Exploring relationships between in-hospital mortality and hospital case volume using random forest: results of a cohort study based on a nationwide sample of German hospitals, 2016–2018
Abstract Background Relationships between in-hospital mortality and case volume were investigated for various patient groups in many empirical studies with mixed results. Typically, those studies relied on (semi-)parametric statistical models like logistic regression. Those models impose strong assu...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2022-01-01
|
| Series: | BMC Health Services Research |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12913-021-07414-z |
| _version_ | 1818940601967575040 |
|---|---|
| author | Martin Roessler Felix Walther Maria Eberlein-Gonska Peter C. Scriba Ralf Kuhlen Jochen Schmitt Olaf Schoffer |
| author_facet | Martin Roessler Felix Walther Maria Eberlein-Gonska Peter C. Scriba Ralf Kuhlen Jochen Schmitt Olaf Schoffer |
| author_sort | Martin Roessler |
| collection | DOAJ |
| description | Abstract Background Relationships between in-hospital mortality and case volume were investigated for various patient groups in many empirical studies with mixed results. Typically, those studies relied on (semi-)parametric statistical models like logistic regression. Those models impose strong assumptions on the functional form of the relationship between outcome and case volume. The aim of this study was to determine associations between in-hospital mortality and hospital case volume using random forest as a flexible, nonparametric machine learning method. Methods We analyzed a sample of 753,895 hospital cases with stroke, myocardial infarction, ventilation > 24 h, COPD, pneumonia, and colorectal cancer undergoing colorectal resection treated in 233 German hospitals over the period 2016–2018. We derived partial dependence functions from random forest estimates capturing the relationship between the patient-specific probability of in-hospital death and hospital case volume for each of the six considered patient groups. Results Across all patient groups, the smallest hospital volumes were consistently related to the highest predicted probabilities of in-hospital death. We found strong relationships between in-hospital mortality and hospital case volume for hospitals treating a (very) small number of cases. Slightly higher case volumes were associated with substantially lower mortality. The estimated relationships between in-hospital mortality and case volume were nonlinear and nonmonotonic. Conclusion Our analysis revealed strong relationships between in-hospital mortality and hospital case volume in hospitals treating a small number of cases. The nonlinearity and nonmonotonicity of the estimated relationships indicate that studies applying conventional statistical approaches like logistic regression should consider these relationships adequately. |
| first_indexed | 2024-12-20T06:42:15Z |
| format | Article |
| id | doaj.art-db32b6d48c8d4485a29c3b0883978f65 |
| institution | Directory Open Access Journal |
| issn | 1472-6963 |
| language | English |
| last_indexed | 2024-12-20T06:42:15Z |
| publishDate | 2022-01-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Health Services Research |
| spelling | doaj.art-db32b6d48c8d4485a29c3b0883978f652022-12-21T19:49:49ZengBMCBMC Health Services Research1472-69632022-01-0122111110.1186/s12913-021-07414-zExploring relationships between in-hospital mortality and hospital case volume using random forest: results of a cohort study based on a nationwide sample of German hospitals, 2016–2018Martin Roessler0Felix Walther1Maria Eberlein-Gonska2Peter C. Scriba3Ralf Kuhlen4Jochen Schmitt5Olaf Schoffer6Center for Evidence-based Healthcare, University Hospital Carl Gustav Carus and Medical Faculty at the Technische Universität DresdenCenter for Evidence-based Healthcare, University Hospital Carl Gustav Carus and Medical Faculty at the Technische Universität DresdenQuality and Medical Risk Management, University Hospital Carl Gustav Carus DresdenIQM Initiative Qualitätsmedizin e.V.IQM Initiative Qualitätsmedizin e.V.Center for Evidence-based Healthcare, University Hospital Carl Gustav Carus and Medical Faculty at the Technische Universität DresdenCenter for Evidence-based Healthcare, University Hospital Carl Gustav Carus and Medical Faculty at the Technische Universität DresdenAbstract Background Relationships between in-hospital mortality and case volume were investigated for various patient groups in many empirical studies with mixed results. Typically, those studies relied on (semi-)parametric statistical models like logistic regression. Those models impose strong assumptions on the functional form of the relationship between outcome and case volume. The aim of this study was to determine associations between in-hospital mortality and hospital case volume using random forest as a flexible, nonparametric machine learning method. Methods We analyzed a sample of 753,895 hospital cases with stroke, myocardial infarction, ventilation > 24 h, COPD, pneumonia, and colorectal cancer undergoing colorectal resection treated in 233 German hospitals over the period 2016–2018. We derived partial dependence functions from random forest estimates capturing the relationship between the patient-specific probability of in-hospital death and hospital case volume for each of the six considered patient groups. Results Across all patient groups, the smallest hospital volumes were consistently related to the highest predicted probabilities of in-hospital death. We found strong relationships between in-hospital mortality and hospital case volume for hospitals treating a (very) small number of cases. Slightly higher case volumes were associated with substantially lower mortality. The estimated relationships between in-hospital mortality and case volume were nonlinear and nonmonotonic. Conclusion Our analysis revealed strong relationships between in-hospital mortality and hospital case volume in hospitals treating a small number of cases. The nonlinearity and nonmonotonicity of the estimated relationships indicate that studies applying conventional statistical approaches like logistic regression should consider these relationships adequately.https://doi.org/10.1186/s12913-021-07414-zHospital mortalityVolume-outcome relationshipCohort studyRisk factorsRandom ForestNonparametric modelling |
| spellingShingle | Martin Roessler Felix Walther Maria Eberlein-Gonska Peter C. Scriba Ralf Kuhlen Jochen Schmitt Olaf Schoffer Exploring relationships between in-hospital mortality and hospital case volume using random forest: results of a cohort study based on a nationwide sample of German hospitals, 2016–2018 BMC Health Services Research Hospital mortality Volume-outcome relationship Cohort study Risk factors Random Forest Nonparametric modelling |
| title | Exploring relationships between in-hospital mortality and hospital case volume using random forest: results of a cohort study based on a nationwide sample of German hospitals, 2016–2018 |
| title_full | Exploring relationships between in-hospital mortality and hospital case volume using random forest: results of a cohort study based on a nationwide sample of German hospitals, 2016–2018 |
| title_fullStr | Exploring relationships between in-hospital mortality and hospital case volume using random forest: results of a cohort study based on a nationwide sample of German hospitals, 2016–2018 |
| title_full_unstemmed | Exploring relationships between in-hospital mortality and hospital case volume using random forest: results of a cohort study based on a nationwide sample of German hospitals, 2016–2018 |
| title_short | Exploring relationships between in-hospital mortality and hospital case volume using random forest: results of a cohort study based on a nationwide sample of German hospitals, 2016–2018 |
| title_sort | exploring relationships between in hospital mortality and hospital case volume using random forest results of a cohort study based on a nationwide sample of german hospitals 2016 2018 |
| topic | Hospital mortality Volume-outcome relationship Cohort study Risk factors Random Forest Nonparametric modelling |
| url | https://doi.org/10.1186/s12913-021-07414-z |
| work_keys_str_mv | AT martinroessler exploringrelationshipsbetweeninhospitalmortalityandhospitalcasevolumeusingrandomforestresultsofacohortstudybasedonanationwidesampleofgermanhospitals20162018 AT felixwalther exploringrelationshipsbetweeninhospitalmortalityandhospitalcasevolumeusingrandomforestresultsofacohortstudybasedonanationwidesampleofgermanhospitals20162018 AT mariaeberleingonska exploringrelationshipsbetweeninhospitalmortalityandhospitalcasevolumeusingrandomforestresultsofacohortstudybasedonanationwidesampleofgermanhospitals20162018 AT petercscriba exploringrelationshipsbetweeninhospitalmortalityandhospitalcasevolumeusingrandomforestresultsofacohortstudybasedonanationwidesampleofgermanhospitals20162018 AT ralfkuhlen exploringrelationshipsbetweeninhospitalmortalityandhospitalcasevolumeusingrandomforestresultsofacohortstudybasedonanationwidesampleofgermanhospitals20162018 AT jochenschmitt exploringrelationshipsbetweeninhospitalmortalityandhospitalcasevolumeusingrandomforestresultsofacohortstudybasedonanationwidesampleofgermanhospitals20162018 AT olafschoffer exploringrelationshipsbetweeninhospitalmortalityandhospitalcasevolumeusingrandomforestresultsofacohortstudybasedonanationwidesampleofgermanhospitals20162018 |