[Zurück zur Trefferliste]

KBOLDT-2015 - Identifying patients receiving polypharmacy who are in need of pharmaceutical care–development and validation of a predictive model


Projektkennung VfD_KBOLDT-2015_10_003750
Laufzeit von 02/2010 bis 02/2015
Status des Projekts abgeschlossen



Fragestellung(en) question to be answered:
Can patients receiving polypharmacy who are at risk of hospitalization and in need of pharmaceutical care be identified by using routine data and data mining techniques?
Hintergrund / Ziele objectives:
Polypharmacy is associated with adverse drug events, which can lead to hospitalization and death. Providing pharmaceutical care can increase drug safety and reduce hospital admissions.
Methodik methode: A retrospective database analysis of health insurance data from 2005-2008 and 2007-2010 was performed. Patients aged 18-85 years on continuous polypharmacy receiving 5 or more drugs per quater for 4 quaters were included. Seriously ill patients were excluded. From the first data set (n=44.108) a predictive model was derived from 64 variables in a stepwise approach (CRISP-DM 1.0) using SPSS 19.0 and logistic regression.The final model containing 24 variables was validated using the second data set (n=45.739).
Datenbasis Sekundärdaten
   Krankenkassen - GKV  (Stichprobengröße: 1.700.000)
Studiendesign Querschnittstudie
retrospektive Datenbankanalyse
Untersuchte Geschlechter weiblich und männlich
Untersuchte Altersgruppen von 18 bis 85 Jahre
Ergebnisse results: Of 45.739 patients on polypharmacy 39% were admitted to hospital within one year, 88% were on medium polypharmacy using 5-8 drugs per quarter for at least one year. Compared to using the number of drugs as a solely predictor(>13 drugs: n=489, PPV=59.9%) the model identified a larger group of patients with a higher probability of hospitalization and a presumed need of pharmaceutical care (n=1.161, PPV=71.6%). The quality of the predictive model was acceptabel (AUC=65.2%, 95% CI 64.7-65.7%) and stabel over a two year period. The strongest predictors for hospitalization among patients on polypharmacy appeared to be number of different drugs per year, age, drug costs and the use of metamizol, opioids, loop-diuretics, phenprocoumon und clopidogrel.

discussion: Cut off points for polypharmacy and the need for pharmaceutical care are defined unevenly and information on adverse drug events and to drug related hospitalization are predominantly missing in routine data. Therefore surrogates like hospitalization in general have to be used to direct analysis. For this reason the derived predictive model identifies patients on polypharmacy at risk of hospitalization rather than patients on polypharmacy causing drug related hospitalization. Although it still provides a general tendency which drugs should be monitored more carefully. In addition routine data are lacking important clinical information to identify the need for pharmaceutical care and to fully use the potential of data mining techniques. Therefore quality of data should be improved to increase quality of prediction. This could be achieved by extended linkage of primary and secondary data records and intensified documentation of adverse drug events and drug related hospitalization by drug experts.

practical implications: Using data mining techniques help to narrow down mass data and to generated a predictive model providing interpretable and significant results. The most common definition of polypharmacy using 5 or more drugs as a cut off point can be confirmed but is of limited use. The total number of prescribed drugs per year seems to be a better predictor for hospitalization. Still the derived predictive model improves identification of patients on polypharmacy at risk of hospitalization and helps addressing phramaceutical care more exactly to patients in need. The model has been successfully used in a subsequent study where about 2.000 patients had to be identified for enrollment in a coaching project on the safe use of drugs.

Forschende und kooperierende Einrichtungen






Stand: 21.04.2016