Trudy Mallinson

A comparison of the separation ratio and coefficient alpha in the creation of minimum item sets.

BackgroundShort-form outcomes measures are becoming common in response to demands for increased efficiency in health care. This study examines Rasch measurement as an aid to selecting items for short form tests. The focus of this paper is on maintaining test quality while reducing items. The separation ratio (SR) aids item reduction by indicating how removing items impacts measurement precision. Results of the SR and coefficient &agr; are compared. ObjectivesTo demonstrate the use of Rasch measurement to shorten clinical outcomes measures and to compare the separation ratio and coefficient &agr; in evaluating when item reduction improved efficiency without sacrificing measurement precision. Research DesignRetrospective analysis of existing health outcomes data. SubjectsA convenience sample of 58 patients receiving cataract surgery. MeasuresThe 14 items of the VF-14 (a measure of visual functioning), the published subset of items from this test (the VF-7), and 5 other 7-item combinations of the items. ResultsThe largest coefficient &agr; was obtained from the VF14 (.84) while the largest separation ratio (2.67) was obtained from the 7-item subtest with the reduced rating scale. ConclusionsThis study demonstrated one way that Rasch measurement can be helpful in selecting items for minimum item sets while maintaining test precision. Both &agr; and the separation ratio provide information about how a sample performed with a given test although variations in measurement precision may not always be detected with &agr;.

Impact of Medicare's Prospective Payment System for Inpatient Rehabilitation Facilities on Stroke Patient Outcomes

Dobrez D, Heinemann AW, Deutsch A, Manheim L, Mallinson T: Impact of Medicares prospective payment system for inpatient rehabilitation facilities on stroke patient outcomes. Objective:To estimate the effect of Medicares prospective payment system for inpatient rehabilitation facilities on discharge functional status, community discharge, and length of stay. Design:Secondary analysis using data drawn from the American Medical Rehabilitation Providers Association subscription database. Eligible patients were Medicare and non-Medicare stroke patients discharged from inpatient rehabilitation facilities from 1998 through the first two quarters of 2006. Random effects panel data models were used to estimate the impact of prospective payment on motor and cognitive discharge function, the probability of discharge to the community and inpatient length of stay, controlling for patient, and facility characteristics. Results:The introduction of prospective payment was associated with small, statistically significant reductions in Functional Independence Measure discharge motor (−1.10) and cognitive (−0.15) scores and in the probability of discharge to the community (adjusted odds ratio: 0.87) for Medicare fee-for-service patients. Length of stay was substantially lower for both Medicare (−1.86 days) and (−2.16) non-Medicare fee-for-service patients. Conclusions:Further research is needed to determine whether the small reductions in patient function are persistent over time. This short-term evaluation of prospective payment system suggests minimal negative impact on stroke patient function at discharge because of the change in Medicare reimbursement but a decrease in likelihood of discharge to the community.

Trends in the Supply of Inpatient Rehabilitation Facilities Services : 1996 to 2004

OBJECTIVES Describe the supply of inpatient rehabilitation facilities (IRFs) services in 1996 and examine changes between 1996 and 2004, including the impact of the IRF prospective payment system (PPS) in 2002 on organizational trends. DESIGN Retrospective pre-post design. SETTING Freestanding and subprovider (distinct-part units) IRFs. PARTICIPANTS IRFs (N=1424), including 257 freestanding IRFs and 1167 IRF units reported in the Healthcare Cost Report Information System database, from years 1996 to 2004. INTERVENTIONS Not applicable. MAIN OUTCOME MEASURES Number of IRF openings, IRF closures, beds, and inpatient days. RESULTS The number of IRFs grew from 1037 to 1183 between 1996 and 2001 and grew to 1235 between 2001 and 2004. The likelihood of IRF closures trended lower after PPS, and there was a significant increase in the likelihood of openings when PPS was introduced. For-profit, rural, and small IRFs were more likely to open over the entire period. There was a 12.9% increase in the number of total inpatient days, somewhat less than the 15.7% growth in IRF beds over the period. There was no impact of PPS on beds available but a significant decline in total inpatient days after PPS. CONCLUSIONS Inpatient days rose under the Tax Equity and Fiscal Responsibility Act and declined after 2002. Yet the likelihood of openings and closures did not appear to respond to these changes, perhaps because they were modest compared with changes in local IRF markets. The IRF PPS did little to affect service distribution in less well-served areas, although we did find growth in rural areas. Occupancy rates in 2004 were close to rates at the start of the period (70%). This observation implies that IRFs were implementing strategies to recruit a sufficient number of patients, even though bed numbers were increasing and length of stay was declining. Consequently, policy that limits the potential of IRFs to increase patient admissions, such as the limits on admissions to IRFs of patients with conditions other than those included in the 75% rule, is likely to produce substantial decreases in total inpatient days.

Confidence Levels Can Broaden the Application of Clinical Research Findings and Promote Evidence-Based Practice

The traditional method of presenting quantitative research findings based on the statistical results of null hypothesis testing can limit the ability of clinicians and consumers to translate research findings into meaningful information for treatment planning and evaluation. In this article, the authors compare results from an illustrative study based on traditional statistical methods, including a t test and p values, to a more clinically useful approach using confidence levels. The process of calculating confidence levels using the basic quantitative information included in most clinical research articles is described. The argument is made that confidence levels do not replace traditional (null) hypothesis testing. Rather, they facilitate the interpretation of statistical findings, improve clinical decision making, and support the development of evidence-based occupational therapy practice.