| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2006;37:2573.)
© 2006 American Heart Association, Inc.
Original Contributions |
From the Coordination Office and Central Data Centre of the ADSR, Stroke Register Northwest-Germany (P.U.H., K.B.), Institute of Epidemiology and Social Medicine (P.U.H., K.B., K.K.), University of Muenster, Stroke Register Rhineland-Palatinate (M.K.B., A.G., K.L.), Regional Office for Quality Assurance Rhineland-Palatinate (M.K.B.), SQMed gGmbH, Mainz, Department of Neurology (A.G., K.L.), Klinikum Ludwigshafen; German Stroke Society (O.B., E.B.R.), Commission "Stroke Units and Acute Stroke Treatment" of the German Stroke Society (E.B.R.), Department of Neurology (E.B.R., D.G.N.), University of Muenster, Commission "Health Care Research" of the German Stroke Society (P.K.R.), Unit for Stroke Research and Public Health Medicine (P.K.R.), University of Erlangen-Nuremberg; German Society of Neurology (O.B.), Minden, Commission "Quality Management" of the German Society of Neurology (R.W.C.J.), Frankfurt; Stroke Register Baden-Wuerttemberg (S.E.), Regional Office for Quality Assurance Baden-Wuerttemberg (GeQik), Stuttgart; Institute for Neurological-Neurosurgical Rehabilitation Research (U.H.), University of Magdeburg; Stroke Register Bavaria (P.H.), Regional Office for Quality Assurance Bavaria (BAQ), Munich; Stroke Register Hesse (B.M., G.M.v.R., R.W.C.J.), Regional Office for Quality Assurance Hesse (GQH) (B.M.), Eschborn, Department of Neurology (G.M.v.R.), Asklepios Klinik Bad Salzhausen, Nidda; Stroke Register North-Rhine (K.O.), Institute for Quality in Health Care North-Rhine (IQN), Duesseldorf; Department of Geriatric Medicine (L.P.), Ruhr University, Herne; Department of Neurology (D.S.), University of Munich; German Stroke Foundation (M.W.), Guetersloh, Germany.
Correspondence to Peter U. Heuschmann, MD, MPH, Institute of Epidemiology and Social Medicine, University of Münster, Domagkstr. 3, 48149 Münster. E-mail heuschma{at}uni-muenster.de
| Abstract |
|---|
|
|
|---|
Methods Quality indicators were developed by a multidisciplinary board between November 2003 and December 2005. The process was initiated by the German Stroke Registers Study Group in cooperation with the German Stroke Society, the German Society of Neurology, the German Stroke Foundation, Regional Offices for Quality Assurance and other experts proven in the field. National and international recommendations were considered during the development process. The process was based on a systematic literature review, an independent external evaluation of the process and its results, and a prospective pilot study to evaluate the defined indicators in clinical practice.
Results Overall a set of 24 indicators was developed to measure performance of acute care hospitals in the 3 health care dimensions structure, process and outcome as well as in 3 treatment phases prehospital, in-hospital/acute and postacute. Practicability of the derived indicators was tested in a prospective pilot study. During a 2-month period, 1006 patients in 13 hospitals were documented. Application of the new indicator set was found to be feasible by participating physicians and hospitals. Median time to document the required information for 1 patient was 5 minutes. Nationwide implementation of the new indicator set within regional registers in Germany started since April 2006.
Conclusions The development of indicators to measure hospital performance in stroke care is an important step toward improving stroke care on a national level. The chosen standardized evidence-based approach ensures maximal transparency, acceptance and sustainability of the developed indicators in Germany.
Key Words: acute care quality indicators stroke
| Introduction |
|---|
|
|
|---|
Up to now there is no national consensus about adequate indicators for measuring quality of stroke care in Germany. Therefore, a standardized evidence-based process was established to develop these indicators. The following summary reports theoretical background, used methodology, results and implementation of performance measures for stroke care.
| Methods |
|---|
|
|
|---|
For the development of evidence-based quality indicators in Germany, all professions and organizations involved in acute stroke care, stroke societies, quality improvement organizations as well as patient organizations in Germany were invited to join the Quality Indicator Board of the ADSR. Representatives of the Executive Board and the Commissions Stroke Unit and Health Care Research of the German Stroke Society, the Commission Quality Management of the German Society of Neurology, the German Stroke Foundation, Regional Offices for Quality Assurance, as well as specialists from the disciplines internal and geriatric medicine, and other experts from the field of quality management in acute stroke care agreed to join the Quality Indicator Board.
Methodological Procedures and Definitions
The standardized process for developing quality indicators was initialized within a workshop in November 2003 and ended after 4 additional workshops in December 2005. During the first meeting, methodological procedures to define quality indicators were agreed on. For this purpose published national and international recommendations for developing indicators of quality of health care were reviewed and presented in a standardized way.8,18,19 In addition, experiences from research groups actually developing performance measures for stroke were reported.11,16,20 The board decided to use the recommendations of the First Scientific Forum on Assessment of Quality of Care and Outcomes Research in Cardiovascular Disease and Stroke8 and the Standards for Defining Clinical Performance Measures in Germany18 as the basis of this development.
Definition of the Term Quality Indicator
The following definitions of the term quality indicator were used8,18: "quality indicators" are explicit standards of care against which actual clinical practice is judged; quality indicators should be followed for all suitable patients with the exception of extraordinary circumstances; quality indicators must also define how to practically identify those patients for whom a specific action should be taken; evidence-based guidelines can be used as a basis for their development but quality indicators are no guidelines.
Dimensions of Health Care to be Covered
The Donabedian concept was chosen as principle basis for the dimensions of acute stroke care to be covered by indicators.21 According to this concept, healthcare quality can be separated into structure, process and outcome.21 To consider relevance of different healthcare domains more carefully,8 the board decided to add the different phases prehospital, in-hospital/acute and postacute stroke care to the Donabedian concept, yielding in total 9 components of stroke care to be covered by indicators. These components served as a structural framework for selecting performance measures. For each dimension of health care, indicators were aimed to be defined to provide a profile of indicators without overemphasizing specific components.18
Methodological Requirements of Quality Indicators
Methodological requirements for ensuring validity of quality indicators were defined as follows8,18: a quality indicator must be meaningful; any potential indicator must be either a meaningful outcome to patients and society or closely related to such an outcome; the indicator must be valid and reliable; to serve as a useful marker of healthcare quality, it must be possible to measure structure, process, or outcome of interest; the indicator must be adjustable for patient variability so that observed differences between hospitals are attributed to the hospital performance and not caused by differences in patient characteristics; the indicator can be modified by improvements in processes of care, requiring variability after risk adjustment among hospitals; measuring performance of healthcare providers must be feasible on a routine basis.
Presentation of Quality Indicators
Following a proposal of the Joint Commission on Accreditation of Healthcare Organizations9 it was decided that a standardized report for each quality indicator should be provided. The report included: component of healthcare quality to be covered; rationale for selecting the indicator including references (original publications, guidelines and consensus statements); definition of suitable patients for whom the quality indicator is valid (included and excluded population for numerator and denominator); possible adjustment for patient variability; way of data report (eg, rate or proportion).
Selection of Quality Indicators
The board agreed to define potential indicators for measuring quality of stroke care in 2 steps. First, potential quality indicators for each dimension of health care to be covered were defined. Different working groups were constituted each responsible for a specific healthcare domain. Within each working group clinicians, epidemiologists and experts in quality management were represented. Previously published recommendations and indicator sets as well as individual experience of the members of the working group were regarded for the preselection of indicators. Potential indicators were suggested by the working groups, reviewed and finally selected after consensus of the whole board.
In the second step, the selected indicators were evaluated according to the predefined methodological requirements of quality indicators, as well as published evidence in the literature and recommendations in guidelines or consensus statements. A systematic literature review was performed, and its results were presented in a standardized report for each indicator including rating on level of evidence. The rules of the Oxford Centre for Evidence-Based Medicine were applied for grading evidence.22 The board decided to implement an external review process by peers, as well as a prospective pilot study to further improve the quality of the project.18 Subsequently, the process of quality indicator development and its results was evaluated by independent external reviewers, who were not involved in the actual process. The reviewers were asked to comment on the appropriateness of used methods, practicability of defined indicators in routine care and coverage of all relevant aspects of acute in-hospital stroke care by the indicator set. In addition, a prospective pilot study was planned to investigate methodological properties of selected indicators and practicability of data collection in routine clinical care. Voluntary hospitals within the ADSR registers were invited to take part in the pilot study. Hospitals with different expertise in acute stroke care, defined by the number of patients treated per year, were included.
| Results |
|---|
|
|
|---|
External Review
For the purpose of the external review a detailed preliminary report was drafted for the independent experts, including an extended description of the methodological procedure and the results of the process with an extensive presentation of all indicators. Experts from the field of methodological development of quality indicators, acute stroke care, neurology, internal and geriatric medicine, as well as rehabilitation served as voluntary reviewers. Overall, the indicator set was judged to be appropriate by the reviewers; no relevant aspects of stroke care were missed. One indicator was recommended to be split in 2 indicators (early rehabilitation in early rehabilitation-physiotherapy/occupational therapy and early rehabilitation-speech therapy), resulting now in total 24 potential indicators. Most remarks and comments of reviewers related to the presentation of indicators, especially to inclusion and exclusion criteria, case-mix adjustment and to description of specific definitions. As options in the future development of the indicator set, the experts suggested to also add indicators for prehospital processes as well as indicators adapted to the specific needs of inpatient rehabilitation in Germany. A final consensus decision about the indicator set was made by the board considering the recommendations of the reviewers. The whole indicator set was classified into 3 clusters with 7 indicators related to hospital structure, 14 indicators related to processes and outcome of patients during hospitalization and 3 indicators related to processes and outcome of patients at follow-up (supplemental Table II
, available online at http://stroke.ahajournals.org).
Prospective Pilot Study
A standardized questionnaire was developed that included the new indicator set and the variables needed to adjust for patients variability. In addition, a detailed manual of operation was prepared. Overall, 13 hospitals agreed to participate in the pilot study (92% departments of neurology; 85% provided Stroke Unit services; 31% treated <250, 31% 250 to 500 and 38% >500 stroke patients per year). All consecutively admitted patients with stroke or transient ischemic attack within these hospitals during a 2-month period were aimed to be documented. The expected number of patients treated within the hospitals during this time period was 1175. In total, data on 1006 patients with stroke or transient ischemic attack were collected yielding a documentation rate of 86% (median 96%, range per hospital 25% to 100%). A sufficient variance between hospitals in the performance of the indicators could be documented (Table). Overall the percentage fulfilled for several indicators was high. The largest variation occurred for the indicator screening of patients for swallowing disorders, the smallest for brain imaging in stroke suspicious patients. A standardized quality report for each indicator was prepared and distributed to the hospitals for comments (example in the Figure). The collection of the new indicator set was judged to be practicable by the physicians in the participating hospitals. The median time to document the required information for each patient was 5 minutes. Comments of participating hospitals related mainly to description of variables and presentation of results.
|
|
|
|
|
Implementation of Defined Quality Indicators in Germany
Based on the results of the pilot study, the questionnaire and the manual of operation was revised finally. The complete documentation for each quality indicator will be provided to interested readers on request by the study group. The executives of the participating ADSR registers aim to implement the new indicator set within the regional registers starting from April 2006. To adjust for changes in the German healthcare system or new trends in stroke treatment the board agreed on a periodical update of the quality indicators scheduled in 2-year time intervals.
| Discussion |
|---|
|
|
|---|
Quality improvement programs can actually improve acute stroke care in hospitals as recently demonstrated, eg, by increased rates of tissue plasminogen activator use23 or by lower pneumonia rates in hospitals with formal dysphagia screening protocols.24 One widely used tool to improve quality of hospital care is the regular feedback of the performance in specific indicators.15 Several initiatives to improve quality of stroke care in different countries have defined specific indicators to compare the performance of hospitals.9,11,13,14,12 Some similarities in the defined indicators between theses initiatives and our proposal can be found, especially regarding in-hospital processes. For example, all initiatives include indicators related to antithrombotics at discharge in eligible patients.9,11,13,14,12 The majority of the programs also include indicators related to Stroke Unit or multidisciplinary team care,11,14 to specific aspects of stroke education9,13,14,12 and to tissue plasminogen activator administration.9,13,14,12 Comparable to our results, some initiatives explicitly include indicators on organizational aspects11 or focus on long-term outcome of stroke patients to measure quality of care.14 However, the indicator sets used in the different quality improvement programs demonstrate also substantial variations. These differences can partly be explained by different methods in the development of the indicators and by different aims when developing performance measures. In addition, the characteristics of the specific healthcare system had to be considered because the experience from one country could not be completely transferred to other healthcare settings without adaptations to the specific local needs. For example, the performance measure home visits before discharge from the Royal College of Physicians key indicators for stroke care11 could not be defined as an indicator for acute hospital care in Germany because of the strict separation between outpatient and inpatient care.
Our development process has strengths and limitations. We decided to develop quality indicators using a standardized, evidence-based approach following previously defined recommendations to ensure high methodological quality and maximal transparency of our results.8,18 A systematic literature review and an external evaluation were implemented as quality criteria to avoid the overemphasis of specific aspects of health care. The indicators were developed by a multidisciplinary board to guarantee a wide acceptance of the results by institutions and organizations engaged in acute stroke care in Germany. The whole process was implemented within the context of ongoing stroke registers in Germany to guarantee sustainability of the program and facilitate its implementation. The information required for calculating the defined indicators will be collected prospectively by the treating physicians. There is evidence that data feedback efforts must be perceived as valid by physicians to motivate changes.25 Thus, results of our quality initiative might be more able to change current practice compared with initiatives that use information from administrative records alone because the latter might be perceived as invalid. The annual pooled data analysis of the regional stroke registers within the ADSR will allow to evaluate prospectively the impact of the implementation of the indicator set on quality of acute stroke care in Germany. Our study has also weaknesses. Because the aim of the process was to develop indicators for in-hospital stroke care in Germany, no performance measures on a population level, eg, recurrence rates, could be considered. In addition, in Germany acute care hospitals have no direct influence in processes of prehospital emergency care. Thus, only prehospital indicators related to hospital structure, like training of emergency medical services, were included in the indicator set. Furthermore, the necessary resources for the data collection must further be minimized, eg, by facilitated electronic data capture. One major challenge to be solved is a regular follow-up of all stroke patients in Germany. Currently experience on indicators related to long-term follow-up of stroke patients in Germany is derived from different regional activities only.26 The broad follow-up of patients after discharge is actually hampered by logistics, costs and legal legislations in several federal states in Germany such as requirement of a written informed consent of the patient for a central follow-up. These problems also limit a regular follow-up of stroke patients in quality initiatives in other countries with similar legislation like Germany.27
Conclusion
The development of evidence-based indicators for measuring quality of acute stroke care in Germany is an important step toward a standardized audit of stroke care on a national level. The chosen standardized approach has ensured maximal transparency, acceptance and sustainability of the defined performance measures among all professions and organizations involved in acute stroke care in Germany.
| Acknowledgments |
|---|
Finally, we want to thank the following hospitals for their participation in the pilot phase:
Bad Zwesten: Neurologische Akutklinik; Bonn: Ev. Kliniken Bonn gGmbH Betriebsstätten Johanniter-Krankenhaus; Ebersberg: Abteilung für Innere Medizin der Kreisklinik; Emden: Neurologische Abteilung des Hans-Susemihl-Krankenhauses; Erlangen: Neurologische Klinik des Universitaetsklinikums; Hagen: Neurologische Klinik des St. Johannes Hospitals; Luedenscheid: Neurologische Klinik des Klinikums; Muenchen: Abteilung für Neurologie des Staedtischen Krankenhauses Harlaching; Muenster: Neurologische Klinik des Universitaetsklinikums; Nidda-Bad Salzhausen: Asklepios Neurologische Klinik Bad Salzhausen; Paderborn: Neurologische Klinik des St Vincenz Krankenhauses; Teupitz: Abteilung für Neurologie der Landesklinik; Westerstede: Neurologische Klinik der Ammerland Klinik
Sources of Funding
The study was partly supported by the German Federal Ministry of Research (BMBF) within the Competence Net Stroke. There was no financial or other support by other third parties for the working process or manuscript preparation.
Disclosures
None.
| Footnotes |
|---|
Received June 15, 2006; revision received July 13, 2006; accepted July 21, 2006.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. Lindsay RN PhD, M. Bayley MD, A. McDonald BScPT, I. D. Graham PhD, G. Warner PhD, and S. Phillips MBBS Toward a more effective approach to stroke: Canadian Best Practice Recommendations for Stroke Care Can. Med. Assoc. J., May 20, 2008; 178(11): 1418 - 1425. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. A. Hinchey, T. Shephard, S. T. Tonn, R. Ruthazer, H. P. Selker, and D. M. Kent Benchmarks and Determinants of Adherence to Stroke Performance Measures Stroke, May 1, 2008; 39(5): 1619 - 1620. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Lyons and A. G. Rudd Health Policy and Outcomes 2006 Stroke, February 1, 2007; 38(2): 229 - 231. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Stroke Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2006 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |