Advertisement

Economic Evaluation of Elective Cesarean Section on Maternal Request Compared With Planned Vaginal Birth—Application to Swedish Setting Using National Registry Data

Open AccessPublished:November 14, 2022DOI:https://doi.org/10.1016/j.jval.2022.10.003

      Highlights

      • Over- and underuse of cesarean section (CS) may be associated with increased risks of complications. A lack of consensus on the definition of CS on maternal request and binding guidelines leaves room for practice variations.
      • Previous economic evaluations have mostly focused on countries with high rates of CS or have used specific assumptions. This is an applied cost-effectiveness analysis of CS on maternal request compared with planned vaginal birth using national Swedish data.
      • In Sweden, planned vaginal birth leads to lower healthcare costs and somatic health gains compared with elective CS without medical indication.

      Abstract

      Objectives

      There is a lack of consensus around the definition of delivery by cesarean section (CS) on maternal request, and clinical practice varies across and within countries. Previous economic evaluations have focused on specific populations and selected complications. Our aim was to evaluate the cost-effectiveness of CS on maternal request compared with planned vaginal birth in a Swedish context, based on a systematic review of benefits and drawbacks and national registry data on costs.

      Methods

      We used the results from a systematic literature review of somatic risks for long- and short-term complications for mother and child, in which certainty was rated low, moderate, or high using the Grading of Recommendations Assessment, Development and Evaluation. Swedish national registry data were used for healthcare costs of delivery and complications. Utilities for long-term complications were based on a focused literature review. We constructed a decision tree and conducted separate analyses for primi- and multiparous women. Costs and effects were discounted by 3% and the time horizon was varied between 1 and 20 years.

      Results

      Planned vaginal birth leads to lower healthcare costs and somatic health gains compared with elective CS without medical indication over up to 20 years. Although there is uncertainty around, for example, quality-of-life effects, results remain stable across sensitivity analyses.

      Conclusions

      CS on maternal request leads to increased hospitalization costs in a Swedish setting, taking into account short- and long-term consequences for both mother and child. Future research needs to study the psychological consequences related to different delivery methods, costs in outpatient care, and productivity losses.

      Keywords

      Introduction

      The increase in rates of delivery by cesarean section (CS) has attracted wide interest globally. In certain countries, the share of CS is very high (eg, 44% of births in Latin America), whereas other countries can be considered to have a low share (4% in parts of Africa).
      • Boerma T.
      • Ronsmans C.
      • Melesse D.Y.
      • et al.
      Global epidemiology of use of and disparities in caesarean sections.
      Both over- and underuse of CS may be associated with increased risks for mother and child (eg, complications from surgery, injuries from vaginal birth, and possible long-term effects for the child). The global increase can partly be explained by a lack of midwives and economic incentive systems (eg, private caregivers).
      • Wiklund I.
      • Malata A.M.
      • Cheung N.F.
      • Cadee F.
      Appropriate use of caesarean section globally requires a different approach.
      In Sweden, the share of CS has been stable at around 18% since 2006, although there is a large variation among the 21 regions (13%-21%).
      Socialstyrelsen [National Board of Health and Welfare]
      Kejsarsnitt i Sverige 2008-2017. Kriterier som styr beslut om förlossningssätt, samt kartläggning av komplikationer. [Caesarean section in Sweden 2008-2017. Criteria that govern decisions about method of delivery, and mapping of complications].
      Approximately half of CS are planned, most with medical indications (such as placenta previa or breech presentation), and the other half are acute. Multiparous women constitute the largest group of women who request CS without a medical indication. Reasons for requesting a CS include a previous traumatic delivery, worry about complications, or fear of not receiving adequate support during birth.
      • Jenabi E.
      • Khazaei S.
      • Bashirian S.
      • Aghababaei S.
      • Matinnia N.
      Reasons for elective cesarean section on maternal request: a systematic review.
      In Sweden, socioeconomic factors do not play any major role in the request.
      Socialstyrelsen [National Board of Health and Welfare]
      Kejsarsnitt i Sverige 2008-2017. Kriterier som styr beslut om förlossningssätt, samt kartläggning av komplikationer. [Caesarean section in Sweden 2008-2017. Criteria that govern decisions about method of delivery, and mapping of complications].
      No legislation confers the right to demand a medical procedure in Sweden (except for abortion and sterilization), only to abstain from an intervention. Nevertheless, since medical care should be decided in consultation between caregiver and patient, the question arises regarding how much weight should be given to the woman’s request. This contrasts with the situation in other countries—for instance, in England, where women, for a few years now, have the legal right to demand a CS.
      There is a lack of consensus on what should be considered a CS on maternal request, leading to uncertain statistics in this area. Despite the existence of an International Classification of Diseases Tenth Revision diagnostic code often used in Sweden (O82.8), it is not applied consistently in clinical practice. This is interconnected with a general lack of binding guidelines in Sweden, leaving room for practice variations between clinics and regions. Therefore, the most common approach to evaluating outcomes after planned CS on maternal request is to exclude pregnancies in which a medical indication for planned CS can be discerned. In Sweden, an estimated 1% to 2% of primiparous women deliver with a planned CS on maternal request, compared with approximately 3% to 7% among multiparous women.
      Socialstyrelsen [National Board of Health and Welfare]
      Kejsarsnitt i Sverige 2008-2017. Kriterier som styr beslut om förlossningssätt, samt kartläggning av komplikationer. [Caesarean section in Sweden 2008-2017. Criteria that govern decisions about method of delivery, and mapping of complications].
      In addition to research on the medical implications of planned CS on maternal request compared with vaginal delivery, there has been a sizable amount of research on the costs and cost-effectiveness in different populations and settings. Most of these studies
      • Entringer A.P.
      • Pinto M.
      • Gomes M.
      Cost-effectiveness analysis of natural birth and elective C-section in supplemental health.
      • Gilbert S.A.
      • Grobman W.A.
      • Landon M.B.
      • et al.
      Cost-effectiveness of trial of labor after previous cesarean in a minimally biased cohort.
      • Gilbert S.A.
      • Grobman W.A.
      • Landon M.B.
      • et al.
      Lifetime cost-effectiveness of trial of labor after cesarean in the United States.
      • Liu T.C.
      • Loh C.P.
      • Lin C.L.
      • Chen C.S.
      Do cesarean deliveries on maternal request lead to greater maternal care utilization during six-month, one-year, and two-year postpartum periods? Evidence from NHI claims data in Taiwan.
      have been conducted in contexts with high rates of CS, specifically the United States (33%
      • Betran A.P.
      • Ye J.
      • Moller A.B.
      • Zhang J.
      • Gulmezoglu A.M.
      • Torloni M.R.
      The increasing trend in caesarean section rates: global, regional and national estimates: 1990-2014.
      ), Brazil (56%
      • Betran A.P.
      • Ye J.
      • Moller A.B.
      • Zhang J.
      • Gulmezoglu A.M.
      • Torloni M.R.
      The increasing trend in caesarean section rates: global, regional and national estimates: 1990-2014.
      ), and Taiwan (35%
      • Betran A.P.
      • Ye J.
      • Moller A.B.
      • Zhang J.
      • Gulmezoglu A.M.
      • Torloni M.R.
      The increasing trend in caesarean section rates: global, regional and national estimates: 1990-2014.
      ). The precise features of treatment practices and healthcare systems mean that results from these economic evaluations have low transferability to countries with lower rates of CS, such as Sweden. A few studies from a European context have been conducted, including 1 analysis in primiparous women in the United Kingdom
      Caesarean section (NICE clinical guideline 132). Royal College of Obstetricians & Gynaecologists.
      and 2 studies on multiparous women with a previous CS in Ireland
      • Fawsitt C.G.
      • Bourke J.
      • Greene R.A.
      • Everard C.M.
      • Murphy A.
      • Lutomski J.E.
      At what price? A cost-effectiveness analysis comparing trial of labour after previous caesarean versus elective repeat caesarean delivery.
      and 4 European countries.
      • Fobelets M.
      • Beeckman K.
      • Faron G.
      • Daly D.
      • Begley C.
      • Putman K.
      Vaginal birth after caesarean versus elective repeat caesarean delivery after one previous caesarean section: a cost-effectiveness analysis in four European countries.
      Moreover, these studies used varying time horizons and included different complications and types of costs. When applying a longer timeframe (more than 6 weeks postpartum), the overall conclusions changed with vaginal delivery remaining cost-saving in 2 of 4 countries,
      • Fobelets M.
      • Beeckman K.
      • Faron G.
      • Daly D.
      • Begley C.
      • Putman K.
      Vaginal birth after caesarean versus elective repeat caesarean delivery after one previous caesarean section: a cost-effectiveness analysis in four European countries.
      underlining the importance of the local context in this field.
      This study is part of a health technology assessment on the benefits and drawbacks of a CS on maternal request performed by the Swedish Agency for Health Technology Assessment and Assessment of Social Services (SBU).
      SBU
      Caesarean section on maternal request – A systematic review and assessment of medical, health economic, ethical, and social aspects. NCBI.
      Whereas SBU does not provide policy recommendations, its reports form a basis for decision-making both on a national and regional level. The focus of this article is on the health economic consequences of CS on maternal request compared with planned vaginal delivery, applied to a Swedish context. The results may be used by the government, care providers, and national organizations using knowledge-based management to inform treatment guidelines and recommendations.

      Methods

      Model Structure

      We constructed a decision tree using Microsoft Excel (Microsoft, Redmond, WA) (Fig. 1). The population covered low-risk women and singletons. Outcomes for both mother and child included short-term complications (within 6 weeks of delivery for the mother, 28 days of birth for the child), and long-term complications occurring more than 1 year after delivery or birth, respectively. Complications related to mental health were excluded. More details on the Population, Intervention, Control, and Outcomes format for the systematic review are available in the Supplemental materials (Appendix Table 1 in Supplemental Materials found at https://doi.org/10.1016/j.jval.2022.10.003). For this article, we followed the Consolidated Health Economic Evaluation Reporting Standards 2022.
      • Husereau D.
      • Drummond M.
      • Augustovski F.
      • et al.
      Consolidated health economic evaluation reporting standards 2022 (CHEERS 2022) statement: updated reporting guidance for health economic evaluations.
      A health economic analysis plan was developed following the initiation of the systematic review on effects. There was no patient engagement during the project.
      Figure thumbnail gr1
      Figure 1Model structure.
      CS indicates cesarean section; M, multiparous women; P, primiparous women.
      The planned mode of delivery can result in different actual modes of delivery, as illustrated in the decision tree (Fig. 1). In the case of a planned CS, actual delivery most often occurs as an elective (planned) CS, but may, in some cases, end up as a vaginal delivery (eg, before the planned date) or as an acute CS. A planned vaginal delivery can either end up as vaginal delivery or as an acute CS. Reflecting the selection criteria of the systematic review, short-term complications were assumed to occur during the first year after delivery (given yearly modeling units), whereas long-term complications were assumed to occur during the second year after delivery. Since the systematic review of effects revealed no difference in mortality for the mother or child between the 2 methods of delivery, mortality was not included in the model.

      Perspective, Populations, and Method of Delivery

      The economic analysis has a healthcare perspective, focusing on direct medical costs, reflecting the outcomes investigated in the systematic review. Analyses were conducted for different time horizons: 1 year, 10 years, and 20 years. For time horizons beyond 1 year, costs and effects were discounted by 3% annually.
      TLV
      Tandvårds- och läkemedelsförmånsverkets allmänna råd. TLVAR 2017:1. [The Dental and Pharmaceutical Benefits Agency’s general guidance].
      Analyses were conducted for primiparous and multiparous women separately. The main difference between the populations lies in the probability of the actual method of delivery. An approximation of the planned method of delivery was performed based on data for full-term single vertex deliveries from the Swedish Medical Birth Register from 2015 to 2017.
      Socialstyrelsen [National Board of Health and Welfare]
      Medicinska födelseregistret. [National Medical Birth Register].
      The register includes the type of labor onset (spontaneous, induction, or CS) and mode of delivery (vaginal noninstrumental, vacuum extraction, forceps delivery, or CS). The CS was considered planned if it was performed before the onset of labor, and acute if performed during ongoing labor. An assumption was made that all planned CS ultimately occurred as planned CS.
      Primiparous women were assumed to have a mean age of 29 years and multiparous women, 32 years.
      Socialstyrelsen [National Board of Health and Welfare]
      Kejsarsnitt i Sverige 2008-2017. Kriterier som styr beslut om förlossningssätt, samt kartläggning av komplikationer. [Caesarean section in Sweden 2008-2017. Criteria that govern decisions about method of delivery, and mapping of complications].
      The age affects the utility weights of the general population when using a long-term perspective.

      Risks of Complications

      The risks of complications were derived from the meta-analyses of results for the mother and child from the systematic review.
      SBU
      Caesarean section on maternal request – A systematic review and assessment of medical, health economic, ethical, and social aspects. NCBI.
      Complications were included in the model if the results were statistically significant, and if certainty was rated low, moderate, or high using the Grading of Recommendations Assessment, Development and Evaluation.
      The GRADE working group. GRADE.
      Some complications fulfilling these criteria were excluded to avoid double-counting, for instance, owing to the short duration and low impact (eg, antibiotic treatment) or to a need for more conclusive evidence (see Table 1). Baseline risks for complications after planned vaginal delivery were based on the frequency of cases in the total cohorts of studies included in the meta-analyses, as vaginal deliveries constituted the majority of all deliveries. Baseline and relative risks are presented in Table 1; both types of data were varied in sensitivity analyses.
      Table 1Maternal and child complications used in the model: baseline risks and relative risks with corresponding 95% CI.
      ComplicationBaseline risk (planned VD), %Relative risk (planned CS)95% CI
      Mother
      Notes on maternal complications: antibiotic treatment was not included as related costs were expected to be negligible. The following maternal complications were not included to avoid double-counting of complications: symptoms owing to prolapse and urinary stress incontinence and pelvic surgery. The model includes surgery owing to prolapse and urinary stress incontinence.
      Short-term:
      Excessive bleeding during birth1.406.186.00-6.37
      Wound infection0.602.62.47-2.75
      Pulmonary embolism0.041.721.38-2.14
      Mastitis1.801.531.48-1.59
      Urinary tract infection0.601.411.32-1.52
      Endometritis1.201.121.07-1.19
      Anal sphincter injury
      2.9% for VD, 0% for CS.
      2.9% for VD, 0% for CS.
      N/A
      Long-term:
      Hernia0.303.203.00-3.40
      Adherences0.202.802.60-3.10
      Ileus0.202.252.15-3.00
      Urinary incontinence (surgery)0.500.300.20-0.30
      Prolapse (surgery)0.600.200.10-0.20
      Child
      Notes on child complications: some child complications were only included in sensitivity analyses due to their short duration (hospitalizations for gastroenteritis or respiratory tract infection) or because the project experts thought that further research on the associations with the method of birth was required (rheumatoid arthritis, inflammatory bowel disease, and diabetes). Overweight was not included in the model, as data on its relationship with costs and quality of life in children are inconclusive.
      Short-term:
      Respiratory morbidity1.402.021.49-2.73
      Neonatal intensive care treatment5.601.921.44-2.56
      Long-term:
      Hospitalization for gastroenteritis3.701.211.16-1.25
      Hospitalization for respiratory tract infection5.801.141.09-1.19
      Asthma4.301.191.17-1.21
      Food allergy2.501.161.11-1.21
      Rheumatoid arthritis0.351.171.06-1.28
      Inflammatory bowel disease0.701.161.03-1.30
      Diabetes0.501.111.04-1.17
      CI indicates confidence interval; CS, cesarean section; N/A, not applicable; VD, vaginal delivery.
      Notes on maternal complications: antibiotic treatment was not included as related costs were expected to be negligible. The following maternal complications were not included to avoid double-counting of complications: symptoms owing to prolapse and urinary stress incontinence and pelvic surgery. The model includes surgery owing to prolapse and urinary stress incontinence.
      2.9% for VD, 0% for CS.
      Notes on child complications: some child complications were only included in sensitivity analyses due to their short duration (hospitalizations for gastroenteritis or respiratory tract infection) or because the project experts thought that further research on the associations with the method of birth was required (rheumatoid arthritis, inflammatory bowel disease, and diabetes). Overweight was not included in the model, as data on its relationship with costs and quality of life in children are inconclusive.

      Costs

      The costs of care per delivery method, as well as for short- and long-term complications for the mother were based on data for 2019 from the Cost-per-patient database held by the Swedish Association of Local Authorities and Regions. The database contains information on the healthcare costs per care episode for consecutive care contacts and medical care services in primary and specialist care
      Sveriges Kommuner och Regioner [Swedish Association of Local Authorities and Regions]
      Nationella KPP-principer, version 4. Kostnad per patient. [National Cost Per Patient principles, version 4].
      but does not facilitate tracking of individuals over time. The costs reflect the length of stay, procedures, the time required per occupational category, care provided, and patient-specific costs for pharmaceuticals and materials. When relevant, surgery costs reflect the duration and difficulty of the procedure.
      Cost data was requested for all care episodes with relevant diagnostic or procedural codes during 2019. The codes were identified together with the project’s clinical experts. The codes could refer to main or secondary diagnoses and the care occurrence was handled as a unique episode under the complication with the first match in the list of codes. Costs are reflected in Euros for the year 2019/2020 (conversion rate 1 SEK = 0.088 EUR; other conversions based on the Campbell and Cochrane Economics Methods Group–Evidence for Policy and Practice Information and Coordinating Centre Cost Converter, version 1.6). Table 2 details the number of care episodes and mean costs (including SDs) in 2019 for inpatient care related to delivery and complications.
      Table 2Costs of delivery and inpatient care for maternal and child complications, Swedish registry data (EUR, 2019).
      Type of costICD-10 codeNumber of care episodesMean costSD (costs)
      Method of delivery (singleton births)
      Vaginal deliveryO80, O8191 85532132150
      Acute CSO82.1958274904523
      Planned CSO82.0618753243073
      CS without medical indicationO82.8356252562358
      Mother, short-term complications
      Excessive bleeding during birthO67195772875294
      Wound infectionO86.0, O86.128962876372
      Pulmonary embolismO88.26348394553
      MastitisO91.1, O91.218836876887
      Urinary tract infectionO86.223057525858
      EndometritisO85116951125475
      Anal sphincter injuryO70.2, O70.330755162038
      Mother, long-term complications
      Based on individuals of age ≥ 16 years.
      HerniaK432007752113 253
      Adherences
      Codes include classifications of care procedures.
      K66, N73.6, N99.4; JAP01140988978465
      IleusK56.568913 08918 548
      Urinary incontinence (surgery)N39.315845366291
      Prolapse (surgery)
      Codes include classifications of care procedures.
      N81; LEF00, LEF03, LEF40, LEF50, LEF10, LEF13, LEF96, LDC10, LEF20, LEF23222952783327
      Child, short-term complications
      Respiratory morbidityP22.1, P22.8, P22.9, P24.0372811 45317 751
      Note. Source: Swedish Association of Local Authorities and Regions, Cost-per-patient database.
      CS indicates cesarean section; ICD, International Classification of Diseases.
      Based on individuals of age ≥ 16 years.
      Codes include classifications of care procedures.
      For short-term maternal complications (in particular, mastitis, endometritis, and urinary tract infection), the number of cases is considerably higher in outpatient than inpatient care. Given that the data refer to hospitalizations, this means that these costs are overestimated. Therefore, sensitivity analyses are performed in which the costs for all maternal short-term complications—except embolism and excessive bleeding during delivery (as these generally are treated as part of inpatient care)—are assumed to constitute 10% of the reported mean values. This scenario served to test the impact of lower costs in outpatient care for the named short-term complications.
      The duration of long-term maternal complications was estimated based on the literature and input from the project’s clinical experts. Urinary stress incontinence was assumed to last 5 years (before surgery); hernia, 10 years; adherences and prolapse, 20 years. As the costs were based on surgical procedures, they were only applied once after the respective number of years.
      For child complications, the inpatient costs for treatment of respiratory morbidity were derived from the same database as maternal complications. Neonatal intensive care was calculated based on a daily cost for children aged 32 weeks or more of 2267 EUR
      Karolinska Universitetssjukhuset
      Prislista för utomlänsvård samt för EU/EES och Schweiz 2021. [Price list for care for patients from other regions and EU/ESS and Switzerland 2021].
      and an average length of stay of 2.6 days.
      • Geller E.J.
      • Wu J.M.
      • Jannelli M.L.
      • Nguyen T.V.
      • Visco A.G.
      Neonatal outcomes associated with planned vaginal versus planned primary cesarean delivery.
      Inpatient costs for treatment of respiratory tract infection and gastroenteritis were based on Diagnosis Related Groups

      Socialstyrelsen [National Board of Health and Welfare]. Prospektiva vikter för slutenvårdsgrupper, NordDRG 2020. [Prospective weights for inpatient care groups]. https://www.regionstockholm.se/globalassets/5.-politik/politiska-organ/samverkansnamnden-sthlm-gotland/2021/prislistor-2021/bilaga-2-karolinska-universitetssjukhuset-2021.pdf. Accessed December 18, 2021.

      at 2495 EUR and 2231 EUR, respectively. For the costs of long-term child complications, a focused literature search was performed to identify studies relevant to children and Sweden (Appendix Table 2 in Supplemental Materials found at https://doi.org/10.1016/j.jval.2022.10.003). Except where noted in Appendix Table 3 in Supplemental Materials found at https://doi.org/10.1016/j.jval.2022.10.003, long-term child complications were assumed to last for the whole modeling horizon.

      Utilities

      A focused literature search was performed using terms for long-term complications for mother and child together with a search block for “health state utility values” (based on the precision-maximising filter by Arber et al
      • Arber M.
      • Garcia S.
      • Veale T.
      • Edwards M.
      • Shaw A.
      • Glanville J.M.
      Performance of Ovid Medline search filters to identify health state utility studies.
      ). Searches were conducted in Medline (Ovid) and the Economic Evaluation Database in June 2021. A structured review and selection process was applied, with prioritization based on the relevance of the complication and patient population, and also a direct measurement using EQ-5D in the population of interest (to facilitate comparison across complications). The selected studies and values are detailed in Appendix Table 3 in Supplemental Materials found at https://doi.org/10.1016/j.jval.2022.10.003. In the health economic model, the utility weights were applied during the assumed duration of the complications. For women and children without complications, the mean values for the Swedish general population were used
      • Burström K.
      • Johannesson M.
      • Diderichsen F.
      Health-related quality of life by disease and socio-economic group in the general population in Sweden.
      ,
      • Åstrom M.
      • Persson C.
      • Linden-Boström M.
      • Rolfson O.
      • Burström K.
      Population health status based on the EQ-5D-Y-3L among adolescents in Sweden: results by sociodemographic factors and self-reported comorbidity.
      : 0.87 for women aged 30 to 39 years, 0.84 for women aged 40 to 49, and 0.95 for children and adolescents.

      Sensitivity Analyses and Model Validation

      Sensitivity analyses were performed for all relevant input parameters, including complication risks, costs, utility weights, and duration of long-term complications. When possible, scenario analyses around model assumptions were conducted, such as the inclusion or exclusion of certain types of complications or costs. The details of the analyses and rationale can be found in Table 3. The results of the sensitivity analyses are presented using a time horizon of 10 years as a reference.
      Table 3Description of sensitivity and scenario analyses.
      ParameterSensitivity analyses
      Complications: relative risks95% CI (lower respectively upper limit), for all complications (short and long-term)
      Risk of complications in comparator arm (VD)Variation ± 20% in baseline risk for complications where the relative risk was < 1 or > 2 (ie, lower risk with CS or halved risk with VD, respectively), testing the impact of complication risks that could change the direction of the results
      Inclusion of yearly physician visits for prolapse and urinary incontinence, respectivelyAssumption of 2 physician visits in primary care per year (352 EUR) for the respective complication, testing the role of complications mostly treated in outpatient care
      Costs long-term complications for the childVariation ± 20% in yearly costs for asthma and food allergy
      Costs short-term complications for the mother that are treated in outpatient careAssumption that the mean cost for the complications such as wound infection, mastitis, urinary tract infection, endometritis, and anal sphincter injury is reduced by 90% given that the costs from national statistics focused on hospital care and these complications are often handled in outpatient care
      Duration long-term complicationsAcute: 2 respectively 4 weeks.

      Urinary incontinence: 3 respectively 7 years.

      Hernia: 5 respectively 15 years.

      Adherences, prolapse: 10 respectively 19 years.

      Asthma: 10 years (20 years in base case).

      Food allergy: 7 respectively 13 years.
      Inclusion of additional chronic long-term complications for the childInclusion of relative risks for rheumatoid arthritis, inflammatory bowel disease, and diabetes, together with related costs and utility weights (see Table 1 footnote)
      Inclusion of costs for temporary long-term complications for the childCosts for hospitalization for respiratory tract infection and gastroenteritis during 2 and 3 years, respectively; 1 hospitalization per year (see Table 1 footnote)
      Utility weightsVariation ± 0.1 (utility weight for normal population in the relevant age group being upper limit)
      Exclusion of acute long-term complications for the motherExclusion of relative risks for ileus, testing the impact of removing acute and costly complications of CS
      Discount rateCosts 0%, QALYs 0%

      Costs 5%, QALYs 5%
      CI indicates confidence interval; CS, cesarean section; QALY, quality-adjusted life-year; VD, vaginal delivery.
      Model validation was performed in several ways. The medical experts gave feedback on face validity regarding model structure, data sources, assumptions, and results. A health economist external to the project team checked the model structure and coding. The model results were cross-validated through comparison with results from other modeling studies,
      Caesarean section (NICE clinical guideline 132). Royal College of Obstetricians & Gynaecologists.
      • Fawsitt C.G.
      • Bourke J.
      • Greene R.A.
      • Everard C.M.
      • Murphy A.
      • Lutomski J.E.
      At what price? A cost-effectiveness analysis comparing trial of labour after previous caesarean versus elective repeat caesarean delivery.
      • Fobelets M.
      • Beeckman K.
      • Faron G.
      • Daly D.
      • Begley C.
      • Putman K.
      Vaginal birth after caesarean versus elective repeat caesarean delivery after one previous caesarean section: a cost-effectiveness analysis in four European countries.
      ,
      • Rogers A.J.
      • Rogers N.G.
      • Kilgore M.L.
      • Subramaniam A.
      • Harper L.M.
      Economic evaluations comparing a trial of labor with an elective repeat cesarean delivery: a systematic review.
      while being mindful of differences in methods and assumptions. External validity was aimed for through the use of data from national registries and comparison of model-predicted events to observed events from the systematic review. Finally, all SBU assessments, including the health economic analyses, undergo structured internal, and external review processes.

      Results

      Methods of Delivery and Associated Costs

      Analyses of the Swedish National Birth Register revealed that, among primiparous women, approximately 90% (SD 0.08%) of those who had a planned vaginal delivery eventually gave birth vaginally, whereas the remainder had an acute CS. For multiparous women, the corresponding proportion was 95% (SD 0.05%).
      Based on the Cost-per-patient database, the cost of care for a planned CS and CS without medical indication was around 5300 EUR (SD 3073 and 2358, respectively) in 2019. This compares to about 3200 EUR (SD 2150) for a vaginal delivery and 7500 EUR (SD 4523) for an acute CS.

      Base Case Cost-Effectiveness Results

      When using a 1-year time horizon, the incremental cost for a planned CS compared with a planned vaginal delivery is around 2530 EUR in primiparous women (Table 4) and around 2820 EUR in multiparous women (Appendix Table 4 in Supplemental Materials found at https://doi.org/10.1016/j.jval.2022.10.003). Going beyond 1 year, the utility losses associated with long-term complications, especially for the child, lead to an average total loss of 0.004 quality-adjusted life-years (QALYs) with planned CS over 10 years (Table 4). Over 20 years, there is a utility gain of 0.003 for the woman, which is largely due to the lower risk of prolapse following a CS. Nevertheless, when the child perspective is weighed in, there is an overall loss of 0.002 QALYs over 20 years. Both in primiparous (Table 4) and multiparous women (Appendix Table 4 in Supplemental Materials found at https://doi.org/10.1016/j.jval.2022.10.003), planned vaginal delivery is, thus, associated with cost savings and somatic health gains in the long-term.
      Table 4Cost-effectiveness of planned CS versus planned VD over different time horizons, primiparous women.
      Result typePlanned CSPlanned VDIncremental differenceICER (EUR/QALY)
      Time horizon: 1 year (discount rate 0%)
      Costs
      Mother627541672108
      Child958539419
      Total723347062527Not calculated
      Because no health effects were estimated using utilities for the short-term perspective.
      Time horizon: 10 years (discount rate 3%)
      Costs
      Mother633842142124
      Child25932123471
      Total893263372595
      QALYs
      Mother7.6297.6300.000VD dominant
      That is, planned VD leads to cost savings and health gains.
      Child8.2968.300−0.004VD dominant
      That is, planned VD leads to cost savings and health gains.
      Total15.92515.930−0.004VD dominant
      That is, planned VD leads to cost savings and health gains.
      Time horizon: 20 years (discount rate 3%)
      Costs
      Mother642442632161
      Child36453158487
      Total10 06974212648
      QALYs
      Mother13.14113.1380.003815 477
      Child14.48614.490−0.004VD dominant
      That is, planned VD leads to cost savings and health gains.
      Total27.62727.628−0.002VD dominant
      That is, planned VD leads to cost savings and health gains.
      CS indicates cesarean section; ICER, incremental cost-effectiveness ratio; QALY, quality-adjusted life-year; VD, vaginal delivery.
      Because no health effects were estimated using utilities for the short-term perspective.
      That is, planned VD leads to cost savings and health gains.

      Sensitivity Analyses

      The results were robust to changes in input parameters and assumptions around modeling scenarios (Fig. 2 and Appendix Fig. 1 in Supplemental Materials found at https://doi.org/10.1016/j.jval.2022.10.003 for primiparous and multiparous women, respectively). The incremental costs of planned CS ranged between 2330 and 2950 EUR for primiparous women, and 2600 and 3260 EUR for multiparous women. Except for 1 scenario in which the utility weights for long-term complications were assumed to be 0.1 higher than in the base case, planned CS led to utility losses between 0.001 and 0.015 for primiparous women (between 0.001 and 0.016 for multiparous women). Relative complication risks had the highest impact on incremental costs, whereas utility weights for long-term complications carried the most weight for incremental effects (Appendix Fig. 2 in Supplemental Materials found at https://doi.org/10.1016/j.jval.2022.10.003).
      Figure thumbnail gr2
      Figure 2Cost-effectiveness plane illustrating results of different sensitivity and scenario analyses for planned CS versus planned vaginal delivery (primiparous women, time horizon 10 years).
      CS indicates cesarean section; QALY, quality-adjusted life-year.
      Model predictions for clinical outcomes for primiparous women and their children are detailed in Appendix Table 5 in Supplemental Materials found at https://doi.org/10.1016/j.jval.2022.10.003.

      Discussion

      We evaluated the cost-effectiveness of planned CS on maternal request compared with planned vaginal delivery in a Swedish setting. Our results revealed that planned vaginal birth leads to lower healthcare costs and somatic health gains compared with elective CS without medical indication over up to 20 years. The results are not generalizable to settings with other treatment practices, particularly in countries where the rates of CS are high.
      Given the importance of the local context to the cost-effectiveness results in this area, it is a strength of our analysis that it is based not only on a comprehensive and systematic literature review and meta-analysis of the risk of complications but also Swedish registry data for method of delivery and healthcare costs. The model includes both short- and long-term complications for mother and child and has been analyzed using different time horizons. Utility weights for long-term complications are based on a focused literature search using structured selection criteria.
      Uncertainty exists around the relative risks of complications, utilities, and duration of several long-term complications. We conducted a range of sensitivity analyses around parameter uncertainty and important model assumptions, which indicated stable overall results. For multiparous women, larger cost differences were seen than for primiparous women, owing to the lower share of acute CS and associated complications for planned vaginal delivery in this population. When comparing different time horizons, the cost difference between planned birth methods is relatively stable. This can be explained by the largest cost difference being incurred during the first year, arising from the method of delivery and different risks for neonatal intensive care.
      According to results from the systematic review of effects, the risks of several maternal complications are reduced with planned CS, including urinary stress incontinence and prolapse. Nevertheless, in the health economic model, these positive effects are offset by other maternal complications with a higher risk, and consistently increased complication risks for the child. Given the long duration of and utility losses associated with urinary stress incontinence and prolapse, which weigh in favor of a planned CS, assumptions around these complications play an important role in the model. In our analysis, we only included complications that need to be treated surgically, to avoid double-counting with symptomatic prolapse and urinary incontinence. We did not include any annual costs for maintenance treatment of urinary stress incontinence and prolapse, as nonsurgical treatment is generally limited to pelvic floor exercises, treatment with estrogen or a pessary, and absorbent pads for urinary stress incontinence. Overall results were not affected when we conducted sensitivity analyses assuming an annual cost of 350 EUR for both complications, which corresponds to 2 physician visits in primary care.
      Previous economic evaluations have led to mixed results when urinary stress incontinence was included,
      • Rogers A.J.
      • Rogers N.G.
      • Kilgore M.L.
      • Subramaniam A.
      • Harper L.M.
      Economic evaluations comparing a trial of labor with an elective repeat cesarean delivery: a systematic review.
      in some cases reducing the probability of planned vaginal delivery being cost-effective, and in others not affecting the results to any large extent. The economic evaluation for primiparous women by the National Institute for Health and Care Excellence
      Caesarean section (NICE clinical guideline 132). Royal College of Obstetricians & Gynaecologists.
      included short- and long-term complications for the mother and child and focused on healthcare costs. In the base case, which did not include urinary stress incontinence, vaginal delivery was cost-saving and associated with a QALY gain of 0.03. When urinary stress incontinence was included (using annual maintenance costs of 497 EUR, mainly for absorbing materials and cleaning, based on Xu et al
      • Xu X.
      • Ivy J.S.
      • Patel D.A.
      • et al.
      Pelvic floor consequences of cesarean delivery on maternal request in women with a single birth: a cost-effectiveness analysis.
      ), the results changed to 495 EUR/QALY for planned CS versus planned vaginal delivery, in contrast to the results of our sensitivity analysis described in the previous paragraph.
      Two studies have evaluated the cost-effectiveness of planned CS versus planned vaginal delivery in women with a previous CS in a European setting. Fawsitt et al
      • Fawsitt C.G.
      • Bourke J.
      • Greene R.A.
      • Everard C.M.
      • Murphy A.
      • Lutomski J.E.
      At what price? A cost-effectiveness analysis comparing trial of labour after previous caesarean versus elective repeat caesarean delivery.
      performed an analysis from a healthcare perspective in low-risk women with previous CS in Ireland up to 6 weeks post-delivery and found planned vaginal delivery to be dominant, which is in line with the results of our 1-year analysis. Fobelets et al
      • Fobelets M.
      • Beeckman K.
      • Faron G.
      • Daly D.
      • Begley C.
      • Putman K.
      Vaginal birth after caesarean versus elective repeat caesarean delivery after one previous caesarean section: a cost-effectiveness analysis in four European countries.
      conducted a cost-effectiveness analysis using a societal perspective for low-risk women with previous CS in Belgium, Ireland, Italy, and Germany, with the latter 2 countries having high rates of CS, and thus, not being comparable to Sweden. Over a time horizon of 6 weeks, planned vaginal delivery was again dominant across all countries. When using a lifetime perspective, the overall result remained unchanged in Ireland (similar to our results), whereas in Belgium, the cost for a planned vaginal delivery was 3931 EUR per QALY compared with planned CS.
      There are certain limitations that need to be kept in mind when using the study results. In our analysis of the data from the Swedish Medical Birth Registry, some of the deliveries that were planned as CS will have been misclassified as planned vaginal deliveries if the delivery occurred before the date of the planned CS. This is related to a general challenge in this field: studies tend not to have an intention-to-treat design, as it is usually the actual rather than the planned method of delivery that is registered in charts and databases. Consequently, the costs of planned CS tend to be overestimated and the costs of vaginal deliveries underestimated, as the costs of preparedness (eg, surgery, neonatal care) are allocated to patients who actually use the resources.
      Swedish Society for Obstetrics and Gynecology (SFOG)
      Kejsarsnitt. Rapport nr 65. [Caesarean section. Report no 65].
      In a sensitivity analysis (not presented) we tested the extreme assumption that the costs of planned CS and vaginal deliveries are the same. Although the difference in total costs became much smaller, the overall results remained unchanged.
      In our model, we have not been able to consider all possible subpopulations and complications reflecting, for example, the number of previous deliveries, previous type of delivery, and earlier complications. Such a model quickly becomes complex, and we developed our model based on an assessment of accessible data, a need for methodological transparency, and clear communication to stakeholders in different parts of the healthcare system.
      An important limitation of our study lies in the exclusion of possible complications related to mental health associated with different methods of delivery. To the best of our knowledge, this aspect has not been included in any previous health economic evaluations either. The possible implications of this limitation are difficult to assess. A systematic review by Olieman et al
      • Olieman R.M.
      • Siemonsma F.
      • Bartens M.A.
      • Garthus-Niegel S.
      • Scheele F.
      • Honig A.
      The effect of an elective cesarean section on maternal request on peripartum anxiety and depression in women with childbirth fear: a systematic review.
      on the effect of planned CS on maternal request on peripartum anxiety and depression among women with childbirth fear only found 3 studies of at least satisfactory study quality. Women who received a planned CS had higher antepartum depression scores compared with women who delivered vaginally.
      • Garthus-Niegel S.
      • von Soest T.
      • Knoph C.
      • Simonsen T.B.
      • Torgersen L.
      • Eberhard-Gran M.
      The influence of women’s preferences and actual mode of delivery on posttraumatic stress symptoms following childbirth: a population-based, longitudinal study.
      Women who preferred a planned CS, but delivered vaginally, had significantly higher levels of childbirth fear, higher levels of antepartum anxiety, and higher scores of posttraumatic stress syndrome 2 months after delivery compared with women who did not have such a preference and who gave birth vaginally.
      • Garthus-Niegel S.
      • von Soest T.
      • Knoph C.
      • Simonsen T.B.
      • Torgersen L.
      • Eberhard-Gran M.
      The influence of women’s preferences and actual mode of delivery on posttraumatic stress symptoms following childbirth: a population-based, longitudinal study.
      Moreover, post hoc analyses revealed that women whose request for a CS was not granted had higher levels of depression postpartum compared with women who did not request a CS and gave birth vaginally. In contrast, women whose request for a planned CS was granted reported normal levels of depression after delivery.
      • Olieman R.M.
      • Siemonsma F.
      • Bartens M.A.
      • Garthus-Niegel S.
      • Scheele F.
      • Honig A.
      The effect of an elective cesarean section on maternal request on peripartum anxiety and depression in women with childbirth fear: a systematic review.
      This suggests that there may be psychological aspects that influence the overall effect of planned CS on maternal request compared with planned vaginal delivery, which could increase the utility gain of a planned CS. Depression is costly in terms of both resource use and reduced quality of life. In a Swedish study, the annual cost per patient with moderate depression treated in primary care was estimated at 4645 EUR.
      • Sobocki P.
      • Ekman M.
      • Agren H.
      • et al.
      Resource use and costs associated with patients treated for depression in primary care.
      Moreover, depression has been associated with a utility weight of 0.44, compared with an average utility weight of 0.87 for women aged 30 to 39 years in the general Swedish population.
      • Burström K.
      • Johannesson M.
      • Diderichsen F.
      Health-related quality of life by disease and socio-economic group in the general population in Sweden.
      An argument that is sometimes put forward in favor of CS concerns complications that can occur if the child gets stuck in the birth canal (eg, clavicle fracture or nerve damage affecting the arms), and that these cannot occur with a planned CS. According to data from the Swedish Medical Birth Registry, the risk of a child getting stuck with its shoulders is 1/350 (0.3%) with planned vaginal delivery for pregnancies that have lasted 39 weeks or more. In a Swedish context, there is an important knowledge gap regarding the consequences of a child getting stuck and more research is needed.
      Our health economic evaluation includes only direct medical costs as the systematic review of effects focused on the medical consequences of somatic complications. The registry data we used for costs of complications primarily contained information on hospital care episodes, which means that costs in outpatient care are not included. Nevertheless, sensitivity analyses in which outpatient care costs are included for complications with a reduced risk in the case of planned CS indicate that this type of cost likely does not have any major impact on results. Finally, the impact of different methods of delivery on productivity losses is also an area requiring further research.

      Conclusions

      Our economic evaluation indicates that, in a Swedish context, planned vaginal birth compared with CS on maternal request leads to lower hospital care costs and somatic health gains over a time horizon of up to 20 years. This includes short- and long-term consequences for both mother and child. Although there is uncertainty around the relative risks of complications, the duration of several long-term complications, and their impact on quality of life, the overall results remained stable across a range of sensitivity and scenario analyses.

      Article and Author Information

      Author Contributions: Concept and design: Berg, Källén, Andolf, Hellström-Westas, Ekéus, Alvan, Vitols
      Acquisition of data: Berg, Källén, Andolf, Hellström-Westas, Ekéus, Alvan, Vitols
      Analysis and interpretation of data: Berg, Källén, Andolf, Alvan, Vitols
      Drafting of the manuscript: Berg
      Critical revision of the paper for important intellectual content: Källén, Andolf, Hellström-Westas, Ekéus, Alvan, Vitols
      Statistical analysis: Berg, Källén, Alvan
      Conflict of Interest Disclosures: The authors reported no conflicts of interest.
      Funding/Support: SBU is a government agency under the Ministry of Health and Social Affairs in Sweden and the study was performed as part of a governmental research assignment in the area of women’s health.
      Role of Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

      Acknowledgment

      The authors thank Hanna Olofsson, SBU, who performed the literature research on utilities.

      Supplemental Material

      References

        • Boerma T.
        • Ronsmans C.
        • Melesse D.Y.
        • et al.
        Global epidemiology of use of and disparities in caesarean sections.
        Lancet. 2018; 392: 1341-1348
        • Wiklund I.
        • Malata A.M.
        • Cheung N.F.
        • Cadee F.
        Appropriate use of caesarean section globally requires a different approach.
        Lancet. 2018; 392: 1288-1289
        • Socialstyrelsen [National Board of Health and Welfare]
        Kejsarsnitt i Sverige 2008-2017. Kriterier som styr beslut om förlossningssätt, samt kartläggning av komplikationer. [Caesarean section in Sweden 2008-2017. Criteria that govern decisions about method of delivery, and mapping of complications].
        • Jenabi E.
        • Khazaei S.
        • Bashirian S.
        • Aghababaei S.
        • Matinnia N.
        Reasons for elective cesarean section on maternal request: a systematic review.
        J Matern Fetal Neonatal Med. 2020; 33: 3867-3872
        • Entringer A.P.
        • Pinto M.
        • Gomes M.
        Cost-effectiveness analysis of natural birth and elective C-section in supplemental health.
        Rev Saúde Publ. 2018; 52: 91
        • Gilbert S.A.
        • Grobman W.A.
        • Landon M.B.
        • et al.
        Cost-effectiveness of trial of labor after previous cesarean in a minimally biased cohort.
        Am J Perinatol. 2013; 30: 11-20
        • Gilbert S.A.
        • Grobman W.A.
        • Landon M.B.
        • et al.
        Lifetime cost-effectiveness of trial of labor after cesarean in the United States.
        Value Health. 2013; 16: 953-964
        • Liu T.C.
        • Loh C.P.
        • Lin C.L.
        • Chen C.S.
        Do cesarean deliveries on maternal request lead to greater maternal care utilization during six-month, one-year, and two-year postpartum periods? Evidence from NHI claims data in Taiwan.
        Health Policy. 2014; 117: 39-47
        • Betran A.P.
        • Ye J.
        • Moller A.B.
        • Zhang J.
        • Gulmezoglu A.M.
        • Torloni M.R.
        The increasing trend in caesarean section rates: global, regional and national estimates: 1990-2014.
        PLoS One. 2016; 11e0148343
      1. Caesarean section (NICE clinical guideline 132). Royal College of Obstetricians & Gynaecologists.
        • Fawsitt C.G.
        • Bourke J.
        • Greene R.A.
        • Everard C.M.
        • Murphy A.
        • Lutomski J.E.
        At what price? A cost-effectiveness analysis comparing trial of labour after previous caesarean versus elective repeat caesarean delivery.
        PLoS One. 2013; 8e58577
        • Fobelets M.
        • Beeckman K.
        • Faron G.
        • Daly D.
        • Begley C.
        • Putman K.
        Vaginal birth after caesarean versus elective repeat caesarean delivery after one previous caesarean section: a cost-effectiveness analysis in four European countries.
        BMC Pregnancy Childbirth. 2018; 18: 92
        • SBU
        Caesarean section on maternal request – A systematic review and assessment of medical, health economic, ethical, and social aspects. NCBI.
        https://www.ncbi.nlm.nih.gov/books/NBK580955/
        Date: 2021
        Date accessed: June 1, 2022
        • Husereau D.
        • Drummond M.
        • Augustovski F.
        • et al.
        Consolidated health economic evaluation reporting standards 2022 (CHEERS 2022) statement: updated reporting guidance for health economic evaluations.
        Value Health. 2022; 25: 3-9
        • TLV
        Tandvårds- och läkemedelsförmånsverkets allmänna råd. TLVAR 2017:1. [The Dental and Pharmaceutical Benefits Agency’s general guidance].
        • Socialstyrelsen [National Board of Health and Welfare]
        Medicinska födelseregistret. [National Medical Birth Register].
      2. The GRADE working group. GRADE.
        https://www.gradeworkinggroup.org/
        Date accessed: March 1, 2022
        • Sveriges Kommuner och Regioner [Swedish Association of Local Authorities and Regions]
        Nationella KPP-principer, version 4. Kostnad per patient. [National Cost Per Patient principles, version 4].
        • Karolinska Universitetssjukhuset
        Prislista för utomlänsvård samt för EU/EES och Schweiz 2021. [Price list for care for patients from other regions and EU/ESS and Switzerland 2021].
        • Geller E.J.
        • Wu J.M.
        • Jannelli M.L.
        • Nguyen T.V.
        • Visco A.G.
        Neonatal outcomes associated with planned vaginal versus planned primary cesarean delivery.
        J Perinatol. 2010; 30: 258-264
      3. Socialstyrelsen [National Board of Health and Welfare]. Prospektiva vikter för slutenvårdsgrupper, NordDRG 2020. [Prospective weights for inpatient care groups]. https://www.regionstockholm.se/globalassets/5.-politik/politiska-organ/samverkansnamnden-sthlm-gotland/2021/prislistor-2021/bilaga-2-karolinska-universitetssjukhuset-2021.pdf. Accessed December 18, 2021.

        • Arber M.
        • Garcia S.
        • Veale T.
        • Edwards M.
        • Shaw A.
        • Glanville J.M.
        Performance of Ovid Medline search filters to identify health state utility studies.
        Int J Technol Assess Health Care. 2017; 33: 472-480
        • Burström K.
        • Johannesson M.
        • Diderichsen F.
        Health-related quality of life by disease and socio-economic group in the general population in Sweden.
        Health Policy. 2001; 55: 51-69
        • Åstrom M.
        • Persson C.
        • Linden-Boström M.
        • Rolfson O.
        • Burström K.
        Population health status based on the EQ-5D-Y-3L among adolescents in Sweden: results by sociodemographic factors and self-reported comorbidity.
        Qual Life Res. 2018; 27: 2859-2871
        • Rogers A.J.
        • Rogers N.G.
        • Kilgore M.L.
        • Subramaniam A.
        • Harper L.M.
        Economic evaluations comparing a trial of labor with an elective repeat cesarean delivery: a systematic review.
        Value Health. 2017; 20: 163-173
        • Xu X.
        • Ivy J.S.
        • Patel D.A.
        • et al.
        Pelvic floor consequences of cesarean delivery on maternal request in women with a single birth: a cost-effectiveness analysis.
        J Womens Health (Larchmt). 2010; 19: 147-160
        • Swedish Society for Obstetrics and Gynecology (SFOG)
        Kejsarsnitt. Rapport nr 65. [Caesarean section. Report no 65].
        • Olieman R.M.
        • Siemonsma F.
        • Bartens M.A.
        • Garthus-Niegel S.
        • Scheele F.
        • Honig A.
        The effect of an elective cesarean section on maternal request on peripartum anxiety and depression in women with childbirth fear: a systematic review.
        BMC Pregnancy Childbirth. 2017; 17: 195
        • Garthus-Niegel S.
        • von Soest T.
        • Knoph C.
        • Simonsen T.B.
        • Torgersen L.
        • Eberhard-Gran M.
        The influence of women’s preferences and actual mode of delivery on posttraumatic stress symptoms following childbirth: a population-based, longitudinal study.
        BMC Pregnancy Childbirth. 2014; 14: 191
        • Sobocki P.
        • Ekman M.
        • Agren H.
        • et al.
        Resource use and costs associated with patients treated for depression in primary care.
        Eur J Health Econ. 2007; 8: 67-76