|Home | About | Journals | Submit | Contact Us | Français|
To estimate the relationship between the presence of vulvodynia, fibromyalgia, interstitial cystitis, and irritable bowel syndrome.
Validated questionnaire-based screening tests for the four pain conditions were completed by women with and without vulvodynia who were participating in the Michigan Woman to Woman Health Study -- a longitudinal population-based survey in southeastern Michigan. Weighted population-based estimates of the prevalence and characteristics of participants with these chronic comorbid pain conditions were calculated using regression analyses.
Of 1,940 women who completed the survey containing all four screening tests, 1,890 (97.4%) answered all screening questions and were included. The prevalences of the screening-based diagnoses ranged from 7.5% (95% CI 6.2, 9.0) for interstitial cystitis, 8.7% (95% CI 7.3, 10.4) for vulvodynia, 9.4% (95% CI 8.1, 11.0) for irritable bowel syndrome, to 11.8% (95% CI 10.1, 13.7) for fibromyalgia, with 27.1% screening positive for multiple conditions. The presence of vulvodynia was associated with the presence of each of the other comorbid pain conditions (p<0.001, OR = 2.3 to 3.3). Demographic risk factors for each condition varied. Increasing age was not associated with greater numbers of comorbid conditions, and only low socioeconomic status was associated with having multiple comorbid conditions concurrently.
Chronic pain conditions are common, and a subgroup of women with vulvodynia is more likely than those without vulvodynia to have one or more of the three other chronic pain conditions evaluated.
Chronic pain conditions including vulvodynia, fibromyalgia, interstitial cystitis, and irritable bowel syndrome are now known to be frequently underdiagnosed and are much more prevalent than previously estimated (1–5). Despite their disparate clinical presentations, studies of specific pain disorders report similarities in their findings – such as systemic sensitivity as illustrated with peripheral pain testing, (6–9) similar alterations on functional MRI scanning of the pain network associated sites in the brain, (6, 8, 10) and differences in descending pain modulation in the central nervous system (11–13). This realization has led to an interest in understanding the pattern of co-occurrence of these chronic pain conditions as well as in investigating common mechanisms that may underlie these conditions and explain their shared features.
We report on data from a population-based cohort study of women with and without vulvodynia southeast Michigan that incorporated validated questionnaire-based screening scales for vulvodynia, fibromyalgia, interstitial cystitis, and irritable bowel syndrome. This analysis estimates the prevalence and co-occurrence of these conditions and their associated demographic risk factors.
This cross-sectional analysis uses data from the 6-month follow-up survey of the Michigan Woman to Woman Study, a population-based, ethnically diverse cohort study of women with and without vulvodynia in southeast Michigan. The study was approved by the University of Michigan Medical Institutional Review Board. Recruitment and survey methodology has been previously described (2). In brief, women ages 18 and older were recruited using random digit dialing by trained interviewers (The Institute of Social Research, University of Michigan, Ann Arbor, MI). Following random selection of one eligible woman within a household, informed consent was obtained, and a brief telephone interview conducted. Women were then sent a more extensive online or written survey (based on participant choice), which assessed demographic characteristics, exposures, urogenital symptoms, and medical history. Follow-up surveys were sent every 6 months to assess changes in symptoms/diagnoses and potential risk factors.
Previously validated survey-based screening scales for several pain conditions were incorporated into the 6-month follow-up survey including a validated screen for vulvodynia, (14) the Fibromyalgia Impact Questionnaire, (15) the Rice High Specificity definition for interstitial cystitis, (16) and the Rome II criteria for irritable bowel syndrome (17). The validated criteria for vulvodynia included having had more than 3 months of vulvar discomfort, located at the opening of the vagina, that had not resolved (provoked or unprovoked, primary or secondary). (14). The criteria for interstitial cystitis included a combination of symptoms of urinary frequency and dysuria, (16) and that for fibromyalgia included the presence of chronic pain or discomfort in four quadrants of the body (15). Irritable bowel syndrome was predicted by a combination of abdominal pain associated with altered bowel frequency and/or function (17).
Initial weighting of the baseline sample to reflect the population of the four-county area of southeast Michigan was conducted. The weights incorporated a) the probability of selection in the sample, b) the non-response rate, and c) a post-stratification component using the specific age-county distribution of the population (see reference (2) for complete details on the weight construction)(2). The weighted prevalence and 95% confidence intervals (CI) were estimated for each condition. These weights were also utilized in the two-sample t-tests and chi-square tests in order to investigate any differences between the women who filled out the 6-month survey vs. those who did not. Weighted t-tests and chi-square tests replace the sample means, standard deviations and percentages by the corresponding weighted versions in the relevant formula and are carried out routinely in standard statistical software packages. We used the Complex Samples Module of the PASW (Version 18.0, Armonk, New York), for our purposes.
Odds of having vulvodynia, fibromyalgia, interstitial cystitis, and irritable bowel syndrome were investigated by means of separate weighted multiple logistic regression models which have as primary independent variables the indicators denoting presence or absence of each of the other three comorbidities. Each model was further controlled for age, marital status, ethnicity, education, and an indicator for difficulty paying for basics -- the variables previously found to be associated with the prevalence of vulvodynia in our previous manuscript.(2) These analyses were repeated replacing the individual comorbidity indicators by their sum, which denotes the frequency of other comorbidities. A similar logistic regression analysis was carried out with the presence/absence of multiple (2 or more) comorbid pain conditions simultaneously. The relationships between the presence or absence of vulvodynia and the presence of the other comorbid pain conditions, singly and in specific combinations of two or three, were estimated using a weighted chi-square test.
Between September 2008 and September 2009, 2542 women, ages 18 and older, were recruited randomly from the four-county area of southeast Michigan. Of these, 1,940 (76.3%) completed both the baseline and 6-month follow-up surveys. Women who completed the 6-month follow-up survey were similar in age to those who did not (50.7 years vs. 51.1 years, p=0.62), but respondents were more likely to be White (78.2% vs. 58.2%, p<0.001), married (66.1% vs. 50.4% married, p<0.001), have at least a high school education (95.8% vs. 90.0%, p<0.001), have a household income greater than $60,000 (53.5% vs. 39.3%, p<0.001), to rate their health as excellent or very good (53.5% vs. 44.6%, p< 0.003), and to indicate they have little or no pain (65.4% vs. 57.3%, p=0.004). Of those completing the 6-month follow-up survey, 1,890 (97.4%) completed all screening questions needed to predict the presence of vulvodynia, fibromyalgia, interstitial cystitis, and irritable bowel syndrome, and were included in this analysis. Their mean age was 50.4 (SD of 16.2) years; 76.8% were White, 16.2% Black, 2.4% Hispanic and 4.5% other ethnicities; 66.4% were married or living as married; 96.0% were high school graduates; 38.6% found it very hard or somewhat hard to pay for basics; and 64.1% had had intercourse in the past 6 months.
The prevalence of the four chronic comorbid pain conditions, using weighted data, ranged from 7.5% (95% CI 6.2, 9.0) for interstitial cystitis, 8.7% (95% CI 7.3, 10.4) for vulvodynia, 9.4% (95% CI 8.1, 11.0) for irritable bowel syndrome, to 11.8% (95% CI 10.1, 13.7) for fibromyalgia. (Figure 1) This translates into 72,000 to 114,000 women in the four-county area with each of the conditions. The proportion of women with each diagnosis that screened positive for any of the conditions is shown in Figure 1. Overall, 72.9% (95% CI 70.2, 75.4) of the 1890 women with complete data on the four chronic pain conditions screened negative for all four conditions, and 19.1% (95% CI 17.0, 21.4) reported having only one of the four conditions.
The adjusted odds ratios of having each of the chronic comorbid pain conditions when each of the other three was present are shown in Table 1. The odds of screening positive for vulvodynia was increased 2.3 to 3.3 fold when a woman reported having any of the three other chronic pain conditions. As the number of other chronic pain conditions increased from 0 to 3, the odds of having vulvodynia increased over 5 fold. Similar associations were seen with the other three comorbid disorders, with the odds of multiple comorbid conditions occurring simultaneously being highest for those with fibromyalgia.
As suggested by Figure 2, the combinations of multiple disorders are not uniformly distributed. For example, the combinations of other disorders among women with and without vulvodynia vary, with vulvodynia cases being more likely to have interstitial cystitis, or irritable bowel syndrome, but not fibromyalgia, as a sole other comorbidity. However, those with fibromyalgia in combination with interstitial cystitis or irritable bowel syndrome were seen more commonly in women with vulvodynia than in those without this disorder. (Figure 2).
Figure 3 shows the weighted prevalence rates for all four comorbid conditions across specific age-decades. The associations between demographic characteristics of women with each disorder are shown in Table 2. For vulvodynia, fibromyalgia, and interstitial cystitis, the prevalence rates exhibit an inverted U-shaped pattern as a function of age, with maximal prevalence in middle age. In contrast, the pattern of change with age for irritable bowel syndrome is roughly U-shaped, with an increase noted in older age groups. Vulvodynia varied with age, was increased in married women and in those with lower education, and was less prevalent in Black women. Conversely, fibromyalgia was associated with increasing age, lower education, and greater difficulty paying for basics, and was the only disorder with a suggestion of an increased prevalence among Black women. When the relationship between these demographic factors and increasing numbers of chronic comorbid pain conditions was evaluated (data not shown), no association was noted with age or ethnicity, marital status, or higher education. However, screening positive for two or more conditions was associated with difficulty paying for basics. Furthermore, the odds of increasing numbers of disorders rose progressively among those having difficulty paying for basics such as food, clothing, or housing (OR for difficult paying for basics of 1.4, 1.8 and 2.4 for those with 1, 2, or 3–4 conditions, p<0.05 for each).
We further assessed other characteristics of vulvar pain that might predict the co-occurrence of multiple comorbid pain conditions. Among those screening positive for vulvodynia, there was no association between age at onset of vulvodynia, duration of vulvodynia symptoms, or maximum severity of vulvar pain they ever had experienced with the presence of any of the three other chronic comorbid pain conditions (data not shown).
Over the past 10 years the substantial prevalence of chronic comorbid pain conditions has become increasingly evident. This is one of the first population-based studies to screen for and examine the association between vulvodynia and three other chronic comorbid pain conditions. Compared to women without vulvodynia, those with this condition are two to three times more likely to have one or more other chronic comorbid pain conditions.
The Institute of Medicine reported that at least 116 million persons in the United States suffer from chronic pain, with national annual cost estimates of over ½ billion dollars (18). This report stresses the need for better data on the incidence and prevalence of pain in order to reduce the impact of pain and the resultant suffering it causes. Studies on a number of pain conditions, such as fibromyalgia, temporomandibular joint disorder, interstitial cystitis, and irritable bowel syndrome indicate the high likelihood of co-occurrence of multiple chronic pain disorders, (19–24), and that the costs associated with a specific disorder are often substantially increased due to the other comorbid pain conditions present (25).
Vulvodynia has not routinely been included in comorbidity studies of chronic pain conditions, (26) despite the similarities in systemic sensitivity, as illustrated with peripheral pain testing, (6–9) similar alterations on fMRI scanning of the pain network associated sites in the brain, (6, 8, 10) and differences in descending pain modulation in the central nervous system (11–13). This lack of inclusion of vulvodynia in many of the previous studies may be due in part to the previous lack of validated screening tests for vulvodynia (14, 27). Chronic pain conditions, in general, are often underdiagnosed; thus, comorbidity is underestimated when only those who have been previously diagnosed clinically are included. In this population-based cohort of adult women in southeast Michigan, we determined the presence of vulvodynia and of fibromyalgia, interstitial cystitis, and irritable bowel syndrome, using previously validated criteria (14–17). While the prevalence based on these criteria may differ from that estimated using only in-office diagnoses, the criteria do provide an estimate of women who report symptoms characteristic of the disorders who may never have been diagnosed or sought treatment.
In a review of previous research on the overlap between chronic pain disorders, Rodriguez and colleagues reported on four substantial methodological shortcomings that undermined the strength of conclusions (24). These included 1) participants were drawn primarily from tertiary care clinics limiting generalizability, 2) inconsistencies in assessment, including use of case definitions, self-reported physician diagnoses, review of ICD-9 codes, etc., 3) predominance of women in studies, and 4) variability in control groups, including healthy individuals, those with other chronic illnesses, or those with their health status non-specified. In previous studies, vulvodynia has been shown to be associated with other chronic pain conditions, (28–32) however, these studies were conducted in clinical practices, (22, 29–31) and/or were based on reported comorbidities which may have underestimated comorbid diagnoses (31). The current study used a population-based sample reflecting the female population in southeast Michigan and verified screening-based criteria, and hence provides further evidence of the substantial associations between these chronic pain disorders in the general population.
We found that although women with vulvodynia were more likely to have any one of the other comorbid disorders, 72.9% screened negative for all four entities, and only approximately one half of those with any one disorder met criteria for at least one additional condition. This finding that having multiple comorbid pain conditions is a characteristic of only a subset of those with a pain condition has been previously described (19, 20). The fact that some women present with multiple conditions and others report only one suggests differing pathophysiology, differences in the time of onset of (other disorders destined to occur later), or a lack of “central sensitivity” that may be the relevant neurologic condition that underlies this predominant grouping of comorbid pain conditions. Co-occurrence appears not to be universal, but rather may indicate a characteristic of a subset of those affected.
We found that the association between vulvodynia and the other chronic comorbid pain conditions was not randomly distributed. Others have found that as the number of comorbid pain conditions increases, the distribution of specific pain conditions is not uniform, (33, 34) with the types of pain conditions present forming distinct patterns. Further study of these variations, and the order in which these disorders appear may further clarify shared and differing pathophysiologic mechanisms.
Strengths and limitations of this study exist. The availability of data derived from previously validated screening inventories allowed us to assess prevalence and co-occurrences of these typically underdiagnosed disorders. In addition, these data were available on a general community-based population minimizing the bias commonly encountered when enrolling from health care facilities. However, the sensitivity and specificity of screening tests are imperfect – and hence some misclassification is likely to remain. However, validation studies suggest their use maximizes the ability to predict these conditions in large numbers of women in the community, despite the known clinical underdiagnosis and misdiagnosis rates inherent in the evaluation of these conditions to date.
In conclusion, vulvodynia is associated with other chronic comorbid pain conditions such as fibromyalgia, interstitial cystitis and irritable bowel syndrome, individually and in combination, and the presence of vulvodynia or any of the other comorbid pain conditions increases the likelihood that a woman will have one or more of the other chronic pain conditions. Further study on the relationships between these disorders, and the differences between those with multiple comorbid pain conditions and those with no or a solitary pain disorder is anticipated to identify previously unrecognized, evidence-based subgroups that may indicate a common pathophysiology, natural history, treatment response, or a combination of these.
Funded by the National Institute of Child Health and Human Development, The National Institutes of Health, R01 HD 054767.
The authors thank Laurie Legocki, PhD and Susan Countryman for project management; Jill Bowdler for assistance with participant correspondence and secretarial support; and Margaret Helmuth for data management. Salary support for these individuals was provided by grant NIH R01 HD 054767 (Susan Countryman and Margaret Helmuth) and by the Department of Family Medicine at the University of Michigan (Jill Bowdler).
Financial Disclosure: The authors did not report any potential conflicts of interest.