跳到正文
原文
Nature 健康护理主题·· 18 天前AI 评分35

结构化肿瘤训练疗法的健康保险利用和成本:一项回顾性分析

Healthcare utilization and costs associated with structured oncological training therapy: a retrospective analysis in cancer patients

AI 导读

一项回顾性研究分析了德国结构化肿瘤训练疗法(OTT)的健康保险利用情况。在152名注册患者中,38.1%的患者至少参加了一次训练,67.2%的训练组患者参加了至少12次训练。

正文

Abstract

Exercise is an essential component of supportive cancer care, improving physical function and quality of life. In Germany, oncological training therapy (OTT) provides a structured way to incorporate exercise into routine oncology care. This retrospective study examined the healthcare utilization of OTT, focusing on differences in healthcare utilization among active participants (T), interested non-participants (I), and a control group (C). Clinical and billing data from a health insurance company were analyzed. The control cohort was selected using predefined demographic, oncological, and regional criteria. The interested group included patients who enrolled but did not participate. The final study population comprised 289 patients: 39 OTT participants, 94 interested non-participants, and 156 controls. A total of 152 patients initially registered for OTT between 2015 and 2019. Outcomes covered inpatient treatment days, outpatient visits and one-year survival. Between-group comparisons were performed using Kruskal–Wallis tests with Dunn’s post hoc tests for continuous or ordinal outcomes and chi-square tests for categorical variables. Of the 152 registered patients, 58 (38.1%) attended at least one session, and 39 (67.2%) attended at least 12 of the 24 scheduled sessions. The training group had significantly more outpatient visits than the control group during the second observation period (t2; median: 13 vs. 7 visits; p = 0.0005), accompanied by higher outpatient costs (median: €639.29 vs. €267.89; p = 0.0006). The interested non-participant group had the highest comorbidity burden and significantly higher inpatient costs than the training group during the first observation period (t1; median: €5666.89 vs. €0; p < 0.0001). Although estimated one-year survival was numerically higher in the training group (94.9%) than in the control group (85.8%), no statistically significant difference in overall survival was observed (log-rank p = 0.945). This real-world analysis showed that participation in OTT was associated with higher outpatient healthcare utilization and costs, while no reduction in inpatient utilization was observed. Further prospective studies are needed to determine whether these utilization patterns translate into a clinical benefit.

Introduction

Malignant tumors vary greatly in terms of their clinical presentation, progression and therapeutic requirements1. Depending on the tumor type, patients receive a combination of treatments, including chemotherapy, radiotherapy, immunotherapy and surgery. These treatments improve patient prognosis1. However, these therapies and the malignancies themselves are often associated with many side effects such as fatigue, nausea, functional impairments, pain, and diminished quality of life2,3. These side effects place a burden on patients and their families. Supportive therapies, for example strength and endurance training, can reduce cancer and treatment side effects while maintaining autonomy. Studies demonstrate that these interventions effectively alleviate symptoms3,4,5,6. Exercise-based interventions have been shown to significantly improve benefits in quality of life, reducing pain, enhancing functional capacity and decreasing fatigue in patients with various tumor types4,6. Despite these documented benefits, there is a lack of comprehensive evaluations of the cost and time efficiency of integrating oncological training therapy (OTT) into standard oncological care. This limits its routine implementation in clinical practice7,8,9,10. Cancer treatment remains a major challenge for the healthcare system, not only regarding treatment efficacy and the development of further treatment approaches for different tumor entities1,8, but also in terms of the associated costs for healthcare companies, patients, and insurers11,12,13,14. In Germany, for instance, an estimated 261,800 men and 231,400 women were diagnosed with cancer in 202015,16.Lung cancer accounted for 34,100 cases in men and 22,590 in women in 202015, with an average treatment cost of €20,000 per patient15,17,18. Expenses peak in the first six months after diagnosis, primarily due to hospital stays15. Global cancer treatment costs are projected to rise to $458 billion by 2030, making cancer care a major driver of international healthcare costs8,14.These costs arise from the high prevalence of the disease15, and its inherent complexity1,8. Besides direct medical expenses, cancer incurs indirect costs, such as loss of income, employment disability, caregiver burden and psychological stress8.This retrospective study addresses the limited evidence on healthcare utilization and associated costs among patients participating in structured oncological training therapy (OTT). Healthcare utilization and costs were compared among OTT participants, a control cohort selected using predefined matching criteria and interested non-participants. The primary outcomes included the number of outpatient visits, outpatient treatment costs, inpatient treatment days, and inpatient treatment costs. Overall survival was assessed as an exploratory secondary outcome. The study aimed to investigate whether OTT participation was associated with differences in healthcare utilization, healthcare costs, and survival between the three groups. By examining healthcare utilization and associated costs, the study seeks to provide insights into the potential economic implications of OTT and inform future research on its integration into routine cancer care.

Methods

Study design

This retrospective cohort study examined the healthcare utilization and associated costs of patients with cancer during or after treatment, as well as demographic and clinical characteristics, and participation in Oncological Training Therapy (OTT), a standardized German exercise. Data were provided by a German statutory health insurance provider operating across multiple regions and nearly 27 million customers, making it one of the largest statutory health insurance companies in Germany. They pay up to €271 million a day for health promotion19. A total of 152 patients registered for OTT between February 2015 and March 2019.

OTT participation

Adult patients insured by the participating health insurance provider and diagnosed with cancer within the preceding 24 months were approached by their oncologist regarding potential OTT participation. Interested patients were interviewed and examined by a physician to determine their physical and psychosocial capability to participate in a structured training program. After signing the contract, patients were enrolled following positive evaluation. Participation was not limited by cancer type, prognosis or treatment intent. Patients receiving palliative cancer treatment were eligible for participation if they met the clinical and functional requirements.

Description of the therapy program and cohorts

The OTT program consisted of 24 supervised exercise sessions, each 50 min long, incorporating aerobic and resistance training delivered by qualified exercise professionals. Before analysis, attendance at a minimum of 12 of the 24 scheduled sessions was defined as the criterion for inclusion in the per-protocol training group. The training group (T) included patients, who did at least twelve units of OTT and met the pre-protocol criteria. Patients who attended one to twelve sessions were excluded due to low attendance. Patients who enrolled in OTT but never participated, were defined as the interested group (I). The control group (C) was generated by the health insurance company according to predefined matching criteria. Matching included tumor entity based on ICD-10 diagnosis (ICD-10 diagnosis, version 9, German modification) during the same observation period (Fig. 1: Formation of Cohorts), age, sex, time since initial cancer diagnosis, regional structure type based on postal code, and antineoplastic treatment status. To ensure a balanced representation of patients receiving active systemic cancer treatment, controls were selected to include patients with and without coded antineoplastic medication (ATC-L1). In addition, patients with selected severe psychiatric disorders or severe heart failure were excluded from the control population to reduce differences resulting from the medical eligibility assessment required for OTT participation. According to the predefined evaluation protocol, the health insurance company generated a control cohort with a planned three times the size of the group of patients who attended at least one OTT session. As 58 patients attended at least one training session, the initial control cohort comprised 174 patients. Tumor stage and overall comorbidity burden were not available as matching criteria and therefore could not be accounted for during control-group generation. However, 18 of the 174 patients were excluded from the primary control group due to the absence of a cancer diagnosis, resulting in a final control group of 156 patients for data analysis.

Figure 1

Formation of Cohorts.

Diagram to present the OTT participants and matched control group. Study participants: patients enrolled to the OTT; interested group: enrolled patients who never took part in OTT; dropout group: participation in less than 12 training units (the predetermined intervention); training group: participation in at least 12 training units.

Ethical approval and data protection

The ethics commission of the University Hospital Essen approved the study (19-8807-BO) prior to patient enrollment and data analysis in accordance with the Declaration of Helsinki and applicable federal and state law, as well as the medical professional code of conduct and the requirements of relevant national and international recommendations and guidelines (GCP, ICH). All participants of the OTT gave their written informed consent. Patients could withdraw from the study at any time before pseudonymization, after which data were permanently deleted. Data were pseudonymized by the health insurance company before being transferred to the data analysis team. Health insurance data are highly confidential. Hence, the analysis team was not involved in patient screening and selection to preserve patient anonymity. Similarly, the screening and matching team at the health insurance company was not involved in data analysis to ensure separation between cohort selection and statistical analysis.

Model to standardize observation periods

To facilitate comparison of the outpatient visits and inpatient treatment days and costs between the groups over the four-year study period, the model for standardized observation periods was divided into three-month intervals. Each patient was observed for three consecutive three-month periods. For the training group observation began on the date of the first training session. The date of enrollment was the start of the observation period for the interested group. For the control group, a reference date was assigned by the health insurance company. This date was defined as the day on which the same amount of time had elapsed since the initial cancer diagnosis as the mean interval between the initial cancer diagnosis and enrollment in the intervention group. The first three-month period (t1) covered the OTT-program (up to 12 weekly training sessions). The second observation period (months four to six; t2) covered training periods that lasted longer than planned and lasted until all physical activity programs were completed. In the final observation period (t3) patients were followed for three months after the last planned training session to assess healthcare utilization and costs after the scheduled intervention period. The number of outpatient visits and inpatient treatment days was determined retrospectively from health insurance data. Due to the pseudonymized nature of the data, visits to multiple specialists or changes in departments per day could not be deconvoluted, leading to several treatment days greater than the time under risk in some cases. The treatment days should therefore be interpreted as a relative measure of treatment density per patient / case.

Data collection

Demographic data, cancer diagnosis, diagnosis of comorbidities, and overall survival were documented and analyzed. The Charlson Comorbidity Index (CCI)20 was used to calculate the CCI for each patient´s comorbidities. All diagnoses were scanned and transmitted as ICD-10-diagnoses (version 9, German modification).

Throughout four-year study period, all outpatient visits and inpatient treatment days, treatment costs and medication costs were documented and analyzed according to a standardized observation period model. Costs and treatment days for each patient were collected and summarized over a nine- month period from the first training date or the date of enrollment. To compare the intervention and the treatment costs, all days and costs were pooled to detect all effects, including those on other comorbidities and the general medical efforts.

The primary outcomes were outpatient healthcare utilization, measured by the number of outpatient visits recorded, and inpatient healthcare utilization, measured by the number of inpatient treatment days, as well as their associated healthcare costs. Overall survival was assessed as an exploratory secondary outcome. Survival time was measured in months from the group-specific starting points to death or censoring.

Statistical analysis

A priori sample size and power analysis was not performed due to the exploratory nature of the study. Control-group selection and matching were performed by the health insurance company before transfer of the pseudonymized dataset. The analysis team did not perform an additional propensity-score or individual pair-matching procedure.

Statistical analysis was performed using SPSS (Version 29, International Business Machines, Armonk, New York, United States of America). Categorical data are presented as absolute numbers and percentages. Continuous variables were summarized using means, medians, interquartile ranges (IQRs), and minimum and maximum values. Categorical variables were reported as frequencies and percentages. Tukey-style boxplots with whiskers extending to 1.5x IQR, near outliers to 3x IQR, and far outliers >3x IQR, as per SPSS standard settings were used. Multi-group hypothesis-testing of ordinal- or higher-scaled data was performed using the Kruskal-Wallis test (nonparametric F-statistics) and Dunn’s post-tests with Holm-Bonferroni correction for multiple testing.

Nominal-scaled data were compared by Pearson`s Chi2 test. P-values for individual differences were generated from the residuals corrected for multiple testing by the Holm-Bonferroni method. Survival curves were generated using the Kaplan-Meier method and compared using the log-rank test. All statistical tests were two-sided, and p-values ≤ 0.05 were considered statistically significant. In addition to the primary per-protocol analyses, supplementary descriptive comparisons were performed using the broader cohort of patients who initiated OTT (≥ 1 session; n = 58) and the control group (n = 156). These analyses included baseline characteristics, comorbidity burden, overall survival, outpatient and inpatient healthcare utilization, and associated treatment costs. Unadjusted and Holm–Bonferroni-adjusted p-values were reported where applicable. The supplementary analyses were intended to provide additional descriptive information and were not considered a substitute for confounder-adjusted analyses.

Results

OTT participation

Of the 152 patients enrolled in OTT, 58 (38.1%) participated in at least one training session, while 94 patients (61.8%) were enrolled in the study, but never attended (interested group, I).

Among the 58 training patients, 67.2% (n = 39; training group, T) participated in at least 12 of 24 training sessions, while 32.7% (n = 19) were not included in the analysis because they did not meet the predefined attendance threshold and were excluded from the primary analysis (Fig. 2: OTT participation).

Figure 2

OTT Participation.

Percentage of patients who completed at least n training units out of all patients who completed at least one unit (all study participants minus interested patients without participation).

The training program was designed to consist of 24 sessions over 6 months. The interval between the first and last session of OTT ranged from 1 to 372 days among all patients who initiated training (median: 109 days; interquartile range (IQR: 116)), and from 46 to 372 days for the training group (median: 147 days; IQR: 78). This heterogeneity prompted the development of a model based on standardized observation periods after enrollment (for the control group and interested group) or after the first training session (for the training group and dropouts).

Patient characteristics

The training group had a median age of 59 years (IQR: 12) and included 20 women (51.3%) and 19 men (48.7%). Hematologic malignancies, lung cancer, and breast cancer were the most common cancer types (n = 8 each; 20.5%). The patient characteristics are also displayed in Table 1 “patient characteristics” below.

The control group showed a similar age distribution to the training group (median age: 57 (IQR: 18)) and showed a similar sex distribution (91 women (58.3%), 65 men (41.7%)). However, there were significant differences in cancer type distribution between the training and control groups. Specifically, 26.9% of patients in the control group had breast cancer (p = 0.038), as opposed to 20.5% in the training group. Distributions of other cancer types showed no significant differences. Likewise, the CCI did not differ significantly (T: 5 | C: 6). During the observation period, 42 of 156 (26.9%) patients in the control group died, compared to 11 of 39 patients (28.2%) in the training group.

There were no significant differences in age or sex. During the observation period, 38 of 94 patients died (40.4%) in the interested group. The interested group had a significantly higher CCI compared to the training group (median: 5 (T) vs. median:7 (I), p = 0.0308; IQR: 7 (T) vs. IQR: 4 (I)). Age and sex distributions were similar across groups, differences in comorbidity burden and cancer-type distribution indicated incomplete baseline comparability. These differences were considered potential confounders in the interpretation of healthcare utilization and survival outcomes. Gastrointestinal cancer was the most common malignancy in the interested group (n = 19; 20.2%), followed by lung (n = 14 | 14.9%), and urogenital cancer (n = 13 | 13.8%). In comparison with the training group, the interested group had a higher proportion of patients with head and neck cancer (I: n = 9, 9.6% vs. T: 1, 2.6%; p = 0.0095) and a lower proportion of patients with breast cancer (I: n = 8, 8.5% vs. T: n = 8, 20.5%; p = 0.0195). Detailed patient characteristics, including tumor-type distribution and comorbidity burden across the study cohorts, as well as additional comparisons of the CCI and age-adjusted CCI between the training and control groups, are presented in Supplementary Tables S1–S4.

Table 1 Patient's characteristics

Full size table

Survival time analysis

No statistical differences in overall survival were observed between the training and the control groups (p = 0.945). This was also observed when comparing the training group with the interested group (p = 0.215). Estimated 1-year survival rates were 94.9% in the training group and 78.7% in the interested group. Mean survival times were 30.72 months (T) and 26.52 months (I) (Fig. 3: Survival time analysis). Additional descriptive survival statistics, including mean survival time and estimated 1-year and 2-year survival rates for the training and control groups (full groups), are provided in Supplementary Table S5.

Figure 3

Survival time analysis.

Kaplan–Meier curves of the overall survival of the training group (n = 39), the control group (n = 156), and the interested group (n = 94) after study start. Dotted lines mark 12 and 24 months after enrollment, respectively. Tables to present the patients at risk.

Outpatient and inpatient treatment

Patients in the training group had significantly more outpatient visits than the control group across all three observation periods following the first training session (median differences: +5.5, + 6, and + 8 visits; p = 0.0045 (t1), p = 0.0005 (t2), p = 0.0009 (t3)). This was also reflected in outpatient treatment costs (median difference: €453.04 (t1), €371.40 (t2), €629.67 (t3); p = 0.0046(t1), p = 0.0006 (t2), p < 0.0001 (t3)) compared between the groups.

While there was no significant difference in inpatient treatment days between the training and control group, the interested group required significantly more inpatient treatment days than the training group at least for the entire active training period (medians at 0; mean differences: 17.77 d (t1), 3.25 d (t2), 0.54 d (t3); p < 0.0001 (t1); p = 0.0404 (t2); not significant (t3)). This was also reflected in higher inpatient treatment costs for the interested group (mean differences: €12,650.51 (t1), €2,224.39 (t2), €38.79 (t3); p < 0.0001 (t1), p < 0.0346 (t2), n.s. (t3)). The observed mean differences in inpatient treatment days between the interested and training groups decreased across the three observation periods, while the comparison was not statistically significant during t3 (Fig. 4: Outpatient and inpatient treatment).

Figure 4

Outpatient and inpatient treatment. (a) Model to standardize the observation time into three three-month periods t1, t2, t3 after the first training unit (training group), the study enrollment (interested group) or on an assigned reference date for the control group. (b) Outpatient and inpatient treatment costs (Euro, €) and effort (days in inpatient care or with outpatient visits) during the observation periods (months 0–3; 4–6; 7–9). Kruskal-Wallis Test for the F-statistics (f) with Dunn`s post-test and corrected for multiple testing using the Holm-Bonferroni. Only p-values ≤ 0.05 are shown. Data presented as SPSS-Boxplots: Mild outliers (°), 1.5-fold to 3-fold interquartile range beyond the quartiles; extreme outliers (*) > 3.0-fold interquartile range beyond the quartiles.

Detailed descriptive statistics and between-group comparisons of outpatient and inpatient healthcare utilization and associated costs across the three observation periods are provided in Supplementary Tables S6–S9.

Contrasting intervention costs and treatment costs

A single training unit costs €27.50, while the total cost for all 24 training units is €660.00. By contrast, the outpatient treatment costs per capita were up to €4558.30 and inpatient treatment costs per capita were €3377.20 (average for the training group). This is reflected in the combined outpatient and inpatient treatment costs over nine months.

The training group had significantly higher outpatient treatment costs in comparison to the control group (median T: €2968.79; C: €1336.56; p < 0.0001). In contrast, the interested group had significantly higher inpatient treatment costs than the training group (median T: €0; I: €11,621.21; p < 0.0001). Notably, the control group stood out as a heterogeneous group by a wide range of costs in contrast to the training group (outpatient treatment cost range T: €33,910.99; C: €82,549.99; inpatient treatment costs range T: €25,550.54; C: €42,989.64). The control group showed a wider range of recorded outpatient and inpatient treatment costs than the training group.

Discussion

This study contributes to the evaluation of healthcare utilization patterns associated with the incorporation of OTT into standard cancer care. Patients who actively participated in OTT exhibited increased outpatient care utilization compared to the control group. This pattern may reflect greater engagement with outpatient healthcare services, but the available data do not allow a clear determination of whether these contacts represented preventive or supportive care or could be translated into a clinical benefit. These findings align with existing international research on benefits of physical activity in enhancing the positive effects on the general well-being of cancer patients4,5,7,8.

Key findings and interpretation

A key finding of this study is the differential use of healthcare services among the training, interested and control groups. The training group had more recorded outpatient visits and higher outpatient costs than the control group. However, the available data do not indicate whether these visits were for preventive, supportive, or other healthcare services. This may be associated with regular participation in OTT. No statistically significant difference in inpatient treatment costs was observed between the training and control groups. Similar cost analyses, for German patients with diffuse large B-cell lymphoma, found that hospitalizations accounted for over 70% of total costs14. In contrast, median inpatient costs were zero in several observation periods in this study´s training cohort, suggesting heterogeneity and indicating that at least half of the recorded cost values were zero in these periods. Since OTT-related costs were incurred in the outpatient setting, the costs for inpatient treatment days were recorded separately. These findings highlighting the importance of distinguishing intervention costs from other healthcare expenditures when assessing healthcare utilization and costs. This is consistent with the growing evidence on the effectiveness of supportive oncology interventions23. One study found that multidisciplinary supportive care reduced symptom intensity, the number of hospital admissions, and overall healthcare costs. Although OTT is a more targeted therapy than comprehensive supportive oncology models, our results indicate that participation in structured exercise therapy is associated with a different pattern of healthcare utilization, characterized by greater outpatient care use.

Patients who were initially enrolled but ultimately did not participate had the highest inpatient treatment costs. The interested group had a higher comorbidity burden, as measured by the CCI20,21, and higher inpatient treatment costs than the training group. These differences may reflect variation in baseline health status, although the reasons for non-participation could not be determined from the available data. These findings suggest that health status may influence participation in structured exercise programs. While studies demonstrate the importance of tailored and inclusive OTT programs for patients with advanced disease stages3,4,6,24, the data imply that a threshold exists beyond which inclusion and participation become less feasible for certain patients.

No statistically significant differences in overall survival were observed between OTT participants and the comparison groups. These findings should be interpreted cautiously because of the observational design, baseline differences, and potential selection bias. Although this finding was not significant, it is noteworthy that the study included patients in a palliative care setting. OTT interventions may help maintain functional capacity and quality of life, aligning with international evidence linking exercise to improved survival and quality of life, especially in breast and colorectal cancer patients25,26,27. These results underscore the potential value of exercise as a supportive therapy for patients with limited prognoses across the entire spectrum of oncological care. Earlier research shows improved overall survival in the exercise therapy group compared to the health education group25.

Previous studies have reported beneficial effects of exercise interventions on survival and quality of life in selected cancer populations22,24,25,26,27,28,29,30. However, the present study did not provide statistically significant evidence of differences in overall survival between the study groups. Future research with larger cohorts, longer follow-up and appropriate adjustment for potential confounders are needed to further investigate the association between structured exercise therapy and survival outcomes.

Limitations

Several limitations must be acknowledged. This retrospective study design does not permit the establishment of a cause-and-effect relationship. Although matching procedures were used to improve group comparability, residual confounding factors cannot be ruled out. The heterogeneity of tumor types and treatment regimens within the cohort may have influenced patterns of healthcare use. Even though the control group was matched for age, sex, and oncological diagnosis (full matching described in chapter “Description of the Therapy Program and Cohorts”), differences in disease progression and treatment intensity may have contributed to cost differences. Because the cohort included multiple cancer types and the number of patients within individual tumor entities was small, tumor-specific subgroup analyses were not considered sufficiently robust. This heterogeneity limits the interpretation and generalizability of the observed healthcare utilization patterns. Lastly, selection bias is an inherent challenge in exercise intervention studies because individuals with better baseline fitness and motivation are more likely to participate in structured training programs. Despite the predefined demographic, oncological, treatment-related and regional matching criteria tumor stage and overall comorbidity burden were not included in the matching algorithm. Therefore, differences in disease severity and health status may have influenced both healthcare utilization and survival outcome. In addition, no a priori and power analysis was performed, and the relatively small training cohort may have limited the statistical power to detect between-group differences, particularly for survival and other secondary outcomes. The exclusion of participants with low training attendance after the initial control-group selection may have affected comparability between groups. As no additional matching was performed following these exclusions, potential selection bias and residual confounding cannot be ruled out. For fining the training group according to attendance at a minimum of 12 sessions may have introduced additional selection bias, as the ability to meet this threshold could depend on baseline health status, disease progression, hospitalization, or survival. Consequently, the final training group may represent a selected subgroup of patients with more favorable clinical characteristics or greater treatment adherence.

Certain patient groups appear to be underrepresented in this study. Patients with head and neck tumors were less frequently represented in the training group than in the interested group. Differences in disease-related symptoms and supportive care needs may have influenced participation, although the reasons of these were not assessed30,31. Head and neck cancer tends to affect older patients, with the highest incidence between ages 60 and 7032,33. The prognosis is often poor in advanced stages, with a 5-year survival rate between 40% and 63%, depending on tumor stage and response to therapy34. Common advanced-stage symptoms can severely impair everyday life31. Average treatment costs for advanced head and neck cancer are around €6606.33 to €39,459.00 per patient per year35,36.

The control group included more patients with breast cancer, possibly due to their earlier diagnoses through screening programs and the younger age at diagnosis. The underlying reasons for this difference could not be determined from the available data. Breast cancer is a leading cause of death among women in their 40s with the highest rate of new diagnoses between ages 70 and 7415,37. Advanced breast cancer often leads to bone pain, lymphoedema, fatigue and emotional distress, affecting physical activity and quality of life38,39,40,41. In advanced tumor stages, the prognosis for patients with breast cancer drops from over 80 to 25%15,42. The cost of treating metastatic breast cancer can vary widely, averaging between €20,000 and over €100,000 per year depending on disease progression and treatment43.

The differences in age distribution, functional limitations and treatment costs among tumor types can impact healthcare utilization and the effectiveness of exercise interventions. Consequently, heterogeneity in tumor biology, disease characteristics, and patient needs limits the generalizability of the findings. Future studies should include stratified analyses by tumor type, disease stage and comorbidity to enable precise and differentiated cost assessments.

Conclusion

The findings suggest that OTT participation is associated with increased outpatient healthcare utilization and treatment costs compared with the control group. No statistically significant difference in overall survival was observed. Given the retrospective design and potential confounding, the findings do not establish causal effects or clinical benefits of OTT participation. Further prospective studies are needed to validate these findings to investigate the long-term effects of exercise therapy on health outcomes and cost-effectiveness in oncology. This includes assessing of direct and indirect medical costs.

Data availability

The data that support the findings of this study are available from AOK Rhineland/Hamburg 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 AOK Rhineland/Hamburg.

References

  1. Debela, D. T. et al. New approaches and procedures for cancer treatment: Current perspectives. SAGE Open. Med. 9, 20503121211034366 (2021).

    Article  MathSciNet  PubMed  PubMed Central  Google Scholar 

  2. Al Maqbali, M., Al Sinani, M., Al Naamani, Z., Al Badi, K. & Tanash, M. I. Prevalence of fatigue in patients with cancer: A systematic review and meta-analysis. J. Pain Symptom Manage. 61, 167–189e14 (2021).

    Article  PubMed  Google Scholar 

  3. Tewes, M., Baumann, F., Teufel, M. & Ostgathe, C. Symptoms during outpatient cancer treatment and options for their management. Dtsch. Ärztebl Int. https://doi.org/10.3238/arztebl.m2021.0028 (2021).

    Article  PubMed  PubMed Central  Google Scholar 

  4. Fernandez, S. et al. Physical activity and cancer: A cross-sectional study on the barriers and facilitators to exercise during cancer treatment. Can. Oncol. Nurs. J. Rev. Can. Soins Infirm En Oncol. 25, 37–42 (2015).

    Article  Google Scholar 

  5. Alzahrani, N. The effect of hospitalization on patients’ emotional and psychological well-being among adult patients: An integrative review. Appl. Nurs. Res. ANR. 61, 151488 (2021).

    Article  PubMed  Google Scholar 

  6. Toohey, K. et al. The effects of physical exercise in the palliative care phase for people with advanced cancer: A systematic review with meta-analysis. J. Cancer Surviv Res. Pract. 17, 399–415 (2023).

    Article  Google Scholar 

  7. Beaudart, C., Drost, R. M. W. A., Evers, S. M. A. A., Paulus, A. T. G. & Hiligsmann, M. Associations between muscle mass/strength and healthcare costs/use for patients with cancer: A systematic literature review. Cancer Treat. Res. Commun. 33, 100633 (2022).

    PubMed  Google Scholar 

  8. Wang, Y. et al. Economic evaluation of exercise interventions for individuals with cancer: A systematic review. Prev. Med. 172, 107491 (2023).

    Article  PubMed  Google Scholar 

  9. Gubler-Gut, B. E., Pöhlmann, J., Flatz, A., Schwenkglenks, M. & Rohrmann, S. Cost-effectiveness of physical activity interventions in cancer survivors of developed countries: a systematic review. J. Cancer Surviv Res. Pract. 15, 961–975 (2021).

    Article  Google Scholar 

  10. Gallagher, A., Shersher, V., Mortimer, D., Truby, H. & Haines, T. The Cost-Effectiveness of Adjunctive Lifestyle Interventions for the Management of Cancer: A Systematic Review. Appl. Health Econ. Health Policy. 21, 225–242 (2023).

    Article  PubMed  Google Scholar 

  11. Krankheitskosten pro Kopf gleichen sich zwischen Männern. und Frauen weiter an. Statistisches Bundesamt https://www.destatis.de/DE/Presse/Pressemitteilungen/2022/07/PD22_316_236.html

  12. Luengo-Fernandez, R., Leal, J., Gray, A. & Sullivan, R. Economic burden of cancer across the European Union: A population-based cost analysis. Lancet Oncol. 14, 1165–1174 (2013).

    Article  PubMed  Google Scholar 

  13. Schlander, M., Hernandez-Villafuerte, K. & Thielscher, C. Kosten der Onkologie in Deutschland. Forum (Genova). 33, 330–337 (2018).

    Article  Google Scholar 

  14. Borchmann, P. et al. Healthcare resource utilization and associated costs of German patients with diffuse large B-cell lymphoma: A retrospective health claims data analysis. Oncol. Ther. 11, 65–81 (2023).

    Article  PubMed  Google Scholar 

  15. Ronckers, C. & Spix, C. Trübenbach. Krebs in Deutschland für 2019/2020. Ergeb Nach ICD. 10, 164 (2023).

    Google Scholar 

  16. Bray, F. et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J. Clin. 74, 229–263 (2024).

    PubMed  Google Scholar 

  17. McGuire, A., Martin, M., Lenz, C. & Sollano, J. A. Treatment cost of non-small cell lung cancer in three European countries: Comparisons across France, Germany, and England using administrative databases. J. Med. Econ. 18, 525–532 (2015).

    Article  CAS  PubMed  Google Scholar 

  18. Jovanoski, N. et al. Systematic literature review to identify cost and resource use data in patients with early-stage non-small cell lung cancer (NSCLC). PharmacoEconomics 41, 1437–1452 (2023).

    Article  PubMed  PubMed Central  Google Scholar 

  19. AOK. Wir über uns: die AOK auf einen Blick in Zahlen und Fakten. (2025).

  20. Charlson, M. E., Pompei, P., Ales, K. L. & MacKenzie, C. R. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J. Chronic Dis. 40, 373–383 (1987).

    Article  CAS  PubMed  Google Scholar 

  21. Malveiro, C. et al. Effects of exercise training on cancer patients undergoing neoadjuvant treatment: A systematic review. J. Sci. Med. Sport. 26, 586–592 (2023).

    Article  PubMed  Google Scholar 

  22. Worster, B., Zhu, Y., Garber, G., Kieffer, S. & Smith-McLallen, A. The impact of outpatient supportive oncology on cancer care cost and utilization. Cancer (2024) https://doi.org/10.1002/cncr.35332

    Article  PubMed  Google Scholar 

  23. Chase, P. et al. A review of the cost-effectiveness of supervised exercise therapy for adults with chronic conditions in the United States. Transl J. Am. Coll. Sports Med 10, (2025).

  24. Courneya, K. S. et al. Structured exercise after adjuvant chemotherapy for colon cancer. N Engl. J. Med.

  25. Holmes, M. D., Chen, W. Y., Feskanich, D., Kroenke, C. H. & Colditz, G. A. Physical activity and survival after breast cancer diagnosis. JAMA 293, 2479–2486 (2005).

    Article  CAS  PubMed  Google Scholar 

  26. Ratjen, I. et al. Postdiagnostic physical activity, sleep duration, and TV watching and all-cause mortality among long-term colorectal cancer survivors: A prospective cohort study. BMC Cancer. 17, 701 (2017).

    Article  PubMed  PubMed Central  Google Scholar 

  27. Friedenreich, C. M., Stone, C. R., Cheung, W. Y. & Hayes, S. C. Physical activity and mortality in cancer survivors: A systematic review and meta-analysis. JNCI Cancer Spectr. 4, pkz080 (2020).

    Article  PubMed  Google Scholar 

  28. Cannioto, R. A. et al. Physical activity before, during, and after chemotherapy for high-risk breast cancer: Relationships with survival. J. Natl. Cancer Inst. 113, 54–63 (2021).

    Article  PubMed  PubMed Central  Google Scholar 

  29. Talwar, B., Donnelly, R., Skelly, R. & Donaldson, M. Nutritional management in head and neck cancer: United Kingdom National Multidisciplinary Guidelines. J. Laryngol Otol. 130, S32–S40 (2016).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Jansen, F. et al. Effectiveness of a guided self-help exercise program tailored to patients treated with total laryngectomy: Results of a multi-center randomized controlled trial. Oral Oncol. 103, 104586 (2020).

    Article  PubMed  Google Scholar 

  31. Vahl, J. M. et al. Increasing mean age of head and neck cancer patients at a German tertiary referral center. Cancers 13, 832 (2021).

    Article  PubMed  PubMed Central  Google Scholar 

  32. Barsouk, A. et al. Risk factors, and prevention of head and neck squamous cell carcinoma. Med. Sci. 11, 42 (2023).

    Article  CAS  Google Scholar 

  33. Jansen, L. et al. Mundhöhlen- und Pharynxkarzinome: Inzidenz, Mortalität und Überleben in Deutschland. Bundesgesundheitsblatt - Gesundheitsforschung - Gesundheitsschutz. 64, 941–950 (2021).

    Article  PubMed  PubMed Central  Google Scholar 

  34. van der Linden, N. et al. Treatments and costs for recurrent and/or metastatic squamous cell carcinoma of the head and neck in the Netherlands. Eur. Arch. Otorhinolaryngol. 273, 455–464 (2016).

    Article  PubMed  Google Scholar 

  35. Schernberg, A., Sagaon-Teyssier, L. & Schwarzinger, M. Clinical and economic burden of head and neck cancer: A nationwide retrospective cohort study from France. Clin. Outcomes Res. CEOR. 11, 441–451 (2019).

    Google Scholar 

  36. Michaels, E., Worthington, R. O., Rusiecki, J. B. & Cancer risk assessment, screening, and primary prevention. Med. Clin. North. Am. 107, 271–284 (2023).

    Article  PubMed  Google Scholar 

  37. McLaughlin, S. A., Brunelle, C. L. & Taghian, A. Breast cancer–related lymphedema: Risk factors, screening, management, and the impact of locoregional treatment. J. Clin. Oncol. 38, 2341–2350 (2020).

    Article  PubMed  PubMed Central  Google Scholar 

  38. Venetis, K. et al. Breast cancer with bone metastasis: Molecular insights and clinical management. Cells 10, 1377 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Medeiros Torres, D., Koifman, J., da Silva Santos & R. & Impact on fatigue of different types of physical exercise during adjuvant chemotherapy and radiotherapy in breast cancer: systematic review and meta-analysis. Support Care Cancer. 30, 4651–4662 (2022).

    Article  PubMed  Google Scholar 

  40. Cardoso, F. et al. 6th and 7th International consensus guidelines for the management of advanced breast cancer (ABC guidelines 6 and 7). The Breast 76, (2024).

  41. Katalinic, A., Eisemann, N., Kraywinkel, K., Noftz, M. R. & Hübner, J. Breast cancer incidence and mortality before and after implementation of the German mammography screening program. Int. J. Cancer. 147, 709–718 (2020).

    Article  CAS  PubMed  Google Scholar 

  42. Khan, S. A., Hernandez-Villafuerte, K., Hernandez, D. & Schlander, M. Estimation of the stage-wise costs of breast cancer in Germany using a modeling approach. Front. Public. Health. 10, 946544 (2023).

    Article  PubMed  PubMed Central  Google Scholar 

  43. Wickmann, A. et al. Cost-benefit evaluation of advanced therapy lines in metastatic triple-negative breast cancer in Germany. Cost Eff. Resour. Alloc CE. 22, 21 (2024).

    Article  Google Scholar 

Download references

Acknowledgements

We acknowledge support by the Open Access Publication Fund of the University of Duisburg-Essen.

Funding

Open Access funding enabled and organized by Projekt DEAL. The authors received no financial support for this article’s research, authorship, or publication.

Author information

Authors and Affiliations

  1. Department of Palliative Medicine, West German Cancer Center, University Hospital Essen, University of Duisburg-Essen, Essen, Germany

    Louisa Zimmermann, Alena-Sophie Fahl, Mitra Tewes & Nico De Lazzari

  2. Center for Integrated Oncology Aachen Bonn Cologne Düsseldorf, Clinic I for Internal Medicine Oncology, University of Cologne, Cologne, Germany

    Freerk Baumann & Sarah Man

  3. Praxiszentrum Elisengalerie, 52062, Aachen, Germany

    Christoph Meyer

  4. AOK Rhineland/Hamburg, 40213, Düsseldorf, Germany

    Marion Hammling

Authors

  1. Louisa Zimmermann
  2. Alena-Sophie Fahl
  3. Freerk Baumann
  4. Sarah Man
  5. Christoph Meyer
  6. Marion Hammling
  7. Mitra Tewes
  8. Nico De Lazzari

Contributions

Authorship: L. Z. and A. F. wrote the main body of the manuscript. L.Z. and N.D.L. conducted the statistical analysis and created the figures. M.H. is our correspondent in the AOK. N.D.L. and M.T. supervised and guided the writing, publication and submission processes. All authors have read, reviewed and approved the final version of the manuscript. The list of authors should include all those who can legitimately claim authorship. This is all those who:1. Made a substantial contribution to the concept and design, acquisition of data or analysis and interpretation of data (A)2. Drafted the article or revised it critically for important intellectual content (B),3. Approved the version to be published (C)Authorship credit is based on substantial contributions to A, B and C. [International Committee of Medical Journal Editors, 1997]. Each author participated sufficiently in the work to take public responsibility for appropriate portions of the content. AuthorAll authors agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Corresponding author

Correspondence to Nico De Lazzari.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

This article is published under an open access license. Please check the 'Copyright Information' section either on this page or in the PDF for details of this license and what re-use is permitted. If your intended use exceeds what is permitted by the license or if you are unable to locate the licence and re-use information, please contact the Rights and Permissions team.

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zimmermann, L., Fahl, AS., Baumann, F. et al. Healthcare utilization and costs associated with structured oncological training therapy: a retrospective analysis in cancer patients. Sci Rep 16, 36600 (2026). https://doi.org/10.1038/s41598-026-78170-5

Download citation

  • Received: 28 March 2026

  • Accepted: 23 October 2026

  • Published: 02 November 2026

  • Version of record: 02 November 2026

  • DOI: https://doi.org/10.1038/s41598-026-78170-5

Keywords

来源:Nature 健康护理主题 · nature.com