Musculoskeletal pain in polio survivors and strength-matched controls

Musculoskeletal pain in polio survivors and strength-matched controls

1679 Musculoskeletal Pain in Polio Survivors and Strength-Matched Controls Mary G. Klein, PhD, Mary Ann Keenan, MD, Alberto Esquenazi, MD, Roberta Co...

97KB Sizes 2 Downloads 89 Views


Musculoskeletal Pain in Polio Survivors and Strength-Matched Controls Mary G. Klein, PhD, Mary Ann Keenan, MD, Alberto Esquenazi, MD, Roberta Costello, BSN, RN, Marcia Polansky, ScD ABSTRACT. Klein MG, Keenan MA, Esquenazi A, Costello R, Polansky M. Musculoskeletal pain in polio survivors and strength-matched controls. Arch Phys Med Rehabil 2004;85:1679-83. Objectives: To determine whether a significant difference exists between musculoskeletal symptoms of polio survivors and those of older adults with no history of polio, and to determine if activity level and strength predict pain in either group. Design: Matched research design. Setting: A research laboratory in a rehabilitation setting. Participants: Fifty-four polio survivors and 54 adults with no history of polio were matched for gender, race, and bilateral knee extensor strength and selected from a cohort of 316 subjects who participated in a study on the relation between activity level and health status. Interventions: Not applicable. Main Outcome Measures: Location and severity of musculoskeletal pain, activity frequency and intensity level, maximum voluntary isometric strength, and physical performance measures. Results: Polio survivors reported significantly more symptoms than the matched controls (P⬍.05). Symptom status among the polio survivors was strongly associated with performance strain, perceived exertion, and activity intensity. Although the polio survivors had activity frequencies and habitual walking speeds that were similar to those from the matched controls, there was evidence that they performed activities at higher intensity levels. Conclusions: Activity level is a factor in the development of musculoskeletal symptoms in polio survivors. Polio survivors who perform at higher intensity levels are more likely to have moderate to severe pain and more mobility difficulties. Key Words: Postpoliomyelitis syndrome; Rehabilitation © 2004 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation LTHOUGH AN INCREASING number of aches and pains are generally accepted as part of the normal aging process, A there is evidence that polio survivors develop muscle and joint

pain problems at an accelerated rate compared with their peers.1 It has been hypothesized that the risk of polio survivors developing musculoskeletal symptoms is dependent on the severity of motoneuron damage, which is often associated with muscle weakness.2 Lower-extremity weakness, in particular, has been identified as a key factor in predicting the presence of muscle and joint pain in the postpolio population.3,4 However, at least 1 study5 on pain and disability in postpolio patients reported no correlation between daily pain ratings and muscle strength. Instead, a significant correlation between pain severity and daily exertion was documented. These results led us to theorize that it might actually be the polio survivors who are more active who experience more pain. Polio survivors are often portrayed as overachievers who push themselves to overcome their muscle weakness and to perform their activities as well as or better than their peers.6,7 As a result, polio survivors are often encouraged by their physicians to decrease their activity levels and to conserve their strength as a way to reduce their musculoskeletal symptoms. However, these clinical decisions are often based on anecdotal evidence that polio survivors with higher activity levels have more symptoms than those with lower activity levels. It is clear that additional research is needed to understand the development of musculoskeletal pain in polio survivors. Research is also needed to determine whether polio survivors are at greater risk for muscle and joint pain than are persons with similar strength levels but no history of polio. The purpose of the present study was to compare the number and severity of musculoskeletal symptoms in a postpolio population with those of a control group who had no history of polio. Both groups were adults matched for gender, race, and bilateral knee extensor strength. The control group represented people with muscle weakness due to the effects of normal aging. The objectives were to determine whether a significant difference existed in the proportion of subjects in each group affected by musculoskeletal symptoms and to determine to what degree activity level and strength predicted pain in either group. Other factors commonly associated with musculoskeletal symptoms, such as weight and age, were also investigated. We hypothesized that the polio survivors would be more likely to report muscle and joint pain and would also have higher activity levels than the matched controls, and also that physical activity level would be a significant predictor of pain in the postpolio population. METHODS

From Moss Rehabilitation Research Institute (Klein, Costello); Albert Einstein Medical Center (Keenan); MossRehab Hospital (Esquenazi); Thomas Jefferson University (Keenan, Esquenazi); and MCP-Hahnemann University (Polansky), Philadelphia, PA. Keenan is currently affiliated with the University of Pennsylvania, Philadelphia, PA. Supported by the US Department of the Army (grant no. DAMD17-00-5033) and the National Institute on Disability and Rehabilitation Research, US Department of Education (grant no. H133A000101). No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. Reprint requests to Mary G. Klein, Korman 204-B, Moss Rehabilitation Research Institute, 1200 W Tabor Rd, Philadelphia, PA 19141, e-mail: [email protected] 0003-9993/04/8510-8453$30.00/0 doi:10.1016/j.apmr.2004.01.041

Participants Data from 54 polio survivors (30 men, 24 women) and 54 gender- and race-matched adults with no history of polio were analyzed for this study. Bilateral knee extensor strength was also matched between the 2 groups and differed by 4lb (1.8kg) or less within each matched pair. All subjects in this study were recruited as part of a larger project on activity level and functional performance in polio survivors and older adults without a history of polio (M Klein et al, unpublished data; 2001). The inclusion criteria were: (1) no major disabilities, such as stroke, amputation, rheumatoid Arch Phys Med Rehabil Vol 85, October 2004


PAIN IN POLIO SURVIVORS AND MATCHED CONTROLS, Klein Table 1: Strength Testing Protocol

Muscle Group

Body Position

Limb Position

HHD Placement

Hip flexion

Side lying*

Proximal to superior side of patella

Pelvis (S)

Hip extension Hip abduction

Side lying* Supine Side lying*

Proximal to popliteal crease Proximal to superior pole of patella on lateral aspect of thigh Proximal to malleoli on posterior aspect of calf

Pelvis (S) Hip (T)

Knee flexion

Hip flexed to 30°; knee flexed to 60° Hip neutral; knee extended Hip abducted to 45°; contralateral hip neutral Hip flexed to 10°; knee flexed to 30°

Stabilization (T/S)

Knee extension

Side lying*

Hip neutral; knee flexed to 45°

Proximal to malleoli on anterior aspect of tibia

Tibia (S) Pole placed anteriorly, just proximal to patella† Femur (S) Pole placed posteriorly, just proximal to popliteal fossa†

Abbreviations: HHD, hand-held dynamometer; S, stabilization provided by a second staff member; T, stabilization provided by therapist. *Leg positioned on raised powder board. † Where appropriate, a padded pole (diameter, ⬇1.5in [3.8cm]) was placed in a predrilled hole in the powder board to help minimize compensatory actions that might be attempted by a subject. Pole placement was standardized; the staff member simply held the pole in place, ensuring that contact was made with the defined area as described above.

arthritis, or peripheral neuropathy, that could cause muscle weakness; (2) no symptoms of uncontrolled or unstable cardiovascular or respiratory conditions, such as difficulty breathing with exertion, chest pain with activity or at rest, or history of a recent heart attack, which might make a maximal strength test or a walking test unsafe; and (3) no fractures or surgeries within the past 6 months. Also, subjects could not be undergoing treatment for cancer (other than skin) and had to be able to ambulate a minimum distance of 30ft (9m) with or without the use of an assistive device. The study protocol received institutional review board approval and written informed consent was obtained from all subjects. Procedure All study participants completed a standardized medical history form that included questions on current health status and medications taken on a regular basis. Polio survivors also completed a questionnaire that included questions about their original polio infection (eg, how old they were at the time, what limbs were affected). The presence and absence of mobility impairment was determined by asking the subjects to rate their ability to perform 5 mobility-related tasks on a scale from 0 (unable) to 4 (no difficulty). The tasks were walking 3 to 4 city blocks, climbing and descending a flight of 10 to 12 stairs, pushing a large object such as a small sofa or loveseat, stooping down to pick up a small object such as a pen or pencil from the floor, and carrying a 10lb (4.5kg) bag of groceries. Subjects who rated themselves as a 2 or lower on any of these tasks were considered to have a mobility impairment. Height and weight were measured on a standard scale. To assess musculoskeletal pain, we used a human body diagram on which subjects were asked to shade in the areas where they experienced muscle and/or joint pain on a regular basis. For those areas where pain was indicated, pain severity was assessed using a scale that ranged from 0 (no pain) to 10 (pain bad as it can be). Activity level was quantified in terms of frequency and intensity. Frequency of physical activity was assessed using the Physical Activity Scale for the Elderly8 (PASE). This scale was developed for measuring activity level among adults at or over the age of 65 years. It includes questions about occupational, household, and leisure activities performed during the past week. It has been used previously in studies with polio survivors younger than 65 years, with the assumption that this population is more sedentary than their peers without polio.9 Arch Phys Med Rehabil Vol 85, October 2004

Scores can range from 0 to over 400, with higher scores indicating higher activity levels. Activity intensity was estimated by rating walking speed. Habitual and maximal walking speed were measured as subjects walked through a 30-foot (9-m) section of hallway. An estimate of activity intensity level was then computed by dividing habitual walking speed by maximal walking speed. To provide an objective measure of performance strain during each walking task, subjects wore a heart rate monitor while they walked. Subjects also rated their perceived exertion after each walking task on the Borg Rating of Perceived Exertion (RPE) Scale.10 Using a Microfet2a hand-held dynamometer, a physical therapist measured maximum voluntary isometric strength in the bilateral hip flexor, hip extensor, hip abductor, knee flexor, and knee extensor muscle groups in gravity-minimized positions. The testing position, stabilization point(s), and dynamometer placement were standardized (table 1). Intrarater reliability was evaluated prior to data collection and the average percentage difference for all muscle groups was less than 10%. A second person helped to stabilize the subject when the larger muscle groups were tested (table 1). For each strength test, the subject pushed against the padded dynamometer forceplate, which was held stationary by the physical therapist. Subjects were told to build to a maximal force over a period of 2 to 3 seconds and then hold this maximal effort for 3 to 4 seconds. The peak force was recorded for each trial (in pounds). A minimum of 2 trials was performed for each muscle group. Additional trials were performed only if the first 2 varied by more than 10%. For very weak muscles with peak strengths of less than 10lb (4.5kg), additional trials were performed only if the difference between the first 2 measurements was greater than 1lb (2.2kg). To prevent fatigue, subjects were given a minimum of 30 seconds to rest between trials, and the maximum number of trials for a single muscle group was 4. If a subject reported pain during testing, those trials were considered invalid. The mean of the valid trials recorded for peak strength of each muscle group was used in all further analyses. Data Analysis Data were analyzed using Systat9b software. Body mass index (BMI) was calculated by multiplying weight (in pounds) by 705 (constant) and dividing by height (inches) squared.11 Performance strain was calculated for each walking task (ha-


PAIN IN POLIO SURVIVORS AND MATCHED CONTROLS, Klein Table 2: Descriptive Statistics Variable

Demographic Age (y) Height (in) Weight (lb) BMI (kg/m2) Pain No. of sites Severity Activity level PASE Habitual speed (m/s) Maximum speed (m/s) Intensity level Task performance Strain (habitual test) Strain (maximum test) Perceived exertion (habitual test) Perceived exertion (maximum test)

Postpolio Group (mean ⫾ SD)

Control Group (mean ⫾ SD)


61.85⫾7.6 66.73⫾3.8 177.13⫾40.6 27.76⫾4.6

74.3⫾7.8 64.9⫾3.8 160.2⫾39.0 26.6⫾5.9

⬍.001 NS .002 .046

4.92⫾4.8 5.14⫾1.8

1.9⫾1.2 4.3⫾1.6

⬍.001 NS

153.17⫾90.5 1.02⫾0.3 1.39⫾0.4 0.74⫾0.1

135.8⫾67.3 1.0⫾0.3 1.5⫾0.3 0.7⫾0.1

0.95⫾0.3 1.01⫾0.3 7.70⫾1.8 9.72⫾2.6

0.9⫾0.3 1.0⫾0.3 6.9⫾1.3 7.5⫾2.0

NS NS .038 .002 NS NS ⬍.001 ⬍.001

Abbreviation: NS, not significant.

bitual vs maximal) by dividing heart rate reserve (HRR; calculated using equation: HRR⫽220⫺age⫺resting heart rate) by the maximum heart rate captured during task performance.12 Overall lower-extremity strength was also calculated for each subject using the sum of the mean strength from all 10 hip and knee muscle groups (5 from each leg). Symptom status was quantified using a modified version of the American College of Rheumatology criteria for musculoskeletal pain.13 The objective was to differentiate between subjects who reported mild or no pain versus those with moderate to severe pain. Subjects were grouped according to whether their visual analog scale (VAS) rating was greater than 3 in at least 1 area of the body (ie, group 1, mild or no pain [VAS, ⬍3]; group 2, moderate to severe pain in 1 or more areas [VAS, ⬎3]). Similar criteria were used previously to group subjects in a study of pain in older disabled women.14 We used Wilcoxon matched-pairs signed-rank tests to examine the degree of difference in the group responses in terms of activity frequency, activity intensity, combined hip and knee strength, perceived exertion, and performance strain. Spearman correlational analyses were also performed to examine the relation between the number of reported symptoms with each of these variables. The independent associations of walking strain, perceived difficulty, activity frequency, and activity intensity on symptom status were examined separately for each group, using logistic regression. Demographic variables, including age, gender, weight, BMI, and combined hip and knee strength were also included as potential predictors. Because of the relatively large number of potential predictor variables, we performed univariate logistic regression analyses with a cutoff value of .15. Our intention was to eliminate some terms before doing a multivariate stepwise logistic analysis for each group. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for each independent variable. These analyses were performed using ranked data. No corrections were made for multiple testing. Statistical significance for all analyses was defined as P less than .05.

RESULTS Among the polio survivors, age at the time of the initial polio infection ranged from 0.75 to 27.5 years (mean ⫾ standard deviation [SD], 7.45⫾6.4y) and length of time since the original infection ranged from 40 to 72 years (50.81⫾7.8y). A total of 48 (88.9%) polio survivors reported experiencing some difficulty with mobility-related activities. By comparison, only 32 (59.3%) matched controls reported any mobility-related difficulties. Overall, the matched controls were significantly older than the polio survivors (table 2). The range in age was 46 to 77 years for the polio survivors and 60 to 94 years for the controls. In terms of weight, the polio survivors were significantly heavier than the matched controls. However, the difference in BMI between groups was only marginally significant (table 2). On average, subjects in both groups were overweight, but not obese. There was no significant difference in the mean strength of the hip and knee muscle groups between the 2 groups. However, there was a greater degree of asymmetry between the left and right sides for the polio survivors than the controls (table 3). Among the 54 polio survivors, 36 (66.7%) reported experiencing moderate to severe musculoskeletal pain in 1 or more sites on a regular basis, compared with only 18 (33.3%) of the matched controls. In terms of gender, 66.7% of men and 66.7% of women in the postpolio group reported moderate to severe symptoms. In the control group, the percentages were 23.3% for the men and 45.8% for the women. The shoulders, knees, lower back, and ankles/feet were the most frequently reported sites for moderate to severe pain in both groups. A significant difference existed in the mean number of pain sites reported, with polio survivors having more symptoms on average. Polio survivors were also more likely to report difficulty with mobility-related tasks (fig 1). In terms of activity level and functional performance ability, the polio survivors had a lower mean maximum walking speed (1.39m/s vs 1.49m/s) and, as a result, a higher mean activity intensity level (.74 vs .68) than the controls. There was also evidence of a weak association between the combined hip and Arch Phys Med Rehabil Vol 85, October 2004


PAIN IN POLIO SURVIVORS AND MATCHED CONTROLS, Klein Table 3: Summary of Strength Results Muscle Group

Postpolio Group (mean ⫾ SD)

Control Group (mean ⫾ SD)

L hip abductors R hip abductors Difference between L and R hip abductors

27.3⫾10.5 23.2⫾9.0 6.5⫾4.8

21.78⫾8.1 25.69⫾9.6 3.39⫾4.8

L hip flexors R hip flexors Difference between L and R hip flexors

30.2⫾12.5 28.9⫾10.3 8.3⫾7.9

32.79⫾11.7 32.76⫾11.4 3.76⫾5.0

L hip extensors R hip extensors Difference between L and R hip extensors

14.7⫾6.5 14.4⫾6.1 5.6⫾4.8

16.81⫾8.7 16.16⫾7.9 2.40⫾2.7

L knee flexors R knee flexors Difference between L and R knee flexors

26.6⫾16.0 23.2⫾12.4 11.2⫾10.9

23.22⫾10.5 27.19⫾12.4 4.41⫾6.4

L knee extensors R knee extensors Difference between L and R knee extensors

32.1⫾20.3 34.6⫾15.7 16.7⫾15.2

32.50⫾15.0 34.37⫾14.7 5.20⫾7.2

Abbreviations: L, left; R, right.

knee strength and activity frequency in polio survivors (Spearman ␳⫽.293; P⫽.032), but not in controls. The perceived exertion levels reported by the postpolio group were significantly higher than those reported by the matched control group for both walking tasks (all Pⱕ.001). Although there was a trend for polio survivors to have higher activity frequency levels than the controls, as indicated by the PASE score (mean PASE scores, 153.17 for polio survivors vs 135.76 for controls), the difference between groups was not significant. Number of symptoms was associated with performance strain for the habitual and maximum-speed walking tasks in the postpolio group. The correlation values ranged from .45 to .49 (all Pⱕ.001). Within the control group, it was perceived exertion during the walking task at habitual speed that correlated with symptom number (␳⫽.310, P⫽.030). However, number of symptoms did not correlate with strength, activity frequency (PASE), or activity intensity in either the postpolio or control groups. Polio survivors with moderate to severe symptoms had higher strain, higher perceived exertion ratings, and higher activity intensity levels than did subjects with mild or no pain. The results of the stepwise multivariate analysis showed that for the maximal walking task strain (OR⫽229.2; 95% CI, 1449.4 –5.1; P⫽.005), activity intensity levels (OR⫽6766.8; 95% CI, 21664.5–2.1; P⫽.032), and RPE score for the habitual walking task (OR⫽1.9; 95% CI, 3.3–1.0; P⫽.037) were the only significant predictors of moderate to severe musculoskeletal symptoms in the postpolio group. Similarly, controls with moderate to severe pain reported significantly higher perceived exertion ratings (OR⫽1.7; 95% CI, 2.8 –1.1; P⫽.030) than did controls with mild or no pain. DISCUSSION The present study provides evidence that the presence of moderate to severe musculoskeletal symptoms among polio Arch Phys Med Rehabil Vol 85, October 2004

survivors is related to their activity level. Polio survivors who had strain levels above .95 during walking at maximum speed and perceived exertion ratings above 8.2 for walking at normal speed, and who performed the walking tasks at activity intensity levels greater than .750 were at high risk for moderate to severe musculoskeletal symptoms. These findings are consistent with those of a previous study,5 which documented that polio survivors with higher levels of pain severity also reported increased levels of daily exertion during their normal daily activities. Of note, number of symptoms was not associated with strength or activity level in either group. Therefore, although polio survivors who performed at higher intensity levels and reported more effort with activity were more likely to have greater pain severity, they did not necessarily report more symptom sites than those who had lower activity intensity levels. In general, musculoskeletal pain, both in terms of the proportion of affected subjects (66.7 % vs 33.3%) and the number of pain sites reported (4.92 vs 1.90), was more prevalent among the polio survivors than among the matched controls. This finding supports the hypothesis that polio survivors are at higher risk for pain due to overuse than are controls with similar lower-extremity strength levels. Unlike persons who age normally, individuals affected by the polio virus did not have all their muscles affected to the same degree, leaving survivors with some muscles that had age-appropriate strength levels and others that were significantly weakened. Although the overall lower-extremity strength values were similar, a closer examination of the strengths of similar muscle groups in opposite legs revealed differences ranging from 5.6 to 16.7lb (2.5–7.5kg) among the polio survivors versus only 2.4 to 5.6lb (1.1–2.5kg) differences among the controls. The greater asymmetry in strength among polio survivors might have increased their risk for developing muscle and joint problems. Although the polio survivors on average were heavier (177lb vs 160lb [79.7kg vs 72kg]) and taller (66.7in vs 64.85in [169.4cm vs 164.7cm]) than the controls, the results did not show any evidence that body size was related to symptom status in either group. This is in contrast to a previous study4 that showed a strong association between weight and shoulder pain among female polio survivors. It is possible that weight may play a greater role in the development of some types of

Fig 1. Percentage of subjects reporting difficulty with mobilityrelated tasks.


musculoskeletal symptoms than in others. However, in the present study, all symptoms were grouped together for analysis because we did not have enough matched pairs to do an in-depth analysis of 1 specific symptom-type or site. The present study showed that polio survivors with higher muscle strength tended to be more active than those who were weaker. However, we found no evidence that stronger polio survivors experienced more pain even though they were more active. One possibility is that duration of weakness may be more important than current level of muscle strength. With normal aging, muscle strength does not begin to decline significantly until after age 60.15 However, polio survivors have been living with some degree of muscle weakness since their original polio infection onset more than 50 years prior in most cases. It is not surprising that their bodies would show more signs of stress and strain than older adults with no history of polio, even if both groups have similar levels of strength on average. In this study, there was no way to determine how long each subject had been at his/her current strength level. We were able to look at duration of time since the onset of polio as a possible confounding factor, but found no evidence of a relation between this variable and number or severity of musculoskeletal symptoms within the postpolio group. Contrary to what was expected, we found no significant difference in activity frequency between the polio survivors and the controls. One possibility is that the similar group scores reflect the common baseline strength needed to perform basic activities of daily living. We might have been more likely to see a significant difference had we focused on each group’s ability to respond to a physical challenge, especially if it was a challenge where endurance was needed. The results indicated that polio survivors performed activities at a higher intensity level than the controls, which may mean that their ability to maintain a certain activity level over an extended period may be lower. There were some limitations associated with this study. Many of the variables were assessed by self-report and were, therefore, prone to measurement error from bias associated with memory, recall, and other personal factors. Also, the data were crosssectional, which affected our ability to make any definitive conclusions about the cause and effect relationships between variables. Because musculoskeletal pain is intermittent in many cases, a longitudinal study would have enabled us to get a better picture of the changes in symptom status, strength, and activity level over time. It is also possible that long-term activity patterns may play a greater role in the development of musculoskeletal problems than short-term activity patterns. The reference period for the activity frequency measure used in this study was relatively short (ie, previous week). CONCLUSIONS The risk for having muscle and joint pain was higher in polio survivors than in controls with similar lower-extremity strength levels, even though the controls were significantly older than the polio survivors. Findings suggested that the polio survivors and controls had similar activity frequency levels, but the polio survivors performed their daily activities at a level closer to their maximal capacity. It was activity intensity rather than


frequency that had a significant effect on polio survivors’ symptom status. In turn, that symptom status was strongly associated with their level of mobility. Raising awareness of the relation between activity level and pain may encourage polio survivors to modify their daily routines so that muscle strain is reduced and pain becomes less of a factor in their lives. Acknowledgments: We thank Jennifer Kuehl, PT, and Carolyn Szumal, PT, for their assistance with data collection. We also thank John Whyte, MD, PhD, for his critical review of the study proposal and his editorial assistance with this manuscript. References 1. Perry J, Barnes G, Gronley JK. The postpolio syndrome: an overuse phenomenon. Clin Orthop 1988;Aug(233):145-62. 2. Rekand T, Albrektsen G, Langeland N, Aarli JA. Risk of symptoms related to late effects of poliomyelitis. Acta Neurol Scand 2000;101:153-8. 3. Vasiliadis HM, Collet JP, Shapiro S, Venturini A, Trojan DA. Predictive factors and correlates for pain in postpoliomyelitis syndrome patients. Arch Phys Med Rehabil 2002;83:1109-15. 4. Klein MG, Whyte J, Keenan MA, Esquenazi A, Polansky M. The relation between lower extremity strength and shoulder overuse symptoms: a model based on polio survivors. Arch Phys Med Rehabil 2000;81:789-95. 5. Willen C, Grimby G. Pain, physical activity and disability in individuals with late effects of polio. Arch Phys Med Rehabil 1998;79:915-9. 6. Creange S, Bruno RL. Compliance with treatment for post-polio sequelae: effect of type A behavior, self-concept and loneliness. Am J Phys Med Rehabil 1997;76:378-82. 7. Farbu E. Polio survivors—well educated and hard working. J Neurol 2001;248:500-5. 8. Washburn RA, Smith KW, Jette AM, Janney CA. The physical activity scale for the elderly (PASE): development and evaluation. J Clin Epidemiol 1993;46:153-62. 9. Willen C, Cider A, Sunnerhagen KS. Physical performance in individuals with late effects of polio. Scand J Rehabil Med 1999; 31:244-9. 10. Borg G. Perceived exertion as an indicator of somatic stress. Scand J Rehabil Med 1970;23:92-8. 11. Seidel HM, Ball JW, Dains JE, Benedict GW. Mosby’s guide to physical examination. 4th ed. St. Louis: Mosby; 1999. 12. Astrand PO, Rodahl K. Textbook of work physiology. 3rd ed. New York: McGraw-Hill; 1986. 13. Wolfe F, Smythe HA, Yunus MB, et al. The American College of Rheumatology 1990 Criteria for the Classification of Fibromyalgia. Report of the Multicenter Criteria Committee. Arthritis Rheum 1990;33:160-72. 14. Leveille SG, Ling S, Hochberg MC, et al. Widespread musculoskeletal pain and the progression of disability in older disabled women. Ann Intern Med 2001;135:1038-46. 15. Asmussen E, Heeboll-Nielsen K. Isometric muscle strength in relation to age in men and women. Ergonomics 1962;5:167-9. Suppliers a. Empi Microfet 2; Hoggan Health Industries Inc, PO Box 957, Draper, UT 84020. b. SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.

Arch Phys Med Rehabil Vol 85, October 2004