Cost of hospital admission after autologous breast reconstruction and factors associated with increased cost: a population-based cohort study Authors: Oluwatobi R Olaiya, Beraki Abraha, Xi Ming Zhu, Lucas Gallo, Sophocles H Voineskos, Christopher J Coroneos, Zak Mireku, Lawrence Mbuagbaw, Mark McRae Published: 2026 ABSTRACT AbstractBackground:Demand for reconstructive surgery postmastectomy has risen, prompting evaluation of its economic impact in a publicly funded health care system. We sought to explore factors associated with increased costs of hospital admission following autologous breast reconstruction.Methods:We conducted a retrospective cohort study involving patients who underwent autologous breast reconstruction in Ontario, Canada (2005 to 2020). Patients received either a pedicled transverse rectus abdominis myocutaneous (TRAM) flap or free tissue transfer. We analyzed costs from the hospital perspective, from admission to discharge. We used generalized linear models with a γ distribution to identify cost-associated factors.Results:Of 2634 patients, 378 (14.4%) underwent pedicled TRAM and 2256 (85.7%) underwent free flap reconstruction. The median cost of hospital admission was $8816 (interquartile range $3383 to $90 974). Univariate analysis identified cost drivers including length of stay, reconstruction type, age, income quintile, laterality, and emergency reoperation. Multivariable analysis showed increased costs associated with free flap reconstruction (13% increase, 95% confidence interval [CI] 5% to 22%; p < 0.001), age older than 70 years (31% increase, 95% CI 7% to 61%; p < 0.01), lower income quintile (10% increase, 95% CI 3% to 18%; p = 0.01), and bilateral reconstruction (11% increase, 95% CI 1% to 23%; p = 0.03). Hospital type, geographic region, timing of reconstruction, comorbidities, and rurality were not significantly associated with cost.Conclusion:Free flap reconstruction, bilateral procedures, and advanced age are key drivers of hospital admission costs. These findings may inform policy and resource-allocation strategies in publicly funded systems to optimize value in breast reconstruction care. Breast cancer is the most diagnosed cancer in females, with an estimated 29 700 new cases in 2023 in Canada.1,2 It is also the second leading cause of cancer-related death in females, following lung cancer.3 About 1 in 8 women are expected to be diagnosed with breast cancer in their lifetime.4 Breast reconstruction is a surgical intervention undertaken after a mastectomy to reshape and restore the breasts’ form.5–7 The primary techniques for breast reconstruction include tissue expanders with implant reconstruction, pedicled latissimus dorsi flaps (with or without an implant), pedicled transverse rectus abdominis myocutaneous (TRAM) flaps, and deep inferior epigastric perforator (DIEP) flaps.8 Among these, DIEP free flaps have grown in popularity because of their reliability, natural appearance, and long-term patient satisfaction. 9,10 Each reconstruction method presents distinct advantages and challenges, including variability in complication rates, duration of hospital stays, and technical demands of the procedures.7,10 Whereas the clinical outcomes of breast reconstruction are well documented, data are limited on the economic impact of hospital admissions following these procedures, particularly in the context of a publicly funded health care system. The economic burden of breast cancer treatment in Canada varies significantly by disease stage, with costs ranging from $29 938 for patients with stage I disease to $66 627 for those with stage IV or metastatic disease.11 These costs in a single-payer public health care system are largely driven by hospital admissions for both initial treatments and terminal care.12 As advances in breast cancer treatment and screening continue, the underlying assumption is that demand for reconstructive surgery will increase, placing additional strain on the health care system. However, comprehensive data on hospital admission costs following reconstructive procedures remain scarce. We aimed to fill this gap by investigating the factors that contribute to increased hospital admission costs after autologous breast reconstruction, focusing on patient- and procedure-related variables. Unlike prior research that has relied on data from the United States, where health care costs are often inflated,13 this study examines costs within Ontario’s single-payer, publicly funded health care system. Understanding these cost drivers is critical for supporting more efficient resource allocation and cost management in breast reconstruction surgery. Methods We conducted a retrospective population-based cost analysis from 2005 to 2020 using health care administrative data sets from Ontario, Canada. We identified all adult patients who underwent mastectomy and then received breast reconstruction within the subsequent 10 years. Patients received either a pedicled TRAM flap or free tissue transfer. Ontario’s health care is provided under a single-payer system through the publicly funded Ontario Health Insurance Plan (OHIP), and care is paid through the provincial government based on a patient’s health care number. ICES is an independent, nonprofit organization that links several provincial health care administrative databases through secure, encrypted OHIP numbers of all Ontario residents. The database includes information from the Ontario Cancer Registry,14 the Canadian Institute for Health Information (CIHI) Discharge Abstract Database,15 the Registered Persons Database,16 and the OHIP database.16 Through these databases, we collected information on patient demographic characteristics, procedures performed, and hospital admission costs. We included only data from the admission associated with the index reconstruction. We excluded patients who had admissions with OHIP codes indicating both pedicled and free flap reconstructions. We undertook the cost analysis from the hospital perspective, from admission to discharge, after a patient’s index autologous breast reconstruction. We derived inpatient acute care costs using person-level costing through the Ontario Case Costing Initiative (OCCI), a program within the Health Data Branch of the Ministry of Health.17–19 The specifics of the costing methodology employed by OCCI are available in other sources, but briefly, it classifies patient encounters into case-mix groups based on diagnoses or procedures performed. Hospital admission costs were estimated using the resource intensity weight (RIW) methodology, which provides an average cost per hospital admission for each assigned case-mix group.19 We did not perform inflation adjustment, as our focus was on identifying cost drivers rather than assessing cost trends over time. Furthermore, by incorporating the year of reconstruction as a variable in the analysis, we accounted for the temporal variations in cost. Patient factors included age, income quintile, Charlson Comorbidity Index (CCI) score, residence in a rural or urban setting, diabetes, laterality (unilateral v. bilateral), type of reconstruction (free v. pedicled), Local Health Integration Network (LHIN), and preoperative radiation status. Surgical factors included year of surgery, delayed or immediate reconstruction, emergency reoperation, and type of autologous reconstruction (pedicled v. free flap). We identified emergency reoperations using OHIP emergency operation billing codes within 1 week of the reconstruction. We used previously validated algorithms to identify patients with diabetes at the time of their index procedure.20,21 As no variables had missing data exceeding 10%, we performed a complete case analysis, with missing values handled with casewise deletion. Statistical analysis We summarized patient demographic data and outcomes using descriptive statistics. Given that cost data are typically positively skewed, we used a generalized linear model (GLM) with a γ distribution to model the positively skewed cost data. We employed this model to analyze the association between surgical and patient factors and the cost of index breast reconstruction. Exponentiated coefficients represent the percent change in cost for each 1-unit increase in the predictor. Values greater than 1 indicate an increase; values less than 1 indicate a decrease. We reported this study in accordance with Reporting of Studies Conducted Using Observational Routinely-collected Health Data statement, which extends the Strengthening the Reporting of Observational Studies in Epidemiology checklist for observational studies to administrative health care data. One author (O.R.O) had direct access to the data. We considered a p value less than 0.05 to be statistically significant. We performed the statistical analysis with STATA 16 (StataCorp LLC. Stata Statistical Software: Release 16). Ethics approval Ethics approval was provided for this study by the Hamilton Integrated Research Ethics Board. Results Between 2005 and 2020, 2634 patients underwent autologous breast reconstruction in Ontario. In this cohort, 378 (14.4%) underwent pedicled TRAM flap reconstruction, and 2256 (85.6%) underwent free flap reconstruction (Table 1). The number of pedicled TRAM flap reconstructions decreased exponentially over time (Figure 1). Most of the patients included in this study were younger than 60 years (n = 2251; 85.5%). In this sample, most women underwent immediate autologous breast reconstruction (n = 1777; 67.5%) and unilateral reconstruction (n = 2343; 89.0%). Most patients resided in an urban setting (n = 2353; 89.3%), and surgery was performed primarily at teaching hospitals (n = 2086; 79.2%). Table 1Patient characteristicsCharacteristicNo. (%) of patientsn = 2634Age, yr < 501312 (49.8) 50–59939 (35.6) 60–69338 (12.8) > 7045 (1.7)Type of flap Pedicled TRAM378 (14.4) Free TRAM or DIEP2256 (85.6)Timing of autologous breast reconstruction Delayed857 (32.5) Immediate1777 (67.5)Laterality of autologous breast reconstruction Unilateral2343 (89.0) Bilateral291 (11.0)Preoperative radiation Yes714 (27.1) No1920 (72.9)Diabetes Yes47 (1.8) No2587 (98.2)Charlson Comorbidity Index score 0711 (27.0) 1–21283 (48.7) 3–457 (2.2) 5–6555 (21.1) > 728 (1.1)Income quintile* 1 (lowest)316 (12.0) 2430 (16.3) 3486 (18.5) 4615 (23.3) 5 (highest)779 (29.6)Place of residence Urban2353 (89.3) Rural276 (10.5)Type of hospital Teaching2086 (79.2) Non-teaching548 (20.8)Local Health Integration Network Central888 (33.7) East503 (19.1) North93 (3.5) Toronto352 (13.4) West798 (30.3)DIEP = deep inferior epigastric perforator; TRAM = transverse rectus abdominis muscle.*Population divided into 5 groups based on income levels: quintile 1 = bottom 20%, quintile 5 = top 20%. Patient characteristics DIEP = deep inferior epigastric perforator; TRAM = transverse rectus abdominis muscle. Population divided into 5 groups based on income levels: quintile 1 = bottom 20%, quintile 5 = top 20%. The different types of breast reconstructions performed by percentage (pedicled transverse rectus abdominis muscle [TRAM] flap v. free flap) between 2005 and 2020 in Ontario. *The 2020 data cover only the first half of the year, and no pedicled TRAM procedures were performed to that point. The median cost of hospital admission was $8816 (interquartile range $3383 to $90 974). The median length of hospital stay for patients was 4 days in the group receiving pedicled TRAM flap reconstruction and 4 days in the group receiving free flap reconstruction. Univariate GLM analysis revealed there were several preoperative factors that were associated with higher cost of hospital admission (Table 2). Key drivers included length of admission; each additional day was associated with a 9% increase in the cost of hospital admission (95% confidence interval [CI] 7% to 10%; p < 0.001). Undergoing free flap reconstruction was associated with a 9% increase in costs (95% CI 0% to 17%; p = 0.04) relative to a pedicled TRAM. Patients older than 70 years had a 26% increase in costs compared with patients aged 18 to 40 years (95% CI 3% to 52%; p = 0.02). Patients in lower income quintiles (quintiles 1 and 2) had costs that were 8% higher (quintile 1, 95% CI 0% to 16%; p = 0.04) and 11% higher (quintile 2, 95% CI 3% to 19%; p = 0.01) than those in the highest quintile. Bilateral reconstruction was associated with an 11% increase in the cost of hospital admission compared with unilateral reconstructions (95% CI 1% to 23%; p = 0.03). Emergency reoperation within 7 days resulted in a 100% cost increase, effectively doubling the expenses of hospital admission (95% CI 84% to 118%; p < 0.001). Table 2Association between variables and cost of hospital admission after autologous breast reconstruction based on a univariate regression analysisVariableExponentiated coefficient95% CIp valueLength of admission1.091.07–1.10< 0.001Free flap reconstruction1.091.0–1.170.04Year of operation1.000.99–1.01> 0.9Age, yr < 40 (reference category)––– 45–490.990.94–1.050.8 50–541.771.00–1.140.06 55–591.050.98–1.120.2 60–640.990.93–1.060.8 65–691.020.92–1.140.7 > 701.261.03–1.520.02Charlson Comorbidity Index score 0 (reference category)––– 1–20.980.93–1.030.4 3–41.080.90–1.270.4 5–60.990.94–1.050.8 > 71.160.90–1.480.2Type of hospital Teaching (reference category)––– Non-teaching0.990.93–1.050.6Income quintile* 11.081.00–1.160.04 21.111.03–1.190.01 31.030.97–1.100.3 41.040.99–1.090.08 5 (reference category)–––Local Health Integration Network Central0.990.94–1.050.7 East1.030.96–1.090.4 North1.090.92–1.300.3 West1.060.99–1.120.09Laterality of autologous breast reconstruction Unilateral (reference category)––– Bilateral1.111.01–1.230.03Timing of autologous breast reconstruction Delayed (reference category)––– Immediate1.030.99–1.070.2Preoperative radiation No (reference category)––– Yes0.990.95–1.030.6Diabetes No (reference category)––– Yes1.130.93–1.360.2Place of residence Urban (reference category)––– Rural1.020.96–1.090.5Emergency reoperation No (reference category)––– Yes2.001.84–2.18< 0.001CI = confidence interval.*Population divided into 5 groups based on income levels; quintile 1 = bottom 20%, quintile 5 = top 20%. Association between variables and cost of hospital admission after autologous breast reconstruction based on a univariate regression analysis CI = confidence interval. Population divided into 5 groups based on income levels; quintile 1 = bottom 20%, quintile 5 = top 20%. Our multivariable GLM analysis revealed factors that were independently associated with higher hospital admission costs (Table 3). Undergoing free flap reconstruction resulted in a 13% increase in cost relative to a pedicled TRAM (95% CI 5% to 22%; p < 0.001). Patients older than 70 years had an average 31% increase in costs compared with younger patients (95% CI 7% to 61%; p < 0.01). Income quintile was a significant predictor of costs, with a 10% increase in costs for patients in income quintile 2 compared with those in income quintile 5 (95% CI 3% to 18%; p = 0.01) and a 7% increase when comparing quintile 1 with quintile 5 (95% CI 0% to 15%; p = 0.05). Additionally, bilateral reconstruction was significantly associated with higher costs. Table 3Association between variables (excluding emergency reoperation) and cost of hospital admission after autologous breast reconstruction based on a multivariable logistic regression analysisVariableExponentiated coefficient95% CIp valueFree flap reconstruction1.131.05–1.22< 0.001Year of operation1.000.99–1.010.7Age, yr 18–44 (reference category)– 45–490.990.95–1.040.8 50–541.061.00–1.130.06 55–591.040.98–1.110.2 60–640.980.92–1.040.6 65–691.030.92–1.150.6 > 701.311.07–1.61< 0.01Charlson Comorbidity Index score 0 (reference category)–– 1–20.990.94–1.040.6 3–41.050.90–1.210.5 5–61.010.95–1.080.7 > 71.140.88–1.470.3Type of hospital Teaching (reference category)–– Non-teaching1.010.96–1.070.7Income quintile* 11.071.00–1.150.05 21.101.03–1.180.01 31.030.98–1.090.3 41.051.00–1.100.06 5 (reference category)––Local Health Integration Network Toronto (reference category)–– Central0.980.93–1.040.6 East1.000.94–1.070.9 North1.080.93–1.250.3 West1.040.97–1.100.3Laterality of autologous breast reconstruction Unilateral (reference category)–– Bilateral1.111.02–1.220.02Timing of autologous breast reconstruction Delayed (reference category)–– Immediate1.030.99–1.080.1Preoperative radiation No (reference category)–– Yes1.000.95–1.05> 0.9Diabetes No (reference category)–– Yes1.070.89–1.280.5Place of residence Urban (reference category)–– Rural0.990.94–1.060.8CI = confidence interval.*Population divided into 5 groups based on income levels; quintile 1 = bottom 20%, quintile 5 = top 20%. Association between variables (excluding emergency reoperation) and cost of hospital admission after autologous breast reconstruction based on a multivariable logistic regression analysis CI = confidence interval. Population divided into 5 groups based on income levels; quintile 1 = bottom 20%, quintile 5 = top 20%. We included additional factors in our univariate and multivariate analyses, such as hospital type (community v. teaching), place of residence (urban v. rural), timing of reconstruction (delayed v. immediate), preoperative radiation, CCI score, and LHIN. However, none of these factors were significantly associated with increased hospital admission costs. Discussion This study provides a comprehensive examination of the factors influencing hospital admission costs for autologous breast reconstruction in Ontario, Canada, between 2005 and 2020. We found that reconstruction type, age, income quintile, and reoperation were associated with increased costs, whereas CCI score, diabetes, and hospital type (academic v. community) were not linked to increased costs. These findings underscore the complex interplay of patient demographic characteristics, surgical choices, and socioeconomic variables in shaping autologous breast reconstruction costs. In contemporary practice, free muscle-sparing TRAM and DIEP flaps are generally preferred over pedicled TRAM flaps in centres with microsurgical expertise, given lower abdominal wall morbidity and comparable or superior reconstructive outcomes.22,23 A broader shift away from muscle-sacrificing abdominal flaps toward perforator-based techniques has further reduced the use of pedicled TRAM flaps.24,25 Consequently, flap choice is often driven more by surgeon experience, intraoperative findings, and institutional resources than by patient comorbidity alone, with pedicled TRAM flaps typically reserved for situations such as absent suitable donor vessels or limited microsurgical capacity.26,27 Because our study was retrospective, based on administrative data, and not designed to compare free flap and pedicled TRAM reconstructions, we could not capture all the risk factors that could act as confounding variables, such as surgeon experience or institutional resources. Our study illustrates that free flap reconstruction has emerged as the dominant method, accounting for 85.7% (n = 2256) of the 2634 procedures performed. This is consistent with a broader trend favouring techniques that preserve muscle integrity, as they are associated with improved long-term outcomes and reduced donor-site morbidity.9,10,28 The muscle-sparing approach of free flaps also addresses aesthetic concerns by preserving abdominal contour, which has been shown to positively impact patients’ quality of life after surgery.29,30 Whereas the median length of stay was similar between groups, the higher cost associated with free flap reconstruction likely reflects greater procedural complexity. Free flaps require longer operative time, specialized microsurgical teams and equipment, and more intensive perioperative monitoring. These factors increase overall resource use and are captured in the RIW-based cost estimates, independent of length of stay. Although free flaps incur higher initial hospital admission costs compared with pedicled TRAM flaps, the reduced need for subsequent health care interventions likely offset these initial costs over time.31 Age emerged as a notable factor, with costs of hospital admissions increasing by 26% for patients older than 70 years. This aligns with existing literature suggesting that older patients require more intensive perioperative care owing to age-related comorbidities, reduced physiologic reserves, and a slower healing process.32,33 However, patients aged 60 to 64 years and 65 to 69 years did not show increased costs relative to those younger than 40 years. The oldest patients incur additional costs owing to unique challenges, but they represent a minority of the population. Overall, adult females, apart from those older than 70 years, tend to do well postoperatively and do not place a disproportionate burden on the health care system. Our findings suggest that age alone should not be a barrier to autologous reconstruction, as patients aged 60 to 69 years did not exhibit significantly increased costs of hospital admission. As expected, hospital length of stay was the primary cost driver in this study, with each additional day in the hospital being associated with a 9% average increase in expenses. This finding reinforces the importance of reducing hospital stays through effective perioperative management and discharge planning. Enhanced Recovery After Surgery protocols have been shown to reduce length of stay without compromising patient safety, by optimizing pain management, early mobilization, and nutritional support.34,35 Adopting such protocols in reconstructive surgery could provide a cost-effective solution to reduce resource use and enhance patient recovery. Prolonged hospital stays are often indicative of complications or challenges in post-surgical recovery, which emphasizes the need for proactive strategies to mitigate these risks. For example, identifying high-risk patients preoperatively and providing additional support may prevent extended hospital stays, which contribute substantially to health care costs. The association between socioeconomic status and hospital admission costs was an inconsistent finding in this study. Socioeconomic status has been shown to affect postoperative outcomes, as lower-income patients are more likely to experience complications that prolong hospital stays.36 Whereas Canada’s publicly funded health care systems aim to provide equitable access, our findings underscore that income-related disparities still play a role in patient outcomes and resource utilization.37,38 Addressing these disparities through enhanced social support, multidisciplinary care pathways, and preoperative optimization programs targeted at lower-income patients could reduce costs of hospital admission and improve health outcomes for vulnerable populations.36,39,40 Charlson Comorbidity Index scores were not associated with increased costs of hospital admissions. In this study, most of the patients had relatively low comorbidity scores. This may reflect that reconstruction after mastectomy is not required and those with high comorbidity scores may be guided toward implant-based reconstruction, may choose to forgo reconstruction, or may be counselled to avoid immediate autogenous breast reconstruction. The elective nature of this study population and the fact that patients typically undergo preoperative assessments to ensure they are well optimized for surgery may mitigate the impact of comorbidities on costs.2 In addition, the CCI consists of categories that do not include smoking or obesity, both of which are known risk factors for flap failure.41,42 As a result, this measure may not fully capture all contributing factors affecting outcomes of free flap procedures and potential drivers of increasing hospital admission costs. Bilateral breast reconstructions were associated with an average increase in costs. This illustrates the resource demands of more extensive surgeries. Bilateral procedures require more operative time, increased perioperative care, and longer recovery, which contribute to higher hospital admission costs.43,44 However, the increased complexity of these procedures underscores the need for meticulous surgical planning and effective postoperative management to control costs. Emergency reoperation was associated with a 100% increase in hospital admission costs, marking it as a critical cost driver. This finding aligns with previous research demonstrating that surgical complications requiring reoperation contribute significantly to the financial burden on health care systems.45–47 Preventing complications through careful surgical planning and robust postoperative monitoring can play a vital role in reducing these costs. Not only do reoperations extend hospital stays, but they also increase the risk of additional iatrogenic adverse events, further escalating resource use. Limitations Our study was designed as a population-based retrospective cohort, and we encountered common challenges associated with administrative database research.9,48 One limitation was the lack of important information related to hospital admission costs, such as body mass index, breast reconstruction operative volume, and surgeon experience. As is done with many studies using CIHI data, we estimated hospital admissions costs using provincial mean unit costs, and we determined the amount of resources consumed using the patient’s RIW. Consequently, the reported costs are estimates and do not reflect the exact costs of the resources used. Since this study focused specifically on hospital admission costs, we excluded physician billing data, as these data fall beyond the perspective of the hospital. It is also important to note that physician remuneration for procedures varies among provinces. Conclusion This study elucidates the financial implications of various patient demographic characteristics and clinical factors within a publicly funded health care system. Free flap reconstructions, which are increasingly favoured, comprised the majority of procedures analyzed. The primary cost drivers identified — reoperation, type of reconstruction, age older than 70 years, and bilateral reconstruction — highlight potential areas for preoperative cost optimization. To mitigate health care costs, especially in a single-payer, government-funded system, targeted strategies are essential to comprehend and address risk factors. This research highlights the need for effective management strategies to control costs of hospital admission while ensuring the provision of high-quality patient care. Footnotes Competing interests: None declared. Contributors: Oluwatobi Olaiya, Lucas Gallo, Christopher Coroneos, and Mark McRae contributed to the conception and design of the study. Oluwatobi Olaiya contributed to acquisition of data. Oluwatobi Olaiya, Beraki Abraha, Xi Ming Zhu, Sophocles Voineskos, Lawrence Mbuagbaw, and Zak Mireku contributed to the analysis and interpretation of data. Oluwatobi Olaiya, Beraki Abraha, and Christopher Coroneos contributed to writing the article. Oluwatobi Olaiya, Xi Ming Zhu, Lucas Gallo, Sophocles Voineskos, Zak Mireku, Lawrence Mbuagbaw, and Mark McRae contributed to critical review of the article. All authors gave final approval of the version to be published and agreed to be accountable for all aspects of the work. Disclaimer: This study contracted ICES Data & Analytic Services (DAS) and used deidentified data from the ICES Data Repository, which is managed by ICES with support from its funders and partners: Canada’s Strategy for Patient-Oriented Research (SPOR), the Ontario SPOR Support Unit, the Canadian Institutes of Health Research, and the Government of Ontario. The opinions, results, and conclusions reported are those of the authors. No endorsement by ICES or any of its funders or partners is intended or should be inferred. Parts of this material are based on data and information compiled and provided by the Canadian Institute for Health Information (CIHI). However, the analyses, conclusions, opinions, and statements expressed herein are those of the authors, and not necessarily those of CIHI. Parts of this material are based on data and information provided by Cancer Care Ontario (CCO). The opinions, results, views, and conclusions reported in this paper are those of the authors and do not necessarily reflect those of CCO. No endorsement by CCO is intended or should be inferred. References