«Thomas K. Bauer Katharina Lindenbaum Magdalena A. Stroka Susanne Ahrens Roland Linder Frank Verheyen Fall Risk Increasing Drugs: The Eﬀect on ...»
Thomas K. Bauer
Magdalena A. Stroka
Fall Risk Increasing Drugs:
The Eﬀect on Injuries of
the Frail Elderly Estimated
from Administrative Data
Ruhr Economic Papers
Ruhr-Universität Bochum (RUB), Department of Economics
Universitätsstr. 150, 44801 Bochum, Germany Technische Universität Dortmund, Department of Economic and Social Sciences Vogelpothsweg 87, 44227 Dortmund, Germany Universität Duisburg-Essen, Department of Economics Universitätsstr. 12, 45117 Essen, Germany Rheinisch-Westfälisches Institut für Wirtschaftsforschung (RWI) Hohenzollernstr. 1-3, 45128 Essen, Germany Editors Prof. Dr. Thomas K. Bauer RUB, Department of Economics, Empirical Economics Phone: +49 (0) 234/3 22 83 41, e-mail: firstname.lastname@example.org Prof. Dr. Wolfgang Leininger Technische Universität Dortmund, Department of Economic and Social Sciences Economics – Microeconomics Phone: +49 (0) 231/7 55-3297, email: W.Leininger@wiso.uni-dortmund.de Prof. Dr. Volker Clausen University of Duisburg-Essen, Department of Economics International Economics Phone: +49 (0) 201/1 83-3655, e-mail: email@example.com Prof. Dr. Christoph M. Schmidt RWI, Phone: +49 (0) 201/81 49 -227, e-mail: firstname.lastname@example.org Editorial Oﬃce Joachim Schmidt RWI, Phone: +49 (0) 201/81 49 -292, e-mail: email@example.com Ruhr Economic Papers #302 Responsible Editor: Christoph M. Schmidt All rights reserved. Bochum, Dortmund, Duisburg, Essen, Germany, 2011 ISSN 1864-4872 (online) – ISBN 978-3-86788-347-4 The working papers published in the Series constitute work in progress circulated to stimulate discussion and critical comments. Views expressed represent exclusively the authors’ own opinions and do not necessarily reﬂect those of the editors.
Ruhr Economic Papers #302 Thomas K. Bauer, Katharina Lindenbaum, Magdalena A. Stroka, Susanne Ahrens, Roland Linder, and Frank Verheyen
Fall Risk Increasing Drugs:
The Eﬀect on Injuries of the Frail Elderly Estimated from Administrative Data Bibliograﬁsche Informationen der Deutschen Nationalbibliothek
Die Deutsche Bibliothek verzeichnet diese Publikation in der deutschen Nationalbibliograﬁe; detaillierte bibliograﬁsche Daten sind im Internet über:
ISSN 1864-4872 (online) ISBN 978-3-86788-347-4 Thomas K. Bauer, Katharina Lindenbaum, Magdalena A.
Stroka, Susanne Ahrens, Roland Linder, and Frank Verheyen1 Fall Risk Increasing Drugs: The Eﬀect on Injuries of the Frail Elderly Estimated from Administrative Data Abstract Society beneﬁts on a large scale from improved medical care and pharmaceuticals.
The prescription of pharmaceuticals, however, also carries risks such as the possibility of an increased risk of falls, which may lead to severe injuries and increased health expenditures associated with these injuries. This study investigates the inﬂuence of several fall risk increasing drugs (FRIDs) on the number of injuries of elderly persons using multivariate regression models. Routine data from the Techniker Krankenkasse (TK) of frail elderly persons aged ≥ 65 years is analyzed for the year 2009 by estimating count data models, in order to take the data generating process of the number of injuries into account. The results of the count data model are compared to those from logistic regressions, which is the default regression model in this ﬁeld of research.
The empirical results suggest that antidepressants, anxiolytics, hypnotics and sedatives, antiarrhythmics, and drugs from the Priscus-list have a signiﬁcant positive eﬀect on the number of injuries, while antihypertensives and anti-parkinsonian agents show no and neuroleptics a signiﬁcant negative eﬀect. As recurrent injuries are common, the analysis of the number of injuries rather than just the probability of having an injury provides a more informative analysis of FRIDs.
JEL Classiﬁcation: I12, I19 Keywords: Fall risk increasing drugs; Priscus-List; frail elderly; multivariate regression;
count data models December 2011 1 Thomas K. Bauer, RWI, Essen, Ruhr-Universität Bochum (RUB), IZA, Bonn; Katharina Lindenbaum, RUB; Magdalena A. Stroka, RWI, RUB, Wissenschaftliches Institut der TK für Nutzen und Eﬃzienz im Gesundheitswese (WINEG), Hamburg; Susanne Ahrens, WINEG; Roland Linder, WINEG; Frank Verheyen, WINEG. – The authors are grateful to Julia Bredtmann and Sebastian Otten for helpful comments and suggestions. All correspondence to Magdalena Stroka, RuhrUniversität Bochum, Universitätsstr. 150, 44801 Bochum, E-Mail: Magdalena.Stroka@rub.de.
1 IntroductionFalls can cause serious injuries, disabilities and might even lead to death, in particular in case of elderly persons. About percent of the community-dwelling people aged years or older experience at least one fall every year. People being older than eighty years even fall with a probability of percent per year. The injuries resulting from falls often require extensive medical treatment and – let alone the threat to quality of life of the persons concerned – represent a non-negligible economic factor as these injuries often entail long and expensive rehabilitation periods.1 Against this background, the influence of FRIDs on falls or fall-related injuries of elderly persons has been examined extensively throughout the last decades. Meta-analyses and reviews reveal that the literature focuses almost exclusively on injuries of certain body regions, such as hip fractures, or on certain drugs such as psychotropic agents. Furthermore, existing studies –7 predominantly rely on cohort and case-control designs. Both designs should be considered with caution, as they often suffer from a lack of validity due to sample selection bias. Only a few studies are bas or representative cross-sectional data, mainly because of a lack of appropriate data. Furthermore, most existing studies rely on data for the US and Canada. Only one study investigates the effects of FRIDs using German data.13 Finally, existing empirical studies predominantly apply logistic regressions or Cox proportional .
Despite these differences in the data and empirical methods used, most studies conclude that it is inevitable to be cautious when prescribing drugs to the elderly, as they can increase the risk of injuries. In particular the following pharmaceuticals are classified as fall-risk increasing drugs (FRIDs): anxiolytics, hypnotics, sedatives, antidepressants, neuroleptics, analgetics, antihypertensives, antiarrhythmics and antiparkinsonian agents as well as drugs from the Beers criteria medication list describing potentially inappropriate medication consumption of older adults. Howev these results and stating that there are other factors causing fall-related injuries such as morbidity or diseases that were not considered in the analyses. After controlling for these factors, hardly any significant increase in fall risk due to drug consumption can be stated.11 Using routine data from a German statutory health insurance fund, this study investigates the specific risk of injuries due to the prescription of FRIDs in older ages. We estimate the influence of FRIDs on the number of injuries using a count data model. This regression model takes the specific data generating process of our dependent variable (the number of fall-related injuries) into account, which results in a non-negative variable with integer values, and hence avoids several limitations of the widely used logistic regression approach.
2.1. Data from the Techniker Krankenkasse (TK) The empirical analysis is based on data – one of the largest social health insurance funds in Germany. The data are part of the statutory administrative and medical records. older, who are in need of medical care at least on care level 1. * In contrast to most previous studies our data include both, frail elderly living at their homes and those living in institutions.
To avoid a potential source of self-selection bias, we only consider individuals who were insured by the TK for the entire With these insurants remain for the empirical analysis, i.e. .17
2.1.1. Sample Description
Table 1 provides definitions and descriptive statistics of the variables used in our analysis. On average, the individuals in our sampl being in need of care at level 1, followed by 37 percent at care level to the one reported by the German Federal Statistical Office . Concerning the health status of the frail elderly in our sample, Table 1 shows that most individuals in our sample suffer from
As dependent variable we employ injuries and fractures of all body parts, i.e., injuries with the - - frail elderly experienced at least one injury in the considered period of time. Even though the maximum number of recorded injuries is, . As the majority of injuries, such as fractures and softIn Germany, the long-term care insurances which are part of the health insurances distinguish between three care levels with increasing severity of care which are formally assessed by an independent Medical Review Board of the Statutory Health Insurance Funds.
tissue injuries of the elderly are related to falls, the number of recorded injuries can also be considered as a proxy for falls.
Neuroleptics Cardiovascular drugs Antihypertensives Antiarrhythmics Miscellaneous drugs Antiparkinsonian agents sum of prescribed DDDs of antiparkinsonian agents in
2.1.3. Fall Risk Increasing Drugs (FRIDs) As potentially FRIDs we consider the annual sums of prescribed daily defined doses (DDDs), an internationally comparable statistical unit of measurement for drug consumption. The sums of DDDs are calculated for the following drugs from the Anatomical Therapeutic Chemical (ATC)
Attention is also paid to drugs that are included in the Priscus-list, a list of potentially inappropriate medication for frail elderly people in Germany that resembles similar international lists such as the Beers-list. We exclude selected drugs from the Priscus-list having no systemic effects, such as, for example, Priscus-list
2.1.4. Descriptive Analysis