Finis Medicinis Pendet: Author Response to Commentary

Letter to the Editor Finis Medicinis Pendet: Author Response to Commentary We appreciate the opportunity to respond to the commentary by Black-Schaffer following the publication of our research article, “Postacute Care and Ischemic Stroke Mortality: Findings from an Integrated Health Care System in Northern California,” in the August 2011 issue of PM&R [1,2]. The study was an observational study with the usual limitations. However, we believe that the findings are not easily dismissed and highlight the need for further research in postacute care for stroke populations. We agree with Black-Schaffer that one explanation for lower mortality in the inpatient rehabilitation facility (IRF) cohort is the characteristics of that population, that is, they were younger and had fewer comorbidities. As we stated in our article: “It is likely that those admitted to an IRF were, from the outset, more likely to survive than those admitted to an SNF.” However, rather than being content with a simple explanation that finis origine pendet (ie, the end depends on the beginning), we believe that our findings pose more complex questions. We are not willing to accept a therapeutic nihilistic view. It is too reminiscent of William Osler’s statement in 1892 about stroke patients [3]: “[T]he friends should at onset be frankly told that the chances of recovery are slight.” Outcomes, including survival, are complex variables that depend not only on the patient’s medical condition or conditions but, it is hoped, also on the care provided. We suggest a Latin phrase for this argument: finis medicinis pendet (ie, the end depends on medical care). Originally, we included all the patients who had a stroke who survived acute hospital care and followed them up to 365 days. We identified the postacute care (PAC) services received during the 1-year follow-up, grouped them by the highest level of PAC services received, that is, inpatient rehabilitation hospital (IRH), skilled nursing facility (SNF), home health (HH), and outpatient therapies. We did not restrict the period when the PAC services were received. If a patient was admitted to the IRH at the end of follow-up, for example, at 300 days of follow-up, the patient was in the IRH group. We found large differences in survival among PAC service groups, even after we controlled for patient characteristics, comorbidities, and other variables under study. The Kaplan-Meier survival curves showed a steep decrease at the early stage of follow-up for patients who received SNF as the highest level of PAC treatment. We were concerned that the patients in this group might be more severe cases than those who received IRH as the highest level of PAC treatment. We then regrouped PAC services within 14 days and 61 days of acute care hospital discharge. The PAC 14-day PM&R 1934-1482/11/$36.00 Printed in U.S.A.

group provided a 2-week window to allow patients to be transferred to different PAC settings, representing the “early stage” of rehabilitation treatment. The PAC 61-day group was created on the basis of the observation that 90% of patients who were admitted to the IRH during the follow-up in the study sample were admitted to the IRH within 61 days of follow-up. We then disregarded the PAC treatment after the indexed dates of grouping. To further minimize the selection bias by severity, we excluded all the deaths that occurred within 14 days or 61 days, respectively, in our analyses for the PAC 14-day and PAC 61-day groups; the majority of these deaths occurred in SNFs. We believe the most important question posed by our study is whether we are triaging the most appropriate patients to IRHs, that is, those in need of longer hospital stays because of the severity of their stroke and/or the existence of more comorbidities that require greater physician contact and higher nurse staffing ratios. In fact, research just published suggests that the hospitalist model may be associated with higher readmission rates than older hospital staffing models [4]. The IRH model is best used to extend hospital-based care for patients who have more severe strokes and more comorbid conditions and who are at risk to be readmitted to the hospital or to die if discharged too soon to a nonhospital setting; these patients are likely the oldest patients. Comprehensive family/ caregiver training in addition to the medical, rehabilitation, and nursing care provided is important for ensuring that care at home is safe and optimal. We also respectfully disagree with Black-Schaffer (who provided no reference for the statement) that the typical patient who has had a stroke has a 2- to 4-week IRH stay and spends 4-8 weeks in an SNF, 3-4 weeks in home care, and 3-4 months in outpatient settings. Most patients in our system who had a stroke and need continued institutional care either go to an IRH or to an SNF, not both, and both IRH and SNF length of stays typically average 16-18 days. After receiving institutional care, patients often receive home health care and/or outpatient care for a variable period, depending on the severity of their condition. Patients may receive different rehabilitation treatment in terms of type, timing, duration, number, and intensity. It is a challenge to capture and group PAC services for research because of the complex and varied trajectories of care. In terms of national statistics, an estimated 795,000 new or recurrent strokes occurred in the United States in 2009 [5], and approximately 59% of patients who had a stroke were admitted to IRHs and SNFs [6]. According to the Report to the Congress on Medicare Payment Policy, there were only 35,757 IRH beds (hospital-based or free-

© 2011 by the American Academy of Physical Medicine and Rehabilitation Vol. 3, 1155-1156, December 2011

1155

1156

Sandel et al

LETTER TO THE EDITOR

standing) in the United States in 2009, and only 25% of the beds were used for stroke populations [7]. In 2006, 23.5% of patients admitted to an SNF were rehospitalized within 30 days [8]. Clinical trials that reported positive associations of IRHs and survival after stroke were conducted mainly with patients who were admitted to an IRH a short time after acute care hospital discharge. In a review paper on randomized trials of organized inpatient multidisciplinary rehabilitation and stroke outcomes, Langhorne and Duncan [9] reported that inpatient rehabilitation was associated with a reduced risk of death, death or institutionalization, and death or dependency. Further investigation of the many variables contributing to both rehospitalization and mortality in patients with stroke during the postacute care period is clearly warranted. Elizabeth Sandel, MD Michelle Camicia, MSN, CRRN Hua Wang, PhD Physical Medicine and Rehabilitation Kaiser Foundation Rehabilitation Center Vallejo, CA E.S. 8B, NINDS grant, NIH clinical center contract M.C Disclosure: nothing to disclose

H.W. 8A, Kaiser co-investigator on NIH studies DOI: 10.1016/j.pmrj.2011.10.004

REFERENCES 1. Wang H, Sandel, ME, Terdiman J, et al. Postacute care and ischemic stroke mortality: Findings from an integrated health care system in Northern California. PM R 2011;3:686-694. 2. Black-Schaffer RM. Finis origine pendet: Commentary on “Postacute care and ischemic stroke mortality: Findings from an integrated health care system in Northern California.” PM R 2011;3:695-696. 3. Osler W. Principles and Practice of Medicine. New York, NY: D Appleton and Co; 1892. 4. Kuo YF, Goodwin JS. Association of hospitalist care with medical utilization after discharge: Evidence of cost shift from a cohort study. Ann Intern Med 2011;155:152-159. 5. Lloyd-Jones D, Adams R, Carnethon M, et al. Heart disease and stroke statistics—2010 update: A report from the American Heart Association. Circulation 2010;121:e46-e215. 6. Conroy BE, DeJong G, Horn SD. Hospital-based stroke rehabilitation in the United States [review]. Top Stroke Rehabil 2009;16:34-43. 7. Medicare Payment Advisory Commission. Report to the Congress: Medicare Payment Policy. March 2011. Available at http://medpac.gov/ documents/Mar11_EntireReport.pdf. Accessed October 18, 2011. 8. Ouslander JG, Berenson RA. Reducing unnecessary hospitalizations of nursing home residents. N Engl J Med 2011;365:1165-1167. 9. Langhorne P, Duncan P. Does the organization of postacute stroke care really matter? Stroke 2001;32:268-274.