- Email: [email protected]
Effect of light and season on pain and depression in subjects with rheumatic disorders
227
Pain, 59 (19941227-234 0 1994 Elsevier Science B.V. AI1 rights reserved 0304-3959/94/$07.00
PAIN 2591
Effect of light and season on pain and depression in subjects with rheumatic disorders 1 Donna J. Hawley ‘To,*and Frederick Wolfe bs,d a School of Nursing, Wichita State UniLjersity,Wichita. KS 67260-041 (USA), b 7’he Arthritis Research Center, St. Francis Research Institute, Wichita, KS (USA), ’ University of Kansas School of Medicine, Wichita, KS (USA), and dArthritis, Rheumatism, and Aging Medical Information System (AR&US), Palo Alto, CA (USA) (Received 3 January 1994, revision received 28 February 1994, accepted 8 March 1994)
The clinical and neurochemical association between depression and season noted in seasonal Summa~ affective disorder (SAD) has suggested that clinical pain might also be linked to season, perhaps through similar neurochemistry or the known association of depression with pain. We investigated the pain-light season/dark season hypothesis in 2523 rheumatic disease outpatients by examining VAS Pain and VAS Global Severity scores, as well as levels of depression and functional disability. No clinically significant difference in pain severity between season (or individual month) was noted for the consecutive outpatients at their first clinic visit, nor in sub-analyses using paired light and dark season visits. A slight trend toward increased pain severity in lighter months by about 3% compared to darker months was identified. No effect of season was seen on depression scores. In a subset of patients with high depression scores, rheumatoid arthritis and osteoarthritis patients, respectively, reported 16% and 7% greater pain scores in light compared to dark months, but fibromyalgia patients had stable scores. Season does not appear to play an important role in pain and/or depression in rheumatic disorders. Key words: Depression; Pain; Season; Rheumatoid
arthritis; Fibromya~gia; Osteoarthritis
Introduction Day to day changes in weather have been shown to affect rheumatic disease pain (Patberg et al. 198.5; Dequeker and Wuestenraed 1986; Laborde et al. 1986; Patberg 1987; Guedj and Weinberger 1990; Falkenbach et al. 1992). Similarly it is known that length of daylight is associated with depression (Neuwirth and Faust 1975; Rosenthal et al. 1984; Hansen et al. 1991; Hardin et al. 1991; Lam et al. 1991; Booker and Hellekson 1992; Wicki et al. 1992; Faedda et al. 1993; Rudorfer et al. 1993), which in turn links to pain (Hawley and Wolfe
* Corresponding author: Donna J. Hawley, R.N., Ed.D. Professor,
School of Nursing, Box 41, Wichita State University, Wichita, KS 67260-041, USA. Tel.: (316) 689-3610; FAX: (316) 689-3094. * These studies were supported in part by grants from the National Institutes of Arthritis, Diabetes, Digestive, and Kidney Diseases (AM21393) to the ~hrit~s, Rheumatism, and Aging Medical Information System (ARAMISI, Palo Alto, CA, USA. aSSDI 0304-3959(94)00067-O
1991). Seasonal effects (as manifested by length of daylight) on cluster and migraine headache as well as on colic have also been demonstrated (Kudrow 1987; Brewerton and George, 1990; Cugini et al. 1990; Manzoni et al. 1991; Weissbluth and Weissbluth, 1992). Finally, both season and light duration are interrelated to serotonin metabolism (Rudorfer et al. 1993). One painful rheumatic disease condition in particular, fibromyalgia, is hypothesized to be related to serotonin abnormality (Kravitz et al. 1992; Russell et al. 1992) and is known to be associated with high depression scores (Yunus et al. 1981; McCain and Scudds 1988; Hawley and Wolfe 1993.) and pain scores (Hawley et al. 1988; Hawley and Wolfe 1993.). In addition, descriptive criteria for fibromyalgia have emphasized improvement in warm weather and worsening in cold, migraine headaches, and cramping abdominal pair (Yunus et al. 1981). Fibromyalgia is a generalized pain syndrome which differs substantially from local pain disorders such as the myofascial pain syndrome by
virtue of pain severity, pain extent, functional disability. and psychological distress (Woffe et al. 1993). Although weather changes over the short term have been investigated in rheumatic disorders, no study has compared seasonal and light-associated effects on rheumatic disease pain nor as stratified by different disorders. Such information would be useful in understanding if pain is affected by season and whether pain and depression together are linked to season generally, and specifically within the construct of fibromyalgia. To clarify the reIationships between season and fight and pain, we investigated pain and its correlates in rheumatic disease by evaluating VAS Pain, Depression, Global Severity, and Functional Disability scores in 2523 consecutive patients attending an outpatient rheumatic disease clinic from 1981 to 1992 who were diagnosed as having either fibromyalgia, rheumatoid arthritis, or osteoarthritis of the knee or hip.
Methods Study sample Participants in this study were consecutive outpatients seen at the Wichita Arthritis Center, a private rheumatic disease clinic in Wichita, Kansas, from 1980 through 1992. At each clinic visit, as part of routine care, patients complete self-report questionnaires related to their pain, mood, and functional ability. In addition, patients with fibromyalgia, rheumatoid arthritis @A), and osteoarthritis (OA) may elect to participate in a separate, additional outcome study related to their illness. Those participants complete detailed questionnaires mailed to them every January and July. In general, those electing to complete the questionnaires (as opposed to non-participants) tend to be younger, better educated, and have lower levels of pain, depressive symptoms, and functional disability. They also differ from the ‘clinic’ patient in that the clinic patient is often assessed at the time of a disease flare-up while the mailed questionnaire assessments are not related to disease activity. Patients with a diagnosis of either rheumatoid arthritis (RA), fibromyalgia or osteoarthritis (OA) of the knee or hip were classified into 3 mutually exclusively groups. Rheumatoid arthritis patients met either the ARA (now ACR) 1958 criteria for definite or classical RA (Ropes et al. 1959) or the ACR 1987 criteria (Amen et al. 19881. They were included in the RA group regardless of secondary diagnoses Fibromyalgia was determined using the, 1990 ACR classification criteria (Wolfe et al. 1990) or previous criteria (Wolfe and Cathey 1983). For this study patients were classified as having fibromyalgia if they met classification criteria and did not have an in~ammato~ rheumatic disorder. Patients with OA of the knee or hip had clinical and radiographic evidence of OA of the knee or hip.
Study variables Four clinical variables, known to be sensitive to change (Wolfe et al. 1991; Hawley and Wolfe 1992; Wolfe and Hawley 19931, were utilized. Pain. Pain during the last week and global disease severity were assessed with self-administered VAS. The VAS for pain is a 15-cm double-anchored line, indexed to a continuous scale ranging from 0 = no pain to 3.0 = very severe pain, and scored in 0.1 increments. In some patients seen early in the course of data collection, pain scales were scored ordinally (0, 1, 2, 3). Such scales might be less sensitive to change. Although the results using the latter method did
not differ from those using interval scoring, the scales are not directly compatible. In this study we report both scales when available. For the purpose of understanding and comparisons, means and standard errors for the ordinal as well as the interval pain data are presented in Tables II and III. Globul ser~wity Global severity was assessed with the question ‘considering all the ways that your illness affects you, rate how you are doing’. and was scored from 0 = Very Well to 100 = Very Poor. Lkprrssion. Depression was evaluated with the AIMS Depression scale, 1 of the 9 scales from the Arthritis Impact Measurement Scale (AIMS-11 (Meenan et al. 1980). It is scored from 0 to 4.9. The AIMS Depression scale has 6 items relating to mood (i.e.. enjoyment of activities, low spirits, etc.) during the past month. Somatic items such as fatigue, appetite, or weight loss or gain are not part of the scale. The AIMS scale is strongly correlated with the Center for Epidemiological Studies-Depression (CES-D1 scale (r = 0.81) (Blalock et al. 198Yl. At the level of probable clinical depression (CES-D = 23). AIMS classification agreement with CES-D was 88% (Hawley and Wolfe 1993). Funcriond ability. Functional ability (disability) was assessed with the Stanford Health Assessment Questionnaire (HAQ) Functional Disability scale. The HAQ measures ability to do specific activities of daily living with and without assistance from others or devices. The questionnaire asks 3 questions in each of 8 areas of activities of daily living (ADL): dressing and grooming, arising, eating, walking, hygiene, reach, grip, and activities. For each area a score of 0 = No Difficulty, I = Some Difficulty, 2 = Much Djffi~l~ or with assistance, and 3 = Unable to Perform. Any activity that requires assistance from another individual or requires the use of an assistive device receives a score of 2. The highest score for each of the 8 areas is summed (range: O-24) and divided by 8 to yield, on a continuous scale. a O-3 disability index.
Data and statistical analyses To examine the possible relationships between season and the study variables of pain, depression, global disease severity, and functional ability, we performed several analyses. Analysis I. We extracted data for all first clinic visits for patients with RA, OA of the knee or hip, and ~bromy~gia for each month of the year. We compared mean scores for the study variables for each month. We also compared combined scores for the 3 dark months (November, December, and January) with scores in the 3 light months (May, June, July). The average hours of daylight per day in Kansas during these 2 periods are 10.2 and 14.4, respectively. A total of 2523 patient visits were used in this first analysis. These analyses test whether patients seen in the clinic differ in pain, mood, or function by month or by season. .4nalysis II. Patients in the clinic who had at least 2 visits, in which 1 visit was in a dark month and 1 in a light month, were similarly studied. A total of 647 patients had such paired visits, including 535 with RA, 47 with OA and 65 with fibromyalgia. If there was more than 1 visit that was paired, the last pair (most recent pair) was used. These analyses test whether scores for the study variables for patients having paired visits differed between seasons. ilnaiysis Zil. This analysis was simiiar to Analysis II except that data from mailed questionnaires rather than ciinic visits were used. It is expected that questionnaire participants would have less severe and clinic participants more severe manifestations of their illness. In Analysis I the number of visits was equally divided among the months. In Analyses II and III the number of persons having their first observation in a ‘light’ month was not significantly different from the number having a first observation in a ‘dark’ month. Data were analyzed with SAS Statistical Software version 6.04 for the PC (SAS Institute 1992). Statistical significance was set at P = 0.05, and all tests were f-tailed. ANOVA was used to compared differences between months of the year. T tests for independent
229 groups were used to compared scores for dark and light months. Paired t tests were employed for the paired data in Analyses II and III. For the ordinal VAS Pain scale data, the Kruskal-Wallis test was
used to compare months of the year and the chi-square (Ridit) analysis for comparison of the dark and light months. For the paired data from the clinic sample for the ordinal Pain scale, the signed rank test was used.
Results Demographics Demographic characteristics of the study subjects are shown in Table I. RA patients were first seen in their mid-fifties with a disease duration of slightly less than 7 years. OA patients were older (mid-sixties) than those in both other groups and, as expected, had less education and were more often widowed (8.1%) than the fibromyaigia 11.3%) and the RA (5.3%) patients. The fibromyalgia group had a higher percentage of women (90%) than did the other 2 groups. Unpaired monthly and seasonal scores (first clinic visit)
Table II displays means and standard errors for each study variable, by month as well as by dark (November, December, and January) and light (May, June, and July) 3-month periods. In this analysis each patient contributes one observation. Data displayed were results obtained at the first clinic visit. Scores for VAS interval Pain, Depression, GlobaI Severity, and Functional Disability scales did not differ either by month or season. In the chi-square analysis of the ordinal pain scale, there was a significant difference between light versus dark months (x2 = 9.60, P = 0.021, with more abnormal scores being found in light months. Differences between months were not significant (Kruskal-Wallis). Separate analyses for men and women yielded similar results (data not shown). These observations indicate that month or season of presentation generally does not influence the score of study variabIes, except for pain where slightly higher scores are seen in light months for the ordinal scale.
Paired seasonal scores: clinic patients (last paired clinic visits)
Data in Table II reflect first clinic data where the severity of the medical problem might outweigh possible seasonal effects. To eliminate this possibility, 647 patients having 2 or more visits to the clinic occurring in each of the 2 light and dark 3-month periods were studied and are presented in Table III. Milder abnormalities are noted when comparing these visits with those in Table II, reflecting differences between initial and follow-up visit data. No differences between seasons are noted for Depression, Global Severity, and Functional Disability scores, but Pain scores are slightly worse in the light than in the dark months for the group as a whole. Such significant differences, however, reflect sample size more than clinically important difference. Both the VAS interval and ordinal Pain scores were statistically significantly higher in the light months compared to the dark months for the RA patient group, although this difference was only about 3%. The OA patient group also had a 3% higher mean score for VAS interval Pain scores in the light versus dark months, but this difference was not statistically significant probably owing to the small sample size for this group (n = 47) compared to the RA group (n = 535). Depression, Global Disease Severity and HAQ Functional Disability scores did not change for any group.
Paired seasonal scores: analysis of mailed questionnaires (last paired mailed questionnaires)
Table IV presents paired questionnaire data. Results shown in this table are less abnormal than data in Tables II and III, reflecting observation taken outside of the clinic. Using responses to questionnaires completed in both a dark and light month, differences similar to those noted with the paired clinical data were found. Statistically significant differences were noted for Pain and Global Severity scores, with worse scores being obtained in the light months, Even so, these differences were small, ranging from 1.6 to 3%.
TABLE I DEMOGRAPHIC
CHARACTERISTICS
Sex (% women) Age in years: mean (SD) Education (% high school graduates) Marital status (% married) Disease duration in years: mean (SD) Ethnic origin (% White)
OF PARTICIPANTS
AT INITIAL CLINIC VISIT
pn”= 1029)
OA (n = 791)
Fibromyalgia (n = 703)
All patients (n = 2523)
70.2 54.9 (14.62) 81.3 75.8 6.8 (8.83) 95.0
72.7 63.9 (12.58) 78.2 65.3 8.7 (9.45) 93.0
90.3 48.3 (13.55) 83.1 77.1 9.2 (10.57) 95.4
76.6 55.9 (14.98) 80.8 72.6 8.1 (9.59) 94.5
II SCORES
FOR
2573 CONSEC‘C”I‘IVtl
RHEUMATI(
DISEASF:
PATIENTS
.4T THE
FIRST
<‘I lNlC_
_
1.8
CO.071 2.8 CO.151 49.5 (1.73) 1.0 (0.05)
(1.58) 1.0 fo.05)
1.8
48.9 (0.97) 1 .o (0.03)
10.08)
(0.04) 3.0
1.7 (0.07) 1.9 iO.061 3.2 (0.14) 52.2 (1.56) 1.1 (0.05)
2.0 (0.08) 2.0 (0.06) 3.0 (0.14) 51.6 (1.63) 1.1 (0.05 1
(0.11)) 2.0 (0.07) 3.2 (0.17) 51.6 (1.72) 1.0 (0.05)
I.6
May 1.8 (0.09) 1.8 (O.QXf 3.1 0.18) 50.0 (1.80) 1.0 (0.05f
1.8 (0.05)
1.8 (0.08) 1.8 (0.08) 2.9 (0.14) 49.7
1.8 (0.08)
1.x (O.UY) 1.9 (Q.Olt} 3.0 (0.15) 48.5 (1.56) 1.0 (0,04)
1.8 tu.08) (0.07) 3.0 (0.16) 48.5 (I .62) I.0 (Q.05,
2.0
July
--____
June
_____.--._._ Lightest months
1.7 (0.10) 1.9 (0.061 3.2 (0.14) 47.7 (1.73) 1.1 (0.05)
* Continuous VAS Pain scale (n = 958). ** Ordinal VAS Pain scale (n = 1565).
VAS Pain * * (O-3) AIMS Depressian (O-3) VAS Global Severity (O-100) HAQ Functional Disability (O-3)
(O-3)
VAS Pain *
Apr
ALL-D
Mar
Jan
Feb
Dee
mi&ths
Nov
Darkest
(0.92) 1.0 CO.031
1.x ~0.05) 2.0 f0.05) 3.0 (0.08) 49.6
ALL-L
__--_-_
1.9 (0.07) I.9 (0.07) 3.2 (0.17) 51.1 (I.761 1.1 (C~.O5)
Sept
__._ ~_._
1.6 (0.07) i.8 (il.06 ) 3.3 (0.131 48.Y (I.501 1.0 (0.04)
hug
(I .h8) 1.1 ~0.1151
I .8 iQ.OXf 37.C (0.14) 51.7
(0.0X)
I.8
Ott
ALL-D are combined scoreb for the 3-month (dark) period: November. December, Januar>. ALL-L are combined ‘Icores for the 3-month (light) period: Ma>. June .luly. Monthl! scores tor all continuous variahh+s were not significanily different by ANOVA: and ALL-I. and ALL-D did not differ significantly by r test. Monthly scores for VAS Pain. ol-din;11 ~c;IIc. did not diffcl significantly by Kruskat-Wallis test. ALL-D and ALL-L, differed significantly by chi-square (Kidit) analysis (x” = 9.hO. P = 0.02).
PAIN. DEPRESSION. GLOBAL SEVERITY AND FUNCTLONAL DISABILITY VISIT BY MONTH AND BY DARKEST AND LIGHTEST j-MONTH PERIODS
TABLE
231
TABLE III PAIRED SCORES FQR PARTICIPANTS
SEEN IN THE CLINIC IN 1 LIGHT AND 1 DARK MONTH
Light
Dark VAS Pain * (O-3) VAS Pain ** (O-3) AIMS Depression (O-10) VAS Global severity (0-100~ HAQ Functional Disability (O-3)
1.5 + (0.03) 1 St+’ (0:04) 2.6 (0.07) 48.6 (0.98) 1.5 (0.03)
1.5
(0.04) 1.6 (0.04) 2.6 CO.071 47.5 (0.931 1.5 (0.03)
All Patients n = 647
Fibromyalgia n = 65
Osteoarthritis n = 47
RA n = 535
Dark
Light
(E4f
(0.13) 1.8 fO.14)
2.1 (0.08) 1.9 (0.12)
3.0 (0.28) 49.9 (3.60) 1.3 (0.101
&, 51.8 (3.49) 1.2 (0.081
327, 57.2 (2.441 1.4 f0.081
1.7
Dark
Light
Dark
Light 1.6” (0.03) 1.8 +++i (0.04) 2.7 (0.071 49.8 (0.90) 1.5 CO.031
Continuous VAS Pain scale (n = 457) Ordinal VAS Pain scale (n = 295) Mean difference = 0.081, P = 0.032 by paired t test. (Rounding of mean values makes values appear equal.) Mean difference = 0.065, P = 0.05 by paired t test. P = O.OU5by signed rank test. P = 0.005 by signed rank test.
As shown in Table IV, HAQ Functional Disability scores did not change by season for any diagnostic group. VAS Global Severity scores were 1.7% higher in the light versus dark months for the total group (42.1 vs. 40.41, 1.6% and 2% higher in the light months for the OA group (40.5 vs. 38.9) and RA group (41.1 vs. 39.1) respectively. Essentially no change was noted for the fibromyalgia patients for this variable. Pain scores showed the most change. Pain scores were about 3% higher in the light rather than dark months for the totai sample (1.3 vs. 1.2) and for RA 11.2 vs. 1.1) and the OA (1.3 vs. 1,2). The fibromyalgia group also had a 3% difference in VAS Pain scores; however, the fibromyalgia patients reported higher Pain
scores in the dark rather light months (1.7 vs. 1.6). This difference was not statistically significant. AIMS Depression scores did not vary for the total sample or for RA patients, and only minimally for the OA group (1%; 2.4 vs. 2.3). However, the fibromyalgia patients had on average 3% higher Depression scores in the light vs. dark months (3.1 vs. 2.8). Nevertheless scores for depression differences were not statistically significant for any subgroup or for the group as a whole, Paired ~~estio~~aire scores: examination in depressed patients
To examine more fully the Depression scores and their relationship to pain, as well as to the seasons, we
TABLE IV PAIRED SCORES USING QUESTIONNAIRE
RESPONSES FROM 1 LIGHT AND 1 DARK MONTH
Scores are mean f SEM. Paired t tests were used for all comparisons. RA (n = 5891 Dark VAS Pain (O-31 AIMS Depression (O-10) VAS Global Severity (O-100) HAQ Functional Disability (O-3) * ** * ** +
Difference Difference Difference Difference ++ Difference
P P P P P
= = = = =
0.006. 0.017, 0.005. 0.001. 0.006.
Light
Dark 1.2 (0.05)
fiti7, 39.1 (0.96) 1.3 (0.03)
Fibromyalgia (n = 130)
tonA_260)
2.3 {O.ll) 38.9 (1.401 1.0 (0.04)
All patients (II = 979)
Light
Dark
Light
Dark
Light
1.3 *** (0.051
&7)
1.6 (0.07) 3.1 (0.19) 49.9 (2.101
1.2 (0.03) 2.4 f0.06) 40.4 (2.341
1.3’ (0.03) 2.4 (0.061 42.1 ++ (0.77) 1.2 (0.03)
& 40.5 (1.46) 1.0 (0.041
2.8 to.171 49.0 (2.08) 1.1 (0.06)
TABLE V PAIRED SCORES FOR QUESTIONNAIRE EITHER DARK OR LIGHT MONTHS
RESPONSES FOR PARTICIPANTS REPORTING
AIMS DEPRESSION SCORE
)* 4.00 IN
Score5 are mean + SEM. Paired I tests were used for all comparisons. RA (n = X9)
VAS Pain ((l-31 AIMS Depression (O-10) VAS Global Severity (O-100) HAQ Functional Disability (O-3)
OA (n = 46)
Fibromyalgia (n = 34)
Dark
Light
Dark
1.7 (0.081 4.5 (0.22) 57.1 (2.64)
2.0 * (0.07) 5.6 * (0.14) 63.9 a** 12.49) 1.9 (0.08)
I.6 (0.11 f 4.2 (0.78) 51.5 13.62)
(0.11)
(0.11)
5.4 * (0.17) 52.x 13.21)
I .3
I ..s
(0.09)
(0.09)
4.6 (0.39) 58.7 14.38) I.4 (0.13)
1.8 (0.0X)
Light
Dark
1X1
2.0
All patients (n = 169) Light
Dark
Light
2.0 0.13) 5.9 ** (0.30) 61.3 (4.01 J 1.4 (0.14)
1.7 (0.061 4.4 (0.16) 55.9 ( 1.92) 1.6 (0.06)
2.0 * (0.54) 5.6 * 10.1I) 60.4 ****
(I .80) 1.7 (O.06)
* Difference PO.001,
** Difference PO.004. *** Difference PO.008. **** Difference PO.006. ‘ Difference PO.063.
selected a sub-sample of persons with Depression scores greater than or equal to 4.0 in either the light or dark months. We hypothesized that within the group of depressed patients we might be more likely to detect an effect of season. A score of 4.0 on the AIMS Depression scale is consistent with serious depressive symptoms (Hawley and Wolfe 1993). For this analysis questionnaire data was used since the sample number of depressed persons was greater than in the paired clinic data. Table V shows the means, standard errors, and the significance levels for all variables. Depression scores differed by season for all diagnostic groups. The difference between light and dark months ranged from 1.1 points (11% change) for the RA sub-sample to 1.3 points (13% change) for the fibromyalgia patients. Regardless of group, scores were higher in the light months. Pain scores showed similar changes for the total group and for the RA and OA groups. RA patients had 16% higher scores in light rather than dark months and OA patients scored 6.7% higher. Fijromyalgia scores were constant across season, remaining high with a mean of 2.0. For Global Severity no differences exceeded 7 points on the lo-point scale (the light months were favored); and the HAQ Functional Disability Scale was essentiaily stable.
Discussion
Our study finds little relationship between season and depression and between season and rheumatic disease pain, global severity and functional ability. What difference there is links lighter months (May
through July) with severity. Our study differs from previous efforts by its large sample, unselected patients with regard to symptoms of pain or depression, and reliance on data actually collected during the season or month being studied. In addition, we have stratified subjects by rheumatic disease diagnosis to obtain a clearer picture of the effect of season. There were a number of reasons to suspect that season might be associated with pain and/or depression in rheumatic disease patients in colder, darker months. Rheumatic disease patients have disabilities that might be particularly limiting during these months, and cold may exacerbate symptoms of stiffness and pain while warmth and heat is used therapeutically. In addition, a subset of rheumatic disease patients with a widespread chronic pain syndrome (fibromyalgia) have higher levels of depression and pain than do other rheumatic disease patients (Hawley et al. 1988; Hawley and Wolfe 19931, disturbed diurnal neurohypophyseal function (McCain and Tilbe 1989), seratonin and other CNS mediator abnormalities (Russell 1989; Russell et al. 1992a,b), a disturbed diurnal sleep pattern, and a demonstrated worsening in cold, dark seasons (by history) (Yunus et al. 1981). In spite of these considerations, we did not find evidence for worse pain or depression in darker or colder months in any of the rheumatic disorders studied, in accord with observations that others have noted in osteoarthritis (Harris 1986; Laborde et al. 1986). The link between depression and season has been clearly demonstrated in SAD (Rosenthal et al. 1984). In this syndrome patients suffer recurring depressions in dark months and have day-time somnolence, weight gain, and carbohydrate craving (Rosenthal et al 1984.
233
1987). Our study, however, did not investigate SAD features (except for depressive symptoms), and we can not draw conclusions concerning SAD. Even so, if a clinically important number of our patients had SAD we would have suspected it to have been reflected in higher Depression scores in the dark months, something we did not see. We considered the possibility that studying patients who were not depressed could have influenced our results. We therefore also studied a subset of patients with high Depression scores. Within this group we found even more evidence for generalized worsening in light months. In an additional attempt to investigate a possible link with SAD, we studied correlations between changes in weight and changes in Depression scores in the patients with paired clinic visits. No statistical association was seen in the group as a whole, in the 3 diagnostic groups separately, and in those who gained more than 2 kg or 4 kg in the dark months. Finally, we applied exploratory data smoothing algorithms, including FFT, Lowess, Savitsky-Golay, and sequential polynomial interpolation (TableCurve, version 1.0, Jandel Scientific, San Rafael, CA) to attempt to identify seasonal patterns. Unlike Terman et al. (1989) who demonstrated a seasonal pattern for SAD symptoms in a random survey of New York residents, we were unable to identify such patterns among depression and pain scores utilized in our study. One factor that might have confounded our results is the finding (in a coastal population of RA patients) that increasing temperature is associated with higher levels of pain, a fact that is possibly explained by a stronger relationship between pain and vapor pressure in summer than winter (Patberg et al. 1985). Other studies, however, in non-coastal regions (similar to the geographic area of our report) did not find relationships between weather and rheumatic disease pain (Sibley 1985). Another possible confounding effect could be the general increase in depressive symptoms among those with chronic illness, a factor which could conceivably mask seasonal depression (there have been no studies investigating this point, however) (Wells et al. 1988; Wolfe et al. 1993). Finally, it is possible that an association between season and pain and depression not seen here might be found in areas such as Scandinavia where greater differences exist between light and dark seasons. Thus the results of this investigation do not find associations between dark, colder months and depressive symptoms; we find slightly the opposite. Nor do we find increased pain in these months, regardless of the rheumatic conditions studied. What trend there is favors severity in lighter months by about 3% compared with darker months. Season does not appear to play an important role in pain and/or depression in rheumatic diseases,
References Arnett, F.C., Edworthy, SM., B&h, D.A., McShane, D.J., Fries, J.F., Cooper, N.S., Healy, L.A., Caplan, S.R., Liang, M.H., Luthra, H.S., Medsger, T.A., Jr., Mitchell, D.M., Neustadt, D.H., Pinals, RX, Schaller, J.G., Sharp, J.T., Wilder, R.L. and Hunder, G.G., The American Rheumatism Association, 1987 revised criteria for the classification of rheumatoid arthritis, Arthr. Rheum., 31 (1988) 315-324. Blalock, S.J., de’lrellis, R.F., Brown, G.K. and Wallston, K.A., Vahdity of the Center for Epidemiological Studies depression scale in arthritis populations, Arthr. Rheum., 32 (1989) 991-997. Booker, J.M. and Heltekson, C.J., Prevalence of seasonal affective disorder in Alaska, Am. J. Psychiat., 149 (1992) ‘1176-1182. Brewerton, T.D. and George, MS., A study of the seasonal variation of migraine, Headache, 30 11990) 511-513. Cugini, P., Romit, A., Di Palma, L. and Giacovazzo, M., Common migraine as a weekly and seasonal headache, Chronobiol. Int., 7 (1990) 467-469. Dequeker, J. and Wuestenraed, L., The effect of biometeorological factors on Ritchie articular index and pain in rheumatoid arthritis, Stand. J. Rheumatol., 15 (1986) 280-284. Faedda, G.L., Tondo, L., Teicher, M.H., Baldessarini, R.J., Gelbard, H.A. and Floris, G.F., Seasonal mood disorders. Patterns of seasonal recurrence in mania and depression, Arch. Gen. Psychiat., 50 (1993) 17-23. Fa~enbach, A., Gottschalk, R., Hadji, F. and Kaltwasser, J.P., Sensitivity to weather in patients with rheumatoid arthritis or psoriatic arthritis, Sem. Hop. Paris, 68 (1992) 781-783. Fries, J.F., Spitz, P.W. and Kraines, R.G., Measurement of patient outcome in arthritis, Arthr. Rheum., 23 (1980) 137-145. Guedj, D. and Weinberger, A., Effect of weather conditions on rheumatic patients, Ann. Rheum. Dis., 49 (1990) 158-159. Hansen, V., Jacobsen, B.K. and Husby, R., Mental distress during winter. An epidemiologic study of 7759 adults north of Arctic Circle, Acta Psychiat. Stand., 84 (1991) 137-141. Hardin, T.A., Wehr, T.A., Brewerton, T., Kasper, S., Berrettini, W., Rabkin, J. and Rosenthal, N.E., Evaluation of seasonal&y in six clinical populations and two normal populations, J. Psychiat. Res., 25 (1991) 75-87. Harris, CM., Further observations on seasonal variation. 1. Osteoarthritis, J. Roy. COIL Gen. Pratt., 36 (1986) 316-318. Hawley, D.J. and Wolfe, F., Pain, disability, and pain/disability relationships in seven rheumatic disorders: a study of 1522 patients, J. Rheumatol., 18 (1991) 1552-1557. Hawley, D.J. and Wolfe, F., Sensitivity to change of the health assessment questionnaire (HAQ) and other clinical and health status measures in rheumatoid arthritis: results of short-term clinical trials and observational studies versus long-term observational studies, Arthr. Care Res., 5 (1992) 130-136. Hawley, D.J. and Wolfe, F., Depression is not more common in rheumatoid arthritis: a 10 year longitudinal study of 6,608 rheumatic disease patients, J. Rheumatol., in press. Hawley, D.J., Wolfe, F. and Cathey, M.A., Pain, functional disability, and psychological status: a 12-month study of severity in fibromyalgia, J. Rheumatol., 15 (1988) 1551-1556. Kravitz, H.M., Katz, R., Kot, E., Helmke, N. and Fawcett, J., Biochemical clues to a fibromyalgia-depression link - imipramine binding in patients with fibromyalgia or depression and in healthy controls, J. Rheumatol., 19 (1992) 1428-1432. Kudrow, L., The cyclic relationship of natural illumination to cluster period frequency, Cephalalgia, 7 (Suppl. 6) (1987) 76-78. Laborde, J.M., Dando, W.A. and Powers, M.J., Influence of weather on osteoarthritics, Sot. Sci. Med., 23 (1986) 549-554. Lam, R.W., Soiyom, L. and Tompkins, A., Seasonal mood symptoms in bulimia nervosa and seasonal affective disorder, Comp. Psychiat., 32 (1991) 5.52-558.
Manzoni. G.C., Micieli, G., Zanferrari, C., Sandrini, G.. Bizzi, P. and Nappi, G., Cluster headache. Recent developments in clinical characterization and pathogenesis. Acta Neurol. (Napoli), 13 (1991) 506-513.
SAS Institute, SAS/STAT Users Guide: Version Gary, NC, 1992. Sibley, J.T., Weather and arthritis symptoms. (1985) 707-710.
McCain, G.A. and Scudds, R.A., The concept of primary fibromyalgia (fibrositis): clinical value, relation and significance to other chronic musculoskeletal pain syndromes, Pain, 33 (1988) 273-287. McCain, G.A. and Tilbe, K.S., Diurnal hormone variation in fibromyalgia syndrome: a comparison with rheumatoid arthritis, J. Rheumatol. (SuppI.), 19 (1989) 154-157. Meenan, R.F., Gertman, P.M. and Mason, J.H., Measuring health status in arthritis: the arthritis impact measurment scales, Arthr. Rheum.. 23 (1980) 146-152. Neuwirth, R. and Faust, V., Neue medizinmeteorologische ergebnisse an psychisch kranken, Munch. Med. Wochenschr., 117 (1975) 431-436. Patberg, W.R., Effect of weather on daily pain score in rheumatoid arthritis (letter). Lancet, ii (1987) 386-387. Patberg, W.R.. Nienhuis, R.L. and Veringa, F., Relation between meteorological factors and pain in rheumatoid arthritis in a marine climate, J. Rheumatol., 12 (1985) 711-715. Ropes, M.W., Bennett, G.A., Cobb, S., Jacox, R. and Jessar, R.A., 1958 Revision of diagnostic criteria for rheumatoid arthritis, Arthr. Rheum., 2 (1959) 16-20. Rosenthal, N.E., Sack, D.A., Gillin, J.C., Levy, A.J., Goodwin, F.K., Davenport, Y., Mueller, P.S., Newsome, D.A. and Wehr, T.A., Seasonal affective disorder: a description of the syndrome and preliminary findings with light therapy, Arch. Gen. Psychiat., 41 (1984) 72-80. Rosenthal, N.E., Genhart, M.J., Jacobson, F.M., Skwerer, R.G. and Wehr. T.A., Disturbances of appetite and weight regulation in seasonal affective disorder, Ann. NY Acad. Sci., 499 (1987) 216-230. Rudorfer, M.V., Skwerer, R.G. and Rosenthal, N.E., Biogenic amines in seasonal affective disorder: effects of light therapy, Psychiat. Res., 46 (1993) 19-28. Russell, I.J., Neurohormonal aspects of the fibromyalgia sundrome, Rheum. Dis. Clin. N. Am., 15 (1989) 149-168. Russell, I.J.. Michalek, J.E., Vipraio, G.A., Fletcher, E.M., Javors, M.A. and Bowden, C.A., Platelet H-3-imipramine uptake receptor density and serum serotonin levels in patients with fibromyalgia/fibrositis syndrome, J. Rheumatol., 19 (1992a) 104-109. Russell, I.J., Vaeroy, H., Javors, M. and Nyberg, F., Cerebrospinal fluid biogenic amine metabolites in fibromyalgia/fibrositis syndrome and rheumatoid arthritis, Arthr. Rheum., 35 (1992b) 550556.
Terman, M., Botticelli, S.R., Link, B.G., Link, M.J., Quitkin, F.M.. Hardin, T.E. and Rosenthal, N.E.. Seasonal symptom patterns in New York: patients and population. In: C. Thompson and T. Silverstone (Eds.), Seasonal Affective Disorder, Clin. Neurosci. Publ., London, 1989, pp. 77-95. Weissbluth, M. and Weissbluth, L.. Colic, sleep inertia, melatonin and circannual rhythms, Med. Hypotheses, 38 (1992) 224-228. Wells, K.B., Golding, J.M. and Burnam, M.A.. Psychiatric disordelin a sample of the general population with and without chronic medical conditions, Am. J. Psychiat., 145 (1988) 976-981. Wicki, W., Angst, J. and Merikangas. K.R., The Zurich Study. XIV. Epidemiology of seasonal depression, Eur. Arch. Psychiat. Clin. Neurosci., 241 (1992) 301-306. Wolfe, F. and Cathey, M.A., Prevalence of primary and secondary fibrositis, J. Rheumatol., IO (1983) 965-968. Wolfe. F. and Hawley, D.J., The relationship between clinical activity and depression on rheumatoid arthritis, J. Rheumatol. 20, (1993) 2032-2037. Wolfe, F., Smythe, H.A.. Yunus, M.B., Bennett, R.M.. Bombardier. C.. Goldenberg, D.L., Tugwell, P., Abeles, M., Campbell, S.M., Clark, P., Fam, A.G., Farber, S.J.. Fiechtner, J.J., Franklin. C.M., Gatter, R.A., Hamaty, D., Lessard, J., Lichtbroun, A.S., Masi, A.T.. McCain, G.A., Reynolds, W.J., Romano, T.J., Russell. I.J. and Sheon, R.P., The American College of Rheumatology, 1990 criteria for the classification of fibromyalgia: report of the multicenter criteria committee, Arthr. Rheum., 33 (1990) 160-172. Wolfe, F.. Hawley, D.J. and Cathey, M.A., The assessment and prediction of functional disability in RA, J. Rheumatol., 18 (1991) 1298-1306. Wolfe, F., Anderson, J., Ross, K. and Russell, I.J., The prevalence and characteristics of fibromyalgia in the general population, Arthr. Rheum., 36 (1993) S48 (abstr.). Wolfe, F., Hawley. D.J., Ross, K. and Anderson, J., Depression is increased two-fold in RA and clinically significant arhtritis, Arthr. Rheum., 20 (1993) 2025-2031. Yunus, M.B., Masi, A.T., Caiabro, J.J., Miller, K.A. and Feigenbaum, S.L., Primary fibromyalgia (fibrositis): clinical study of 50 patients with matched normal controls, Semin. Arthr. Rheum.. 11 (1981) 151-171.
6.08. SAS Institute. J. Rheumatol..
12
Pain, 59 (19941227-234 0 1994 Elsevier Science B.V. AI1 rights reserved 0304-3959/94/$07.00
PAIN 2591
Effect of light and season on pain and depression in subjects with rheumatic disorders 1 Donna J. Hawley ‘To,*and Frederick Wolfe bs,d a School of Nursing, Wichita State UniLjersity,Wichita. KS 67260-041 (USA), b 7’he Arthritis Research Center, St. Francis Research Institute, Wichita, KS (USA), ’ University of Kansas School of Medicine, Wichita, KS (USA), and dArthritis, Rheumatism, and Aging Medical Information System (AR&US), Palo Alto, CA (USA) (Received 3 January 1994, revision received 28 February 1994, accepted 8 March 1994)
The clinical and neurochemical association between depression and season noted in seasonal Summa~ affective disorder (SAD) has suggested that clinical pain might also be linked to season, perhaps through similar neurochemistry or the known association of depression with pain. We investigated the pain-light season/dark season hypothesis in 2523 rheumatic disease outpatients by examining VAS Pain and VAS Global Severity scores, as well as levels of depression and functional disability. No clinically significant difference in pain severity between season (or individual month) was noted for the consecutive outpatients at their first clinic visit, nor in sub-analyses using paired light and dark season visits. A slight trend toward increased pain severity in lighter months by about 3% compared to darker months was identified. No effect of season was seen on depression scores. In a subset of patients with high depression scores, rheumatoid arthritis and osteoarthritis patients, respectively, reported 16% and 7% greater pain scores in light compared to dark months, but fibromyalgia patients had stable scores. Season does not appear to play an important role in pain and/or depression in rheumatic disorders. Key words: Depression; Pain; Season; Rheumatoid
arthritis; Fibromya~gia; Osteoarthritis
Introduction Day to day changes in weather have been shown to affect rheumatic disease pain (Patberg et al. 198.5; Dequeker and Wuestenraed 1986; Laborde et al. 1986; Patberg 1987; Guedj and Weinberger 1990; Falkenbach et al. 1992). Similarly it is known that length of daylight is associated with depression (Neuwirth and Faust 1975; Rosenthal et al. 1984; Hansen et al. 1991; Hardin et al. 1991; Lam et al. 1991; Booker and Hellekson 1992; Wicki et al. 1992; Faedda et al. 1993; Rudorfer et al. 1993), which in turn links to pain (Hawley and Wolfe
* Corresponding author: Donna J. Hawley, R.N., Ed.D. Professor,
School of Nursing, Box 41, Wichita State University, Wichita, KS 67260-041, USA. Tel.: (316) 689-3610; FAX: (316) 689-3094. * These studies were supported in part by grants from the National Institutes of Arthritis, Diabetes, Digestive, and Kidney Diseases (AM21393) to the ~hrit~s, Rheumatism, and Aging Medical Information System (ARAMISI, Palo Alto, CA, USA. aSSDI 0304-3959(94)00067-O
1991). Seasonal effects (as manifested by length of daylight) on cluster and migraine headache as well as on colic have also been demonstrated (Kudrow 1987; Brewerton and George, 1990; Cugini et al. 1990; Manzoni et al. 1991; Weissbluth and Weissbluth, 1992). Finally, both season and light duration are interrelated to serotonin metabolism (Rudorfer et al. 1993). One painful rheumatic disease condition in particular, fibromyalgia, is hypothesized to be related to serotonin abnormality (Kravitz et al. 1992; Russell et al. 1992) and is known to be associated with high depression scores (Yunus et al. 1981; McCain and Scudds 1988; Hawley and Wolfe 1993.) and pain scores (Hawley et al. 1988; Hawley and Wolfe 1993.). In addition, descriptive criteria for fibromyalgia have emphasized improvement in warm weather and worsening in cold, migraine headaches, and cramping abdominal pair (Yunus et al. 1981). Fibromyalgia is a generalized pain syndrome which differs substantially from local pain disorders such as the myofascial pain syndrome by
virtue of pain severity, pain extent, functional disability. and psychological distress (Woffe et al. 1993). Although weather changes over the short term have been investigated in rheumatic disorders, no study has compared seasonal and light-associated effects on rheumatic disease pain nor as stratified by different disorders. Such information would be useful in understanding if pain is affected by season and whether pain and depression together are linked to season generally, and specifically within the construct of fibromyalgia. To clarify the reIationships between season and fight and pain, we investigated pain and its correlates in rheumatic disease by evaluating VAS Pain, Depression, Global Severity, and Functional Disability scores in 2523 consecutive patients attending an outpatient rheumatic disease clinic from 1981 to 1992 who were diagnosed as having either fibromyalgia, rheumatoid arthritis, or osteoarthritis of the knee or hip.
Methods Study sample Participants in this study were consecutive outpatients seen at the Wichita Arthritis Center, a private rheumatic disease clinic in Wichita, Kansas, from 1980 through 1992. At each clinic visit, as part of routine care, patients complete self-report questionnaires related to their pain, mood, and functional ability. In addition, patients with fibromyalgia, rheumatoid arthritis @A), and osteoarthritis (OA) may elect to participate in a separate, additional outcome study related to their illness. Those participants complete detailed questionnaires mailed to them every January and July. In general, those electing to complete the questionnaires (as opposed to non-participants) tend to be younger, better educated, and have lower levels of pain, depressive symptoms, and functional disability. They also differ from the ‘clinic’ patient in that the clinic patient is often assessed at the time of a disease flare-up while the mailed questionnaire assessments are not related to disease activity. Patients with a diagnosis of either rheumatoid arthritis (RA), fibromyalgia or osteoarthritis (OA) of the knee or hip were classified into 3 mutually exclusively groups. Rheumatoid arthritis patients met either the ARA (now ACR) 1958 criteria for definite or classical RA (Ropes et al. 1959) or the ACR 1987 criteria (Amen et al. 19881. They were included in the RA group regardless of secondary diagnoses Fibromyalgia was determined using the, 1990 ACR classification criteria (Wolfe et al. 1990) or previous criteria (Wolfe and Cathey 1983). For this study patients were classified as having fibromyalgia if they met classification criteria and did not have an in~ammato~ rheumatic disorder. Patients with OA of the knee or hip had clinical and radiographic evidence of OA of the knee or hip.
Study variables Four clinical variables, known to be sensitive to change (Wolfe et al. 1991; Hawley and Wolfe 1992; Wolfe and Hawley 19931, were utilized. Pain. Pain during the last week and global disease severity were assessed with self-administered VAS. The VAS for pain is a 15-cm double-anchored line, indexed to a continuous scale ranging from 0 = no pain to 3.0 = very severe pain, and scored in 0.1 increments. In some patients seen early in the course of data collection, pain scales were scored ordinally (0, 1, 2, 3). Such scales might be less sensitive to change. Although the results using the latter method did
not differ from those using interval scoring, the scales are not directly compatible. In this study we report both scales when available. For the purpose of understanding and comparisons, means and standard errors for the ordinal as well as the interval pain data are presented in Tables II and III. Globul ser~wity Global severity was assessed with the question ‘considering all the ways that your illness affects you, rate how you are doing’. and was scored from 0 = Very Well to 100 = Very Poor. Lkprrssion. Depression was evaluated with the AIMS Depression scale, 1 of the 9 scales from the Arthritis Impact Measurement Scale (AIMS-11 (Meenan et al. 1980). It is scored from 0 to 4.9. The AIMS Depression scale has 6 items relating to mood (i.e.. enjoyment of activities, low spirits, etc.) during the past month. Somatic items such as fatigue, appetite, or weight loss or gain are not part of the scale. The AIMS scale is strongly correlated with the Center for Epidemiological Studies-Depression (CES-D1 scale (r = 0.81) (Blalock et al. 198Yl. At the level of probable clinical depression (CES-D = 23). AIMS classification agreement with CES-D was 88% (Hawley and Wolfe 1993). Funcriond ability. Functional ability (disability) was assessed with the Stanford Health Assessment Questionnaire (HAQ) Functional Disability scale. The HAQ measures ability to do specific activities of daily living with and without assistance from others or devices. The questionnaire asks 3 questions in each of 8 areas of activities of daily living (ADL): dressing and grooming, arising, eating, walking, hygiene, reach, grip, and activities. For each area a score of 0 = No Difficulty, I = Some Difficulty, 2 = Much Djffi~l~ or with assistance, and 3 = Unable to Perform. Any activity that requires assistance from another individual or requires the use of an assistive device receives a score of 2. The highest score for each of the 8 areas is summed (range: O-24) and divided by 8 to yield, on a continuous scale. a O-3 disability index.
Data and statistical analyses To examine the possible relationships between season and the study variables of pain, depression, global disease severity, and functional ability, we performed several analyses. Analysis I. We extracted data for all first clinic visits for patients with RA, OA of the knee or hip, and ~bromy~gia for each month of the year. We compared mean scores for the study variables for each month. We also compared combined scores for the 3 dark months (November, December, and January) with scores in the 3 light months (May, June, July). The average hours of daylight per day in Kansas during these 2 periods are 10.2 and 14.4, respectively. A total of 2523 patient visits were used in this first analysis. These analyses test whether patients seen in the clinic differ in pain, mood, or function by month or by season. .4nalysis II. Patients in the clinic who had at least 2 visits, in which 1 visit was in a dark month and 1 in a light month, were similarly studied. A total of 647 patients had such paired visits, including 535 with RA, 47 with OA and 65 with fibromyalgia. If there was more than 1 visit that was paired, the last pair (most recent pair) was used. These analyses test whether scores for the study variables for patients having paired visits differed between seasons. ilnaiysis Zil. This analysis was simiiar to Analysis II except that data from mailed questionnaires rather than ciinic visits were used. It is expected that questionnaire participants would have less severe and clinic participants more severe manifestations of their illness. In Analysis I the number of visits was equally divided among the months. In Analyses II and III the number of persons having their first observation in a ‘light’ month was not significantly different from the number having a first observation in a ‘dark’ month. Data were analyzed with SAS Statistical Software version 6.04 for the PC (SAS Institute 1992). Statistical significance was set at P = 0.05, and all tests were f-tailed. ANOVA was used to compared differences between months of the year. T tests for independent
229 groups were used to compared scores for dark and light months. Paired t tests were employed for the paired data in Analyses II and III. For the ordinal VAS Pain scale data, the Kruskal-Wallis test was
used to compare months of the year and the chi-square (Ridit) analysis for comparison of the dark and light months. For the paired data from the clinic sample for the ordinal Pain scale, the signed rank test was used.
Results Demographics Demographic characteristics of the study subjects are shown in Table I. RA patients were first seen in their mid-fifties with a disease duration of slightly less than 7 years. OA patients were older (mid-sixties) than those in both other groups and, as expected, had less education and were more often widowed (8.1%) than the fibromyaigia 11.3%) and the RA (5.3%) patients. The fibromyalgia group had a higher percentage of women (90%) than did the other 2 groups. Unpaired monthly and seasonal scores (first clinic visit)
Table II displays means and standard errors for each study variable, by month as well as by dark (November, December, and January) and light (May, June, and July) 3-month periods. In this analysis each patient contributes one observation. Data displayed were results obtained at the first clinic visit. Scores for VAS interval Pain, Depression, GlobaI Severity, and Functional Disability scales did not differ either by month or season. In the chi-square analysis of the ordinal pain scale, there was a significant difference between light versus dark months (x2 = 9.60, P = 0.021, with more abnormal scores being found in light months. Differences between months were not significant (Kruskal-Wallis). Separate analyses for men and women yielded similar results (data not shown). These observations indicate that month or season of presentation generally does not influence the score of study variabIes, except for pain where slightly higher scores are seen in light months for the ordinal scale.
Paired seasonal scores: clinic patients (last paired clinic visits)
Data in Table II reflect first clinic data where the severity of the medical problem might outweigh possible seasonal effects. To eliminate this possibility, 647 patients having 2 or more visits to the clinic occurring in each of the 2 light and dark 3-month periods were studied and are presented in Table III. Milder abnormalities are noted when comparing these visits with those in Table II, reflecting differences between initial and follow-up visit data. No differences between seasons are noted for Depression, Global Severity, and Functional Disability scores, but Pain scores are slightly worse in the light than in the dark months for the group as a whole. Such significant differences, however, reflect sample size more than clinically important difference. Both the VAS interval and ordinal Pain scores were statistically significantly higher in the light months compared to the dark months for the RA patient group, although this difference was only about 3%. The OA patient group also had a 3% higher mean score for VAS interval Pain scores in the light versus dark months, but this difference was not statistically significant probably owing to the small sample size for this group (n = 47) compared to the RA group (n = 535). Depression, Global Disease Severity and HAQ Functional Disability scores did not change for any group.
Paired seasonal scores: analysis of mailed questionnaires (last paired mailed questionnaires)
Table IV presents paired questionnaire data. Results shown in this table are less abnormal than data in Tables II and III, reflecting observation taken outside of the clinic. Using responses to questionnaires completed in both a dark and light month, differences similar to those noted with the paired clinical data were found. Statistically significant differences were noted for Pain and Global Severity scores, with worse scores being obtained in the light months, Even so, these differences were small, ranging from 1.6 to 3%.
TABLE I DEMOGRAPHIC
CHARACTERISTICS
Sex (% women) Age in years: mean (SD) Education (% high school graduates) Marital status (% married) Disease duration in years: mean (SD) Ethnic origin (% White)
OF PARTICIPANTS
AT INITIAL CLINIC VISIT
pn”= 1029)
OA (n = 791)
Fibromyalgia (n = 703)
All patients (n = 2523)
70.2 54.9 (14.62) 81.3 75.8 6.8 (8.83) 95.0
72.7 63.9 (12.58) 78.2 65.3 8.7 (9.45) 93.0
90.3 48.3 (13.55) 83.1 77.1 9.2 (10.57) 95.4
76.6 55.9 (14.98) 80.8 72.6 8.1 (9.59) 94.5
II SCORES
FOR
2573 CONSEC‘C”I‘IVtl
RHEUMATI(
DISEASF:
PATIENTS
.4T THE
FIRST
<‘I lNlC_
_
1.8
CO.071 2.8 CO.151 49.5 (1.73) 1.0 (0.05)
(1.58) 1.0 fo.05)
1.8
48.9 (0.97) 1 .o (0.03)
10.08)
(0.04) 3.0
1.7 (0.07) 1.9 iO.061 3.2 (0.14) 52.2 (1.56) 1.1 (0.05)
2.0 (0.08) 2.0 (0.06) 3.0 (0.14) 51.6 (1.63) 1.1 (0.05 1
(0.11)) 2.0 (0.07) 3.2 (0.17) 51.6 (1.72) 1.0 (0.05)
I.6
May 1.8 (0.09) 1.8 (O.QXf 3.1 0.18) 50.0 (1.80) 1.0 (0.05f
1.8 (0.05)
1.8 (0.08) 1.8 (0.08) 2.9 (0.14) 49.7
1.8 (0.08)
1.x (O.UY) 1.9 (Q.Olt} 3.0 (0.15) 48.5 (1.56) 1.0 (0,04)
1.8 tu.08) (0.07) 3.0 (0.16) 48.5 (I .62) I.0 (Q.05,
2.0
July
--____
June
_____.--._._ Lightest months
1.7 (0.10) 1.9 (0.061 3.2 (0.14) 47.7 (1.73) 1.1 (0.05)
* Continuous VAS Pain scale (n = 958). ** Ordinal VAS Pain scale (n = 1565).
VAS Pain * * (O-3) AIMS Depressian (O-3) VAS Global Severity (O-100) HAQ Functional Disability (O-3)
(O-3)
VAS Pain *
Apr
ALL-D
Mar
Jan
Feb
Dee
mi&ths
Nov
Darkest
(0.92) 1.0 CO.031
1.x ~0.05) 2.0 f0.05) 3.0 (0.08) 49.6
ALL-L
__--_-_
1.9 (0.07) I.9 (0.07) 3.2 (0.17) 51.1 (I.761 1.1 (C~.O5)
Sept
__._ ~_._
1.6 (0.07) i.8 (il.06 ) 3.3 (0.131 48.Y (I.501 1.0 (0.04)
hug
(I .h8) 1.1 ~0.1151
I .8 iQ.OXf 37.C (0.14) 51.7
(0.0X)
I.8
Ott
ALL-D are combined scoreb for the 3-month (dark) period: November. December, Januar>. ALL-L are combined ‘Icores for the 3-month (light) period: Ma>. June .luly. Monthl! scores tor all continuous variahh+s were not significanily different by ANOVA: and ALL-I. and ALL-D did not differ significantly by r test. Monthly scores for VAS Pain. ol-din;11 ~c;IIc. did not diffcl significantly by Kruskat-Wallis test. ALL-D and ALL-L, differed significantly by chi-square (Kidit) analysis (x” = 9.hO. P = 0.02).
PAIN. DEPRESSION. GLOBAL SEVERITY AND FUNCTLONAL DISABILITY VISIT BY MONTH AND BY DARKEST AND LIGHTEST j-MONTH PERIODS
TABLE
231
TABLE III PAIRED SCORES FQR PARTICIPANTS
SEEN IN THE CLINIC IN 1 LIGHT AND 1 DARK MONTH
Light
Dark VAS Pain * (O-3) VAS Pain ** (O-3) AIMS Depression (O-10) VAS Global severity (0-100~ HAQ Functional Disability (O-3)
1.5 + (0.03) 1 St+’ (0:04) 2.6 (0.07) 48.6 (0.98) 1.5 (0.03)
1.5
(0.04) 1.6 (0.04) 2.6 CO.071 47.5 (0.931 1.5 (0.03)
All Patients n = 647
Fibromyalgia n = 65
Osteoarthritis n = 47
RA n = 535
Dark
Light
(E4f
(0.13) 1.8 fO.14)
2.1 (0.08) 1.9 (0.12)
3.0 (0.28) 49.9 (3.60) 1.3 (0.101
&, 51.8 (3.49) 1.2 (0.081
327, 57.2 (2.441 1.4 f0.081
1.7
Dark
Light
Dark
Light 1.6” (0.03) 1.8 +++i (0.04) 2.7 (0.071 49.8 (0.90) 1.5 CO.031
Continuous VAS Pain scale (n = 457) Ordinal VAS Pain scale (n = 295) Mean difference = 0.081, P = 0.032 by paired t test. (Rounding of mean values makes values appear equal.) Mean difference = 0.065, P = 0.05 by paired t test. P = O.OU5by signed rank test. P = 0.005 by signed rank test.
As shown in Table IV, HAQ Functional Disability scores did not change by season for any diagnostic group. VAS Global Severity scores were 1.7% higher in the light versus dark months for the total group (42.1 vs. 40.41, 1.6% and 2% higher in the light months for the OA group (40.5 vs. 38.9) and RA group (41.1 vs. 39.1) respectively. Essentially no change was noted for the fibromyalgia patients for this variable. Pain scores showed the most change. Pain scores were about 3% higher in the light rather than dark months for the totai sample (1.3 vs. 1.2) and for RA 11.2 vs. 1.1) and the OA (1.3 vs. 1,2). The fibromyalgia group also had a 3% difference in VAS Pain scores; however, the fibromyalgia patients reported higher Pain
scores in the dark rather light months (1.7 vs. 1.6). This difference was not statistically significant. AIMS Depression scores did not vary for the total sample or for RA patients, and only minimally for the OA group (1%; 2.4 vs. 2.3). However, the fibromyalgia patients had on average 3% higher Depression scores in the light vs. dark months (3.1 vs. 2.8). Nevertheless scores for depression differences were not statistically significant for any subgroup or for the group as a whole, Paired ~~estio~~aire scores: examination in depressed patients
To examine more fully the Depression scores and their relationship to pain, as well as to the seasons, we
TABLE IV PAIRED SCORES USING QUESTIONNAIRE
RESPONSES FROM 1 LIGHT AND 1 DARK MONTH
Scores are mean f SEM. Paired t tests were used for all comparisons. RA (n = 5891 Dark VAS Pain (O-31 AIMS Depression (O-10) VAS Global Severity (O-100) HAQ Functional Disability (O-3) * ** * ** +
Difference Difference Difference Difference ++ Difference
P P P P P
= = = = =
0.006. 0.017, 0.005. 0.001. 0.006.
Light
Dark 1.2 (0.05)
fiti7, 39.1 (0.96) 1.3 (0.03)
Fibromyalgia (n = 130)
tonA_260)
2.3 {O.ll) 38.9 (1.401 1.0 (0.04)
All patients (II = 979)
Light
Dark
Light
Dark
Light
1.3 *** (0.051
&7)
1.6 (0.07) 3.1 (0.19) 49.9 (2.101
1.2 (0.03) 2.4 f0.06) 40.4 (2.341
1.3’ (0.03) 2.4 (0.061 42.1 ++ (0.77) 1.2 (0.03)
& 40.5 (1.46) 1.0 (0.041
2.8 to.171 49.0 (2.08) 1.1 (0.06)
TABLE V PAIRED SCORES FOR QUESTIONNAIRE EITHER DARK OR LIGHT MONTHS
RESPONSES FOR PARTICIPANTS REPORTING
AIMS DEPRESSION SCORE
)* 4.00 IN
Score5 are mean + SEM. Paired I tests were used for all comparisons. RA (n = X9)
VAS Pain ((l-31 AIMS Depression (O-10) VAS Global Severity (O-100) HAQ Functional Disability (O-3)
OA (n = 46)
Fibromyalgia (n = 34)
Dark
Light
Dark
1.7 (0.081 4.5 (0.22) 57.1 (2.64)
2.0 * (0.07) 5.6 * (0.14) 63.9 a** 12.49) 1.9 (0.08)
I.6 (0.11 f 4.2 (0.78) 51.5 13.62)
(0.11)
(0.11)
5.4 * (0.17) 52.x 13.21)
I .3
I ..s
(0.09)
(0.09)
4.6 (0.39) 58.7 14.38) I.4 (0.13)
1.8 (0.0X)
Light
Dark
1X1
2.0
All patients (n = 169) Light
Dark
Light
2.0 0.13) 5.9 ** (0.30) 61.3 (4.01 J 1.4 (0.14)
1.7 (0.061 4.4 (0.16) 55.9 ( 1.92) 1.6 (0.06)
2.0 * (0.54) 5.6 * 10.1I) 60.4 ****
(I .80) 1.7 (O.06)
* Difference PO.001,
** Difference PO.004. *** Difference PO.008. **** Difference PO.006. ‘ Difference PO.063.
selected a sub-sample of persons with Depression scores greater than or equal to 4.0 in either the light or dark months. We hypothesized that within the group of depressed patients we might be more likely to detect an effect of season. A score of 4.0 on the AIMS Depression scale is consistent with serious depressive symptoms (Hawley and Wolfe 1993). For this analysis questionnaire data was used since the sample number of depressed persons was greater than in the paired clinic data. Table V shows the means, standard errors, and the significance levels for all variables. Depression scores differed by season for all diagnostic groups. The difference between light and dark months ranged from 1.1 points (11% change) for the RA sub-sample to 1.3 points (13% change) for the fibromyalgia patients. Regardless of group, scores were higher in the light months. Pain scores showed similar changes for the total group and for the RA and OA groups. RA patients had 16% higher scores in light rather than dark months and OA patients scored 6.7% higher. Fijromyalgia scores were constant across season, remaining high with a mean of 2.0. For Global Severity no differences exceeded 7 points on the lo-point scale (the light months were favored); and the HAQ Functional Disability Scale was essentiaily stable.
Discussion
Our study finds little relationship between season and depression and between season and rheumatic disease pain, global severity and functional ability. What difference there is links lighter months (May
through July) with severity. Our study differs from previous efforts by its large sample, unselected patients with regard to symptoms of pain or depression, and reliance on data actually collected during the season or month being studied. In addition, we have stratified subjects by rheumatic disease diagnosis to obtain a clearer picture of the effect of season. There were a number of reasons to suspect that season might be associated with pain and/or depression in rheumatic disease patients in colder, darker months. Rheumatic disease patients have disabilities that might be particularly limiting during these months, and cold may exacerbate symptoms of stiffness and pain while warmth and heat is used therapeutically. In addition, a subset of rheumatic disease patients with a widespread chronic pain syndrome (fibromyalgia) have higher levels of depression and pain than do other rheumatic disease patients (Hawley et al. 1988; Hawley and Wolfe 19931, disturbed diurnal neurohypophyseal function (McCain and Tilbe 1989), seratonin and other CNS mediator abnormalities (Russell 1989; Russell et al. 1992a,b), a disturbed diurnal sleep pattern, and a demonstrated worsening in cold, dark seasons (by history) (Yunus et al. 1981). In spite of these considerations, we did not find evidence for worse pain or depression in darker or colder months in any of the rheumatic disorders studied, in accord with observations that others have noted in osteoarthritis (Harris 1986; Laborde et al. 1986). The link between depression and season has been clearly demonstrated in SAD (Rosenthal et al. 1984). In this syndrome patients suffer recurring depressions in dark months and have day-time somnolence, weight gain, and carbohydrate craving (Rosenthal et al 1984.
233
1987). Our study, however, did not investigate SAD features (except for depressive symptoms), and we can not draw conclusions concerning SAD. Even so, if a clinically important number of our patients had SAD we would have suspected it to have been reflected in higher Depression scores in the dark months, something we did not see. We considered the possibility that studying patients who were not depressed could have influenced our results. We therefore also studied a subset of patients with high Depression scores. Within this group we found even more evidence for generalized worsening in light months. In an additional attempt to investigate a possible link with SAD, we studied correlations between changes in weight and changes in Depression scores in the patients with paired clinic visits. No statistical association was seen in the group as a whole, in the 3 diagnostic groups separately, and in those who gained more than 2 kg or 4 kg in the dark months. Finally, we applied exploratory data smoothing algorithms, including FFT, Lowess, Savitsky-Golay, and sequential polynomial interpolation (TableCurve, version 1.0, Jandel Scientific, San Rafael, CA) to attempt to identify seasonal patterns. Unlike Terman et al. (1989) who demonstrated a seasonal pattern for SAD symptoms in a random survey of New York residents, we were unable to identify such patterns among depression and pain scores utilized in our study. One factor that might have confounded our results is the finding (in a coastal population of RA patients) that increasing temperature is associated with higher levels of pain, a fact that is possibly explained by a stronger relationship between pain and vapor pressure in summer than winter (Patberg et al. 1985). Other studies, however, in non-coastal regions (similar to the geographic area of our report) did not find relationships between weather and rheumatic disease pain (Sibley 1985). Another possible confounding effect could be the general increase in depressive symptoms among those with chronic illness, a factor which could conceivably mask seasonal depression (there have been no studies investigating this point, however) (Wells et al. 1988; Wolfe et al. 1993). Finally, it is possible that an association between season and pain and depression not seen here might be found in areas such as Scandinavia where greater differences exist between light and dark seasons. Thus the results of this investigation do not find associations between dark, colder months and depressive symptoms; we find slightly the opposite. Nor do we find increased pain in these months, regardless of the rheumatic conditions studied. What trend there is favors severity in lighter months by about 3% compared with darker months. Season does not appear to play an important role in pain and/or depression in rheumatic diseases,
References Arnett, F.C., Edworthy, SM., B&h, D.A., McShane, D.J., Fries, J.F., Cooper, N.S., Healy, L.A., Caplan, S.R., Liang, M.H., Luthra, H.S., Medsger, T.A., Jr., Mitchell, D.M., Neustadt, D.H., Pinals, RX, Schaller, J.G., Sharp, J.T., Wilder, R.L. and Hunder, G.G., The American Rheumatism Association, 1987 revised criteria for the classification of rheumatoid arthritis, Arthr. Rheum., 31 (1988) 315-324. Blalock, S.J., de’lrellis, R.F., Brown, G.K. and Wallston, K.A., Vahdity of the Center for Epidemiological Studies depression scale in arthritis populations, Arthr. Rheum., 32 (1989) 991-997. Booker, J.M. and Heltekson, C.J., Prevalence of seasonal affective disorder in Alaska, Am. J. Psychiat., 149 (1992) ‘1176-1182. Brewerton, T.D. and George, MS., A study of the seasonal variation of migraine, Headache, 30 11990) 511-513. Cugini, P., Romit, A., Di Palma, L. and Giacovazzo, M., Common migraine as a weekly and seasonal headache, Chronobiol. Int., 7 (1990) 467-469. Dequeker, J. and Wuestenraed, L., The effect of biometeorological factors on Ritchie articular index and pain in rheumatoid arthritis, Stand. J. Rheumatol., 15 (1986) 280-284. Faedda, G.L., Tondo, L., Teicher, M.H., Baldessarini, R.J., Gelbard, H.A. and Floris, G.F., Seasonal mood disorders. Patterns of seasonal recurrence in mania and depression, Arch. Gen. Psychiat., 50 (1993) 17-23. Fa~enbach, A., Gottschalk, R., Hadji, F. and Kaltwasser, J.P., Sensitivity to weather in patients with rheumatoid arthritis or psoriatic arthritis, Sem. Hop. Paris, 68 (1992) 781-783. Fries, J.F., Spitz, P.W. and Kraines, R.G., Measurement of patient outcome in arthritis, Arthr. Rheum., 23 (1980) 137-145. Guedj, D. and Weinberger, A., Effect of weather conditions on rheumatic patients, Ann. Rheum. Dis., 49 (1990) 158-159. Hansen, V., Jacobsen, B.K. and Husby, R., Mental distress during winter. An epidemiologic study of 7759 adults north of Arctic Circle, Acta Psychiat. Stand., 84 (1991) 137-141. Hardin, T.A., Wehr, T.A., Brewerton, T., Kasper, S., Berrettini, W., Rabkin, J. and Rosenthal, N.E., Evaluation of seasonal&y in six clinical populations and two normal populations, J. Psychiat. Res., 25 (1991) 75-87. Harris, CM., Further observations on seasonal variation. 1. Osteoarthritis, J. Roy. COIL Gen. Pratt., 36 (1986) 316-318. Hawley, D.J. and Wolfe, F., Pain, disability, and pain/disability relationships in seven rheumatic disorders: a study of 1522 patients, J. Rheumatol., 18 (1991) 1552-1557. Hawley, D.J. and Wolfe, F., Sensitivity to change of the health assessment questionnaire (HAQ) and other clinical and health status measures in rheumatoid arthritis: results of short-term clinical trials and observational studies versus long-term observational studies, Arthr. Care Res., 5 (1992) 130-136. Hawley, D.J. and Wolfe, F., Depression is not more common in rheumatoid arthritis: a 10 year longitudinal study of 6,608 rheumatic disease patients, J. Rheumatol., in press. Hawley, D.J., Wolfe, F. and Cathey, M.A., Pain, functional disability, and psychological status: a 12-month study of severity in fibromyalgia, J. Rheumatol., 15 (1988) 1551-1556. Kravitz, H.M., Katz, R., Kot, E., Helmke, N. and Fawcett, J., Biochemical clues to a fibromyalgia-depression link - imipramine binding in patients with fibromyalgia or depression and in healthy controls, J. Rheumatol., 19 (1992) 1428-1432. Kudrow, L., The cyclic relationship of natural illumination to cluster period frequency, Cephalalgia, 7 (Suppl. 6) (1987) 76-78. Laborde, J.M., Dando, W.A. and Powers, M.J., Influence of weather on osteoarthritics, Sot. Sci. Med., 23 (1986) 549-554. Lam, R.W., Soiyom, L. and Tompkins, A., Seasonal mood symptoms in bulimia nervosa and seasonal affective disorder, Comp. Psychiat., 32 (1991) 5.52-558.
Manzoni. G.C., Micieli, G., Zanferrari, C., Sandrini, G.. Bizzi, P. and Nappi, G., Cluster headache. Recent developments in clinical characterization and pathogenesis. Acta Neurol. (Napoli), 13 (1991) 506-513.
SAS Institute, SAS/STAT Users Guide: Version Gary, NC, 1992. Sibley, J.T., Weather and arthritis symptoms. (1985) 707-710.
McCain, G.A. and Scudds, R.A., The concept of primary fibromyalgia (fibrositis): clinical value, relation and significance to other chronic musculoskeletal pain syndromes, Pain, 33 (1988) 273-287. McCain, G.A. and Tilbe, K.S., Diurnal hormone variation in fibromyalgia syndrome: a comparison with rheumatoid arthritis, J. Rheumatol. (SuppI.), 19 (1989) 154-157. Meenan, R.F., Gertman, P.M. and Mason, J.H., Measuring health status in arthritis: the arthritis impact measurment scales, Arthr. Rheum.. 23 (1980) 146-152. Neuwirth, R. and Faust, V., Neue medizinmeteorologische ergebnisse an psychisch kranken, Munch. Med. Wochenschr., 117 (1975) 431-436. Patberg, W.R., Effect of weather on daily pain score in rheumatoid arthritis (letter). Lancet, ii (1987) 386-387. Patberg, W.R.. Nienhuis, R.L. and Veringa, F., Relation between meteorological factors and pain in rheumatoid arthritis in a marine climate, J. Rheumatol., 12 (1985) 711-715. Ropes, M.W., Bennett, G.A., Cobb, S., Jacox, R. and Jessar, R.A., 1958 Revision of diagnostic criteria for rheumatoid arthritis, Arthr. Rheum., 2 (1959) 16-20. Rosenthal, N.E., Sack, D.A., Gillin, J.C., Levy, A.J., Goodwin, F.K., Davenport, Y., Mueller, P.S., Newsome, D.A. and Wehr, T.A., Seasonal affective disorder: a description of the syndrome and preliminary findings with light therapy, Arch. Gen. Psychiat., 41 (1984) 72-80. Rosenthal, N.E., Genhart, M.J., Jacobson, F.M., Skwerer, R.G. and Wehr. T.A., Disturbances of appetite and weight regulation in seasonal affective disorder, Ann. NY Acad. Sci., 499 (1987) 216-230. Rudorfer, M.V., Skwerer, R.G. and Rosenthal, N.E., Biogenic amines in seasonal affective disorder: effects of light therapy, Psychiat. Res., 46 (1993) 19-28. Russell, I.J., Neurohormonal aspects of the fibromyalgia sundrome, Rheum. Dis. Clin. N. Am., 15 (1989) 149-168. Russell, I.J.. Michalek, J.E., Vipraio, G.A., Fletcher, E.M., Javors, M.A. and Bowden, C.A., Platelet H-3-imipramine uptake receptor density and serum serotonin levels in patients with fibromyalgia/fibrositis syndrome, J. Rheumatol., 19 (1992a) 104-109. Russell, I.J., Vaeroy, H., Javors, M. and Nyberg, F., Cerebrospinal fluid biogenic amine metabolites in fibromyalgia/fibrositis syndrome and rheumatoid arthritis, Arthr. Rheum., 35 (1992b) 550556.
Terman, M., Botticelli, S.R., Link, B.G., Link, M.J., Quitkin, F.M.. Hardin, T.E. and Rosenthal, N.E.. Seasonal symptom patterns in New York: patients and population. In: C. Thompson and T. Silverstone (Eds.), Seasonal Affective Disorder, Clin. Neurosci. Publ., London, 1989, pp. 77-95. Weissbluth, M. and Weissbluth, L.. Colic, sleep inertia, melatonin and circannual rhythms, Med. Hypotheses, 38 (1992) 224-228. Wells, K.B., Golding, J.M. and Burnam, M.A.. Psychiatric disordelin a sample of the general population with and without chronic medical conditions, Am. J. Psychiat., 145 (1988) 976-981. Wicki, W., Angst, J. and Merikangas. K.R., The Zurich Study. XIV. Epidemiology of seasonal depression, Eur. Arch. Psychiat. Clin. Neurosci., 241 (1992) 301-306. Wolfe, F. and Cathey, M.A., Prevalence of primary and secondary fibrositis, J. Rheumatol., IO (1983) 965-968. Wolfe. F. and Hawley, D.J., The relationship between clinical activity and depression on rheumatoid arthritis, J. Rheumatol. 20, (1993) 2032-2037. Wolfe, F., Smythe, H.A.. Yunus, M.B., Bennett, R.M.. Bombardier. C.. Goldenberg, D.L., Tugwell, P., Abeles, M., Campbell, S.M., Clark, P., Fam, A.G., Farber, S.J.. Fiechtner, J.J., Franklin. C.M., Gatter, R.A., Hamaty, D., Lessard, J., Lichtbroun, A.S., Masi, A.T.. McCain, G.A., Reynolds, W.J., Romano, T.J., Russell. I.J. and Sheon, R.P., The American College of Rheumatology, 1990 criteria for the classification of fibromyalgia: report of the multicenter criteria committee, Arthr. Rheum., 33 (1990) 160-172. Wolfe, F.. Hawley, D.J. and Cathey, M.A., The assessment and prediction of functional disability in RA, J. Rheumatol., 18 (1991) 1298-1306. Wolfe, F., Anderson, J., Ross, K. and Russell, I.J., The prevalence and characteristics of fibromyalgia in the general population, Arthr. Rheum., 36 (1993) S48 (abstr.). Wolfe, F., Hawley. D.J., Ross, K. and Anderson, J., Depression is increased two-fold in RA and clinically significant arhtritis, Arthr. Rheum., 20 (1993) 2025-2031. Yunus, M.B., Masi, A.T., Caiabro, J.J., Miller, K.A. and Feigenbaum, S.L., Primary fibromyalgia (fibrositis): clinical study of 50 patients with matched normal controls, Semin. Arthr. Rheum.. 11 (1981) 151-171.
6.08. SAS Institute. J. Rheumatol..
12