Fibromyalgia is a common disorder characterized by widespread musculoskeletal pain. However, the exact cause of this prevalent condition is not entirely understood. The data from cross-sectional studies have revealed an association of various psychological, somatic and social factors with fibromyalgia, but, only a few prospective studies support these observations. Therefore, most of the reviews have to rely solely on the current data for etiology, which otherwise can only be confirmed by the prospective studies. The incidence of fibromyalgia varies from 1.2% to 5.4% as per the various definitions of fibromyalgia. Chronic widespread pain (CWP) is more prevalent, hitting nearly 14.2% of the UK population. As per the Forseth study, CWP is two times more prevalent than that of fibromyalgia. 

One systematic review has depicted physical and psychological traumatic situations in the development of fibromyalgia while the prospective studies have included people with whiplash or other injuries. The increased prevalence of fibromyalgia in patients with diabetes, irritable bowel syndrome and patients undergoing hemodialysis has been reported by another review.  The results of a recent review about the transformation of regional pain to CWP were found to be uncertian due to the inclusion of only a few studies. Another systematic review established a bidirectional association between CWP and sleep disturbances.

The present review covered the studies of CWP and fibromyalgia. The authors estimated the incident rate separately while the estimation of risk factors has been done collectively to increase the number of included studies.

Rationale behind research:

There is a lack of systematic review describing the fibromyalgia onset in population-based samples. The present study was conducted to fulfill this area of unmet need.

Objective:

The objective of this review was to identify all studies depicting incidence of CWP and fibromyalgia in population based samples and to assemble the risk factors for new onset of both CWP and fibromyalgia.

In this study, two reviews were performed. The first review was conducted to include the articles indicating incidence of CWP and fibromyalgia in the general population. The second review was to combine the risk factors for new-onset of CWP or fibromyalgia. The overlapping of the two reviews occurred and many studies included were found to be common for both the reviews and hence they are performed collectively.  For both reviews, the method of the search was comparable.

Literature Search:

The articles of interest published amid January 1998 and July 2019 were systematically searched from Web of Science, Ovid Medline and Cochrane Database of Systematic Reviews. The terms used for searching the database were onset, incidence, population and epidemiology firstly in context to "fibromyalgia" and after that with "chronic widespread pain". The handpicked searching for the relevant articles resulted in the finding of good-quality studies. For additional relevant studies, systematic reviews published in the past were also explored. The author performed all the searches.

Inclusion criteria:

The criteria for inclusion were similar for both reviews. The studies were considered for inclusion if:

• the prospective study included a population-based cohort or if it was a high-quality retrospective cohort so that the risk factors were reported well ahead of the onset of fibromyalgia
• the study included the control group
• the study included patients with a confirmed diagnosis of fibromyalgia by a physician or as per standard questionnaire preferably ACR criteria or self-report
• the study mentioned that patients diagnosed with fibromyalgia prior to the date of study initiation were excluded

The case studies, case reports, cross-sectional studies or case control studies were not included in the review.

Data Extraction:

For each article, the following were recorded: first author/year, study design, base population, control group, mode of fibromyalgia diagnosis, number of new cases/duration of the study, incidence, risk factors in univariate, and in multivariate analyses and quality. The rate of incidence was standardized to new cases per 1000 per annum to tabulate the results.

Assessment of Quality:

Newcastle-Ottawa Scale for cohort studies was utilized to estimate the quality of the chosen studies. The quality was termed to be high if the selection domain had 3 to 4 stars, comparability domain 1 or 2 stars and outcome/exposure has 2 to 3 stars, fir when selection domain has 2 stars, comparability domain had 1 to 2 stars and outcome/exposure domain had 2 to 3 stars, other termed as a low-quality study.

Outcome Measure:

The diagnosis of fibromyalgia or CWP was the primary outcome. The research database with all patient contacts for all diagnosis was searched for the clinical diagnosis of fibromyalgia. In most of the studies, ICD-diagnosis or a GP diagnostic code was used to diagnose fibromyalgia. The patients who had a minimum of two medical encounters during the reporting of fibromyalgia diagnosis were included by a few studies. Self-reporting was also utilized by two studies. Most of the other studies utilized a questionnaire to obtain widespread pain as per ACR criteria. The recording of the incidence rates was done separately, as different methods of diagnosis were used.

Several studies compared the large population group with a cohort with a specific medical disease under examination as a possible risk factor for fibromyalgia. An adjustment to represent the total population had to be made for these studies. Three methods were utilized to estimate the strength of association between risk factor and onset of fibromyalgia or CWP: Adjusted odds or hazard ratio, an exposure-response relationship and a reported bidirectional association.

Outcomes:

Incidence of fibromyalgia and CWP

The median incidence was found to be 12.5 per 1000 person-years, data from the 11 cohorts of new-onset CWP that involved both male and female patients. The three studies were of a short follow up duration with the incidence of 58, 61 and 81.6 per 1000 person-years while the remaining had a long follow up of 3-13 years and median incidence of 9.4 per 1000 person-years. No considerable female predominance was noted in half of the studies.

The incidence rate of fibromyalgia was reported by 12 articles. The 4 articles were included due to the predominance of one sex. The median incidence of physician-diagnosed fibromyalgia the remaining 8 articles containing general population were found to be 3.6 per 1000 person-years. Post-adjustment, in which the patients with some specific medical condition were excluded, the median incidence was found to be 4.3 per 1000 person-years. The higher incidence of fibromyalgia in women than men was reported in all studies except 2. The median incidence of physician diagnosis fibromyalgia was found to be 14 in six population-based cohorts with a specific medical condition. Also, no female predominance was noted in these cohorts. However, one study showed an increased risk of fibromyalgia in men with inflammatory bowel disease (IBD) compared to women with IBD.

The separate data of patients with some or severe regional pain at baseline was provided by six of the CWP studies. The median incidence of these six studies was found to be 67 per 1000 person-years as compared to 14 per 1000 person-years of the pain-free patients at the baseline.

Risk factors for CWP and Fibromyalgia

The risk factors for fibromyalgia and CWP development were provided by 12 and 22 studies, respectively. Studies estimating physician-diagnosed fibromyalgia included limited information on demographics but with a full range of data on medical disorders, including depression and sleep disorders. On the contrary, CWP studies estimated a more extensive range of risk factors. The data reporting similar risk factors for CWP and fibromyalgia was found to be insufficient.

A total of 12 and 11 studies reported female sex and middle or older age as risk factors, respectively. However, for both the variables, four studies were found indicating these variables not linked with the onset of the disease. Thirteen studies reported musculoskeletal disorders as a risk factor. In contrast, 2, 1, 2, 3, 3, 4, 1, 1, 1, and 3 studies reported IBD, gastroesophageal reflux disease, peptic ulcer, hypertension, diabetes, hyperlipidemia, multiple sclerosis, stroke, coronary heart disease and medical comorbidities as risk factors, respectively. A total of 13 and 11 studies frequently reported sleep disorders and anxiety/depression as risk factors, respectively. But, there was less frequent reporting of somatic symptoms, illness behavior, and stress as risk factors as only a few studies have considered these for inclusion.

High BMI as a lifestyle risk factor was reported in 7 studies, while two studies have reported this factor to be protective. 8 studies have reported smoking to be a risk factor. Two studies have reported no drinking of alcohol as a risk factor, and three studies have reported moderate alcohol consumption to be protective. The massive physical task had been reported to be a risk factor in 4 studies. One study reported physical inactivity as a risk factor while it was not reported in another study.

Strength of the association between risk factor and onset

The pattern of risk factors in the multivariate analyses, including all appropriate variables was comparable to that of univariate analyses; however, there were only a few variables which were considerably linked to the onset, mentioned below in the descending order of frequency:

Musculoskeletal disorders, including pain (11 studies)

Medical disorders, Sleep disorder (9)

Smoking (6)

Raised BMI, Older/middle age (5)

Female, Anxiety/depression, Illness behavior (4)

Somatic symptoms (3),

Heavy work, Allergy (2)

Stress, a few years of formal education (1)

Three studies reported moderate alcohol to be protective. The highest odds ratios were recorded for headaches, pain disorders and somatic symptoms. Also, the additive effect of multiple risk factors (somatic symptoms, pre-existing regional pain, sleep, heavy work and illness behavior) has been reported by three studies. Fibromyalgia was found to have a bidirectional relationship with each of the following: headache, GERD, insomnia, migraine, depression and IBD. Fibromyalgia was predicted by each of these six disorders, and that is why fibromyalgia is a risk factor for them.

As per the multivariate analysis, an exposure-response association was noted between the severity of risk factor and the possibility of developing CWP for the following: number of sleep problems, headache, number of somatic symptoms, number of pain sites at the baseline, illness behaviour score, severity of osteomyelitis, allergy and number of serious illness when young. A linear relationship was found between the IQ at age 11 and the risk of CWP at 45 years. An exposure-response relationship was noted between the lack of physical activity and increased possibility of CWP only in women, as per the Tromso study. In patients, less than 60 years, an association between fibromyalgia onset and osteomyelitis was noted, the most powerful was found in less than 35 years. Likewise, for patients under 50-60 years, IBD was a risk factor, not for others.

The present review shows a range of risk factors for CWP and fibromyalgia, which can be best, suited for multiple causal pathways into CWP/fibromyalgia. The review also highlights number of questions which could be answered in future research; like evaluating the difference in the risk factor based on age and sex or estimating the difference in the causal pathway of older people with comorbidities compared to young healthy people. The risk factors for fibromyalgia common for everyone include sleep disturbance, fatigue and somatic symptoms; however, these factors have now been considered as a part of fibromyalgia syndrome and have been incorporated in the latest ACR diagnostic criteria.  

The study’s biggest strength was its restriction to the population based cohort and exclusion of the people who had fibromyalgia before the study was initiated. Also, the risk factors were measured well ahead of the fibromyalgia onset so that there is no confusion about the association of risk factor and onset.

For this CWP/fibromyalgia review, the inclusion criteria were met by a total of 37 studies, which represents a significant set of information not reported in previous studies. The incidence of CWP was found to be higher than that of physician-diagnosed fibromyalgia. Patients with preexisting medical disorders had the highest prevalence of fibromyalgia. The higher prevalence of fibromyalgia as expected was reported in females, the lower incidence was reported in CWP and not visible in fibromyalgia patients with some other medical conditions. The wide range of risk factors like adverse lifestyle, regional or other pain disorders, preexisting medical conditions, difficulties during childhood, sleep disturbances, negative health perception, depression and somatic symptoms have been revealed in this review. These factors are linked to the fibromyalgia, which has been presented in the studies of prevalent cases with hepatic, rheumatic, GIT and psychological disorders. The authors of the study feel that the consideration of these risk factors as true factors lies prevails in utilizing them for the development of causal pathways.