Archives of Physical Medicine and Rehabilitation
Volume 85, Issue 10 , Pages 1599-1601, October 2004

Significance of poor patient participation in physical and occupational therapy for functional outcome and length of stay

Presented in part to the American Academy of Physical Medicine and Rehabilitation, November 2002, Orlando, FL.

  • Eric J. Lenze, MD

      Affiliations

    • Intervention Research Center in Late-Life Mood Disorders, Department of Psychiatry, University of Pittsburgh School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA
    • Department of Physical Medicine and Rehabilitation, University of Pittsburgh School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA
    • Department of Occupational Therapy, University of Pittsburgh, Pittsburgh, PA, USA
    • Corresponding Author InformationReprint requests to Eric Lenze, MD, Western Psychiatric Institute and Clinic, Rm E1124, 3811 O’Hara St, Pittsburgh, PA 15213
  • ,
  • Michael C. Munin, MD

      Affiliations

    • Department of Physical Medicine and Rehabilitation, University of Pittsburgh School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA
  • ,
  • Tanya Quear, BS

      Affiliations

    • Intervention Research Center in Late-Life Mood Disorders, Department of Psychiatry, University of Pittsburgh School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA
  • ,
  • Mary Amanda Dew, PhD

      Affiliations

    • Intervention Research Center in Late-Life Mood Disorders, Department of Psychiatry, University of Pittsburgh School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA
  • ,
  • Joan C. Rogers, PhD

      Affiliations

    • Department of Occupational Therapy, University of Pittsburgh, Pittsburgh, PA, USA
  • ,
  • Amy E. Begley, MA

      Affiliations

    • Intervention Research Center in Late-Life Mood Disorders, Department of Psychiatry, University of Pittsburgh School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA
  • ,
  • Charles F. Reynolds III, MD

      Affiliations

    • Intervention Research Center in Late-Life Mood Disorders, Department of Psychiatry, University of Pittsburgh School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA

Article Outline

Abstract 

Lenze EJ, Munin MC, Quear T, Dew MA, Rogers JC, Begley AE, Reynolds CF III. Significance of poor patient participation in physical and occupational therapy for functional outcome and length of stay. Arch Phys Med Rehabil 2004;85:1599–601.

Objectives

To determine the frequency of poor patient participation during inpatient physical (PT) and occupational therapy (OT) sessions and to examine the influence of poor participation on functional outcome and length of stay (LOS).

Design

Prospective observational study.

Setting

University-based, freestanding acute rehabilitation hospital.

Participants

Two hundred forty-two inpatients, primarily elderly (age range, 20–96y), with a variety of impairment diagnoses (eg, stroke), who were admitted for inpatient rehabilitation.

Interventions

Not applicable.

Main outcome measures

The Pittsburgh Rehabilitation Participation Scale, the 13 motor items from the FIM instrument (FIM motor), LOS, and discharge disposition.

Results

We categorized the sample into 3 groups: “good” participators were those for whom all inpatient PT and OT sessions were rated 4 or greater (n=139), “occasional poor” participators were those with less than 25% of scores rated below 4 (n=53), and “frequent poor” participators were those with 25% or more of scores rated below 4 (n=50). Change in FIM motor scores during the inpatient rehabilitation stay was significantly better for good and occasional poor participators, compared with frequent poor participators (mean FIM improvement: 23.2, 22.8, and 17.6, respectively; repeated-measures analysis of variance group by time interaction, P<.002). LOS was significantly longer for occasional poor participators, compared with good and frequent poor participators controlling for admission FIM differences (adjusted means: 13.9d, 11.0d, and 10.9d, respectively; analysis of covariance, P<.001).

Conclusions

Poor participation in therapy is common during inpatient rehabilitation and has important clinical implications, in terms of lower improvement in FIM scores and longer LOS. These results suggest that poor inpatient rehabilitation participation and its antecedents deserve further attention.

Key words:  Motivation , Patient participation , Rehabilitation , Treatment outcome

 

INPATIENT REHABILITATION INVOLVES high-intensity therapy efforts, typically 3 hours a day of physical (PT) and occupational therapy (OT). It has been argued that such intensive therapy may lead to improved functional outcomes, as compared with lower-intensity therapy delivered in nursing homes,1, 2 although other studies have found that rehabilitation setting does not affect outcome.3, 4

Although such studies typically describe intensity of therapy in terms of total hours per day, the quality of interaction between patient and therapist may also play a role in rehabilitation intensity and effectiveness. However, no studies have measured the degree to which patient participation influences functional outcomes during rehabilitation training nor is it known whether sporadic poor participation during the rehabilitation stay would negatively affect outcomes, or whether only pervasively low participation is clinically meaningful.

Our group developed the Pittsburgh Rehabilitation Participation Scale (PRPS), the first published rating of patient participation.5 This scale measures participation by patients in their PT and OT. In the present study, we sought to show the frequency of poor participation in a typical inpatient rehabilitation facility, as well as the clinical significance of poor participation in terms of overall functional improvement, length of stay (LOS), and discharge disposition. Our overarching aim was to stimulate more research assessing the quality of therapy during inpatient rehabilitation, and ultimately to design interventions that improve the quality and intensity of acute rehabilitation.

Back to Article Outline

Methods 

We studied a sample of 242 patients admitted between November 2001 and February 2002 to a university-associated, freestanding rehabilitation hospital. The study was approved by the university’s institutional review board. Subjects were consecutive admissions to the orthopedic and neurologic units. All admissions during the enrollment were included (ie, there were no refusals or deaths during this period).

Subjects had 3 hours of rehabilitation therapies per day, typically a 1-hour session of OT and two 1-hour sessions of PT. Although a few subjects also received speech-language therapy, these therapy sessions were not assessed in this study. Therapists were instructed to rate a patient’s participation after each therapy session by using the PRPS, a 6-point Likert-type measure of therapist observation of patient participation in their therapy session. The PRPS, including reliability and validity characteristics, is described in detail elsewhere.5

For this analysis, we categorized each subject’s PRPS score for each therapy session as either “good” (score of 4–6, indicating good, very good, or excellent participation) or “poor” (score of 1–3, indicating fair, poor, or no participation). Our rationale for this dichotomization was that the PRPS is constructed to distinguish between good and poor participation at this cutoff. To achieve a score of 4 (good) for a therapy session, a patient must participate in all exercises or activities with good effort and finish most exercises, whereas a score of 3 (fair) is given to those who do not show maximal effort, do not finish most exercises, and/or require much encouragement. Subjects were then characterized based on the frequency of PRPS scores in the 1 to 3 range. We determined cutoffs for the groups to be 5% or less (good participators), 5% to 25% (occasional poor), and 25% or more (frequently poor). These cutoffs were based on the visual examination of the distribution and natural breaks in the data.

We used the FIM instrument,6 a validated measure for clinician-rated assessment of disability in the inpatient rehabilitation setting, to rate patients’ admission and discharge functional status. Only the 13 motor items were used to calculate the FIM motor score. Covariates included sociodemographics (age, gender, ethnicity), primary diagnosis (eg, stroke, hip fracture), LOS, and medical comorbidity count measured as the number of current, comorbid medical conditions (excluding psychiatric diagnoses) that were documented in the patient’s chart. Impairment group was categorized as 1 of the following 8 groups: hip and knee replacement (n=56), hip fracture or multifracture including hip (n=22), other fracture or multitrauma (n=15), ischemic or hemorrhagic cerebrovascular accident (CVA) (n=38), debility (n=73), other neurologic conditions (n=22), amputee (n=9), and other (n=7).

The main analytic strategy was a comparison of the good, occasional poor, and frequent poor groups in terms of change in FIM motor score and LOS. We controlled for admission FIM (as the only baseline variable that differed between the participation groups) by using a repeated-measures analysis of variance (ANOVA) for FIM and analysis of covariance (ANCOVA) for LOS. LOS was transformed (natural log transformation) because of nonnormal distribution. The Fisher exact test was used to compare groups at discharge disposition; data were dichotomized as home versus institutional setting (those discharged to other settings were removed from the analysis).

Back to Article Outline

Results 

In the overall sample of 242 subjects, we found that 57% (n=139) were good participators (ie, <5% of therapy sessions rated ≤3 on the PRPS), whereas 22% of subjects (n=53) were occasional poor participators (5%–25% of sessions rated ≤3), and 21% of subjects (n=50) were frequent poor participators (≥25% of sessions rated ≤3). In the good group, all subjects had a good participation score in all therapy sessions. In the occasional poor group, we found decreased participation scores in an average ± standard deviation (SD) of 14.1%±4.8% (range, 6.3%–23.1%) of sessions, whereas in the frequent poor group, lower scores were seen in an average of 49.5%±22.5% (range, 25%–100%) of sessions. The mean participation score throughout the inpatient stay was 5.1±0.5 for the good participator group, 4.7±0.4 for the occasional poor group, and 3.7±0.6 for the frequent poor group (ANOVA: F2,239=164.57, P<.001, Tukey post hoc significant differences: good> occasional poor>frequent poor).

Table 1 compares the good, occasional poor, and frequent poor rehabilitation participator groups by baseline sociodemographic and clinical characteristics. This table shows that the only significant difference among the 3 groups was admission motor FIM score (lower in the frequent poor participator group). Table 2 displays the rates of good, occasional poor, and frequent poor rehabilitation participator groups by impairment diagnosis, showing no significant differences in impairment diagnosis among the 3 groups.

Table 1. Comparison of Baseline Characteristics of the Good, Occasional Poor, and Frequent Poor Participator Groups
GoodOccasional PoorFrequent PoorANOVA or χ2P Value (Tukey post hoc group differences)
Mean age ± SD70.4±13.970.6±14.372.2±17.5NS
% male37% (n=52)30% (n=16)40% (n=20)NS
% white77% (n=107)81% (n=43)78% (n=39)NS
Mean no. of comorbidities ± SD (median)3.3±1.5 (4)3.2±1.5 (3)3.3±1.6 (3.5)
Mean admission FIM motor score ± SD48.3±8.547.5±8.439.9±10.6<.001 (good, occas poor>freq poor)

Abbreviations: freq, frequently; NS, not significant; occas, occasionally.

*Kruskal-Wallis test done due to distribution and count.

Table 2. Participation in Therapy by Impairment Diagnosis
ImpairmentParticipation Group
Good (n=139)Occasional Poor (n=53)Frequent Poor (n=50)
Hip/knee replacement, elective (n=56)41 (73.2%)*9 (16.1%)6 (10.7%)
Hip fracture (n=22)8 (36.4%)8 (36.4%)6 (27.3%)
Other fracture or multiple trauma (n=15)10 (66.7%)2 (13.3%)3 (20.0%)
CVA (n=38)22 (57.9%)4 (10.5%)12 (31.6%)
Amputee (n=9)7 (77.8%)1 (11.1%)1 (11.1%)
Debility (n=73)34 (46.6%)22 (30.1%)17 (23.3%)
Other neurologic diagnoses (n=22)12 (54.5%)6 (27.3%)4 (18.2%)
Other (n=7)5 (71.4%)1 (14.3%)1 (14.3%)

NOTE. The Fisher exact test showed no significant group differences in distribution of impairment diagnosis (Monte Carlo estimate of P value, .077).

* Percentages are cell percentages of row totals.

The change in FIM motor score was significantly better for good and occasional poor participator groups (mean change score: 23.2 and 22.7, respectively), compared with frequently poor participators, (mean, 18.0; repeated-measures ANOVA group by time interaction: F2,239=6.84, P=.001). Additionally, the occasional poor participator group had significantly longer LOS (adjusted mean, 13.9d) than the good and poor participator groups (11.0d and 10.9d, respectively; ANCOVA: F2,238=8.29, P=.001).

In the good participator group, 94.0% (126/134) of subjects were discharged home, rather than to an institution (skilled nursing facility or personal care home). In the occasional participator group, 85.7% (42/49) of subjects were discharged home, whereas in the frequent poor participator group, 84.1% (37/44) were discharged home (Fisher exact test, P=.056). There were also no group differences in a logistic regression controlling for admission FIM motor score (data not shown).

Back to Article Outline

Discussion 

This study is the first to examine the extent and clinical effects of poor patient participation in PT and OT during inpatient rehabilitation. By using the PRPS, we found that subjects who occasionally participated in the fair to poor range for therapy sessions (≈20% of the sample) had the same overall FIM improvement as those who participated well but took an additional 3 days of inpatient rehabilitation on average to attain that same improvement. Those who frequently participated in the fair to poor range (≈20% of the sample) had about 25% less improvement, as measured by the FIM. Poor participation was also associated with a greater likelihood of discharge to an institution, rather than home, although this association was not significant when controlling for baseline functional status.

These results suggest that it may be advantageous for clinicians to assess patients’ participation during their rehabilitation stay to maximize functional performance. We have argued previously that such efforts might include identifying and managing antecedents of poor participation or implementing efforts to improve participation regardless of the cause.5 The present study was not designed to determine antecedents or causes of poor participation. Our group7 and others8 have previously shown that patient characteristics such as depression, cognitive impairment, and poor self-efficacy predict poorer participation, and such characteristics may have led to poor participation as well as lower admission FIM scores in the present sample. Also, the patient-therapist relationship may be an important determinant of patient participation.9 Such research to identify the causes of poor rehabilitation participation could lead to intervention studies that improve participation and maximize rehabilitation intensity.10

Regarding the effects of poor participation on LOS, our findings apply mainly to the lower overall rehabilitation efficiency for people with occasional or frequent poor participation. Inpatient rehabilitation facilities are under financial pressure to decrease overall LOS, in response to the prospective payment system that has emerged, increasing the need to evaluate factors that influence rehabilitation efficiency as well as overall functional outcome. Our study suggests that many inpatients have longer stays associated with poor participation; therefore, efforts to decrease LOS will require interventions to reduce poor participation, even if it occurs only sporadically. Anecdotally, our group has noted that clinicians in rehabilitation typically discount such sporadic participation as normal or of minimal concern; our findings suggest that, at least in terms of inpatient rehabilitation efficiency, any amount of poor participation may be of concern.

Limitations 

Our findings may be somewhat limited in their generalizability because our study was done in 1 rehabilitation hospital and our sample did not include patients with spinal cord injury or traumatic brain injury, in whom poor participation may have different clinical implications. It is also possible that therapists’ observations of participation with this scale were biased and may have reflected other factors such as poor progression in therapy or poor quality of patient-therapist relationship.

Back to Article Outline

Conclusions 

We found that poor participation in inpatient therapy sessions was common and was associated with longer inpatient rehabilitation stay; lower likelihood of discharge to home; and, in those with frequent poor participation, poorer functional outcome. These findings highlight the need to improve rehabilitation participation.

Back to Article Outline

Acknowledgments 

We thank the PT and OT staff, and the medical records department, of the UPMC Rehabilitation Hospital for their time and effort participating in this study.

Back to Article Outline

References 

  1. Kane RL , Chen Q , Finch M , Blewett L , Burns R , Moskowitz M . The optimal outcomes of post-hospital care under Medicare . Health Serv Res . 2000;35:615–661
  2. Kane RL , Chen Q , Finch M , Blewett L , Burns R , Moskowitz M . Functional outcomes of posthospital care for stroke and hip fracture patients under Medicare . J Am Geriatr Soc . 1998;46:1525–1533
  3. Kramer A , Steiner JF , Schlenker RE , et al.   Outcomes and costs after hip fracture and stroke. A comparison of rehabilitation settings . JAMA . 1997;277:396–404
  4. Levi SJ . Posthospital setting, resource utilization, and self-care outcome in older women with hip fracture . Arch Phys Med Rehabil . 1997;78:973–979
  5. Lenze EJ , Munin MC , Quear T , et al.   The Pittsburgh Rehabilitation Participation Scale (reliability and validity of a clinician-rated measure of participation in acute rehabilitation) . Arch Phys Med Rehabil . 2004;85:380–384
  6. Linacre JM , Heinemann AW , Wright BD , Granger CV , Hamilton BB . The structure and stability of the functional independence measure . Arch Phys Med Rehabil . 1994;75:127–132
  7. Lenze EJ , Munin MC , Quear T , Dew MA , Rogers JC , Reynolds CF . Cognitive and affective predictors of rehabilitation outcome in patients with hip fracture . [abstract] Arch Phys Med Rehabil . 2003;84:E18
  8. Resnick B , Zimmerman SI , Magaziner J , Adelman A . Use of the Apathy Evaluation Scale as a measure of motivation in elderly people . Rehabil Nurs . 1998;23:141–147
  9. Maclean N , Pound P . A critical review of the concept of patient motivation in the literature on physical rehabilitation . Soc Sci Med . 2000;50:495–506
  10. Resnick B . Efficacy beliefs in geriatric rehabilitation . J Gerontol Nurs . 1998;24:34–44

 Supported by the National Institute of Mental Health (grant nos. K23 MH64196, P30 MH52247).No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors(s) or upon any organization with which the author(s) is/are associated.

PII: S0003-9993(04)00430-7

doi:10.1016/j.apmr.2004.03.027

Archives of Physical Medicine and Rehabilitation
Volume 85, Issue 10 , Pages 1599-1601, October 2004