Physical inactivity is associated with more frequent exacerbations, hospitalizations, and increased mortality in patients with chronic obstructive pulmonary disease (COPD) (1–3). As important as it is to assess and document patients' level of physical activity (PA) during routine care to tailor therapy and behavioral counseling accordingly, this is currently not performed in most health care settings. Kaiser Permanente, a large integrated health system, is the only system we are aware of that routinely measures and documents one dimension of PA in the form of an “exercise vital sign” (EVS) across all outpatient encounters in its electronic medical record (EMR) system. Although the EVS has been validated across the health plan population (4) and has construct validity with its ability to predict all-cause readmission and mortality in COPD (2,3), it has not been validated using commonly accepted clinical measures of COPD disease severity. Therefore, the purpose of this study is to further establish the construct validity of the EVS in persons with COPD. The EVS may be an efficient proxy measure to assess PA during routine care compared with lengthy questionnaires or objectively measured PA.
Design and Sample
Data for this cross-sectional analysis were drawn from baseline measures of patients with COPD who are included in an ongoing pragmatic clinical trial of a PA coaching intervention (Walk On!) that was approved by the institutional review board (5). Written informed consent was obtained from patients who contributed patient-reported and functional performance measures, and a waiver of consent was approved by the institutional review board for data obtained from the EMR. We used EMR, claims, and administrative records to automatically identify patients who had been hospitalized, an observational stay, or emergency department visit for a COPD exacerbation in the previous 12 months. Patients included in the analyses were ≥40 yr old; were on at least a bronchodilator or steroid inhaler before the encounter or, if not on an inhaler, had a previous COPD diagnosis; and were continuously enrolled in the health plan in the 12 months before study inclusion. We further excluded patients if they (a) had a ratio of forced expiratory volume in 1 s to forced vital capacity (FEV1/FVC ratio) >0.70 at any point in the past 24 months for those with spirometry data; (b) were discharged to hospice, a skilled nursing facility, long-term care, or another acute care hospital during their index admission; (c) were nonambulatory at admission or discharge; (d) had Alzheimer's disease, dementia, or metastatic cancer; (e) were morbidly obese (body mass index of >40 kg · m−2); or (f) completed pulmonary rehabilitation in the 6 months before cohort selection.
Self-reported PA (EVS)
All patients in our health system are asked two questions to capture their PA during the intake process for all outpatient visits. The EVS is administered by medical staff who ask: 1) “On average, how many days per week do you engage in moderate to strenuous (vigorous) exercise (like a brisk walk)?” and 2) “On average, how many minutes do you engage in exercise at this level?” Response choices for days are categorical (0–7). Minutes are recorded as 0, 10, 20, 30, 40, 50, 60, 90, 120, and 150 min or greater. The EMR software multiplies the two responses to display total minutes per week of moderate or vigorous PA. The EVS is summarized into three categories according to established cutoffs used in the National Health and Examination Survey: inactive (0 min · wk−1), insufficiently active (1–149 min · wk−1), or active (≥150 min · wk−1) (4). Patients with COPD in our health system have on average 15 ambulatory visits in a year, with approximately 50% of those visits having usable EVS data. We used all available data from the 12 months before study enrollment to classify patients into their usual pattern of PA on the basis of the mode/median (most frequently occurring or median value captured during the study period) or best (the highest EVS captured during the study period) EVS values.
Spirometry (airflow obstruction), Charlson comorbidity index (6), use of supplemental oxygen, and use of long-term steroid use in the previous 12–24 months before study enrollment were automatically captured from the EMR for all patients in the study sample. The Global Initiative for Obstructive Lung Disease (GOLD) (7) was used to classify patients into mild, moderate, severe, and very severe COPD on the basis of the predicted FEV1 (FEV1% predicted) values from the most recent spirometry test. Spirometry data were available in the EMR for 75% of the study sample.
The 6-min walk test (6MWT) and directly measured step counts were only obtained in a subset of patients randomized to the Walk On! intervention and who completed a baseline visit at the start of the program. Mean step counts were obtained in the 2–7 d before the baseline visit using two previously validated devices, a waist mounted pedometer (Omron Alvita HJ325) (8) or an ankle worn accelerometer (Tractivity, Kineteks) (9). Sitting time was measured via self-report: “In the last 7 d, please estimate the time (hours) you spent watching TV or videos (streaming or renting movies, etc.) on a typical day.”
Symptoms and Quality of Life
Assessment of symptoms and quality of life were also completed at the start of the study using the Medical Research Council (MRC) dyspnea scale, COPD Assessment Test (CAT) (10), Personal Health Questionnaire (PHQ-8) (11), and General Anxiety Disorder (GAD-7) (12) surveys and the Patient-Reported Outcomes Measurement Information System (PROMIS)-10 Global scale that produces a physical and mental subscale, respectively (13). Similar to the functional performance measures, these patient-reported outcome measures were collected specifically for the trial and only available for a subset of the study sample.
We used chi-square and one-way ANOVA tests to compare categorical or continuous variables across the EVS categories, respectively, using IBM-SPSS 22.0 (Armonk, NY). A P value of <0.05 was considered statistically significant.
The sample size for this analysis ranged from 172 to 324 for patient-reported (symptoms and quality of life) and functional performance measures that were collected from willing participants for study purposes and 1396 to 1834 for disease severity measures that were available from routine care in the EMR. This variability in the available data across measures was due to our efforts to maximize the use of existing EMR-based data for all patients enrolled in the main pragmatic trial. The sample had a mean age of 72 ± 10 yr (n = 1834), 54% female patients, and an FEV1/FVC ratio of 54.8 ± 14.5 and 60.4 ± 22.7 FEV1% predicted.
Construct Validity of the EVS
Associations between the median/mode and best EVS (inactive, insufficiently active, and active) with disease severity, functional performance, symptoms, and quality of life are shown in Table 1. The best EVS classification resulted in a more balanced distribution of patients across the three levels of activity with findings similar to the median/mode EVS, and thus for simplicity, we focus our reporting of the results on the best EVS data hereinafter.
Patients with higher levels of PA tended to have higher FEV1% predicted (P < 0.01), less severe COPD per GOLD criteria (P = 0.03), and fewer comorbidities (P < 0.001) and were less likely to use supplemental oxygen (P < 0.001). The functional performance measures were also associated with the best EVS categories. Patients who were more active had incrementally higher 6MWT distance (P < 0.001) and higher step counts (P = 0.04). Overall symptom burden (CAT, P = 0.01) and depressive symptoms (PHQ-8, P < 0.01) were significantly associated with level of PA, but dyspnea (Medical Research Council dyspnea scale, P = 0.30) or anxiety (General Anxiety Disorder-7, P = 0.48) was not. Finally, patients with higher levels of PA had higher quality of life (PROMIS-10 physical, P = 0.04; mental, P = 0.05) compared with less active patients.
Our findings suggest that the EVS, captured in a commonly used US-based EMR system by medical staff during routine ambulatory care encounters, could provide a valid and efficient approach to measuring PA in patients with COPD and should be given serious consideration by other health systems given the ease of its administration by frontline outpatient staff. These results further extend our earlier findings, with the EVS being a significant and independent predictor of 30-d readmission and mortality in patients with COPD (2,3). These previous studies strictly relied on available EMR-based data, whereas in this study, we were able to capture other relevant COPD-specific functional performance and patient-reported outcome measures as part of an ongoing trial (5). The positive associations between PA and measures of disease severity, functional performance, symptoms, and quality of life observed in this study are consistent with previous studies that used either lengthy PA surveys or accelerometers (14–18) and therefore further support the validity and utility of the EVS to efficiently quantify PA during routine clinical care.
Because this is a cross-sectional descriptive study, we attempted to describe associations between the primary measure of PA and other related constructs with no assumptions of directionality. However, disease severity is presumably a strong determinant of PA, whereas symptoms and quality of life may be either a cause or a consequence of PA levels. We presented two different EVS summary measures because the mode/median EVS indicator allowed very few patients to be classified as being active, thus potentially reducing our power to detect differences across the EVS categories on several of the symptom and quality of life measures.
Several weaknesses of this study include our heavy reliance on existing data from the EMR for construct validation versus using gold standards such as accelerometers, which would have been very costly; the timing of when the EVS was assessed relative to the other constructs; the thresholds used to categorize patients as insufficiently active and active; and use of moderate to vigorous exercise as a “proxy” for level of PA. Those weaknesses aside, it is notable that as a health system, we have been able to translate the research evidence regarding the importance and need to routinely assess PA levels across our diverse population. In addition, we acknowledge limitations to using the EVS, being a “noisy” unidimensional measure versus a more comprehensive measure like PRO-Active (19) or objective monitoring. However, use of lengthier surveys primarily designed for research does not fit within the time constraints of most existing clinical workflows. Although use of objective monitoring is ideal, standardized interfaces and algorithms are still needed to uniformly process data from disparate, patient-owned PA sensors.
We believe that use of a simple measure like the EVS during routine clinical practice provides a valid tool both for screening and monitoring population level PA and to inform individual-level recommendations for lifestyle behavioral interventions, as was recommended by a 2014 report by the Institute of Medicine on capturing social and behavioral domains in the EMR (20). This article provides additional evidence regarding the construct validity of the EVS in a vulnerable, older clinical population with chronic lung disease.
Research reported in this publication was funded through a Patient-Centered Outcomes Research Institute award (1403-14117).
The authors thank Ms. Leah Maddock, Kathy Silva, and Janet Lee for their assistance with data collection and management.
All statements in this report, including its findings and conclusions, are solely those of the authors and do not necessarily represent the views of the Patient-Centered Outcomes Research Institute, its Board of Governors, or its Methodology Committee.
None of the authors have a conflict of interest related to the present study. The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Waschki B, Kirsten A, Holz O, et al. Physical activity is the strongest predictor of all-cause mortality in patients with COPD: a prospective cohort study. Chest
2. Nguyen HQ, Chu L, Liu IL, et al. Associations between physical activity and 30-day readmission risk in chronic obstructive pulmonary disease. Ann Am Thorac Soc
3. Moy ML, Gould MK, Liu IA, Lee JS, Nguyen HQ. Physical activity assessed in routine care predicts mortality after a COPD hospitalisation. ERJ Open Res
4. Coleman KJ, Ngor E, Reynolds K, et al. Initial validation of an exercise “vital sign” in electronic medical records. Med Sci Sports Exerc
5. Nguyen HQ, Bailey A, Coleman KJ, et al. Patient-centered physical activity coaching in COPD (Walk On!): a study protocol for a pragmatic randomized controlled trial. Contemp Clin Trials
6. Katz JN, Chang LC, Sangha O, Fossel AH, Bates DW. Can comorbidity be measured by questionnaire rather than medical record review? Med Care
7. Vogelmeier CF, Criner GJ, Martinez FJ, et al. Global strategy for the diagnosis, management, and prevention of chronic obstructive lung disease 2017 report. GOLD Executive Summary. Am J Respir Crit Care Med
8. Danilack VA, Okunbor O, Richardson CR, Teylan M, Moy ML. Performance of a pedometer to measure physical activity in a U.S. cohort with chronic obstructive pulmonary disease. J Rehabil Res Dev
9. Sallis R, Roddy-Sturm Y, Chijioke E, et al. Stepping toward discharge: level of ambulation in hospitalized patients. J Hosp Med
10. Jones PW, Harding G, Berry P, Wiklund I, Chen WH, Kline Leidy N. Development and first validation of the COPD Assessment Test. Eur Respir J
11. Kroenke K, Spitzer R. The PHQ-9: a new depression diagnostic and severity measure. Psychiatr Ann
12. Spitzer RL, Kroenke K, Williams JB, Lowe B. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med
13. Hays RD, Bjorner JB, Revicki DA, Spritzer KL, Cella D. Development of physical and mental health summary scores from the Patient-Reported Outcomes Measurement Information System (PROMIS) global items. Qual Life Res
14. Pitta F, Takaki MY, Oliveira NH, et al. Relationship between pulmonary function and physical activity in daily life in patients with COPD. Respir Med
15. Watz H, Waschki B, Boehme C, Claussen M, Meyer T, Magnussen H. Extrapulmonary effects of chronic obstructive pulmonary disease on physical activity: a cross-sectional study. Am J Respir Crit Care Med
16. Garcia-Aymerich J, Serra I, Gómez FP, et al. Physical activity and clinical and functional status in COPD. Chest
17. Altenburg WA, Bossenbroek L, de Greef MH, Kerstjens HA, Ten Hacken NH, Wempe JB. Functional and psychological variables both affect daily physical activity in COPD: a structural equations model. Respir Med
18. Gimeno-Santos E, Frei A, Steurer-Stey C, et al. Determinants and outcomes of physical activity in patients with COPD: a systematic review. Thorax
19. Gimeno-Santos E, Raste Y, Demeyer H, et al. The PROactive instruments to measure physical activity in patients with chronic obstructive pulmonary disease. Eur Respir J
20. Capturing Social and Behavioral Domains in Electronic Health Records, Phase I and Phase II
. Washington (DC): Institute of Medicine; 2014.