Correspondence to: Dr H. B. El-Serag, The Houston Veterans Affairs Medical Center (152), 2002 Holcombe Blvd., Houston, TX 77030, USA. E-mail: email@example.com
Health-related quality of life among persons with irritable bowel syndrome: a systematic review
Article first published online: 27 MAY 2002
Alimentary Pharmacology & Therapeutics
Volume 16, Issue 6, pages 1171–1185, June 2002
How to Cite
El-Serag, H. B., Olden, K. and Bjorkman, D. (2002), Health-related quality of life among persons with irritable bowel syndrome: a systematic review. Alimentary Pharmacology & Therapeutics, 16: 1171–1185. doi: 10.1046/j.1365-2036.2002.01290.x
- Issue published online: 27 MAY 2002
- Article first published online: 27 MAY 2002
- Accepted for publication 9 March 2002
- Top of page
- Materials and methods
- Summary and discussion
Aim : To perform a systematic review of the literature with three objectives: (1) to compare the health related quality of life (HRQoL) of patients with irritable bowel syndrome with that of healthy controls; (2) to compare the HRQoL of irritable bowel syndrome patients to those with other diseases; and (3) to examine therapy-associated changes in HRQoL of irritable bowel syndrome patients.
Methods : Searches of all English and non-English articles from 1980 to 2001 were performed in Medline and Embase, and two investigators performed independent data abstraction.
Results : Seventeen articles met our selection criteria. 13 studies addressed objective no. 1; 11 showed a significant reduction in HRQoL among irritable bowel syndrome patients. Of these, only one study was considered of high quality. Four studies addressed objective no. 2, none of which was considered to be high quality in addressing this objective. Four trials (three of high quality) addressed objective no. 3. One showed that symptomatic improvement with Leupron compared to placebo was accompanied an improvement only in the comparative health domain of the HRQoL. The second study reported significant positive changes in HRQoL after 12 weeks of cognitive behavioural therapy. The third report of two placebo-controlled studies indicated significant improvement with alosetron on most domains of Irritable Bowel Syndrome Quality of Life Questionnaire.
Conclusions : (i) There is reasonable evidence for a decrease in HRQoL in patients with moderate to severe irritable bowel syndrome; however, the data are conflicting regarding the impact of irritable bowel syndrome on HRQoL in population-based studies of nonconsulters. (ii) HRQoL in irritable bowel syndrome patients is impaired to a degree comparable to other chronic disorders such as GERD and depression. (iii) A therapeutic response in irritable bowel syndrome-related pain has a corresponding improvement in HRQoL. (iv) Limitations of the literature include focusing on moderate-severe irritable bowel syndrome in referral centres, and lack of appropriate controls
- Top of page
- Materials and methods
- Summary and discussion
Irritable bowel syndrome is a disorder best described by the presence of chronic or recurrent symptoms of abdominal pain, diarrhoea, constipation and/or abdominal distension.1 It is a complex disorder involving the brain-gut axis, the aetiology of which is incompletely defined. irritable bowel syndrome is generally a non-life threatening disorder that is associated with neither the need for surgery nor a reduction in survival. These facts might have led to a potential underestimation of the impact of irritable bowel syndrome as a health problem. Health Related Quality of Life (HRQoL) is becoming recognized as an important outcome for patients with chronic diseases.2 The measurement of HRQoL seeks to encompass the emotional and social dimensions of the patient's illness, in addition to that of physical function.3 Irritable bowel syndrome has several features that make HRQoL measurement of particular importance in studying this disorder. Irritable bowel syndrome affects relatively younger people and is associated with psychological distress, sexual dysfunction and disturbance of work and sleep.4 Measuring only symptoms, such as pain ratings or number of bowel movements alone, may present an incomplete and inadequate measure of a patient's overall well being, burden of illness and response to therapeutic interventions.5
Three types of HRQoL measurements are available: global assessment, generic or general instruments, and disease-specific instruments.6, 7 Global assessment in the form of a graded summary (good, fair, poor), or using a visual analogue scale, is often inadequate for complex hypothesis testing. Generic instruments use generalized questions that are not specific to any particular disease; which allows one to compare HRQoL across a broad spectrum of diseases. Generic HRQoL instruments, however, may not be responsive to a clinically important change for the particular disease. Prior to the 1990s, symptom reports and `global assessments' were the major outcome measures for epidemiological as well as clinical trials of irritable bowel syndrome. Subsequently, both generic as well as disease-specific instruments were developed to measure HRQoL in patients with irritable bowel syndrome.8 Examples of commonly used generic questionnaires in irritable bowel syndrome are the Medical Outcomes Study Short Form 36 (SF-36),9 the Psychological General Well Being Index (PGWBI)10 and the Sickness Impact Profile (SIP).11 For the SF-36, there are published normative data for the general population of the USA and several other countries, facilitating comparisons between subjects with disease and healthy controls.9 These results are reported in specific domains. For example, the SF-36, reports quality of life in the following domains: physical functioning (PF), role physical (RP), bodily pain (BP), general health (GH), vitality (V), social functioning (SF), role emotional (RE), and mental health (MH).
Lastly, disease-specific questionnaires are designed to detect HRQoL changes that may not be targeted in generic instruments, but are important manifestations of a certain disorder.3, 6 Disease-specific instruments may not be applied widely to other diseases because these symptoms may not be present in other disorders. There are at least four disease-specific instruments. These are the Irritable Bowel Syndrome-Quality of Life measure (irritable bowel syndrome-QoL), the Irritable Bowel Syndrome Quality of Life Questionnaire (IBSQoL), the Irritable Bowel Syndrome Questionnaire (IBSQ), and the Functional Digestive Disorders Quality of Life Questionnaire (FDDQoL). The characteristics and validation of these instruments were reviewed elsewhere.12, 13
We have performed this systematic review of the literature with three primary and two secondary objectives. The three primary objectives are: (i) to compare the health related quality of life of irritable bowel syndrome patients with that of healthy controls; (ii) to compare the health related quality of life of irritable bowel syndrome patients to patients with that of other diseases; and (iii) to examine therapy-associated changes in health related quality of life of irritable bowel syndrome patients. The two secondary objectives are: (iv) to compare HRQoL between patients with constipation dominant-irritable bowel syndrome and patients with diarrhoea-dominant irritable bowel syndrome; and (v) to examine the correlation between HRQoL and severity of bowel symptoms among patients with irritable bowel syndrome.
Materials and methods
- Top of page
- Materials and methods
- Summary and discussion
In order to identify the relevant articles to be included in this systematic review, the following literature search was employed. Separate searches of all English and non-English articles from 1980 to 2001 were performed in Medline, Current Contents, Embase, and the Cochrane library with different combinations of the following search terms: ‘irritable bowel syndrome’ and ‘quality of life’. Alternatively ‘health status’ and ‘health related quality of life’ were each paired with ‘irritable bowel syndrome’. A manual search for other relevant articles was performed using the annotated references of selected review articles and book chapters on irritable bowel syndrome.
Study selection criteria
Due to the need to subject the methodologies and results to careful review, only full-length published manuscripts were eligible for inclusion. One investigator reviewed for relevance the titles and abstracts of all the citations identified by the literature search. The full papers for potentially relevant abstracts were obtained. The following inclusion criteria had to be fulfilled: (a) patients specifically described as having irritable bowel syndrome diagnosis, but not functional dyspepsia only; (b) HRQoL measured using a validated generic or irritable bowel syndrome-specific instruments, but not global assessment alone. For studies that examine objective no. 3, which assesses the effect of therapies on health related quality of life of irritable bowel syndrome patients, an additional selection criteria was: (c) randomized trials only. Studies that dealt with development of HRQoL instruments were excluded.
A standardized comprehensive data abstraction form was designed for the purpose of this review. Two investigator (H.E.S. and D.B.) performed independent data abstractions. Several articles displayed HRQoL scores in a graphic form only. The corresponding authors listed in these articles were contacted to obtain the numerical values for HRQoL. The studies were then analysed to see if they addressed any of the primary or secondary objectives listed above. In some instances, the studies might have addressed more than one objective. Studies that did not answer any of the pre-specified objectives were excluded.
Assessment of study quality
All potential articles were then rated on ‘study quality assessment scoring sheet’. For studies that addressed our first and second objectives, we have compiled the following quality criteria: (i) use of age and sex matched controls or adjustment for these variables (2 = yes, 0 = no); and (ii) disease definition of irritable bowel syndrome (2 = Rome and/or Manning criteria, 1 = modification of the original criteria, 0 = not defined), (iii) severity of irritable bowel syndrome bowel symptoms assessed and either qualified or quantified (1 = yes, 0 = no); and (iv) psychological assessment performed and reported (1 = yes, 0 = no). A study with a score of 4 or higher was considered to be of high quality.
For the third objective, we decided a priori on five criteria to assess the quality of each trial based on previously published guidelines for evaluating the quality of randomized controlled trials14 and for therapeutic trials in irritable bowel syndrome.15, 16 These criteria were: (a) randomization, (b) double-blinding, (c) completeness of follow-up, (d) baseline similarity of treatment and control groups in parallel trials, and (e) the use of an HRQoL instrument with validated responsiveness. Scores of 1–2 were given for their randomization (2 = appropriate methods of randomization detailed, 1 = study described as ‘randomized’) 0–2 for blinding (2 = subjects and investigators blinded by appropriate method including the use of identical placebos or ‘double-dummy’ technique, 1 = study described as ‘double-blind’, 0 = not double-blind), 0–1 for the frequency of withdrawals (1 = the number and reason for withdrawals were stated, 0 = no statement on withdrawals), 0–1 for similarity of baseline severity of symptoms (1 = similar, 0 = not similar), and 0–1 for the use of HRQoL instrument with validated responsiveness (1 = previously validated, 0 = not previously validated). The total score ranged from 1 to 6. A score of 0 was not included in the randomization category because our inclusion criteria required randomized trials. Studies with a score of 4 or higher were considered to be high quality studies.
A standard meta-analytic approach to aggregate the reported data was precluded because of substantial variations between the studies in the identification of severity and duration of irritable bowel syndrome, the identification of co-morbid psychological illness and the measurement method for the HRQoL. The data were presented in tabular and graphical form.
For studies that used the SF-36 on irritable bowel syndrome patients but did not report the comparison to general population controls, we calculated a two-sided t-test to examine for the differences between the mean score for irritable bowel syndrome patients and those previously reported in the Medical Outcomes Study.9
- Top of page
- Materials and methods
- Summary and discussion
The initial Medline search (‘quality of life’ and ‘irritable bowel syndrome’) revealed 73 potentially relevant English and non-English citations. A review of the titles and abstracts yielded 32 potentially relevant articles that were reviewed in full. One additional potentially relevant article was identified and reviewed in full in a subsequent Medline search of the term ‘Health Status’, which yielded a total of 32 citations. No additional potentially relevant articles were identified in the third search of the term ‘health-related quality of life’, which yielded a total 20 articles. Three additional potentially relevant citations were identified in Embase, which yielded a total of 65 citations. No additional potentially relevant articles were identified in Current Contents, or Cochrane library. In total, a review of the titles and abstracts yielded 36 relevant articles that were reviewed in full; of these, we excluded 7 validation studies, 3 with no definition for irritable bowel syndrome, 6 nonoriginal studies, and 3 studies that did not use an HRQoL instrument. Seventeen articles met our selection criteria, and were included in the data abstraction step.17–33
Selected characteristics of the 17 articles are presented in Tables 1–4. In 11 studies, irritable bowel syndrome was defined according to either Manning and/or Rome criteria; in three studies, operational criteria that used a modification of the original criteria was used; and in two studies, no definition was given. An adequate work-up to rule out organic disease (e.g. haemoglobin/haematocrit, stool examination for blood, sigmoidoscopy) was reported in only three of 17 studies.
|Ref., country||Year published||Study design||Patients with IBS (n)||Controls without IBS (n)||QoL instrument||QoL score in IBS patients mean (SD)||QoL score in controls, mean (SD)||P-value||Severity of IBS assessment|
|17 USA||1996||Cross-sectional, population- based||165 IBS non-consulters||122||SF-36||PF 92 (12) RP 83 (29) BP 56 (14) GH 71 (10) V 52 (20) SF 84 (17) RE 69 (39) MH 66 (17)||PF 97 (8) RP 93 (23) BP 66 (14) GH 86 (15) V 66 (15) SF 93 (12) RE 87 (27) MH 75 (14)||All < 0.05||No|
|17 USA||1996||Cross-sectional, population- based||83 IBS consulters||122||SF-36||PF 91 (12) RP 68 (37) BP 52 (14) GH 62 (20) V 46 (12) SF 78 (19) RE 59 (39) MH 64 (18)||PF 97 (8) RP 93 (23) BP 66 (14) GH 86 (15) V 66 (15) SF 93 (12) RE 87 (27) MH 75 (14)||All < 0.05||No|
|18 USA||1995||Cross-sectional, population- based||73 (>65 years of age)||173 (>65 years of age)||SF-36||PF 52 (IQR* 18–83) RF 54 (0–100) SF 78 (60–100) PH 49 (29–72) BP 51 (25–75) MH 74 (64–88)||PF 70 (50–100) RF 77 (50–100) SF 91 (50–100) PH 74 (57–92) BP 78 (50–100) MH 84 (76–92)||> 0.05 for each domain in regression analyses||No|
|19 Ireland||1991||Cross-sectional, referral||20||42||SEIQoL||68.6 (12.7)||77.4 (9.5)||< 0.001||No|
|Ref., country||Year published||Study design||Patients with IBS (n)||Controls without IBS (n)||QoL instrument||QoL score in IBS patients, mean (SD)||QoL score in historically healthy controls, mean (SD)||P-value||Severity of IBS|
|20 UK||2001||Cross-sectional,||257||0||SF-36||PF 69 (28)||PF 93||All||Moderate to|
|referral||RP 37 (39) BP 39 (20) GH 42 (20) V 38 (23) SF 50 (26) RE 52 (43) MH 56 (21) Physical score 41(13) Mental score 38(11)||RP 91 BP 86 GH 79 V 64 SF 91 RE 86 MH 75||< 0.0001 < 0.05||severe|
|21 USA||2000||Cross sectional,||156||0||SIP||Total 6.5 (SE 6.7)||1.4||< 0.0001||Moderate to|
|22 USA||2000||Cross-sectional,||877||2474||SF-36||PF 79 (22)||PF 84 (9)||All < 0.001||NA|
|referral||(general population, historical controls from MOS)||RP 50 (42) BP 54 (25) GH 55 (23) V 44 (23) SF 63 (28) RE 65 (41) MH 65 (20)||RP 81 (9) BP 74 (5) GH 72 (6) V 61 (3) SF 83 (3) RE 81 (3) MH 74 (2)|
|23 USA||2000||Cross sectional,||144||0||SIP||5.5 (SE 0.5)||1.4||< 0.0001||Moderate|
|23 USA||2000||Cross sectional,||67||0||SIP||9.3 (SE 0.8)||1.4||< 0.0001||Severe|
|24 USA||1999||Cross-sectional,||342 IBS||0||SF-36||RF 74||PF 84 (9)||NA||Severe|
|referral||clinic patients||RP 44 RE 62 V 43||RP 81 (9) RE 81 (3) V 61 (3)|
|24 USA||1999||Cross-sectional,||315 IBS||0||SF-36||PF 81||PF 84 (9)||NA||Moderate|
|referral||advertisement||RP 58 RE 64 V 48||RP 81 (9) RE 81 (3) V 61 (3)|
|25 USA||1999||Cross-sectional,||140 IBS-C||0||SF-36||PF 78 (2)||PF 84 (9)||All||Moderate to|
|referral||RP 52 (4) BP 49 (2) GH 57 (2) V 44 (2) SF 64 (3) RE 60 (4) MH 63 (2)||RP 81 (9) BP 74 (5) GH 72 (6) V 61 (3) SF 83 (3) RE 81 (3) MH 74 (2)||< 0.0001||severe|
|25 USA||1999||Cross-sectional, referral||216 IBS-D||0||SF-36||PF 82 (1) RP 55 (3) BP 54 (2) GH 59 (2) V 48 (2) SF 63 (2) RE 69 (3) MH 67 (1)||PF 84 (9) RP 81 (9) BP 74 (5) GH 72 (6) V 61 (3) SF 83 (3) RE 81 (3) MH 74 (2)||All < 0.0001||Moderate to severe|
|26 USA||1999||Cross-sectional,||287||0||SF-36||PF 76||PF 84 (9)||< 0.05||42%|
|IBS support||RP 45||RP 81 (9)||Moderate to|
|group||BP 54 GH 53 V 46 SF 57 RE 57 MH 59||BP 74 (5) GH 72 (6) V 61 (3) SF 83 (3) RE 81 (3) MH 74 (2)||severe, rest mild|
|26 UK||1999||Cross-sectional,||342||0||SF-36||PF 67||PF 93||< 0.05||52%|
|members of||RP 37||RP 91||Moderate to|
|foundations of||BP 48||BP 86||severe, rest|
|functional GI||GH 45||GH 79||mild|
|disorders||V 39 SF 54 RE 48 MH 55||V 64 SF 91 RE 86 MH 75|
|27 USA||1998||Cross-sectional,||156||0||SF-36||PF 82 (22)||PF 93||All||Moderate to|
|referral||RP 51 (41) BP 49 (19) GH 57 (24) V 38 (20) SF 62 (22) RE 50 (42) MH 57 (19) Physical score 56 (19) Mental score 52 (19)||RP 91 BP 86 GH 79 V 64 SF 91 RE 86 MH 75||< 0.0001 except PF > 0.05||severe|
|28 USA||1997||Cross-sectional,||7||0||SF-36||RP 75||RP 91||> 0.05||Mild|
|referral||BP 66 GH 79 V 63 SF 83 RE 85 MH 77||BP 86 GH 79 V 64 SF 91 RE 86 MH 75|
|28 USA||1997||Cross-sectional,||28||0||SF-36||RP 58||All < 0.05||Moderate|
|referral||BP 60 GH 64 V 51 SF 73 RE 60 MH 64||except GH >0.05|
|28 USA||1997||Cross-sectional,||51||0||SF-36||RP 42||All < 0.05||Severe|
|referral||BP 50 GH 53 V 37 SF 55 RE 58 MH 56|
|28 USA||1997||Cross-sectional,||25||0||SF-36||RP 24||All < 0.05||Very severe|
|referral||BP 33 GH 53 V 40 SF 32 RE 39 MH 53|
|Ref., country||Year published||Study design||Patients with IBS (n)||Controls with other disorders (n)||QoL instrument||QoL score in IBS patients, mean (SD)||QoL score in controls, mean (SD)||P-value||Severity of IBS|
|22 USA||2000||Cross-sectional,||877||165 ESRD*||SF-36||PF 79 (22)||PF 52 (30)||< 0.001||NA|
|referral||RP 50 (42) BP 54 (25) GH 55 (23) V 44 (23) SF 6 (28) RE 65 (41) MH 65 (20)||RP 33 (40) BP 59 (30) GH 44 (25) V 46 (24) SF 64 (30) RE 60 (44) MH 70 (20)||0.002 NS < 0.001 0.002 NS NS NS|
|22 USA||2000||Cross-sectional,||877||516 GERD*||SF-36||PF 79 (22)||PF 80 (23)||< 0.001 all||NA|
|referral||RP 50 (42) BP 54 (25) GH 55 (23) V 44 (23) SF 63 (28) RE 65 (41) MH 65 (20)||RP 72 (38) BP 58 (21) GH 68 (21) V 57 (20) SF 79 (23) RE 78 (35) MH 71 (19)||except PF|
|22 USA||2000||Cross-sectional,||877||541||SF-36||PF 79 (22)||PF 68 (29)||< 0.001||NA|
|referral||Diabetes mellitus*||RP 50 (42) BP 54 (25) GH 55 (23) V 44 (23) SF 63 (28) RE 65 (41) MH 65 (20)||RP 57 (42) BP 69 (27) GH 56 (21) V 56 (22) SF 82 (25) RE 76 (37) MH 77 (18)||0.002 < 0.001 NS < 0.001 < 0.001 < 0.001 < 0.001|
|22 USA||2000||Cross-sectional,||877||502||SF-36||PF 79 (22)||PF 72 (7)||< 0.001||NA|
|referral||Depression*||RP 50 (42) BP 54 (25) GH 55 (23) V 44 (23) SF 63 (28) RE 65 (41) MH 65 (20)||RP 44 (40) BP 59 (27) GH 53 (23) V 40 (21) SF 57 (28) RE 39 (40) MH 46 (21)||NS < 0.001 NS < 0.001 < 0.001 0.002 < 0.001|
|29 USA||1996||Cross-sectional,||43||21 UC||SIP||12.2||5.0 UC||NA||NA|
|referral||46 CD 29 PUD 29 Liver dis||(overall score)||7.4 CD 8.3 PUD 8.3 Liver dis|
|19 Ireland||1991||Cross-sectional,||20||20 PUD||SEIQoL||68.6 (12.7)||70.4 (12.4)||> 0.05||NA|
|Ref., country||Patients (n)||QoL instrument||Intervention||Pre-treatment score, mean (SD)||Post-treatment score, mean (SD)||P-value|
|21 USA||156||SIP||Cognitive therapy vs. Despiramine or placebo||Total 6.5 (6.7) Physical 3.0 (4.4) Psychosocial 8.8 (10.8) Eating 2.8 (4.4)||Total 5.0 (7.1) Physical 2.6 (5.5) Psychosocial 6.0 (9.9) Eating 2.8 (4.7)||0.0001 0.0346 0.0001 0.9969|
|21 USA||156||IBSQoL||Cognitive therapy vs. Despiramine or placebo||65.8 (19.9)||76.0 (18.8)||0.0001|
|31 USA||626||IBSQoL||1 mg alesterone p.o. b.d.s. vs. placebo for 12 weeks||*||*||All < 0.05 except MH > 0.05|
|31 USA||647||IBSQoL||1 mg alesterone p.o. b.d.s. vs. placebo for 12 weeks||*||*||< 0.05|
|30 USA||33, 32||SF-36||Leuprolide acetate 7.5 mg, 3.75 mg q 4 weeks for 16 weeks||NA||NA||0.004|
|30 USA||35||SF-36||Placebo for 16 weeks||NA||NA||0.019|
|32 Germany||12||GIQLI||Behavioural therapy + standard medical treatment for 14 weeks||91.3 (17.8)||104.8 (17.6)||0.001|
|32 Germany||12||GIQLI||Standard medical treatment for 14 weeks||85.3 (15.5)||83.1 (13.9)||< 0.05|
Summary results for objective no. 1
‘To compare the health related quality of life of irritable bowel syndrome patients to healthy controls.’
There were in total 12 published studies that had at least one measurement addressing the objective outlined above.17–28 Of these, three studies used a generic HRQoL instrument to compare subjects with irritable bowel syndrome to a sample of healthy controls without irritable bowel syndrome (Table 1).17–19 The other nine studies had no controls but used the SF-36, which has normative values for the general healthy population (Table 2).20–28 In six of the latter nine studies, the investigators presented a statistical comparison between the SF-36 scores of irritable bowel syndrome patients and those of historical general population control subjects reported in the Medical Outcomes Study.20–24, 28 In two other studies, we performed these calculations using data from the article or from correspondence with the authors.25, 27 In one study a statistical comparison could not be made on basis of the available data.24 Most studies (10 out of 12) were carried out among patients with irritable bowel syndrome who presented to a referral setting, while only two were population-based.17, 18
Eleven studies were ‘positive’ in showing a significant reduction in HRQoL among patients with irritable bowel syndrome compared to healthy controls.17, 19–28 Of the positive studies, only one was considered to be high quality; this was a population based study among university students (mean age 23 years) in which SF-36 scores were compared among three groups: irritable bowel syndrome patients, persons with irritable bowel syndrome identified on a symptom questionnaire but who had never sought medial attention, and a group of asymptomatic students.17 Irritable bowel syndrome patients had worse HRQoL scores than irritable bowel syndrome nonpatients, and both groups were significantly worse than asymptomatic controls. Two positive studies reported a summary mental score as well as a physical score; both scores were equally low.20, 27 In general, the score on physical domains were similar to those on mental domains. Two studies were ‘negative’ and did not show a significant reduction in HRQoL in nonconsulters and patients with mild irritable bowel syndrome.18, 28 The first negative study was a high quality population-based mailed survey of elderly persons (> 65 years old) in which respondents were identified as having irritable bowel syndrome on the basis of responses to a symptom questionnaire.18 In this study, there were no statistically significant differences in any of the SF-36 domains between persons with irritable bowel syndrome and asymptomatic persons. The investigators, however, reported that the overall score from all domains was significantly reduced in irritable bowel syndrome patients compared to controls. Such a reporting approach is not standard. In the second negative study, a total of only seven patients with mild irritable bowel syndrome were reported to have had SF-36 scores that were not different from those in the general population.28
In summary, there is good evidence to support the proposition that HRQoL is decreased among irritable bowel syndrome patients having moderate to severe symptoms who are seen in a referral setting. There is conflicting evidence from population-based studies regarding the impact of irritable bowel syndrome on the HRQoL of nonconsulters.
Summary results for objective no. 2
‘Is HRQoL reduced among patients with irritable bowel syndrome compared to patients with other medical disorders?’
There were four studies that had at least one instance of HRQoL comparison between irritable bowel syndrome and other diseases.17, 19, 22, 23 None of these were considered to be high quality in addressing this objective. Two studies compared the SF-36 scores of persons with irritable bowel syndrome seen at referral settings with previously published scores for patients with other disorders who were seen at different practice settings.17, 22 The two other studies carried out a direct comparison between irritable bowel syndrome patients and controls examined at the same institution;19, 23 however, the number of patients included in each group was too small to have adequate statistical power to avoid a type II error. In addition, irritable bowel syndrome symptom severity was not reported in any of these studies. Details were available for three studies (Table 3). The fourth study,17 which compared SF-36 between irritable bowel syndrome patients and those previously published for patients with CHF, did not contain enough information to be presented in Table 2. In latter study, the investigators surmised that irritable bowel syndrome patients showed as great or greater impairment as patients with CHF on all scales of SF-36, except for physical functioning, role physical and general health.
The largest study to address this objective was a recent cross-sectional survey in which SF-36 was administered to 877 patients with irritable bowel syndrome who were seen at UCLA, either as referrals or who responded to an advertisement.22 The HRQoL of patients with irritable bowel syndrome was compared in SF-36 scores to those previously published for persons with gastro-oesophageal reflux disease (GERD), type 2 diabetes, depression, and end stage renal disease (ESRD). None of the patients with these other conditions was seen at the same institute. Patients with irritable bowel syndrome had lower scores (i.e. a worse HRQoL) in most domains of SF-36 compared to patients with moderate to severe GERD symptoms. Patients with irritable bowel syndrome also had lower scores than patients with diabetes on the role physical and general health domains, but not the rest of the domains, where there was a trend toward lower HRQoL in patients with diabetes. Similarly, patients with irritable bowel syndrome had lower vitality scores than those with ESRD; however, the scores from the other domains were either similar to or worse in those patients with ESRD. Lastly, patients with depression had significantly lower scores on all SF-36 domains compared to patients with irritable bowel syndrome.
In the study by Drossman et al.23 the overall score of the Sickness Impact Profile (SIP), a generic instrument, was used to compare HRQoL among several small groups of patients with different digestive and hepatic disorders. Patients with irritable bowel syndrome had a worse HRQoL (higher score) on the overall score of SIP compared with patients with peptic ulcer disease, ulcerative colitis, Crohn's disease, liver disease, or hepatic or biliary disorders. The test of significance was not given in the paper and the available data were not sufficient to calculate one.
In summary, it appears that patients with irritable bowel syndrome have a quality of life that is impaired to a degree which is comparable to other disorders. In some cases, HRQoL is worse in patients with irritable bowel syndrome, and in some cases worse in the comparator disease. The precise relationship depends upon the disease compared, and the domain of interest.
Summary results for objective no. 3
‘To examine therapy-associated changes in health related quality of life for irritable bowel syndrome patients’
There were four published therapeutic trials of irritable bowel syndrome that met our inclusion criteria (Table 4).29–32 Three of these were of high quality. HRQoL was measured as a secondary outcome of therapy in all of them. The measured outcomes in all studies were relatively short-term (12–16 weeks). Two studies had ‘cognitive’ therapy as part the treatment arms;21, 32 however, the description of such therapy was poor.
The first high quality study was a multicentre randomized double-blind placebo controlled trial of leuprolide acetate depot (Leupron) in the treatment of abdominal pain and nausea in pre-menopausal women with functional bowel disease.30 Patients received either Leupron Depot via the intramuscular route at a dose of 7.5 mg (33 patients) or 3.75 mg (32 patients) or placebo (35 patients) every 4 weeks for 16 weeks. Follow-up data were available for 90% of the patients. Only the group receiving Leupron Depot 7.5 mg showed significant improvements for abdominal pain and nausea compared to placebo (P < 0.001). Concomitantly, there was only an improvement in the comparative health domain of the HRQoL questionnaire (P = 0.004 vs. placebo). No significant changes were detected on the patient's perception of general health, physical and emotional role functioning, bodily pain or overall mental health.
The second high quality study conducted by Drossman et al. reported changes in HRQoL measured by SIP and irritable bowel syndrome-QoL after 12 weeks of treatment with cognitive behavioural therapy vs. desipramine or placebo.21 Responders, as defined by an improvement in pain scores, daily function and days in bed, had a statistically significant improvement in HRQoL scores. Partial responders had a smaller improvement in HRQoL scores.
The third high quality publication reported the results of two large multicentre randomized, placebo controlled, double-blind studies comparing alosetrone 1 mg b.d. to treatment with placebo.31 A total of 626 and 647 irritable bowel syndrome patients were enrolled in these two studies. Patients completed a validated disease-specific HRQoL questionnaire (IBSQoL) at baseline and at the end of a 12-week treatment period. Approximately 70% of the patients in each study had diarrhoea-predominant irritable bowel syndrome. In patients with diarrhoea-predominant irritable bowel syndrome, there was a statistically significant improvement with alosetron vs. placebo on all nine domains of IBSQoL in one study, and in all domains but mental health in the second study. The greatest improvement was in the domains of role-physical, social functioning and food/diet. A greater proportion of patients treated with alosetrone experienced a perceived meaningful difference on role-physical, food/diet and social functioning scales in each of the studies.
In the fourth study, 24 patients with irritable bowel syndrome were randomized to receive either ‘standard medical therapy’ alone or combined with behavioural therapy.32 Standard medical therapy included antispasmodic, analgesics, low dose antidepressants, antidiarrhoeals, and regular visits to the gastroenterologist every other week. Neither the patients nor the investigators were blinded to the assigned treatments and there was no placebo treated control arm. There was a greater irritable bowel syndrome bowel symptom reduction in the group that received behavioural therapy combined with standard medical treatment (P < 0.01). In addition, the HRQoL on GLQLI improved significantly in the combination group whereas it remained unchanged in the standard medical therapy group.
The results of the review suggest that patients who have a therapeutic response to therapy for irritable bowel syndrome have a corresponding improvement in HRQoL.
Summary results for objective no. 4
‘To compare HRQoL between patients with constipation dominant-irritable bowel syndrome and patients with diarrhoea-dominant irritable bowel syndrome’
Irritable bowel syndrome is classified according to the pattern of bowel symptoms, as constipation-dominant, diarrhoea-dominant, or an alternating pattern. In the present search, three studies compared HRQoL measurement in constipation-dominant irritable bowel syndrome patients with those with diarrhoea-dominant constipation or alternating-irritable bowel syndrome. All the studies utilized validated generic questionnaires. A study from Sweden used PGWBI to compare HRQoL among three groups of irritable bowel syndrome patients: 119 with diarrhoea-dominant irritable bowel syndrome, 93 with constipation-dominant irritable bowel syndrome and 131 patients with alternating-pattern irritable bowel syndrome.33 There were no significant differences among these groups either on the total score of PGWB or on the individual domains of anxiety, depressed mood, positive well-being, self-control, general health, or vitality. In the second study from the USA, there were no differences in the SF-36 scores between 140 patients with constipation-dominant diarrhoea and 216 patients with diarrhoea-dominant irritable bowel syndrome.25 Similar results were reported in third study from the UK, in which all patients had severe irritable bowel syndrome.20
It therefore appears that the expression of irritable bowel syndrome as either diarrhoea-predominant, constipation-predominant, or of alternating symptoms does not affect the degree of impact of the disease on HRQoL.
Summary results for objective no. 5
‘To examine the correlation between HRQoL and severity of bowel symptoms among patients with irritable bowel syndrome’
The severity of irritable bowel syndrome varied within individual studies and between different studies. In the 12 studies that addressed objective no. 1, irritable bowel syndrome was characterized as severe and refractory to medical therapy in one study, a mixture of mild, moderate and severe symptoms in two studies and mixture of moderate, severe in five studies. The severity of irritable bowel syndrome was not specified in the other four. All studies of irritable bowel syndrome patients with moderate to severe symptoms showed a statistically significant reduction in all domains of SF-36.20, 23–25, 27, 28 Only one study presented separate a analysis for patients with mild irritable bowel syndrome (n = 7);28 those patients had SF-36 scores that were not statistically different from controls. None of the studies addressing objective no. 2 measured the severity of irritable bowel syndrome symptoms.
The available results suggest that the severity of the bowel symptoms in irritable bowel syndrome patients is associated with a corresponding impact on the HRQoL. Patients with worse bowel symptoms have a diminished quality of life, while those with milder symptoms may have a less significant impact.
Summary and discussion
- Top of page
- Materials and methods
- Summary and discussion
The findings of this systematic review in relation to objectives outlined above are:
Objective no. 1:
Compared to healthy controls, HRQoL is significantly impaired among irritable bowel syndrome patients who seek consultation for their symptoms. However, this may not the case for ‘nonconsulters’ persons with irritable bowel syndrome detected in the general population. The mental as well as physical scores were similarly reduced in most studies examined in this review.
Objective no. 2:
There is weak evidence to suggest that patients with irritable bowel syndrome seen at referral centres have HRQoL scores similar to, or less than, those of patients with other common disorders including GERD, diabetes, peptic ulcer disease and inflammatory bowel disease. The relative impact of irritable bowel syndrome compared to other disorders depends upon the specific disease under comparison and the domains of interest. Some diseases have worse HRQoL in specific domains than irritable bowel syndrome, while they may be similar or better than irritable bowel syndrome in other domains.
Objective no. 3:
There is good evidence to suggest that HRQoL scores correlate well with other therapy-associated changes in symptoms and disability.
Objectives no. 4 and no. 5:
The severity but not the pattern of bowel symptoms seems to be related to the degree of HRQoL impairment in patients with irritable bowel syndrome.
The methodology of this systematic review closely adhered to established methods for the performance of systematic reviews, as outlined by Oxman et al.34 A dual abstraction of the information from the papers minimized the possible bias of a single reviewer influencing the results. It should be noted that the reviewers achieved virtually complete agreement in their independent reviews. Even so, excluding abstracts in the systematic review has the potential to miss some relevant information. The reason for not searching for abstracts was that they are not subjected to rigorous peer review. In addition, our aim of comparing between the scores of physical and psychological domains required a level of detail in the methods and results that were felt to be unlikely to appear in abstracts. Nonetheless, it is possible that this resulted in a publication bias in our results. We attempted to overcome this by doing extensive searches, in both English and other languages, for all published papers. As outlined in the methods section, additional manual searches of references was also performed. We feel that it was unlikely that a published article of sufficient size to alter our conclusions was missed in our search.
The nature of the literature on HRQoL in irritable bowel syndrome patients is limited by several factors. First, most of the data focus on patients with moderate to severe irritable bowel syndrome who were referred to tertiary care medical centres. Thus, these results may have a limited generalizability to irritable bowel syndrome patients with mild disease or those who do not seek out medical care for other reasons. Second, the co-morbid physical illness or psychological illnesses of the study patients were not characterized in most of these studies. Future studies need to examine this factor, quantify it and adjust for it when comparing HRQoL for irritable bowel syndrome patients to healthy controls or controls with other medical disorders. Third, most studies compared the HRQoL of irritable bowel syndrome patients to historical controls seen at different institutions. Ideally, studies should prospectively identify an appropriate sample of control subjects without irritable bowel syndrome and match these patients by age and gender with the irritable bowel syndrome patients. Both groups of patients would then be assessed for HRQoL. Finally, the use of both general and disease specific questionnaires will allow a comparison of the impact of irritable bowel syndrome on HRQoL, both to other diseases, and to other studies of similar populations.
- Top of page
- Materials and methods
- Summary and discussion
Dr El-Serag is a VA HSR&D awardee (Grant no. RCD00-013-2). This study was partially supported by Novartis Pharmaceuticals. Dr El-Serag and Dr Bjorkman have served as consultants to Novartis Pharmaceuticals Corporation.
- Top of page
- Materials and methods
- Summary and discussion
- 9The SF-36 Health Survey. In: Spilker, B, ed. Quality of Life and Pharmacoeconomics in Clinical Trials, 2nd edn. Philadelphia: Lippincot-Raven, 1996: 337–45.
- 10Psychological aspects of health related quality of life measurement. Tests and scales. In: Spilker, B, ed. Quality of Life and Pharmacoeconomics in Clinical Trials, 2nd edn. Philadelphia: Lippincot-Raven, 1996: 117–31., , , et al.
- 15Irritable bowel syndrome: toward optimal design of controlled treatment trials. Gastroenterol Int 1993; 6: 189–211., , , et al.
- 18Bowel disorders impair functional status and quality of life in the elderly: a population-based study. J Gerontol A Biol Sci Med Sci 1995; 50: 184–9., , , .