Subjective Measurement of Activity in Chronic Obstructive Pulmonary Disease

Abstract

Activity is an essential part of life. Daily activities include caring for one's self and one's environment, moving about the home and community, and fulfilling various social roles. Personal fulfillment activities, such as interaction with others, hobbies, and attending social or entertainment events, play an important role in life satisfaction and well-being. Physical activity in the form of movement and exercise is essential for optimal fitness and health. Given these varied perspectives, what constitutes “activity” and how is it measured, both generally and in chronic obstructive pulmonary disease specifically? How should “activity” be measured as an outcome of treatment—pharmacologic, surgical, or rehabilitation? This paper discusses patient-reported measurement of activity, with specific reference to patients with chronic obstructive pulmonary disease. It begins with an overview of key measurement issues to consider when selecting or developing instruments, followed by clarification of two perspectives or classes of patient-reported activity assessment: functional and physical activity. Examples of measures designed to capture these concepts are provided. The paper concludes with recommendations for evaluating activity-related outcomes in patients with chronic obstructive pulmonary disease.

Keywords::

• Activity
• Functional Status
• Chronic Obstructive Pulmonary Disease
• COPD
• Outcomes
• Outcome Measurement
• Instruments
• Measurement
• Clinical Trial

INTRODUCTION

Selecting or developing a patient-reported measure of activity to test outcomes of treatment is not quite as simple as it might first appear. This is due in large part to the different paradigms, terms, and measures for evaluating this surprisingly elusive concept. Clearly, activity is an essential part of life. Daily activities include caring for one's self and one's environment, moving about the home and community, and fulfilling various social roles. In the rehabilitation literature, the term “activities of daily living” (ADL) refers to dressing, eating, ambulating, and taking care of personal hygiene, while the phrase “instrumental activities of daily living” (I-ADL) refers to activities such as cooking, shopping, housework, personal accounting, and transport around the community. Although these types of activities are important, people also perform activities for personal satisfaction or fulfillment, such as hobbies, interacting with others, and attending various types of social or entertainment events. Patients with chronic obstructive pulmonary disease (COPD) commonly describe problems performing ADL, I-ADL, and activities for personal fulfillment for a variety of reasons, such as dyspnea, cough, and fatigue, among other concerns.

The term “activity” should not be restricted to ADL- and I-ADL-related tasks and activities performed for personal fulfillment. Physical activity in the form of movement and exercise is important for achieving or maintaining fitness and has been closely linked to morbidity and mortality. For those with COPD, lower levels of physical activity have been linked to higher risk of hospital admission and shorter survival ([1], [2]). An important goal of pulmonary rehabilitation programs is to increase muscle strength and endurance, with the intent of having patients continue their program of physical activity at home in order to maintain optimal levels of cardiopulmonary and musculoskeletal fitness over time.

The purpose of this paper is to discuss the measurement of activity as a patient-reported outcome (PRO) in clinical trials of COPD. The paper begins with a brief review of measurement considerations, followed by a discussion of two perspectives or classes of activity: functional activity associated with ADL/I-ADL tasks and personal fulfillment activities; and physical activity linked to movement and exercise for health promotion. This forms the conceptual basis for a discussion of measurement strategy with examples of instruments designed to capture these two classes of “activity.” The paper concludes with a discussion of daily activity prescriptions for optimal health and a recommendation for developing a consensus around patient-reported activity measurement in clinical trials of COPD.

MEASUREMENT

Measurement has been defined as the “rules for assigning symbols to objects to either represent quantities of attributes numerically (scaling) or define whether the objects fall in the same or different categories with respect to a given attribute (classification)” ([3]). The measurement of “activity,” therefore, refers to a set of rules for assigning numbers to quantify the concept of “activity,” not unlike assigning numbers to represent levels of pulmonary function or exercise capacity. The numbers assigned may be expressed in terms of time spent in activity, exertion associated with a given activity, or the number, ease or difficulty with which activities are performed. Assigning numeric values to types of activities, such as calories expended or metabolic equivalents (METS), would be another way to measure “activity.”

Reliability, validity, responsiveness

Clearly a research instrument must be accurate to be useful; on the other hand, no instrument is perfectly accurate. An important question for clinical trial design, therefore, is the extent to which a given instrument's measurement error is known and tolerable. “Error” in validity creates a form of bias; the numeric values represent something other than the intended content, which may lead to the potential for Type I error, i.e., concluding an effect was present when, in fact, it was not. Error in reliability creates noise, attenuating effects and leading to Type II error, i.e., concluding an effect was not present, when, in fact, the effect simply could not be detected.

In addition to conceptual accuracy and precision (validity and reliability), the instrument must be responsive, that is, sensitive to change and interpretable. Simply put, the numbers on the scale must move up or down with change in the underlying concept, and the values assigned to individuals or groups of patients and those associated with within-patient and within-group change and between-group differences must have meaning. Confidence in an instrument and the meaning associated with the values attributed to it build over time, as the tool is used in various studies or trials and a history and context for interpretation is developed. Table 1 summarizes criteria that can be used to evaluate the validity, reliability, and responsiveness of an instrument being considered for use in a descriptive study or clinical trial.

Table 1 Criteria used to evaluate an instrument with reference to activity measurement in COPD ([3])

Validity:

• Contents. Does the instrument (item) represent the underlying concept? Do the items represent an adequate and balanced sampling of the concept? •

• Concurrent. Are the scores or values related to different instruments of the same construct? Does this activity measure correlate with other activity measures?

• • Discriminant. Can the instrument discriminate and differentiate groups known to differ on the construct? Can a particular activity instrument differentiate healthy patients from COPD patients and distinguish patients with different levels of disease severity?

• • Convergent. Are the scores/values related to those of other instruments of related construct? Can the activity measure correlate with, at least to a certain degree, other related outcomes such as psychological well-being health-related quality of life, or exercise tolerance?

• • Divergent. Are the scores/values independent of scores on instruments measuring unrelated constructs? Are the scores weakly related to scores from measures of weakly related concepts, such as FEV1?

• • Predictive. Can the instrument predict a future condition or event? For example, can baseline activity score predict number of exacerbations in the ensuing year? Or life expectancy?

Reliability:

• Internal consistency. Do multiple items measuring the same construct/dimension behave in similar ways?

• • Reproducibility. When the condition is replicated, are the scores replicated?

Responsiveness:

• Changes. Can the instrument detect signals or change when the underlying conditions change? For example, will a change in function or activity level occur with improvements in exercise tolerance? •

• Interpretation. Are the scores interpretable? Does the metric have meaning?

*For more information on testing subjective measures, see Nunnally JC, Bernstein IH. Psychometric Theory 3rd ed. New York: McGraw-Hill; 1994.

Defining the outcome

Selecting an existing instrument or developing a new tool for a clinical trial requires not only a clear definition of the underlying construct, in this case “activity,” but also an understanding of its proximal or distal relationship to the treatment's mechanism of action. For example, changes in expiratory flow rates are used to evaluate bronchodilators; exercise tolerance tests can be used to evaluate the effectiveness of exercise programs, such as cycling or walking. These are examples of a proximal relationship, where there is a direct link between the treatment's mechanism of action and the outcome measure. Distal outcomes can also occur with treatment. Symptom relief and reduction in rescue medication use, for example, are distal outcomes of bronchodilator therapy, providing important information on the effects of treatment. Improvements in patient's level of activity may be proximal or distal outcomes, depending on the intervention.

A desired goal or outcome of treatment for patients with COPD is to maintain or increase patient activity. This outcome can be translated into several different empirical questions. Does the treatment improve functional activity, i.e., patient performance of ADL, I-ADL, and activities for personal fulfillment? Does the patient perform more of these activities, or perform the same activities with less difficulty? Does the treatment increase the patient's overall level of physical activity? Are patients moving around more–getting up, down and around, or exercising regularly? Do patients naturally increase their functional or physical activity as their health (e.g., airflow obstruction or endurance) improves with pharmacologic or rehabilitation interventions? Or are additional, “adjuvant therapies,” such as ongoing and regular encouragement or reward programs, required to realize improvements in activity?

Given this description of measurement and the types of activity outcomes that may be associated with treatment–What constitutes the patient's report of “activity”? And what patient-reported measures are available to capture this outcome?

MEASURING ACTIVITY

The patient's perspective

Qualitative research is a study design that uses systematic empirical methods and inductive reasoning to understand a phenomenon from the perspective of the person experiencing it. During the early stages of the instrument development process, qualitative data gathered through patient interviews and focus groups are used to understand the patient's perception of the concept of interest and inform the content and structure of the instrument. Qualitative methods can also be used to evaluate an existing measure in order to determine the adequacy of an instrument for its intended purpose (content validity). In the case of activity measurement, qualitative research can be used to understand the meaning of “activity” to patients with COPD, inform the underlying content of an outcome measure, substantiate its content validity, and ultimately enhance its sensitivity to change.

Despite the extensive empirical literature related to physical functioning in COPD and the number of instruments purported to measure various aspects of functioning in this population, few qualitative studies have been reported that describe activity from the patient's perspective ([4], [5], [6], [7], [8], [9]). In a phenomenological study of patient experiences, Barnett discussed lost social activity, diminished role within the family, and lost intimacy in personal relationships ([4]). Results from Jeng and colleagues' study of patient experience with daily activity after hospital discharge suggest patients slow and simplify their activities, act according to their perceived abilities, and protect themselves as they strive for an independent life ([5]). McBride found that patients explored various methods to increase their activity while maintaining feelings of “control”([9]).

One qualitative study has been reported that specifically addressed day-to-day activity from the perspective of patients with COPD ([7], [8]). This study involved one-on-one interviews with 24 patients (12 men and 12 women) with moderate-to-severe COPD. Thematic analyses were employed in order to: ([1]) identify specific activities these patients performed each day and those they could no longer perform; ([2]) understand the kinds of problems these individuals experienced as they performed or attempted to perform specific activities and how they overcame the problem or let go of the activity; and ([3]) understand the purpose and meaning of activity in their lives ([7], [8]).

The dimensions of activity patients identified are shown in Table 2. According to these patients, the decision to perform a given activity is a function of the personal satisfaction that accompanies the activity, the presence of “intruders,” that are expected to adversely affect the experience (largely symptoms), and the “enablers” that can be used to help make performance possible (planning or pacing, using assisted devices, personal assistance) ([7]). Patients also described their determination to continue to perform certain activities and their sadness in giving up others, both in terms of their desire to preserve their personal integrity, i.e., maintain a sense of wholeness as they encountered a variety of physical changes that interfered with day-to-day activity ([8]). For these patients, activity played an essential role in feelings of effectiveness (being able to perform activities for themselves and others) and feelings of connectedness (being able to be with others) ([8]).

Table 2 Categories of functional activities valued by patients with COPD ([7])

• Household maintenance—Cleaning, preparing meals and doing repair work

• Movement – Carrying, walking and exercise

• Family activities—Physical activities with spouse, children or pets

• Social activities—Holding conversations, playing cards, dinner, entertaining

• Work—Physical, full-time or part-time

• Altruistic avocation—Contributing to the welfare of family, friends or the community

• Recreation—Having fun, e.g., traveling, hobbies and gardening

Together, the results of these qualitative studies suggest that patients with COPD perceive and describe “activity” as far more than ADLs or I-ADLs. They need and want to perform activities that enable them to not only care for themselves, but care for themselves effectively, and participate in activities that are personally satisfying and help them stay connected to others. These activities include family, social, work, altruistic, and recreational activities that give their lives meaning. The data also suggest patients would increase their activity with treatment if the treatment alters factors, such as breathlessness or fatigue that interfere with desired activities. Further, treatment effects could be detected if the instrument used to test the effect included activities that are meaningful to patients and thus likely to change with treatment.

The health promotion perspective

The literature on subjective measures of physical activity addresses self-reporting of involvement in a purposeful program of exercise to enhance physical fitness and promote health. These physical activities include running or jogging, swimming, cycling, walking, or team sports such as basketball, soccer, or tennis, quantified in terms of time, intensity, exertion, and frequency of activity to achieve optimal levels of fitness, i.e., cardiopulmonary endurance, muscle strength, flexibility or range of motion, and body composition.

The physical activity paradigm is different from, yet related to, the functional performance paradigm outlined here. In fact, one of the dimensions of activity shown in Table 2 is Movement, defined as “activity characterized by deliberative motion” associated with purposeful activities, therapeutic movement, and exercise. In the qualitative study described previously, some patients discussed movement as part of their daily activity, motivated by the associated feelings of release or the satisfaction they felt participating in activities to improve their general health ([7]). It is possible, therefore, that the physical activity of patients with COPD may improve with treatment. Like the general population, however, only a sub-group of patients with COPD find a physical exercise program sufficiently rewarding or satisfying to integrate it into the normal course of their lives. This may underlie the public health interest in encouraging otherwise sedentary older adults to increase their involvement in functional activities requiring movement or exertion, such as vacuuming, gardening, or walking briskly during errands, in order to enhance fitness.

SUBJECTIVE MEASURES OF ACTIVITY IN COPD

The two perspectives of activity outlined here frame the discussion of subjective or patient-reported measures of activity in COPD. To capture the concept adequately, measures assessing patient performance of functional activities should capture day-to-day activity patients need and want to perform. The term functional performance may be useful to further clarify this class of activity for measurement purposes. This term grew out of a review of the literature and analytical framework to clarify the concept of “functional status” for purposes of instrument development and outcomes assessment ([10], [11], [12], [13]). Functional performance is defined as the “physical, psychological, social, occupational, and spiritual activities that people choose to do in the normal course of their lives to meet basic needs, fulfill usual roles, and maintain their health and well-being” ([10], [11]). These are activities people believe they need or want to perform, subject to the limits imposed by capacity and generally requiring less than maximal capacity to accomplish ([10], [11]). In contrast, measures of physical activity assess the frequency, duration, and intensity of activity to enhance or maintain physical fitness. It is important to note that activity measures, whether functional or physical, do not – nor should they – measure symptoms, health-related quality of life, health status, or well-being.

The following summary of patient-reported measures of activity is for illustrative purposes, rather than a comprehensive review of the literature, with select instruments provided as examples of the types of measures that might be considered for evaluating activity outcomes in clinical trials of COPD.

Functional performance measures

Generic instruments

A number of instruments have been developed to measure function in the general population or in specific groups, such as the elderly or those with a cardiovascular condition (Table 3). Because they are “generic,” i.e., designed for use across populations, they are not designed to capture specific activities patients with COPD have described as important and difficult to perform, and therefore may be less sensitive to change in clinical trial settings. The advantage of these measures, however, is the ability to compare scores in COPD samples with those of other samples, giving the values context and meaning.

Table 3 Patient-reported instruments that capture one or more dimensions of functional performance

Generic

• Functional Status Questionnaire (FSQ)

• Physical Function (PF) & Role Physical (RP) Subscales of the Short-Form 36

COPD-Specific

• Functional Performance Inventory

• Manchester Respiratory ADL Questionnaire Health Outcomes Measures with Functional Performance Items or Sub-Scales

• Clinical COPD Questionnaire (CCQ)

• Pulmonary Functional Status & Dyspnea Questionnaire (PFSDQ-M)

• Seattle Obstructive Lung Disease (SOLD)

• St. George's Respiratory Questionnaire (SGRQ)

The 34-item Functional Status Questionnaire (FSQ) was designed to assess the functional status of patients seen in primary care settings – to both detect problems and track patients over time ([14], [15]). It includes specific reference to, and assessment of, ADLs (3 items) and I-ADLs (6 items) together with items assessing role, social, and psychological functioning. A “warning zone” score was developed to help clinicians interpret the measure, although these guidelines have not been assessed or validated in patients with COPD. Validation of the measure in primary care included correlation with bed disability days, restricted activity days, satisfaction with health, and social contact ([14]).

Items comprising the Physical Function and Role Physical subscales of the short-form-36 (SF-36)([16]) are consistent with the definition of functional performance outlined here ([17]). The widespread use of the SF-36, together with the availability of normative data, facilitate the interpretation of baseline, end of study, and change scores on this instrument.

COPD-specific instruments

COPD-specific instruments to assess functional performance should address the activities these patients consider meaningful, are motivated to perform, and find difficult to execute particularly as their disease progresses. These attributes reflect the content validity of the instrument and enhance its likelihood of sensitivity to change with treatment. Two known measures specifically designed to assess functional performance in the COPD population are shown in Table 3 and described briefly next.

The Functional Performance Inventory (FPI) was designed to assess functional performance in patients with COPD across the 7 types of activities that are both important to these patients and difficult to perform due to their disease (see Table 2) ([18]). Initially developed as a 65-item questionnaire, the measure was later reduced to a 32-item short form (FPI-SF) and a 12-item “FPI-Mini” intended for clinical use. The initial validation study included tests of internal consistency reliability (α = 0.96, Total Score) and reproducibility (ICC = 0.87, Total) and correlations with the Duke Activity Status Index (DASI), FSQ, perception of disease severity, pulmonary function (FEV₁), and a family member's assessment of the patient's function (Katz Adjustment Scale for Relatives). Clinical validation included replication of correlations with the DASI, FSQ, and disease-related variables as well as a test of its relationship to 12-minute walk distance and movement around the home, the latter captured through a wrist-worn actigraph previously validated to capture movement associated with daily activity ([19], [20]). Additional studies have shown consistent evidence of the instrument's reliability and validity ([21]).

The Manchester Respiratory ADL Questionnaire was designed to measure the functional effects of respiratory-specific impairment ([22]). This 21-item instrument is based on a composite of discriminative questions from the Nottingham Extended ADL Questionnaire (NEADL) and the Breathing Problems Questionnaire (BPQ) and assesses mobility (7 items), kitchen activities (4 items), domestic tasks (6 items), and leisure (4 items). The instrument has been tested for reliability and validity in a COPD sample with high levels of internal consistency reliability (Cronbach's α of 0.91) and the ability to discriminate between COPD patients and controls ([22]).

COPD-specific health status measures with functional dimensions

Most COPD-specific health status or health outcome measures include assessment of daily functioning, together with other dimensions of health specific to the COPD population. Examples are shown in Table 3. The 10-item Clinical COPD Questionnaire (CCQ) includes 4 items related to patient's daily activity ([23]). The 40-item Pulmonary Functional Status & Dyspnea Questionnaire (PFSDQ-M) measure includes 10 items to assess daily activity specifically and 30 to evaluate dyspnea or fatigue ([24]). The 29-item Seattle Obstructive Lung Disease (SOLD) includes 18 items related to activity, while the 50-item St. George's Respiratory Questionnaire (SGRQ) includes the 16-item Activity subscale ([24], [25]). Because these instruments are not dedicated exclusively to activity assessment, items or subscales generally do not cover the depth and breadth of assessment required to fully capture functional performance and establish a treatment's effect on activity, particularly as a primary outcome in a clinical trial.

As one might expect, these different questionnaires contain many common content areas, providing support for the types of activities considered most problematic in this patient population. For example, all of the measures include personal care items, such as dressing or washing, as well as reference to climbing – either stairs, hills, or incline—and walking speed or distance.

Physical activity measures

Generic instruments of physical activity

Physical activity measures have been used in epidemiologic, population-based surveys to assess fitness in populations and to determine or assess cardiovascular risk. Unfortunately, most of these instruments involve 60- to 90-minute one-on-one interviews that ask patients to report their activity over the past year, which is clearly not a suitable approach for a clinical trial. Many are also heavily weighted toward vigorous activity and include items such as jogging, running, tennis/squash, gymnastics, ballet, and figure skating—activities which are generally not relevant for patients with COPD.

COPD-specific instruments

Pitta and colleagues reviewed instruments that have been commonly used to quantify physical activity in COPD patients and described selected measures that may be useful for this patient population ([26]). Representative measures are shown in Table 4. Pitta and colleagues noted that although there are several subjective instruments for assessing physical activity in COPD patients (i.e., diaries, questionnaires), evidence concerning reliability, validity, and responsiveness of these instruments, generally and for clinical trials specifically, are lacking. Among the instruments with some documented validation in the elderly and other populations are the Minnesota LTPA Questionnaire, the Baecke questionnaire, and the Physical Activity Scale for the Elderly (PASE) questionnaire ([27], [28], [29]).

Table 4 Measures of physical activity ([26])

• Baecke Physical Activity Questionnaire

• Minnesota Leisure Time Physical Activity Questionnaire

• Perspiration Score

• Physical Activity Scale for the Elderly (PASE)

• Specific Activity Score

• Tecumseh Occupational Physical Activity Questionnaire

• Yale Physical Activity Survey

• Zuiphen Physical Activity Questionnaire (ZPAC)

CONSIDERATIONS

Instruments designed to assess functional performance capture day-to-day activities people choose to perform in the normal course of their lives to meet their basic needs and fulfill their usual roles. Response scales on these measures reflect the extent and ease with which these activities are performed. In contrast, physical activity measures assess the extent to which people are involved in a program of exercise to enhance physical fitness. In this case, response scales reflect frequency, intensity, and duration of movement. Each perspective has the potential to enhance the understanding of activity in patients with COPD and the impact of treatment. Clearly, functional performance is a concern for this patient population. Physical activity should be a concern as well. Not in terms of the traditional definition of physical exercise to achieve cardiovascular fitness, but in terms of movement around the home and community to prevent the deconditioning, muscle weakness, and joint stiffness associated with a habitual sedentary lifestyle, and that can lead to disability disproportionate to the severity of the underlying airway obstruction.

To date, the functional and physical activity literature and measures have been presented in relative isolation, with each perspective offering its own unique measurement approaches. Perhaps it is time to bring these two perspectives together. One way to do this would be through a new patient-reported instrument capturing the extent to which patients with COPD are performing activities that meet their ADL/I-ADL/satisfaction and social needs under the definition of “functional performance” and, at the same time, determine the extent to which they are engaged in sufficient physical activity for “optimal fitness,” i.e., to keep them conditioned and mobile.

The structure of such a new measure would include the seven areas of daily activity that patients with COPD describe, value, and find challenging (Table 2) juxtaposed with a scaling system commonly associated with physical activity, such as calories, METS, or exertional ratings. Each item/activity in the tool would be assigned an intensity code that reflects the level of physical activity required for this task. By adding frequency and duration to the assessment, it may be possible to arrive at a score that reflects the magnitude of activity performed and the ease or difficulty with which it is performed, simultaneously capturing both perspectives of activity. A second approach would be to use a functional performance measure together with an activity monitor, such as an actigraph, that quantifies physical activity by tracking movement.

Longitudinal, naturalistic studies capturing functional and physical activity parameters simultaneously could help determine the relationship between type and level of daily activity and disease trajectory, morbidity, and mortality in COPD. These data would inform the estimation of minimal levels of daily physical activity necessary for “optimal fitness” in this population. Patient activity recommendations could be developed that are framed within the context of daily activities, rather than physical exercise regimes with their correspondingly poor levels of compliance. These “daily activity prescriptions” would help patients meet their basic needs and fulfill usual roles while maintaining optimal “fitness and health” appropriate to their level of disease severity.

To date, there has been no consensus on the best measure with which to evaluate activity in clinical trials of COPD from either the functional performance or the physical activity perspective. Different studies have used different instruments with different metrics, making the development of context for activity scores—descriptive, change scores, and even effect sizes—slow to evolve. This has also made systematic reviews or meta-analysis very difficult, limiting the understanding of the impact of COPD on activity and the potential benefits of treatment. It would be very useful, therefore, if consensus could be reached regarding both the definition of activity and the best approach for the subjective measurement of activity in COPD to hasten the development of scientific evidence upon which to base therapeutic decisions.

CONCLUSION

Adequately and accurately understanding the effect of treatment on activity begins with the definition of activity and the certainty that the instrument reflects this definition. The purpose of this paper was to describe two classes or perspectives of activity and approaches for the measurement of these perspectives in clinical trials of COPD. Functional performance measures can be used to evaluate the extent to which treatment is affecting day-to-day activities that are important to life satisfaction and well-being, including caring for one's self and environment, interacting with others, participating in family or work activities and hobbies, and attending social or entertainment events. Physical activity measures assess movement and exercise essential for optimal fitness and overall health. Examples of each perspective were provided, with consideration given to a measurement approach that would combine the two perspectives, leading to a more comprehensive understanding of the effect of treatment on activity and to the development of functional performance-based activity prescriptions for health promotion in patients with COPD. The paper concludes with a recommendation that consensus on the measurement of activity in COPD be reached in order to build a coherent body of evidence on the impact of disease and treatment on this important outcome.

Notes

*Items cannot be taken out of the questionnaire context to form an independent instrument without permission from the original instrument's author/copyright holder and establishing that the measurement properties have not changed when the structure has been altered.