Letters to the Editor
Some Points of Criticism
The authors analyzed billing data exclusively from members of the AOK statutory health insurance carrier and showed a survival advantage for different cancers if initial treatment was given in a certified hospital. The survival advantages, shown as absolute risk reductions, varied between 0.62 (lung cancer) and 4.61 (cervical cancer). The authors estimate that by referring all patients with incident cancers to certified centers could save about 33 000 life years every year (1).
My criticisms:
- The article is based on a secondary analysis of billing data (DRG system) that are compiled primarily for billing purposes (so called routine data). The German Federal Constitutional Court states: “Because of the complexity of this classification system (DRG (diagnosis related groups) system), a substantial amount of miscoding for the diagnoses and procedures that are crucial for billing undoubtedly happens.” And: “The problem is to a lesser extent intentional miscoding.” Furthermore, the Federal Audit Office [Bundesrechnungshof] criticizes in the context of coding that one in every two invoices is incorrect.
- This inevitably prompts the question of whether it is feasible on the basis of these non-validated AOK billing data to achieve comparability of both groups (certified versus non-certified) by means of risk adjustment using the Elixhauser score as undertaken by the authors.
- In addition to the survival period/length of survival, patient reported outcome measures should be included in assessing the quality of hospitals.
- The authors address the problem of causality because of their study design (regression analysis): “it seems unlikely to us that the association between initial treatment at a certified center and longer overall survival is not causal […]” The objection of a non-causal association remains valid, in my opinion, and could have been resolved by using propensity score analysis (or other causal inference methods, such as target trial emulation), which can absolutely reach the potential of a randomized trial.
DOI: 10.3238/arztebl.m2023.0263
Prof. Dr. med. Dr. med. dent. Ernst Hanisch
Arzt für Chirurgie, Schwerpunkt Viszeralchirurgie,
Chirurgische Intensivmedizin, Notfallmedizin,
Klinische Akut- und Notfallmedizin
e.hanisch@em.uni-frankfurt.de
Conflict of interest statement
The author declares that no conflict of interest exists.
| 1. | Schmitt J, Klinkhammer-Schalke M, Bierbaum V et al., on behalf of the WiZen Study Group: Initial cancer treatment in certified versus non-certified hospitals: results of the WiZen comparative cohort study. Dtsch Arztebl Int 2023; 120: 647–54 VOLLTEXT |
