Berlowitz, D.R., Christiansen, C.L., Brandeis, G.H., and others (2002). "Profiling nursing homes using Bayesian hierarchical modeling." (AHRQ grant HS09768). Journal of the American Geriatrics Society 50, pp. 1126-1130.
It is particularly difficult to profile the health care performance of nursing homes using statistical techniques. For example, many nursing homes are relatively small, so extreme values of the outcome are likely to arise by chance. However, a new statistical method, Bayesian hierarchical modeling, can successfully profile nursing home performance, according to these authors. They analyzed nursing home data from 1997 and 1998 to compare nursing home performance on risk-adjusted rates of pressure ulcer development (a marker of poor quality care) calculated using standard statistical techniques and Bayesian hierarchical modeling. The range of risk-adjusted rates among nursing homes was 0 to 14.3 percent using standard methods and 1 to 4.8 percent using Bayesian analysis. Fifteen nursing homes were designated as outliers based on their z scores, but only two were outliers using Bayesian modeling.
Birkmeyer, N.J., Weinstein, J.N., Tosteson, A., and others (2002). "Design of the spine patient outcomes research trial (SPORT)." (AHRQ grant K02 HS11288). Spine 27(12), pp. 1361-1372.
This article describes the rationale and design of the Spine Patient Outcomes Research Trial (SPORT). The objective is to assess the relative efficacy and cost-effectiveness of surgical and nonsurgical approaches to the treatment of common conditions associated with low back and leg pain. The SPORT project is being conducted at 11 clinical centers around the United States. It involves three simultaneous multicenter, randomized, controlled clinical trials. The study includes patients with the three most common diagnoses for which spine surgery is performed: intervertebral disc herniation, spinal stenosis, and degenerative spondylolisthesis. It compares the most commonly used standard surgical and nonsurgical treatments for these patients. Patients are being followed for at least 2 years, with visits scheduled at 6 weeks and at 3, 6, 12, and 24 months.
Cegala, D.J., McClure, L., Marinelli, T.M., and Post, D.M.
(2002). "The effects of communication skills training on patients' participation during medical interviews." (AHRQ grant HS09520). Patient Education and Counseling 41, pp. 209-222.
Patients often do not ask questions during doctor visits, even
though virtually all patients claim they want as much information as possible. This study examined the effectiveness of a training booklet designed to enhance patients' communication skills in information exchange with doctors. A total of 25 doctors each saw six patients, two patients in each of three communication skills interventions (trained, informed, and control). Trained patients received a work booklet with examples and practice opportunities to describe symptoms and related medical history for various illnesses. The informed group received a brief summary of the major points covered in the booklet, and the control group received no intervention. Trained patients engaged in more effective and efficient information seeking, provided doctors with more detailed information about their medical condition, and used more summarizing statements to verify information they received from physicians. They also demonstrated a more patient-controlled style of communication compared with informed and untrained patients.
Deb, P. and Trivedi, P.K. (2002). "The structure of demand for health care: Latent class versus two-part models." (AHRQ grant HS09843). Journal of Health Economics 21, pp. 601-625.
These authors examine empirical strategies for modeling the demand for health services, which has importance for areas such as the price sensitivity of the demand for medical services and predicted likelihood of being extensive users of services. They contrast the two-part model (TPM) that distinguishes between users and non-users of health care with a latent class model (LCM) that distinguishes between infrequent and frequent users. In model comparisons using data accounts of use from the RAND Health Insurance Experiment, the investigators found strong evidence in favor of the LCM. They showed that individuals in the infrequent and frequent user latent classes may be described as being healthy and ill, respectively. Although sample averages of price elasticities, conditional means, and event probabilities were not statistically different, the estimates of these policy-relevant measures were substantially different when calculated for hypothetical individuals with specific characteristics.
Finkelstein, J.A., Lozano, P., Streiff, K.A., and others (2002, June). "Clinical effectiveness research in managed-care systems: Lessons from the pediatric asthma care PORT." (AHRQ grant HS08368). Health Services Research 37(3), pp. 775-789.
These authors use the Pediatric Asthma Care Patient Outcomes Research Team (PORT) to highlight the unique challenges of evaluative research to change medical practice in the "real world" settings of contemporary managed care organizations (MCOs). The Pediatric Asthma Care PORT is a 5-year initiative to study strategies for asthma care improvement in three MCOs in Chicago, Seattle, and Boston. It compares two care improvement strategies with usual care: first, a targeted physician education program using practice-based peer leaders (PLs) as change agents, and second, the PL intervention plus a planned asthma care intervention incorporating joint asthma check-ups by nurse-physician teams. For this article, the researchers reviewed the organization and implementation challenges in conducting the PORT (as described by investigators at each health plan) for common themes and lessons that might be useful to investigators planning interventional research in similar care delivery settings.
Freburger, J.K., and Konrad, T.R. (2002, June). "The use of Federal and State databases to conduct health services research related to physical and occupational therapy." (National Research Service Award training grant T32 HS00032). Archives of Physical and Medical Rehabilitation 83, pp. 837-845.
With the aging of the population and increasing prevalence of chronic disease, the roles of physical therapy (PT) and occupational therapy (OT) are likely to become increasingly important. These researchers identified State and Federal databases maintained by the National Center for Health Statistics, the Agency for Healthcare Research and Quality, and the Centers for Medicare and Medicaid Services as having potential for answering questions related to the use and effectiveness of PT and OT. They explored these databases to determine if PT and OT had sufficient representation, and they identified potential questions that could be answered by examining the databases in more detail. They also identified some of the advantages, disadvantages, and methodologic issues of using the databases.
Hermann, R.C., Finnerty, M., Provost, S., and others (2002). "Process measures for the assessment and improvement of quality of care for schizophrenia." (AHRQ grant HS10303). Schizophrenia Bulletin 28(1), pp. 95-104.
Schizophrenia care studies have revealed gaps between actual and recommended care practices in the dosing of antipsychotic medications, use of depot medications, and the provision of effective psychosocial interventions. This report reviews the status of 42 process-of-care measures for schizophrenia care, which these researchers identified as part of a national inventory of mental health quality measures. A greater proportion of measures assessed pharmacotherapy than assessed psychosocial interventions or other clinical processes, such as assessment, continuity of care, or coordination of care. Twenty-five measures (60 percent) were based on research evidence linking measure conformance with improved patient outcomes, while 17 (40 percent) were based on clinical consensus or opinion. Only 12 measures (29 percent) were fully operationalized. Few were tested for validity or reliability. Several demonstration projects currently under way should expand the pool of well-developed and tested measures.
Newgard, C.D., and Lewis, R.J. (2002). "Statistical methods for prehospital care research." (AHRQ National Research Service fellowship F32 HS00148). Prehospital Emergency Care 6(2), pp. S9- S17.
These authors explore concepts underlying the statistical design and analysis of clinical research in the out-of-hospital setting. They describe type I error, type II error and power, power analysis and sample size determination, and several common parametric and nonparametric statistical tests. Parametric tests are used to analyze numerical data and require that the data follow a normal distribution and that variances are equal for data from all groups. Nonparametric tests can be used for normally or non-normally distributed data and provide a more robust estimate of the p value, that is, an estimate less affected by the underlying distribution of the data than a parametric test. They also describe the use of confidence intervals, for example, to determine whether a new dispatch system results in less total out-of-hospital run time; multiple comparisons of different groups; interim data analysis, subgroup analysis, intention-to-treat analysis, and multivariate analyses; and the use of statistical consultants.
Return to Contents
Current as of September 2002
AHRQ Publication No. 02-0043
Internet Citation:
Research Activities newsletter. September 2002, No. 265. AHRQ Publication No. 02-0043. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/research/sep02/
540 Gaither Road Rockville, MD 20850