COMPARISON OF PAIN PROPERTIES IN FIBROMYALGIA PATIENTS AND RHEUMATOID ARTHRITIS PATIENTS

Transcription

1 775 COMPARISON OF PAIN PROPERTIES IN FIBROMYALGIA PATIENTS AND RHEUMATOID ARTHRITIS PATIENTS FRANK LEAVITT, ROBERT S. KATZ, HARVEY E. GOLDEN, PAUL B. GLICKMAN, and LAWRENCE F. LAYFER Pain properties of 50 fibromyalgia patients were examined and compared with pain properties of 50 rheumatoid arthritis patients. In both fibromyalgia and rheumatoid arthritis, pain was bilateral, involved multiple sites, and was of equal intensity (60.8 versus 58.7, respectively, on a scale of 100). Fibromyalgia pain, however, was less localized to the joints and suggested greater spatial diffusion. It involved more kinds of pain experiences (radiating, steady, spreading, spasms, gnawing, unlocalized, pricking, crushing, shooting, pressing, splitting, cramping, nagging, and pins and needles), and was dispersed over larger areas of the body. The anatomic sites best for discrimination between patients with fibromyalgia and patients with rheumatoid arthritis were the lower back, thigh, abdomen, head, and hips for fibromyalgia, and wrist, foot, and fingers for rheumatoid arthritis. The traditional clinical description of aching and stiffness does not appear to accurately describe the complexity of the fibromyalgia pain syndrome. - From the Department of Psychology and Social Sciences and the Department of Medicine (Section of Rheumatology), Rush Medical College, Rush Presbyterian-St. Luke s Medical Center, Chicago, Illinois. Frank Leavitt, PhD: Associate Professor, Department of Psychology and Social Sciences; Robert S. Katz, MD: Associate Professor, Department of Medicine (Section of Rheumatology); Harvey E. Golden, MD: Associate Professor, Department of Medicine (Section of Rheumatology); Paul B. Glickman, MD: Associate Professor, Department of Medicine (Section of Rheumatology); Lawrence F. Layfer, MD: Assistant Professor, Department of Medicine (Section of Rheumatology). Address reprint requests to Frank Leavitt, PhD, Department of Psychology and Social Sciences, Rush Presbyterian-St. Luke s Medical Center, 1750 West Harrison Street, Chicago, IL Submitted for publication April 11, 1985; accepted in revised form December 5, Fibromyalgia is a nonarticular rheumatic disorder without a known pathologic basis (1). Its most outstanding clinical symptoms are chronic diffuse musculoskeletal aching and stiffness, and tenderness in specific anatomic sites. Laboratory tests have not been of value in establishing the diagnosis: blood and urine test results are routinely normal, as are chest and skeletal roentgenograms (2). Abnormalities in stage 4 sleep (3) have been reported, but this laboratory finding could not be replicated in recent studies of alpha-delta sleep in patients with fibromyalgia (4). Of 13 patients satisfying criteria for fibromyalgia (2), only 1 patient exhibited an alpha-delta electroencephalogram. Therefore, a diagnosis of fibromyalgia is based primarily on musculoskeletal pain and positive areas of tenderness, with the application of associated clinical criteria. Despite its general acceptance as a diagnostic criterion for fibromyalgia, very little is known about the nature of pain experienced by patients in this group. Generalized aching and stiffness of a nonspecific nature typically is reported, but the description by Ellman et a1 40 years ago of a dull ache comparable to a toothache (5) is characteristic of the imprecision which characterizes current clinical documentation of the fibromyalgia pain syndrome. This study characterizes the pain of fibromyalgia along qualitative and quantitative dimensions. Because of the resemblance of some of its aspects to that seen in patients who have fibromyalgia, pain in rheumatoid arthritis patients was also studied (6,7). Comparison of the characterization of pain in fibromyalgia and in rheumatoid arthritis may help clarify properties that are more specific for these 2 rheumatic syndromes, and may help in the develop- Arthritis and Rheumatism, Vol. 29, No. 6 (June 1986)

2 776 LEAVITT ET AL ment of pain scales to delineate and monitor fibromyalgia. PATIENTS AND METHODS Patients. Hospitalized patients, ages -69, who were admitted primarily for the diagnosis and treatment of rheumatic disease, were the subjects. All were patients of 4 cooperating rheumatologists, and all were informed of the nature of the study and were asked to participate. Those patients who had less than a ninth-grade education or who could not demonstrate adequate reading ability were excluded. Sixty-four patients who had classic rheumatoid arthritis (RA), according to American Rheumatism Association criteria (8), agreed to participate. All had fewer than 4 tender points (1,2); all rheumatoid arthritis patients who had 4 or more tender points were excluded from the study. Fifty patients who had fibromyalgia diagnosed by independent examinations by at least 2 rheumatologists, according to the criteria listed below, formed the second study sample. The fibromyalgia patients in this study had experienced diffuse musculoskeletal pain for at least 3 months. All patients with fibromyalgia had exaggerated bilateral tenderness over the midportion of the trapezius muscles. They also had tenderness over many muscles and trigger points, including at least 4 of those described by Smythe and Moldofsky (1) and Moldofsky et al (3). These sites of tenderness included the lateral epicondyle of the humerus, the chest wall (including the second and third costochondral junctions), the region overlying the medial collateral ligament of the knee, the L4-S 1 interspinous ligaments, the ligaments joining the transverse processes of C4 to C6, the medial border of the scapula, and the upper outer quadrant of the buttock. No demonstrable synovial inflammatory findings, muscle weakness, neurologic changes, tendinitis, or bursitis were noted. All patients had full range of musculoskeletal motion on objective testing. The following laboratory studies were performed on all patients with fibromyalgia, and the results were negative or normal: complete blood count, Westergren erythrocyte sedimentation rate, rheumatoid factor, antinuclear antibodies, complement studies, chest and skeletal roentgenograms, creatine phosphokinase, serum glutamic oxaloacetic transaminase, serum glutamic pyruvic transaminase, electrolytes, serum protein electrophoresis, blood urea nitrogen, creatinine, calcium, alkaline phosphatase, urinalysis, and electrocardiogram. After clinical examination and laboratory tests, the rheumatologists filled out a diagnostic classification form for each patient and forwarded it to the investigators. None of the rheumatologists had knowledge of the patients scores on the 2 pain measurement questionnaires. The average age of the RA group was 8.1 years older than that of the fibromyalgia group (53.2 versus 45.1); the 14 oldest rheumatoid arthritis patients were excluded in order to match the groups for size and age. The 2 groups were then quite similar in respect to the demographic variables listed in Table 1. None of the differences listed were significant. The subjects in both samples were predominantly female, white, Table 1. Description of study participants arthritis Fibromyalgia (n = 50) (n = 50) Age (average, years) SD Education (average, years) SD Sex (%) Male Female Race (%) White Black Marital status (%) Single Mamed Other Duration of pain (%) 0-6 months months years years Duration of pain (months), mean 2 SD and married, and had experienced pain for more than 2 years. Pain questionnaire. Our pain questionnaire was an adaptation of the McGill Pain Questionnaire (MPQ), which has been reported by Melzack (9). The MPQ is a rationally derived, general pain scale consisting of 78 words categorized along 20 subscales (Table 2). Each subscale contains from 2-6 words. The makeup of the subscales is based on similarity of pain qualities, with the pain words arranged in increasing order of pain intensity. Patients were limited to 1 word choice in each subscale. They could omit word groups which were not suitable. A score was assigned to each subscale, ranging from 0 (no word chosen) to 6 (the highest ranking word in a 6-word subscale); a score of 3 indicated that the word with the third highest intensity rating was chosen. The 20 subscales are also grouped along 4 dimensions: sensory, affective, evaluative, and mixed (see Table 2). The sensory word group contains 10 subscales that describe pain in terms of temporal, spatial, pressure, and thermal properties. The affective group of words contains 5 subscales that describe pain in terms of tension, autonomic reaction, fear, and punishment properties. The evaluative dimension is a single, S-word subscale that describes the subjective, overall intensity of the total pain experience. The mixed word category involves 3 sensory subscales and 1 affective subscale. Our adaptation of the MPQ lists the 78 pain words in random order, to reduce the selection bias that is inherent in responding to groups of pain words of similar meaning. The number of words was increased by 9, with the words steady, intermittent, momentary, stiff, spasms, localized, unlocal-

3 PAIN IN FIBROMYALGIA AND RA 777 Table 2. Melzack s general pain model - Sensory words 1. 1 Flickering 2. I Jumping 3. 1 PRICKING? 4. 1 Sharp 5. 1 Pinching 6. 1 Tugging 7. I Hot 8. I Tingling 9. 1 Dull Tender 2 Quivering 2 Flashing 2 Boring 2 Cutting 2 PRESSING 2 Pulling 2 Burning 2 Itching 2-2 Taut 3 Pulsing 4 Throbbing* 3 SHOOTING 3 Drilling 4 Stabbing 3 Lacerating 3 GNAWING 4 CRAMPING 3 Wrenching 3 Scalding 4 Searing 3 Smarting 4 Stinging 3 Hurting 4Aching 3 Rasping 4 SPLITTING 5 Beating 6 Pounding 5 Lancinating 5 CRUSHING 5 Heavy Affective words Tiring Sickening Fearful Punishing Wretched 2 Exhausting 2 Suffocating 2 Frightful 2 Grueling 2 Blinding 3 Terrifying 3 Cruel 4 Vicious 5 Killing Evaluative words Annoying 2 Troublesome 3 Miserable 4 Intense 5 Unbearable Mixed words SPREADING. 1 Tight Cool NAGGING 2 RADIATING 2 m 2 Cold 2 Nauseating 3 Penetrating 4 Piercing 3 Drawing 4 Squeezing 3 Freezing 3 Agonizing 4 Dreadful * ljnderlined = words used by more than 40% of patients in the fibromyalgia group. t Boldface capital letters = words differentiating pain of fibromyalgia and pain of rheumatoid arthritis. 5 Tearing 5 Torturing ized, moving, and pins and needles added. To avoid the possible loss of information by restricting choice to 1 word per subscale, patients were instructed to check every word that described their pain. Pain location and pain intensity measurement procedure. All patients who met the criteria and agreed to participate were administered the Pain Questionnaire, the Pain Intensity measure, and the Pain Location sheet in individual sessions, usually within the first 4 days of admission to the hospital. The printed instructions given for filling out the Pain Questionnaire were: Some of the words describe your present pain. Check those words that best describe how your pain typically feels. You may check as many boxes as you wish that describe your typical pain these days. The Pain Intensity measure was a linear scale with self-ratings from 0 (no pain) to 100 (severest pain imaginable). The instructions for this measure were: We need a more accurate idea of how severe your pain is. On a scale of 0 to 100, in which 0 is no pain at all, and I00 is the most severe pain that one can possibly imagine, what number would you give your average pain? On the Pain Location sheet, pain was categorized for 25 body sites (see Table 3), and patients indicated with a check mark all areas of the body in which pain was currently being felt. RESULTS Following the categorization of the 20 subscales suggested by Melzack, sensory, affective, evaluative, and mixed scores were obtained by summing the values of all words checked, for each of the 4 categories of pain. The number of words chosen by each patient was also tallied for each subcategory, as well as for the total pain questionnaire. Table 4 shows the mean values of the sensory, affective, evaluative, and mixed scores for the 2 patient groups. The group of patients diagnosed as having fibromyalgia produced higher mean scores in each of Melzack s subcategories; however, none of these differences reached statistical significance. Fibromyalgia patients selected an average of 4.5 more words to describe their pain on a pain questionnaire than did patients with RA (Table 5). A t-test showed that this difference was significant at the 0.05 level. The increased pain word usage appears to be accounted for by increased word usage across all the categories (sensory, affective, evaluative, and mixed). For each pain subcategory, the mean number of pain

4 778 LEAVITT ET AL Table 3. Percentage of fibromyalgia patients and rheumatoid arthritis patients reporting pain in the listed locations - Pain Fibromyalgia arthritis location (n = 50) (n = 50) Z P Lower back 78 Thigh Abdomen 28 Wrist Foot Fingers Shins Hips 68 Head Left leg 76 Hands 62 Chest 30 Upper back Ankle Shoulder 62 Toes Right leg 14 S houlder-bac k 76 Knees 70 Elbows Genitals 24 Face Left arm 58 Right arm 58 Neck 70 * NS = not significant ~~ NS* 0.65 NS 0.62 NS 0.50 NS 0.20 NS 0.19 NS 0. NS 0.00 NS words used was higher in the fibromyalgia group, and approached statistical significance. P values ranged from The specific pain words used by both groups were evaluated by frequency analysis to further clarify characteristic features of the pain reported by the 2 groups. The most frequent responses used by fibromyalgia patients are listed in Table 6. Eighteen of the 87 descriptive words were used by more than 40% of the fibromyalgia patients to describe their pain. Twelve of the same words were used by more than 40% of the patients with RA. This indicates a high degree of similarity in pain description between the 2 Table 4. Melzack general pain model scores (mean C SD) by arthritis Fibromyalgia (n = 50) (n = 50) P Sensory score 13.8 f NS* Affective score f 3.6 NS Evaluative score 3.0 f NS Mixed score f 4.8 NS * NS = not significant. Table 5. Number of pain words (mean 2 SD) used by fibromyalgia patients and rheumatoid arthritis patients arthritis Fibrom yalgia (n = 50) (n = 50) P Total words (range) 14.4 f 9.7 (2-34) (5-43) 0.05 Sensory words 7.5 f f Affective words 1.7 f f Evaluative words f Mixed words f groups. Aching was the most frequently reported pain quality in both groups, and occurred in 82% of the patients. The word used next most frequently by fibromyalgia patients was exhausting (62%), followed by nagging (60%) and hurting (60%). In comparison, stiff (64%) and moving (64%) were the second and third most frequently used words by patients who had RA. Both of these pain words were used with 16% less frequency by patients with fibromyalgia. Actual differences in pain word usage were analyzed by log linear statistics which are appropriate for cross-tabulated categoric data involving multiple comparisons (10). Fourteen comparisons of frequency differences were significant at the 0.05 level, and are shown in Table 7. The pain descriptors used more often in fibromyalgia were radiating, steady, spreading, spasms, gnawing, unlocalized, pricking, crushing, shooting, pressing, splitting, cramping, nagging, and pins and needles. There were no significant differences in the frequency of usage by the 2 groups for the remaining 74 pain qualities. To further elucidate the nature of words used by fibromyalgia patients to describe pain, responses were mapped on Melzack s general pain classification model. This model characterizes pain into 20 subclasses which are grouped according to 4 dimensions: sensory, affective, evaluative, and mixed. In Table 2, the words used most often by patients with fibromyalgia to describe their pain are underlined. The words which best differentiate the pain of fibrornyalgia from that of rheumatoid arthritis are entered in boldface. All words in the evaluative category (class 16) were used by at least 40% of the fibromyalgia sample; these words were used to describe the patient s overall emotional response to pain, as distinguished from their physical perception of pain. These percentages were similar to those found for patients with RA (Table 4), which suggests that a strong emotional response colors the pain language of both groups. In addition, pain words used more frequently

5 PAIN IN FIBROMYALGIA AND RA 779 by patients with fibromyalgia were not randomly dis- Table 7. Pain words differentiating patients with fibromyalgia tributed. According to this model, 7 of the words used from Patients with hmnatoid arthritis more commonly by fibromyalgia patients were sen- sory; 3 were from the mixed category (Table 2). Four Fibromyalgia arthritis other differentiating words (pins and needles, steady, patients patients (%I (%I Z P unlocalized, and spasms) do not appear in the Melzack list and cannot be mapped on the scale. Two clear Radiating measure of constrictive pressure. Two of the 3 mixed words belonged to subclass 17, which is a spatial factor. The mean pain intensity scores of the 2 diagnostic groups were markedly similar. Patients with fibromyalgia reported a mean (? SD) score of 60.8 * 22.3 on a scale of 100, whereas rheumatoid arthritis patients reported a score of 58.7? The differences were not significant. The actual painful body sites reported by the 2 groups are shown in Table 3. The 2 groups did not dliffer in the number of body sites which were painful. E ibromyalgia patients complained of pain in body sites; RA patients complained of pain in 11.9 * 4.5 sites. The mean difference of 1.0 sites was not significant at the 0.05 level. However, the 2 groups differed in the variety of locations at which pain was reported to occur. The most common sites of pain in the RA group were related to the joints. The fingers, hand, wrist, shoulder, knee, ankle, foot, and toes were ad frequently mentioned sites of pain, and each site was checked by at least 70% of the sample. However, Table 6. Most frequently used pain descriptors, by patient group Aching Exhausting Nagging ldurting!sore.4nnoying Shooting Troublesome Miser a b I e Radiating Unbearable Stiff Moving Throbbing Intense Penetrating Constant Numb Fibrom yalgia arthritis patients patients (%) (%) Steady patterns emerged from mapping. Four of the 7 sensory Spreading words belonged to subclass 5, which purports to be a SDasms Gnawing Unlocalized Pricking Crushing Shooting Pressing Splitting Cramping Nagging Pins and needles the 6 most common pain areas reported by the fibromyalgia patients were nonarticular sites: the low back, shoulder-upper back area, left leg, right leg, neck, and hips. Differences between the groups were statistically significant for 17 of the 25 anatomic sites listed. The sites that provided the best discrimination between patients with fibromyalgia and those with RA were: the lower back, thigh, abdomen, shin, hips, and head (fibromyalgia); wrist, foot, and fingers (RA). DISCUSSION From the analysis of the pain data, it is clear that there are significant similarities and differences in the way the 2 groups report, and therefore, presumably in the way they experience, pain. Patients identified as having fibromyalgia share many pain properties with patients with rheumatoid arthritis. They reported 74 of 87 pain qualities with similar frequency, and report pain intensity of equal severity. The differences are also noteworthy in several ways. Pain reported by patients with fibromyalgia is significantly more diffuse and radiating than pain reported by patients with RA, even though the 2 groups suffered the same degree of pain intensity for approximately the same duration. An average of 4.5 more words are used in describing fibromyalgia pain, which suggests that a more complex complaint is clinically presented by this group. The increase in word usage seems to be accounted for by increased word usage across all 4 dimensions of the Melzack index of pain. Of all the pain words used, radiating is the single best discriminator. It was used 4 times as often (%

6 LEAVITT ET AL versus 12%) in the pain repertoire of the fibromyalgia sample as it was used in the RA sample. This may, in part, reflect the spatially diffuse and spreading nature of the fibromyalgia pain experience. Additionally, patients with fibromyalgia seem to point out several other aspects of their pain. First, there is an emphasis on constrictive pressure, characterized by sensations of pressing, gnawing, cramping, and crushing. Approximately one-third of patients with fibromyalgia described pain as steady, unlocalized, spasmodic, or pins and needles. This contrasted with RA patients, only 15% of whom used the latter 4 descriptors. While there are no compelling explanations for the emphasis on these particular sensations by fibromyalgia patients, they may prove clinically useful in distinguishing fibromyalgia from rheumatoid arthritis. The diagnostic value of pain language in the clinical situation is complicated by several factors. First, there was a high degree of overlap in the use of common pain descriptors by the 2 groups, which may mask the differences noted. Of the pain words selected by over 40% of the patients with fibromyalgia, 12 were also chosen by over 40% of the RA group. The overlapping of pain properties may make it difficult to differentiate the 2 diseases on the basis of spontaneous pain descriptions. Second, whereas pain data from study subjects were elicited using a standardized questionnaire, pain information on clinical subjects is elicited from their descriptions using their own words. These different methods of data collection could influence the nature of the clinical information obtained. Formal pain measurement, using standardized pain questionnaires that evaluate pain in its quantitative form, may be necessary to detect the distinguishing differences. Based on these results, it is clear that fibromyalgia pain is more than generalized aching and stiffness. It involves multiple sensory and affective pain sensations, some of an intense nature described by patients as miserable and unbearable. This may make the distinction between fibromyalgia and psychogenic rheumatism difficult if pain of an intense or dramatic nature is used as the differentiating criteria, as has been suggested by some investigators (1 1). In fact, descriptions of the pain experience varied greatly; as indicated, pain words were used by more than 40% of the sample. Aching was the pain sensation mentioned most frequently by both groups, appearing in the pain repertoires of 82% of each group. It is not, therefore, a differentiating symptom, nor one that is particularly unique to arthritis, inasmuch as it is an extremely common descriptor of other pain syndromes (12,13). In a recent study (14), 65% of patients with suspected disc disease used aching to describe their pain. Exhausting was used by 62% of the fibromyalgia sample and ranks second in selection; stiff (64%) and moving (64%) rank second in selection by RA patients. In comparison, stiff and moving are used by % of fibromyalgia patients; 12 other pain words are used as often in the description of fibromyalgia pain. Both groups show a similar pattern of communality and difference in respect to anatomic site. Pain is bilateral, and multiple pain sites are common to both groups. Differences between the groups relate to the distribution of pain. The location of pain in fibromyalgia patients, as would be expected, is frequently nonarticular and involves large anatomic areas. The lower back, thigh, head, and abdomen are the anatomic areas providing the greatest degree of differentiation. These pain sites are approximately 4 times more common in fibromyalgia patients. In contrast, pain in patients with RA is localized to joint areas, particularly those involving the fingers, wrist, shoulder, knee, and ankle. Fibromyalgia pain is less localized and there is a greater spatial diffusion of the pain. In some sense, fibromyalgia involves more of the body. The widespread distribution of pain described by our patients was in accord with what had been described as fibromyalgia pain in the past (2). The percentage of fibromyalgia patients who experienced pain in the lower extremities was higher than that reported in the clinical literature. Yunus et a1 (2), for example, found leg pain in only 14% of their patient sample. The differences in reports of abdominal pain (28% for fibromyalgia versus 0% for RA) correlated with the high incidence of irritable bowel symptoms found in patients with fibromyalgia (2). This finding may be helpful in differentiating the musculoskeletal pain of fibromyalgia from the musculoskeletal pain of RA early in the disease process, since abdominal pain was unique to fibromyalgia. Because the fibromyalgia patients involved in the study were hospitalized, they may not be entirely representative of the general population of patients with fibromyalgia. Application of our results to outpatients must be done with caution. Hospitalized patients may have more severe disease. However, it should be remembered that patients in the RA group were also hospitalized. As with all research, greater

7 PAIN IN FIBROMYALGIA AND RA 78 1 confidence in the findings will only emerge with studies of other samples. A common process may be involved in the pain properties that characterize the fibromyalgia syndrome. The symptoms appear linked by a diffusion of the qualitative and spatial experience, which is evidenced by a multiplication of the number of sensory and affective pain sensations and the spread of these sensations across large areas of the body. Although it is still subject to debate, many clinicians stress the role of muscle tenderness in fibromyalgia pain (7). The anatomic sites where patients most often experience fibromyalgia involve the large areas, as opposed to the joints, which are mentioned by patients with RA. This can be interpreted as supporting the hypothesis that palin of fibromyalgia is muscle pain. In several studies, psychologic disturbance has been associated with pain patterns characterized by diffuse pain language (15,16). In each study, pain language used by medical patients with psychologic disturbance appeared more complex, because more words were used to describe pain symptomatology. Similarities in pain word usage by fibromyalgia patients raise the possibility of more psychologic disturbance in this group. It is conceivable that if psychologic disturbance exists, it is specific to subgroups of patients with fibromyalgia who use complex, diffuse descriptions of pain. In our study, 12 patients with fibromyalgia used 25 or more words and would be suspected of having possible psychologic disturbance. Sixteen patients used fewer than 13 words, the mean of the RA group; their description of pain was more specific and limited. Our interpretations are in accord with the work of several investigators (17-19) who found pathologic elevations on the Minnesota Multiphasic Personality Inventory for approximately % of patients meeting strict criteria for the diagnosis of fibromyalgia. In summary, the traditional, clinical description of aching and stiffness does not appear to adequately deline the complexity of the fibromyalgia pain syndrome. Diffusion of pain, in terms of quality and location, appears to be a better clinical marker of the disorder than a limited set of pain descriptors. REFERENCES 1. Smythe HA, Moldofsky H: Two contributions to understanding of the fibrositis syndrome. Bull Rheum Dis 28:92&931, Yunus MB, Masi AT, Calabro JJ, Miller KA, Fei- genbaum SL: Primary fibromyalgia (fibrositis): clinical study of 50 patients with matched normal controls. Semin Arthritis Rheum 11: , Moldofsky H, Scarisbrick P, England R, Smythe H: Musculoskeletal symptoms and non-rem sleep disturbance in patients with fibrositis syndrome and healthy subjects. Psychosom Med 37: , Golden H, Weber SM, Bergen D: Sleep studies in patients with fibrositis syndrome (abstract). Arthritis Rheum (suppl) 26:S32, Ellman P, Savage OA, Wittkower E, Rodger TF: Fibrositis: a biographical study of fifty civilian and military cases, from the rheumatic unit, St. Stephen s Hospital and a military hospital. Ann Rheum Dis 3:5676, Committee of the American Rheumatism Association Section of the Arthritis Foundation: Primer on the Rheumatic Diseases. Seventh edition. New York, The Arthritis Foundation, Beetham WP: Diagnosis and management of fibrositis syndrome and psychogenic rheumatism. Med Clin North Am 63:433439, Ropes MW, Bennett GA, Cobb S, Jacox R, Jessar RA: 1958 revision of diagnostic criteria for rheumatoid arthritis. Bull Rheum Dis 9: , Melzack R: The McGill Pain Questionnaire: major properties and scoring methods. Pain 1: , Norusis MJ: SPSSx Advanced Statistics Guide. New York, McGraw-Hill, Yunus MB: Differential diagnosis of primary fibromyalgia syndrome. Intern Med Specialist 43-74, Burckhardt CS: The use of the McGill Pain Questionnaire in assessing arthritis pain. Pain 19: , Dubuisson D, Melzack R: Classification of clinical pain descriptions by multiple group discriminant analysis. Exp Neurol 51:0-7, Leavitt F, Garron DC, Whisler WW, Sheinkop MD: Affective and sensory dimensions of back pain. Pain 4: , Atkinson JH, Kremer EF, Ignelzi RJ: Diffusion of pain language with affective disturbance confounds differential diagnosis. Pain 12: , Leavitt F, Garron DC: Psychological disturbance and pain report differences in both organic and non-organic low back pain patients. Pain 7: 7-195, Payne TC, Leavitt F, Garron DC, Katz RS, Golden HE, Glickman PB, Vanderplate C: Fibrositis and psychologic disturbance. Arthritis Rheum 25: , Ahles TA, Yunus MB, Riley SD, Bradley JM, Masi AT: Psychological factors associated with primary fibromyalgia syndrome. Arthritis Rheum 27: , Wolfe F, Cathey MA, Kleinheksel SM, Amos SP, Hoffman RG, Young DY, Hawley DJ: Psychological status in primary fibrositis and fibrositis associated with rheumatoid arthritis. J Rheumatol 11:5W506, 1984