SEARCH

SEARCH BY CITATION

Keywords:

  • Activity limitation;
  • Grip force;
  • Health Assessment Questionnaire;
  • Evaluation of Daily Activities Questionnaire;
  • Rheumatoid arthritis;
  • Sex

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Objective

To evaluate activity limitations 3 years after diagnosis of early rheumatoid arthritis (RA) in relation to grip force and sex.

Methods

A total of 217 patients, 153 women and 64 men, with recent-onset RA were included. Activity limitations were reported using the Health Assessment Questionnaire (HAQ) and the Evaluation of Daily Activities Questionnaire (EDAQ). The relationships between activity limitations versus grip force (measured by the Grippit), walking speed, functional impairment, grip ability, pain, plasma C-reactive protein, the 28-joint disease activity score and its components, the physician's global assessment of disease activity, and sex were analyzed by partial least squares (PLS).

Results

Women had significantly lower grip force and more activity limitations (HAQ and EDAQ) than men. The PLS analyses demonstrated that grip force was the strongest regressor of activity limitation, closely followed by walking speed. However, within subgroups based on grip force (group 1 = grip force <114 N, group 2 = 116–206 N, group 3 = 214–321 N, group 4 = grip force >328 N) and including sexes, women and men had corresponding degrees of activity limitation as reported by the HAQ and EDAQ.

Conclusion

Our results indicate that the more pronounced activity limitations seen in women with RA, as compared with men, may be explained by lower grip force rather than sex.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Rheumatoid arthritis (RA) is a chronic inflammatory disease that ultimately leads to joint destruction and permanent disability. However, early in the disease course, the inflammatory process leads to a multitude of functional limitations (1–5). Several studies have also shown that RA is associated with a shortened life span, mainly due to coronary vascular disease (6–8). Wolfe et al recently reported that patients' self-assessed functional impairments, using the Health Assessment Questionnaire (HAQ), are the best predictors of life expectancy compared with laboratory, radiographic, and physical examination data (9). Early use of potent antirheumatic therapy is advocated to improve the outcome (10).

Typically, disability in RA diminishes within the first year after diagnosis and start of intervention, and then either remains stable but still affected (11) or slowly worsens over time (2, 12). Two recent Swedish studies on early RA have found that women on average have more pronounced difficulties in performing daily activities than men as reflected by the HAQ (13, 14). Apart from sex, factors such as age, disease activity, and joint destruction have been found to correlate with HAQ score (4, 15–18). Also, by using another instrument designed to evaluate activity limitations in patients with RA, the Evaluation of Daily Activities Questionnaire (EDAQ), women have been found to be more disabled than men (5, 19). Similar to healthy individuals, women with RA have a substantially lower average grip force than men with RA (5, 20–23). Several studies have also found that grip force is associated with disability (5, 20, 24–26).

In a review of sex-based differences in patients with RA, Harrison pointed out the need for studies that highlight sex differences to optimize intervention strategies in women and men (27). The purpose of the present study was to investigate possible differences between women and men concerning grip force and activity limitation 3 years after the diagnosis of recent-onset RA, and to determine if activity limitation correlates with grip force.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

TIRA study.

The present study is part of the Swedish early intervention in RA (TIRA) project, in collaboration between 10 Swedish rheumatology units (Eskilstuna, Jönköping, Kalmar, Lindesberg, Linköping, Motala, Norrköping, Oskarshamn, Västervik, and Örebro) (13). A total of 320 patients with recent-onset RA (≤1 year after the first self-reported signs of joint swelling) were included in the Swedish TIRA, 215 women (median age 56 years, interquartile range [IQR] 24) and 105 men (median age 62 years, IQR 20). The first signs of arthritis (joint swelling) were observed at least 6 weeks, but not more than 1 year, before inclusion. All patients either fulfilled at least 4 of 7 criteria for RA as defined by the 1987 revised American College of Rheumatology (ACR) classification criteria (18) or experienced morning stiffness (≥60 minutes as judged by the patients), symmetric arthritis, and arthritis in small joints (fingers, hands, wrists, feet, and toes). The mean ± SD number of ACR criteria was 4.5 ± 0.8, with no differences between women and men. The 44 patients who dropped out during the period from inclusion to the 3-year followup, 24 women (median age 69 years, IQR 23 at inclusion) and 20 men (median age 70 years, IQR 15 at inclusion), were significantly older (P < 0.001) than the rest of the study group. There were no significant differences regarding the baseline Disease Activity Score in 28 joints (DAS28) (28) or HAQ score (29).

Study group.

Of the 276 patients (86%) remaining in the TIRA study at the 3-year followup, 191 (69%) were women (median age 55 years at inclusion, IQR 24) and 85 (31%) were men (median age 60 years at inclusion, IQR 20). Latex-agglutinating rheumatoid factor (RF) was analyzed, and 62% of the women and 60% of the men were RF seropositive at the time of inclusion. The 217 (78%) patients who completed the EDAQ (19) at the 3-year followup were included in the present study. Of these, 153 (71%) were women (median age 56 years, IQR 24 at inclusion) and 64 (29%) were men (median age 60 years, IQR 21 at inclusion). RF was present in 61% of the women and 58% of the men at the time of inclusion. Of the 59 patients who dropped out and did not complete the EDAQ, 37 were women (median age 50 years, IQR 27 at inclusion) and 22 were men (median age 59 years, IQR 23 at inclusion). Age, erythrocyte sedimentation rate (ESR), plasma C-reactive protein (CRP), DAS28 (28), grip force (21), and HAQ score (29) did not differ between the study group and the dropouts, whereas the dropouts reported significantly more general pain on a 0–100-mm visual analog scale (VAS) (P = 0.023) and more tender joints (P = 0.04).

The patients were offered medication and multiprofessional intervention when considered adequate, and ongoing medication was registered at followup. At the 3-year followup, treatment with disease-modifying antirheumatic drugs was ongoing in 65% of the women and 64% of the men. No differences were found between women and men regarding the proportion of patients who were prescribed nonsteroidal antiinflammatory drugs, analgesics, or oral corticosteroids.

Patient characteristics.

ESR and CRP levels were analyzed at the local hospital laboratories. DAS28 was calculated (21) and the physician's global assessment of disease activity (PGA) was estimated on a 5-degree scale (range 0–4, where 0 represents no activity and 4 represents high activity). The patients estimated the duration of morning stiffness (minutes). Patients also graded average pain intensity and average wellbeing during the past week using 0–100-mm VAS scales (where 0 represents no pain or best possible wellbeing and 100 represents worst possible pain/wellbeing). Grip ability was tested using the Grip Ability Test (GAT), which consists of 3 items; the score (range 10–276) is based on time consumption (high scores indicating decreased hand function). The reliability and internal consistency of the GAT are high (30). Signals of Functional Impairment (SOFI) was used to assess hand function (4 items scored 0–16), upper limb function (3 items scored 0–12), and lower limb function (4 items scored 0–16) (31). The scoring of each item ranged from 0 (full function) to 2 (cannot perform) for the right and left side, respectively. The reliability, validity, and sensitivity of SOFI have been reported to be adequate (31). Walking speed was defined as the time (seconds) it took to walk 20 meters as fast as possible indoors; the patients used their own assistive devices if necessary.

Grip force.

Maximum, momentary, and average grip force (N) over a period of 10 seconds were measured in the right and the left hand separately using a digital electronic device (Grippit; Detektor AB, Göteborg, Sweden) (21). In this study, the average grip force of 10 seconds in the right hand was used. Three percent of the patients were left handed, with no significant difference in grip force in the left or right hand in comparison with the right-handed patients.

Activity limitation.

The Swedish version of the HAQ (29) was used to report difficulties in performing activities. The HAQ includes 20 items divided into 8 groups. The HAQ score is achieved by adding the highest score within each group divided by the number of completed groups, and ranges from 0 to 3 (where 0 = without any difficulty and 3 = unable to do). Difficulties performing daily activities were also measured by the EDAQ, which consists of 102 items grouped into 11 dimensions: eating/drinking, transfer, toileting, dressing, bathing, cooking, mobility indoors, cleaning, washing/clothes care, mobility outdoors/shopping, and communication (19). Each dimension contains 4–13 items, which in this study are termed “activities.” Difficulties performing activities were represented by the answer to the question, “How do you perform the following activities without assistive devices or altered methods?” The scale ranged from 0 to 3 (where 0 = without any difficulty and 3 = unable to do). The occupational therapist gave instructions regarding the EDAQ at the visit. The questionnaire was completed by the patient at home and was sent back within 2 weeks. An EDAQ score is calculated by adding the highest score within each dimension and then by dividing by the number of completed dimensions, and the score ranges from 0 to 3 (where 0 = without any difficulty and 3 = unable to do).

Statistics.

All statistics were calculated using SPSS for Windows version 11.5 (SPSS, Chicago, IL) or SIMCA-P version 10.0, and P values ≤0.05 were considered significant. Median values and IQRs are presented. The Kruskal-Wallis test was performed to determine if significant differences occurred between subgroups, and if so, the Mann-Whitney U test was used to test differences between independent subgroups and to test differences between women and men. Partial least squares or projection to latent structures (32) was used to investigate the relationships between variables representing patient characteristics and grip force versus variables representing activity limitation. The variable influence on projection (VIP) provides information about the relevance of each X variable and each Y variable pooled over all dimensions, and VIP >1.0 is considered significant. Components with eigenvalues ≥2.00 were considered nontrivial components. Multiple linear regression is an alternative method for the prediction, but it assumes that the regressors (X variables) are mathematically independent, and only 1 Y variable at a time can be predicted. If multicolinearity (high correlations) occurs among the X variables, the calculated regression coefficients become unstable and their interpretability breaks down. Outliers were identified using the 2 methods available: score plots in combination with Hotelling's T2 (identifies strong outliers) and distance to model in X space (identifies moderate outliers). Subgroups of patients, based on grip force (means), were identified by K-means cluster analysis. All data except EDAQ score were collected at the regular visit.

Ethics.

All patients gave written informed consent to participate. The local ethics committees of the participating units approved the protocol for the study.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Patient characteristics.

The majority of the variables did not differ significantly between women and men in the study group (Table 1). A small but significant difference was found in SOFI assessments of the upper limb, where men had more functional impairment than women (Table 1).

Table 1. Comparison of patient characteristics, grip force, and activity limitation between women and men*
VariableWomen (n = 153)Men (n = 64)P
  • *

    The median missing rate in the variables was 4% (range 0–16%). Values are the median (interquartile range) unless otherwise indicated. ESR = erythrocyte sedimentation rate; NS = not significant; CRP = C-reactive protein; DAS28 = disease activity score 28 joint count; PGA = physicians global assessment of disease activity; VAS = visual analog scale; GAT = Grip Ability Test; SOFI = Signals of Functional Impairment; HAQ = Health Assessment Questionnaire; EDAQ = Evaluation of Daily Activities Questionnaire.

Patient characteristics   
 ESR, mm/hour15 (20)16 (26)NS
 CRP, mg/liter5 (9)5 (10)NS
 Swollen joints (range 0–28)2 (5)3 (7)NS
 Tender joints (range 0–28)2 (5)1 (4)NS
 DAS28 score3.45 (2.23)3.42 (2.08)NS
 PGA (range 0–4)1 (0.25)1 (1)NS
 Morning stiffness, minutes30 (55)30 (87)NS
 Pain (0–100-mm VAS)30 (38)26 (48)NS
 Wellbeing (VAS, mm)31 (36)28 (49)NS
 GAT (range 0–276)18 (9)20 (9)NS
 SOFI hand (range 0–16)1 (3)2 (4)NS
 SOFI upper limb (range 0–12)0 (1)1 (3)< 0.001
 SOFI lower limb (range 0–16)2 (3)1 (2)NS
 Walking speed, seconds13 (4)12 (5)NS
Grip force and activity limitation   
 Grip force, N114 (104)221 (154)< 0.001
 HAQ (range 0–3)0.63 (1.0)0.25 (0.75)< 0.001
 EDAQ (range 0–3)0.66 (0.91)0.29 (0.66)< 0.001

Grip force.

The mean ± SD grip force in men (222 ± 108) was significantly higher than in women (126 ± 74) (Table 1).

Activity limitation.

On average, women had a significantly higher HAQ score than men (Table 1). Women reported significantly more pronounced difficulties than men in 11 of the 20 activities in the HAQ, whereas in 9 activities there were no differences (Table 2). Of the 20 activities in the HAQ, 11 were mainly performed using the upper extremities (Table 2), and 6 (55%) of these 11 activities were significantly more difficult for women to perform compared with men. Of the other 9 HAQ activities, which were not mainly dependent on the upper extremities, 5 activities (55%) were significantly more difficult for women to perform compared with men. The activities most frequently reported as difficult to perform by both women and men were “open jars which have been previously opened” and “reach and get down a 5 pound object such as a bag of sugar from just above your head.”

Table 2. Percentage distribution in the Health Assessment Questionnaire (HAQ) in 20 items*
 0123 
HAQ itemsW (M)W (M)W (M)W (M)P
  • *

    Test between women (W) (n = 153) and men (M) (n = 64). P values <0.05 are considered significant. The median missing rate in items was 16% (range 15–51%). 0 = without any difficulty; 1 = with some difficulty; 2 = with much difficulty; 3 = unable to do; NS = not significant.

  • Activities mainly performed by using upper extremity.

Shampoo your hair76 (85)21 (15)21NS
Dress yourself including tying shoelaces and doing buttons53 (69)42 (28)5 (4) NS
Stand up from an armless straight chair67 (74)26 (26)52NS
Get in and out of bed75 (80)22 (21)3 NS
Cut your meat55 (80)30 (19)14 (2)10.001
Open a new milk carton82 (84)15 (14)4 (2)NS
Lift a full glass to your mouth75 (89)23 (9)2 (2)0.045
Climb up 5 steps64 (82)33 (19)30.016
Walk outdoors on flat ground72 (83)22 (15)6 (2)NS
Take a tube bath55 (71)20 (27)15 (2)100.018
Get on and off the toilet74 (87)18 (11)8 (2)0.049
Wash and dry your entire body69 (78)30 (19)2 (4)NS
Reach and get down a 5 pound object such as a bag of sugar from just above your head42 (63)43 (32)1 (6)0.006
Bend down to pick up clothing from the floor68 (78)26 (19)5 (4)NS
Open car doors72 (85)23 (15)50.045
Open jars that have been previously opened28 (61)39 (26)31 (13)2< 0.001
Turn faucets on and off65 (80)26 (17)9 (4)0.048
Do chores such as vacuuming and yardwork49 (83)38 (17)122< 0.001
Run errands and shop50 (84)31 (12)18 (4)2< 0.001
Get in and out of a car62 (74)31 (24)7 (2)NS

On average, women had a significantly higher EDAQ score than men (Table 1). Women stated significantly more marked difficulties than men in 46 of the 102 activities in the EDAQ; however, there were no differences in the remaining 56 activities (Table 2). Of the 102 activities in the EDAQ, 77 were mainly performed using the upper extremities, and in 39 (51%) of these activities, women had significantly more difficulties than men (Table 3). Of the remaining 25 EDAQ activities, which were not mainly dependent on the upper extremities, 7 activities (28%) were significantly more difficult for women to perform compared with men. The activities most frequently reported as difficult to perform by women belonged to the dimensions eating/drinking (“opening glass jar” and “opening juice bottle”), cooking (“lifting frying pan” and “reaching for sugar”), and activities related to mobility outdoors/shopping (“taking long walks,” “bringing home groceries,” and “shopping on a large scale”) (Table 3). For men, the most frequently reported difficulties were “opening glass jar” and “taking long walks.” For women, 71% of the 102 activities within the EDAQ were performed “without any difficulty,” and for men, 84% were performed “without any difficulty.”

Table 3. Percentage distribution in Evaluation of Daily Activities Questionnaire in 102 activities grouped in 11 dimensions*
 0123P
WMWMWMWM
  • *

    Test between women (W) (n = 153) and men (M) (n = 64). P values <0.05 are considered significant. The median missing rate in the activities was 2% (range 0–48%). 0 = without any difficulty; 1 = with some difficulty; 2 = with much difficulty; 3 = unable to do; NS = not significant.

  • Activities mainly performed by using upper extremity.

Eating/drinking         
 Lifting glass8091208 2  NS
 Lifting cup789221612  0.017
 Using knife, fork778820133   NS
 Cutting bread5476412252  0.002
 Slicing cheese5671382563  0.034
 Bringing down milk carton678629144   0.003
 Pouring milk698626145   0.008
 Opening milk carton466940301123 0.001
 Opening bottle4978352213 3 < 0.001
 Opening glass jar20524638246114< 0.001
 Opening juice bottle266150361539 < 0.001
 Opening can426839271357 < 0.001
Toileting         
 Getting to toilet939773    NS
 Sitting on and rising from toilet77872011221 NS
 Drying oneself after toilet visit879211622  NS
 Washing back919793    NS
 Arranging clothes88921181   NS
 Washing hands96984  2  NS
 Brushing and combing hair848915111   NS
 Brushing teeth8795135    NS
 Using toothpaste838917111   NS
 Using suppository90909101   NS
 Opening medicine bottle688628153 1 0.008
 Putting on makeup8892128    NS
 Putting on jewelery69912795   0.001
Dressing         
 Outdoor clothing788721131   NS
 Clothes over head738126191   NS
 Clothes with front button8184161333  NS
 Buttoning and unbuttoning6876292133  NS
 Clothes over feet727827211   NS
 Zipping and unzipping78922181   0.013
 Putting on tights69702728321 NS
 Putting on shoes7875192432  NS
 Putting on boots697424267 1 NS
 Tying shoelaces727821196212NS
 Putting on finger gloves83931473   0.044
Bathing/shower         
 Getting to bathroom959753  1 NS
 Getting in and out of bathtub5466272666122NS
 Standing up to take a shower88911091 1 NS
 Washing neck and back61733122553 NS
 Drying neck and back70832413452 NS
 Washing and drying feet737521205312NS
 Managing lever80941763 1 0.009
 Washing hair808916112 2 NS
 Putting hair in rollers68932075 7 0.040
 Blow-drying hair749018103 4 NS
 Manicure788315185 2 NS
 Pedicure636922268472NS
Cooking         
 Walking into kitchen9297731 1 NS
 Working in kitchen788917125   NS
 Setting the table88971131   0.049
 Peeling potatoes59763623521 0.015
 Burning on stove77921884 1 0.010
 Lifting frying pan by its handle25654427197122< 0.001
 Emptying potato water487834211523 < 0.001
 Reaching for sugar31664026155143< 0.001
 Making sponge cake7591206431 0.036
 Bread in and out of oven63873311431 0.005
 Washing the dishes85931372 1 NS
 Picking up dishes70872612321 0.011
Mobility indoors         
 Walking indoors88951151 1 NS
 Opening outer door83951611   0.015
 Opening yale lock789120102 1 0.026
 Answering the door8695123121 NS
 Opening balcony door84931471 1 NS
 Walking out on balcony9297631 1 NS
 Getting to the telephone83921661 1 NS
Cleaning         
 Making the bed68732726322 NS
 Dusting8396142121 0.019
 Sweeping the floor768817115 3 NS
 Cleaning kitchen floor63782517953 0.034
 Wringing out cloth466340311173 0.020
 Using vacuum cleaner537930191423 < 0.001
 Opening window7390218521 0.006
Washing clothes care         
 Washing up in bowl79901615   NS
 Putting wash in machine8998621 3 NS
 Hanging up wash67872411623 0.004
 Ironing blouse77851912322 NS
 Turning up hem of a skirt7087199745 NS
 Cutting out material6189274104240.008
 Picking up needles55703723761 NS
 Putting in and taking out plug638528157 1 0.002
 Opening/folding ironing board66922486 4 < 0.001
Transferring         
 Getting into bed929585    NS
 Turning in bed787920212   NS
 Getting out of bed808119191 1 NS
 Rising from chair778118193 2 NS
Communication         
 Using telephone929882    NS
 Holding a bock7289251032  0.009
 Writing a postcard83931373 1 NS
 Taking out money81921583   0.044
Mobility outdoors/shopping         
 Walking on flat ground82921394 1 NS
 Talking long walks4348364411911 NS
 Walking up stairs50663029143620.017
 Going by tram/bus77891510325 NS
 Getting in and out of car637832224 1 0.024
 Driving a car81931471 4 0.031
 Opening entrance doors69932575 1 < 0.001
 Opening elevator door7397224411  < 0.001
 Walking to the shops758918114 4 0.016
 Bringing home groceries305852401029 < 0.001
 Shopping on a large scale306141339520 < 0.001

Regression of activity limitation.

Activity limitations measured by HAQ and EDAQ scores were regressed by 16 variables (Figure 1). Six of these variables were significant and entered into the model (grip force, walking speed, SOFI lower extremities, pain, wellbeing, and PGA) and the explained variation (R2) was 0.55. When the HAQ score alone was regressed by the 16 variables, the same 6 significant variables were entered into the model (grip force [VIP 1.57], walking speed [VIP 1.57], SOFI lower extremities [VIP 1.44], pain [VIP 1.28], wellbeing [VIP 1.09], and PGA [VIP 1.04]) and the explained variation (R2) was 0.61. When the EDAQ score alone was regressed by the 16 variables, the same 6 significant variables were entered into the model, although grip force (VIP 1.54) changed place with walking speed (VIP 1.6), and thereafter SOFI lower (VIP 1.47), pain (VIP 1.32), wellbeing (VIP 1.12), and PGA (VIP 1.03) were entered into the model and the explained variation (R2) was 0.5.

thumbnail image

Figure 1. Partial least squares regression of the Health Assessment Questionnaire and the Evaluation of Daily Activities Questionnaire (Y variables) using the patient characteristics variables and grip force as regressors (X variables). Variable influence of projection (VIP) is given for each variable. Values >1.0 are considered significant, and significant variables are shown in bold. The explained variance (R2) is 0.50. SOFI = Signals of Functional Impairment; PGA = physicians global assessment of disease activity; ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; GAT = Grip Ability Test.

Download figure to PowerPoint

Correlation between activity limitation and grip force.

The absolute average grip force for the study group correlated strongly and negatively with both the HAQ score (Spearman's correlation coefficient = –0.632, P < 0.01) and the EDAQ score (Spearman's correlation coefficient = –0.635, P < 0.01). The HAQ score was missing in 4 of the 209 patients.

Dividing patients into subgroups based on grip force.

In total, 209 of the 217 patients could be divided into 4 subgroups based on the mean grip force values over 10 seconds (Table 4). Grip force values were missing in 8 patients. Subgroup 1 included the 88 patients with the lowest grip force (<114 N), where women were in the majority. Subgroup 2 (116–206 N) included 71 patients, again with women in the majority. Subgroup 3 (214–321 N) included the 19 women with the highest grip force and 19 men. Finally, subgroup 4 (>328 N) included 12 men with the highest grip force (Table 4). The median age was similar in subgroups 1 and 4, and the lowest median age was found in subgroup 3 (Table 4).

Table 4. Patient subgroups by grip force
Subgroup*Total patients (n = 209)Women (n = 148)Men (n = 61)
Age No.; median (range)Age No.; median (range)Grip force Mean; median (range)Age No.; median (range)Grip force Mean; median (range)
  • *

    Grip force in Newtons: 1 = <114; 2 = 116–206; 3 = 214–321; 4 = >328.

   66; 66 (5–114)13; 51 (27–80)81; 85 (39–110)
188; 59 (27–81)75; 59 (35–81)159; 157 (116–206)17; 59 (18–76)169; 173 (134–201)
271; 56 (19–76)54; 55 (19–76)262; 259 (214–312)19; 65 (31–74)264; 263 (221–321)
338; 54 (18–75)19; 37 (18–75)12; 59 (29–69)384; 387 (328–488)
412; 59 (29–69)   

Differences between subgroups.

Grip force measurement revealed significant differences when comparing subgroups (P < 0.001 for all comparisons). There were significant differences in both HAQ score and EDAQ score between subgroup 1 and subgroups 2, 3, and 4, and between subgroups 2 and 3, whereas there was no significant difference between subgroups 3 and 4 (Table 5).

Table 5. Test of differences between subgroups regarding patient characteristics, grip force, and activity limitation*
CharacteristicSubgroup 1 (n = 88)Subgroup 2 (n = 71)Subgroup 3 (n = 38)§Subgroup 4 (n = 12)ANOVA1 vs 2 P1 vs 3 P1 vs 4 P2 vs 3 P2 vs 4 P3 vs 4 P
  • *

    P values <0.05 are considered significant. Values are the median (interquartile range) unless otherwise indicated. ANOVA = analysis of variance; vs = versus; see Table 1 for additional abbreviations.

  • Grip force <114 N.

  • Grip force 116–206 N.

  • §

    Grip force 214–321 N.

  • Grip force >328 N.

Women           
 HAQ (range 0–3)1.0 (0.75)0.38 (0.75)0.13 (0.38)< 0.001< 0.001< 0.0010.005
 EDAQ (range 0–3)0.91 (0.83)0.50 (0.72)0.16 (0.33)< 0.001< 0.001< 0.0010.002
Men           
 HAQ (range 0–3)0.75 (1.0)0.5 (0.81)0.13 (0.38)0.0 (0.0)< 0.0010.0430.0020.0010.0490.003NS
 EDAQ (range 0–3)1.0 (0.72)0.58 (0.75)0.16 (0.33)0.12 (0.31)< 0.0010.0090.0010.0010.0020.007NS
Women and men           
 Grip force, N70 (44)161 (41)262 (48)387 (53)< 0.001< 0.001< 0.001< 0.001< 0.001< 0.001< 0.001
 HAQ (range 0–3)1.0 (0.75)0.38 (0.75)0.13 (0.38)0.0 (0.0)< 0.001< 0.001< 0.001< 0.001< 0.001< 0.001NS
 EDAQ (range 0–3)0.91 (0.83)0.50 (0.74)0.16 (0.33)0.12 (0.31)< 0.001< 0.001< 0.001< 0.001< 0.0010.002NS
Patient characteristics           
 ESR, mm20 (31)15 (21)12 (12)14 (18)0.008NS0.025NSNSNSNS
 CRP, mg/liter5 (15)5 (8)5 (5)5 (5)0.029NS0.0270.038NSNSNS
 Swollen joints (range 0–28)3 (9)2 (4)1 (5)3 (7)NS
 Tender joints (range 0–28)2 (7)2 (4)0 (1)1 (1)< 0.001NS< 0.0010.0480.006NSNS
 DAS-28 score3.83 (2.08)3.46 (1.87)2.30 (1.13)3.33 (2.26)< 0.0010.014< 0.001NS0.001NSNS
 PGA (range 0–4)1 (1)1 (1)1 (1)1 (1.25)< 0.001< 0.001< 0.001NS0.026NSNS
 Morning stiffness, minutes60 (75)30 (58)5 (45)70 (120)0.0010.009< 0.001NSNSNSNS
 Pain (0–100-mm VAS)43 (44)24 (39)10 (30)21 (29)< 0.0010.001< 0.0010.0450.016NSNS
 Wellbeing35 (39)28 (32)17 (37)26 (18)< 0.0010.004< 0.001NSNSNSNS
 GAT (range 0–276)22 (8)17 (10)16 (6)15 (5)< 0.0010.001< 0.0010.001NSNSNS
 SOFI hand (rang 0–16)2 (3)1 (2)0 (1)1 (2)< 0.001< 0.001< 0.0010.042NSNSNS
 SOFI upper limb (range 0–12)0 (2)0 (2)0 (1)1 (3)NS
 SOFI lower limb (range 0–16)2 (3)1 (3)0 (2)1 (2)0.0010.0080.0010.0190.010NSNS
 Walking speed, seconds14 (6)12 (4)10 (3)10 (2)0.0010.0010.0010.0010.0040.015NS

Separate analysis of women and men revealed, in both instances, significant differences between subgroups concerning grip force (P < 0.001 for all comparisons) (Figure 1). In the women, there were significant differences regarding EDAQ and HAQ scores between subgroups 1 and 2, 1 and 3, and 2 and 3. In the men, there were also significant differences between all subgroups regarding EDAQ and HAQ scores, apart from EDAQ and HAQ comparisons between subgroups 3 and 4 (Table 5).

When testing for differences in patient characteristics between subgroups, 2 variables, the number of swollen joints and SOFI upper limb, did not differ significantly between subgroups (Table 5). The majority of the patient characteristics differed significantly between subgroups 1 versus 2 and between 1 versus 3. Differences between subgroups 2 and 3 were seen concerning some variables. The only difference between subgroups 2 and 4 concerned walking time, whereas no differences were found between subgroups 3 and 4 (Table 5). Therefore, the women and men in subgroup 1, having a mean grip force <114 N, had higher disease activity and more pronounced disability compared with the women and men in subgroup 2, with a grip force between 116 and 206 N. At the same time, women and men in subgroup 2, with a grip force between 116 and 206 N, did not differ significantly regarding patient characteristics compared with the men in subgroup 4, with a grip force >328 N, although the men in subgroup 4 had significantly more pronounced activity limitations. The women and men in subgroup 3, with a grip force of 214–321 N, did not differ significantly regarding any variables (except the grouping variable grip force).

Differences within subgroups.

The mean grip force and the median HAQ score for women and men distributed in the 4 subgroups are shown in Figure 2A, and the mean grip force and the median EDAQ score are shown in Figure 2B. Within subgroups 1, 2, and 3, where women and men were grouped together according to the absolute grip force, there were no significant differences between women and men regarding difficulties in performing activities as judged by HAQ or EDAQ scores (Figure 2).

thumbnail image

Figure 2. Top, Health Assessment Questionnaire (HAQ) scores and number of women and men within subgroups. Bottom, Evaluation of Daily Activities Questionnaire (EDAQ) scores and number of women and men within subgroups.

Download figure to PowerPoint

Of the 122 activities/items in the EDAQ and HAQ, 110 (90%) did not differ between women and men within the 3 subgroups (data not presented). However, differences were found in 12 activities (8 in the EDAQ and 4 in the HAQ). In subgroup 1, women had significantly more pronounced difficulties in 2 activities in the EDAQ, i.e., opening a glass jar (P = 0.001) and opening a juice bottle (P = 0.004), and in 2 activities in the HAQ, i.e., getting in and out of a bed (P = 0.02) and opening jars (P = 0.009). Men had significantly more pronounced difficulties in 2 activities in the EDAQ, i.e., getting to the bathroom (P = 0.045) and turning in bed (P = 0.004). In subgroup 2, women had significantly more pronounced difficulties in 3 activities in the EDAQ, i.e., emptying potato water (P = 0.032), opening a bottle (P = 0.020), and using a vacuum cleaner (P = 0.048), and in 1 activity in the HAQ, i.e., vacuuming (P = 0.02). In subgroup 3, women had significantly more pronounced difficulties, as assessed by the EDAQ, concerning the lifting of a frying pan by its handle (P = 0.002), whereas men in subgroup 3 had significantly more difficulty, according to the EDAQ, buttoning and unbuttoning (P = 0.032).

Within the subgroups, men generally displayed signs of more pronounced inflammation than women. Therefore, within subgroup 1, men had a significantly more elevated ESR than women (P = 0.033), and within subgroup 2, men had a significantly higher number of swollen joints (P = 0.015) and pain (P = 0.034) than women. Furthermore, PGA (P = 0.01), GAT (P = 0.004), SOFI hand (P = 0.001), and SOFI upper extremity (P < 0.001) were more affected in men in subgroup 2. In subgroup 3, men had a significantly higher DAS28 (P = 0.025). In addition, GAT (P = 0.004), SOFI hand (P = 0.009), and walking speed (P = 0.029) were more affected compared with women in subgroup 3. Women in subgroup 3 were substantially younger than the men, whereas the opposite was found in subgroup 1.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

In the present study, we investigated the relationship between grip force and difficulty performing activities assessed by the HAQ and EDAQ. The HAQ was used on a routine basis at all followups. The EDAQ was used at 3 visits to identify difficulties in performing daily activities with the intention to facilitate planning and to evaluate the effects of assisstive devices (5). On average, the HAQ as well as the EDAQ, which both include many activities that are dependent on the upper extremities, revealed more pronounced difficulties for women. Using regression analysis, we found that grip force was the strongest regressor when HAQ and EDAQ were analyzed together. Also, when we regressed HAQ separately, grip force was the strongest regressor, closely followed by walking speed, SOFI lower extremity, and pain. The same regressors were significant when EDAQ was regressed separately, although here walking speed was strongest, closely followed by grip force. Sex and age were not identified as significant regressors of HAQ and EDAQ. The degree of activity limitation, as measured by the HAQ and the EDAQ, was associated with grip force. Therefore, we found that both women and men with low average grip force (<114 N) had substantial activity limitations, indicating a need for multiprofessional intervention. Conversely, women and men with a grip force corresponding to that of healthy women (>214 N) (33) had very low HAQ scores. Our results regarding the relationship between activity limitation and grip force corroborate the results of a recently published study where increased values in performance-based functional measures (grip strength, timed button test, and walking velocity) displayed a proportional increase in self-reported disability (34).

The finding of lower grip force values in women with RA, as compared with men with RA, is not surprising because this is also the case in healthy individuals. Women and men with RA had approximately the same relative degree of grip force reduction, resulting in grip force values of men comparable with those of healthy women in most cases. Because many of the functional abilities assessed by the HAQ and EDAQ require strong hands, it is not surprising that women with RA on average proved to be more affected than men by means of these instruments. However, when women and men with RA were subgrouped together with regard to grip force, it became evident that poor outcome assessed by the HAQ and EDAQ was related to low grip force, but not to sex (or age) per se. The activity limitations seen 3 years after the diagnosis of RA were of approximately the same magnitude as those seen after 1 and 2 years in both women and men (5). Several investigators, using the HAQ, have reported more pronounced activity limitations in women than men (2, 4, 13–15, 18, 35), and this finding, together with a majority of other self-reported outcome measures, has been considered an indication of a more severe disease course in women (16). The HAQ score has also been found to increase with increasing age in patients with RA as well as in healthy individuals (35), and Kuiper et al have argued that the postmenopausal state may be responsible for the difference in outcome between women and men (17). As judged by our results, however, grip force under or above a critical level, rather than sex, age, or postmenopausal state, may be the most important explanation to activity limitation. Therefore, women and men with comparable grip force values also had comparable outcome measures by means of the HAQ and EDAQ. Interestingly, however, when women and men were subgrouped according to grip force, the men appeared to be more severely affected with regard to measures of systemic inflammation.

In line with our findings, Krishnan et al recently reported that there were no significant differences in the HAQ score between women and men in a general population in Finland with a mean age of 55 years (33). However, besides differences in grip force, HAQ score, and EDAQ score in our cohort, there was a small but significant difference in SOFI upper limb, where men had higher scores indicating more functional impairment. This small difference is also seen in a Swedish reference population, which is age and sex matched to the TIRA cohort (unpublished observations), indicating that this difference is dependent on sex.

Grip force and walking speed from inclusion through 1-year followup in the Swedish TIRA cohort were also found to be significant regressors of self-reported variables (36). Compared with a healthy Swedish reference population (21), the grip force reduction in patients with RA with a disease duration of 12 years was ∼75% in women. Men, having approximately double the average grip force of women, end up with a normal female grip force after a 50% reduction. Therefore, the 50% grip force reduction in patients with RA, as seen in both women and men in this study, could be expected to have less impact on the HAQ and EDAQ outcomes in men. The common observation that women have a more severe form of RA than men could possibly, at least in part, be misleading due to the female-biased design of the HAQ and EDAQ, which to a great extent evaluate traditional female activities. However, Wolfe et al reported that HAQ score is an even better predictor of life expectancy in men than in women (9). After subdividing RA patients into 4 groups with respect to absolute values of grip force, the sex differences concerning HAQ and EDAQ outcomes disappeared. This, however, does not exclude that grip force/hand function and other functional abilities may also be associated with disease activity (12). Our findings highlight the fact that the HAQ score in a study group is dependent on the percentage distribution of women and men, respectively. This needs to be taken into consideration when comparing HAQ scores between study groups with different distributions of female and male patients.

Several studies have reported that grip force increases as a result of interventions (30, 37–42). The results in the present study also indicate that increasing the grip force may result in reduced activity limitations. However, further studies are needed to evaluate interventions directed to reduce disabilities and to analyze the relationships between different aspects of disability.

In conclusion, activity limitations are closely related to grip force, followed by walking speed and SOFI lower extremity, regardless of sex. By grouping patients with respect to grip force, we found that men and women with similar grip force had equal degrees of self-reported activity limitation. This finding offers an explanation as to why women report more pronounced activity limitation than men. Therefore, low grip force is closely related to activity limitation regardless of sex. At the same time, we found that subgroups of patients with low or no activity limitation had an average grip force corresponding to the average grip force of healthy women. In clinical practice, grip force measurement is simple and rapidly performed, and the absolute grip force can be used as an indicator to identify patients with activity limitations and a need for multiprofessional intervention.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Supported by grants from the Medical Research Council of Southeast Sweden, the National Board of Health and Welfare, the County Council of Östergötland, the Swedish Rheumatism Association, King Gustaf V 80-Year Foundation, and the Swedish Research Council (project K2003-74VX-14594-01A)

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES