Abstract
Background:
Persons with multiple sclerosis (MS) experience co-occurring symptoms termed “symptom clusters” that can be distinguished based on mild, moderate, or severe symptom severity termed “symptom cluster severity.” Physical activity (PA) may be an approach for improving co-occurring symptoms.
Objective:
To examined if PA and social cognitive theory (SCT) variables differed by symptom cluster groups, and if associations between SCT variables and PA were moderated by symptom cluster groups.
Methods:
Secondary analysis of participants with MS (N=205) enrolled in a cross-sectional study. Trend analyses were conducted to determine if device-measured and self-reported PA and SCT variables (i.e., social support, self-efficacy, outcome expectations, goal setting, planning, and impediments) decreased with increased symptom cluster severity. Spearman rho rank-order correlations were conducted between PA measures and SCT variables within each symptom cluster group.
Results:
Linear trend analyses indicated that self-reported PA declined with increased symptom cluster severity groups (F=4.90,p=0.03). Linear trend analyses indicated significant differences among symptom cluster severity groups in social support (F=31.43,p=0.001), exercise self-efficacy (F=22.55,p=0.001), barrier self-efficacy (F=11.48,p=0.001), outcome expectations (F=6.98,p=0.009), and impediments (F=34.41,p=0.001). There were differential associations of moderate magnitude in correlations, such that three SCT variables were associated with PA in the mild group (i.e., self-efficacy, goal setting and planning), two in the moderate group (i.e., social support and goal setting), and four in the severe group (i.e., self-efficacy, outcome expectations, planning, and social support).
Conclusions:
Further research is warranted examining the use of SCT-based behavior change techniques for promoting PA and improving symptom clusters in persons with MS.
Keywords: Multiple Sclerosis, Physical Activity, Social Cognitive Theory, Symptom Clusters
Introduction
Multiple sclerosis (MS) is a chronic neurodegenerative disease of the brain, brain stem, spinal cord, and optic nerves with a current estimated prevalence of 1 million adults in the United States.1 Persons with MS experience a range of symptoms including depression, anxiety, fatigue, and sleep problems.2-4 These symptoms rarely occur in isolation, and the co-expression can be described by the presence of a symptoms cluster. The term “symptom cluster” describes the presence of three or more concurrent symptoms that are inter-related.5 The study of symptom clusters originated in cancer research, and there are reasons for the examination of symptom clusters in MS, including incidence of common, co-occurring, and potentially synergistic symptoms.6-11 Of note, the established symptom clusters among persons with MS include various combinations of pain, fatigue, depression, anxiety, cognitive functioning, sleep quality, and/or irritability; cluster groups further have been distinguished based on experience of mild, moderate, or severe MS symptom severity.6-11 Symptom cluster severity groups in this context specifically represent the recognized phenomena wherein individuals experience a consistent group of symptoms and each symptom within the group is experienced in a similar magnitude among a cluster of individuals.
There are existing theories for conceptualizing symptom clusters and the putative influence on meaningful outcomes. The Theory of Unpleasant Symptoms (TUS), in particular, is a middle-range theory that captures the dynamic impact of influencing factors (i.e., physiological, psychological, and situational), symptom dimensions (i.e., clusters), and performance/behavioral outcomes.12 One central tenet of TUS emphasizes the existence and impact of concurrent symptoms (i.e., clusters) such that symptoms in a cluster can render vicious cycles that are more impactful on outcomes than a single symptom in isolation.12 Symptom cluster research in MS grounded in TUS has established the presence of symptom cluster groups by severity of fatigue, pain, depression, cognitive dysfunction, and sleep quality and reported associations with performance and behavioral consequences (i.e., physical activity and quality of life).6-9 For example, one study established mild, moderate, and severe symptom cluster groups based on fatigue, depression, and pain, and the clusters were associated with self-reported levels of physical activity behavior (i.e., individuals with worse symptoms were less physically active).6
TUS-based examinations of symptom clusters and physical activity as a performance outcome in MS might consider other theory-based constructs that guide behavior change approaches for promoting physical activity. Social Cognitive Theory (SCT) is an appropriate, commonly applied theoretical foundation for behavior change in persons with MS.13 SCT was founded in social learning theory wherein individual behaviors is a product of dynamic interactions among the individual, environment, and subsequent behavior.14 Constructs within SCT that are associated with physical activity behavior include social support, self-efficacy, outcome expectations, goal setting, planning, and impediments.15-17 Social support within SCT focuses on individual perceptions that assistance is available for behavior as well as the strength of support from their social network. Self-efficacy refers to an individual’s confidence that they are capable to perform behavior and outcome expectations encompass the expected consequences of performing the behavior. Goal setting and planning are related to an individual’s process for identifying appropriate objectives and methods for performing behavior and impediments encompass barriers to engaging in behavior.
The consideration of symptom cluster severity groups, rather than symptoms in isolation, offers an opportunity for examining real-world manifestations of symptoms in the context of theory-informed constructs of physical activity in MS. Indeed, SCT has been applied for examining physical activity and symptoms in persons with MS.15-17 One recent study examined the individual contributions of anxiety, depression, and fatigue as moderators of the relationship between self-efficacy and physical activity.18 Results from that study indicated that only levels of anxiety significantly moderated the relationship between self-efficacy and physical activity.18 Further examination is warranted regarding the influence of symptoms on SCT variables and physical activity in MS with an emphasis on symptom clusters.
The present study first examined if physical activity and SCT variables differed by symptom cluster severity groups using an established symptom cluster in MS that includes depression, anxiety, fatigue, and sleep problems which are among the most common and debilitating symptoms reported by persons with MS.19 Then we examined if the associations between SCT variables and physical activity differed by symptom cluster severity groups (i.e., does symptom cluster group serve as moderator). To that end, we first examined expected differences in self-reported physical activity, device-measured physical activity, and SCT variables (i.e., social support, self-efficacy, outcome expectations, goal setting, planning, and impediments) among mild, moderate and severe symptoms cluster groups. We then examined whether expected associations between physical activity behavior and SCT variables differed among the three symptom cluster severity groups. We hypothesized that rates of physical activity and SCT variables would be lower with increased symptom cluster severity groups, specifically the severe and moderate symptom cluster groups would report lower levels of physical activity and SCT variables compared with mild symptom cluster group that would be consistent with a negative linear trend in physical activity and SCT outcomes of interest with increased symptom severity. Such an analysis is important for the design of behavior change interventions targeting SCT variables for physical activity as a function of symptom cluster severity, and perhaps using physical activity as an approach for managing concurrent symptom expression.
Methods
Participants
This paper involved a secondary analysis of data from a large cross-sectional study of persons with MS.19 The sample included 205 persons with MS between 20-79 years of age who were recruited through flyers advertising the study and distributed in the local community, the National MS Society, local MS chapter, the Lakeshore Foundation, and through the University of Alabama at Birmingham (UAB) i2b2 database. Interested participants were instructed to contact the research team for a telephone screening. Inclusion criteria assessed during telephone screening included: (1) age between 20–79 years; (2) ambulation with or without assistance; (3) willingness to complete all testing procedures; (4) relapse-free for at least 30 days.
Procedures
The procedure was approved by the Institutional Review Board at The University of Alabama at Birmingham, and all participants provided written informed consent before participating in study procedures. The data were collected in a one-day testing session at the research facility and additional take-home packet that included device-measured physical activity measurements for 7 days and a questionnaire battery. Participants were provided a pre-stamped, pre-addressed envelope at baseline to return materials following the 7-day data collection period. Participants who returned packets with missing data were contacted via phone to limit missing data; however, not all participants could be reached. Participants received a gift card for participation.
Measures
Symptoms.
The measures of symptoms have been described in full detail elsewhere.19 Briefly, measures of symptoms included fatigue, depression, anxiety, and sleep measured during the in-person testing session. Fatigue was measured using the 9-item Fatigue Severity Scale (FSS).20 Depression and anxiety were measured using the 14-item Hospital Anxiety and Depression Scale (HADS), which includes 7-items for assessment of depressive symptoms and 7-items for assessment of anxiety symptoms.21 Sleep quality was assessed using the Pittsburgh Sleep Quality Index (PSQI). The PSQI includes seven component scores for calculating a global sleep quality score.22 Higher scores on these measures represent greater severity of symptoms. In the primary analyses, scores on the four symptom measures were used to categorize participants into three subgroups (mild, moderate, and severe) based on reported experiences with concurrent symptoms using K-means cluster analyses.19 K-means cluster analyses were utilized in the primary study given it has been identified as a suitable approach for analyzing larger sample sizes.23 This secondary analysis focuses on further comparing outcomes among the three symptom cluster severity groups.
Physical Activity.
The study protocol included self-report and device-measured physical activity. Self-reported physical activity was measured using the Godin Leisure Time Exercise Questionnaire (GLTEQ) during the in-person testing session.24 The GLTEQ represents a reliable and valid measure of self-reported physical activity in persons with MS.25 The three-item survey inquires about engagement in light, moderate, and vigorous physical activity in bouts of 15 minutes or more during the previous 7-day period. The health contribution score (HCS) is calculated by multiplying the recorded number of moderate and strenuous bouts of physical activity by weights of 5 and 9 respectively, into an overall score that ranges between 0 and 98.26
Physical activity was device-measured using accelerometry (ActiGraph, model GT3X+, ActiGraph LLC, Pensacola, FL). We followed a standard waist-worn procedure, wherein the accelerometer was placed in a pouch on an elastic belt. Participants were asked to wear the device above the non-dominant hip during waking hours over a 7-day period, except while showering, bathing, and swimming. Participants recorded the date and time that the accelerometer was worn in a daily log, and this log was used to verify wear time during data processing. Accelerometer data were downloaded, processed into 60-second epochs and then scored for minute-by-minute activity counts and vector magnitude. The focus of this study involved time spent in light and moderate-vigorous physical activity (LPA; MVPA) based on a MS-specific cut-point of 1,584 counts·min−1.27 Days with 10 or more hours were considered valid and included in analyses.
Social Cognitive Theory Variables.
Seven SCT variables were measured using validated questionnaires and included in a take-home packet that was returned with the accelerometer. The SCT variables included social support for exercise, self-efficacy for exercise and barriers for exercise, outcome expectations, goal setting, planning, and impediments. The 6-item Social Provisions Scale (SPS) measured perceived social support for exercise.28 Two measures were used to assess self-efficacy for exercise. The six-item Exercise Self-Efficacy (EXSE) scale was used to assess beliefs in individual’s ability to continue exercising three times per week at a moderate intensity monthly ranging from 1-6 months.29 The 13-item Barriers for Self-Efficacy (BARSE) scale assessed perceived capabilities to exercise three times per week for 40 minutes over the next two months in the face of barriers (e.g., weather).30 Outcome expectations was measured using the 19-item Multidimensional Outcome Expectations for Exercise Scale (MOEES).31 The 10-item Exercise Goal Setting Scale (EGS) measured goal setting for exercise.32 The 10-item Exercise Plans Scale measured planning behaviors for exercise.32 One common impediment for exercise is functional limitations, and this was measured using the 15-item Functional Limitations component of the abbreviated Late-Life Function and Disability Instrument (LL-FDI).33
Demographic and Clinical Characteristics.
Participants reported current age, biological sex, marital status, employment status, race, education level, annual household income, MS disease duration, and MS clinical course. The Patient Determined Disease Steps (PDDS) measured self-reported ambulatory disability status.34
Data Analysis
Data were analyzed using SPSS Version 25 (IBM Corporation, Armonk, NY). Descriptive characteristics of all measures are presented in tables and text as mean ± standard deviation (SD) or median interquartile range (IQR) where appropriate. Data were examined for significant outliers, and one participant’s physical activity value was winsorized to the next closest value given it was ≥ 4 SD from the mean.35 Symptom cluster groups were adapted from a previous study to conduct secondary analyses, and differences between symptom cluster groups regarding demographic and clinical characteristics were confirmed using Chi-square and Kruskal–Wallis tests, as appropriate. Trend analyses were conducted to determine if physical activity and SCT variables decreased with increased symptom cluster severity. Both linear and quadratic models were assessed to determine the best fit, and linear models were the most appropriate across variables. We then conducted Spearman rho rank-order correlations (ρ) between physical activity measures and SCT variables within each symptom cluster group. Non-parametric correlations were most appropriate given the majority of variable distributions were not normal based on Shapiro-Wilks tests. Values for correlation coefficients of 0.1, 0.3, and 0.5 were interpreted as weak, moderate, and strong, respectively.36 Lastly, we conducted multiple linear regression analyses with stepwise entry of variables whereby we examined significant SCT variables as independent correlates of physical activity measures within each symptom severity group.
Results
Participants
Participant demographic and clinical characteristics are fully described elsewhere.19 Briefly, the mean age among participants was 49.4±13.2 years and years since MS diagnosis was 12.8±9.4 years (Table 1). The median(IQR) PDDS score was 1.0(3.0) indicating mild to moderate ambulatory disability and the most participants reported relapsing remitting MS clinical course (92%). The majority of the sample identified as female (76%), married (61%), employed (57%), and Caucasian (65%). There were significant differences among symptom cluster severity groups only in marital status and employment status. The mild symptom cluster group had a higher proportion of participants who were married and employed. Symptom cluster groups did not differ in age, MS duration, biological sex, race, education, income, or MS clinical course.
Table 1.
Demographics and Clinical Characteristics of the Overall Sample and Three Symptom Cluster Severity Groups
| Variable, units | Overall Sample N=205 |
Mild Symptom Cluster N=96 |
Moderate Symptom Cluster N=73 |
Severe Symptom Cluster N=36 |
|---|---|---|---|---|
| Mean(SD) | Mean(SD) | Mean(SD) | Mean(SD) | |
| Age, year | 49.4(13.2) | 49.7(13.2) | 49.3(13.0) | 48.6(13.7) |
| MS Duration, years | 12.8(9.4) | 13.4(9.5) | 13.0(9.7) | 10.5(8.5) |
| Median(IQR) | Median(IQR) | Median(IQR) | Median(IQR) | |
| PDDS | 1.0(3.0) | 1.0(3.0) | 2.0(3.0) | 2.0(4.0) |
| Biological Sex | n(%) | n(%) | n(%) | n(%) |
| Female | 154(75.5) | 75(78.9) | 50(68.5) | 29(80.6) |
| Male | 50(24.5) | 20(21.1) | 23(31.5) | 7(19.4) |
| Marital Status* | ||||
| Married | 125(61.0) | 70(72.9) | 38(52.1) | 17(47.2) |
| Single/Divorced/Widowed | 80(39.0) | 26(27.1) | 35(47.9) | 19(52.8) |
| Employed** | ||||
| Yes | 116(57.1) | 64(66.7) | 40(55.6) | 12(34.3) |
| No | 87(42.9) | 32(33.3) | 32(44.4) | 23(65.7) |
| Race | ||||
| Caucasian | 133 (65.2) | 66(69.5) | 46(63.0) | 21(58.3) |
| Other | 61(34.9) | 29(30.6) | 7(36.9) | 15(41.7) |
| Education | ||||
| High School-Some College | 88(42.9) | 31(32.2) | 36(42.1) | 21(57.1) |
| College Graduate or More | 117(57.1) | 65(67.7) | 37(57.9) | 15(42.9) |
| Annual Household Income | ||||
| $40,000 or Less | 70(35.0) | 17(18.3) | 33(45.8) | 20(58.3) |
| Greater than $40,000 | 130(65.0) | 76(81.7) | 39(54.2) | 15(41.7) |
| MS Clinical Course | ||||
| RRMS | 183(92.0) | 91(94.8) | 61(83.6) | 31(86.1) |
| Progressive | 16(8.0) | 3(5.2) | 8(16.4) | 5(13.9) |
Note. PDDS: Patient Determined Disease Steps; RRMS: Relapsing Remitting Multiple Sclerosis
p <.05
p <.01
Physical Activity
Device-measured and self-reported physical activity (GLTEQ) are reported in Table 2. Valid accelerometer data for at least one day (≥600 minutes) were available for 184 participants (n=86 mild symptom cluster group, n=67 moderate symptom cluster group, n=31 severe symptom cluster group). The average wear time between mild, moderate, and severe symptom cluster groups was nearly statistically significantly different (F= 5.32, p= 0.06), and the mean±SD values were 830.46±88.99, 794.48±84.70, and 780.93±77.85 minutes, respectively. Linear trend analyses for device-measured physical activity indicated no significant differences among groups in LPA (F= 2.43, p= 0.12) and MVPA (F= 0.40, p= 0.52), nor did groups differ in percentage of time spent in MVPA and LPA (not reported). Therefore, linear trend analyses did not indicate a significant change in levels of device-measured physical activity with increased symptom cluster severity group status. GLTEQ HCS data were available for 202 participants. Linear trend analyses indicated significant differences in GLTEQ HCS (F= 4.90, p= 0.03), whereby mean levels of health-promoting physical activity declined across mild, moderate and severe symptom cluster groups. Specifically, linear trend analyses indicated a significant difference in self-reported physical activity wherein mean physical activity was lowest among the severe symptom cluster group (13.94±23.38), intermediate among the moderate symptom cluster group (22.63±27.08), and highest among the mild symptom cluster group (25.07±24.84).
Table 2.
Results of a Linear Trend Analysis Comparing Differences in Outcomes Across Three Levels of Symptom Cluster Severity
| Variable (n) | Mild Symptom Cluster (n=96) |
Moderate Symptom Cluster (n=73) |
Severe Symptom Cluster (n=36) |
F Value |
P Value |
|---|---|---|---|---|---|
| Physical Activity Variables | |||||
| Accel LPA (184) | 295.14±77.78 | 298.74±84.14 | 268.64±83.66 | 2.43 | 0.12 |
| Accel MVPA (184) | 20.92±17.89 | 21.26±22.77 | 18.40±18.84 | 0.40 | 0.52 |
| GLTEQ HCS (202) | 25.07±24.84 | 22.63±27.08 | 13.94±23.38 | 4.90 | 0.03 |
| Social Cognitive Variables | |||||
| SPS (186) | 19.03±2.90 | 18.13±2.91 | 15.61±3.00 | 31.43 | 0.001 |
| EXSE (186) | 74.32±31.22 | 57.26±35.63 | 41.21±33.94 | 22.55 | 0.001 |
| BARSE (185) | 54.19±23.81 | 46.80±27.09 | 36.03±26.91 | 11.48 | 0.001 |
| MOEES (187) | 59.94±7.37 | 58.41±7.19 | 56.03±6.49 | 6.98 | 0.009 |
| EGS (183) | 23.17±10.50 | 23.97±10.63 | 19.55±11.35 | 2.62 | 0.11 |
| EPS (186) | 25.16±8.06 | 24.18±9.41 | 23.63±8.40 | 0.74 | 0.39 |
| LL-FDI Functional Sum (184) | 63.18±10.90 | 56.91±12.92 | 47.97±14.68 | 34.41 | 0.001 |
Note. Values represent mean ± standard deviation; Accel = accelerometer; LPA = Light Physical Activity; MVPA = Moderate to Vigorous Physical Activity; GLTEQ HCS = Godin Leisure Time Exercise Questionnaire Health Contribution Score; SPS = Social Provision Scale; EXSE = Exercise Self-Efficacy Scale; BARSE = Barriers for Self-Efficacy Scale; MOEES = Multidimensional Outcome Expectations for Exercise Scale; EGS = Exercise Goal Setting Scale; EPS = Exercise Plans Scale; LLF-DI= Late-Life Function and Disability Instrument
SCT Variables
SCT variables are presented in Table 2. SCT variables were returned as part of the take-home packet, and therefore the number of participants with valid data per measure ranged between 183 and 186 participants. Linear trend analyses indicated significant differences in social support (SPS; F= 31.43, p= 0.001), exercise self-efficacy (EXSE; F= 22.55, p= 0.001), barriers self-efficacy (BARSE; F= 11.48, p= 0.001), outcome expectations (MOEES; F= 6.98, p= 0.009), and functional limitations (LL-FDI; F= 34.41, p= 0.001). All differences were in the expected direction such that scores declined across groups by symptom cluster severity (i.e., lower self-efficacy with worse symptom cluster severity). Linear trend analyses indicated no significant differences in planning (EGS; F= 2.62, p= 0.11) or planning (EPS; F= 0.74, p= 0.39).
Bivariate Correlations between Physical Activity and SCT Variables
The results from Spearman’s rho correlation analyses examining the association between physical activity and SCT variables by symptom cluster severity group are provided in Table 3. This report is focused on associations of moderate strength or greater (i.e., 0.3 or above) because such effect sizes are most likely to be replicable and clinically meaningful.
Table 3.
Correlations Among Social Cognitive Theory Variables and Physical Activity
| Mild Symptom Cluster | Moderate Symptom | Severe Symptom | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Group (n=83) | Cluster Group (n=65) | Cluster Group (n=30) | |||||||
| Measure | 1 | 2 | 3 | 1 | 2 | 3 | 1 | 2 | 3 |
| 1. Accel LPA | – | – | – | ||||||
| 2. Accel MVPA | .23* | – | .18 | – | .48** | – | |||
| 3. GLTEQ HCS | .02 | .22* | – | .20 | .26* | – | .43* | .15 | – |
| 4. SPS | .02 | .22* | .18 | .22 | .003 | .30* | .39* | .13 | .20 |
| 5. EXSE | −.02 | .399*** | .30** | .07 | .46*** | .60*** | .37* | .33 | .18 |
| 6. BARSE | −.01 | .16 | .23* | .03 | .16 | .54*** | .08 | .42* | .08 |
| 7. MOEES | −.02 | .05 | .25* | −.01 | .12 | .21 | .12 | .37* | .19 |
| 8. EGS | .16 | .13 | .46*** | −.07 | .33** | .40** | .15 | .15 | .27 |
| 9. EPS | −.03 | .12 | .40*** | −.09 | .13 | .21 | .09 | .07 | .37* |
| 10. LL-FDI Functional Sum | −.11 | .55*** | .31** | .02 | .51*** | .36** | .34 | .50** | .41* |
Note. Accel = accelerometer; LPA = Light Physical Activity; MVPA = Moderate to Vigorous Physical Activity; GLTEQ HCS = Godin Leisure Time Exercise Questionnaire Health Contribution Score; SPS = Social Provision Scale; EXSE = Exercise Self-Efficacy Scale; BARSE = Barriers for Self-Efficacy Scale; MOEES = Multidimensional Outcome Expectations for Exercise Scale; EGS = Exercise Goal Setting Scale; EPS = Exercise Plans Scale; LLF-DI= Late-Life Function and Disability Instrument
p <.001
p <.01
p <.05
Among the mild symptom cluster group, device-measured MVPA was moderately associated with exercise self-efficacy (EXSE; ρ= 0.39) and strongly associated with functional limitation (LL-FDI; ρ= 0.55), whereas LPA was not moderately or strongly associated with any SCT variables. GLTEQ HCS was moderately associated with exercise self-efficacy (EXSE; ρ= 0.30), goal setting (EGS; ρ= 0.46), planning (EPS; ρ= 0.40), and functional limitation (LL-FDI; ρ= 0.31). Among the moderate symptom cluster group, device-measured MVPA was moderately associated with exercise self-efficacy (EXSE; ρ= 0.46) and goal setting (EGS; ρ= 0.33) and strongly associated with functional limitation (LL-FDI; ρ= 0.51), whereas LPA was associated with any SCT variables. GLTEQ HCS was moderately associated with social support (SPS; ρ= 0.30), goal setting (EGS; ρ= 0.40), and functional limitations (LL-FDI; ρ= 0.36) and strongly associated with exercise self-efficacy (EXSE; ρ= 0.60) and barriers self-efficacy (BARSE; ρ= 0.54). Among the severe symptom cluster group, device-measured MVPA was moderately associated with exercise self-efficacy (EXSE (ρ= 0.33), barriers self-efficacy (BARSE; ρ= 0.42), and outcome expectations (MOEES; ρ= 0.37) and strongly associated with functional limitations (LL-FDI; ρ= 0.50). Device-measured LPA was moderately associated with social support (SPS; ρ= 0.39), exercise self-efficacy (EXSE; ρ= 0.37), and functional limitations (LL-FDI; ρ= 0.34). GLTEQ HCS was moderately associated with planning (EPS; ρ= 0.37) and functional limitations (LL-FDI; ρ= 0.41).
Multiple Linear Regression
Multiple linear regression analyses were conducted for each symptom cluster severity group which involved seven separate analyses regressing SCT variables on physical activity measures to identify which SCT variables were independent correlates of physical activity behavior (Table 4). Only significant variables from the bivariate correlation analyses were included and a stepwise approach was utilized as it is a robust technique for handling multicollinearity between SCT variables.
Table 4.
Results of Multiple Linear Regression Analyses with Stepwise Entry of Variables for Identifying Independent Social Cognitive Correlates of Physical Activity Within Each Symptom Cluster Severity Group
| Step | Variable | B | β | t | p |
|---|---|---|---|---|---|
| Mild Svmotom Cluster | |||||
| MVPA | |||||
| Step 1 | LL-FDI | .91 | .51 | 5.37 | .001 |
| Note: R2=.26 | |||||
| GLTEQ HCS | |||||
| Step 1 | EGS | .80 | .36 | 3.44 | .001 |
| Note: R2=.13 | |||||
| Moderate Svmotom Cluster | |||||
| MVPA | |||||
| Step 1 | EXSE | .21 | .37 | 3.19 | .002 |
| Step 2 | EXSE | .16 | .28 | 2.32 | .02 |
| EGS | .51 | .27 | 2.20 | .03 | |
| Note: R2=.14 Step 1, R2=.20 Step 2 | |||||
| GLTEQ HCS | |||||
| Step 1 | BARSE | .53 | .52 | 4.86 | .001 |
| Step 2 | BARSE | .47 | .47 | 4.42 | .001 |
| SPS | 2.31 | .24 | 2.26 | .03 | |
| Step 3 | BARSE | .26 | .25 | 1.71 | .09 |
| SPS | 2.32 | .24 | 2.32 | .02 | |
| EXES | .23 | .30 | 2.08 | .04 | |
| Note: R2=.27 Step 1, R2=.33 Step 2, R2=.37 Step 3 | |||||
| Severe Svmotom Cluster | |||||
| LPA | |||||
| Step 1 | SPS | 11.61 | .39 | 2.19 | .04 |
| Note: R2=.15 | |||||
| MVPA | |||||
| Step 1 | LL-FDI | .75 | .57 | 3.61 | .001 |
| Note: R2=.33 | |||||
| GLTEQ-HCS | |||||
| N/A | |||||
Note. LPA = Light Physical Activity; MVPA = Moderate to Vigorous Physical Activity; GLTEQ HCS = Godin Leisure Time Exercise Questionnaire Health Contribution Score; SPS = Social Provision Scale; EXSE = Exercise Self-Efficacy Scale; BARSE = Barriers for Self-Efficacy Scale; EGS = Exercise Goal Setting Scale; LL-FDI= Late-Life Function and Disability Instrument
Among the mild symptom cluster group, the results of the regression indicated that functional limitations (LL-FDI) explained 26% of the variance in MVPA (R =.26, F(1,81)=28.85, p<.001). Regarding GLTEQ HCS, the results of the regression indicated that goal setting (EGS) explained 13% of the variance (R2=.13, F(1,81)=11.80, p<.001).
Among the moderate symptom cluster group, two predictors were identified that explained variance in MVPA. In Step 1, exercise self-efficacy (EXSE) explained 14% of variance in MVPA (R2 =.14, F(1,63)=10.18, p=.002) and addition of goal setting (EGS) in Step 2 accounted for 20% of variance in MVPA (R2 =.20, F(2,62)=7.82, p<.001). Regarding GLTEQ HCS, three predictors were identified with barriers self-efficacy (BARSE) explaining 27% of variance in Step 1 (R2 =.27, F(1,63)=23.64, p<.001). Step 2 included barriers self-efficacy (BARSE) and social support (SPS) explaining 33% of variance in GLTEQ HCS (R2 =.33, F(2,62)=15.13, p<.001). Lastly, in Step 3 exercise self-efficacy (EXSE) was added in the model accounting for 37% of the variance in GLTEQ HCS (R2 =.37, F(3,61)=12.08, p<.001) and notably the inclusion of exercise self-efficacy (EXSE) attenuated the significant contribution of barriers self-efficacy (BARSE) in the overall model.
Among the severe symptom cluster group, the results of the regression indicated that social support (SPS) explained 15% of the variance (R2 =.15, F(1,27)=4.81, p=.04) in LPA. Regarding MVPA, the results of the regression indicated that functional limitations (LL-FDI) explained 33% of the variance (R2 =.33, F(1,27)=13.04, p<.001). Lastly, there were no significant SCT correlates of GLTEQ HCS.
Discussion
This study provides a comprehensive overview regarding symptom cluster groups, physical activity and SCT variables. Such research is particularly opportune given that symptom cluster group severity is associated with meaningful outcomes such as quality of life in persons with MS.7,8,19 Though physical activity is an evidence-based approach for improving symptoms when examined independently,37,38 there is a need for interventions focusing on established SCT approaches for changing physical activity that might account for symptom cluster severity. Such approaches may strategically focus on increasing positive outcome expectations related to symptoms, guidance regarding exercise planning, identification of social supports for exercise, and increasing self-efficacy for exercise among individuals with co-occurring severe symptoms.
The primary results from this study indicated that self-reported physical activity and levels of SCT variables decreased with worsening symptom cluster groups. Participants in the group with severe co-occurring depression, anxiety, fatigue, and sleep problems reported lower levels of physical activity and social support, self-efficacy, outcome expectations, and functional status than participants in the moderate or mild symptom cluster groups. Physical activity and SCT variables were significantly associated in all groups; however, possible meaningful differences were noted between symptom cluster severity groups such that there were fewer moderate or large associations between self-reported physical activity and SCT variables in the severe symptom cluster group than mild and moderate groups.
Physical activity was a primary variable of interest in the current study based on established benefits of health promoting physical activity in persons with MS, yet adults with MS engage in low rates of health-promoting physical activity.37,38 We observed a significant linear trend for declines in physical activity by symptom cluster severity group, wherein participants in the moderate and severe symptom cluster groups reported less physical activity than participants in the mild symptom cluster group. These results are consistent with previous research examining associations between physical activity and (i) number of symptoms, (ii) worsening of symptoms, and (iii) symptom cluster group severity.9,39 Our results indicate that symptoms are associated with physical activity, but it is unclear if worsening of symptoms are driving reduced physical activity or if reduced physical activity is yielding worsening symptoms. We underscore the critical need for longitudinal data collection that is initiated early in MS disease course to help elucidate the causal mechanism underlying the relationship between physical activity and symptoms.
This study further investigated if SCT variables, commonly associated with physical activity behavior in persons with MS, differed by symptom cluster severity groups. We observed a significant linear trend for declines in social support, self-efficacy, outcome expectations, and functional limitations, wherein participants in the moderate and severe symptom cluster groups reported less social support, self-efficacy, outcomes expectations and functional status when compared with participants in the mild symptom cluster group. These results are striking given preliminary research highlighting self-efficacy as an SCT-variable of interest partially accounting for the relationship between symptoms and physical activity.40 The current study provides a more comprehensive picture by incorporating additional SCT variables that are associated with physical activity behavior and effects of physical activity behavioral interventions in persons with MS.15-17,41 Future interventions applying behavior change techniques based on SCT for promoting physical activity in persons with MS may require greater support based on symptom cluster severity.
The current study included subsequent analyses that examined whether the associations between SCT variables and physical activity differed by symptom cluster severity groups. Across all symptom cluster groups, device-measured MVPA was strongly associated with self-reported functional status (i.e., impediments). Alternatively, self-reported health-promoting physical activity (i.e., GLTEQ HCS) was strongly associated with self-efficacy exclusively in the moderate symptom cluster group. There were differential associations of moderate magnitude across groups, such that three SCT variables were associated with physical activity in the mild group (i.e., self-efficacy, goal setting and planning), two in the moderate group (i.e., social support and goal setting), and four in the severe group (i.e., self efficacy, outcome expectations, planning, and social support). One previous study examining inactive adults with MS reported that self-efficacy was the only significant variable associated with device-measured physical activity,16 whereas another study of inactive adults with MS reported that goal setting was the primary SCT variables associated with self-reported physical activity.42 Such differences in associations between SCT variables and measures of physical activity may be the nature of the SCT measures; specifically measurements of self-efficacy assess individual perceptions whereas measurement of goal setting is often an assessment of current behavior and intentions. The current study provides further precision regarding recurrent associations between SCT variables and various measures of physical activity that may guide further research illuminating factors underlying inconsistencies between self-reported and device-measured physical activity.
The current study involved a secondary analysis with inherent limitations given the primary research question was not explicitly to examine the associations between physical activity, SCT variables, and symptom cluster groups. Specifically, focal studies examining these associations longitudinally would be beneficial, as the directionally of associations cannot be surmised from data collected using the present cross-sectional design. This study focused on four prevalent self-reported symptoms among persons with MS, however additional research is warranted examining other symptoms of interest such as mobility disability and cognitive dysfunction. There was no association between device-measured physical activity and symptom cluster group in this sample, though previous research has established that device-measured and self-reported physical activity are strongly correlated in persons with MS.25 Further focal research is warranted regarding the utility of device-measured physical activity measurement in persons with MS given growing concerns across research groups regarding adherence to protocols and validity with increased walking impairment.43
Conclusion
Results from this study indicated that physical activity, social support, self-efficacy, outcome expectations, and impediments are negatively associated with symptom cluster group severity. Associations between physical activity and SCT variables differ by symptom cluster groups. In the severe and moderate symptom cluster groups, physical activity was associated with both individual-level and environmental factors, whereas in the mild symptom cluster groups, physical activity was only associated with individual-level factors. Future research is warranted examining the use of various theory-based strategies for promoting physical activity as an approach for improving symptom clusters in persons with MS.
Funding:
Funding for this study was provided by the National Multiple Sclerosis Society [CA-1708-29059], the Eunice Kennedy Shriver National Institute Of Child Health & Human Development of the National Institutes of Health [F32HD101214; F31HD097903] and the National Heart, Lung, and Blood Institute of the National Institutes of Health [T32HL105349].
Footnotes
Conflict of Interest: The authors report no conflicts of interest.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Wallin MT, Culpepper WJ, Campbell JD, et al. The prevalence of MS in the United States: A population-based estimate using health claims data. Neurology. 2019;92(10):e1029–e1040. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Krupp L Fatigue is intrinsic to multiple sclerosis (MS) and is the most commonly reported symptom of the disease. Mult Scler. 2006;12(4):367–368. [DOI] [PubMed] [Google Scholar]
- 3.Marrie RA, Reingold S, Cohen J, et al. The incidence and prevalence of psychiatric disorders in multiple sclerosis: A systematic review. Mult Scler. 2015;21(3):305–317. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Sakkas GK, Giannaki CD, Karatzaferi C, Manconi M. Sleep Abnormalities in Multiple Sclerosis. Curr Treat Options Neurol. 2019;21(1):4. [DOI] [PubMed] [Google Scholar]
- 5.Dodd MJ, Miaskowski C, Lee KA. Occurrence of symptom clusters. JNCI Monographs. 2004;2004(32):76–78. [DOI] [PubMed] [Google Scholar]
- 6.Motl RW, McAuley E. Symptom cluster as a predictor of physical activity in multiple sclerosis: preliminary evidence. J Pain Symptom Manage. 2009;38(2):270–280. [DOI] [PubMed] [Google Scholar]
- 7.Motl RW, McAuley E. Symptom cluster and quality of life: preliminary evidence in multiple sclerosis. J Neurosci Nurs. 2010;42(4):212. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Motl RW, Suh Y, Weikert M. Symptom cluster and quality of life in multiple sclerosis. J Pain Symptom Manage. 2010;39(6):1025–1032. [DOI] [PubMed] [Google Scholar]
- 9.Motl RW, Weikert M, Suh Y, Dlugonski D. Symptom cluster and physical activity in relapsing-remitting multiple sclerosis. Res Nurs Health. 2010;33(5):398–412. [DOI] [PubMed] [Google Scholar]
- 10.Newland PK, Fearing A, Riley M, Neath A. Symptom clusters in women with relapsing-remitting multiple sclerosis. J Neurosci Nurs. 2012;44(2):66–71. [DOI] [PubMed] [Google Scholar]
- 11.Shahrbanian S, Duquette P, Kuspinar A, Mayo NE. Contribution of symptom clusters to multiple sclerosis consequences. Qual Life Res. 2015;24(3):617–629. [DOI] [PubMed] [Google Scholar]
- 12.Lenz ER, Pugh LC. The theory of unpleasant symptoms. Middle range theory for nursing. 2003:69–90. [DOI] [PubMed] [Google Scholar]
- 13.Motl RW, Pekmezi D, Wingo BC. Promotion of physical activity and exercise in multiple sclerosis: Importance of behavioral science and theory. Mult Scler J Exp Transl Clin 2018;4(3):2055217318786745. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Bandura A Health promotion by social cognitive means. Health Educ Behav. 2004;31(2):143–64. [DOI] [PubMed] [Google Scholar]
- 15.Suh Y, Weikert M, Dlugonski D, Balantrapu S, Motl RW. Social cognitive variables as correlates of physical activity in persons with multiple sclerosis: findings from a longitudinal, observational study. Behav Med. 2011;37(3):87–94. [DOI] [PubMed] [Google Scholar]
- 16.Suh Y, Weikert M, Dlugonski D, Sandroff B, Motl RW. Social cognitive correlates of physical activity: findings from a cross-sectional study of adults with relapsing-remitting multiple sclerosis. J Phys Act Health. 2011;8(5):626–635. [DOI] [PubMed] [Google Scholar]
- 17.Uszynski MK, Casey B, Hayes S, et al. Social cognitive theory correlates of physical activity in inactive adults with multiple sclerosis. Int J MS Care. 2018;20(3):129–135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Casey B, Uszynski M, Hayes S, Motl R, Gallagher S, Coote S. Do multiple sclerosis symptoms moderate the relationship between self-efficacy and physical activity in people with multiple sclerosis? Rehabil Psychol. 2018;63(1):104–110. [DOI] [PubMed] [Google Scholar]
- 19.Silveira SL, Cederberg KL, Jeng B, et al. Symptom clusters and quality of life in persons with multiple sclerosis across the lifespan. Qual Life Res. 2020:1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Krupp LB, LaRocca NG, Muir-Nash J, Steinberg AD. The fatigue severity scale: application to patients with multiple sclerosis and systemic lupus erythematosus. Arch Neurol. 1989;46(10):1121–1123. [DOI] [PubMed] [Google Scholar]
- 21.Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. 1983;67(6):361–370. [DOI] [PubMed] [Google Scholar]
- 22.Buysse DJ, Reynolds CF III, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28(2):193–213. [DOI] [PubMed] [Google Scholar]
- 23.Sarsted M, Mooi E. Cluster analysis. In Wainrib BR (Ed.), A concise guide to market research. The process, data, and methods using IBM SPSS statistics (pp. 274–324). New York: Springer; 2014. [Google Scholar]
- 24.Shephard R Godin leisure-time exercise questionnaire. Med Sci Sports Exerc. 1997;29(6):S36–S38. [Google Scholar]
- 25.Motl RW, McAuley E, Snook EM, Scott JA. Validity of physical activity measures in ambulatory individuals with multiple sclerosis. Disabil Rehabil. 2006;28(18):1151–1156. [DOI] [PubMed] [Google Scholar]
- 26.Motl RW, Bollaert RE, Sandroff BM. Validation of the Godin Leisure-Time Exercise Questionnaire classification coding system using accelerometry in multiple sclerosis. Rehabil Psychol. 2018;63(1):77. [DOI] [PubMed] [Google Scholar]
- 27.Sandroff BM, Motl RW, Suh Y. Accelerometer output and its association with energy expenditure in persons with multiple sclerosis. J Rehabil Res Dev. 2012;49(3). [DOI] [PubMed] [Google Scholar]
- 28.Chiu C-Y, Motl RW, Ditchman N. Validation of the Social Provisions Scale in people with multiple sclerosis. Rehabil Psychol. 2016;61(3):297. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.McAuley E The role of efficacy cognitions in the prediction of exercise behavior in middle-aged adults. J Behav Med. 1992;15(1):65–88. [DOI] [PubMed] [Google Scholar]
- 30.McAuley E. Self-efficacy and the maintenance of exercise participation in older adults. J Behav Med. 1993;16(1):103–113. [DOI] [PubMed] [Google Scholar]
- 31.McAuley E, Motl RW, White SM, Wójcicki TR. Validation of the multidimensional outcome expectations for exercise scale in ambulatory, symptom-free persons with multiple sclerosis. Arch Phys Med Rehabil. 2010;91(1):100–105. [DOI] [PubMed] [Google Scholar]
- 32.Rovniak LS, Anderson ES, Winett RA, Stephens RS. Social cognitive determinants of physical activity in young adults: a prospective structural equation analysis. Ann Behav Med. 2002;24(2):149–156. [DOI] [PubMed] [Google Scholar]
- 33.McAuley E, Konopack JF, Motl RW, Rosengren K, Morris KS. Measuring disability and function in older women: psychometric properties of the late-life function and disability instrument. J Gerontol A Biol Sci Med Sci. 2005;60(7):901–909. [DOI] [PubMed] [Google Scholar]
- 34.Learmonth YC, Motl RW, Sandroff BM, Pula JH, Cadavid D. Validation of patient determined disease steps (PDDS) scale scores in persons with multiple sclerosis. BMC Neurol. 2013;13(1):37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Dixon WJ. Simplified estimation from censored normal samples. Ann Math Stat. 1960:385–391. [Google Scholar]
- 36.Cohen J. Statistical power analysis for the behavioral sciences. Academic press; 2013. [Google Scholar]
- 37.Motl RW, Sandroff BM. Benefits of exercise training in multiple sclerosis. Curr Neurol Neurosci. 2015;15(9):62. [DOI] [PubMed] [Google Scholar]
- 38.Motl RW, Sandroff BM, Kwakkel G, et al. Exercise in patients with multiple sclerosis. Lancet Neurol. 2017;16(10):848–856. [DOI] [PubMed] [Google Scholar]
- 39.Motl RW, Arnett PA, Smith MM, Barwick FH, Ahlstrom B, Stover EJ. Worsening of symptoms is associated with lower physical activity levels in individuals with multiple sclerosis. Mult Scler. 2008;14(1):140–142. [DOI] [PubMed] [Google Scholar]
- 40.Motl RW, Snook EM, McAuley E, Gliottoni RC. Symptoms, self-efficacy, and physical activity among individuals with multiple sclerosis. Res Nurs Health. 2006;29(6):597–606. [DOI] [PubMed] [Google Scholar]
- 41.Silveira SL, Motl RW. Do Social Cognitive Theory constructs explain response heterogeneity with a physical activity behavioral intervention in multiple sclerosis? Contemp Clin Trials Commun. 2019;15:100366. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Dlugonski D, Wójcicki TR, McAuley E, Motl RW. Social cognitive correlates of physical activity in inactive adults with multiple sclerosis. Int J Rehabil Res. 2011;34(2):115–120. [DOI] [PubMed] [Google Scholar]
- 43.Coote S, O'Dwyer C. Comparative Validity of Accelerometer-Based Measures of Physical Activity for People With Multiple Sclerosis. Arch Phys Med Rehabil. 2012;93(11):2022–2028. [DOI] [PubMed] [Google Scholar]