The primary goal of rehabilitation is to optimize function and promote functional recovery for patients. Outcome measures are the objective tests and measures used to determine functional status in a standard and consistent way, and can quantify the effects of rehabilitation.
Outcome measures are also a method by which one can validate the efficacy of treatments. Such measures are becoming increasingly important as providers and payors focus more on value as opposed to volume in health care.
The systematic collection of outcome measures may help a provider identify functional limitations during an initial patient evaluation, select appropriate interventions, and establish goals for future planning (e.g., discharge from inpatient care).1 Communication about the results of standardized outcomes among the interdisciplinary team establishes a common language for a comprehensive plan of care. Standardized outcomes are essential for high methodological quality of clinical trials.2 Ongoing review of outcomes at an institutional level may be used for quality improvement, comparison across settings, analysis of cost-effectiveness, or health services research.3,4
In spite of the many advantages to the use of outcome measures, there are many barriers to the practical use of outcome measures in clinical practice. Studies have shown a high degree of heterogeneity of outcome measures used in rehabilitation centers, even within the same diagnostic groups.5 This is true even in research studies; a review of outcome measures reported in randomized clinical trials (RCTs) in stroke rehabilitation found that 489 different measurement tools were used in 491 RCTs.2 In 2009, less than 50% of physical therapists surveyed reported using standardized outcome measures.6 Barriers to the use of outcome measures have been reported to include lack of time, need for special equipment, provider knowledge, and perceived value of information.6–8 Concerns have also been raised about the validity of patient- and clinician-reported outcomes and the potential need for stronger theory-referenced measurement tools.9
Outcome measures may be collected through a variety of ways, for example, self-reported answers to standardized questions, direct observation of performance, equipment recordings, or review of medical records. Outcome measures have the potential to be influenced by both the patient’s and clinician’s motivation and judgement.10 To provide useful information, all measures should have adequate reliability and validity documented with a sufficient sample size for a specific population.11
Reliability of a measure indicates the degree of consistency with repeated measurements, with as little error or variation as possible.12 All measurements have potential for error, from the individual taking the measurement to the measurement tool itself. The degree of variability of the characteristic being measured can also be a source of variation. For example, the reliability of pain measures may be difficult to determine because pain can fluctuate between measurement sessions due to many factors. It is important to interpret research studies of measurement reliability in the context of the experience of the rater, the population tested, and the inherent variability of the characteristic being tested. Reliability coefficients (e.g., interclass correlation coefficients, or ICC) above a level of 0.75 are generally considered to have adequate reliability for clinical measures.12
The validity of a measure indicates the usefulness of the measure for what it is intended. It is relatively easy to establish validity for physiological tests such as body temperature and muscle force using a thermometer and dynamometer, respectively. However, it is more difficult to establish validity for constructs such as fatigue, fall risk, perception, and intelligence. Validation of measures may include a content validity assessment of a collection of questions into a score, or criterion-related validity of a measure in comparison to an established gold standard.12 It is important to establish content validity of clinician-reported outcomes for drug development studies, in particular as required by the U.S. Food and Drug Administration (FDA) and recently highlighted by an International Society for Pharmacoeconomics and Outcomes Research (ISPOR) task force.10
In addition to reliability and validity, practitioners should consider the clinical utility of a measure. This indicates the anticipated usefulness in describing change in a patient’s condition over time, the time and cost to administer the test, and the ability of a patient to tolerate the test.13,14
The World Health Organization’s (WHO) International Classification of Functioning, Disability, and Health (ICF) model provides a multidisciplinary classification of health and health-related domains.15 ICF was endorsed by all 191 WHO member organizations in 2001 as an international standard. This model may be used as a conceptual framework to categorize different types of problems as illustrated in Fig. 9–1 for a patient with a recent hip fracture.
Figure 9–1
The ICF model to describe functional problems in a patient with osteoarthritis and new hip fracture. (Reproduced with permission from Brown CJ, Clark D. Rehabilitation. In: Halter JB, Ouslander JG, Studenski S, High KP, Asthana S, Supiano MA, Ritchie C, eds. Hazzard’s Geriatric Medicine and Gerontology, 7e New York, NY: McGraw-Hill; 2017.)
The category of Body Functions and Structures includes the following problems that relate directly to rehabilitation: mental function; sensory function and pain; hearing and vestibular function; voice and speech; and the neuromusculoskeletal, cardiovascular, and respiratory systems. The category of Activities includes learning, communication, mobility, and self-care. The category of Participation includes domestic life, relationships, employment, community life, recreation, and leisure.
Outcome measures may be indicated to support clinical decision making for this patient at the level of Body Function and Structures (e.g., hip and knee pain); Activity (e.g., walking distance); or Participation (e.g., quality of life). This framework was applied in a similar way in a series of articles related to stroke rehabilitation outcome measures.14,16,17
It is beyond the scope of this chapter to present a comprehensive review of all outcome measures that may be used in rehabilitation. Several published articles have provided descriptions of outcome measures and recommendations based on known psychometric properties and clinical utility for patients with stroke,7,14,16,17 multiple sclerosis,18 and traumatic brain injury.19 Table 9–1 presents several common outcome measures in rehabilitation, organized by category using the ICF model. Additional detailed information is available in searchable online resources that provide updated information on reliability and validity, instructions, list of citations, and often links to the measures themselves. Some examples of these resources are listed in Table 9–2.
Quality-of-Life Measures | |
Reintegration to Normal Living Index | Assessment of reintegration into social activities |
Medical Outcomes Study Short Form 36 | Generic measure of health-related quality of life (patient-reported) |
Community Integration Questionnaire | Assessment of community interactions/social roles |
Multiple Errands Test (MET) | Evaluates function through the performance of everyday tasks |
Sickness Impact Profile | Global assessment of dysfunction and quality of life secondary to an illness, measures efficacy of rehabilitative treatments |
Neurorehabilitation | |
NIH Stroke Scale | Measurement of symptoms associated with stroke deficits |
Rancho Los Amigos | Assessment of the continuum of cognition and behavior changes in brain injury recovery |
Mini Mental | Screen for dementia/cognitive decline |
Short Orientation-Memory-Concentration | Cognitive impairment measurement (elderly/dementia) |
Brief Cognitive Assessment Tool | Global cognitive assessment in elderly (orientation, verbal recall, visual recognition and processing, recall, executive function) |
Glasgow Coma Scale | Level of consciousness assessment after brain injury |
Agitated Behavior Scale | Agitation assessment in brain injury recovery |
Fugl-Meyer Assessment | Assessment of motor impairment after stroke |
Stroke Impact Scale | General assessment quality of life in stroke patients |
Galveston Orientation Amnesia Test | Measurement of attention/orientation, indication of posttraumatic amnesia in brain injury patients |
Modified Ashworth Scale | Quantitative spasticity measurement |
Coma Recovery Scale | Assessment prognosis of consciousness |
Mayo Portland Adaptability Inventory | General rehabilitative assessment of brain injury patients helpful in prognosis and planning |
Activities of Daily Living | |
Functional Independence Measure | Global measurement for disability, determines assistance level required to perform ADLs |
Patient-Specific Functional Scale | Assessment of functional ability to finish specific ADLs, utilized in musculoskeletal/spine injuries |
Barthel Index | Assesses the ability to care for oneself/perform own ADLs (used in neurorehabilitation) |
Aerobic Assessment | |
Six-Minute Walk Test | Assessment of the distance walked over 6 minutes (determines endurance and capacity) |
Timed Up and Go | Assessment of the time to transfer to sit to stand, determines fall risk and balance |
Balance Outcome Measures | |
Functional Gait Assessment | Assessment of postural stability with ambulation (used in geriatrics/neurorehabilitation) |
Multidirectional Reach Test | Assessment of patient stability with reach in all directions (geriatric and neurorehabilitation) |
Functional Reach Test | Assessment of postural stability with ambulation by measuring the maximal forward reach while standing |
Berg Balance Test | Assessment of balance and fall risk (geriatric and neurorehabilitation) |
Pain Scales/Outcomes | |
Oswestry Disability Index | Assessment of the inability to performing ADLs in low back pain patients |
Numeric Pain Rating Scale | Assessment of subjective intensity of pain, patient self-reported |
West-Haven-Yale Multidimensional Pain Inventory | Assessment of cognitive, emotional, and behavioral factors leading to disability in patients with chronic pain |
McGill Pain Questionnaire | Assessment of both the intensity and quality of pain, patient self-reported |
Other Rehabilitation Outcome Measures | |
Beck Depression Index | Self-reported severity of depression assessment |
Braden Scale | Assessment of the likelihood of pressure ulcers developing |
WHO Disability Assessment Scale (WHODAS 2.0) | A generic assessment instrument for health and disability inclusive of cognition, mobility, vocation, and community integration |