The epidemiology and health-related quality of life associated with functional gastrointestinal disorders are reviewed, with particular emphasis on irritable bowel syndrome and functional dyspepsia. The literature supports the significant world-wide prevalence of functional gastrointestinal disorders, including irritable bowel syndrome (IBS), functional dyspepsia and chronic constipation. An increased female prevalence has been demonstrated in most studies in patients with IBS and chronic constipation, but not functional dyspepsia. The female to male ratio appears to be greater in the health care-seeking population than in community populations. However, some differences in the reported general prevalence and gender-related prevalence of functional gastrointestinal disorders may be due to cultural factors and study methodology. A significant health care burden is associated with IBS, with increased out-patient services, abdominal and pelvic surgeries, and gastrointestinal- and non-gastrointestinal-related physician visits and health care costs. Health-related quality of life is impacted significantly in patients with functional gastrointestinal disorders, such as functional dyspepsia and IBS, compared with the general healthy population, as well as patients with other chronic medical conditions, such as gastro-oesophageal reflux disease and asthma. Impaired health-related quality of life has been demonstrated, in particular, in patients with moderate to severe disease seen in referral settings. The health-related quality of life appears to improve in treatment responders, or correlates with symptom improvement, with at least some treatment modalities studied in functional gastrointestinal disorders, but further studies are needed. Predictors of health-related quality of life in patients with functional gastrointestinal disorders include psychosocial factors, such as early adverse life events, and symptoms related to visceral perception, e.g. pain and chronic stress. The presence of extra-intestinal symptoms appears to have a major if not greater impact on health care visits, excess health care costs and health-related quality of life in patients with functional gastrointestinal disorders.
General prevalence of functional gastrointestinal disorders (FGIDs)
Functional bowel disorders are amongst the most common medical conditions seen in primary care as well as in gastroenterology speciality clinics. Irritable bowel syndrome (IBS) is the most common functional bowel disorder, has a prevalence in the range 5–25% and accounts for 36% of all visits to gastroenterologists. If other non-IBS FGIDs are included, they account for 41% of diagnoses in speciality practices.1, 2 With regard to dyspepsia, there is a prevalence of up to 25% in Western countries, even after the exclusion of individuals with typical gastro-oesophageal reflux disease (GERD) symptoms.3–5 The annual incidence of dyspepsia is approximately 9–10%,6, 7 and 15% of patients have chronic (> 3 months in a year), frequent (> 3 episodes per week) and often severe symptoms. Functional dyspepsia is the most common cause of dyspeptic symptoms.4 Chronic constipation is a common condition with a prevalence rate in the range 1.2–27%.8–12
The first presentation of IBS patients to a physician usually takes place between the ages of 30 and 50 years, with a decrease in reporting frequency in older subjects. However, not all individuals with IBS seek medical care for their symptoms. Based on epidemiological studies conducted in different countries, 20–75% of subjects who meet the diagnostic criteria for IBS will seek medical attention at some point in their lives.13, 14 Cultural factors, as well as differences in health care systems and access to medical care, may influence the wide range of these rates. Although most individuals with IBS do not consult a physician, there are still between 2.4 and 3.5 million physician visits annually for IBS in the US. The burden of illness of this disorder has recently been estimated to be $1.6–10 billion in direct and $19.2 billion in indirect costs.15–17
Gender differences in the prevalence of FGIDs
Several population-based studies have found women to be at least twice as frequently affected as men, with prevalence ratios ranging from 2 : 1 to 3 : 1.18 Only a limited number of studies have compared the prevalence of FGIDs in men and women. Locke et al. compared the prevalence of non-cardiac chest pain in male and female community residents using mailed questionnaires.19 Non-cardiac chest pain was defined as ‘chest pain, any pain or discomfort felt inside the chest but not including heartburn or any pain that is primarily in the abdomen’. No significant gender differences in the prevalence of ‘non-cardiac chest pain’ were found. However, there appears to be a female predominance in tertiary care referral populations with functional chest pain of oesophageal origin. This condition represents a subgroup of those with non-cardiac chest pain, which has an equal gender prevalence in the general population but a higher female to male ratio in tertiary care referral centres.20
The prevalence of functional dyspepsia has been reported to be slightly higher in men than in women overall.21 However, a recent multicentre study conducted in Denmark and Sweden demonstrated that, in patients referred for upper abdominal symptoms who completed self-administered questionnaires and underwent an upper endoscopic examination, female gender was a significant predictor of functional dyspepsia compared with organic causes of dyspepsia. Patients with a history of GERD, peptic ulcer disease or a predominant symptom of heartburn or acid regurgitation were excluded. Female gender has also been associated with delayed gastric emptying in patients with functional dyspepsia22, 23 and in those with lower tolerance to the water-loading test.24 In the latter study, gender differences were seen in both healthy controls and patients with functional dyspepsia, and therefore these differences may not be specific to functional dyspepsia. Nonetheless, there seems to be little relationship between these physiological measures and symptom severity.
Two-thirds of individuals with IBS are female, with an estimated prevalence in women in the range of 14–24%.18 Of those who seek health care services, including tertiary and ambulatory care, for IBS and other functional bowel disorders, women lead men by a ratio of 2–2.5 : 1,8 whilst others estimate the rate to be higher at 3–4 : 1.25 However, the gender distribution appears to be less than 2 : 1 in IBS non-patients in the community.26 Moreover, studies conducted in Asian countries, such as India27 and Taiwan,28 have reported a higher male or equal gender prevalence of IBS. Cultural factors as well as study methodology may explain in part these gender differences. For example, the Rome II diagnostic criteria were used to assess the prevalence of IBS in the Taiwan study. Compared with earlier versions of the Rome diagnostic criteria, Rome II is more restrictive for pain-associated symptoms. In a Western population in Olmstead County, Minnesota, Saito et al. demonstrated that the age-adjusted prevalence of IBS using the Rome II criteria was higher in men than in women,29 which is in contrast with the observations made with the Rome (1990) and Rome I (1992) criteria where the female gender was predominant. These differences may be explained in part by the summarized findings of studies evaluating gender differences in the clinical symptoms of IBS:30–33 (i) there is a greater female predominance in the non-pain-associated symptoms of constipation, bloating and extra-intestinal manifestations; (ii) men and women report similar pain-related bowel symptoms and levels of abdominal pain; (iii) men report higher levels of diarrhoea; and (iv) women report higher levels of constipation.
Chronic constipation has been self-reported in 20.8% of women and 8.0% of men.34 A more recent study, which conducted nationwide telephone interview surveys of more than 10 000 individuals, reported the prevalence of constipation to be 16% in women and 12% in men.35 The female to male ratio was increased for both the ‘outlet’ type (1.65) and the combined ‘irritable bowel syndrome–outlet’ type (2.27) of functional constipation. In another study, women with functional constipation were found to be more likely to seek medical care than men (35.6% vs. 19.5%) at all ages, except for 50–64 years where the probability rates were similar.12
Health care costs
A number of studies have demonstrated increased health care costs of IBS patients compared with non-IBS patients, with indirect and direct annual costs estimated to total up to 30 billion dollars.16, 17 Three recently published studies, two performed in health maintenance organizations36, 37 and one in two different Medicaid populations,38 found that the total costs for health care were approximately 50% higher in patients with IBS than in non-IBS controls. These studies similarly found that the costs were increased particularly with regard to out-patient rather than in-patient services. Interestingly, Levy et al. found that the majority of excess total health care costs were due to non-lower gastrointestinal-related services.36 This finding is supported by the increased physician visits by IBS patients not only for gastrointestinal symptoms but also for non-IBS-related reasons.8 A recent systematic review of studies evaluating the prevalence of abdominal and pelvic surgeries in IBS patients supported the increased prevalence of cholecystectomy and hysterectomy in population and referral settings.39 In a health maintenance organization population of 4587 patients with physician-diagnosed IBS, Longstreth and Yao reported three-fold higher rates of cholecystectomy, two-fold higher rates of appendectomy and hysterectomy and 50% higher rates of back surgery compared with non-IBS controls.40
The literature supports the significant world-wide prevalence of FGIDs. Most studies have reported an increased female prevalence in patients with IBS and chronic constipation, but not functional dyspepsia. The female to male ratio appears to be greater in the health care-seeking population than in community populations. Although men and women with IBS report similar pain-related symptoms, there is a greater female predominance of the non-pain-associated symptoms of constipation, bloating and extra-intestinal manifestations. However, the general and gender-related prevalence of FGIDs may vary due to cultural factors and study methodology. A significant health care burden is associated with IBS, with increased out-patient services, abdominal and pelvic surgeries, and gastrointestinal- and non-gastrointestinal-related physician visits and health care costs.
Health-related quality of life (HRQOL)
General HRQOL in FGIDs
A number of studies have demonstrated reductions in HRQOL measures as well as general well-being in patients with FGIDs with moderate to severe symptoms relative to healthy controls, particularly those seen in referral settings. In IBS patients, HRQOL has been found to be lower than that in patients with GERD and asthma,41 and similar (and lower in some domains) to that in patients with other chronic illnesses, such as diabetes mellitus or end-stage renal disease.42 Based on these findings, a recent American College of Gastroenterology guideline suggests routine HRQOL screening in patients with IBS and recommends the initiation of treatment when the symptoms of IBS are found to reduce functional status and diminish overall HRQOL.43 In a systematic review evaluating studies that compared HRQOL in patients with functional dyspepsia,44 one of four comparisons demonstrated a decrease in HRQOL in patients compared with normative healthy population values,45 whilst the other three showed only trends for some of the domains of the generic HRQOL measure, SF-36.46, 47 Although these studies suggest that HRQOL is significantly reduced in patients with FGIDs in referral settings, there is a lack of studies performed in population settings. However, the limited evidence from population-based studies supports a decrease in HRQOL in patients with FGIDs.48–50
Changes in HRQOL with treatment
There is evidence to support improvements in HRQOL with treatment of FGIDs, which may correlate with an improvement in symptomatic response. Greater improvements in HRQOL were demonstrated in patients with functional dyspepsia who received pharmacological (e.g. the proton pump inhibitor, omeprazole)51 and non-pharmacological (e.g. hypnotherapy)52 treatment, which were found to provide more effective relief of symptoms than the respective control treatment arms. In another study, there were no significant symptom or HRQOL improvements with the prokinetic agent ABT-299; however, strong and significant correlations were observed between changes measured using a disease-specific HRQOL measure and visual analogue symptom scales.53 In a multicentre, 3-month, prospective observational study, gastrointestinal symptom (using the Gastrointestinal Symptom Rating Scale) and HRQOL (using SF-36) changes associated with the prokinetic agent, cisapride, given as single-drug therapy, were measured in 328 out-patients with functional dyspepsia.45 Both the Gastrointestinal Symptom Rating Scale and HRQOL scores improved with therapy, with significant correlations between the total symptom score and all HRQOL domains found at each visit. A major limitation of this study was the lack of a placebo arm.
The literature also suggests that HRQOL can be improved with effective treatments for IBS.54 In a multicentre, randomized, placebo-controlled trial evaluating the efficacy of a leuprolide acetate depot in relieving abdominal pain and nausea in premenopausal women with FGIDs, the general HRQOL measure showed improvement in the health domain only, but not in the other domains.55 Heymann-Monnikeset al. compared symptoms, visceral perception and HRQOL in 24 IBS patients who received multicomponent behavioural therapy plus standard medical treatment (SMBT) compared with patients who received standard medical treatment alone (SMT).56 SMBT included IBS information and education, progressive muscle relaxation, training in illness-related cognitive coping strategies, problem solving and assertiveness training in 10 sessions over 10 weeks. SMT included standardized symptom-orientated medical treatment and regular visits to a gastroenterologist every second week. Symptoms and HRQOL (measured by a gastrointestinal quality of life index) significantly improved in the SMBT group, but remained unchanged in the SMT group.
A few studies have also demonstrated significant improvements in HRQOL in IBS patients using a disease-specific IBS instrument. The 5-HT3 antagonist, alosetron, has been shown in several multicentre, randomized, placebo-controlled trials to be effective in the treatment of global symptoms in women with diarrhoea-predominant IBS compared with placebo.57–60 HRQOL was assessed as part of two randomized, double-blind, placebo-controlled IBS studies comparing alosetron at a dose of 1 mg b.d. with placebo.61 Patients completed a validated disease-specific quality of life questionnaire, the IBS Quality of Life Questionnaire (IBSQOL), at baseline and at the 12-week or final visit. Diarrhoea-predominant IBS patients treated with alosetron showed statistically significant improvements on IBSQOL scales compared with those given placebo; these improvements were observed on all nine IBSQOL scales, except for one (mental health) in one of the two clinical trials. In both studies, a significantly greater percentage of patients treated with alosetron experienced clinically meaningful improvement on three of the IBSQOL scales (food/diet, social functioning and role-physical) compared with patients treated with placebo. In another high-quality study, Drossman et al. demonstrated the efficacy of desipramine and cognitive behavioural therapy compared with control conditions in patients with moderate to severe FGIDs, most of whom had IBS.62 There were trends for HRQOL improvement with treatment using the disease-specific IBS-QOL instrument. In addition, there were significant correlations between the global IBS measure (composite score) and end-of-treatment HRQOL (r = 0.70, P < 0.0001) and the change in HRQOL (r = 0.24, P < 0.0001) (D. A. Drossman et al., personal communication, 2004).
Clinical predictors of HRQOL
In the clinical research setting, HRQOL has been measured using a standardized generic measurement tool, such as SF-36 or the Sickness Impact Profile, or using disease-specific questionnaires, such as IBS-QOL, IBSQOL and the Nepean Dyspepsia Index. However, in clinical practice, HRQOL is more challenging to measure if these standardized QOL questionnaires are not used. Recent studies have been conducted to determine the clinical predictors of HRQOL in order to understand the risk factors and disease variables which are important influences on outcome, and to help guide physicians to assess the impact of disease and to deliver more efficient health care.
Psychosocial stressors have been associated with exacerbations of symptoms in functional and inflammatory disorders of the gastrointestinal tract. Naliboff et al. performed a longitudinal study to assess whether chronic life stressors enhanced symptoms in 60 patients with chronic heartburn.63 The presence of a severe, sustained life stress over the previous 6 months significantly predicted increased heartburn symptoms over the following 4 months. Affective and subjective stress ratings were not strongly related to heartburn severity. However, anxiety was the strongest predictor of impaired HRQOL, and depression was the strongest predictor of heartburn medication use.
A recent UK study assessed whether childhood and current adversities were more prevalent and predicted HRQOL in clinic out-patients with functional and organic disorders with symptoms referable to the chest and upper abdomen.64 One hundred and thirty-three patients with functional (40 with non-cardiac chest pain and 43 with functional dyspepsia) and organic (29 with GERD and 21 with ischaemic heart disease) disorders were included in the study. Thirty per cent reported one or more forms of childhood adversity, with no difference between the organic and functional groups. In addition, 40% reported ongoing social stress, 27% had experienced a severely threatening life event in the previous year and 14% lacked a close confidant; however, there were no differences between the functional and organic groups. The distress score, which was measured by the total (anxiety plus depression) score based on the Hospital Anxiety and Depression scale, was the primary independent predictor for both the physical composite score and mental composite score of HRQOL for organic and functional disorders. However, distress and the lack of a close confidant were the two independent predictors of HRQOL in patients with organic disease, while distress, age and reported childhood adversity were the three predictors in patients with functional disorders. The negative effect of childhood adversity on HRQOL in functional disorders was found to be mediated mainly through distress, but also had some direct effect.
Although HRQOL has been shown to be impaired in patients with IBS, there are a lack of data identifying specific clinical factors that predict HRQOL in this condition. Spiegel et al. recently evaluated the determinants of both mental and physical HRQOL in 770 consecutive Rome-positive IBS patients recruited from both advertisement and a tertiary referral clinic at a university-based centre.65 Seven additive factors independently predicted the physical composite score of HRQOL (in order of decreasing significance): (i) more than five physician visits per year; (ii) tiring easily; (iii) low in energy; (iv) severe symptoms; (v) predominantly painful symptoms; (vi) feeling that there is ‘something seriously wrong with body’; and (vii) symptom flares > 24 h. Eight additive factors independently predicted the mental composite score of HRQOL: (i) feeling tense; (ii) feeling nervous; (iii) feeling hopeless; (iv) difficulty sleeping; (v) tiring easily; (vi) low sexual interest; (vii) IBS symptom interference with sexual function; and (viii) low energy. Whereas mental HRQOL was associated with abnormalities in sexuality, mood and anxiety, physical HRQOL was associated with symptom severity, symptom periodicity and pain. Both physical and mental HRQOL shared an association with symptoms of chronic stress (exhibited by symptoms of vital exhaustion),66 and neither was determined by traditionally elicited gastrointestinal symptoms, including stool frequency, stool characteristics or IBS bowel habit subtype.
HRQOL is significantly impacted in patients with FGIDs, such as functional dyspepsia and IBS, compared with the general healthy population, as well as patients with other chronic conditions, such as GERD and asthma. The impaired HRQOL has been particularly demonstrated in patients with moderate to severe disease seen in referral settings. Less information is available about non-consulters and community populations with FGIDs. With regard to treatment response, HRQOL appears to improve in responders, or correlates with symptom improvement, with at least some treatment modalities studied in FGIDs. However, more studies are needed to further elucidate this association. In addition, studies have demonstrated the importance and predictive value of psychosocial factors and symptoms related to visceral perception and chronic stress in HRQOL in patients with organic and functional gastrointestinal disorders. They further support the conceptual model that early adverse life events influence later psychosocial experiences, physiological functioning and susceptibility to the development of IBS and other FGIDs.67 The combined effects of altered gut function and psychosocial factors integrated within brain–gut interactions affect symptom perception, illness behaviour, outcome and HRQOL.
A summary of the epidemiological and HRQOL findings related to FGIDs is shown in Table 1. Based on the literature, a conceptual model related to HRQOL can be proposed (see Figure 1). The important disease variables predicting HRQOL in FGIDs appear to be symptom severity, particularly pain, and psychological distress. These two main disease variables can also influence each other. Risk factors which affect these disease variables may include gastrointestinal-related factors, such as visceral hypersensitivity and altered gastrointestinal motility, which primarily affect gastrointestinal symptom severity; however, other risk factors include psychosocial factors, such as childhood adversity, chronic life stressors, psychological conditions and lack of social support, which primarily affect psychological distress. Extra-intestinal symptoms, such as fatigue, are associated with both disease variables of psychological distress and gastrointestinal symptom severity. Although gastrointestinal-related symptoms are obviously important, non-gastrointestinal symptoms appear to have a major, if not greater, effect on health care visits, health care costs and HRQOL in IBS.
|Prevalence of FGIDs: IBS, 5–25%; chronic constipation, 1.2–27%; dyspepsia, 25% (functional dyspepsia is most common)|
|FGIDs account for 41% of diagnoses in GI speciality practices|
|IBS comprises 12% of the diagnoses in primary care and 28% of the diagnoses in GI practices|
|Female predominance in IBS (particularly constipation subgroup) and chronic constipation, but not functional dyspepsia|
|Similar reports of pain in IBS, but greater reporting of non-pain-related symptoms in women|
|Health care burden in IBS|
|Increased GI- and non-GI-related physician visits|
|Increased abdominal and pelvic surgeries|
|Increased health care costs, particularly out-patient services and excess costs for non-GI-related services|
|Decreased HRQOL in IBS, functional dyspepsia and chronic constipation|
|Improvement in HRQOL can be associated with positive symptom response to treatment|
|Predictors of HRQOL include early adverse life events, psychological distress, stress-related symptoms and pain|
Further studies are needed to address several issues of HRQOL in patients with FGIDs: (i) assessment in non-consulters vs. consulters; (ii) assessment in a primary care population vs. speciality patient population; (iii) further analysis of risk factors and disease variables to determine the primary mediating factor predicting HRQOL (e.g. does gastrointestinal symptom severity primarily affect HRQOL through distress?); (iv) the importance and predictive value of non-gastrointestinal symptoms on health care seeking and costs and HRQOL; and (v) the determination of whether the predictive factors of HRQOL predict other clinically significant outcomes, including resource utilization, satisfaction with care and specific treatment responses.