ORIGINAL ARTICLE
Mobility Limitations and Fall-Related Factors Contribute to the Reduced Health-Related Quality of Life in Older Adults With Chronic Musculoskeletal Pain Brendon Stubbs, MSc, MCSP*; Pat Schofield, PhD*; Sandhi Patchay, PhD† *School of Health and Social Care, University of Greenwich, London; †School of Psychology, University of Greenwich, London, U.K.
& Abstract Purpose: To investigate (1) the prevalence of chronic musculoskeletal pain (CMP) among a sample of communitydwelling older adults and (2) health-related quality of life (HRQOL) in people with CMP, particularly the association with mobility limitations and falls-related factors. Method: Overall, 295 (response rate 73.5%) communitydwelling older adults were recruited across 10 sites. CMP was assessed using recognized criteria. In the sample of people with CMP, a hierarchical multiple regression analysis was conducted with HRQOL as the dependent variable and a number of independent variables were then inserted into the model. After controlling for demographic and medical variables, mobility (timed up and go (TUG), walking aid use, sedentary behavior) and fall-related factors (falls history, balance confidence, concerns about consequences of falling) were inserted into the model at the second step and changes in adjusted R2 noted.
Address correspondence and request to: Brendon Stubbs, School of Health and Social Care, Avery Hill Road, Eltham, London SE9 2UG, U.K. E-mail: [email protected]. Disclosures: SP and PS have no conflict of interest to declare in relation to this work. BS has no direct conflict of interest. However, BS is supported by a vice chancellors scholarship, but this did not influence the study at any stage nor the decision to publish. Submitted: May 19, 2013; Revision accepted: September 22, 2014 DOI. 10.1111/papr.12264
© 2014 World Institute of Pain, 1530-7085/16/$15.00 Pain Practice, Volume 16, Issue 1, 2016 80–89
Results: Within our sample of older adults, 52% had CMP (154/295). Compared to the group without CMP of similar age (n = 141), those with CMP had reduced HRQOL and profound mobility limitations and more falls risk factors (P < 0.001). The mobility and falls explanatory variables increased the variance explained within HRQOL from 14% to 36% (adjusted R2 change 20%) in those with CMP. Sedentary behavior, pain interference, concerns about the consequences of falling, falls history, TUG scores, and balance confidence all remained significant predictors of HRQOL in the fully adjusted model in the CMP sample. Conclusion: Older adults with CMP have pronounced mobility limitations and increased falls risk factors, and these are associated with a marked reduction in HRQOL. Future prospective research is required to build on this crosssectional study. & Key Words: musculoskeletal pain, pain in the elderly, mobility limitations, fall risk, health-related quality of life
INTRODUCTION Life expectancy is rising across the world, while at the same time, people are experiencing more years with disability.1 One of the largest contributors to years lived with disability, particularly in the Western world, is chronic musculoskeletal conditions (eg, osteoarthritis, chronic low back pain).1 One common sequela of chronic musculoskeletal conditions is chronic musculoskeletal pain (pain lasting 3 or more months).2
Chronic Musculoskeletal Pain and Quality of Life 81
Chronic musculoskeletal pain affects approximately 50% of community-dwelling older adults,3,4 and it remains one of the most common reasons that older adults seek medical attention.5 The financial impact of chronic musculoskeletal pain is also profound. For instance, in 2010, it was estimated that the costs associated with chronic pain in the United States were between 560 and 635 billion dollars per annum.6 Previous research has demonstrated that musculoskeletal pain is associated with disability,7 mobility limitations,8,9 and functional decline10 in older adults. In addition, chronic musculoskeletal pain is related to an increased risk of falls,11,12 lower levels of physical activity,13 and also fractures.14 Such mobility difficulties are pertinent because physical activity is an important nonpharmacological approach to manage chronic musculoskeletal pain.13 More recently, research has started to consider the influence of chronic musculoskeletal pain on “psychological concerns related to falls” such as balance confidence and excessive concerns about the adverse consequences of falling over.15,16 In light of these factors, it is not surprising that older adults with chronic musculoskeletal pain often experience a reduction in their health-related quality of life (HRQOL).14 However, to the best of our knowledge, no author has previously investigated the impact of mobility limitations and fall-related factors on the HRQOL of older adults with chronic musculoskeletal pain. HRQOL is an important patient-reported outcome and is a measure of the impact an illness has upon the functional health status as perceived by the patients themselves.17,18 In addition, HRQOL is an important outcome among policymakers, researchers, and clinicians.19 Indeed, in response to the increasing age, a number of policies have been developed to specifically promote HRQOL in older age.20,21 A range of HRQOL measures currently exist, but one measure, the European Quality of Life Instrument (EuroQoL EQ-5D),22 is commonly used in clinical practice and research.23 The EQ-5D asks the participant to rate their overall perceived health state from 0 (worst imaginable health state possible) to 100 (best imaginable health state possible), and this provides a summary of their HRQOL.23 The EQ-5D is advocated as a measure to ascertain HRQOL in older adults in international guidelines.24 For this study, we defined HRQOL as the summary score from the EQ-5D (ie, their own rating of their overall health state). Despite the aforementioned reasons for concern, research investigating the contribution of mobility limitations, falls, and psychological concerns related to
falls and HRQOL in older adults with chronic musculoskeletal pain is sparse. Identifying determinants for HRQOL is important so that clinicians can seek to develop appropriate interventions. From a clinical perspective, it seems plausible that experiencing mobility limitations (eg, difficulty with balance), having a heightened risk of falls, and more sedentary behavior (sitting for longer periods) could contribute to a reduced HRQOL. If this is true, then from a theoretical perspective, it seems that these mobility factors could impact the 2 key domains of current HRQOL identified by the International Classification of Functioning, Disability, and Health (ICF).25 First, mobility difficulties, if present, are key body tasks that may influence the older adults functioning.26 Second, mobility limitations and increased fall risk have the potential to affect the individual’s participation in wider society.26 With the global demographic changes and high numbers of older adults affected by chronic musculoskeletal pain,3,4 it is essential that research is conducted to investigate the prevalence and impact of chronic musculoskeletal pain on HRQOL in community-dwelling older adults. The purpose of our study was to 1) investigate the prevalence of chronic musculoskeletal pain among community-dwelling older adults in our sample and 2) investigate the impact of CMP on HRQOL and, in particular, the contribution of mobility limitations and falls-related factors. We hypothesized that mobility limitations (reduced lower limb function, sedentary behavior) and fall-related factors (including falls history and psychological concerns related to falls) would significantly contribute to a reduced HRQOL in older adults with chronic musculoskeletal pain.
METHOD Study Design and Participants A multisite cross-sectional study was conducted in the United Kingdom in 2013. Data were collected from 10 different sites (5 day centers, 2 sheltered housing schemes, and 3 community “clubs” for older adults) over an 8-month period. Permission to undertake the study was obtained from a scheme manager at each participating site, and only older adults deemed suitable by the manager were approached to take part. Inclusion criteria included the following: (1) community-dwelling older adults (60 years old and above); (2) mobile over 10 meters with or without a walking aid; and (3) able to understand written and verbal English. People were
Chronic Musculoskeletal Pain and Quality of Life 85
Table 4. Summary of Hierarchical Multiple Regression Analysis Investigating the Predictors of Health-Related Quality of Life (Dependent Variable) in Older Adults With Chronic Musculoskeletal Pain (n = 154) Step 1 Independent Variables Constant Age Gender Number of comorbidities Number of medications Duration of pain Mean BPI interference score Walking aid use Sedentary behavior TUG score History of falls CoF scale ABC scores
B
SE B
70.640 0.086 0.064 1.253 0.708 0.378 4.196
19.083 0.230 3.517 1.764 1.061 0.220 0.992
Step 2 b 0.031 0.001 0.072 0.069 0.142 0.368***
B
SE B
112.365 0.353 5.178 2.761 1.093 0.366 4.031 6.643 0.037 0.941 6.990 1.282 0.290
23.405 0.220 3.246 1.603 0.951 0.199 1.046 4.078 0.011 0.326 1.920 0.381 0.122
b 0.128 0.119 0.158 0.107 0.138 0.353*** 0.155 0.366*** 0.271** 0.285*** 0.330*** 0.296*
*P < 0.05; **P < 0.01, ***P < 0.001, B and SE (standard error) B = unstandardized coefficients, b = standardized beta coefficients. Independent variables entered into the model at step 1: age, gender, number of comorbidities, number of medications, duration of pain (mean years), and mean BPI (brief pain inventory) interference score. Independent variables entered into the model at step 2: all the variables at step 1 and walk aid use, sedentary behavior, TUG score (timed up and go), history of falls in past 12 months, CoF scale (consequences of falling scale), and ABC (activities balance confidence scale). HRQOL = dependent variable.
factors are associated with a reduced HRQOL in older adults with chronic musculoskeletal pain. Specifically, our study demonstrated that not only pain interference but also sedentary behavior, mobility limitations (measured by timed up and go scores), a history of falls, and increased concerns about the consequences of falling were all significant negative predictors of HRQOL. We also demonstrated that better balance confidence was positively associated with improved HRQOL. Approximately half (52%) of our sample reported experiencing chronic musculoskeletal pain according to recognized pain assessment guidelines.4,29 This is much in line with previous work and further underscores the extent of the problem in the expanding older adult community. For instance, in a recent population-based cohort study, Leveille et al.28 established that 64% experienced chronic musculoskeletal pain according to the criteria that we adopted (40% had multisite and 24% had single site pain). The present study also demonstrates that older adults with chronic musculoskeletal pain had a significantly lower HRQOL compared to a group of similar age and gender without chronic musculoskeletal pain. Despite the relative paucity of research specifically investigating HRQOL and its determinants in community-dwelling older adults with chronic musculoskeletal pain, this is not surprising as musculoskeletal pain is known to have a substantial impact on older adults.43 In fact, our study established that the chronic musculoskeletal pain group had significantly more mobility limitations, spent more time being sedentary, had more psychological concerns related to falls and actual falls than the comparison
group. This is similar with earlier work which has also highlighted such profound mobility deficits in those with chronic musculoskeletal pain. For instance, Karttunen et al.9 reported that older adults with musculoskeletal pain were significantly more likely to report mobility limitations according to the TUG. This finding was exemplified by Peraira et al.44 who found older adults with chronic pain had significantly poorer physical performance. Recently, in a large nationally representative sample, Patel et al.43 reported that those with multisite pain are at greatest risk of experiencing mobility limitations such as a slower gait speed. The authors found that up to 80% experienced difficulties undertaking fundamental activities of daily living. Eggermont et al.45 found that pain interference measured by the BPI interference subscale is a predictor of mobility limitations and onset of difficulties in ADL. Earlier work has also established that musculoskeletal pain may be predictive of functional decline and disability,10 and possibly an early marker for frailty syndrome.14 With this mounting evidence, that chronic musculoskeletal pain and pain interference causes mobility limitations, it seems important to identify those with chronic musculoskeletal pain and offer appropriate interventions.43 Much in line with Patel and colleagues,43 the older adults with chronic musculoskeletal pain in our sample also expressed more self-reported difficulties in their functional abilities. Specifically, our sample with chronic musculoskeletal pain experienced lower balance confidence and had higher concerns about the consequence of falling over. Collectively, these “psychological concerns related to falls” may, in part, explain why the
Chronic Musculoskeletal Pain and Quality of Life 83
individual’s perceived ability to undertake a range of functional activities of daily living (ADL) and not lose their balance or fall over. It consists of 16 questions in which the individual rates their confidence from 0% (= no confidence) to 100% (= complete confidence) undertaking each task.38 A mean overall score is calculated across the 16 items. The ABC is a reliable and valid tool to measure balance confidence in community-dwelling older adults.15,39,40 In addition, all participants completed the consequences of falling scale (CoF).41 The CoF is a 12-item scale that measures concerns about the consequences of falling among community-dwelling older adults. Previous research40 within the general older adult population has established that the ABC and CoF are important determinants of HRQOL. Data Analysis All data analyses were conducted using SPSS (version 20; IBM Corp, Armonk, NY, USA). Continuous data were assessed for normality with a visual inspection of PP plots and the calculation of skew and kurtosis to ensure normal ranges.42 Independent t-tests and chisquare tests were used to compare the continuous and categorical variables, respectively, between those with and without chronic musculoskeletal pain. When tests for normality and equality of variance were not satisfied, nonparametric equivalents were employed. The relationship between HRQOL (dependent variable) and demographic, medical, pain, mobility, and fall-related factors were assessed with Pearson’s product–moment correlation coefficient (Pearson’s r). Next, a hierarchical multiple regression was conducted with the mean HRQOL score as the dependent variable and independent variables being inserted into the model in 2 steps. Within the first step of the model, we inserted the following independent variables: demographic (mean age, gender), medical factors (mean number of comorbidities, mean number of medications), and pain factors (duration of pain (years), mean BPI interference subscale) into the model. Next, we inserted the mobility (walking aid use, TUG scores, sedentary behavior) and falls-related factor independent variables (history of falls, CoF and ABC scale) into the model. Changes in adjusted R2 were noted to investigate their unique contribution on the variance of the HRQOL.42 At each step of the model, we report the standardized beta coefficients to see the unique contribution of each independent variable in the fully adjusted model. Multicollinearity was assessed by calculation of the variation
inflation factor (VIF), and tolerance for each model ensuring this was within satisfactory ranges (VIF < 10 and not much higher than 1; and tolerance > 0.2).42 All analyses conducted were two-tailed, and significance was set at P < 0.05. Sample Size Calculation An a priori sample size calculation was conducted using G*Power software for the regression analyses. Based upon on an R2 increase with 12 predictors in the model, a power of 0.8, significance level at 0.05, and a medium effect size (F2 = 0.15), a total sample size of 127 was required in the chronic musculoskeletal pain group. Therefore, this study was adequately powered.
RESULTS Participant Characteristics In total, 295 older adults of a possible 401 agreed to take part in the study (response rate 73.5%). Most nonparticipants were not interested in taking part in the study (75/401, 18.7%), and a further minority (31/401, 7.7%) met 1 or more of the exclusion criteria. The mean age of the 295 participants was 77.5 years (8.1 years), 196 were female (66.4%), and 268 were Caucasian (90.8%). Prevalence and Impact of Chronic Musculoskeletal Pain Overall, 154 participants (52.2%) met the criteria for chronic musculoskeletal pain and 141 (47.8%) did not and formed the comparison group. Of those with chronic musculoskeletal pain, almost two-thirds (90/ 154, 58.4%) had pain across multiple sites. The most common primary sites of pain were the knee (n = 64/ 154, 41.6%), back (n = 36/154, 23.4%), or foot (n = 18/154, 11.7%). There was no statistically significant difference in the mean age or proportion of females in the chronic musculoskeletal pain group and the comparison group (Table 1). However, the chronic musculoskeletal pain group had a higher number of comorbidities and took more medications. Lastly, older adults with chronic musculoskeletal pain reported a significantly lower perceived overall HRQOL, compared to the comparison group. Full details are summarized in Table 1. As shown in Table 2, the chronic musculoskeletal pain group experienced pronounced mobility limitations and increased fall risk factors (all P < 0.001). For
84
STUBBS ET AL.
Table 1. Comparison of Demographic and Medical Factors and Quality of Life of the Group with Chronic Musculoskeletal Pain and the Comparison Group
Variable Age (years, SD) Female n (%) Live in warden accommodation n (%) Current smoker yes (%) Wear glasses yes (%) Number of comorbidities (SD) Number of medications (SD) Health-related quality of life How good or bad is your health? (0–100, SD) Chronic pain classification Duration years median (range) Single site n (%) Multisite n (%) BPI severity (SD) BPI interference (SD)
Comparison group (n = 141)
Chronic musculoskeletal pain (n = 154)
P value
76.6 ( 8.5) 95 (67.4) 42 (29.8)
78.3 ( 7.8) 101 (65.6) 72 (46.8)
0.08 0.80 < 0.001
15 105 3.9 3.6
0.30 0.22 < 0.001 < 0.001
11 (7.8) 77 (54.6) 2.8 ( 1.3) 2.9 ( 2) 79.8 ( 15.6)
(9.7) (68.2) ( 1.2) ( 2)
58.3 ( 20.6)
Variable
< 0.001
3 (0.4–50) 64 90 5.6 4.8
Table 2. Comparing the Mobility and Fall-Related Factors Between the Chronic Musculoskeletal Pain and Comparison Group
Variable Walking Aid use yes (%) In the past year have you had a fall? Yes (%) Sedentary behavior (hours per day, SD) Timed get up and go scores (sec, SD) Timed up and go > 13.5 seconds (%) Balance confidence (Total ABC, %) CoF scale (SD)
Chronic musculoskeletal pain (n = 154)
P value
97 (62.9%) 91 (59%)
< 0.001 < 0.001
7.9 ( 3.7)
11.0 ( 3.36)
< 0.001
10.9 ( 4.5)
14.6 ( 4.6)
< 0.001
26 (18.4)
72 (47.1)
< 0.001
71.3 ( 22)
48.3 ( 20.7)
< 0.001
37 (26.2%) 48 (34%)
25.7 5.9
31.8 5.3
Pearson r
Age (years) Female n (%) Number of comorbidities Number of medications Walking Aid use (%) History of falls (%) Sedentary behavior (IPAQ-SF) Timed get up and go scores (seconds) BPI pain severity BPI pain interference Balance confidence (ABC scale) Concerns about the consequences of falling (CoF scale)
P value
0.04 0.03 0.08 0.10 0.17 0.13 0.31 0.26
0.58 0.68 0.31 0.18 0.02 0.10 < 0.001 < 0.001
0.32 0.29 0.28 0.40
< < <
Mobility Limitations and Fall-Related Factors Contribute to the Reduced Health-Related Quality of Life in Older Adults With Chronic Musculoskeletal Pain Brendon Stubbs, MSc, MCSP*; Pat Schofield, PhD*; Sandhi Patchay, PhD† *School of Health and Social Care, University of Greenwich, London; †School of Psychology, University of Greenwich, London, U.K.
& Abstract Purpose: To investigate (1) the prevalence of chronic musculoskeletal pain (CMP) among a sample of communitydwelling older adults and (2) health-related quality of life (HRQOL) in people with CMP, particularly the association with mobility limitations and falls-related factors. Method: Overall, 295 (response rate 73.5%) communitydwelling older adults were recruited across 10 sites. CMP was assessed using recognized criteria. In the sample of people with CMP, a hierarchical multiple regression analysis was conducted with HRQOL as the dependent variable and a number of independent variables were then inserted into the model. After controlling for demographic and medical variables, mobility (timed up and go (TUG), walking aid use, sedentary behavior) and fall-related factors (falls history, balance confidence, concerns about consequences of falling) were inserted into the model at the second step and changes in adjusted R2 noted.
Address correspondence and request to: Brendon Stubbs, School of Health and Social Care, Avery Hill Road, Eltham, London SE9 2UG, U.K. E-mail: [email protected]. Disclosures: SP and PS have no conflict of interest to declare in relation to this work. BS has no direct conflict of interest. However, BS is supported by a vice chancellors scholarship, but this did not influence the study at any stage nor the decision to publish. Submitted: May 19, 2013; Revision accepted: September 22, 2014 DOI. 10.1111/papr.12264
© 2014 World Institute of Pain, 1530-7085/16/$15.00 Pain Practice, Volume 16, Issue 1, 2016 80–89
Results: Within our sample of older adults, 52% had CMP (154/295). Compared to the group without CMP of similar age (n = 141), those with CMP had reduced HRQOL and profound mobility limitations and more falls risk factors (P < 0.001). The mobility and falls explanatory variables increased the variance explained within HRQOL from 14% to 36% (adjusted R2 change 20%) in those with CMP. Sedentary behavior, pain interference, concerns about the consequences of falling, falls history, TUG scores, and balance confidence all remained significant predictors of HRQOL in the fully adjusted model in the CMP sample. Conclusion: Older adults with CMP have pronounced mobility limitations and increased falls risk factors, and these are associated with a marked reduction in HRQOL. Future prospective research is required to build on this crosssectional study. & Key Words: musculoskeletal pain, pain in the elderly, mobility limitations, fall risk, health-related quality of life
INTRODUCTION Life expectancy is rising across the world, while at the same time, people are experiencing more years with disability.1 One of the largest contributors to years lived with disability, particularly in the Western world, is chronic musculoskeletal conditions (eg, osteoarthritis, chronic low back pain).1 One common sequela of chronic musculoskeletal conditions is chronic musculoskeletal pain (pain lasting 3 or more months).2
Chronic Musculoskeletal Pain and Quality of Life 81
Chronic musculoskeletal pain affects approximately 50% of community-dwelling older adults,3,4 and it remains one of the most common reasons that older adults seek medical attention.5 The financial impact of chronic musculoskeletal pain is also profound. For instance, in 2010, it was estimated that the costs associated with chronic pain in the United States were between 560 and 635 billion dollars per annum.6 Previous research has demonstrated that musculoskeletal pain is associated with disability,7 mobility limitations,8,9 and functional decline10 in older adults. In addition, chronic musculoskeletal pain is related to an increased risk of falls,11,12 lower levels of physical activity,13 and also fractures.14 Such mobility difficulties are pertinent because physical activity is an important nonpharmacological approach to manage chronic musculoskeletal pain.13 More recently, research has started to consider the influence of chronic musculoskeletal pain on “psychological concerns related to falls” such as balance confidence and excessive concerns about the adverse consequences of falling over.15,16 In light of these factors, it is not surprising that older adults with chronic musculoskeletal pain often experience a reduction in their health-related quality of life (HRQOL).14 However, to the best of our knowledge, no author has previously investigated the impact of mobility limitations and fall-related factors on the HRQOL of older adults with chronic musculoskeletal pain. HRQOL is an important patient-reported outcome and is a measure of the impact an illness has upon the functional health status as perceived by the patients themselves.17,18 In addition, HRQOL is an important outcome among policymakers, researchers, and clinicians.19 Indeed, in response to the increasing age, a number of policies have been developed to specifically promote HRQOL in older age.20,21 A range of HRQOL measures currently exist, but one measure, the European Quality of Life Instrument (EuroQoL EQ-5D),22 is commonly used in clinical practice and research.23 The EQ-5D asks the participant to rate their overall perceived health state from 0 (worst imaginable health state possible) to 100 (best imaginable health state possible), and this provides a summary of their HRQOL.23 The EQ-5D is advocated as a measure to ascertain HRQOL in older adults in international guidelines.24 For this study, we defined HRQOL as the summary score from the EQ-5D (ie, their own rating of their overall health state). Despite the aforementioned reasons for concern, research investigating the contribution of mobility limitations, falls, and psychological concerns related to
falls and HRQOL in older adults with chronic musculoskeletal pain is sparse. Identifying determinants for HRQOL is important so that clinicians can seek to develop appropriate interventions. From a clinical perspective, it seems plausible that experiencing mobility limitations (eg, difficulty with balance), having a heightened risk of falls, and more sedentary behavior (sitting for longer periods) could contribute to a reduced HRQOL. If this is true, then from a theoretical perspective, it seems that these mobility factors could impact the 2 key domains of current HRQOL identified by the International Classification of Functioning, Disability, and Health (ICF).25 First, mobility difficulties, if present, are key body tasks that may influence the older adults functioning.26 Second, mobility limitations and increased fall risk have the potential to affect the individual’s participation in wider society.26 With the global demographic changes and high numbers of older adults affected by chronic musculoskeletal pain,3,4 it is essential that research is conducted to investigate the prevalence and impact of chronic musculoskeletal pain on HRQOL in community-dwelling older adults. The purpose of our study was to 1) investigate the prevalence of chronic musculoskeletal pain among community-dwelling older adults in our sample and 2) investigate the impact of CMP on HRQOL and, in particular, the contribution of mobility limitations and falls-related factors. We hypothesized that mobility limitations (reduced lower limb function, sedentary behavior) and fall-related factors (including falls history and psychological concerns related to falls) would significantly contribute to a reduced HRQOL in older adults with chronic musculoskeletal pain.
METHOD Study Design and Participants A multisite cross-sectional study was conducted in the United Kingdom in 2013. Data were collected from 10 different sites (5 day centers, 2 sheltered housing schemes, and 3 community “clubs” for older adults) over an 8-month period. Permission to undertake the study was obtained from a scheme manager at each participating site, and only older adults deemed suitable by the manager were approached to take part. Inclusion criteria included the following: (1) community-dwelling older adults (60 years old and above); (2) mobile over 10 meters with or without a walking aid; and (3) able to understand written and verbal English. People were
Chronic Musculoskeletal Pain and Quality of Life 85
Table 4. Summary of Hierarchical Multiple Regression Analysis Investigating the Predictors of Health-Related Quality of Life (Dependent Variable) in Older Adults With Chronic Musculoskeletal Pain (n = 154) Step 1 Independent Variables Constant Age Gender Number of comorbidities Number of medications Duration of pain Mean BPI interference score Walking aid use Sedentary behavior TUG score History of falls CoF scale ABC scores
B
SE B
70.640 0.086 0.064 1.253 0.708 0.378 4.196
19.083 0.230 3.517 1.764 1.061 0.220 0.992
Step 2 b 0.031 0.001 0.072 0.069 0.142 0.368***
B
SE B
112.365 0.353 5.178 2.761 1.093 0.366 4.031 6.643 0.037 0.941 6.990 1.282 0.290
23.405 0.220 3.246 1.603 0.951 0.199 1.046 4.078 0.011 0.326 1.920 0.381 0.122
b 0.128 0.119 0.158 0.107 0.138 0.353*** 0.155 0.366*** 0.271** 0.285*** 0.330*** 0.296*
*P < 0.05; **P < 0.01, ***P < 0.001, B and SE (standard error) B = unstandardized coefficients, b = standardized beta coefficients. Independent variables entered into the model at step 1: age, gender, number of comorbidities, number of medications, duration of pain (mean years), and mean BPI (brief pain inventory) interference score. Independent variables entered into the model at step 2: all the variables at step 1 and walk aid use, sedentary behavior, TUG score (timed up and go), history of falls in past 12 months, CoF scale (consequences of falling scale), and ABC (activities balance confidence scale). HRQOL = dependent variable.
factors are associated with a reduced HRQOL in older adults with chronic musculoskeletal pain. Specifically, our study demonstrated that not only pain interference but also sedentary behavior, mobility limitations (measured by timed up and go scores), a history of falls, and increased concerns about the consequences of falling were all significant negative predictors of HRQOL. We also demonstrated that better balance confidence was positively associated with improved HRQOL. Approximately half (52%) of our sample reported experiencing chronic musculoskeletal pain according to recognized pain assessment guidelines.4,29 This is much in line with previous work and further underscores the extent of the problem in the expanding older adult community. For instance, in a recent population-based cohort study, Leveille et al.28 established that 64% experienced chronic musculoskeletal pain according to the criteria that we adopted (40% had multisite and 24% had single site pain). The present study also demonstrates that older adults with chronic musculoskeletal pain had a significantly lower HRQOL compared to a group of similar age and gender without chronic musculoskeletal pain. Despite the relative paucity of research specifically investigating HRQOL and its determinants in community-dwelling older adults with chronic musculoskeletal pain, this is not surprising as musculoskeletal pain is known to have a substantial impact on older adults.43 In fact, our study established that the chronic musculoskeletal pain group had significantly more mobility limitations, spent more time being sedentary, had more psychological concerns related to falls and actual falls than the comparison
group. This is similar with earlier work which has also highlighted such profound mobility deficits in those with chronic musculoskeletal pain. For instance, Karttunen et al.9 reported that older adults with musculoskeletal pain were significantly more likely to report mobility limitations according to the TUG. This finding was exemplified by Peraira et al.44 who found older adults with chronic pain had significantly poorer physical performance. Recently, in a large nationally representative sample, Patel et al.43 reported that those with multisite pain are at greatest risk of experiencing mobility limitations such as a slower gait speed. The authors found that up to 80% experienced difficulties undertaking fundamental activities of daily living. Eggermont et al.45 found that pain interference measured by the BPI interference subscale is a predictor of mobility limitations and onset of difficulties in ADL. Earlier work has also established that musculoskeletal pain may be predictive of functional decline and disability,10 and possibly an early marker for frailty syndrome.14 With this mounting evidence, that chronic musculoskeletal pain and pain interference causes mobility limitations, it seems important to identify those with chronic musculoskeletal pain and offer appropriate interventions.43 Much in line with Patel and colleagues,43 the older adults with chronic musculoskeletal pain in our sample also expressed more self-reported difficulties in their functional abilities. Specifically, our sample with chronic musculoskeletal pain experienced lower balance confidence and had higher concerns about the consequence of falling over. Collectively, these “psychological concerns related to falls” may, in part, explain why the
Chronic Musculoskeletal Pain and Quality of Life 83
individual’s perceived ability to undertake a range of functional activities of daily living (ADL) and not lose their balance or fall over. It consists of 16 questions in which the individual rates their confidence from 0% (= no confidence) to 100% (= complete confidence) undertaking each task.38 A mean overall score is calculated across the 16 items. The ABC is a reliable and valid tool to measure balance confidence in community-dwelling older adults.15,39,40 In addition, all participants completed the consequences of falling scale (CoF).41 The CoF is a 12-item scale that measures concerns about the consequences of falling among community-dwelling older adults. Previous research40 within the general older adult population has established that the ABC and CoF are important determinants of HRQOL. Data Analysis All data analyses were conducted using SPSS (version 20; IBM Corp, Armonk, NY, USA). Continuous data were assessed for normality with a visual inspection of PP plots and the calculation of skew and kurtosis to ensure normal ranges.42 Independent t-tests and chisquare tests were used to compare the continuous and categorical variables, respectively, between those with and without chronic musculoskeletal pain. When tests for normality and equality of variance were not satisfied, nonparametric equivalents were employed. The relationship between HRQOL (dependent variable) and demographic, medical, pain, mobility, and fall-related factors were assessed with Pearson’s product–moment correlation coefficient (Pearson’s r). Next, a hierarchical multiple regression was conducted with the mean HRQOL score as the dependent variable and independent variables being inserted into the model in 2 steps. Within the first step of the model, we inserted the following independent variables: demographic (mean age, gender), medical factors (mean number of comorbidities, mean number of medications), and pain factors (duration of pain (years), mean BPI interference subscale) into the model. Next, we inserted the mobility (walking aid use, TUG scores, sedentary behavior) and falls-related factor independent variables (history of falls, CoF and ABC scale) into the model. Changes in adjusted R2 were noted to investigate their unique contribution on the variance of the HRQOL.42 At each step of the model, we report the standardized beta coefficients to see the unique contribution of each independent variable in the fully adjusted model. Multicollinearity was assessed by calculation of the variation
inflation factor (VIF), and tolerance for each model ensuring this was within satisfactory ranges (VIF < 10 and not much higher than 1; and tolerance > 0.2).42 All analyses conducted were two-tailed, and significance was set at P < 0.05. Sample Size Calculation An a priori sample size calculation was conducted using G*Power software for the regression analyses. Based upon on an R2 increase with 12 predictors in the model, a power of 0.8, significance level at 0.05, and a medium effect size (F2 = 0.15), a total sample size of 127 was required in the chronic musculoskeletal pain group. Therefore, this study was adequately powered.
RESULTS Participant Characteristics In total, 295 older adults of a possible 401 agreed to take part in the study (response rate 73.5%). Most nonparticipants were not interested in taking part in the study (75/401, 18.7%), and a further minority (31/401, 7.7%) met 1 or more of the exclusion criteria. The mean age of the 295 participants was 77.5 years (8.1 years), 196 were female (66.4%), and 268 were Caucasian (90.8%). Prevalence and Impact of Chronic Musculoskeletal Pain Overall, 154 participants (52.2%) met the criteria for chronic musculoskeletal pain and 141 (47.8%) did not and formed the comparison group. Of those with chronic musculoskeletal pain, almost two-thirds (90/ 154, 58.4%) had pain across multiple sites. The most common primary sites of pain were the knee (n = 64/ 154, 41.6%), back (n = 36/154, 23.4%), or foot (n = 18/154, 11.7%). There was no statistically significant difference in the mean age or proportion of females in the chronic musculoskeletal pain group and the comparison group (Table 1). However, the chronic musculoskeletal pain group had a higher number of comorbidities and took more medications. Lastly, older adults with chronic musculoskeletal pain reported a significantly lower perceived overall HRQOL, compared to the comparison group. Full details are summarized in Table 1. As shown in Table 2, the chronic musculoskeletal pain group experienced pronounced mobility limitations and increased fall risk factors (all P < 0.001). For
84
STUBBS ET AL.
Table 1. Comparison of Demographic and Medical Factors and Quality of Life of the Group with Chronic Musculoskeletal Pain and the Comparison Group
Variable Age (years, SD) Female n (%) Live in warden accommodation n (%) Current smoker yes (%) Wear glasses yes (%) Number of comorbidities (SD) Number of medications (SD) Health-related quality of life How good or bad is your health? (0–100, SD) Chronic pain classification Duration years median (range) Single site n (%) Multisite n (%) BPI severity (SD) BPI interference (SD)
Comparison group (n = 141)
Chronic musculoskeletal pain (n = 154)
P value
76.6 ( 8.5) 95 (67.4) 42 (29.8)
78.3 ( 7.8) 101 (65.6) 72 (46.8)
0.08 0.80 < 0.001
15 105 3.9 3.6
0.30 0.22 < 0.001 < 0.001
11 (7.8) 77 (54.6) 2.8 ( 1.3) 2.9 ( 2) 79.8 ( 15.6)
(9.7) (68.2) ( 1.2) ( 2)
58.3 ( 20.6)
Variable
< 0.001
3 (0.4–50) 64 90 5.6 4.8
Table 2. Comparing the Mobility and Fall-Related Factors Between the Chronic Musculoskeletal Pain and Comparison Group
Variable Walking Aid use yes (%) In the past year have you had a fall? Yes (%) Sedentary behavior (hours per day, SD) Timed get up and go scores (sec, SD) Timed up and go > 13.5 seconds (%) Balance confidence (Total ABC, %) CoF scale (SD)
Chronic musculoskeletal pain (n = 154)
P value
97 (62.9%) 91 (59%)
< 0.001 < 0.001
7.9 ( 3.7)
11.0 ( 3.36)
< 0.001
10.9 ( 4.5)
14.6 ( 4.6)
< 0.001
26 (18.4)
72 (47.1)
< 0.001
71.3 ( 22)
48.3 ( 20.7)
< 0.001
37 (26.2%) 48 (34%)
25.7 5.9
31.8 5.3
Pearson r
Age (years) Female n (%) Number of comorbidities Number of medications Walking Aid use (%) History of falls (%) Sedentary behavior (IPAQ-SF) Timed get up and go scores (seconds) BPI pain severity BPI pain interference Balance confidence (ABC scale) Concerns about the consequences of falling (CoF scale)
P value
0.04 0.03 0.08 0.10 0.17 0.13 0.31 0.26
0.58 0.68 0.31 0.18 0.02 0.10 < 0.001 < 0.001
0.32 0.29 0.28 0.40
< < <
