Kramer AM, Steiner JF, Schlenker RE, Eilertsen TB, Hrincevich CA, Tropea DA, Ahmad LA, Eckhoff DG. Outcomes and costs after hip fracture and stroke: a comparison of rehabilitation settings.

Kramer AM, Steiner JF, Schlenker RE, Eilertsen TB, Hrincevich CA, Tropea DA, Ahmad LA, Eckhoff DG. Outcomes and costs after hip fracture and stroke: a comparison of rehabilitation settings. JAMA. 277(5):396-404, 1997 Feb 5

This article assessed outcomes and costs for patients with hip fractures or strokes who were admitted to rehabilitation facilities, subacute nursing homes (a term used synonymously with "subacute SNFs" by the authors), and traditional nursing homes (synonymous with "traditional SNFs"); therefore, I reviewed only the results from this study pertaining to patients with strokes Complete data were available for 366 of the original 485 patients in this study. In the results, the authors reported data for 450 subjects, including those with incomplete data; however, they did not account for a missing 35 subjects. The unadjusted results suggest that patients who received rehabilitation services in an acute rehabilitation facility (n=271) were more likely to be residing in the community 6 months after admission than patients who received rehabilitation services in a traditional SNF (n=71; odds ratio [OR]=3.3, 95% confidence interval [CI]=1.5-7.2). I continued to be primarily interested in the results pertaining to subacute care, because it is a viable alternative in our health care system. Including patients with incomplete data, those who were admitted to subacute SNFs (n=108) also were more likely to be residing in the community (OR=6.8, 95% CI=2.2-21.4) than those admitted to traditional SNFs. Patients in either type of SNF who had had a stroke had significantly more ADL difficulties at 6 months than patients who were admitted to an acute rehabilitation facility. Unadjusted results showed that patients admitted to subacute SNFs had significantly greater recovery of function for transferring, walking 6 m (20 ft), and toileting than patients admitted to traditional SNFs. After adjusting for age, Barthel Index score, hemiplegia, and depression ([R.sup.2]=.41), differences remained between acute rehabilitation facilities and SNFs, but did not exist between subacute and traditional SNFs.

A limitation of this study was the use of self-reports for an ADL score, which may not represent actual changes in performance of ADLs. A strength of this study was the comparison of patients who declined to participate in the study to the actual study sample. The authors found no significant differences between groups, except that patients with stroke who chose not to participate were more likely to have psychiatric diagnoses. The authors also compared data from subjects who did not complete the study to data from the remaining sample and found no significant difference between groups. I agreed with their conclusion that their data suggested better function as well as a higher likelihood of return to community living for patients with stroke 6 months after admission to an acute rehabilitation facility. Based on the similarity of my patient to the study sample (again he met the inclusion criteria of having an ICD-9-CM code for a diagnosis of stroke, he was covered by Medicare, he was greater than 65 years of age, he was in an acute care hospital from which he would be transferred within 30 days, and he did not have previous SNF or rehabilitation admission for stroke), I felt confident generalizing these findings to his case.

* Clinical decision: I encountered initial difficulty determining the best approach to this literature search. My low success using keywords prompted me to change direction and use subject headings. In the process, I gained insight into useful tools (eg, mapping terms to subject headings) for more efficient and effective searching of the literature. In this situation, however, my search resulted in a small number of relevant citations, which could be the result of a lack of published research (as noted by the authors themselves) comparing the different levels of care. My search did not yield a systematic review or a meta-analysis that could have summarized multiple studies. I also chose to use CINAHL as my database, which might have limited my search results. I decided to use CINAHL because it contained more journals specific to allied health professions than MEDLINE. Because I was not researching a particular physical therapy intervention, I may have found additional articles by including another database. I also made the decision not to include articles with an emphasis on specialized stroke rehabilitation units, because the acute rehabilitation facility in question for this patient did not have a specialized stroke unit (the article by Keith et al, however, did utilize a rehabilitation hospital with a team dedicated to stroke rehabilitation). Additional research exists, such as a meta-analysis by Langhorne and Duncan (5) that shows better outcomes for patients participating in inpatient multidisciplinary rehabilitation in specialized stroke units compared with unspecialized units; however, this was outside the scope of my clinical question, given this particular patient's limited options.

The combined results of my literature search enhanced my decision regarding discharge recommendations for this patient. The literature suggested better functional outcomes at discharge for those patients who received rehabilitation in an acute rehabilitation facility compared with those who had been in subacute or SNF rehabilitation facilities. In addition, the likelihood of return to community living was highest following acute rehabilitation. Although limitations existed, no study suggested a poorer outcome after acute rehabilitation, nor did any study suggest any increase in adverse events in the acute rehabilitation setting versus the subacute setting or the SNF setting. Concluding that acute rehabilitation can promote better outcomes, I then considered the applicability of my search results to my patient.

My patient met the inclusion criteria for each of the studies, suggesting that he had similar characteristics to the patients included in these studies, which enabled me to generalize these findings to him. Based on his potential for recovery, his desire to return home, his family support, and my review of the literature on the difference in outcomes associated with different rehabilitation settings, I recommended that this patient receive rehabilitation services in an acute rehabilitation facility. I was more confident in discussing options with the patient, his family, and his case manager, because I had evidence to support my recommendation. The family accepted my recommendation, and the case manager was able to confirm admission for transfer to an acute rehabilitation facility.

References

(1) Kendall FE McCreary EK, Provance PG. Muscles: Testing and Function, With Posture and Pain. 4th ed. Baltimore, Md: Williams & Wilkins; 1993.

(2) Reding MJ, Potes E. Rehabilitation outcome following initial unilateral hemispheric stroke: life table analysis approach. Stroke. 1988;19:1354-1358.

(3) Domholdt E. Physical Therapy Research: Principles and Applications. 2nd ed. Philadelphia, Pa: WB Saunders Co; 2000.

(4) Fetters L, Figueiredo EM, Keane-Miller D, et al. Critically appraised topics. Pediatric Physical Therapy. 2004; 16:19-21.

(5) Langhorne P, Duncan E Does the organization of postacute stroke care really matter? Stroke. 2001;32:268-274.