11,12 clinical counseling toward intervals closer to 12 or 18 months may be effective in reducing the risks that we have shown are still associated with short intervals among older women. While we believe these findings will be useful to women and clinicians, women’s decisions regarding the optimal timing and spacing of pregnancy are multifactorial, and modest increases in risk associated with short intervals may not outweigh other factors, including those unrelated to health outcomes, that women and families consider in spacing their pregnancies.

    Limitations

    Our findings must be interpreted in the context of several limitations. Because our analyses examined differences in risks according to interpregnancy interval based on comparing outcomes of different women, they may be biased due to confounding by unmeasured factors that differ within and between women, including pregnancy intention11,12 and intimate partner violence.33,34 Previous studies found substantially attenuated associations between interpregnancy interval and adverse pregnancy outcomes using matched designs to examine successive interpregnancy intervals within the same women40-42 compared with between-woman analyses. However, within-woman analyses cannot adequately account for the role of parity2 and may not be generalizable to the overall population owing to restriction to women with at least 3 births and women with discordant health outcomes. While we cannot rule out unmeasured confounding, our sensitivity analyses examining the potential role of unmeasured or residual confounding provide reassurance that our findings are unlikely to be entirely explained by confounding.

    Conclusions

    We present what we believe to be new, robust evidence to guide clinicians counseling women considering short interpregnancy intervals. Our study used a large, population-based cohort with clinical detail based on validated43 medical record abstraction, supplemented through linkages to additional health care and demographic information. This database enabled us to overcome many methodologic limitations of previous studies in this area, including controlling for both stillbirth and neonatal loss in the index birth and accurately measuring intervals between pregnancies by including the 18.4% of subsequent pregnancies that ended in spontaneous or induced abortions at less than 20 weeks in our definition.44,45

    Short interpregnancy intervals (<12 months) are associated with increased maternal and fetal and infant risks for women aged 20 to 34 years and for women 35 years or older at index birth. Our findings indicate a shorter optimal interval than previously thought (12-24 months) for women of all ages. This finding may be reassuring particularly for older women who must weigh the competing risks of increasing maternal age with longer interpregnancy intervals (including infertility and chromosomal anomalies) against the risks of short interpregnancy intervals. We found evidence that confounding does not seem to fully explain the observed association between short interpregnancy interval and increased risks of several outcomes.

    Back to top
    Article Information

    Accepted for Publication: July 21, 2018.

    Corresponding Author: Laura Schummers, SD, Department of Epidemiology, Harvard T. H. Chan School of Public Health, 677 Huntington Ave, Boston, MA 02115 (lauraschummers@mail.harvard.edu).

    Published Online: October 29, 2018. doi:10.1001/jamainternmed.2018.4696

    Author Contributions: Dr Schummers had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Schummers, Hutcheon, Hernandez-Diaz, VanderWeele, Norman.

    Acquisition, analysis, or interpretation of data: Schummers, Hutcheon, Hernandez-Diaz, Williams, Hacker, VanderWeele.

    Drafting of the manuscript: Schummers, VanderWeele.

    Critical revision of the manuscript for important intellectual content: All authors.

    Statistical analysis: Schummers, Hernandez-Diaz, Williams, VanderWeele.

    Obtained funding: Schummers.

    Administrative, technical, or material support: Schummers.

    Supervision: Hutcheon, Hernandez-Diaz, Williams, Hacker, Norman.

    Conflict of Interest Disclosures: None reported.

    Funding/Support: Dr Schummers was supported by National Research Service Award 1F31HD086970-01A1 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, and received grant CPP-137903 from the Canadian Institutes for Health Research and the Public Health Agency of Canada (CIHR-PHAC) Family Planning Public Health Chair Seed Grant to support this project. Dr Hutcheon was supported by New Investigator Awards from the Michael Smith Foundation for Health Research and the Canadian Institutes for Health Research. Dr VanderWeele was supported by National Institutes of Health Award R01CA222147. Dr Norman was supported by a Research Scholar Award from the Michael Smith Foundation for Health Research and as applied Public Health Chair in Family Planning Research from the Canadian Institutes for Health Research and the Public Health Agency of Canada (CPP-137903)

    Role of the Funder/Sponsor: The funding organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Disclaimer: All inferences, opinions, and conclusions drawn in this publication are those of the authors, and do not reflect the opinions or policies of Population Data BC or the data stewards.

    Meeting Presentations: This study was presented at the 2018 Meeting of the Society for Epidemiologic Research; June 19, 2018; Baltimore, Maryland; and 2018 Annual Meeting of the Society for Pediatric and Perinatal Epidemiologic Research; June 18, 2018; Baltimore, Maryland.

    Additional Contributions: We thank Population Data BC for compiling the study data and facilitating data access.

    References
    1.
    Conde-Agudelo  A, Rosas-Bermúdez  A, Kafury-Goeta  AC.  Birth spacing and risk of adverse perinatal outcomes: a meta-analysis.  JAMA. 2006;295(15):1809-1823. doi:10.1001/jama.295.15.1809PubMedGoogle ScholarCrossref
    2.
    Merklinger-Gruchala  A, Jasienska  G, Kapiszewska  M.  Short interpregnancy interval and low birth weight: a role of parity.  Am J Hum Biol. 2015;27(5):660-666. doi:10.1002/ajhb.22708PubMedGoogle ScholarCrossref
    3.
    McKinney  D, House  M, Chen  A, Muglia  L, DeFranco  E.  The influence of interpregnancy interval on infant mortality.  Am J Obstet Gynecol. 2017;216(3):316.e1-316.e9. doi:10.1016/j.ajog.2016.12.018PubMedGoogle ScholarCrossref
    4.
    Smith  GCS, Pell  JP, Dobbie  R.  Interpregnancy interval and risk of preterm birth and neonatal death: retrospective cohort study.  BMJ. 2003;327(7410):313. doi:10.1136/bmj.327.7410.313PubMedGoogle ScholarCrossref
    5.
    Wendt  A, Gibbs  CM, Peters  S, Hogue  CJ.  Impact of increasing inter-pregnancy interval on maternal and infant health.  Paediatr Perinat Epidemiol. 2012;26(1)(suppl 1):239-258. doi:10.1111/j.1365-3016.2012.01285.xPubMedGoogle ScholarCrossref
    6.
    Zhang  L, Shen  S, He  J,  et al.  Effect of interpregnancy interval on adverse perinatal outcomes in southern China: a retrospective cohort study, 2000-2015.  Paediatr Perinat Epidemiol. 2018;32(2):131-140. doi:10.1111/ppe.12432PubMedGoogle ScholarCrossref
    7.
    American College of Obstetricians and Gynecologists.  ACOG Committee opinion 736: optimizing postpartum care.  Obstet Gynecol. 2018;131(5):e140-e150. doi:10.1097/AOG.0000000000002633PubMedGoogle ScholarCrossref
    8.
    Royal College of Obstetricians and Gynaecologists. Best practice in postpartum family planning. Vol. Best Practice Paper No 1. London, UK: Royal College of Obstetricians and Gynaecologists. https://www.rcog.org.uk/globalassets/documents/guidelines/best-practice-papers/best-practice-paper-1---postpartum-family-planning.pdf. Published June 2015. Accessed March 12, 2018.
    9.
    Office of Disease Prevention and Health Promotion. Healthy People.gov. US Healthy people 2020 family planning objective 5. Healthy People 2020 Web site. https://www.healthypeople.gov/2020/topics-objectives/topic/family-planning/objectives. Updated September 7, 2018. Accessed March 3, 2018.
    10.
    World Health Organization. Report of a WHO technical consultation on birth spacing. Geneva, Switzerland, 13-15, 2005. http://apps.who.int/iris/bitstream/10665/69855/1/WHO_RHR_07.1_eng.pdf. Published 2007. Accessed March 3, 2018.
    11.
    Gemmill  A, Lindberg  LD.  Short interpregnancy intervals in the United States.  Obstet Gynecol. 2013;122(1):64-71. doi:10.1097/AOG.0b013e3182955e58PubMedGoogle ScholarCrossref
    12.
    Cheslack Postava  K, Winter  AS.  Short and long interpregnancy intervals: correlates and variations by pregnancy timing among US women.  Perspect Sex Reprod Health. 2015;47(1):19-26. doi:10.1363/47e2615PubMedGoogle ScholarCrossref
    13.
    Lean  SC, Derricott  H, Jones  RL, Heazell  AEP.  Advanced maternal age and adverse pregnancy outcomes: a systematic review and meta-analysis.  PLoS One. 2017;12(10):e0186287. doi:10.1371/journal.pone.0186287PubMedGoogle ScholarCrossref
    14.
    Zhu  BP, Rolfs  RT, Nangle  BE, Horan  JM.  Effect of the interval between pregnancies on perinatal outcomes.  N Engl J Med. 1999;340(8):589-594. doi:10.1056/NEJM199902253400801PubMedGoogle ScholarCrossref
    15.
    Finer  LB, Zolna  MR.  Declines in unintended pregnancy in the United States, 2008-2011.  N Engl J Med. 2016;374(9):843-852. doi:10.1056/NEJMsa1506575PubMedGoogle ScholarCrossref
    16.
    de Weger  FJ, Hukkelhoven  CW, Serroyen  J, te Velde  ER, Smits  LJ.  Advanced maternal age, short interpregnancy interval, and perinatal outcome.  Am J Obstet Gynecol. 2011;204(5):421.e1-421.e9. doi:10.1016/j.ajog.2010.12.008PubMedGoogle ScholarCrossref
    17.
    Perinatal Services BC. Perinatal Data Registry; 2017. Available at: http://www.perinatalservicesbc.ca/health-professionals/data-surveillance/perinatal-data-registry. Updated 2018. Accessed March 12, 2018.
    18.
    Population Data BC. Medical Services Plan (MSP) payment information file. V2; 2016. Data extract. Ministry of Health (2016). https://www.popdata.bc.ca/data/health/msp. Updated May 9, 2018. Accessed June 10, 2018.
    19.
    Population Data BC. PharmaNet. V2. BC Ministry of Health. Data extract. Data Stewardship Committee (2017). https://www.popdata.bc.ca/data/health/PharmaNet. Updated May 9, 2018. Accessed June 10, 2018.
    20.
    Population Data BC. Canadian Institute for Health Information. Discharge abstract database (hospital separations). V2. Data extract. 2015. Ministry of Health (2015). https://www.popdata.bc.ca/data/health/dad. Updated May 9, 2018. Accessed July 12, 2018.
    21.
    Population Data BC. BC Ministry of Health (2015). Consolidation file. V2. Data Extract. Ministry of Health (2016). https://www.popdata.bc.ca/data/population/consolidationfile. Updated May 9, 2018. Accessed May 12, 2018.
    22.
    Population Data BC. Statistics Canada. Small Area and Administrative Data Division (2009). Income band data; catalogue number 13C0016. Data extract (2017). https://www.popdata.bc.ca/data/demographic/incomeband. Updated May 9, 2018. Accessed June 12, 2018.
    23.
    Population Data BC. BC Vital Statistics Agency (2015). Vital statistics deaths. V2. Data extract. BC Vital Statistics Agency (2016). https://www.popdata.bc.ca/data/population/vsdeaths. Updated May 9, 2018. Accessed June 17, 2018.
    24.
    Love  ER, Bhattacharya  S, Smith  NC, Bhattacharya  S.  Effect of interpregnancy interval on outcomes of pregnancy after miscarriage: retrospective analysis of hospital episode statistics in Scotland.  BMJ. 2010;341:c3967. doi:10.1136/bmj.c3967PubMedGoogle ScholarCrossref
    25.
    Geller  SE, Rosenberg  D, Cox  S, Brown  M, Simonson  L, Kilpatrick  S.  A scoring system identified near-miss maternal morbidity during pregnancy.  J Clin Epidemiol. 2004;57(7):716-720. doi:10.1016/j.jclinepi.2004.01.003PubMedGoogle ScholarCrossref
    26.
    Kramer  MS, Platt  RW, Wen  SW,  et al; Fetal/Infant Health Study Group of the Canadian Perinatal Surveillance System.  A new and improved population-based Canadian reference for birth weight for gestational age.  Pediatrics. 2001;108(2):E35. doi:10.1542/peds.108.2.e35PubMedGoogle ScholarCrossref
    27.
    Orsini  N, Greenland  S.  A procedure to tabulate and plot results after flexible modeling of a quantitative covariate.  Stata J. 2011;11(1):1-29.Google Scholar
    28.
    Harrell  FE.  Regression Modeling Strategies with Applications to Linear Models, Logistic Regression, and Survival Analysis. New York, NY: Springer; 2001.
    29.
    Altman  DG, Bland  JM.  Interaction revisited: the difference between two estimates.  BMJ. 2003;326(7382):219. doi:10.1136/bmj.326.7382.219PubMedGoogle ScholarCrossref
    30.
    Carlin  JB, Galati  J, Royston  P.  A new framework for managing and analyzing multiply imputed data in Stata.  Stata J. 2008;8(1):49-67.Google Scholar
    31.
    Corp  S. Stata statistical software: Release 14. College Station, TX: StataCorp, LP; 2015.
    32.
    Hall  JA, Benton  L, Copas  A, Stephenson  J.  Pregnancy intention and pregnancy outcome: systematic review and meta-analysis.  Matern Child Health J. 2017;21(3):670-704. doi:10.1007/s10995-016-2237-0PubMedGoogle ScholarCrossref
    33.
    Maxwell  L, Nandi  A, Benedetti  A, Devries  K, Wagman  J, García-Moreno  C.  Intimate partner violence and pregnancy spacing: results from a meta-analysis of individual participant time-to-event data from 29 low-and-middle-income countries.  BMJ Glob Health. 2018;3(1):e000304. doi:10.1136/bmjgh-2017-000304PubMedGoogle ScholarCrossref
    34.
    Donovan  BM, Spracklen  CN, Schweizer  ML, Ryckman  KK, Saftlas  AF.  Intimate partner violence during pregnancy and the risk for adverse infant outcomes: a systematic review and meta-analysis.  BJOG. 2016;123(8):1289-1299. doi:10.1111/1471-0528.13928PubMedGoogle ScholarCrossref
    35.
    VanderWeele  TJ, Ding  P.  Sensitivity analysis in observational research: introducing the E-value.  Ann Intern Med. 2017;167(4):268-274. doi:10.7326/M16-2607PubMedGoogle ScholarCrossref
    36.
    Lipsitch  M, Tchetgen  E, Cohen  T.  Negative controls: a tool for detecting confounding and bias in observational studies.  Epidemiology. 2010;21(3):383-388. doi:10.1097/EDE.0b013e3181d61eebPubMedGoogle ScholarCrossref
    37.
    Stamilio  DM, DeFranco  E, Paré  E,  et al.  Short interpregnancy interval: risk of uterine rupture and complications of vaginal birth after cesarean delivery.  Obstet Gynecol. 2007;110(5):1075-1082. doi:10.1097/01.AOG.0000286759.49895.46PubMedGoogle ScholarCrossref
    38.
    Ananth  CV, Skjaerven  R, Klunssoyr  K.  Change in paternity, risk of placental abruption and confounding by birth interval: a population-based prospective cohort study in Norway, 1967-2009.  BMJ Open. 2015;5(2):e007023. doi:10.1136/bmjopen-2014-007023PubMedGoogle ScholarCrossref
    39.
    Conde-Agudelo  A, Belizán  JM.  Maternal morbidity and mortality associated with interpregnancy interval: cross sectional study.  BMJ. 2000;321(7271):1255-1259. doi:10.1136/bmj.321.7271.1255PubMedGoogle ScholarCrossref
    40.
    Class  QA, Rickert  ME, Oberg  AS,  et al.  Within-family analysis of interpregnancy interval and adverse birth outcomes.  Obstet Gynecol. 2017;130(6):1304-1311. doi:10.1097/AOG.0000000000002358PubMedGoogle ScholarCrossref
    41.
    Ball  SJ, Pereira  G, Jacoby  P, de Klerk  N, Stanley  FJ.  Re-evaluation of link between interpregnancy interval and adverse birth outcomes: retrospective cohort study matching two intervals per mother.  BMJ. 2014;349:g4333. doi:10.1136/bmj.g4333PubMedGoogle ScholarCrossref
    42.
    Hanley  GE, Hutcheon  JA, Kinniburgh  BA, Lee  L.  Interpregnancy interval and adverse pregnancy outcomes: an analysis of successive pregnancies.  Obstet Gynecol. 2017;129(3):408-415. doi:10.1097/AOG.0000000000001891PubMedGoogle ScholarCrossref
    43.
    Frosst  G, Hutcheon  J, Joseph  KS, Kinniburgh  B, Johnson  C, Lee  L.  Validating the British Columbia Perinatal Data Registry: a chart re-abstraction study.  BMC Pregnancy Childbirth. 2015;15:123. doi:10.1186/s12884-015-0563-7PubMedGoogle ScholarCrossref
    44.
    Conzuelo-Rodriguez  G, Naimi  AI.  The impact of computing interpregnancy intervals without accounting for intervening pregnancy events.  Paediatr Perinat Epidemiol. 2018;32(2):141-148. doi:10.1111/ppe.12458PubMedGoogle ScholarCrossref
    45.
    Naimi  AI.  Studying interpregnancy interval effects using observational data: some cautionary remarks.  BJOG. 2016;123(8):1319. doi:10.1111/1471-0528.13743PubMedGoogle ScholarCrossref
    ×