Error bars indicate 95% CIs.
eMethods. Supplemental Methods
eResults. Supplemental Results
eTable 1. International Classification of Diseases (ICD) Codes Used in the Analyses.
eTable 2. Associations Between Number of Preterm Deliveries and Subsequent Risk of Chronic Hypertension Among Women With at Least Two Singleton Deliveries
eTable 3. Spontaneous or Medically Indicated Delivery and Subsequent Risk of Chronic Hypertension, Sweden, 1990-2015
Customize your JAMA Network experience by selecting one or more topics from the list below.
Crump C, Sundquist J, Sundquist K. Preterm Delivery and Long-term Risk of Hypertension in Women. JAMA Cardiol. 2022;7(1):65–74. doi:10.1001/jamacardio.2021.4127
What are the long-term risks of hypertension in women who deliver preterm?
In this cohort study of more than 2 million women in Sweden, after adjusting for preeclampsia, other hypertensive disorders of pregnancy, and other maternal factors, women who delivered preterm had a greater than 1.6-fold risk of hypertension and women who delivered extremely preterm had a 2.2-fold risk within the next 10 years compared with those who delivered full term. These risks decreased but remained significantly elevated 40 years later and were not explained by shared familial factors.
These results suggest that preterm delivery is a lifelong risk factor for hypertension in women.
Preterm delivery has been associated with future cardiometabolic disorders in women. However, the long-term risks of chronic hypertension associated with preterm delivery and whether such risks are attributable to familial confounding are unclear. Such knowledge is needed to improve long-term risk assessment, clinical monitoring, and cardiovascular prevention strategies in women.
To examine the long-term risks of chronic hypertension associated with preterm delivery in a large population-based cohort of women.
Design, Setting, and Participants
This national cohort study assessed all 2 195 989 women in Sweden with a singleton delivery from January 1, 1973, to December 31, 2015. Data analyses were conducted from March 8, 2021, to August 20, 2021.
Pregnancy duration identified from nationwide birth records.
Main Outcomes and Measures
New-onset chronic hypertension identified from primary care, specialty outpatient, and inpatient diagnoses using administrative data. Cox proportional hazards regression was used to compute hazard ratios (HRs) while adjusting for preeclampsia, other hypertensive disorders of pregnancy, and other maternal factors. Cosibling analyses were assessed for potential confounding by shared familial (genetic and/or environmental) factors.
In 46.1 million person-years of follow-up, 351 189 of 2 195 989 women (16.0%) were diagnosed with hypertension (mean [SD] age, 55.4 [9.9] years). Within 10 years after delivery, the adjusted HR for hypertension associated with preterm delivery (gestational age <37 weeks) was 1.67 (95% CI, 1.61-1.74) and when further stratified was 2.23 (95% CI, 1.98-2.52) for extremely preterm (22-27 weeks of gestation), 1.85 (95% CI, 1.74-1.97) for moderately preterm (28-33 weeks of gestation), 1.55 (95% CI, 1.48-1.63) for late preterm (34-36 weeks of gestation), and 1.26 (95% CI, 1.22-1.30) for early-term (37-38 weeks of gestation) compared with full-term (39-41 weeks of gestation) delivery. These risks decreased but remained significantly elevated at 10 to 19 years (preterm vs full-term delivery: adjusted HR, 1.40; 95% CI, 1.36-1.44), 20 to 29 years (preterm vs full-term delivery: adjusted HR, 1.20; 95% CI, 1.18-1.23), and 30 to 43 years (preterm vs full-term delivery: adjusted HR, 95% CI, 1.12; 1.10-1.14) after delivery. These findings were not explained by shared determinants of preterm delivery and hypertension within families.
Conclusions and Relevance
In this large national cohort study, preterm delivery was associated with significantly higher future risks of chronic hypertension. These associations remained elevated at least 40 years later and were largely independent of other maternal and shared familial factors. Preterm delivery should be recognized as a lifelong risk factor for hypertension in women.
Cardiovascular disease (CVD) is the leading cause of death in women and men worldwide but remains understudied and underrecognized in women.1,2 The most common modifiable CVD risk factor is hypertension, which accounts for most cases of stroke, ischemic heart disease (IHD), and other forms of CVD.3,4 Nearly 1 in 2 women worldwide have hypertension according to the 2017 American College of Cardiology/American Heart Association diagnostic criteria (mean blood pressure ≥130/80 mm Hg or antihypertensive medication use).1,5 Major risk factors for hypertension include advancing age, obesity, unhealthy diet, excessive alcohol use, and physical inactivity.1,6,7 Certain pregnancy complications, especially preeclampsia and other hypertensive disorders of pregnancy, also have been linked with subsequent development of chronic hypertension.8-10 These complications also are commonly associated with preterm delivery.11 However, the independent contribution of preterm delivery to long-term risks of chronic hypertension is unclear. A better understanding of the long-term hypertension risks associated with preterm delivery is needed to improve risk stratification, clinical monitoring, and CVD prevention in women.
Preterm delivery is a major public health problem, affecting nearly 11% of all births or 15 million deliveries worldwide each year.12 Women who deliver preterm have been reported to have higher long-term risks of stroke13 and IHD.14 A few studies15-17 also have suggested that preterm delivery, even without preeclampsia or other hypertensive disorders of pregnancy, is associated with modestly increased risks of developing chronic hypertension, with relative risks in the 1.1 to 1.5 range. However, prior evidence has been limited by self-reported pregnancy history and outcomes15 and the unavailability of data from primary care settings15-17 where most hypertension cases are diagnosed.18 Furthermore, it is unclear whether previously reported associations are attributable to confounding by familial (genetic and/or environmental) factors that are shared determinants of both preterm delivery and hypertension. To our knowledge, this has not been examined; thus, the relative contributions of shared familial factors vs direct effects of preterm delivery remain unknown.
To address these knowledge gaps, we conducted a national cohort study of more than 2 million women in Sweden. Our goals were to (1) examine the risks of chronic hypertension associated with preterm delivery in the largest population-based cohort to date, (2) examine changes in such risks later in life with up to 43 years of follow-up, and (3) assess for potential confounding by shared genetic or early-life environmental factors within families by performing cosibling analyses. We hypothesized that preterm delivery is associated with long-term increased risks of chronic hypertension and that these risks are largely independent of shared familial factors.
This national cohort study used data from the Swedish Medical Birth Register, which contains prenatal and birth information for nearly all deliveries in Sweden since 1973.19 Using this register, we identified 4 331 668 singleton deliveries among 2 203 746 women from January 1, 1973, to December 31, 2015. Singleton deliveries were selected to improve internal comparability given the higher prevalence of preterm delivery and its different underlying causes in multiple-gestation pregnancies.20 Women with preexisting hypertension at the onset of their first pregnancy (5625 [0.3%]), a subset of those with hypertensive disorders of pregnancy,21 were excluded to enable assessment of new-onset chronic hypertension associated with preterm delivery. We also excluded 8812 deliveries that lacked information on pregnancy duration, resulting in exclusion of 2132 women (0.1%). A total of 2 195 989 women (99.6% of the original cohort) with 4 308 286 deliveries remained for inclusion in the study. Data analyses were conducted from March 8, 2021, to August 20, 2021. This study was approved by the Regional Ethical Review Board in Lund, Sweden. Participant consent was not required because this study used only pseudonymized, registry-based secondary data. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Pregnancy duration was identified from the Swedish Medical Birth Register based on maternal report of last menstrual period in the 1970s and ultrasonographic estimation starting in the 1980s and onward (>70% of the cohort). Pregnancy duration was examined alternatively as a continuous variable or categorical variable with 6 groups based on the number of completed weeks: extremely preterm (22-27 weeks of gestation), moderately preterm (28-33 weeks of gestation), late preterm (34-36 weeks of gestation), early-term (37-38 weeks of gestation), full-term (39-41 weeks of gestation, used as the reference group), and postterm (≥42 weeks of gestation) delivery.11,13,14,22,23 In addition, the first 3 groups were combined to provide summary estimates for preterm delivery (<37 weeks of gestation).
The study cohort was followed up for hypertension from first delivery through December 31, 2015 (follow-up time: median, 23.7 years; range, 0.2-43.0 years). The outcome was defined as any hypertension diagnosis that occurred more than 12 weeks post partum to distinguish chronic from gestational hypertension.21 Hypertension diagnoses were identified using International Classification of Diseases, Eighth Revision (ICD-8), International Classification of Diseases, Ninth Revision (ICD-9), and International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) assigned at any health care encounter in inpatient, specialty outpatient, or primary care settings as recorded in the Swedish hospital, outpatient, and primary care registers and all deaths attributed to hypertension in the Swedish Death Register (see eTable 1 in the Supplement for ICD codes).23 The Swedish Hospital Discharge Register contains all primary and secondary hospital discharge diagnoses from 6 populous counties in southern Sweden starting in 1964 and with nationwide coverage starting in 1987; these diagnoses are currently more than 99% complete, and positive predictive values greater than 95% have been reported for hypertension and other major CVDs.24 The Swedish Outpatient Register contains all diagnoses from specialty clinics nationwide starting in 2001. The Swedish Primary Care Register initially included primary care diagnoses that covered 20% of the national population starting in 1998 and was expanded to cover approximately 45% of the national population in 2001 and more than 75% by 2008 and onward.25 The Swedish Death Register includes all deaths and causes of death for all persons registered in Sweden since 1960, with compulsory reporting nationwide.
Throughout the study period (January 1, 1973, to December 31, 2015), the predominant criteria for diagnosis of hypertension were mean systolic blood pressure of 140 mm Hg or higher and/or mean diastolic blood pressure of 90 mm Hg or higher or receiving antihypertensive drug therapy based on World Health Organization and European Society of Hypertension/European Society of Cardiology guidelines during this period.26,27 However, specific blood pressure measurements were unavailable for analysis; thus, hypertension was assessed based on ICD codes as described above.
Other maternal characteristics that may be associated with pregnancy duration and hypertension risk were identified using the Swedish Medical Birth Register and national census data, which were linked using a pseudonymous version of the personal identification number. The following maternal variables were included in adjusted models, with all time-varying factors updated for each pregnancy and modeled as time-dependent covariates: age at delivery (continuous and categorical in 5-year groups), calendar year of delivery (continuous and categorical by decade), parity (1, 2, 3, 4, or ≥5), educational level (≤9, 10-12, or >12 years), employment status (yes/no), and income (quartiles) in the year before delivery; country or region of origin (Sweden; other European countries, US, and Canada; Asia and Oceania; Africa; Latin America; and other or unknown); prenatal smoking (0, 1-9, or ≥10 cigarettes/d) and body mass index (BMI [calculated as weight in kilograms divided by height in meters squared]; continuous and categorical [<18.5, 18.5-24.9, 25.0-29.9, or ≥30.0]); and preeclampsia, other hypertensive disorders of pregnancy, and diabetes (gestational or pregestational types 1 or 2) before delivery (see eTable 1 in the Supplement for ICD codes).
Maternal smoking and BMI were assessed at the beginning of prenatal care starting in 1982, and maternal smoking data were available for 74.7% and BMI data for 61.8% of all deliveries. Data were greater than 99% complete for all other variables. Missing data for each covariate were multiply imputed with 10 imputations using all other covariates and hypertension as independent variables.28 As alternatives to multiple imputation, sensitivity analyses were performed that were restricted to deliveries with complete data (2 600 924 deliveries among 1 559 495 women) or coded missing data for each covariate as a separate category.
Cox proportional hazards regression was used to compute hazard ratios (HRs) and 95% CIs for associations between pregnancy duration and subsequent risk of chronic hypertension. These associations were examined across the maximum possible follow-up (up to 43 years) and in narrower intervals of follow-up (<10, 10-19, 20-29, and 30-43 years) among women still living in Sweden without a prior diagnosis of hypertension at the beginning of the respective interval. Pregnancy duration was modeled as a time-dependent variable, with the exposure category determined by the shortest pregnancy to date. For example, if a woman’s first delivery was full term and her second was preterm, she entered the preterm exposure category at the date of her second delivery.13,14,22 In all models, elapsed time since delivery was used as the Cox proportional hazards regression model time axis. Models were performed both unadjusted and adjusted for covariates (as defined above). Women were censored at death from causes other than hypertension as identified in the Swedish Death Register (53 633 [2.4%]) or emigration as determined by absence of a Swedish residential address in census data (44 896 [2.0%]). Absolute risk differences and 95% CIs were computed for each pregnancy duration group compared with full term.29 The proportional hazards assumption was assessed by examining log-log plots,30 and no substantial departures were found.
Cosibling analyses were performed to assess for potential confounding by unmeasured shared familial (genetic and/or early-life environmental) factors.13,14,22,23 Shared environmental factors in families may include lifestyle factors, such as diet and physical activity, or ambient exposures, such as passive smoking. These analyses included all 1 190 226 women with at least 1 sister who had a singleton delivery. Stratified Cox proportional hazards regression was used with a separate stratum for each set of sisters as identified by their mother’s pseudonymous identification number. In the stratified Cox proportional hazards regression model, each set of sisters had its own baseline hazard function that reflected their shared genetic and environmental factors, thus controlling for their shared exposures. In addition, these analyses were further adjusted for the same covariates as in the main analyses.
Other secondary and sensitivity analyses are described in the eMethods in the Supplement. All statistical tests were 2-sided and used an α level of .05. All analyses were conducted using Stata software, version 15.1 (StataCorp LLC).
In 46.1 million person-years of follow-up, 351 189 of 2 195 989 women (16.0%) were diagnosed with hypertension (mean [SD] age, 55.4 [9.9] years). Table 1 gives the maternal characteristics by women’s shortest pregnancy duration. Women who delivered preterm were more likely than other women to be younger than 20 years, have a low educational level, be unemployed, smoke, or have high prenatal BMI, preeclampsia, other hypertensive disorders of pregnancy, or diabetes. A total of 71.2% of hypertension cases were diagnosed only in primary care settings and thus would have been missed if primary care records were unavailable. Mean (SD) ages were 27.6 (5.2) years at first delivery, 55.4 (9.9) years at hypertension diagnosis, and 50.6 (13.3) years at end of follow-up. Incidence rates by pregnancy duration and follow-up time are reported in Table 2.
Across the entire follow-up period (0-43 years after first delivery), the adjusted HRs for hypertension associated with pregnancy duration were 1.42 (95% CI, 1.35-1.49) for extremely preterm delivery, 1.25 (95% CI, 1.23-1.26) for any preterm delivery, and 1.11 (95% CI, 1.10-1.12) for early-term delivery compared with full-term delivery (Table 2). Each additional week of pregnancy was associated with a mean 3% lower risk of hypertension (adjusted HR, 0.97; 95% CI, 0.97-0.97; P < .001).
The HRs were highest within the first 10 years after delivery and then subsequently decreased, whereas risk differences (ie, excess hypertension cases) associated with preterm delivery increased with longer follow-up (Table 2 and Figure 1). Within the first 10 years after delivery, the adjusted HR for hypertension associated with preterm delivery was 1.67 (95% CI, 1.61-1.74) and when further stratified was 2.23 (95% CI, 1.98-2.52) for extremely preterm, 1.85 (95% CI, 1.74-1.97) for moderately preterm, 1.55 (95% CI, 1.48-1.63) for late preterm, and 1.26 (95% CI, 1.22-1.30) for early-term delivery compared with full-term delivery (Table 2). Each additional week of pregnancy was associated with a mean 7% lower risk (adjusted HR, 0.93; 95% CI, 0.93-0.94; P < .001).
After longer follow-up, the risk of hypertension associated with preterm delivery decreased but remained significantly elevated at 10 to 19 years (adjusted HR, 1.40; 95% CI, 1.36-1.44), 20 to 29 years (adjusted HR, 1.20; 95% CI, 1.18-1.23), and 30 to 43 years (adjusted HR, 1.12; 95% CI, 1.10-1.14) after delivery. Early-term delivery (37-38 weeks of gestation) also was associated with slightly increased risks even more than 20 years later compared with full-term delivery (adjusted HR, 20-29 years: 1.12; 95% CI, 1.11-1.14; P < .001; 30-43 years: adjusted HR, 1.06; 95% CI, 1.04-1.07; P < .001). Figure 1 shows adjusted HRs and unadjusted risk differences for hypertension by time since delivery for different pregnancy durations using spline curves. Figure 2 shows adjusted HRs and 95% CIs for hypertension by pregnancy duration and time since delivery.
As an alternative to adjusting for preeclampsia or other hypertensive disorders of pregnancy, exclusion of women with these disorders resulted in little change in risk estimates. For example, comparing preterm with full-term delivery, the adjusted HRs for chronic hypertension were 1.63 (95% CI, 1.54-1.72) at less than 10 years of follow-up, 1.42 (95% CI, 1.37-1.46) at 10 to 19 years of follow-up, 1.19 (95% CI, 1.17-1.22) at 20 to 29 years of follow-up, and 1.11 (95% CI, 1.09-1.13) at 30 to 43 years of follow-up (P < .001 for each).
Cosibling analyses to control for unmeasured shared familial factors resulted in partial attenuation of most but not all risk estimates (Table 3). For example, comparing preterm with full-term delivery, the adjusted HRs for hypertension were 1.67 (95% CI, 1.61-1.74) in the primary analysis vs 1.70 (95% CI, 1.39-2.07) in the cosibling analysis within 10 years after delivery, 1.40 (95% CI, 1.36-1.44) in the primary analysis vs 1.29 (95% CI, 1.15-1.45) in the cosibling analysis at 10 to 19 years after delivery, 1.20 (95% CI, 1.18-1.23) in the primary analysis vs 1.11 (95% CI, 1.04-1.19) in the cosibling analysis at 20 to 29 years after delivery, and 1.12 (95% CI, 1.10-1.14) in the primary analysis vs 1.07 (95% CI, 1.00-1.15) in the cosibling analysis at 30 to 43 years after delivery. Overall, these findings suggest that the associations observed in the main analyses were only partially attributable to shared genetic or early-life environmental factors in families.
Associations between the number of preterm deliveries and risk of hypertension among women with at least 2 singleton deliveries are given in eTable 2 in the Supplement. Women with recurrent preterm delivery had further increases in risk. The adjusted HRs per each additional preterm delivery were 1.51 (95% CI, 1.47-1.56) at less than 10 years of follow-up, 1.28 (95% CI, 1.26-1.31) at 10 to 19 years of follow-up, 1.12 (95% CI, 1.10-1.14) at 20 to 29 years of follow-up, and 1.10 (95% CI, 1.08-1.11) at 30 to 43 years of follow-up (P < .001 for each).
Both spontaneous and medically indicated delivery at preterm or early term were associated with increased risks of chronic hypertension compared with full-term delivery (eg, spontaneous preterm: adjusted HR, 1.25; 95% CI, 1.21-1.30; P < .001; medically indicated preterm: adjusted HR, 1.46; 95% CI, 1.42-1.50; P < .001) (eTable 3 in the Supplement). However, medically indicated delivery was the stronger risk factor (P < .001 for difference in these HRs). Medically indicated preterm delivery that was specifically related to preeclampsia or other hypertensive disorders of pregnancy was associated with nearly a 1.7-fold subsequent risk (adjusted HR, 1.67; 95% CI, 1.62-1.73; P < .001), whereas preterm delivery for other indications (predominantly diabetes) was associated with a 1.4-fold risk (adjusted HR, 1.41; 95% CI, 1.35-1.47; P < .001) of chronic hypertension.
Among women with 2 deliveries, the subsequent risk of hypertension did not vary significantly by whether the first vs second delivery occurred preterm. In addition, all results were negligibly changed in other analyses that further adjusted for oral contraceptive use, assessed alternatives to multiple imputation for missing data, ascertained hypertension based on 2 or more (rather than ≥1) diagnoses, or restricted to each woman's first delivery. These analyses are reported in the eResults in the Supplement.
In this large national cohort study of women, shorter pregnancy duration was associated with significantly higher future risks of chronic hypertension, even after adjusting for preeclampsia, other hypertensive disorders of pregnancy, and other potential confounders. Women who delivered preterm had greater than 1.6-fold and those who delivered extremely preterm had 2.2-fold risks of hypertension within the next 10 years compared with those who delivered full term. These risks subsequently decreased but remained slightly elevated (approximately 1.1-fold) even 30 to 43 years after delivery. Early-term delivery (37-38 weeks of gestation) also was associated with modestly increased risks of chronic hypertension. Cosibling analyses suggested that these findings were only partially explained by familial (genetic and/or early-life environmental) factors that are shared determinants of both preterm delivery and hypertension.
To our knowledge, this is the largest study to date of preterm delivery in relation to future risks of hypertension and the first to assess for potential confounding by shared familial factors using a cosibling design. Most of the risk estimates were consistent with or slightly higher than those from previous smaller studies.15-17 For example, a US study15 based on self-reported data in 57 904 women found that a first delivery that was preterm was associated with a 1.4-fold risk of hypertension within the next 10 years (HR, 1.43; 95% CI, 1.20-1.70) and a 1.1-fold risk after a median follow-up of 28 years (HR, 1.11; 95% CI, 1.06-1.17). A Danish cohort study16 based on hospital and death data for 427 775 women (average follow-up, 28 years) reported a nearly 1.3-fold risk of hypertension associated with preterm delivery (adjusted HR, 1.27; 95% CI, 1.21-1.34). In an overlapping Danish cohort of 782 287 women (median follow-up, 14.6 years), the adjusted HRs for hypertension were 1.49 (95% CI, 1.09-2.04) for those who delivered at less than 28 weeks of gestation, 1.50 (95% CI, 1.25-1.80) for those who delivered at 28 to 31 weeks of gestation, and 1.30 (95% CI, 1.20-1.40) for those who delivered at 32 to 36 weeks of gestation.17 However, each of those studies lacked data from primary care settings; thus, hypertension likely was substantially underreported. A prospective US cohort study31 of 4484 women reported that preterm delivery was associated with a 2.2-fold risk of hypertension (95% CI, 1.6-3.1) after a mean follow-up of 3.2 years. Smaller clinical studies32-35 also reported that preterm delivery was associated with slightly higher systolic blood pressure (2-5 mm Hg higher mean levels) and diastolic blood pressure (1-4 mm Hg higher mean levels) after follow-up that ranged from 3 to 20 years.
The current study extends prior evidence by incorporating primary care, specialty outpatient, and inpatient data for a large national cohort with up to 43 years of follow-up and more than 12-fold as many hypertension cases as prior studies.15-17 In this cohort, more than 70% of hypertension cases would have been missed without primary care records. The findings indicated that hypertension risks after preterm delivery were highest within the next 10 years yet remained significantly elevated for at least 40 years. The relative risks decreased with longer follow-up, likely because of the increasing incidence of hypertension at older ages that is unrelated to preterm delivery. However, the excess hypertension cases (ie, risk differences) associated with preterm delivery, and thus the public health impacts, increased with longer follow-up to ages when hypertension is more likely to manifest. Women who delivered at early term (37-38 weeks of gestation) also had persistent, slightly increased risks compared with those who delivered at full term. Hypertension risks were significantly increased even after spontaneous preterm delivery and in women without preeclampsia or other hypertensive disorders of pregnancy. Women with recurrent preterm delivery had further increases in risk.
A previously unanswered key question was whether associations between preterm delivery and CVD are related to preexisting shared determinants of these outcomes or whether preterm delivery itself contributes to or modulates the pathophysiologic mechanisms that lead to CVD.17,36 Our findings for hypertension, as well as previous findings for stroke,13 IHD,14 and CVD mortality,22 suggest that the latter pathway is likely relevant. We found that associations between preterm delivery and hypertension were only partially attenuated after adjusting for unmeasured shared familial (genetic and/or early-life environmental) factors in cosibling analyses. Preeclampsia and other hypertensive disorders of pregnancy also are common causes of preterm delivery11 and linked with higher subsequent risks of hypertension8-10 and CVD mortality.10,37 However, even adjusting for or excluding these conditions, women with preterm delivery still had persistently elevated long-term risks of hypertension. Recent evidence suggests that preterm delivery may be a key event that triggers endothelial-specific inflammation that is undetectable before pregnancy.38 The resulting endothelial dysfunction is associated with functional changes in the microvasculature, including impaired ability to release endothelium-derived relaxing factors, leading to increased constrictive tone.39 These functional changes may contribute to subsequent development of hypertension and CVD.40-42 Further delineation of these mechanisms may eventually reveal new targets for intervention to prevent both preterm delivery and development of hypertension and CVD.
The current findings are consistent with previously reported associations between preterm delivery and long-term risks of stroke,13 IHD,14 and all-cause and CVD mortality.22,43 These findings have important clinical implications. Preterm delivery should now be recognized by clinicians and patients as a lifelong major risk factor for hypertension and CVD. Cardiovascular risk assessment in women should routinely include reproductive history that covers preterm delivery and other adverse pregnancy outcomes. This history should be a required element of electronic health records and linked with primary care to facilitate access by clinicians across patients’ life course.44 A history of preterm delivery may help identify women at high risk of hypertension and CVD potentially long before the onset of these conditions. Such women need early preventive actions to reduce other modifiable risk factors, including obesity, unhealthy diet, excessive alcohol use, smoking, and physical inactivity.1,6,7 These interventions should be implemented soon after preterm delivery, followed by long-term clinical monitoring for hypertension and other CVD risk factors. Ambulatory blood pressure monitoring may be needed for better detection of hypertension in high-risk women.45,46 Our findings also underscore the importance of public health strategies to help prevent preterm delivery, including better access to high-quality preconception and prenatal care,47 especially in the US and other populations with a high prevalence of preterm birth.12
This study has several strengths. A major strength is its large national cohort design, with up to 43 years of follow-up. The use of highly complete nationwide prenatal and birth records helped minimize potential selection biases. The availability of primary care, specialty outpatient, and inpatient diagnoses enabled, for the first time, highly complete ascertainment of hypertension that included primary care settings, where most cases are diagnosed.18 The large size of this cohort afforded high statistical power to assess narrowly defined pregnancy durations and key subgroups. Many maternal factors that are potential confounders were controlled for, as were unmeasured shared familial factors using cosibling analyses.
The study also has limitations, including the lack of detailed clinical records needed to verify hypertension diagnoses. High positive predictive values have been reported using Swedish registry data.24 However, to our knowledge, these values and the sensitivity and specificity have not been estimated specifically using outpatient records. In the current study, their sensitivity is likely greatly enhanced by the inclusion of primary care data to enable more complete capture of cases.48 Maternal report of last menstrual period (used mainly in the 1970s) may overestimate gestational age by approximately 2 days,49 causing a slight conservative bias to risk estimates. From the 1980s onward, gestational age was determined by ultrasonography, reducing potential misclassification. The 2017 American College of Cardiology/American Heart Association diagnostic criteria for hypertension5 have lower blood pressure cutoffs than those used in the current study period, which would result in a higher prevalence of hypertension than reported in this study. Lifestyle factors later in life may be important risk modifiers after preterm delivery and should be examined in future studies with access to this information. The current study also was limited to one country and will need replication in other populations when feasible. A priority for future research is to extend these findings to racial and ethnic subgroups that are at highest risk of preterm delivery and hypertension. Additional follow-up will be needed to examine these associations in older adulthood when hypertension increasingly and disproportionately affects women.5,50
In this national cohort of more than 2 million women, preterm delivery was associated with higher future risks of chronic hypertension even after adjusting for preeclampsia, other hypertensive disorders of pregnancy, other maternal factors, and unmeasured shared familial factors. These risks were highest within the first 10 years but remained significantly elevated at least 40 years later. Preterm delivery should now be recognized as a risk factor for hypertension across the life course. Women with a history of preterm delivery need early preventive evaluation and long-term risk reduction and monitoring for hypertension.
Accepted for Publication: August 24, 2021.
Published Online: October 13, 2021. doi:10.1001/jamacardio.2021.4127
Corresponding Author: Casey Crump, MD, PhD, Department of Family Medicine and Community Health, Icahn School of Medicine at Mount Sinai, One Gustave L. Levy Place, PO Box 1077, New York, NY 10029 (firstname.lastname@example.org).
Author Contributions: Dr J. Sundquist had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Crump.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Crump.
Obtained funding: Crump, K. Sundquist.
Administrative, technical, or material support: J. Sundquist, K. Sundquist.
Supervision: All authors.
Conflict of Interest Disclosures: Drs Crump and K. Sundquist reported receiving grants from the National Heart, Lung, and Blood Institute during the conduct of the study. No other disclosures were reported.
Funding/Support: This work was supported by grant R01 HL139536 from the National Heart, Lung, and Blood Institute at the National Institutes of Health (Drs Crump and K. Sundquist), the Swedish Research Council, the Swedish Heart-Lung Foundation, and an Avtal om Läkarutbildning och Forskning (Agreement on Medical Training and Research) (ALF) project grant from Region Skåne/Lund University.
Role of the Funder/Sponsor: The funding sources 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.