Tomorrow's challenge: Bridging the information gap
Making data real for front-line caregivers
By Patrice Spath, ART
Consultant in Health Care Quality and Resource Management
Forest Grove, OR
[Editor's note: In addition to being the author of The Quality-Cost Connection column in Hospital Peer Review (HPR) since 1985, Spath has presented more than 200 educational programs on quality improvement, utilization review, case management/clinical paths, and outcomes management topics. For HPR's anniversary issue, she presents specific tips on data management and statistical analysis -- skills that she says quality professionals need to have.]
A schism between research and application of that research exists in health care today. For example, patients with localized breast cancer can undergo lumpectomy and postoperative irradiation with the same outcomes as those having a mastectomy, and yet the more radical procedure continues to be performed.
Studies have shown that practitioners often fail to prescribe adequate analgesics for cancer pain treatment, even though the Geneva-based World Health Organization and the Rockville, MD-based Agency for Health Care Policy and Research have disseminated cancer pain management guidelines.
These situations should cause us to ask the question, "Why have hospital quality improvement activities failed to put clinical trial results and guidelines into practice?" Although there is no one answer to this question, some of the responsibility may lie with the quality management professional (myself included). Perhaps we've been too focused on meeting the accreditation standards of the Joint Commission on Accreditation of Healthcare Organizations and failed to obtain the skills we need to effect important changes in patient care practices.
Caregivers must be provided with information that motivates them to improve their practices, and quality management professionals must be able to provide that information. To discover more about the current level of sophistication of today's quality management programs, I conducted an informal survey of 15 hospital-based quality management professionals throughout the country to determine how many are using more advanced scientific research tools. Although not a statistically valid study, the findings nonetheless suggest that hospital quality programs lack the sophistication considered to be important in most research studies.
Statistical tools not used
In response to the question "What summarizing numbers and statistical analysis techniques are used when evaluating measures of clinical performance at your facility?" more than half the respondents indicated that mean, range, and median are commonly reported for a data set. Four of the 15 respondents said they didn't use any summarizing numbers, not even the mean, range, or median. Only two of the respondents said they occasionally report the confidence intervals for a set of data, and none of the survey participants reported using linear regression, t-tests, ANOVA tests, Chi-square, or other statistical tests.
Quality management professionals must become more like researchers, regularly using statistical analysis techniques like Bayesian decision analysis, coefficient of variation, confidence intervals, and F-tests, among others. These analysis tools commonly are used in research studies to assist the data reviewers in determining the significance of the results. In many instances, these same tools should be used by hospital-based quality management professionals to tell the story in their data. (See related column on z-scores in main issue, p. 34.)
When designing guidelines, paths, or other action plans aimed at improving patient care, quality management professionals should promote a sound development methodology. The steps in clinical practice guideline development detailed by the Agency for Health Care Policy and Research are considered to be the gold standard, and yet many hospital-based guideline developers fail to follow these steps.
Of the 15 survey respondents, only six indicated that published research articles were even considered when developing paths or guidelines, and of those six, five indicated their clinicians use an evidence-grading system to evaluate the strength of published research articles. As facilitators of many guideline and path development initiatives, quality management professionals must help clinicians design clinically valid treatment protocols and the best way to do this is by following a systematic process which includes an analysis of scientific evidence.
Stimulate change with good data
Five years ago, health care institutions were data-poor. In many organizations, this is no longer the situation. Being data-rich, however, does not ensure caregivers receive the information they need to make good decisions.
Quality management professionals should learn where to find data in their institution and how to interpret the information once it is found. Quality management professionals must have a basic understanding of coding classification principles (ICD-9-CM, CPT-4, etc.) and the data element definitions used by the data gatherers. Without this knowledge, quality professionals are more likely to create reports that distort what the data have to say.
Conducting outcomes research in any environment involves the steps of: a) theory development; b) creation of hypotheses from theory; c) experimental design; and d) gathering data to test and support hypotheses. Traditional quality management activities have not followed this research model. It's time quality management professionals began to conduct outcomes management research to evaluate the relationship between specific health care services and the health of the population served by the provider.
One way to transition to a scientific model is to acquaint yourself with health services research literature. Journals such as Medical Care, Health Affairs, The Journal of Outcomes Management, and Clinical Performance and Quality Health Care offer readers examples of well-designed studies and suggestions for recreating these research-based activities in their own institutions. (See editor's note at the end of story for contact information on publishers.) By adding research journals to their reading list, quality management professionals can begin to blend epidemiology and biostatistics with their continuous quality improvement activities.
Gap between research, reality
There is a significant gap between research, clinical trials, scientific evidence and the day-to-day practice of performance improvement in hospitals. To help in narrowing this gap, quality management professionals must learn new skills and assert their information management expertise.
Quality management professionals must help clinicians clarify the relationship between service and outcomes, and provide caregivers with an understanding of how to manipulate variables to effect a desirable outcome. If the quality management profession cannot meet this challenge, the responsibility will be assumed by other specialists.
[Editor's note: For information on guidelines published by the Agency for Health Care Policy and Research, contact the AHCPR Publications Clearinghouse, P.O. Box 8547, Silver Spring, MD 20907, or call (800) 358-9295. Access AHCPR information on the Internet at http://text.nlm.nih.gov.
The publishers of the journals mentioned above are: Medical Care, J.B. Lippincott Co., Downsville Pike, Route 3, Box 20-B, Hagerstown, MD 21740 (800) 638-3030; Health Affairs, School of Public Health, The University of Michigan, Health Administration Press, Ann Arbor, MI 48109; The Journal of Outcomes Management, Porter Healthcare Communications, P.O. Box 4339, Burton, SC 29903, (803) 986-9500; and Clinical Performance and Quality Health Care, Slack Inc., 6900 Grove Road, Thorofare, NJ 08086, (609) 848-1000.] *
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content