Effect of hospital delivery volume and travel time on maternal postpartum health service use in Norway: a nationwide population‑based cohort study

Effect of hospital delivery volume and travel time on maternal postpartum health service use in Norway: a nationwide population‑based cohort study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Effect of hospital delivery volume and travel time on maternal postpartum health service use in Norway: a nationwide population‑based cohort study Tingting Feng, Andreas Asheim, Sara Marie Nilsen, Signe Opdahl, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8796064/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract BACKGROUND We investigated the effect of hospital delivery volume and travel time on maternal postpartum health service use in Norway. METHODS We conducted a population-based study including women giving birth from 2008 to 2021. Exposures were ( 1 ) annual delivery volume at the observed delivery hospital and travel time by car from the municipality center to the observed delivery hospital, and ( 2 ) expected delivery volume and expected travel time based on hospital catchment patterns. The primary outcome was maternal postpartum health service use. We used pseudo–maximum likelihood Poisson regression with adjustment for maternal characteristics, timing of birth, and pre-pregnancy health service use; within-woman analyses used fixed effects. RESULTS The study included 792,330 childbirths by 492,080 women. In the first approach, higher hospital volume (2000 vs. 500 births/year) was associated with 15% more days of GP contacts with a psychiatric diagnosis code (RR 1.15, 95% CI 1.05–1.26). Longer travel time (120-min vs. 30-min) was associated with a 20% longer delivery admission (RR 1.20, 95% CI 1.12–1.28), 5% more days of GP contact (RR 1.05, 95% CI 1.03–1.07), and 10% more days of GP contacts for conversation therapy (RR 1.10, 95% CI 1.01–1.20). In the within-woman analysis, these associations largely disappeared, except for a 13–19% increase in out-of-hours GP contacts at higher volume hospitals. CONCLUSIONS We found little evidence of a causal effect of hospital volume or travel time on overall postpartum health service use. Associations in conventional analyses may reflect residual confounding. postpartum health hospital delivery volume travel time maternal health health service use Norway BACKGROUND Centralising health care to larger facilities has occurred in most countries including Norway ( 1 ). Debate continues regarding whether centralisation improves or compromises patient outcomes ( 2 ). Previous studies have shown conflicting results regarding the association between hospital delivery volume and postpartum outcomes such as severe maternal morbidity and maternal death ( 3 – 8 ). Less is known about general postpartum outcomes and health service use. For associations between travel distance to delivery institutions and postpartum outcomes, the evidence suggests that travel time of more than one hour may increase risk of severe maternal morbidity ( 2 , 9 – 14 ). In Norway, pregnant women with conditions associated with increased risk of complications during labour are referred to give birth at specialized regional hospitals according to established selection criteria ( 15 ). Therefore, women with high‑risk pregnancies living in catchment areas of small hospitals often give birth at larger institutions. Most previous research compared deliveries across hospitals while adjusting for available risk factors ( 1 , 2 ), but such approaches assume that all relevant confounders are measured. Given increased travel distances following centralisation, understanding impacts on maternal postpartum health service use is important for planning maternity services. Observed associations between hospital volume, travel time and patient outcomes do not inevitably indicate that volume or travel time has causal effects on postpartum outcomes ( 16 – 21 ). Residual confounding may remain because hospital choice is influenced by factors not fully captured in registries. Evidence on causal effects of hospital delivery volume and travel time on postpartum health service use in high‑income settings is scarce. Only one study—the Sichuan instrumental variable analysis of 810,049 deliveries, investigated maternal outcomes after delivery and suggested benefits of higher volume, though generalisability is uncertain ( 6 ). To bridge this gap, we used two complementary study designs to investigate effects of hospital delivery volume and travel time on postpartum health service use. Following a previous Norwegian study ( 22 ), we compared ( 1 ) maternal postpartum health service use across hospitals of different sizes/travel times after adjusting for confounders, and ( 2 ) postpartum health service use among women who moved between catchment areas, controlling for time‑invariant confounding. METHODS Study setting This nationwide population‑based cohort study used linked Norwegian registry data. Norway provides publicly funded antenatal and obstetric care ( 23 ). Each hospital has responsibility for the population in its catchment area. Women with high-risk pregnancies are recommended to deliver at larger, specialized hospitals that manage a higher volume of complex and high-risk births. Study population Data on all births in Norway between 2008 and 2021 were retrieved from the Norwegian Patient Registry ( 24 ). Births were identified using Diagnosis-Related Group (DRG) codes 370–375 based on ICD‑10 diagnoses and relevant procedures. Data were linked with Statistics Norway (maternal education, residence), the Norwegian Control and Payment of Health Reimbursements Database (primary care) ( 24 ), and the Norwegian Patient Registry (specialist care) ( 24 ). Analytical design We used two separate analytical approaches in our study. First, comparison of maternal health service use was made between women who gave birth at hospitals with different delivery volumes, and between women with different travel times to the hospitals where they gave birth, respectively. This approach was done with adjustments for a range of observed potential confounders, including characteristics of the woman, time of birth, and previous contacts with health care services. The premise of this analysis is that all relevant differences among women who gave birth at high-volume hospitals compared to those at low-volume hospitals can be measured and accounted for. Second, the data were analysed as a natural experiment, comparing births among women who moved between hospital catchment areas. By comparing births within the same woman, this is effectively a quasi-experimental sibling design. This approach has the potential to control for fixed maternal characteristics (e.g., genetics, chronic conditions) by using each woman as her own control for observed and unobserved time-invariant confounders. Based on their residential relocation, we assumed these women had distinctly different probability of giving birth at different hospitals before and after their move. The analysis used the volume and travel time to the expected delivery hospitals as the exposures, rather than the volume and the travel time to the observed delivery hospitals. The definition of the expected hospitals is described below and was based on hospital catchment areas and the mother’s municipality of residence at delivery. Thus, this approach could be seen as analogous to an intention to treat analysis in a randomized trial and could avoid biased from selection of more severe cases to larger hospitals (confounding by indication) ( 21 ). Previous research investigated effects of hospital delivery volume and travel time on perinatal mortality with a similar approach ( 3 ). Exposures Observed-hospital analysis: delivery volume (annual births per hospital) and travel time (minutes by car from municipality center). Within-woman analysis: expected delivery volume and expected travel time, calculated using probability-weighted hospital use by municipality-year ( 22 ). When determining expected hospitals based on women’s residence, hospitals with less than 10% of the municipality's annual births were excluded due to the possibility of births occurring away from home, without an official address change, or referrals. Only births during the previous and following year were considered, so that the index birth did not affect the definition of catchment area. For women in a given municipality and year, the proportion of births expected to take place in each of the municipality's candidate hospitals, is therefore assumed to be the probability of delivery at that hospital for women from the same municipality in the previous and following year. Travel time was calculated as the probability-weighted mean travel time by car from the municipality to delivery institutions based on data from Norwegian Road Administration databases ( 25 ). Expected travel time was weighted the same way as expected hospital volume. Outcomes While comprehensive clinical data on maternal postpartum outcomes was unavailable, we used healthcare use outcomes that serve as indicators of maternal outcomes after birth. The following health service use were included: 1. Days with general practitioner (GP) contact within 1 year after delivery 2. Days with out of hours GP contact within 1 year after delivery 3. Days with emergency hospital contithin 1 year after delivery 4. Duration of hospital stay within 1 year after delivery excluding the initial delivery stay (including somatic and mental health care) 5. Days of GP contacts with psychiatric diagnosis code within 1 year after birth 6. Days with contacts for mental health care in the specialist health service within 1 year after delivery 7. Days of GP contacts for conversation therapy within 1 year after birth 8. Duration of stay in hospital for delivery admission These outcomes reflect common somatic and mental health–related postpartum care needs. Statistical analyses Pseudo-maximum likelihood Poisson regression estimated relative risks ( 26 ). Non-hierarchical clustering of errors was allowed by using the ppmlhdfe modules for Stata ( 27 , 28 ). Women who moved between catchment areas were identified and within-woman associations were estimated by using a fixed effects estimator with the mother’s id-variable as the fixed effect ( 27 , 28 ). Restricted cubic splines with 5 knots were estimated by using Stata’s mkspline-function to explore non-linear associations ( 29 ). The cubic splines were visualized as comparisons with a reference level at 500 births per year and 30 minutes travel time, respectively. Analyses were adjusted for the following maternal factors: age in categories (< 20 years, 20–24 years, 25–29 years, 30–34 years, 35 years and above), parity (1, 2, 3, or more previous births), maternal education level (primary, secondary, higher), immigration status, days with contact with health care services in the year prior to birth (sum of days with any GP contact and contacts with specialist mental health care). We also adjusted for year, month and weekday of delivery. Finally, we adjusted for the sum of Diagnostic-Related Groups weights, which may indicate treatment intensity in specialist healthcare. When the exposure was expected hospital delivery volume, expected travel time was adjusted using the restricted cubic spline. Vice versa, expected hospital delivery volume was adjusted when expected travel time was the exposure. Precision was estimated as 95% confidence intervals (CI) with robust standard errors, clustered by the woman’s id-variable and municipality of residence. All statistical analyses were conducted using Stata/MP 18.0 for Windows. Ethical approval Approved by REK‑Midt (2016/2159; 01.02.2017). Data were de‑identified; informed consent was waived. This study was conducted in accordance with the principles of the Declaration of Helsinki. Results In the period 2008–2021, we identified 792,330 childbirths by 492,080 women; 106,305 births (13.4%) occurred among 58,226 women who moved between catchment areas.Table 1 provides descriptive characteristics of the women who moved across a catchment area border between births. In the first approach, there was little evidence of a substantial association between hospital volume and the postpartum maternal health service use, see Table 2. The exception was days of GP contacts with psychiatric diagnosis code within 1 year after birth. The number of days with GP contact with a psychiatric diagnosis code was 15% higher (rate ratio RR 1.15 with 95% CI 1.05 - 1.26) when the hospital volume was 2000 births/year compared to the hospital volume with 500 births/year. Higher volumes ≥4000 showed no clear associations. For travel time, women with 120-minute travel time had longer delivery admission (RR = 1.20, 95% CI: 1.12 -1.28), slightly more GP contacts (RR = 1.05, 95% CI: 1.03 - 1.07), more GP conversation therapy contacts (RR = 1.10, 95% CI: 1.01-1.20), and slightly higher emergency contacts (RR = 1.05, 95% CI: 0.97 – 1.14),compared with a 30-minute travel time, see Table 3. And a similar change when the travel time increased to 180 minutes and 240 minutes, suggesting the effect of travel time begins to plateau beyond 120 minutes. Absolute numbers are shown in Supporting Information Tables S1-S2. In the second approach, most associations attenuated, see Table 2. We found that there were 13% more days with out of hours GP contact within 1 year after birth (RR = 1.13, 95% CI: 1.01 -1.26) when the hospital volume was 2000 births/year compared to a hospital volume of 500 births/year (Table 2), and 19% (RR = 1.19, 95% CI: 1.06-1.32) more days of contact when the hospital volume was 4000 births/year. Absolute numbers are also presented in Supporting Information Table S1. For travel time, we found little evidence of substantial associations between travel time and postpartum health service use, see Table 3 and Supporting Information Table S2. Spline models (Supporting Information Figures S1-S4) show similar patterns. Discussion In this nationwide population-based cohort study, we address a knowledge gap in the ongoing debate about centralizing maternity care in Norway. While previous studies report conflicting results (3-8), Our dual design helps distinguish associations from potential confounding (16). Our conventional analyses suggested small associations between delivery volume or travel time and some postpartum service outcomes. However, within‑woman analyses showed little evidence of causal effects. This divergence suggests that conventional analysis may reflect residual confounding, such as socioeconomic disparities or other unmeasured confounders, rather than causal effects of hospital volume or travel time. Our findings align with prior evidence of weak or inconsistent associations. For instance, a US cohort study (3) found that lower hospital volume was not associated with severe maternal morbidity among high-risk women. A population-based Australian study (8) demonstrated that lower hospital volume was not associated with adverse outcomes that included postpartum hemorrhage, perineal trauma (3rd/4th degree tears), maternal sepsis/infection and thromboembolic events for low risk women in Australia. In addition, A systematic review of 13 observational studies from the U.S., Europe, Canada, and Australia emphasized that no consistent relation was found between hospital delivery volume and maternal outcomes including maternal mortality and severe maternal morbidity for low-risk women (7). However, other observational studies from North America, Europe, and Asia indicate associations between hospital volume/travel time and maternal outcomes (3-6). A Chinese population-based study (6) found that higher volume in China’s tiered healthcare system was associated with reduced risk of severe maternal morbidity after birth. This is in contrast to our study, where higher hospital volume was not associated with improved maternal outcomes in Norway’s universal healthcare system when accounting for constant, unobserved factors. Such discrepancies might reflect that differences in healthcare systems could impact maternal outcomes. For travel time, A 2022 systematic review (9) found minimal effects on maternal outcomes (e.g., maternal mortality, postpartum hemorrhage, severe perineal tears) in high-income countries. A 2020 systematic review (10) found weak evidence that longer travel time was associated with worse maternal outcomes in high-income countries, including outcomes such as maternal mortality, severe perineal trauma, postpartum hemorrhage, maternal admission to intensive care units and maternal blood transfusion. In contrast, subsequent studies showed associations between longer travel time and increased risk of severe maternal postpartum complications in Sweden, Canada and the US (2, 11-14, 30). Critically, to our knowledge, our study is the first to use a within-woman quasi-experimental design to study effects of travel time on maternal health service use after childbirth, and this approach offers more reliable causal evidence by considering unobserved risk factors. Using a quasi-experimental design, we found weak or null associations between hospital delivery volume or travel time and postpartum healthcare use. Some associations observed in conventional analyses may reflect residual confounding. Therefore, when making policy decisions about centralization or decentralization on birth institutions, caution is essential regarding evidence based only on associations from non-causal analytical approaches. Also, differences in healthcare systems should be carefully considered when comparing results across studies. Future research should use methods that address confounding to better understand factors influencing maternal postpartum health service use. Our study has several strengths. As a nationwide population-based study using comprehensive registry data, we minimized selection bias and ensured high generalizability to high-income countries with universal healthcare systems. Our approach comparing the same women before and after moving between catchment areas reduced unmeasured confounding, thereby strengthening causal inference. Use of administrative records for outcome measures (e.g., hospital stays, GP contacts) eliminated recall bias that is common in self-reported studies. The registry linkages enabled complete tracking of maternal postpartum health service use without loss to follow-up. Our study examined a comprehensive range of maternal postpartum health service use including both somatic and mental health indicators. Unlike previous studies that focused only on severe morbidity or delivery outcomes alone, our outcomes provide a more complete assessment of maternal postpartum health outcomes within one year after birth. Lastly, we compared conventional adjusted analyses with a quasi-experiment design to address unmeasured confounding to increase the reliability of causal inference. Meanwhile, several limitations should be considered. Although moving between hospital catchment areas may control for time-invariant confounders (e.g., genetics), time-varying factors (e.g., job loss, new diagnoses) might bias our estimates. The registry linkage lacked detailed socioeconomic information and data on behavioral factors that may change between births. Furthermore, healthcare use does not capture all aspects of maternal postpartum health, particularly conditions such as undiagnosed depression or unreported complications, and other important outcomes such as level of satisfaction or pain. Health service use reflects a complex interplay between the condition itself, characteristics of the patient and the doctor, and societal factors. Hence, we cannot rule out that measuring maternal outcomes in other ways could change the results. Our findings are specific to Norway's well-developed universal healthcare system and might not be generalizable to low-income countries with fragmented healthcare systems. Conclusion Our study used both a conventional and a quasi-experimental approach to investigate associations between hospital volume or travel time and maternal postpartum health service use. The conventional analyses showed some associations between hospital volume/travel time with GP visits or hospital stays. However, these associations were not seen in the quasi-experimental approach, which found weak evidence of a causal effect. This discrepancy suggests that the associations observed in the conventional analyses may reflect residual confounding rather than causal effects. Consequently, policy decisions regarding centralization of birth institutions should not rely solely on conventional analyses. Declarations Ethics approval and consent to participate The study and data linkage were approved by the Regional Committee for Ethics in Medical Research (REK-Midt 2016/2159; 01.02.2017). Data were de-identified before transfer to the researchers, and informed consent was waived in accordance with applicable regulations. This study was conducted in accordance with the principles of the Declaration of Helsinki. Consent for publication Not applicable. Availability of data and materials The data that support the findings of this study are available from the Norwegian Patient Registry and Statistics Norway, but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of the Norwegian Patient Registry and Statistics Norway. Competing interests The authors declare that they have no competing interests. Funding This work was supported by the Research Council of Norway (grant number 295989). Authors’ contributions TF conducted the data analysis and drafted the manuscript. JHB supervised the project and contributed to study design and manuscript revision. AA contributed to the statistical methodology and data analysis. SMN supported project planning and manuscript revision. SO and HNB provided critical input on epidemiological aspects and women’s health. KR contributed clinical expertise and interpretation of findings. FC provided input on interpretation of findings and manuscript revision. All authors read and approved the final manuscript. Acknowledgements We thank the Norwegian Research Council for funding this study. References Polin K, Hjortland M, Maresso A, van Ginneken E, Busse R, Quentin W. Top-Three health reforms in 31 high-income countries in 2018 and 2019: an expert informed overview. Health Policy. 2021;125(7):815–32. Holowko N, Haas J, Ahlberg M, Stephansson O, Örtqvist A. More than time: travel time to the delivery ward and maternal outcomes–onset of labour, postpartum haemorrhage and obstetric anal sphincter injury. Public Health. 2023;217:105–14. Bozzuto L, Passarella M, Lorch S, Srinivas S. Effects of delivery volume and high-risk condition volume on maternal morbidity among high-risk obstetric patients. Obstet Gynecol. 2019;133(2):261–8. Campbell KH, Illuzzi JL, Lee HC, Lin H, Lipkind HS, Lundsberg LS, et al. Optimal maternal and neonatal outcomes and associated hospital characteristics. Birth. 2019;46(2):289–99. Snowden JM, Cheng YW, Emeis CL, Caughey AB. The impact of hospital obstetric volume on maternal outcomes in term, non–low-birthweight pregnancies. Am J Obstet Gynecol. 2015;212(3):380. e1–. e9. Chen N, Pan J. The causal effect of delivery volume on severe maternal morbidity: an instrumental variable analysis in Sichuan, China. BMJ Global Health. 2022;7(5):e008428. Walther F, Kuester D, Bieber A, Malzahn J, Rüdiger M, Schmitt J. Are birth outcomes in low risk birth cohorts related to hospital birth volumes? A systematic review. BMC Pregnancy Childbirth. 2021;21(1):1–16. Tracy SK, Sullivan E, Dahlen H, Black D, Wang YA, Tracy MB. General obstetrics: does size matter? A population-based study of birth in lower volume maternity hospitals for low risk women. BJOG: Int J Obstet Gynecol. 2006;113(1):86–96. Ames HM, Forsetlund L, Larun L, Hval G, Muller A. How far is too far? A systematic review of travel distance to delivery institutions and negative maternal and natal outcomes. Res Square. 2022. Malouf RS, Tomlinson C, Henderson J, Opondo C, Brocklehurst P, Alderdice F, et al. Impact of obstetric unit closures, travel time and distance to obstetric services on maternal and neonatal outcomes in high-income countries: a systematic review. BMJ open. 2020;10(12):e036852. Engjom HM, Morken N-H, Høydahl E, Norheim OF, Klungsøyr K. Risk of eclampsia or HELLP-syndrome by institution availability and place of delivery–A population-based cohort study. Pregnancy Hypertens. 2018;14:1–8. Kozhimannil KB, Interrante JD, Henning-Smith C, Admon LK. Rural-urban differences in severe maternal morbidity and mortality in the US, 2007–15. Health Aff. 2019;38(12):2077–85. Minion SC, Krans EE, Brooks MM, Mendez DD, Haggerty CL. Association of driving distance to maternity hospitals and maternal and perinatal outcomes. Obstet Gynecol. 2022;140(5):812–9. Aubrey-Bassler FK, Cullen RM, Simms A, Asghari S, Crane J, Wang PP, et al. Population‐based cohort study of hospital delivery volume, geographic accessibility, and obstetric outcomes. Int J Gynecol Obstet. 2019;146(1):95–102. Helsedirektoratet. Nasjonal faglig retningslinje for fødselsomsorgen 2024 [Available from: https://www.helsedirektoratet.no/retningslinjer/fodselsomsorgen HernánMA R. Causal inference: what if. Boca Raton: Chapman & Hall/CRC; 2020. Gaynor M, Seider H, Vogt WB. The volume-outcome effect, scale economies, and learning-by-doing. Am Econ Rev. 2005;95(2):243–7. Hentschker C, Mennicken R. The volume–outcome relationship revisited: practice indeed makes perfect. Health Serv Res. 2018;53(1):15–34. Chou YY, Hwang JJ, Tung YC. Optimal surgeon and hospital volume thresholds to reduce mortality and length of stay for CABG. PLoS ONE. 2021;16(4):e0249750. Kahn JM, Ten Have TR, Iwashyna TJ. The relationship between hospital volume and mortality in mechanical ventilation: an instrumental variable analysis. Health Serv Res. 2009;44(3):862–79. Baiocchi M, Cheng J, Small DS. Instrumental variable methods for causal inference. Stat Med. 2014;33(13):2297–340. Asheim A, Nilsen SM, Opdahl S, Risnes K, Balstad Magnussen E, Carlsen F, et al. The Effects of Hospital Delivery Volume and Travel Time on Perinatal Mortality and Delivery in Transit: Causal Inference with Triangulation. Epidemiology. 2025. 10.1097/EDE.0000000000001840 . Ringard Å, Sagan A, Sperre Saunes I, Lindahl AK, Organization WH. Norway: health system review. 2013. Bakken IJ, Ariansen AM, Knudsen GP, Johansen KI, Vollset SE. The Norwegian Patient Registry and the Norwegian Registry for Primary Health Care: Research potential of two nationwide health-care registries. Scand J Public Health. 2020;48(1):49–55. Sand Ø, Johansen BG, Halse AH, Sæther SO. Road Tolls in Norway, 2005–2021. Oslo; 2022. p. 8248019551. Report No. Greenland S. Model-based Estimation of Relative Risks and Other Epidemiologic Measures in Studies of Common Outcomes and in Case-Control Studies. Am J Epidemiol. 2004;160(4):301–5. Correia S, Guimarães P, Zylkin T. ppmlhdfe: Fast Poisson estimation with high-dimensional fixed effects. arXiv org. 2019. Correia S. A feasible estimator for linear models with multi-way fixed effects. Preprint at http://scorreiacom/research/hdfe pdf . 2016. Harrell FE. Regression modeling strategies: with applications to linear models, logistic regression, and survival analysis. Springer; 2001. Luke S, Hobbs A, Mak S, Der K, Pederson A, Schummers L. Travel time to delivery, antenatal care, and birth outcomes: a population-based cohort of uncomplicated pregnancies in British Columbia, 2012–2019. J Obstet Gynecol Can. 2022;44(8):886–94. Tables Tables are available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files supplementarytable.docx Sfig1.pdf Sfig2.pdf Sfig3.pdf Sfig4.pdf Tables.docx Cite Share Download PDF Status: Under Review Version 1 posted Reviewers invited by journal 10 Mar, 2026 Editor invited by journal 11 Feb, 2026 Editor assigned by journal 10 Feb, 2026 Submission checks completed at journal 10 Feb, 2026 First submitted to journal 05 Feb, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8796064","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":590814453,"identity":"b4b09bc5-8d6d-47bf-8a6a-74a5cb4e7276","order_by":0,"name":"Tingting Feng","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA70lEQVRIiWNgGAWjYDACZhBhAGIcABIVcglshHTwgLQcAGlhSwASZ4yJ0AIiQOaDtTC2GScQdJc9O+/h1x8K7tg1sDEfe/h1nkEeHwPvwQf4HcaXZnHA4FlyAxtburHsNoNiNga+ZAP8WnjMDA4YHE5mkO8xk5bc9iexjYHHTII4LWz836Ql5xgQpcX4AVCLHQMbD5vkxwZitBzmMWM4Y3A4gY2NzUya4RhQC9AQvH5h7z9j/KHiz2F7fjbmZ5I/agwS57f3GD7ApwUI2EDOALoHmADAscRMQD1IyQcgYQ9iMf4grHoUjIJRMApGIAAAJoQ+isOClAwAAAAASUVORK5CYII=","orcid":"","institution":"Norwegian University of Science and Technology","correspondingAuthor":true,"prefix":"","firstName":"Tingting","middleName":"","lastName":"Feng","suffix":""},{"id":590814454,"identity":"370c9272-4f43-4941-a22c-490c359bfbf9","order_by":1,"name":"Andreas Asheim","email":"","orcid":"","institution":"St. Olav's University Hospital","correspondingAuthor":false,"prefix":"","firstName":"Andreas","middleName":"","lastName":"Asheim","suffix":""},{"id":590814455,"identity":"d1683be0-fc0a-42de-b382-53c8c60e8cdc","order_by":2,"name":"Sara Marie Nilsen","email":"","orcid":"","institution":"St. Olav's University Hospital","correspondingAuthor":false,"prefix":"","firstName":"Sara","middleName":"Marie","lastName":"Nilsen","suffix":""},{"id":590814456,"identity":"13b77daa-c7de-44c1-97eb-7c79d6aa0c08","order_by":3,"name":"Signe Opdahl","email":"","orcid":"","institution":"Norwegian University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Signe","middleName":"","lastName":"Opdahl","suffix":""},{"id":590814458,"identity":"d48f8b56-009a-4e70-9033-9c39491e0ce6","order_by":4,"name":"Hanne Nissen Bjørnsen","email":"","orcid":"","institution":"Norwegian Institute of Public Health","correspondingAuthor":false,"prefix":"","firstName":"Hanne","middleName":"Nissen","lastName":"Bjørnsen","suffix":""},{"id":590814459,"identity":"500a4045-fc47-424a-9b25-f1185f9088ea","order_by":5,"name":"Kari Risnes","email":"","orcid":"","institution":"Norwegian University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Kari","middleName":"","lastName":"Risnes","suffix":""},{"id":590814460,"identity":"58a25c1a-59df-4807-941d-50b34c333db1","order_by":6,"name":"Fredrik Carlsen","email":"","orcid":"","institution":"Norwegian University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Fredrik","middleName":"","lastName":"Carlsen","suffix":""},{"id":590814461,"identity":"03623dfa-66e7-424c-897d-9cf0c178a488","order_by":7,"name":"Johan Håkon Bjørngaard","email":"","orcid":"","institution":"Norwegian University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Johan","middleName":"Håkon","lastName":"Bjørngaard","suffix":""}],"badges":[],"createdAt":"2026-02-05 11:08:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8796064/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8796064/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":103051274,"identity":"e7bfd749-2775-4df5-ab9a-31aac3955f93","added_by":"auto","created_at":"2026-02-20 07:59:15","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":504224,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8796064/v1/74143ae0-ecea-49fa-b7fa-1107d438a025.pdf"},{"id":102819433,"identity":"24050d80-3258-4ba0-b8b9-9ae3f87bb64e","added_by":"auto","created_at":"2026-02-17 06:59:03","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":25585,"visible":true,"origin":"","legend":"","description":"","filename":"supplementarytable.docx","url":"https://assets-eu.researchsquare.com/files/rs-8796064/v1/497268190c8919b78293b999.docx"},{"id":102819431,"identity":"7a13e6d0-2028-4d0b-9545-4d61ff0c3c0f","added_by":"auto","created_at":"2026-02-17 06:59:03","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":476385,"visible":true,"origin":"","legend":"","description":"","filename":"Sfig1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8796064/v1/55da911b05e5db63ad1776e5.pdf"},{"id":103049199,"identity":"98ce495a-53d7-4f74-b7d8-e4c3d57e9b17","added_by":"auto","created_at":"2026-02-20 07:37:55","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":512963,"visible":true,"origin":"","legend":"","description":"","filename":"Sfig2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8796064/v1/2e04de1507a9a69f91452e67.pdf"},{"id":102819430,"identity":"6b460b71-8ea9-4e6e-bc7c-b3f5318a56e8","added_by":"auto","created_at":"2026-02-17 06:59:03","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":507609,"visible":true,"origin":"","legend":"","description":"","filename":"Sfig3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8796064/v1/ebb6a07156d161dafa97af21.pdf"},{"id":102819434,"identity":"197643f4-e9ed-40d8-bcf0-e37fc2f9024b","added_by":"auto","created_at":"2026-02-17 06:59:03","extension":"pdf","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":519131,"visible":true,"origin":"","legend":"","description":"","filename":"Sfig4.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8796064/v1/7ffe3178356d1a3feddac8b8.pdf"},{"id":102962634,"identity":"4c8462be-8ed3-4d77-8551-4fbd3d4e4d47","added_by":"auto","created_at":"2026-02-19 04:10:12","extension":"docx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":27824,"visible":true,"origin":"","legend":"","description":"","filename":"Tables.docx","url":"https://assets-eu.researchsquare.com/files/rs-8796064/v1/287574420d3464f66c2ab221.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Effect of hospital delivery volume and travel time on maternal postpartum health service use in Norway: a nationwide population‑based cohort study","fulltext":[{"header":"BACKGROUND","content":"\u003cp\u003eCentralising health care to larger facilities has occurred in most countries including Norway (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). Debate continues regarding whether centralisation improves or compromises patient outcomes (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). Previous studies have shown conflicting results regarding the association between hospital delivery volume and postpartum outcomes such as severe maternal morbidity and maternal death (\u003cspan additionalcitationids=\"CR4 CR5 CR6 CR7\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). Less is known about general postpartum outcomes and health service use. For associations between travel distance to delivery institutions and postpartum outcomes, the evidence suggests that travel time of more than one hour may increase risk of severe maternal morbidity (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan additionalcitationids=\"CR10 CR11 CR12 CR13\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn Norway, pregnant women with conditions associated with increased risk of complications during labour are referred to give birth at specialized regional hospitals according to established selection criteria (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). Therefore, women with high‑risk pregnancies living in catchment areas of small hospitals often give birth at larger institutions. Most previous research compared deliveries across hospitals while adjusting for available risk factors (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e), but such approaches assume that all relevant confounders are measured. Given increased travel distances following centralisation, understanding impacts on maternal postpartum health service use is important for planning maternity services.\u003c/p\u003e \u003cp\u003eObserved associations between hospital volume, travel time and patient outcomes do not inevitably indicate that volume or travel time has causal effects on postpartum outcomes (\u003cspan additionalcitationids=\"CR17 CR18 CR19 CR20\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). Residual confounding may remain because hospital choice is influenced by factors not fully captured in registries. Evidence on causal effects of hospital delivery volume and travel time on postpartum health service use in high‑income settings is scarce. Only one study\u0026mdash;the Sichuan instrumental variable analysis of 810,049 deliveries, investigated maternal outcomes after delivery and suggested benefits of higher volume, though generalisability is uncertain (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTo bridge this gap, we used two complementary study designs to investigate effects of hospital delivery volume and travel time on postpartum health service use. Following a previous Norwegian study (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e), we compared (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) maternal postpartum health service use across hospitals of different sizes/travel times after adjusting for confounders, and (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) postpartum health service use among women who moved between catchment areas, controlling for time‑invariant confounding.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\n \u003ch2\u003eStudy setting\u003c/h2\u003e\n \u003cp\u003eThis nationwide population‑based cohort study used linked Norwegian registry data. Norway provides publicly funded antenatal and obstetric care (\u003cspan class=\"CitationRef\"\u003e23\u003c/span\u003e). Each hospital has responsibility for the population in its catchment area. Women with high-risk pregnancies are recommended to deliver at larger, specialized hospitals that manage a higher volume of complex and high-risk births.\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003eStudy population\u003c/h3\u003e\n\u003cp\u003eData on all births in Norway between 2008 and 2021 were retrieved from the Norwegian Patient Registry (\u003cspan class=\"CitationRef\"\u003e24\u003c/span\u003e). Births were identified using Diagnosis-Related Group (DRG) codes 370\u0026ndash;375 based on ICD‑10 diagnoses and relevant procedures. Data were linked with Statistics Norway (maternal education, residence), the Norwegian Control and Payment of Health Reimbursements Database (primary care) (\u003cspan class=\"CitationRef\"\u003e24\u003c/span\u003e), and the Norwegian Patient Registry (specialist care) (\u003cspan class=\"CitationRef\"\u003e24\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eAnalytical design\u003c/h3\u003e\n\u003cp\u003eWe used two separate analytical approaches in our study.\u003c/p\u003e\n\u003cp\u003eFirst, comparison of maternal health service use was made between women who gave birth at hospitals with different delivery volumes, and between women with different travel times to the hospitals where they gave birth, respectively. This approach was done with adjustments for a range of observed potential confounders, including characteristics of the woman, time of birth, and previous contacts with health care services. The premise of this analysis is that all relevant differences among women who gave birth at high-volume hospitals compared to those at low-volume hospitals can be measured and accounted for.\u003c/p\u003e\n\u003cp\u003eSecond, the data were analysed as a natural experiment, comparing births among women who moved between hospital catchment areas. By comparing births within the same woman, this is effectively a quasi-experimental sibling design. This approach has the potential to control for fixed maternal characteristics (e.g., genetics, chronic conditions) by using each woman as her own control for observed and unobserved time-invariant confounders. Based on their residential relocation, we assumed these women had distinctly different probability of giving birth at different hospitals before and after their move. The analysis used the volume and travel time to the expected delivery hospitals as the exposures, rather than the volume and the travel time to the observed delivery hospitals. The definition of the expected hospitals is described below and was based on hospital catchment areas and the mother\u0026rsquo;s municipality of residence at delivery. Thus, this approach could be seen as analogous to an intention to treat analysis in a randomized trial and could avoid biased from selection of more severe cases to larger hospitals (confounding by indication) (\u003cspan class=\"CitationRef\"\u003e21\u003c/span\u003e). Previous research investigated effects of hospital delivery volume and travel time on perinatal mortality with a similar approach (\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eExposures\u003c/h3\u003e\n\u003cp\u003eObserved-hospital analysis: delivery volume (annual births per hospital) and travel time (minutes by car from municipality center).\u003c/p\u003e\n\u003cp\u003eWithin-woman analysis: expected delivery volume and expected travel time, calculated using probability-weighted hospital use by municipality-year (\u003cspan class=\"CitationRef\"\u003e22\u003c/span\u003e). When determining expected hospitals based on women\u0026rsquo;s residence, hospitals with less than 10% of the municipality\u0026apos;s annual births were excluded due to the possibility of births occurring away from home, without an official address change, or referrals. Only births during the previous and following year were considered, so that the index birth did not affect the definition of catchment area. For women in a given municipality and year, the proportion of births expected to take place in each of the municipality\u0026apos;s candidate hospitals, is therefore assumed to be the probability of delivery at that hospital for women from the same municipality in the previous and following year. Travel time was calculated as the probability-weighted mean travel time by car from the municipality to delivery institutions based on data from Norwegian Road Administration databases (\u003cspan class=\"CitationRef\"\u003e25\u003c/span\u003e). Expected travel time was weighted the same way as expected hospital volume.\u003c/p\u003e\n\u003ch3\u003eOutcomes\u003c/h3\u003e\n\u003cp\u003eWhile comprehensive clinical data on maternal postpartum outcomes was unavailable, we used healthcare use outcomes that serve as indicators of maternal outcomes after birth. The following health service use were included:\u003c/p\u003e\n\u003cp\u003e1. Days with general practitioner (GP) contact within 1 year after delivery\u003c/p\u003e\n\u003cp\u003e2. Days with out of hours GP contact within 1 year after delivery\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cspan\u003e3. Days with emergency hospital contithin 1 year after delivery\u003cbr\u003e\u003c/span\u003e\u003cspan\u003e4. Duration of hospital stay within 1 year after delivery excluding the initial delivery stay (including somatic and mental health care)\u003cbr\u003e\u003c/span\u003e\u003cspan\u003e5. Days of GP contacts with psychiatric diagnosis code within 1 year after birth\u003cbr\u003e\u003c/span\u003e\u003cspan\u003e6. Days with contacts for mental health care in the specialist health service within 1 year after delivery\u003cbr\u003e\u003c/span\u003e\u003cspan\u003e7. Days of GP contacts for conversation therapy within 1 year after birth\u003cbr\u003e\u003c/span\u003e\u003cspan\u003e8. Duration of stay in hospital for delivery admission\u003cbr\u003e\u003c/span\u003e\u003c/p\u003e\n\u003cp\u003eThese outcomes reflect common somatic and mental health\u0026ndash;related postpartum care needs.\u003c/p\u003e\n\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\n \u003ch2\u003eStatistical analyses\u003c/h2\u003e\n \u003cp\u003ePseudo-maximum likelihood Poisson regression estimated relative risks (\u003cspan class=\"CitationRef\"\u003e26\u003c/span\u003e). Non-hierarchical clustering of errors was allowed by using the ppmlhdfe modules for Stata (\u003cspan class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e28\u003c/span\u003e). Women who moved between catchment areas were identified and within-woman associations were estimated by using a fixed effects estimator with the mother\u0026rsquo;s id-variable as the fixed effect (\u003cspan class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e28\u003c/span\u003e).\u003c/p\u003e\n \u003cp\u003eRestricted cubic splines with 5 knots were estimated by using Stata\u0026rsquo;s mkspline-function to explore non-linear associations (\u003cspan class=\"CitationRef\"\u003e29\u003c/span\u003e). The cubic splines were visualized as comparisons with a reference level at 500 births per year and 30 minutes travel time, respectively.\u003c/p\u003e\n \u003cp\u003eAnalyses were adjusted for the following maternal factors: age in categories (\u0026lt;\u0026thinsp;20 years, 20\u0026ndash;24 years, 25\u0026ndash;29 years, 30\u0026ndash;34 years, 35 years and above), parity (1, 2, 3, or more previous births), maternal education level (primary, secondary, higher), immigration status, days with contact with health care services in the year prior to birth (sum of days with any GP contact and contacts with specialist mental health care). We also adjusted for year, month and weekday of delivery. Finally, we adjusted for the sum of Diagnostic-Related Groups weights, which may indicate treatment intensity in specialist healthcare. When the exposure was expected hospital delivery volume, expected travel time was adjusted using the restricted cubic spline. Vice versa, expected hospital delivery volume was adjusted when expected travel time was the exposure. Precision was estimated as 95% confidence intervals (CI) with robust standard errors, clustered by the woman\u0026rsquo;s id-variable and municipality of residence. All statistical analyses were conducted using Stata/MP 18.0 for Windows.\u003c/p\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003eEthical approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eApproved by REK‑Midt (2016/2159; 01.02.2017). Data were de‑identified; informed consent was waived. This study was conducted in accordance with the principles of the Declaration of Helsinki.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eIn the period 2008\u0026ndash;2021, we identified 792,330 childbirths by 492,080 women; 106,305 births (13.4%) occurred among 58,226 women who moved between catchment areas.Table 1 provides descriptive characteristics of the women who moved across a catchment area border between births.\u003c/p\u003e\n\u003cp\u003eIn the first approach, there was little evidence of a substantial association between hospital volume and the postpartum maternal health service use, see Table 2. The exception was days of GP contacts with psychiatric diagnosis code within 1 year after birth. The number of days with GP contact with a psychiatric diagnosis code was 15% higher (rate ratio RR 1.15 with 95% CI 1.05 - 1.26) when the hospital volume was 2000 births/year compared to the hospital volume with 500 births/year. Higher volumes \u0026ge;4000 showed no clear associations. For travel time, women with 120-minute travel time had longer delivery admission (RR = 1.20, 95% CI: 1.12 -1.28), slightly more GP contacts (RR = 1.05, 95% CI: 1.03 - 1.07), more GP conversation therapy contacts (RR = 1.10, 95% CI: 1.01-1.20), and slightly higher emergency contacts (RR = 1.05, 95% CI: 0.97 \u0026ndash; 1.14),compared with a 30-minute travel time, see Table 3. And a similar change when the travel time increased to 180 minutes and 240 minutes, suggesting the effect of travel time begins to plateau beyond 120 minutes. Absolute numbers are shown in Supporting Information Tables S1-S2. \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn the second approach, most associations attenuated, see Table 2. We found that there were 13% more days with out of hours GP contact within 1 year after birth (RR = 1.13, 95% CI: 1.01 -1.26) when the hospital volume was 2000 births/year compared to a hospital volume of 500 births/year (Table 2), and 19% (RR = 1.19, 95% CI: 1.06-1.32) more days of contact when the hospital volume was 4000 births/year. Absolute numbers are also presented in Supporting Information Table S1. For travel time, we found little evidence of substantial associations between travel time and postpartum health service use, see Table 3 and Supporting Information Table S2. Spline models (Supporting Information Figures S1-S4) show similar patterns.\u003c/p\u003e\n"},{"header":"Discussion","content":"\u003cp\u003eIn this nationwide population-based cohort study, we address a knowledge gap in the ongoing debate about centralizing maternity care in Norway. While previous studies report conflicting results (3-8), Our dual design helps distinguish associations from potential confounding (16).\u003c/p\u003e\n\u003cp\u003eOur conventional analyses suggested small associations between delivery volume or travel time and some postpartum service outcomes. However, within‑woman analyses showed little evidence of causal effects. This divergence suggests that conventional analysis may reflect residual confounding, such as socioeconomic disparities or other unmeasured confounders, rather than causal effects of hospital volume or travel time.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOur findings align with prior evidence of weak or inconsistent associations. For instance, a US cohort study (3) found that lower hospital volume was not associated with severe maternal morbidity among \u0026nbsp; high-risk \u0026nbsp;women. A population-based Australian study (8) demonstrated that lower hospital volume was not associated with adverse outcomes that included postpartum hemorrhage, perineal trauma (3rd/4th degree tears), maternal sepsis/infection and thromboembolic events for low risk women in Australia. In addition, A systematic review of 13 observational studies from the U.S., Europe, Canada, and Australia emphasized that no consistent relation was found between hospital delivery volume and maternal outcomes including maternal mortality and severe maternal morbidity for low-risk women (7). However, other observational studies from North America, Europe, and Asia indicate associations between hospital volume/travel time and maternal outcomes (3-6). A Chinese population-based study (6) found that \u0026nbsp;higher volume in China\u0026rsquo;s tiered healthcare system was associated with reduced risk of severe maternal morbidity after birth. This is in contrast to our study, where higher hospital volume was not associated with improved maternal outcomes in Norway\u0026rsquo;s universal healthcare system when accounting for constant, unobserved factors. Such discrepancies might reflect that differences in healthcare systems could impact maternal outcomes.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFor travel time, A 2022 systematic review (9) found minimal effects on maternal outcomes (e.g., maternal mortality, postpartum hemorrhage, severe perineal tears) in high-income countries. A 2020 systematic review (10) found weak evidence that longer travel time was associated with worse maternal outcomes in high-income countries, including outcomes such as maternal mortality, severe perineal trauma, postpartum hemorrhage, maternal admission to intensive care units and maternal blood transfusion. In contrast, subsequent studies showed associations between longer travel time and increased risk of severe maternal postpartum complications in Sweden, Canada and the US (2, 11-14, 30). Critically, to our knowledge, our study is the first to use a within-woman quasi-experimental design to study effects of travel time on maternal health service use after childbirth, and this approach offers more reliable causal evidence by considering unobserved risk factors.\u003c/p\u003e\n\u003cp\u003eUsing a quasi-experimental design, we found weak or null associations between hospital delivery volume or travel time and postpartum healthcare use. Some associations observed in conventional analyses may reflect residual confounding. Therefore, when making policy decisions about centralization or decentralization on birth institutions, caution is essential regarding evidence based only on associations from non-causal analytical approaches. \u0026nbsp;Also, differences in healthcare systems should be carefully considered when comparing results across studies. Future research should use methods that address confounding to better understand factors influencing maternal postpartum health service use.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOur study has several strengths. As a nationwide population-based study using comprehensive registry data, we minimized selection bias and ensured high generalizability to high-income countries with universal healthcare systems. Our approach comparing the same women before and after moving between catchment areas reduced unmeasured confounding, thereby strengthening causal inference. Use of administrative records for outcome measures (e.g., hospital stays, GP contacts) eliminated recall bias that is common in self-reported studies. The registry linkages enabled complete tracking of maternal postpartum health service use without loss to follow-up. Our study examined a comprehensive range of maternal postpartum health service use including both somatic and mental health indicators. Unlike previous studies that focused only on severe morbidity or delivery outcomes alone, our outcomes provide a more complete assessment of maternal postpartum health outcomes within one year after birth. Lastly, we compared conventional adjusted analyses with a quasi-experiment design to address unmeasured confounding to increase the reliability of causal inference.\u003c/p\u003e\n\u003cp\u003eMeanwhile, several limitations should be considered. Although moving between hospital catchment areas may control for time-invariant confounders (e.g., genetics), time-varying factors (e.g., job loss, new diagnoses) might bias our estimates. The registry linkage lacked detailed socioeconomic information and data on behavioral factors that may change between births. Furthermore, healthcare use does not capture all aspects of maternal postpartum health, particularly conditions such as undiagnosed depression or unreported complications, and other important outcomes such as level of satisfaction or pain. Health service use reflects a complex interplay between the condition itself, characteristics of the patient and the doctor, and societal factors. Hence, we cannot rule out that measuring maternal outcomes in other ways could change the results. Our findings are specific to Norway\u0026apos;s well-developed universal healthcare system and might not be generalizable to low-income countries with fragmented healthcare systems.\u0026nbsp;\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eOur study used both a conventional and a quasi-experimental approach to investigate associations between hospital volume or travel time and maternal postpartum health service use. The conventional analyses showed some associations between hospital volume/travel time with GP visits or hospital stays. However, these associations were not seen in the quasi-experimental approach, which found weak evidence of a causal effect. This discrepancy suggests that the associations observed in the conventional analyses may reflect residual confounding rather than causal effects. Consequently, policy decisions regarding centralization of birth institutions should not rely solely on conventional analyses.\u0026nbsp;\u003c/p\u003e\n"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study and data linkage were approved by the Regional Committee for Ethics in Medical Research (REK-Midt 2016/2159; 01.02.2017). Data were de-identified before transfer to the researchers, and informed consent was waived in accordance with applicable regulations. This study was conducted in accordance with the principles of the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data that support the findings of this study are available from the Norwegian Patient Registry and Statistics Norway, but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of the Norwegian Patient Registry and Statistics Norway.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Research Council of Norway (grant number 295989).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTF conducted the data analysis and drafted the manuscript. JHB supervised the project and contributed to study design and manuscript revision. AA contributed to the statistical methodology and data analysis. SMN supported project planning and manuscript revision. SO and HNB provided critical input on epidemiological aspects and women\u0026rsquo;s health. KR contributed clinical expertise and interpretation of findings. FC provided input on interpretation of findings and manuscript revision. All authors read and approved the final manuscript.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank the Norwegian Research Council for funding this study.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003ePolin K, Hjortland M, Maresso A, van Ginneken E, Busse R, Quentin W. Top-Three health reforms in 31 high-income countries in 2018 and 2019: an expert informed overview. Health Policy. 2021;125(7):815\u0026ndash;32.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHolowko N, Haas J, Ahlberg M, Stephansson O, \u0026Ouml;rtqvist A. More than time: travel time to the delivery ward and maternal outcomes\u0026ndash;onset of labour, postpartum haemorrhage and obstetric anal sphincter injury. Public Health. 2023;217:105\u0026ndash;14.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBozzuto L, Passarella M, Lorch S, Srinivas S. Effects of delivery volume and high-risk condition volume on maternal morbidity among high-risk obstetric patients. Obstet Gynecol. 2019;133(2):261\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCampbell KH, Illuzzi JL, Lee HC, Lin H, Lipkind HS, Lundsberg LS, et al. Optimal maternal and neonatal outcomes and associated hospital characteristics. Birth. 2019;46(2):289\u0026ndash;99.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSnowden JM, Cheng YW, Emeis CL, Caughey AB. The impact of hospital obstetric volume on maternal outcomes in term, non\u0026ndash;low-birthweight pregnancies. Am J Obstet Gynecol. 2015;212(3):380. e1\u0026ndash;. e9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen N, Pan J. The causal effect of delivery volume on severe maternal morbidity: an instrumental variable analysis in Sichuan, China. BMJ Global Health. 2022;7(5):e008428.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWalther F, Kuester D, Bieber A, Malzahn J, R\u0026uuml;diger M, Schmitt J. Are birth outcomes in low risk birth cohorts related to hospital birth volumes? A systematic review. BMC Pregnancy Childbirth. 2021;21(1):1\u0026ndash;16.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTracy SK, Sullivan E, Dahlen H, Black D, Wang YA, Tracy MB. General obstetrics: does size matter? A population-based study of birth in lower volume maternity hospitals for low risk women. BJOG: Int J Obstet Gynecol. 2006;113(1):86\u0026ndash;96.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAmes HM, Forsetlund L, Larun L, Hval G, Muller A. How far is too far? A systematic review of travel distance to delivery institutions and negative maternal and natal outcomes. Res Square. 2022.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMalouf RS, Tomlinson C, Henderson J, Opondo C, Brocklehurst P, Alderdice F, et al. Impact of obstetric unit closures, travel time and distance to obstetric services on maternal and neonatal outcomes in high-income countries: a systematic review. BMJ open. 2020;10(12):e036852.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEngjom HM, Morken N-H, H\u0026oslash;ydahl E, Norheim OF, Klungs\u0026oslash;yr K. Risk of eclampsia or HELLP-syndrome by institution availability and place of delivery\u0026ndash;A population-based cohort study. Pregnancy Hypertens. 2018;14:1\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKozhimannil KB, Interrante JD, Henning-Smith C, Admon LK. Rural-urban differences in severe maternal morbidity and mortality in the US, 2007\u0026ndash;15. Health Aff. 2019;38(12):2077\u0026ndash;85.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMinion SC, Krans EE, Brooks MM, Mendez DD, Haggerty CL. Association of driving distance to maternity hospitals and maternal and perinatal outcomes. Obstet Gynecol. 2022;140(5):812\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAubrey-Bassler FK, Cullen RM, Simms A, Asghari S, Crane J, Wang PP, et al. Population‐based cohort study of hospital delivery volume, geographic accessibility, and obstetric outcomes. Int J Gynecol Obstet. 2019;146(1):95\u0026ndash;102.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHelsedirektoratet. Nasjonal faglig retningslinje for f\u0026oslash;dselsomsorgen 2024 [Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.helsedirektoratet.no/retningslinjer/fodselsomsorgen\u003c/span\u003e\u003cspan address=\"https://www.helsedirektoratet.no/retningslinjer/fodselsomsorgen\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHern\u0026aacute;nMA R. Causal inference: what if. Boca Raton: Chapman \u0026amp; Hall/CRC; 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGaynor M, Seider H, Vogt WB. The volume-outcome effect, scale economies, and learning-by-doing. Am Econ Rev. 2005;95(2):243\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHentschker C, Mennicken R. The volume\u0026ndash;outcome relationship revisited: practice indeed makes perfect. Health Serv Res. 2018;53(1):15\u0026ndash;34.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChou YY, Hwang JJ, Tung YC. Optimal surgeon and hospital volume thresholds to reduce mortality and length of stay for CABG. PLoS ONE. 2021;16(4):e0249750.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKahn JM, Ten Have TR, Iwashyna TJ. The relationship between hospital volume and mortality in mechanical ventilation: an instrumental variable analysis. Health Serv Res. 2009;44(3):862\u0026ndash;79.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaiocchi M, Cheng J, Small DS. Instrumental variable methods for causal inference. Stat Med. 2014;33(13):2297\u0026ndash;340.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAsheim A, Nilsen SM, Opdahl S, Risnes K, Balstad Magnussen E, Carlsen F, et al. The Effects of Hospital Delivery Volume and Travel Time on Perinatal Mortality and Delivery in Transit: Causal Inference with Triangulation. Epidemiology. 2025. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/EDE.0000000000001840\u003c/span\u003e\u003cspan address=\"10.1097/EDE.0000000000001840\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRingard \u0026Aring;, Sagan A, Sperre Saunes I, Lindahl AK, Organization WH. Norway: health system review. 2013.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBakken IJ, Ariansen AM, Knudsen GP, Johansen KI, Vollset SE. The Norwegian Patient Registry and the Norwegian Registry for Primary Health Care: Research potential of two nationwide health-care registries. Scand J Public Health. 2020;48(1):49\u0026ndash;55.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSand \u0026Oslash;, Johansen BG, Halse AH, S\u0026aelig;ther SO. Road Tolls in Norway, 2005\u0026ndash;2021. Oslo; 2022. p. 8248019551. Report No.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGreenland S. Model-based Estimation of Relative Risks and Other Epidemiologic Measures in Studies of Common Outcomes and in Case-Control Studies. Am J Epidemiol. 2004;160(4):301\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCorreia S, Guimar\u0026atilde;es P, Zylkin T. ppmlhdfe: Fast Poisson estimation with high-dimensional fixed effects. arXiv org. 2019.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCorreia S. A feasible estimator for linear models with multi-way fixed effects. Preprint at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://scorreiacom/research/hdfe pdf\u003c/span\u003e\u003cspan address=\"http://scorreiacom/research/hdfe pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. 2016.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarrell FE. Regression modeling strategies: with applications to linear models, logistic regression, and survival analysis. Springer; 2001.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLuke S, Hobbs A, Mak S, Der K, Pederson A, Schummers L. Travel time to delivery, antenatal care, and birth outcomes: a population-based cohort of uncomplicated pregnancies in British Columbia, 2012\u0026ndash;2019. J Obstet Gynecol Can. 2022;44(8):886\u0026ndash;94.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTables are available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-health-services-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bhsr","sideBox":"Learn more about [BMC Health Services Research](http://bmchealthservres.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/BHSR/default.aspx","title":"BMC Health Services Research","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"postpartum health, hospital delivery volume, travel time, maternal health, health service use, Norway","lastPublishedDoi":"10.21203/rs.3.rs-8796064/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8796064/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBACKGROUND\u003c/h2\u003e \u003cp\u003eWe investigated the effect of hospital delivery volume and travel time on maternal postpartum health service use in Norway.\u003c/p\u003e\u003ch2\u003eMETHODS\u003c/h2\u003e \u003cp\u003eWe conducted a population-based study including women giving birth from 2008 to 2021. Exposures were (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) annual delivery volume at the observed delivery hospital and travel time by car from the municipality center to the observed delivery hospital, and (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) expected delivery volume and expected travel time based on hospital catchment patterns. The primary outcome was maternal postpartum health service use. We used pseudo\u0026ndash;maximum likelihood Poisson regression with adjustment for maternal characteristics, timing of birth, and pre-pregnancy health service use; within-woman analyses used fixed effects.\u003c/p\u003e\u003ch2\u003eRESULTS\u003c/h2\u003e \u003cp\u003eThe study included 792,330 childbirths by 492,080 women. In the first approach, higher hospital volume (2000 vs. 500 births/year) was associated with 15% more days of GP contacts with a psychiatric diagnosis code (RR 1.15, 95% CI 1.05\u0026ndash;1.26). Longer travel time (120-min vs. 30-min) was associated with a 20% longer delivery admission (RR 1.20, 95% CI 1.12\u0026ndash;1.28), 5% more days of GP contact (RR 1.05, 95% CI 1.03\u0026ndash;1.07), and 10% more days of GP contacts for conversation therapy (RR 1.10, 95% CI 1.01\u0026ndash;1.20). In the within-woman analysis, these associations largely disappeared, except for a 13\u0026ndash;19% increase in out-of-hours GP contacts at higher volume hospitals.\u003c/p\u003e\u003ch2\u003eCONCLUSIONS\u003c/h2\u003e \u003cp\u003eWe found little evidence of a causal effect of hospital volume or travel time on overall postpartum health service use. Associations in conventional analyses may reflect residual confounding.\u003c/p\u003e","manuscriptTitle":"Effect of hospital delivery volume and travel time on maternal postpartum health service use in Norway: a nationwide population‑based cohort study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-17 06:58:58","doi":"10.21203/rs.3.rs-8796064/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewersInvited","content":"","date":"2026-03-10T08:43:38+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-02-11T10:13:35+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-10T09:02:48+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-10T09:01:11+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Health Services Research","date":"2026-02-05T10:18:56+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-health-services-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bhsr","sideBox":"Learn more about [BMC Health Services Research](http://bmchealthservres.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/BHSR/default.aspx","title":"BMC Health Services Research","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"cb336e86-d5af-48f3-8e6e-9cc704695b38","owner":[],"postedDate":"February 17th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-03-10T08:53:28+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-17 06:58:58","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8796064","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8796064","identity":"rs-8796064","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}