Secondary Logo

Journal Logo

End-of-Life Care

Impact of Physical Therapy Intervention on Patient-Centered Outcomes in a Community Hospice

Mueller, Karen PhD, PT; Decker, Ilene PhD, RN

Author Information
Topics in Geriatric Rehabilitation: January/March 2011 - Volume 27 - Issue 1 - p 2-9
doi: 10.1097/TGR.0b013e3181ff6808
  • Free


At the current time, several factors have converged to suggest the need for evidence-based studies supporting the value of physical therapy (PT) intervention in a hospice setting. These include trends affecting regulatory, institutional, demographic, and epidemiological elements of hospice care.

At the regulatory level, a compelling mandate for increased PT involvement in hospice emerged from the 2008 Medicare rule change that included a revision of section 418.92, entitled “Conditions of participation-physical therapy, occupational therapy, and speech language pathology.” This revision reads as follows: Physical therapy, occupational therapy, and speech language pathology must be; (1) available, and when provided, offered in a manner consistent with accepted standards of practice; and; (2) furnished by personnel who…. are licensed in the relevant disciplines.” Whereas this mandate suggests that PT is an important component of interdisciplinary hospice care, individual programs are essentially left without guidelines to determine suitable indications for appropriate referrals.

At the institutional level, the 2008 Medicare rules change coincided with a time of rapid growth in the number of US Hospice Programs. According to the National Hospice and Palliative Care Organization's (NHPCO) 2009 Facts and Figures Report on Hospice Care in America, 1.45 million persons received hospice services in 2008, a 142% increase in just 10 years. (In 1999, the NHPCO reported that 600 000 Americans received hospice services). In addition, in 2009, the NHPCO reported that 38.5% of all deaths in the US involved the provision of hospice care.

As the number of patients receiving hospice services has increased, so has the number of programs. Accordingly, in the past 10 years, the number of programs increased by 60% from 3021 programs in 1997 to 4850 in 2008.

At the demographic level, a major impetus for the rapid growth of hospice programs is the increasing number of Americans over the age of 65. The newest members of this cohort will emerge from the nation's largest generation, known as “baby boomers.” The first members of this group, born between 1946 and 1964, will reach the age of 65 in 2011, and as successive members of these 80 million citizens join their ranks, the numbers of elderly Americans is expected to increase dramatically in the next few decades. More importantly, baby boomers, who comprise 20% of the US population, are generally more active than previous generations. Accordingly, many of these individuals have benefitted from PT intervention as part of their comprehensive health care and will likely continue to demand such treatment as they approach their end of life. Thus, as physical therapists continue to gain public recognition as the practitioner of choice for the treatment of movement disorders and the enhancement of quality life throughout the continuum of care, a recognized standard of PT care in the hospice setting is a goal of considerable professional and social relevance.

The epidemiological factor driving the emerging opportunities for PT involvement in a hospice setting relates to changing diagnostic trends in this arena of care. Although cancer has long been one of the primary diagnoses of patients receiving hospice care, the NHPCO's 2009 Edition of Facts and Figures Report on Hospice Care in America indicates that cancer now comprises 38% all diagnoses currently seen. Emerging diagnostic categories reflect an increasing prevalence of chronic diseases and conditions such as general debility (15%), dementia and heart disease (11.7% and 11.1% respectively), lung disease (7%), and stroke (4%). Other diagnoses such as amyotrophic lateral sclerosis (ALS) and other neuromotor degenerative diseases comprise less than 5%, but are increasing in prevalence each year. Because neuromuscular and musculoskeletal impairments are a common source of pain and disability in these conditions, PT intervention can play a critical role in optimizing patient comfort and function. Specifically, patient-centered “rehabilitation in reverse,” the systematic preparation of the patient and caregiver for condition-related functional decline, can assist in the maintenance of optimal function, safety and energy reserve in the presence of disease-related decline.

This growing opportunity for physical therapists to make meaningful contributions to the function and quality of life of people in the hospice setting requires a demonstration of our efficacy. At the current time, the limited research examining the impact of PT in a hospice setting suggests that patients in this setting can benefit from interventions related to improvement of function respiratory status, fatigue stress reduction, and patient/family education. Although based entirely on clinical observation of small numbers of subjects, these studies provide a valuable starting point for the continued exploration of optimal interventions and outcomes.

In all areas of PT practice, standardized outcome measures are critical tools in the advancement of evidence-based care. Indeed, much of the growing recognition of the benefits of PT intervention is derived from the appropriate selection and utilization of such measures. Despite the rapid growth of hospice programs in the United States, insurance coverage for their services is limited. In 2009, NHPCO reported that 96% of US hospices received a Medicare rate of reimbursement of $140.00 per patient/per day. This reimbursement must support the clinical services of the core personnel (nursing, social work, and medicine) along with medications, equipment, and other patient supplies. Thus, the 2008 Medicare Conditions of Participation mandating the availability of PT services provides a valuable opportunity to demonstrate the cost effectiveness through evidence-based outcome measures. Patients in the hospice setting often experience significant multisystem impairments and physical limitations related to their diagnosis. In addition, the approaching end of life presents many emotional, psychological, and spiritual issues for the patient and family. Thus, outcome measures used in the hospice setting must be sensitive to change, sensible to administer, and most importantly, reflect patient and family goals. The goals of many patients in hospice may appear insignificant to those outside of the hospice setting, but contribute immeasurably to a peaceful and meaningful life closure for the patient and family. For example, an elderly man with end stage brain cancer requested a hospice PT consult for gait training. This gentleman desired to ambulate from his bedroom to the family dining room table for Thanksgiving dinner. With PT intervention, this patient was successful in meeting this goal, much to his delight and that of his family. Once having reached that goal, this gentleman refused further PT services and experienced a peaceful death shortly afterwards.

In summary, the status of hospice and the regulatory bodies that govern this growing industry compel physical therapists to not only meet the demand for our services, but also to demonstrate our value in the hospice setting through outcome-based studies. The purpose of this study was to evaluate the impact of PT intervention on functional outcomes and quality of life of patients in a community hospice setting. The Patent-Specific Functional Scale was used as the standardized outcome measure as described in the following section.


The Patient-Specific Functional Scale (PSFS) is a generic outcome measure that quantifies functional limitations resulting from any illness or injury.13,14 The PSFS consists of an introductory statement that is read by the physical therapist or other health practitioner. Patients are then asked to identify up to 5 functional activities with which they are having difficulty performing (or are unable to perform) because of their medical problem. Each activity is then rated on an 10-point ordinal scale (from 0 being “unable to perform” to 10 being “able to perform with no difficulty”). A total score is calculated by dividing the sum of the activity scores by the number of activities identified. A minimal detectable change (at a 90% confidence interval) is either a 2-point change in the total score or a 3-point change in a single activity score. The PSFS has demonstrated excellent test-retest reliability and sensitivity to change with an intraclass correlation coefficient of R = 0.84 and a Pearson's r. = 0.78, respectively. The PSFS is shown in Appendix A.


Setting and sample

The study setting was a hospice and palliative care organization that serves a population of approximately 10000 residents within a 30-mile radius of a rural city. The patient census is approximately 25 patients with about 4 new admissions each week. Services are provided both in the patients' home or in a 10-bed residential hospice home.

The inclusion criteria required that study participants admitted to hospice service express interest in a PT consult during the initial intake assessment and be able to identify at least one goal related to physical function. Additional criteria required that participants be 21 years of age or older and able to read and understand the questionnaires. The criteria allowed for inclusion of patients who had terminal illnesses but excluded those who suffered from profound cognitive and disabling physical impairments.


This study received institutional review board approval from the Committee for Human studies research at Northern Arizona University. All subjects provided written informed consent.

Patients newly admitted to hospice received an in-home visit from a hospice nurse. This visit involved the completion of a history and medical intake, as well as an explanation of hospice services. During this initial visit, the nurse informed the patient and family about the availability and role of PT and if appropriate, explored the level of interest by the patient, caregiver, or family. The Medicare Conditions of Participation mandates that each patient the hospice serves must receive an interdisciplinary plan of care at the time of admission and periodically thereafter. In order to accomplish this, the hospice interdisciplinary team (IDT) met weekly, discussing half of the patient caseload every week (such that every patient is discussed on a biweekly basis). Thus, each new admission was presented for treatment planning upon admission and every 2 weeks thereafter.

The physical therapist and nurse investigators attended the weekly conferences for a period of 1 year (April 1, 2008-April 1, 2009) during which time, 164 consecutive new admissions were screened for inclusion in the study. Participants who met the inclusion criteria were then contacted by the hospice nurse to assess their interest in participating in the study. In addition, as patients were presented at weekly conferences by the hospice nurses, the team also identified patients whom they felt could also benefit from a PT consult (ie, mobility issues that may pose safety concerns for the family, or positioning/comfort issues).

Patients who expressed interest in study participation were contacted by either the physical therapist or nurse investigators to set up a visit. During the visit, the project was explained to the patient and their caregivers, and informed consent was obtained. Depending on the needs of the patient, family, and caregiver, the PT initial evaluation was also conducted during the same visit. If the patient desired that the PT assessment occur at a separate time, a subsequent visit was arranged, optimally within the following 24 to 48 hours.

Either during the initial contact or shortly thereafter the physical therapist investigator conducted an initial home visit on patients who had agreed to participate in the study. A PT assessment was performed and included general strength, ROM, mobility, vital signs (02 saturation, HR, BP, and Respiratory Rate). Pain and fatigue was measured on a 1 to 10 visual analogue scale. Following the assessment, the therapist explained the PSFS to the patient and caregivers, and the PSFS was completed. Because caregivers are often involved in providing physical assistance to the patient, their input was considered in the goal identification process. In cases where there was lack of agreement between the patient and caregiver on the selection of goals, the disagreement was noted on the PSFS form. Any activity that the physical therapist considered unsafe for the patient or caregiver was addressed during PT treatment sessions and was not included in the patient and family home exercise plan. Physical therapy interventions were also provided as appropriate for mobility, energy conservation, equipment usage, etc.

At least 1 PT follow-up visit was conducted for each patient within 72 hours from initial evaluation to assess functional changes on the PSFS. Other interventions were provided as needed, including nonpharmacologic pain control measures, continued mobility training and family education as needed. If a patient did not desire a follow-up visit, the physical therapist contacted the patient or caregiver by telephone within 1 week of initial visit to assess changes in function, pain, and fatigue. The need for further follow-up visits was determined by the patient, caregivers, and physical therapists. Regular nursing follow-up visits occurred per protocol through the duration of the patient's hospice admission.

Demographic data was obtained from the patient's medical record and included age, gender, race, marital status, religion, primary caregiver, diagnosis, and length of stay.


Demographic and numeric data were compiled electronically using a Windows Excel file. Frequencies, means and ranges were tabulated for ratio data (age) of the subject pool. Other demographic data was numerically coded for tabulation of ranges, means, and frequencies.

A minimal detectable change on the PSFS (at a 90% confidence interval) is either a 2-point change in the total score or a 3-point change in a single activity score. The PSFS has demonstrated excellent test-retest reliability and sensitivity to change with an intraclass correlation coefficient of R = 0.84 and a Pearson's r. = 0.78, respectively. t tests were used to group pre- and post-PSFS and pain scores at the .05 significance level.


Response rates for participation in the study were compiled for the 164 patients that were screened. Of the 164 patients screened, 112 (68%) did not meet the inclusion criteria and were excluded from the study. Table 1 shows the reasons for exclusion.

Reasons for Exclusion From the Study

Fifty-two subjects of the initial 164 (32%) met the inclusion criteria. Of the 52 patients who met the inclusion criteria, 47 patients (90%) were referred for PT services.

Of the 47 patients who received a PT referral, 21 (44%) were lost to the study before the PT initial assessment. Six patients died before the initial PT evaluation, 10 decided not to pursue PT intervention (despite expressing initial interest to their nurse case manager during the intake assessment), 3 patients had become too ill to participate in a PT assessment upon scheduling, and 2 could not be reached for scheduling in a timely manner.

26 patients (55% of the included subjects and 15.8% of the entire cohort) received a PT initial consult and intervention. Two of these 26 patients (7%) were excluded from the study at the initial visit due to unreliable cognitive status. Twelve additional patients (46%) were lost to the study before the PT discharge due to death (7 or 27%) or being too ill for a visit (5 or 19%). Twelve patients (46% of the included subjects and 7% of the entire cohort) were able to complete both the pre- and post-PT PSFS.

Table 2 shows the demographic characteristics of patients who were in the initial subject pool along with those who received PT referrals and those who completed both pre- and post-PT PSFS assessments. Table 3 illustrates the diagnostic categories of these groups.

Demographic Profiles
Diagnostic Categories

Problem categories

The 12 patients who completed both pre- and post-PSFS measures were seen for an average of 2.1 visits (range = 1-5), each of which lasted 60 to 80 minutes. During the first visit, the PSFS was reviewed with the patient, and he/she was asked to identify problem areas pertaining to impaired function or decreased comfort. Five problem areas were identified by the subjects, the most frequent being decreased mobility (n = 7, 58%) and pain (n = 5, 38%). The other 3 areas included edema (n = 4, 33%), dyspnea (n = 2, 16%), and anxiety (n = 1, 8%). On average, these patients identified 1.3 problems on the PSFS.


Interventions for each patient were provided to address their goals related to comfort and function. Nine different types of intervention were provided to the 12 patients. The 2 most frequent of these were mobility training and family education. Table 4 illustrates the types and distribution of the interventions provided.

Interventions Provided

Functional outcomes

Seven of the 12 patients-identified functional concerns on the PSFS. Of these, 6 made a clinically detectable change in at least 1 area of functional concern. The mean initial PSFS score for these 7 patients was 2.67 points (SD = 2.93), and the mean discharge score was 5.61 points (SD = 2.91). As a group, the mean change in PSFS was 2.94 points (SD = 2.55). The difference between the overall group means was significant at the 0.05 level (P = 0.0038). Table 5 shows the functional areas identified and the PSFS scores.

Initial and Discharge PSFS Scores

In addition to functional concerns, 5 patients identified pain as a major problem. This was quantified using a 0 to 10 visual analogue scale, where 0 corresponded to no pain, and 10 corresponded to the worst possible pain. The subjects' average pain score before PT intervention was 8.8 (SD = 1.3), and 3 (SD = 2.82) following PT intervention. A t test of the difference in group means showed a significant difference at the 0.05 level (P = .005). Table 6 shows the pre- and postintervention pain scores.

Pre- and Post-PT Pain Scores


In 2008, the NHPCO reported that approximately 35.4% of patients served by hospice died or were discharged in 7 days or less, and 12.1% died or were discharged in 180 days or more. A significant challenge in the completion of this study was the number of patients among the 47 meeting the inclusion criteria who died, either before the initial PT assessment (N = 6, 12.7%) or between the initial and discharge visits (N = 7, 14.8%). In addition, 8 patients (17%) declined to the point of exclusion during this time. This resulted in an overall loss of 44% of the potential subjects in this study.

Whereas death and significant decline are typical outcomes in the hospice setting, it was interesting to note that the time of this study (April 1, 2008 to March 31, 2009) corresponded with the nation-wide economic decline that affected access to health care in the community where this study was conducted. Specifically, the nurse case managers noted that an increasing number of patients were admitted to hospice directly from the hospital emergency department. In several instances, these patients died within 48 hours of hospice admission (1 patient died within 3 hours). Furthermore, it was reported that these patients had experienced significant symptoms for which they had not sought prior due to higher co-pays they could no longer afford. Finally, many of these patients had been under the care of physicians who withdrew from the Medicare payment system during that time period. Because the complete medical histories of such patients were not always available, it is difficult to determine they might have been eligible for hospice at an earlier point in time. Although these observations were anecdotal, it suggests the need for evidence-based support and the possible need for educating health care providers about the indications for admission to hospice.

One of the most important outcomes in hospice care is the reduction of pain to a level of 3 or less on the visual analogue scale for pain within 48 hours of admission. In 2007, the NHPCO reported that this outcome was achieved 84% of the time. Thus, when patients are denied access to hospice care until the point of imminent death, they may spend their final hours in significant pain. This outcome has serious implications for both the patient (who is left to suffer) and their loved ones (who are left with painful memories of that experience). The recent development of prognostication models for patients in hospice will prove increasing helpful by enabling appropriate symptom management for patients earlier in their course of illness.

The rapid decline and deaths of several patients admitted to the hospice where this study was conducted had a significant impact on data collection. The original design of the research methodology was for the patients who met the eligibility criteria to complete PSFS upon referral to PT and 72 hours after PT assessment and intervention. However, by the fourth month of the project, 7 patients had died within the 72-hour period, and another 5 had declined too much to participate in a PT discharge evaluation. Following a consultation with researchers in the APTA oncology section, it was agreed that when possible, both pre- and post-PSFS scores would be obtained during the initial visit in order to determine immediate change. This methodology change enabled complete data to be collected on 4 subjects who otherwise would not have survived the 72 hours between visits. This change resulted in 1 of the patients not making any detectable change in their PSFS change. The other 3 patients reported pain as their primary problem upon initial assessment and each of these reported a significant decrease in their discomfort as a result of PT intervention (in a study by Jensen et al, a decrease of 33% or 3.3 points on the visual analogue scale was determined to be a “meaningful decrease” in pain). Four of the 5 patients who reported pain as a primary problem met this level. The interventions provided for pain relief included massage, gentle passive stretching, and breathing exercises for relaxation. Although the subject pool is too small to generalize these results to the hospice population at large, it is encouraging that PT intervention may have an immediate positive impact on patient comfort. Future studies examining the impact of PT intervention on pain reduction among hospice patients will thus be valuable in providing greater support for our role in this care setting.

The functional concerns identified by patients in this study included mobility skills typically addressed by physical therapists in other care settings. These included bed mobility, transfers, gait, strengthening, standing balance, edema limiting comfort and mobility, and pain. This study identified 9 areas of primary PT intervention that included mobility training, family education, pain control, ace wrapping for edema, massage, breathing exercise to reduce dyspnea, home modification, strengthening, and stretching.

The first 2 interventions, mobility training and family education were provided to 92% and 83% of the subjects, respectively. In the training of mobility skills, an important element of family education involves emphasis on energy conservation, especially among patients with congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD), and end stage cancers. In other words, the qualitative elements of task performance (comfort and decreased physiologic stress) are emphasized over quantitative aspects (time, distance, number of repetitions). For example, 1 of the subjects with end stage COPD (DM) was interested in increasing his ambulation endurance. However, he stated that any attempt to walk more than a few feet would cause him to become dizzy, short of breath, and anxious. Physical therapy interventions consisted of patient and family education about the importance of energy conservation in order to lessen respiratory and physical discomfort. The patient was then provided with a front-wheeled walker for stability and support. In addition, the patient and his adult daughter were instructed in the use of pursed lip breathing emphasizing longer periods of inhalation and exhalation. The patient was provided visual feedback for successful exhalation by timing how long he could blow on a tissue held an inch from his lips. Furthermore, the patient and his daughter were instructed to use slow and regular breathing to reduce the anxiety that accompanied his episodes of SOB. Once this patient understood the importance of controlled breathing, he was able to ambulate throughout his home. He and his family reported less anxiety and feeling of instability during ambulation. Another important approach in the hospice setting, both to mobility training is the concept of “rehabilitation in reverse” that involves preparing the patient and family for the inevitable decline that accompanies end stage disease. In the case of DM, this involved maintaining the emphasis on breathing control to limit his episodes of anxiety during all ADL, even when ambulation was no longer possible.

Edema management was provided to 38% of the subjects. In each of these instances, the patients had end stage abdominal cancers that resulted in significant bilaterally lower quarter edema and a sense of heaviness and immobility that limited function. In each case, the extent of the edema (pre- and posttreatment) was measured by circumferential measurements at the base of the fifth metatarsal, above the malleoli, and at mid-calf (7 inches below the inferior patella). Light tactile sweeping motions were applied to the affected limbs, followed by circumferential ace wrapping. In 3 cases, this treatment provided enough relief that the patients were able to walk with greater comfort. Family education was provided to maintain continuity of treatment as desired.

Perhaps 1 of the most interesting outcomes arose through comments by both patients and family members that PT intervention improved sleep quality (longer periods of sleeping time, less night time awakenings), reduced family stress, increased overall relaxation, and provided emotional support. The finding of improved sleep quality may be worthy of further exploration, both as a valuable patient outcome as well as a potentially cost saving one (reduction in medications for pain and sedation).

The limitations of this study were largely related to the small final pool (n = 12) who were able to complete both pre- and post-PT PSFS measurements. However, the data also indicates that 47/164 patients (28.6%) were identified by the hospice IDT as being appropriate for PT intervention. In addition, 26 (15.8%) of these received an initial PT consult and intervention. Finally the data suggests that PT intervention results in meaningful outcomes in patient-identified areas of concern. Accordingly 11 of 12 subjects experienced minimally detectable changes on the PSFS and 4/5 patients met the criteria for a clinically meaningful reduction in pain.

This study also suggests that regular participation at hospice IDT meetings by physical therapists can assist the identification of patients who could benefit from PT intervention. During the time of this study, the rate of PT referrals increased from 10% to 28%, suggesting an increased team awareness of the benefits of PT intervention. The provision of education about the role of PT intervention to hospice clinical and volunteer staff can also help to increase awareness of our contributions. The results of this study suggest that PT intervention improves patient outcomes, reduces family stress, and reduces pain. Patients also anecdotally reported better sleep quality. The outcomes related to pain and sleep may prove to be cost effective in terms of reducing the need for sedative medication and warrant further exploration with larger subject pools.


The results of this study provide promising evidence for the benefit of PT intervention in improving patient-centered outcomes for patients in hospice. By serving as an active member of the hospice IDT, physical therapists can assist in the identification of appropriate patients for intervention, that may provide cost effective alternatives to medication while improving desired outcomes. This study was limited by a subject pool whose outcomes may not be applicable to the hospice population at large. However, the use of currently emerging prognostic models for determining appropriate candidates for hospice may help assure that patients obtain services before the point of imminent death.


1. Department of Health and Human Services: Centers for Medicare and Medicaid Services. CMS-1385-FC. Revisions to payment policies under the physician fee schedule. Accessed September 5, 2010.
    2. National Hospice and Palliative Care Organization. 009 Edition: NHPCO Facts and Figures: Hospice Care in America. Accessed September 6, 2010.
      3. National Hospice and Palliative Care Organization: Facts and Figures on Hospice Care in America. NHPCO, 1999. Accessed September 6, 2010.
        4. Physical Therapy can help relieve Boomer's back pain. NY February 1, 2007. Accessed September 6, 2010.
          5. Briggs R. Models for physical therapy practice in palliative medicine. Rehabil Oncol. 2000;18(2):18–19.
          6. Mackey KM, Starling JW. Experiences with older women with cancer receiving hospice care: significance for physical therapy. Phys Ther. 2000;80(5):459–469.
          7. Syrett E, Taylor J. Non-pharmacological management of breathlessness: a collaborative nurse-physiotherapist approach. Int J Palliat Nurs. 2003;9(4):150–156.
          8. Donnelly CM. Physiotherapy management of cancer-related fatigue: a survey of UK current practice. Supportive care in cancer. Palliat Suppor Care. 2005;3:4:68–71.
            9. Marcant D, Rapin CH. Role of physiotherapist in palliative care. J Pain Symptom Manag. 1993;8(2):68.
            10. Toot J. Physical therapy and hospice: concept and practice. Phys Ther. 1984;64(5):665.
            11. National Hospice and Palliative Care Organization. Medicare Hospice Benefit. http:// Accessed September 6, 2010.
              12. Reis E. A special place: physical therapy in hospice and palliative care. PT Magazine. 2007;15(3):42.
                13. Physical Therapy Resources: The Patient Specific Functional Scale. Government of Western Australia. Transport Accident Commission. Accessed September 6, 2010.
                14. Chatman AB, Hyams SP, Neel JM, et al. The patient-specific functional scale: measurement properties in patients with knee dysfunction. Phys Ther. 1997;77(8):820.
                15. Stratford P, Gill C, Westaway M, Binkley J. Assessing disability and change on individual patients: a report of a patient specific measure. Physiother Can. 1995;47(4):258.
                16. Jensen MP, Chen C, Brugger AM. Integration of visual analogue scale rating and change scores: a reanalysis. J Pain. 2003;(4)7:407–414.

                  functional outcomes; hospice and palliative care; physical therapy

                  © 2011 Lippincott Williams & Wilkins, Inc.