APTA Cross Sections and Academies Recommendations for COVID-19 Core Outcome Measures.

The novel coronavirus (COVID-19) emerged as a major health concern within the United States in early 2020. Because this is a novel virus, little guidance exists for best practice to evaluate this population within the field of physical therapy.
Methods: An expert task force appointed by the leadership of 9 different academies or sections of the American Physical Therapy Association was formed to develop recommendations for a set of core outcome measures for individuals with or recovering from COVID-19.
Results: This perspective provides guidance on a best practice recommendation to physical therapists and researchers regarding the use of core outcome measures for individuals with or recovering from COVID-19. The process for the selection of core measures for this population is presented and discussed. Conclusions: Core outcome measures improve the ability to track progress and change across the continuum of care at both the patient and population levels.

The novel coronavirus disease (COVID-19) caused by the respiratory virus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), was officially designated a pandemic in the spring of 2020.1 Since the initial case, the number of confirmed individuals with COVID-19 in the United States has climbed rapidly, reaching over 29 million cases and 500 000 deaths by March 2021.2 Patients with COVID-19 present with a comprehensive disease spectrum ranging from asymptomatic infection to critical presentations, including severe hypoxic respiratory failure, shock, and multiorgan failure requiring prolonged care in an intensive care unit.3

The nature of COVID-19, with the potential for respiratory, cardiovascular, and neurologic compromise, increases the risk of short- and long-term physical disability, especially in the frail and elderly and those with comorbidities such as cancer, hypertension, diabetes mellitus, or stroke.3,4 Current literature also suggests an increase in hospitalization and death in racial and ethnic minority groups.4 However, COVID-19 has impacted all ages and all populations ranging from high functioning athletes5 to patients who are dependent and living in extended-stay facilities.6 Patients surviving COVID-19 may be at risk for cardiovascular and neurologic events due to direct insult from the virus, with evidence that the virus can cross the blood-brain barrier causing neurologic damage.7 Additionally, some individuals with COVID-19 develop a severe, systemic, inflammatory response (cytokine storm) that leads to widespread damage and long-term consequences.8 These sequelae can significantly impact patients' function leading to the need for physical therapy services. Referral to rehabilitation providers at the time of awakening, for the critically ill, and at earliest signs of movement deficits for others can mitigate secondary complications and improve patient-specific outcomes.9–11

Physical therapists in all settings use outcome measures to establish a baseline, aid in clinical decision-making, and track progress across all phases of recovery and all settings of the continuum of care.12,13 More recently, our profession has begun to recommend using core outcome measures to reduce unwarranted variation in practice and facilitate research initiatives.14 Core outcomes improve communication between clinicians. Additionally, the use of common outcome measures allows researchers to aggregate the same data of hundreds to thousands of patients to better describe patterns of recovery for patients with a particular disease, such as COVID-19. The purpose of this article is to describe the process used to establish a set of core outcome measures for individuals with or recovering from COVID-19, which serves as a guide for clinical decision-making, reduces unwarranted variation in practice, and facilitates research initiatives.

PROJECT PREMISE: APTA CROSS ACADEMY/SECTION COVID-19 CORE OUTCOME MEASURE TASK FORCE

On April 15, 2020, leaders from the APTA, including representatives from individual academies and sections, formed the “COVID-19 Response Panel” (Panel), to address the rapidly evolving nature of the novel coronavirus and function as an avenue for leaders to share and work together during the pandemic. This panel met weekly through mid-summer, then as needed through March of 2021. The panel consisted of leaders from 17 of the 18 academies and sections of the APTA. The panel served to reduce redundancies, maximize information and resources being developed, and assist with disseminating succinct and important information to APTA members during the initial months of the pandemic. During a meeting, the panel representatives from multiple sections identified and discussed the need for a set of core outcome measures to be used across settings. Members of the panel and component leaders appointed qualified members of their respective components to the Cross Section/Academies COVID-19 Core Outcome Measures Task Force (Task Force). No component was excluded from this process; however, some components chose not to assign representation to the Task Force. The 12-member Task Force included representatives appointed by APTA Acute Care, Cardiovascular and Pulmonary Section, APTA Geriatrics, Home Health Section, Academy of Neurologic Physical Therapy, APTA Oncology, Academy of Orthopaedic Physical Therapy, and the Private Practice Section. APTA Pediatrics also appointed 3 members to a separate but complementary working group to complete a parallel project for the pediatric population. The Task Force included the chair of the Parkinson Evidence Database to Guide Effectiveness (PD EDGE) Task Force, a member of the Home Health Section Outcome Measures Toolbox Team, members of the APTA Acute Care Outcome Measures Team, and experienced outcome measure researchers and speakers from APTA Acute Care, Cardiovascular and Pulmonary Section, the Academy of Neurologic Physical Therapy, APTA Oncology, and the Academy of Orthopaedic Physical Therapy.

Additionally, the Task Force was comprised of researchers, academicians, and clinicians actively working in various clinical settings. During the Task Force work, 1 member was actively engaged in data collection for research within this population, and several were actively treating patients with COVID-19 in a variety of settings. Consultation with the Core Set of Outcome Measures for Adults with Neurologic Conditions Task Force Chair was an integral part of formulating the Task Force processes.

The panel charged the Task Force to review existing literature and recommend a core set of outcome measures applicable to the greatest number of patients with or recovering from COVID-19 across the continuum of care and all clinical settings. The panel determined that a core set of outcome measures for this population would aid clinicians by improving their ability to track progress across settings, discuss clinical cases with peers, and identify appropriate measures when they are less familiar with the diagnosis. The panel also felt that physical therapy research related to COVID-19 would be expedited by using core outcome measures to allow for data pooling.

The Task Force Chair regularly attended panel meetings to provide updates and solicit feedback as appropriate from all academy/section representatives. Individual Task Force Members and panel members were also integral in soliciting further feedback from members of their respective academies/sections at key times throughout the process. The first task was to devise a process to complete the goals to identify and recommend a core set of outcome measures expeditiously, as shown in Figure 1.

FIGURE 1.

Overview of Selection Process.

IDENTIFICATION OF THE CONSTRUCTS

The next objective of the Task Force was to identify the constructs that would be of primary importance for individuals with or recovering from COVID-19. The Task Force used various sources, including published literature, World Health Organization and Centers for Disease Control and Prevention reports, information from various professional organizations, and anecdotal reports to define the breadth of the clinical presentation. The clinical presentation was defined across health care settings and by body systems, severity, type of movement dysfunction, and functional limitations of these patients. This search led to adopting the following constructs: cognition, endurance, functional mobility, health-related quality of life, strength/power, and upper extremity function. The Task Force unanimously voted to adopt these constructs. Cognition was included as a construct for multiple reasons. Based on current research, cognition is often impacted in patients with a moderate to severe presentation of COVID-19.15 Cognition is often underevaluated in similar populations.16,17 To identify the effect of cognitive changes on patients' recovery and to increase the likelihood of identifying neurologic changes, when present, cognitive testing is completed per a recommended schedule as outlined later within this document.

DEVELOPMENT OF INCLUSION AND EXCLUSION CRITERIA

The criteria for the set of core measures needed to have clinical utility for the greatest number of individuals impacted by COVID-19 across the continuum of care and functional level. Perceived and actual barriers to the use of outcomes measures were discussed from the perspectives of all participating section representatives. The Task Force considered the constraints of different practice settings, including telehealth and home health, as well as the effect of isolation precautions when selecting criteria. Measures that required little to no training were prioritized. Cut-off scores related to reliability were included at this stage, but other psychometric properties were considered at a later stage of our process. With this perspective, the inclusion and exclusion criteria were agreed upon and can be found in Table 1.

TABLE 1.

Inclusion and Exclusion Criteria

Inclusion Criteria	Exclusion Criteria
Interrater reliability >0.75 (administered by tester)	Requires a space larger than a typical hospital room
Test-retest reliability ICC >0.80 (questionnaires)	Materials are not readily available in a telehealth setting
Free to use	Materials cannot be easily cleaned for infection control
Training is free
<15 min to complete

ICC, intraclass correlation coefficient.

Due to the time-sensitive nature of the translation of our findings, the Task Force unanimously voted for a 4-stage expedited review process as follows:

Step 1. The Task Force identified documents of high levels of evidence recommended or promoted by their respective academy or section, such as clinical practice guidelines, systematic reviews, meta-analysis papers, or measures that were reviewed and recommended by their academy or section. Additionally, Task Force members completed a full review of the sralab.org/rehabilitation-measures and PTNow.org Web sites to identify outcome measures that may be appropriate for inclusion. Specific literature searches were completed in PubMed, MEDLINE, OVID, Cochran, CINAHL, and Google Scholar to identify relevant articles. Search terms include such terms as physical therapy evaluation and assessment, physiotherapy, exercise, rehabilitation, assessment, post-intensive care syndrome, heart and lung impairment and failure, venous thromboembolism disease, Guillain-Barré, stroke, sepsis, and infection.

Step 2. The Task Force reviewed the documents identified in step 1 and extracted all outcome measures that were recommended after previous rigorous review. These outcome measures were reviewed for inclusion and exclusion criteria and verified the psychometric properties; then, the outcome measures were grouped to identify unrepresented constructs. After review, we determined that all identified measures assessed constructs that are listed earlier, and no other constructs were identified.

Step 3. The Task Force reviewed the measures to add outcome measures to the list for screening based on clinical expertise. The total number of measures identified was 93. Appendix 1 provides details on all considered outcome measures.

Step 4. A public call through APTA Engage asked volunteers to identify any “crucial outcome measures” related to each construct for this population that they believe the Task Force had not yet identified and provide resources as able. The Task Force Chair announced the public call to all academies and sections via the panel. Sections and academies who chose to participate used individualized approaches to make their members aware of the public call. Fifty-two respondents from various clinical practice settings and various specialty areas identified 7 additional measures that fit within the key constructs. These additional measures were screened against the criteria, and those that met inclusion and exclusion criteria were added to the discussion for core outcome measures. The Task Force reviewed all these responses to determine whether the outcome measures addressed the constructs of interest or whether other constructs needed to be included. This process verified that the original 6 constructs covered the spectrum of movement dysfunction that professionals were noting in practice.

The Task Force recognized early that the core set of outcome measures would exhibit floor and ceiling effects in some patients dependent on the patient's functional level at the time of testing. These floor and ceiling effects would need to be addressed through additional, secondary measures. A clinical application algorithm has been developed and published on the APTA Web site, which addresses these and other limitations of the recommended core outcome measure set.18 Additional outcome measures identified through the 4-step expedited review process that did not meet established criteria were considered for inclusion in the algorithm in areas where no measure met all criteria.

SELECTION OF THE CORE OUTCOME MEASURES

Group consensus of the Task Force was reached to consider the outcome measure characteristics listed in Table 1 when determining a final core set. Additional consideration was given to measures that could capture more than one construct while providing the opportunity to evaluate constructs individually via subscales or that could be related to other measures to span the continuum of care. These recommendations are consistent with those made by other outcome measure task forces.14,19,20 For example, the Short Physical Performance Battery (SPPB) was selected because it addresses the constructs of both functional mobility and strength/power by utilizing gait speed and a variation of the 5 Times sit-to-stand test, which is recommended as a core outcome measure within the neurologic population,14 and developed by the National Institute of Aging21 to assess lower extremity functioning in older adults. Moreover, the SPPB has high clinical utility in a diverse range of clinical populations.22–26 The 2-minute step test was selected for endurance because of its utility in acutely ill and active populations and because the testing conditions, like step height, could be varied and exercise capacity and metabolic equivalent of task level performance could be calculated to illustrate progress, as functional capacity improved.20

As the Task Force compiled and sorted the outcome measures into the constructs, the Task Force members had poor agreement in the inclusion of any upper extremity functional outcome measure. The Task Force put forth additional efforts to identify an appropriate measure for this construct. We determined that the accepted constructs captured global deficits of patients and were sufficient to screen for upper extremity deficits. Further recommendations regarding upper extremity testing can be found in the algorithm.18

Finally, all identified measures that met inclusion and exclusion criteria were discussed by construct grouping to identify potential core outcome measures. Candidate measures were compared based on the previously presented criteria. Measures were removed from consideration when a consensus of the Task Force was reached that other measures within the construct category would be more appropriate as a core measure. This process was repeated for each construct until final candidate core outcome measures were identified (see Figure 2). Appendix 1 provides a brief rationale for the removal of each test. Psychometric properties of all final candidate core outcome measures were fully investigated using the original literature (see Table 2).

FIGURE 2.

Identification of Outcome Measures.

TABLE 2.

Psychometric Properties of Core Measures (Continued)

St Louis University Mental Status (SLUMS) examination to assess cognition
Interrater reliability	Excellent interrater reliability (ICC = 0.990–0.998) and intrarater reliability (ICC = 0.968)27
Test-retest reliability	Regression fit: R2, 0.67; r, 0.8228
Validity	When compared to the Mini-Mental Status Examination (MMSE), the SLUMS has similar sensitivities, specificities, and area under the curve in detecting dementia and may be better at detecting mild cognitive decline.29The Montreal Cognitive Assessment (MoCA) and the SLUMS have excellent convergent validity (r = 0.91)30
Cut-off scores	In individuals with at least a high school education, a score <26 indicates mild cognitive impairment and a score <22 indicates possible dementia. In individuals with less than a high school education, scores <24 indicate mild impairment and <20 indicate possible dementia.29
Patient-Reported Outcomes Measurement Information System (PROMIS) Global 10 to assess health-related quality of life
Interrater reliability	N/A (survey)
Test-retest reliability	Test-retest in stroke was high across all items.31 Stroke = Responses on all PROMIS Global-10 items were significantly associated with their prior responses (correlation coefficients ≥0.80).Test-retest reliability for PROMIS Global-10 was high (ICC ≥0.85) in patients with lupus.31
Validity	Correlates with ED-Q5 at 0.7232Correlates with SF-36 = 0.90 (physical function items) = 0.85 (mental health items)33
MDC/MCID	MCID of 1 standard deviation from the mean is recommended34,35
Cut-off scores	PROMIS measures would use the T-score metric, in which scores have a mean of 50 and a standard deviation of 10 compared with the general population33
European-Quality of Life-5D-5L (EQ-5D-5L) to assess health-related quality of life when participating in international studies or when a language barrier prevents the use of PROMIS Global-10
Interrater reliability	N/A (survey)
Test-retest reliability	For individuals with chronic diseases, the ICC was 0.82.36
Validity	Construct validity: for people with hip/knee ostearthritis—high correlation with Western Ontario and McMaster Universities Arthritis Index (WOMAC).37 No correlation with COPD assessment tool/St George's Respiratory Questionnaire for patents participating in pulmonary rehabilitation,38 for patients who have had a cerebrovascular accident, poor validity with modified Rankin tool, and Barthel index39Content validity: For people with ostearthritis, higher with the 5 levels (5L) rather than 3 levels (3L) especially in domains of mobility, usual activities, and pain/discomfort37
MDC/MCID	For individuals with hip/knee ostearthritis MDC = 0.30 in the utility index; MCID = 0.07 pts for improvement, −0.05 patients for becoming worse, 0.32 pts for improvement in patients who did not have symptoms.40 For patients participating in pulmonary rehabilitation MCID = 0.054 for utility index and 6.99 for the visual analog scale question.38
Cut-off scores	No cut-off scores were identified
Medical Research Council-Sum Score (MRC-SS) to assess strength and power
Interrater reliability	The ICC was 0.94 (95% confidence interval (CI): 0.85–0.98) in patients in the intensive care unit.41
Test-retest reliability	Manual muscle testing: For intensive care unit survivors and simulated patients, the ICC was 0.62–100 for the upper extremities and 0.66–1.00 for the lower extremities.42
Validity	MRC-SS has moderate to strong correlations with physical performance measures, predictive of 2-y outcomes, and independently predicted delayed extubation in clients with critical illness.43–45
MDC/MCID	MCID of 1 standard deviation from the mean is recommended35
Cut-off scores	<48/60 on MRC-SS for patients surviving a critical illness that required an intensive care unit admission is considered Intensive Care Unit-Acquired Weakness (ICU-AW)46,47
Short Physical Performance Battery (SPPB) to assess functional mobility
Interrater reliability	Interrater reliability is excellent (ICC = 0 .92) in patients with COPD.48
Test-retest reliability	Test-retest reliability is excellent (ICC ranged from 0.81 to 0.91) in older adults.49–51
Validity	In community-dwelling older adults scores of ≦10 had significantly higher odds of mobility disability at follow-up (odds ratio: 3.38, 95% CI).52 SPPB scores ≦10 predictive of all-cause mortality in meta-analysis of 17 studies.53
MDC/MCID	Range from 0.54 to 2.9 depending on patient population.49,51,54
Cut-off scores	Score of <9: poor physical performance and indicative of physical frailty.55Score of ≤6: associated with a high fall rate; a score of 7–9: identifies high risk for recurrent falls in women56
2-min step test (2MST) to assess endurance
Interrater reliability	Excellent interrater reliability (ICC = 0.999–1.000, P < .0001).57
Test-retest reliability	Excellent relative test-retest reliability (ICC = 0.927–0.934, P < .0001).57
Validity	Convergent = excellent with exercise/activity history or fitness levels. Additional evidence for “relationship between 2MST steps and psycho/cognitive measures, activity performance, training and health status, and age.”58
MDC/MCID	It is recommended metabolic equivalents (METs) are calculated. [MET = VO2/3.5 and VO2 = (0.2 × frequency of stepping rate) + 1.33 × 1.8 (step height in meters)(stepping rate) + 3.5] 20 studies have found that even an increase of 1 MET level can equate to clinically important changes in individuals in cardiac rehabilitation.20,59–61
Cut-off scores	Age-related norms for older adults reported by mean (standard deviation): age 60–64: men 101,21 women 9124; age 65–69: men 101,23, women 9026; age 70–74: men 95,23 women 8425; age 75–79: men 91,27 women 8424; age 80–84: men 87,24 women 7523; age 85–89: men 75,24 women 7022; age 90–94: men 69,26 women 5821,62

COPD, chronic obstructive pulmonary disease; ICC, intraclass correlation coefficient; MCID, minimal clinically important difference; MDC, minimal detectable change.

The measures identified as candidates from each construct were compiled into a list of potential core outcome measures and reviewed for overall time requirements, ability to capture functional levels and be used in all settings, possible overlap of constructs, and ability to capture the core standardized tasks recommended in the movement system diagnosis approach.63 This resulted in identifying a final recommended set of core outcome measures (see Table 3) that was unanimously agreed to on May 29, 2020. A description of domains measured by each measure, scoring, and equipment, space, and time requirements is provided in Table 4. The APTA provided support throughout the process and published the measures on their Web site on June 29, 2020, https://www.apta.org/your-practice/outcomes-measurement/covid-19-core-outcome-measures.18

TABLE 3.

Core Outcome Measure Recommendationsa

Constructs	Core Outcome Measure
Cognition	Saint Louis University Mental Status (SLUMS) Examination
Health-related quality of life	PROMIS Global-10 (health-related quality of life measure) or EQ-5D-5L when completing research for an international audience
Muscle strength and power	Medical Research Council-Sum Score (MRC-SS)
Functional mobility	Short Physical Performance Battery (SPPB) 4-m gait speed subscore 5 times sit-to-stand subscore Balance screen
Endurance	2-min step test

aRecommendations: include all raw data to examine detailed changes in performance.

TABLE 4.

COVID-19 Recommended Core Outcome Measures Described

COVID Core Outcome Measures	Domains Measured	Scoring	Equipment and Space Needed	Time to Complete
2-min step test58	Measures cardiovascular endurance. Highly correlated with the 6-minute walk test.	Number of times the right knee reaches the appropriate height in 2 min	Timing device and tape measure and tape.	5 min
EQ-5D-5L 64	Measures health-related quality of life in the 5 dimensions of mobility, self-care, usual activities, pain/discomfort, and anxiety/depression.	Describes health status in 5 levels from having no problems to being unable/having extreme problems	Paper and pencil or computer based. Can be self-completed.	<5 min
PROMIS Global-1065	Consists of 10 global health items that represent 5 core PROMIS domains (physical function, pain, fatigue, emotional distress, and social health). Four items are used to assess global physical health.	Lower scores indicate more of the construct being measured (eg, more fatigue and more limitations of physical function). Uses a T-score metric in which 50 is the mean of a relevant reference population and 10 is the standard deviation (SD) of that population.	Self-completed either on paper or computer/tablet.	1.8 min
Medical Research Council-Sum Score (MRC-SS)66	Muscle strength in supine position for shoulder abduction, elbow flexion, wrist extension, hip flexion, knee extension, and ankle dorsiflexion	The total MRC sum score ranges from 0 (total paralysis) to 60 (normal strength). The score is the sum of the MRC score of 6 muscles (3 at the upper and 3 at the lower limbs) on both sides, each muscle graded from 0 to 5.	Bed, goniometer to measure bed recline	5–10 min
St Louis University Mental Status (SLUMS)29	Identify individuals with neurocognitive impairment and is sensitive to change over time. Measures attention, immediate and delayed recall, orientation, numeric calculation and registration, visual spatial relations, executive functioning, and extrapolation.	It is a 30-point, 11-question questionnaire that tests orientation, memory, attention, and executive function. Higher scores indicate better performance.	Paper and pencil. Clinician-administered examination.	7 min
Short Physical Performance Battery (SPPB)21	Measures balance in standing in progressively narrowed base of support, 3-m or 4-m walk, and time to rise from a chair 5 times.	Scores range from 0 to 12, with higher scores indicating better performance.	Requires a chair, timer, and 3-m (9.8 ft) area to walk in. May use an assistive device or chair with arms if needed.	10 min

The Task Force recommends using the core set across all clinical settings for patients who have functional impairments secondary to COVID-19 and who have goals to improve in the associated constructs or may have unmonitored cognitive decline that may impact the plan of care. Recommendations for the order of completion of the core measures can be found within the algorithm. A review of these recommendations is encouraged to ensure proper monitoring of cardiovascular and pulmonary tolerance, as the demands of the core outcome measures increase. The algorithm order is progressive in energy demand to decrease fatigue, as patients progress down the pathway.

Clinicians should record the raw data for each measure along with the total score. The raw data allow the clinician and patient to see subtle changes in performance that may be missed when the raw data are converted to an ordinal or nominal score. An example of the use of raw scores on the Medical Research Council-Sum Score (MRC-SS) demonstrates the value of recording raw scores and not just total scores. Consider a patient showing upper extremity strength improvement over the last 6 weeks but is earning a consistent score of 42/60 on the MRC-SS during repeated testing. The raw data illustrate that strength in the upper extremities has been showing continued gains in manual muscle testing. However, an actual loss of muscle performance of the lower extremities has resulted in the same sum score on the MRC-SS. Recording raw scores would be critical to understanding the functional changes of this patient. In addition, the inclusion of raw data for the components of the SPPB, such as recording the gait speed and time for the balance test, may add to the clinical picture for the patient.

We also recommended that clinicians record a score of 0 when a patient is unable to complete a core measure. Having the clinician attempt all the core outcome measures and record a 0 on a test is meaningful information for the next testing cycle and aids in appreciating the trajectory of recovery of the population of patients with and recovering from COVID-19. The importance of 0 can be demonstrated when we consider the SPPB. A patient who is very low functioning due to experiencing the severe effects of COVID-19 may score a 0 on the SPPB for 4 weeks, but on serial testing, the score progresses to a 2/12 at 8 weeks; therefore, the clinician has captured meaningful functional improvements for this patient.54 If we carry this clinical scenario out further, at 3 months, the patient scores a 10/12 at 12 weeks.

As data become available, recommended measures can be reviewed and adjusted. Waiting for research to be completed before recommending a core set of outcome measures for this population would mean that a wealth of potentially beneficial data may be lost due to a lack of consistent outcome measurement across the continuum of care. The panel and the Task Force's goal is to be proactive, to improve our ability as a profession to demonstrate our value in the care of these patients, and to be able to identify early on best practices for rehabilitating these patients.

TESTING INTERVALS

The Task Force recommends that the core set of measures be completed according to the following prescriptive timeline to understand best the effect of therapy on functional performance and the recovery trajectory.14 All testing should be done upon initial entry to physical therapy services. The Task Force recognizes the clinician may need 1 to 2 visits to complete the assessment due to patient tolerance. The core set should be done prior to discharge from the current setting and entry to the next setting. Additionally, the core set should be completed when a significant change in clinical presentation occurs. Finally, completing testing at 30 days, 3, 6 and, 12 months post-diagnosis while under physical therapy's care would be extremely helpful and support researchers' efforts to study this novel disease and the response to interventions thoroughly. The Task Force anticipates a need to consider longer-term testing recommendations such as 2, 3, 5 and 10-year testing intervals based upon longitudinal tracking of other diseases and syndromes.67,68 However, an alternative frequency schedule is recommended for the St Louis University Mental Status examination due to the possible learning effect of repeated testing. The panel recommends completing cognitive testing if no prior score has been attained, at 1 to 2 months following hospital discharge (when transient cognitive decline is expected to have resolved),17 with a decline in condition during postacute care, or at discharge from all physical therapy services when a prior deficit was noted, and the patient is not receiving skilled care from another specialist for cognitive impairments. This frequency schedule has a primary goal to ensure that the therapist is properly adapting the plan of care to the current capabilities of the patient while working to ensure that physical therapists are identifying and referring patients appropriately when cognitive decline may indicate underlying neurologic pathology. Physical therapists should refer the patient when a cognitive decline is initially noted, worsens without explanation, or persists beyond 1 to 2 months after hospital discharge.17

LIMITATIONS

The Task Force recognized and discussed several limitations regarding the process for determining the recommendations for the set of core measures. First, the Task Force relied on work that had already been published, such as clinical practice guidelines, systematic reviews, meta-analysis papers, and Web sites including sralab.org/rehabilitation-measures and PTNow.org, to generate the original outcomes. The Task Force only explored primary articles where gaps existed in the principal sources. The Task Force also relied on selecting measures based upon other diseases and syndromes with a presentation similar to that of patients with or recovering from COVID-19, such as measures recommended for post-ICU syndrome, heart, and lung failure, or Guillain-Barré syndrome. COVID-19 is a novel disease, and the Task Force may have missed key constructs to assess and key outcome measures that may be more sensitive to detect change. Finally, the Task Force recognized and debated the limitations we had imposed on ourselves by our inclusion/exclusion criteria. Measures such as the 6-minute walk test and handgrip strength with well-established value were considered. However, the Task Force appreciated that these measures might be impractical or impossible to administer depending on the limitations of the setting and infection prevention guidelines. We also eliminated some valuable measures because of copyright or license requirements that pose an increased burden to many sites. Thus, many of these common clinical measures were considered, as we developed recommendations for secondary measures within a clinical application algorithm.18 Finally, the Task Force recognizes the clinician may experience a floor or ceiling effect by only using the core set of outcome measures when assessing a patient with either a low or high level of physical function.

NEXT STEPS

To address the issue of floor and ceiling effects, return to work and sports, and community reintegration, and to provide recommendations for screening and additional tests and measures based on the patient's clinical presentation, the Task Force has designed a clinical application algorithm to guide clinicians using the core measures along with secondary measures to address a large range of functional levels. This can be found at https://www.apta.org/your-practice/outcomes-measurement/covid-19-core-outcome-measures.18 At the time of submission, a team was completing a second manuscript that demonstrates the utility of the clinical application algorithm through patient case scenarios in different clinical practice settings and severities of movement impairments and dysfunction. A subgroup of the Task Force is also applying a similar approach to reach recommendations for the pediatric population.

CONCLUSION

Due to the novelty of COVID-19, no outcome measure data specific to COVID-19 could be integrated into our process. These recommendations are considered an early guide to provide clinical recommendations based on expert consensus and to align clinical practice to allow for greater compilation of data for research. This should speed the process of ascertaining what outcome measures may be valid and reliable within this population. As such, updates to these core outcome measures will be required as more is learned about COVID-19.