DÄ internationalArchive7/2013Unanswered Questions
LNSLNS

The study reported by Kutschmann and colleagues (1) of the discriminative ability of a minimum caseload of 30 cases per year shows that minimum caseload reflects differences in the quality of care given by different hospitals to an unsatisfactory degree.

However, the study has methodological weaknesses. In calculating the logistic regression model, the authors did not consider the different levels of data (individual level and hospital level), and cluster effects were not corrected for. This may result in an overestimate of the association between caseload and mortality; for this reason, a multilevel model would have been more suitable (2).

Furthermore, some questions remain unanswered after reading the article. In addition to the effect of caseload on the mortality of preterm infants, it would have been interesting to investigate the extent to which the caseload of < 30 is even of relevance in explaining the different mortality rates at different hospitals. To this end one would have to calculate the proportion of overall mortality that is due to hospital-specific factors and not to individual factors, and determine how much of the proportion of hospital-specific factors is explained by the caseload. This can also be shown in a multilevel model (3). The relevance of several other quality indicators mentioned in the article—such as staffing structures and equipment—could also be captured in such a model.

Furthermore it would have been interesting not only to investigate the discriminative ability of a minimum caseload of 30 cases/year but also the degree to which a useful threshold value can be deduced at all. The possible model for this could be a study by the Institute for Quality and Efficiency in Health Care (Institut für Qualität und Wirtschaftlichkeit im Gesundheitswesen, IQWiG) of the threshold values for total knee joint endoprosthesis (4).

DOI: 10.3238/arztebl.2013.0117a

Dr. med. Melanie Eßer, MPH
Bayerische Arbeitsgemeinschaft für Qualitätssicherung, München

Prof. Dr. med. Rüdiger von Kries, Institut für Soziale Pädiatrie und Jugendmedizin, Ludwig-Maximilians-Universität München

Prof. Dr. rer. nat. Ulrich Mansmann

Institut für Medizinische Informationsverarbeitung, Biometrie und Epidemiologie, Ludwig-Maximilians-Universität München

Conflict of interest statement
The authors declare that no conflict of interest exists.

1.
Kutschmann M, Bungard S, Kötting J, Trümner A, Fusch C, Veit C:
The care of preterm infants with birth weight below 1 250 g: risk-adjusted quality benchmarking as part of validating a caseload-based management system. Dtsch Arztebl Int 2012; 109(31–32): 519–26 VOLLTEXT
2.
Hox JJ: Why do we need special multilevel analysis techniques. Multilevel Analysis: Techniques and Applications. New Jersey: Lawrence Erlbaum Associates 2002.
3.
Snijders TAB, Bosker RJ: Discrete dependent variables: In: Multilevel analysis. An introduction to basic and advanced multilevel modeling. 2nd edition. London: Sage Publications 2012; 289–322.
4.
IQWiG: Entwicklung und Anwendung von Modellen zur Berechnung von Schwellenwerten bei Mindestmengen für die Knie-Totalendoprothese. Abschlußbericht B05/01a. Köln: Institut für Qualität und Wirtschaftlichkeit im Gesundheitswesen (IQWiG), 2005.
1.Kutschmann M, Bungard S, Kötting J, Trümner A, Fusch C, Veit C:
The care of preterm infants with birth weight below 1 250 g: risk-adjusted quality benchmarking as part of validating a caseload-based management system. Dtsch Arztebl Int 2012; 109(31–32): 519–26 VOLLTEXT
2.Hox JJ: Why do we need special multilevel analysis techniques. Multilevel Analysis: Techniques and Applications. New Jersey: Lawrence Erlbaum Associates 2002.
3.Snijders TAB, Bosker RJ: Discrete dependent variables: In: Multilevel analysis. An introduction to basic and advanced multilevel modeling. 2nd edition. London: Sage Publications 2012; 289–322.
4.IQWiG: Entwicklung und Anwendung von Modellen zur Berechnung von Schwellenwerten bei Mindestmengen für die Knie-Totalendoprothese. Abschlußbericht B05/01a. Köln: Institut für Qualität und Wirtschaftlichkeit im Gesundheitswesen (IQWiG), 2005.

Info

Specialities