Brief, Self-Report Fibromyalgia Screener Evaluated in a Sample of Chronic Pain Patients

Abstract

Objective

To validate a self-report fibromyalgia screener in a chronic pain population.

Design

Adults with chronic pain were evaluated with a 6-item, self-report fibromyalgia screening tool, based on revised American College of Rheumatology (ACR) fibromyalgia diagnostic criteria, with fibromyalgia diagnosed when patients experienced chronic pain and scored ≥13 on the ACR Fibromyalgia Symptom severity scale. Patients were independently assigned clinical diagnoses by treating clinicians.

Setting

University-based, tertiary care pain clinic

Subjects

337 mixed chronic pain patients

Results

Agreement between the clinical diagnosis and screener diagnosis was good (P<0.001), with 76% sensitivity and 82% specificity.

Conclusion

A self-administered, brief fibromyalgia screening questionnaire can effectively identify chronic pain patients who will likely have clinical fibromyalgia.

Introduction

Fibromyalgia is a common condition among patients being evaluated for chronic pain, with a recent study identifying fibromyalgia in 41% of patients referred to a tertiary pain clinic (1). Diagnosing fibromyalgia can be challenging, with a multinational survey reporting it took an average of 2.3 years after first presenting fibromyalgia complaints to a physician before a patient was diagnosed with fibromyalgia (2). The American College of Rheumatology (ACR) originally published classification criteria for fibromyalgia in 1990, based on physician assessment and a tender point assessment (3). Applying the original ACR criteria to clinical practice has been problematic. For example, evaluation of 206 consecutive patients identified a clinical diagnosis of fibromyalgia for 49%, while only 29% satisfied ACR classification criteria (4). In particular, it was identified that the ACR tender point requirement hindered the fibromyalgia diagnosis, as many patients with clinical fibromyalgia had fewer positive tender points on their assessment day than the required cut-off for an ACR classification.

Recently, two revisions have been made to the ACR criteria, due to limitations identified with the original physical examination requirements (5). Currently proposed criteria are based on patient reports of number of pain locations that produces the Widespread Pain Index and the Symptom Severity score (based on the severity of fatigue, waking unrefreshed, and cognitive problems plus the presence or absence of headache, lower abdominal pain/cramps, and depression) (6). Scoring from both sets of questions are used for classification. These questions were designed to be asked to patients by their healthcare provider. Responses could be used to calculate a Fibromyalgia Symptoms scale (also called the fibromyalgia-ness or FS) score, with fibromyalgia identified when the calculated score is ≥13. As part of the criteria development process, the new ACR criteria were applied to a sample of 7233 rheumatology patients, 10% of whom were clinically diagnosed with fibromyalgia. Using the new ACR assessment, a FS score cut-off of ≥13 correctly identified 93% of patients, with a sensitivity of 97% and specificity of 92%. Recently, a self-adminstered version of criteria needed to determine FS score and pain chronicity were shown to be a valid in a German cross-sectional survey of individuals reporting fibromyalgia or chronic widepread pain (7).

The current study was designed to expand upon these earlier studies by developing a self-report fibromyalgia screening tool based on updated ACR criteria (6) and validating that instrument in a mixed group of chronic pain patients, This study was designed to the hypothesis that a brief self-report measure that permits calculation of FS will be an accurate fibromyalgia screening tool for clinicians to consider when evaluating patients with chronic pain complaints. This project surveyed chronic pain patients at a tertiary care clinic and compared the accuracy of a brief, patient-completed, self-report fibromyalgia screening questionnaire, based on the new ACR diagnostic criteria for fibromyalgia (6), with a clinical fibromyalgia diagnosis.

Methods

This study was conducted at a university tertiary pain clinic from March 28, 2012 to May 24, 2012. The study protocol was approved by a local Institutional Review Board as an exempt study not requiring informed consent from participants. All participants were informed of the research nature of this project.

A convenience sample of patients being evaluated or treated at a tertiary pain clinic was collected over the two-month study period. All adult patients (≥18 years of age) attending an appointment during the time of study were eligible to participate if they could understand English and complete the questionnaire. Both new and follow-up patients could participate, with patients only allowed to participate during a single visit.

Participants completed a one-page, 6-question self-report fibromyalgia screening questionnaire based on updated ACR fibromyalgia diagnostic criteria (Box 1) prior to meeting with their treating physician or physician assistant. Using the self-report questions, fibromyalgia was diagnosed in patients with pain ≥3 months (Question 6) and FS ≥13, with FS calculated as follows (6):

• Widespread Pain Index: number of items endorsed in question 1 are counted (range 0–19)

• Symptom Severity Score (range 0–12)

  • –
    Questions 2–4: responses to each item are assigned a numeric score using none=0, mild/intermittent=1, moderate/often=2, severe/continuous/disabling=3; the three scores are added (range 0–9)
  • –
    The number of items selected in question 5 are counted (range 0–3) and this number is added to the above score from questions 2–4 to produce the Symptom Severity Score

• FS score = Widespread Pain Index PLUS total Symptom Severity Score (range 0–31)

Box 1. Assessment questionnaires.

A. 6-item self-report fibromyalgia screener

1. Where have you had pain or tenderness over the last week

   (please check all that apply):

   • ◦
     Shoulder, left
   • ◦
     Shoulder, right
   • ◦
     Upper arm, left
   • ◦
     Upper arm, right
   • ◦
     Lower arm, left
   • ◦
     Lower arm, right
   • ◦
     Hip or butt, left
   • ◦
     Hip or butt, right
   • ◦
     Upper leg, left
   • ◦
     Upper leg, right
   • ◦
     Lower leg, left
   • ◦
     Lower leg, right
   • ◦
     Jaw, left
   • ◦
     Jaw, right
   • ◦
     Chest
   • ◦
     Abdomen
   • ◦
     Upper back
   • ◦
     Lower back
   • ◦
     Neck
2. Please rate the severity of your fatigue over the last week

   • ◦
     None
   • ◦
     Mild or intermittent
   • ◦
     Moderate or often troublesome
   • ◦
     Severe, continuous, or disabling
3. Please rate the severity of your waking not feeling refreshed over the last week

   • ◦
     None
   • ◦
     Mild or intermittent
   • ◦
     Moderate or often troublesome
   • ◦
     Severe, continuous, or disabling
4. Please rate the severity of your problems with concentration, memory, or thinking over the last week

   • ◦
     None
   • ◦
     Mild or intermittent
   • ◦
     Moderate or often troublesome
   • ◦
     Severe, continuous, or disabling
5. Over the last 6 months, have you experienced the following (please check all that you have experienced)

   • ◦
     Headache
   • ◦
     Abdominal pain or cramps
   • ◦
     Depression
6. Have you been having pain for at least the last 3 months?

   • ◦
     Yes
   • ◦
     No

B. Complementary questionnaire for patient's treating clinician

Has this patient had pain for at least 3 months?

• ◦Yes
• ◦No

Does this patient report having widespread pain (on left and right sides of the body AND above and below the waist)

• ◦Yes
• ◦No

Please record the results of today's tender point examination, rating each point from 0–10

1. _____

2. _____

3. _____

4. _____

5. _____

6. _____

7. _____

8. _____

9. _____

10. _____

11. _____

12. _____

13. _____

14. _____

15. _____

16. _____

17. _____

18. _____

From your clinical assessment, does this patient have fibromyalgia? ( ) YES ( ) NO

What is this patient's primary pain diagnosis?

• ◦Fibromyalgia
• ◦Myofascial pain
• ◦Arthritis or other mechanical pain
• ◦Neuropathic pain
• ◦Migraine or other headache disorder
• ◦Abdominal pain
• ◦Other: ________________________

Clinician initials: _________________________________

In the original sample evaluated with these criterion (6), fibromyalgia patients had FS scores ranging from 12–31. As noted above, a cut-off of 13 was identified as the best predictor of fibromyalgia and that cut-off was used in the current study. Patient questionnaires were independently scored by an investigator who was not involved in the clinical care of the participants and this information was not provided to the assessing clinicians.

Each patient underwent a clinical evaluating with their treating clinician on the same day on which the patient's completed the self assessment screener. Clinical examinations occurred after the screener had been completed. Clinicians separately completed a clinician assessment questionnaire follwoing their clinical evaluation of each patient to provide information on pain chronicity, widespread nature of the pain, tender point assessment scores, and clinical diagnosis.

Data analysis

A clinical fibromyalgia diagnosis was based on clinical assessment by the examining clinician. Because fibromyalgia symptoms and signs may fluctuate with time and with treatment, patients were not required to meet ACR 1990 diagnostic criteria for the clinician's diagnosis to be positive. Results of the tender point examination were used to calculate a tender point count and score. Positive tender points were determined using a cut-off score of at least 2 on a 0 to 10 severity scale, which was previously shown to discriminate between fibromyalgia patients and other pain patients in a tertiary pain clinic setting (8). Descriptive statistics were used to evaluate fibromyalgia severity among participants with and without fibromyalgia.

The primary analysis was an evaluation of the accuracy of the self-reported diagnosis based on FS ≥13 in patients with chronic pain (duration ≥3 months), using the physician diagnosis as the gold standard and a Chi-square analysis, with additional calculation of sensitivity and specificity for diagnostic agreement. Wolfe recommended using the median FS score of 19 identified in the original validation sample as a severity cutoff (6). Accuracy of diagnostic agreement was additionally assessed to determine the influence on diagnostic accuracy of fibromyalgia severity (FS 13–19 vs. >19) and the individual evaluating clinician.

Results

A total of 337 adult patients under the care of three clinicians were included (Table 1). Participants were evaluated by one of two physicians (n=249) or one physician assistant (n=88), each specializing in pain management. Widespread pain was reported in eight of the 24 patients with a primary diagnosis of neuropathic pain (three had a diagnosis of secondary fibromyalgia) and in two of the 13 patients with primary myofascial pain (one had a secondary diagnosis of fibromyalgia). None of the patients with primary abdominal pain had widespread pain or a secondary fibromyalgia diagnosis. Table 2 provides a comparison of fibromyalgia features based on method of diagnosing fibromyalgia (clinical assessment vs. self-report screener). The screener incorrectly diagnosed fibromyalgia in two patients with myofascial pain (one of whom had widespread pain), one patient with neuropathic pain who had widespread pain, and one patient with abdominal pain.

Table 1.

Participant demographics

Characteristic	Number N=337
Age, years
Mean (SD)	50.9 (12.9)
Range	20–89
Gender, n (%)	(n=327)
Male	108 (33.0)
Female	219 (67.0)
Clinician primary diagnosis
Arthritis or other mechanical pain	152 (45.1)
Fibromyalgia	87 (25.8)
Migraine or other headache disorder	27 (8.0)
Neuropathic pain	24 (7.1)
Myofascial pain	13 (3.9)
Abdominal pain	12 (3.7)
Other	22 (6.5)
Patients with secondary clinical diagnosis of fibromyalgia	57 (16.9)

Table 2.

Comparison of fibromyalgia characteristics based on diagnostic criteria used

	Clinical fibromyalgia diagnosis		Self-report questionnaire fibromyalgia (FS ≥13 plus chronic pain)
	Fibromyalgia n=144	Not fibromyalgia N=193	Fibromyalgia N=145	Not fibromyalgia N=192
Fibromyalgia score (range 0–31)	17.2 (5.9)	9.0 (4.5)	18.7 (4.5)	7.9 (3.1)
	17	9	18	8
	53 (36.8)	6 (3.1)	58 (40.0)	1 (0.5)
Mean (SD)
Median, N
>19, N (%)
Tender point count, mean (SD)	14.6 (3.8)	3.1 (3.1)	12.7 (5.6)	4.5 (5.0)
Tender point score, mean (SD)	91.3 (38.4)	15.3 (16.6)	79.6 (47.2)	23.8 (29.2)
Met ACR 1990 criteria* (%)	111 (77.1)	2 (1.0)	94 (64.8)	19 (9.9)

^*Pain at least 3 months, widespread pain present, and tender point count ≥11

Agreement between the clinical diagnosis of fibromyalgia and the self report screener was good (P<0.001; Table 3). Sensitivity and specificity, respectively, were 76.4% and 81.9%. Positive predictive value was 75.8% and negative predictive value was 82.3%. Separate analyses were performed to determine diagnostic agreement between clinical assessment and the self-report screener for each individual participating clinician, with similarly good agreement for all clinicians. Furthermore, among the 278 participants with FS ≤19, agreement between clinician diagnosis and the screener was good (P<0.001). Among the 59 participants scoring >19, all but one had a clinical diagnosis of fibromyalgia, with 52 of these 58 correctly classified by the screener.

Table 3.

Agreement between clinical assessment and self-report questionnaire

	Clinician diagnosed fibromyalgia	Clinician diagnosed no fibromyalgia	Total
Self-report questionnaire positive for fibromyalgia*	110	35	145
Self-report questionnaire negative for fibromyalgia	34	158	192
Total	144	193	337

^*Fibromyalgia positive required: FS ≥13 and pain present for at least 3 months

Discussion

The self-report fibromyalgia screener, based on updated ACR diagnostic criteria, was an effective tool to screen for fibromyalgia among patients with a variety of chronic pain complaints. Interestingly, although comorbid symptoms included in the updated ACR diagnostic criteria are also common among other types of chronic pain (9–13), the updated ACR criteria successfully differentiated those chronic pain patients with clinically recognized fibromyalgia. These data expand on the earlier report validating an FS score ≥13 among rheumatology patients (3). Furthermore, these data support that ACR questions may be completed independently by patients as a self-report assessment measure.

This study is limited by evaluation of a single center population, with confirmatory results needed from other centers. Furthermore, the subjective nature of the clinical diagnosis also limits study findings. While the self-report screener cannot be used to replace clinical assessment, as other medical conditions may mimic important features of fibromyalgia and diagnosis requires exclusion of other potential conditions, this tool might be beneficial to help identify patients who require a clinician assessment for fibromyalgia.

Conclusion

This brief, self-report fibromyalgia screener is a valid tool for identifying chronic pain patients who are likely to have fibromyalgia. While a clinical evaluation is necessary to confirm a fibromyalgia diagnosis and rule out other potential conditions that might mimic fibromyalgia, the screener may be useful for identifying patients who may have fibromyalgia for both clinical practice and for research.