Predictors of Hypertensive Disorders of Pregnancy in Japan: A Population-Based Retrospective Cohort Study Using the Shizuoka Kokuho Database.

Aim To identify the predictors of the onset of hypertensive disorders of pregnancy (HDP) in pregnant women in Japan.

This population‐based, retrospective cohort study was conducted in Shizuoka Prefecture, Japan, using data from the Shizuoka Kokuho Database from 2012 to 2020. A 1‐year baseline period and a 280‐day follow‐up period were defined for each participant. The outcome was the occurrence of HDP, identified using the International Classification of Diseases, 10th Revision codes. Candidate predictors included maternal age, residential area, and baseline comorbidities. Associations were assessed using multivariable logistic regression and conditional inference tree analysis.

A total of 6362 pregnant women were included. Among them, 222 (3.5%) were diagnosed with HDP. Multivariable logistic regression identified advanced maternal age (odds ratio [OR] 1.05, 95% confidence interval [CI] 1.02–1.07), diabetes mellitus (OR 3.21, 95% CI 1.22–8.47), preexisting hypertension (OR 2.38, 95% CI 1.04–5.46), and systemic lupus erythematosus (OR 5.90, 95% CI 1.41–24.65) as predictors of HDP onset. In sensitivity analyses excluding women with preexisting hypertension, maternal age, diabetes mellitus, systemic lupus erythematosus, and obesity remained associated with HDP onset. Conditional inference tree analysis identified maternal age > 34 years as the primary splitting variable, followed by hyperlipidemia, hypertension, and liver disease. Among women aged > 34 years without hyperlipidemia or hypertension, polycystic ovary syndrome was associated with increased risk of HDP.

We delineated a multifactorial risk profile for HDP. These findings indicate that routinely collected claims data may facilitate earlier risk stratification and support more individualized prenatal care.

conditional inference tree, health insurance claims database, hypertensive disorders of pregnancy, pregnancy complications, risk factors 1. Introduction Hypertensive disorders of pregnancy (HDP) affect 5%–10% of all pregnancies and remain a leading cause of maternal and perinatal morbidity and mortality worldwide [1, 2, 3]. In addition to acute complications, HDP confers long‐term cardiovascular risks to both mother and child [4]. Given these short‐ and long‐term consequences, early recognition of women at risk for HDP is crucial to prevent progression to severe disease. International guidelines emphasize prompt recognition and monitoring to enable preventive interventions and reduce maternal and perinatal complications [3, 5]. Large‐scale epidemiological studies in Western countries have established multiple maternal risk factors and predictors of HDP, including advanced maternal age, obesity, chronic hypertension, and autoimmune conditions [1, 2, 5]. However, population‐based evidence in Japan remains limited. Differences in genetic background, prevalence of comorbidities, and healthcare delivery systems suggest that the risk profiles observed in Western populations may not directly apply to Japanese women. Previous Japanese studies have focused mainly on maternal age or overall incidence trends [6, 7, 8], without systematically evaluating diverse comorbidities or their interactions. Therefore, clarifying the predictors of HDP in Japanese women using large‐scale, real‐world data is essential to inform region‐specific prevention and management strategies. To address this gap, we conducted a population‐based retrospective cohort study using the Shizuoka Kokuho Database (SKDB), a health insurance claims database covering approximately 2.6 million residents across all 35 municipalities in Shizuoka Prefecture [9, 10]. We aimed to identify maternal and comorbidity‐related predictors of HDP in Japan and to explore interactions among these factors as well as characterize clinically meaningful high‐risk subgroups. 2. Methods 2.1. Data Source The SKDB is a health insurance claims database for Shizuoka Prefecture, covering residents of 23 cities and 12 towns enrolled in the National Health Insurance and the Late‐Stage Elderly Medical Care System. Shizuoka Prefecture, which is located in central Japan, has a population of approximately 3.6 million [11]. Shizuoka is considered broadly representative of the national average due to its demographic and socioeconomic characteristics, such as age distribution, occupational structure, consumer prices, and household expenditure [9]. The database is updated frequently; the version used in this study was SKDB ver. 2021 and the analytical data generation system ver. 2.0, which covered the period from April 1, 2012, to September 30, 2020, and registered 2 654 567 individuals. All data were anonymized and processed to ensure no duplication. The survival time data structure was maintained for longitudinal analysis [10]. 2.2. Study Design and Study Population This retrospective cohort study was conducted using the SKDB, and we adopted a historical cohort design following the RECORD guidelines for database studies [12]. Figure S1 illustrates an outline of the study design. In Japan, delivery is not covered by public health insurance. However, when pregnant women seek medical care for conditions other than delivery, the disease or injury may be accompanied, at the physician's discretion, by a gestational‐age modifier (ranging from 4 to 45 weeks). In Japan, insurance claims can include the gestational week (4–45 weeks) recorded alongside the diagnosis—the gestational‐age modifier, which denotes pregnancy status. In this study, patients with this modifier were considered pregnant [13]. We restricted the study population to women aged 12–55 years with a gestational‐age modifier (4–45 weeks), thereby selecting women of reproductive age (Figure 1). The key date (onset of pregnancy) was obtained by subtracting the gestational weeks indicated in the modifier from the date of the first registration of a disease with that modifier. The baseline period was set as 1 year before the key date, and the first day of the baseline period was defined as the participant's entry into the cohort. The follow‐up period was set at 280 days (the duration of pregnancy) from the key date. Participants who could not secure a baseline period of 1 year and a follow‐up period of 280 days were excluded. Participants for whom residential data could not be obtained were also excluded. 2.3. Outcomes The study outcome was the occurrence of HDP during the follow‐up period, identified using the 10th revision of the International Classification of Diseases (ICD‐10) codes corresponding to gestational hypertension, preeclampsia, HELLP syndrome, and eclampsia (O13, O14, and O15). Chronic hypertension (O10), superimposed preeclampsia (O11), non‐hypertensive edema/proteinuria (O12), and unspecified maternal hypertension (O16) were excluded from the outcome definition to focus on new‐onset HDP during pregnancy and to avoid overlap with preexisting hypertension as a predictor. 2.4. Variables We selected the following factors as candidate predictors: age, area of residence, and previously reported risk factors, including miscarriage, antiphospholipid antibody syndrome, benign uterine tumor, blood coagulation disorder, diabetes, endometriosis, Helicobacter pylori infection, hyperlipidemia, hypertension, hyperthyroidism or hypothyroidism, liver disease, obesity, periodontal disease, polycystic ovary syndrome (PCOS), renal failure, rheumatic disease, systemic lupus erythematosus (SLE), and venous thrombosis [1, 2, 3, 14, 15, 16, 17, 18, 19, 20]. In addition, comorbid conditions clinically thought to contribute to the onset of HDP were included. These conditions include amenorrhea, anemia, arrhythmia, dental caries, cervical dysplasia, chlamydial infection, congestive heart failure, depression, hypermenorrhea, gonorrhea infection, infertility, influenza infection, hypotension, pelvic inflammatory disease, peripheral vascular disease, psychosis, recurrent pregnancy loss, valvular disease, and underweight or malnutrition. All candidate predictors were defined based on the ICD‐10 list (Table S1). Residential area was categorized into two groups: government‐designated cities (major urban areas; Shizuoka City and Hamamatsu City) and other cities and towns. 2.5. Statistical Analysis To identify the predictors associated with the onset of HDP, we conducted univariate and multivariable logistic regression analyses, with the presence or absence of HDP during the follow‐up period as the outcome. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated, and p‐values were obtained using the Wald test. After confirming linearity, age was entered into the model as a continuous variable. To address multicollinearity, one of two variables with a Spearman's correlation coefficient of ≥ 0.4 was excluded from the multivariable model. We also performed a conditional inference tree (CIT) analysis using the ctree() function from the partykit package in R, with HDP (absent/present) as the outcome. Candidate predictors included age, residential area, comorbidities that were significant in univariable logistic regression, and additional conditions considered clinically relevant to the onset of HDP. CIT is a recursive partitioning method that accounts for non‐linear relationships and interactions among variables, splitting nodes according to statistically significant associations, starting with the most influential predictors. The stopping criteria were set at mincriterion = 0.80 (p < 0.20), minsplit = 20, and minbucket = 5. For each terminal node, we calculated HDP incidence and 95% CIs to identify high‐risk groups. Sensitivity analyses were additionally performed after excluding women with preexisting hypertension to evaluate the robustness of the observed associations independent of baseline hypertension status. Statistical significance was set at p < 0.05. Statistical analyses were conducted using R version 4.2.2 (via RStudio). 2.6. Use of Artificial Intelligence In writing this paper, we used OpenAI's ChatGPT (GPT‐4o) and Anthropic's Claude (Claude 3 Opus) to improve the readability of the English text and perform proofreading. After using these services, all authors reviewed and edited the content as necessary, and they assume full responsibility for the content of this paper. 3. Results Of the 2 654 567 individuals registered in the SKDB, 25 101 women with a gestational‐age modifier were identified as pregnant. Of these, 6362 women with available residential information and complete data for a baseline period of 1 year and a follow‐up period of 280 days were included in the analysis (Figure 1). Baseline characteristics are shown in Table S2. During the follow‐up period, 222 women were diagnosed with HDP, yielding a proportion of 3.5%. After assessing and addressing multicollinearity among candidate predictors (Table S3), multivariable logistic regression analysis revealed that advanced maternal age (OR 1.05, 95% CI 1.02–1.07), diabetes mellitus (OR 3.21, 95% CI 1.22–8.47), preexisting hypertension (OR 2.38, 95% CI 1.04–5.46), and SLE (OR 5.90, 95% CI 1.41–24.65) were associated with increased risk (Table 1). TABLE 1. | Variable (reference) | Category or unit | Univariate model | Multivariate model | |---|---|---|---| | Crude OR [95% CI] | Adjusted OR [95% CI] | || | Age, mean (SD) (year) | Per 1‐year increase | 1.06 [1.03–1.08] | 1.05 [1.02–1.07] | | Area of residence (Government‐designated cities) | Non‐government‐designated cities | 1.04 [0.79–1.36] | 1.00 [0.76–1.33] | | Amenorrhea (absence) | Presence | 2.03 [0.73–5.65] | 1.37 [0.45–4.15] | | Anemia (absence) | Presence | 1.66 [0.97–2.84] | 1.24 [0.69–2.20] | | Antiphospholipid antibody syndrome (absence) | Presence | 3.96 [0.49–32.36] | 1.00 [0.07–13.52] | | Benign uterine tumor (absence) | Presence | 1.69 [0.90–3.15] | 1.14 [0.59–2.19] | | Diabetes (absence) | Presence | 7.11 [3.07–16.45] | 3.21 [1.22–8.47] | | Endometriosis (absence) | Presence | 1.51 [0.66–3.47] | 1.26 [0.53–3.01] | | Helicobacter pylori infection (absence) | Presence | 3.20 [1.13–9.08] | 2.45 [0.83–7.18] | | Hyperlipidemia (absence) | Presence | 3.21 [1.64–6.25] | 1.32 [0.58–3.00] | | Hypertension (absence) | Presence | 4.67 [2.28–9.58] | 2.38 [1.04–5.46] | | Hyperthyroidism (absence) | Presence | 2.48 [1.06–5.78] | 1.91 [0.78–4.72] | | Hypothyroidism (absence) | Presence | 1.44 [1.06–1.95] | 0.90 [0.62–1.31] | | Liver disease (absence) | Presence | 2.63 [1.31–5.28] | 1.76 [0.84–3.71] | | Obesity (absence) | Presence | 4.66 [1.36–15.93] | 1.96 [0.48–8.07] | | Periodontal disease (absence) | Presence | 1.07 [0.80–1.42] | 0.99 [0.74–1.33] | | Polycystic ovary syndrome (absence) | Presence | 2.55 [1.36–4.80] | 2.00 [0.99–4.07] | | Renal failure (absence) | Presence | 2.60 [1.24–5.43] | 1.32 [0.58–3.04] | | Systemic lupus erythematosus (absence) | Presence | 9.33 [2.51–34.71] | 5.90 [1.41–24.65] | | Venous thrombosis (absence) | Presence | Not estimable a | Not included b | Abbreviations: CI, confidence interval; OR, odds ratio; SD, standard deviation. Complete separation due to absence of HDP events in exposed individuals. Excluded from the multivariable model because of complete separation. In sensitivity analyses excluding women with preexisting hypertension, advanced maternal age (OR 1.05, 95% CI 1.02–1.07), diabetes mellitus (OR 3.03, 95% CI 1.03–8.96), obesity (OR 4.03, 95% CI 1.02–15.92), and SLE (OR 6.30, 95% CI 1.29–30.75) remained independently associated with the onset of HDP. CIT analysis (Figure 2) identified maternal age as the primary discriminator of HDP risk, with a threshold at 34 years. Among women aged > 34 years, hyperlipidemia further increased the risk. Furthermore, among women aged > 34 years without hyperlipidemia, hypertension and PCOS provided additional stratification. Specifically, the proportions of HDP were 2.6% (Node 2, ≤ 34 years), 4.7% (Node 6, > 34 years, no hyperlipidemia, no hypertension, no PCOS), 13.5% (Node 7, > 34 years, no hyperlipidemia, no hypertension, with PCOS), 20.0% (Node 8, > 34 years, no hyperlipidemia, with hypertension), 10.5% (Node 10, > 34 years, with hyperlipidemia, without liver disease), and 45.5% (Node 11, > 34 years, with hyperlipidemia and liver disease). Consistently, logistic regression with women aged ≤ 34 years (Node 2) as the reference confirmed this pattern: moderately increased odds were observed in women aged > 34 years without hyperlipidemia, hypertension, or PCOS (Node 6; OR 1.84, 95% CI 1.38–2.45), whereas markedly increased odds were observed in women with PCOS (Node 7; OR 5.79, 95% CI 2.34–12.32), hypertension (Node 8; OR 9.30, 95% CI 3.05–23.44), hyperlipidemia without liver disease (Node 10; OR 4.38, 95% CI 1.29–11.21), and both hyperlipidemia and liver disease (Node 11; OR 31.01, 95% CI 8.83–104.39). 4. Discussion In this population‐based cohort study using the SKDB, we identified advanced maternal age, diabetes mellitus, preexisting hypertension, and SLE as independent predictors of HDP during pregnancy. The overall proportion of patients with HDP in our study was 3.5%. This figure is lower than the 6.37% reported by Maeda et al. [8], likely reflecting our stricter outcome definition focusing on new‐onset gestational hypertension, preeclampsia, HELLP syndrome, and eclampsia (O13, O14, and O15), while excluding chronic hypertension and superimposed preeclampsia. Notably, the combined incidence of preeclampsia/superimposed preeclampsia and gestational hypertension in the COPE study was 3.45% [6], which is comparable to our findings, further supporting the validity of our outcome definition. Additionally, our proportion (3.5%) was lower than estimates from Western countries ranging from 5% to 10% [1, 2]. This difference may reflect our stricter outcome definition focusing on gestational hypertension, preeclampsia, HELLP syndrome, and eclampsia (O13, O14, and O15), while excluding chronic hypertension and unspecified hypertensive disorders. To our knowledge, this is one of the few population‐based studies to systematically evaluate risk factors for HDP in a Japanese cohort using administrative claims data. Given that most existing prediction models have been developed in Western populations, our findings provide important evidence for risk stratification in Asian women, who may differ in baseline cardiovascular risk profiles and disease prevalence. Advanced maternal age was consistently associated with HDP onset, in agreement with previous studies linking vascular aging, metabolic dysfunction, and impaired placentation to hypertensive complications during pregnancy [1, 20]. Among the comorbidities, diabetes mellitus and preexisting hypertension showed strong associations with HDP risk [1, 2, 20], supporting the importance of preconception cardiovascular and metabolic health. These associations may reflect underlying endothelial dysfunction, vascular maladaptation, and impaired placentation. In addition, SLE was strongly associated with HDP risk [2, 20, 21], consistent with previous evidence implicating chronic inflammation, endothelial dysfunction, and immune‐mediated vascular injury in the pathogenesis of HDP. Buyon et al. additionally identified preexisting hypertension, lupus anticoagulant positivity, and SLE disease activity as important predictors of HDP in pregnancies complicated by SLE [22]. Furthermore, sensitivity analyses excluding women with preexisting hypertension showed a generally consistent pattern of association, with maternal age, diabetes mellitus, obesity, and SLE remaining associated with HDP risk, suggesting that these findings were not solely driven by baseline hypertension status. Notably, obesity emerged as an additional predictor in this analysis. This is consistent with previous findings in Japanese women showing that high body mass index is associated with increased HDP risk [23], and suggests that obesity‐related metabolic dysfunction may contribute to HDP development even in the absence of preexisting hypertension. To further explore interactions among risk factors, CIT analysis illustrated the hierarchical interplay of metabolic and vascular risk factors for HDP. Maternal age > 34 years emerged as the primary splitting variable, followed by hyperlipidemia and hypertension, suggesting that age‐related cardiometabolic vulnerability may play a central role in HDP development. Among older women without hyperlipidemia and preexisting hypertension, PCOS was associated with increased HDP risk [24], highlighting the potential contribution of underlying metabolic and endocrine dysfunction even in the absence of overt cardiovascular disease. In contrast, among older women with hyperlipidemia, concomitant liver disease identified a subgroup with particularly high HDP risk. Moreover, the elevated risk observed in women with hyperlipidemia is consistent with previous evidence linking hypercholesterolemia to increased HDP risk [25]. This finding should be interpreted with caution, as claims‐based liver disease codes are heterogeneous; however, it may partly reflect underlying metabolic liver conditions, which have been associated with gestational hypertension and preeclampsia [26, 27]. Taken together, our findings suggest that women with advanced maternal age, diabetes mellitus, preexisting hypertension, or SLE should be considered candidates for enhanced prenatal surveillance, including early blood pressure monitoring and timely referral to maternal‐fetal medicine specialists. The combination of conventional regression‐based risk profiling and tree‐based statistical approaches such as CIT analysis may enable clinicians to identify high‐risk subgroups that would not be captured by single‐factor screening alone. Given that these risk factors are routinely recorded in administrative claims databases, our findings support the feasibility of population‐level risk stratification for HDP without the need for additional clinical measurements. 4.1. Strengths and Limitations This study has several strengths. First, it was based on a large, population‐based claims database covering approximately 2.65 million residents in Shizuoka Prefecture, which is often regarded as a microcosm of Japan and is broadly representative of the Japanese population. Second, although the requirement for a 1‐year baseline period and a 280‐day follow‐up period reduced the number of eligible participants, the number of pregnant women identified in the SKDB (25101) was similar to the number of births recorded in the prefecture during the same period (26495), suggesting a high capture rate [28, 29]. Given Japan's universal health insurance coverage, a large proportion of pregnant women were likely captured because many sought medical care even for relatively minor conditions during pregnancy. Third, by redefining the outcome to focus on new‐onset HDP during pregnancy (O13, O14, and O15), we reduced potential overlap between the outcome definition and baseline chronic hypertension. Fourth, by combining multivariable logistic regression with CIT analysis, we quantified overall associations while exploring non‐linear interactions and identifying clinically meaningful high‐risk subgroups. Nevertheless, this study has some limitations. First, identifying pregnant women in the SKDB is inherently challenging because maternal and birth records are not directly linked. Consequently, the estimated start and end of pregnancy may be uncertain, and we could not account for competing risks, such as miscarriage or preterm birth, which may terminate pregnancy before HDP can occur. These competing risks mean that some women classified as not developing HDP may have ended their pregnancy earlier for other reasons. Second, many women were unable to secure a full baseline and follow‐up period, which reduced the eligible sample size and may have introduced some degree of selection bias. Third, the accuracy of HDP definitions based on claims data must be considered. Coding variability could not be ruled out. Fourth, our study could not include several important maternal and lifestyle factors that were not available in the claims data (such as gravidity, parity, plurality, height, weight, blood pressure, test results, smoking/alcohol consumption). Therefore, some associations may reflect the influence of unmeasured risk factors. Fifth, the study population consisted of women with at least one recorded medical diagnosis, which may have excluded healthy pregnancies. Consequently, the observed HDP risk might be higher than that in the general population, and some associations may be overestimated. However, recognizing these predictors remains important in clinical practice as they highlight potentially high‐risk groups that warrant closer monitoring and potentially targeted interventions. Sixth, the SKDB covers individuals enrolled in the National Health Insurance and the Late‐Stage Elderly Medical Care System, which primarily includes self‐employed individuals, retirees, and their dependents. Employees covered by workplace‐based health insurance were not included, which may limit the representativeness of the study population and the generalizability of our findings to the broader Japanese pregnant population. In conclusion, this population‐based retrospective cohort study focusing on new‐onset HDP during pregnancy identified advanced maternal age, diabetes mellitus, preexisting hypertension, and SLE as independent predictors of HDP onset. Furthermore, by accounting for non‐linear relationships in the CIT analysis, we identified maternal age > 34 years as the primary discriminator of HDP risk, with additional stratification by hyperlipidemia, hypertension, PCOS, and liver disease. These findings highlight the value of integrating complementary analytical approaches for early identification of high‐risk pregnancies, which may support more personalized prenatal care, timely interventions, and ultimately better maternal and neonatal outcomes. Author Contributions Yoko Sato: conceptualization, methodology, data curation, writing – review and editing, supervision, funding acquisition. Kei Takehara: conceptualization, methodology, data curation, visualization, writing – original draft, writing – review and editing, formal analysis, investigation. Yasuharu Tabara: conceptualization, writing – review and editing, funding acquisition. Takakazu Kawamura: conceptualization, writing – review and editing. Funding This work was supported by Shizuoka prefecture, SGUPH_2021_001_058. Disclosure An earlier version of this article was presented in abstract form at the 60th Annual Congress of the Japan Society of Perinatal and Neonatal Medicine, which was held in Osaka, Japan, on 13–15 July 2024. Ethics Statement This study protocol was reviewed and approved by the Medical Ethics Committee of the Shizuoka Graduate School of Public Health, Shizuoka, Japan (approval number: SGUPH_2021_001_058). Consent Because this was a retrospective study using anonymized administrative claims data, the ethics committee approved a waiver of written informed consent and permitted the use of an opt‐out procedure. An information document describing the study objectives, data to be used, and the right to refuse participation was posted on the university's website, and individuals who expressed their intention not to participate were excluded from the study. Conflicts of Interest The authors declare no conflicts of interest. Supporting information Acknowledgments This study was supported by Shizuoka Prefecture as part of a commissioned research project to promote the health of its citizens. The sponsor approved the publication of the results but played no role in the study design, execution, analysis, interpretation of data, or manuscript writing, and there were no restrictions on publishing results that might be unfavorable to the funding source. This paper has been edited by Editage English proofreading, a service that provides professional editing and proofreading services to help authors improve the quality and clarity of their written work. Data Availability Statement According to Shizuoka Prefecture's data use agreement with local insurers, individual‐level data are not publicly available. Access may be granted to qualified researchers who obtain the necessary approvals from both the data custodian and an accredited ethics committee. Further enquiries can be directed to the corresponding author.

- 1. Duckitt K. and Harrington D., “Risk Factors for Pre‐Eclampsia at Antenatal Booking: Systematic Review of Controlled Studies,” BMJ 330, no. 7491 (2005): 565, 10.1136/bmj.38380.674340.e0. [DOI] [PMC free article] [PubMed] [Google Scholar] - 2. Bartsch E., Medcalf K. E., Park A. L., and Ray J. G., “Clinical Risk Factors for Pre‐Eclampsia Determined in Early Pregnancy: Systematic Review and Meta‐Analysis of Large Cohort Studies,” BMJ 353 (2016): i1753, 10.1136/bmj.i1753. [DOI] [PMC free article] [PubMed] [Google Scholar] - 3. Magee L. A., Brown M. A., Hall D. R., et al., “The 2021 International Society for the Study of Hypertension in Pregnancy Classification, Diagnosis & Management Recommendations for International Practice,” Pregnancy Hypertens 27 (2022): 148–169, 10.1016/j.preghy.2021.09.008. [DOI] [PubMed] [Google Scholar] - 4. Jim B. and Karumanchi S. A., “Preeclampsia: Pathogenesis, Prevention, and Long‐Term Complications,” Seminars in Nephrology 37, no. 4 (2017): 386–397, 10.1016/j.semnephrol.2017.05.010. [DOI] [PubMed] [Google Scholar] - 5. Poon L. C., Shennan A., Hyett J. A., et al., “The International Federation of Gynecology and Obstetrics (FIGO) Initiative on Pre‐Eclampsia: A Pragmatic Guide for First‐Trimester Screening and Prevention,” International Journal of Gynaecology and Obstetrics 145, no. 1 (2019): 1–33, 10.1002/ijgo.12802. [DOI] [PMC free article] [PubMed] [Google Scholar] - 6. Ohkuchi A., Suzuki H., Matsubara K., et al., “Exponential Increase of the Gestational‐Age‐Specific Incidence of Preeclampsia Onset (COPE Study): A Multicenter Retrospective Cohort Study in Women With Maternal Check‐Ups at < 20 Weeks of Gestation in Japan,” Hypertension Research 45, no. 11 (2022): 1679–1689, 10.1038/s41440-022-01054-2. [DOI] [PubMed] [Google Scholar] - 7. Uchinuma H., Tsuchiya K., Horiuchi S., et al., “Advanced Maternal Age Elevates the Prevalence of Hypertensive Disorders in Women of Japanese, Independent of Blood Pressure: A Study From the Japan Environment and Children's Study,” Hypertension Research 48, no. 3 (2025): 904–913, 10.1038/s41440-024-02019-5. [DOI] [PubMed] [Google Scholar] - 8. Maeda N., Koyama M., Takatsuka S., et al., “Maternal Age and the Rising Incidence of Hypertensive Disorders of Pregnancy: A Comprehensive Analysis of National Claims Data From Japan,” PLoS One 20, no. 2 (2025): e0319177, 10.1371/journal.pone.0319177. [DOI] [PMC free article] [PubMed] [Google Scholar] - 9. Makita H., Haraguchi S., and Ueyama H., “Case Study of 35 Municipalities in Shizuoka Prefecture,” Regional Innovation 9 (2016): 3–14. [Google Scholar] - 10. Nakatani E., Tabara Y., Sato Y., Tsuchiya A., and Miyachi Y., “Data Resource Profile of Shizuoka Kokuho Database (SKDB) Using Integrated Health‐ and Care‐Insurance Claims and Health Checkups: The Shizuoka Study,” Journal of Epidemiology 32, no. 8 (2022): 391–400, 10.2188/jea.JE20210187. [DOI] [PMC free article] [PubMed] [Google Scholar] - 11. Shizuoka Graduate University of Public Health , “Shizuoka Kokuho Database (SKDB),” https://skdb.s‐sph.ac.jp/en/. - 12. Benchimol E. I., Smeeth L., Guttmann A., et al., “The REporting of Studies Conducted Using Observational Routinely‐Collected Health Data (RECORD) Statement,” PLoS Medicine 12, no. 10 (2015): e1001885, 10.1371/journal.pmed.1001885. [DOI] [PMC free article] [PubMed] [Google Scholar] - 13. Ishikawa T., Obara T., Miyachi Y., et al., “Identification of Pregnancy Status Within Health Insurance Claims Data in Japan,” Pharmacoepidemiology and Drug Safety 27, no. 7 (2018): 751–762, 10.1002/pds.4412. [DOI] [PubMed] [Google Scholar] - 14. Horton J., Sterrenburg M., Lane S., Maheshwari A., Li T. C., and Cheong Y., “Reproductive, Obstetric, and Perinatal Outcomes of Women With Adenomyosis and Endometriosis: A Systematic Review and Meta‐Analysis,” Human Reproduction Update 25, no. 5 (2019): 592–602, 10.1093/humupd/dmz012. [DOI] [PubMed] [Google Scholar] - 15. Bellos I., Daskalakis G., and Pergialiotis V., “ Helicobacter pylori Infection Increases the Risk of the Onset of Preeclampsia: A Meta‐Analysis of Observational Studies,” International Journal of Clinical Practice 72, no. 2 (2018): e13027, 10.1111/ijcp.13027. [DOI] [PubMed] [Google Scholar] - 16. Nourollahpour Shiadeh M., Riahi S. M., Adam I., et al., “ Helicobacter pylori Infection and Risk of Preeclampsia: A Systematic Review and Meta‐Analysis,” Journal of Maternal‐Fetal & Neonatal Medicine 32, no. 2 (2019): 324–331, 10.1080/14767058.2017.1378331. [DOI] [PubMed] [Google Scholar] - 17. Medici M., Korevaar T. I., Schalekamp‐Timmermans S., et al., “Maternal Early‐Pregnancy Thyroid Function Is Associated With Subsequent Hypertensive Disorders of Pregnancy: The Generation R Study,” Thyroid 24, no. 11 (2014): 1777–1785, 10.1089/thy.2014.0272. [DOI] [PubMed] [Google Scholar] - 18. Ahmed M. A., Hassan N. G., Omer M. E., et al., “Management of Hyperthyroidism During Pregnancy: A Systematic Literature Review,” Journal of Maternal‐Fetal & Neonatal Medicine 36, no. 12 (2023): 2023–2030, 10.1080/14767058.2022.2054482. [DOI] [Google Scholar] - 19. Karimi N., Samiee N., and Moradi Y., “The Association Between Periodontal Disease and the Risk of Adverse Maternal or Neonatal Outcomes: A Systematic Review and Meta‐Analysis,” Health Science Reports 6, no. 10 (2023): e1630, 10.1002/hsr2.1630. [DOI] [PMC free article] [PubMed] [Google Scholar] - 20. Umesawa M. and Kobashi G., “Epidemiology of Hypertensive Disorders in Pregnancy: Prevalence, Risk Factors, and Outcomes,” Hypertension Research 40, no. 3 (2017): 213–220, 10.1038/hr.2016.126. [DOI] [PubMed] [Google Scholar] - 21. Wind M., Fierro J. J., Bloemenkamp K. W. M., et al., “Pregnancy Outcome Predictors in Systemic Lupus Erythematosus: A Systematic Review and Meta‐Analysis,” Lancet Rheumatol 6, no. 10 (2024): e667–e683, 10.1016/S2665-9913(24)00160-7. [DOI] [PubMed] [Google Scholar] - 22. Buyon J. P., Kim M. Y., Guerra M. M., et al., “Predictors of Pregnancy Outcomes in Patients With Lupus: A Cohort Study,” Annals of Internal Medicine 163, no. 3 (2015): 153–163, 10.7326/M14-2235. [DOI] [PMC free article] [PubMed] [Google Scholar] - 23. Poudel K., Kobayashi S., Miyashita C., et al., “Hypertensive Disorders During Pregnancy (HDP), Maternal Characteristics, and Birth Outcomes Among Japanese Women: A Hokkaido Study,” International Journal of Environmental Research and Public Health 18, no. 7 (2021): 3342, 10.3390/ijerph18073342. [DOI] [PMC free article] [PubMed] [Google Scholar] - 24. Palomba S., Santagni S., Falbo A., and La Sala G. B., “Complications and Challenges Associated With Polycystic Ovary Syndrome: Current Perspectives,” International Journal of Women's Health 7 (2015): 745–763, 10.2147/IJWH.S70314. [DOI] [PMC free article] [PubMed] [Google Scholar] - 25. Thadhani R., Stampfer M. J., Hunter D. J., Manson J. E., Solomon C. G., and Curhan G. C., “High Body Mass Index and Hypercholesterolemia: Risk of Hypertensive Disorders of Pregnancy,” Obstetrics and Gynecology 94, no. 4 (1999): 543–550, 10.1016/s0029-7844(99)00400-7. [DOI] [PubMed] [Google Scholar] - 26. Hagström H., Höijer J., Ludvigsson J. F., et al., “Adverse Outcomes of Pregnancy in Women With Non‐Alcoholic Fatty Liver Disease,” Liver International 36, no. 2 (2016): 268–274, 10.1111/liv.12902. [DOI] [PubMed] [Google Scholar] - 27. Zhang L., Gao S., Luan Y., et al., “Predictivity of Hepatic Steatosis Index for Gestational Hypertension and Preeclampsia: A Prospective Cohort Study,” International Journal of Medical Sciences 22, no. 4 (2025): 834–844, 10.7150/ijms.104943. [DOI] [PMC free article] [PubMed] [Google Scholar] - 28. Shizuoka Prefecture , “National Health Insurance Business Status in 2017,” https://www.pref.shizuoka.jp/_res/projects/default_project/_page_/001/024/893/h29jigyojokyo.pdf. - 29. Shizuoka Prefecture , “National Health Insurance Business Status in 2021,” https://www.pref.shizuoka.jp/_res/projects/default_project/_page_/001/024/893/r3kokuhojigyojokyo.pdf. Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Data Availability Statement According to Shizuoka Prefecture's data use agreement with local insurers, individual‐level data are not publicly available. Access may be granted to qualified researchers who obtain the necessary approvals from both the data custodian and an accredited ethics committee. Further enquiries can be directed to the corresponding author.