SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Subject enrolment and study design
  6. Questionnaire
  7. Data handling and statistical methods
  8. Results
  9. Characteristics of the population
  10. Discriminative validity of the WPAI:IBS
  11. Accuracy of the WPAI:IBS
  12. Discussion
  13. Acknowledgement
  14. References

Background : Irritable bowel syndrome is a common chronic functional gastrointestinal disorder characterized by recurrent abdominal pain and discomfort associated with alterations in bowel habit. Irritable bowel syndrome affects patients’ quality of life and increases productivity loss.

Aim : To assess validity and accuracy of the Work Productivity and Activity Impairment questionnaire in irritable bowel syndrome as a tool for quantifying the effects of irritable bowel syndrome on productivity and daily activities.

Methods : Validity and accuracy were evaluated in 135 irritable bowel syndrome patients relative to three measures of irritable bowel syndrome disease severity; a debriefing questionnaire; retrospective diary; Work Limitations Questionnaire, and an activity impairment measure (Dimensions of Daily Activities).

Results : Symptom severity scores, diary scores, Work Limitations Questionnaire and Dimensions of Daily Activities were significant predictors of work productivity and activity impairment questionnaire in irritable bowel syndrome measures of work time missed, and work and activity productivity loss (P = 0.04 to < 0.0001). Impairment due to irritable bowel syndrome was estimated to be 2.9–4.3% for work time missed and 22–32% for impairment at work, the equivalent of 9.7 –14 h lost productivity per week. Activity impairment was 24–41%.

Conclusions : Discriminative validity of the Work Productivity and Activity Impairment questionnaire in irritable bowel syndrome was established, making it the only validated tool for measuring the relative differences between disease severity groups and quantifying work productivity loss and activity impairment in irritable bowel syndrome patients.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Subject enrolment and study design
  6. Questionnaire
  7. Data handling and statistical methods
  8. Results
  9. Characteristics of the population
  10. Discriminative validity of the WPAI:IBS
  11. Accuracy of the WPAI:IBS
  12. Discussion
  13. Acknowledgement
  14. References

Irritable bowel syndrome (IBS) is a common chronic functional gastrointestinal disorder characterized by recurrent abdominal pain and discomfort associated with alterations in bowel habit (constipation, diarrhoea, or alternating periods of both). IBS has a broad spectrum of symptom severity, ranging from mild to severe and intractable. IBS prevalence estimates vary widely, depending on populations studied and diagnostic criteria used, but are generally thought to be between 10 and 15%.1–4 Patients with IBS often experience symptoms for many years, with an average duration of 10 years or more.5 Several reports have described the negative effects of the disease on quality life6–8 and health care costs.9, 10 In the United States alone, it is estimated that the direct annual costs of IBS are $1.7–$10 billion11, 12 while indirect costs, including lost productivity, are estimated at $20 billion.12 Taking into account direct and indirect costs, an employee with IBS costs the employer 50% more than an employee without IBS.13

Health-related work productivity loss includes time lost from work (absenteeism) and reduced on-the-job effectiveness (presenteeism). These outcomes have not been adequately investigated for IBS patients. Previous IBS studies have measured days missed from work,14 missed and reduced activity days,15 or estimated time away from work for health care encounters,16 but hours lost from work of less than a day and presenteeism have not been quantified. Experience with other chronic conditions, such as allergies17 and gastro-oesophageal reflux disease (GERD)18 suggests that missed work hours and impairment at work are common outcomes for patients who may not take days off from work when suffering disease symptoms.

The Work Productivity and Activity Impairment (WPAI) questionnaire measures work time missed and work and activity impairment because of a specified health problem during the past 7 days19 (for questionnaire: http://www.reillyassociates.net/WPAI_SHP.html). The validity of the WPAI has been established in a number of diseases, e.g. allergies,20 dermatitis,21 GERD,22 nocturia,23 and asthma.24 In addition, the WPAI has proven a useful tool to measure the relative difference between treatment groups in clinical trials25–29 and in subjects with and without disease.30, 31

The primary objective of this study was to determine if the WPAI modified for IBS (WPAI:IBS) was valid (able to differentiate between patients with varying IBS symptom severity) and whether it would be useful for quantifying productivity loss in clinical practice and economic analyses.

Subject enrolment and study design

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Subject enrolment and study design
  6. Questionnaire
  7. Data handling and statistical methods
  8. Results
  9. Characteristics of the population
  10. Discriminative validity of the WPAI:IBS
  11. Accuracy of the WPAI:IBS
  12. Discussion
  13. Acknowledgement
  14. References

We enrolled a sample of patients with IBS at five USA sites from September to December 2003. Currently employed patients who met Rome II IBS diagnostic criteria32 were eligible to participate in the study. New England Internal Review Board approved the protocol and all patients signed informed consent.

Qualified patients completed a self-administered questionnaire that included several sections: demographics and disease severity; the WPAI:IBS; the Work Limitations Questionnaire (WLQ);33 debriefing and Dimensions of Daily Activities (DDAI); and a retrospective diary. In order to test the effect of the detailed questions about work productivity in the WLQ on the single-item work productivity question in the WPAI:IBS, patients were randomly assigned to one of two groups, with different order of administration of the questionnaire, as follows (Table 1):

Table 1.  Order of administration of the sections of the questionnaire
Group A (N = 67)Group B (N = 68)
  1. WPAI, Work Productivity and Activity Impairment questionnaire; IBS, irritable bowel syndrome; DDAI, Dimensions of Daily Activities; WLQ, Work Limitations Questionnaire.

Demographics and symptom severityDemographics and symptom severity
WPAI:IBSWLQ
Debriefing and DDAIWPAI:IBS
WLQDebriefing and DDAI
DiaryDiary

Group A: Demographics/symptom severity and WPAI:IBS; debriefing/DDAI: WLQ, diary.

Group B: Demographics/symptom severity and WLQ; WPAI:IBS; debriefing/DDAI; diary.

The questionnaire was given to the subject in sections, so that only one section could be accessed at a time. For the debriefing section, the top copy of the WPAI:IBS was returned to the subject for reference and possible revision, while the study administrator retained the carbonized copy of the original responses. In each section of the questionnaire except the WLQ, additional questions were asked about non-IBS health problems; these results will be reported in a future publication.

Questionnaire

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Subject enrolment and study design
  6. Questionnaire
  7. Data handling and statistical methods
  8. Results
  9. Characteristics of the population
  10. Discriminative validity of the WPAI:IBS
  11. Accuracy of the WPAI:IBS
  12. Discussion
  13. Acknowledgement
  14. References

The IBS symptom severity was self-assessed by patients with three measures: level (none/mild, moderate, severe, or no symptoms); visual analogue scale (VAS) from 0 (no symptoms) to 10 (very severe symptoms); symptom distress because of five symptoms (constipation, gas, abdominal pain or discomfort, bloating, diarrhoea), scored by summing the product of the symptom frequency (from 0, none of the time, to 4 all of the time) and degree of distress (from 1, not at all to 5, extremely). Distress scores ranged from 0, no symptom distress to 100, i.e. all symptoms occurring all the time, with extreme distress. The recall period was 7 days for all measures.

The WPAI:IBS consists of six questions, which elicit the following: employment status; hours missed because of IBS, hours missed because of other reasons; hours actually worked; degree IBS affected productivity while working (VAS from 0 to 10); degree IBS affected regular activities (VAS from 0 to 10). The sum of work time missed and impairment at work yields the overall work impairment (productivity loss) score; scores are expressed as percentage of impairment/productivity loss, with higher scores indicating greater impairment.

During the debriefing, patients referred to their original WPAI:IBS responses (WPAI:IBS second administration) while they answered questions regarding the number of days worked and the number of entire or part days missed as a result of IBS, and described in their own words how IBS symptoms caused any impairment reported. We considered the debriefing response as contradicting the original WPAI:IBS response if the original WPAI:IBS response was changed from some impairment to no impairment; if the original WPAI:IBS response indicated impairment, but the debriefing response indicated there were no days of impairment; or if a statement made was inappropriate, e.g. if a subject indicated missed time from work was the reason for impairment while working. Responses were considered missing if impairment was indicated on the debriefing questionnaire, but the text response was blank. All other debriefing responses were considered as supporting WPAI:IBS responses.

Following the debriefing, patients completed a 25-item DDAI questionnaire assessing the frequency of difficulty performing daily activities other than work as a result of IBS symptoms. The DDAI was created for this study and was modelled on the WLQ. Responses were collected on a 5-point scale from 0 (difficult none of the time) to 100% (difficult all of the time) and were scored in two ways: the percentage of difficult activities (at least a slight bit of the time), from 0 to 100%, and the average difficulty of activities from 1 to 5, with higher scores indicating greater difficulty.

For the retrospective diary, patients were first given a 7-day calendar marked with the day of the week and date of the past 7 days and instructed to write down anything they remembered about each day that would distinguish 1 day from the next. They were then given a questionnaire for each day and asked about the severity of IBS symptoms (VAS) and to complete an abbreviated WPAI:IBS. Each day's work and activity scores were calculated as for the weekly WPAI:IBS. Days ranged from 1 (yesterday) to 7 (a week ago today).

The WLQ consists of 25-items that assess the frequency of difficulty performing work tasks because of health problems during the past week. The WLQ items are scored on a 5-point scale from 0 (difficult none of the time) to 100% (difficult all of the time). Four subscales are scored and then combined to form an overall index of work productivity loss. Scores range from 0 to 28.6, with higher scores indicating greater productivity loss. The highest possible score (28.6) is equivalent to approximately 25% productivity loss.34

Data handling and statistical methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Subject enrolment and study design
  6. Questionnaire
  7. Data handling and statistical methods
  8. Results
  9. Characteristics of the population
  10. Discriminative validity of the WPAI:IBS
  11. Accuracy of the WPAI:IBS
  12. Discussion
  13. Acknowledgement
  14. References

Questionnaire data were entered through independent dual entry by two researchers. The analysis was performed using sas version 8.2.

Gender, employment status, order of questionnaire administration, site, IBS severity (level), occupation, and education were treated as categorical variables; all other variables were continuous. Pearson product moment correlation coefficients (and corresponding 95% confidence intervals) were calculated between the original and reviewed WPAI:IBS scores; between the 7-day WPAI:IBS scores and the average of the WPAI:IBS diary scores; between the severity of IBS symptoms and average diary severity scores; between the WPAI:IBS scores and the most severe diary score; between WPAI:IBS scores and each day's diary scores. Analysis of covariance (ancova) was used to test the relationship between the dependent measures (WPAI:IBS impairment scores) and independent measures of the severity of IBS symptoms, WLQ and DDAI scores. In all analyses, a P < 0.05 was required for significance using two-sided hypothesis tests; no P-value adjustments were made for the analyses of multiple end-points.

Characteristics of the population

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Subject enrolment and study design
  6. Questionnaire
  7. Data handling and statistical methods
  8. Results
  9. Characteristics of the population
  10. Discriminative validity of the WPAI:IBS
  11. Accuracy of the WPAI:IBS
  12. Discussion
  13. Acknowledgement
  14. References

Of the 135 patients in the study, 133 reported being currently employed; two patients were not employed and were excluded from the analyses of work impairment measures, but included in the analyses of activity impairment. Demographics, clinical characteristics and work and activity impairment scores are displayed in Table 2. Patients in the two questionnaire groups were comparable in all characteristics.

Table 2.  Demographics, clinical characteristics, and work and activity impairment scores by order of questionnaire administration
VariableMean (±s.d.) or percentage
Group A (N = 67)Group B (N = 68)Total* (N = 135)
  1. IBS, irritable bowel syndrome; DDAI, Dimensions of Daily Activities; WLQ, Work Limitations Questionnaire; VAS visual analogue scale.

  2. * Sample size may vary due to missing information. Two subjects in group B were not employed and are excluded from work variables. Items may not sum to 100% due to rounding.

Age (years)44.8 ± 12.745.9 ± 11.545.4 ± 12.1
Gender (female, %)939091
Education (some college, %)726970
Occupation (white-collar, %)908488
Type of IBS (%)
 Normal bowel pattern735
 Constipated302427
 Diarrhoea273129
 Alternating constipation/diarrhoea344339
Time since IBS diagnosis (years)13.1 ± 12.514.4 ± 10.813.7 ± 11.7
IBS symptom severity level (%)
 No symptoms444
 Mild242927
 Moderate555153
 Severe161516
IBS symptom severity (VAS)6.1 ± 2.46.0 ± 2.46.0 ± 2.4
IBS symptom distress36.7 ± 16.633.3 ± 17.335.0 ± 17.0
DDAI: average difficulty of activities2.0 ± 0.72.2 ± 0.92.1 ± 0.8
DDAI: frequency of difficult activities (%)59.264.261.7
WLQ index6.5 ± 4.66.9 ± 5.26.7 ± 4.9

Discriminative validity of the WPAI:IBS

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Subject enrolment and study design
  6. Questionnaire
  7. Data handling and statistical methods
  8. Results
  9. Characteristics of the population
  10. Discriminative validity of the WPAI:IBS
  11. Accuracy of the WPAI:IBS
  12. Discussion
  13. Acknowledgement
  14. References

Table 3 displays the results of the validity analysis. Symptom severity level was a significant predictor of percentage work and activity impairment and overall work impairment (P = 0.04 to < 0.0001), and approached significance for percentage missed work time (P = 0.06). Symptom severity, as measured by VAS was a significant predictor of all WPAI:IBS measures (P = 0.03 to < 0.0001), as was symptom distress (P = 0.01 to < 0.0001). WPAI:IBS scores varied proportionately with the level of impairment, with patients in the none/mild, moderate and severe symptoms group reporting 2.5%, 3.7% and 10.1% work time missed, respectively; 22.2%, 36.6% and 38.6% productivity loss, respectively; 25.0%, 45.3%, 59.0% activity impairment, respectively.

Table 3.  Summary of validation results for the WPAI:IBS*
Validation criteriaWPAI impairment due to IBS† (N = 135)
Work time missedImpairment at workOverall work productivity lossActivity impairment
  1. IBS, irritable bowel syndrome; DDAI, Dimensions of Daily Activities; WLQ, Work Limitations Questionnaire; VAS, visual analogue scale; WPAI, Work Productivity and Activity Impairment questionnaire.

  2. * Sample size may vary due to missing information. Two subjects were not employed and are excluded from work variables. Items may not sum to 100% due to rounding.

  3. † P-values derived from ancovas, with order of questionnaire administration, site, type of IBS, time since diagnosis, age, and gender as covariates. For all P-values, an increase in the validation criterion was associated with an increase in WPAI score.

IBS symptom severity level
 None/mild (n = 42, %)2.522.224.725.0
 Moderate (n = 72, %)3.736.637.345.3
 Severe (n = 21, %)10.138.641.759.0
 All patients4.432.434.241.1
Effect of IBS symptom severity (level)P = 0.06P = 0.03P = 0.04P < 0.0001
Effect of IBS symptom severity (VAS)P = 0.03P = 0.0002P = 0.0001P < 0.0001
Effect of IBS symptom severity (distress)P = 0.01P < 0.0001P < 0.0001P < 0.0001
Effect of WLQ indexNAP < 0.0001NANA
Effect of DDAI index – percentage difficult activitiesNANANAP < 0.0001
Effect of DDAI index – average difficultyNANANAP < 0.0001
Effect of average day-by-day scoresP < 0.0001P < 0.0001P < 0.0001P < 0.0001

The average (or sum for missed work time) of the diary scores was a significant predictor of the corresponding WPAI:IBS score (P < 0.0001). As hypothesized, the WLQ score was a significant predictor of WPAI:IBS impairment at work (P < 0.0001) and DDAI scores were significant predictors of activity impairment (P < 0.0001). Order of questionnaire administration was not a significant predictor in any of the analyses.

Accuracy of the WPAI:IBS

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Subject enrolment and study design
  6. Questionnaire
  7. Data handling and statistical methods
  8. Results
  9. Characteristics of the population
  10. Discriminative validity of the WPAI:IBS
  11. Accuracy of the WPAI:IBS
  12. Discussion
  13. Acknowledgement
  14. References

Table 4 displays the results of the analysis of accuracy. Correlation coefficients between WPAI:IBS scores on first administration and second administration ranged from r = 0.97–0.99. Exact agreement between first and second administration was 91% for overall work productivity (absenteeism and presenteeism) and 95% for the other measures. Scores tended to decrease slightly from the first to the second administration.

Table 4.  Summary of accuracy results for the WPAI:IBS*
Accuracy measuresWPAI impairment due to IBS (N = 135)
Percentage or correlation coefficient (confidence interval)†
Work time missed (%)Impairment at work (%)Overall work productivity loss (%)Activity impairment (%)
  1. IBS, irritable bowel syndrome; WPAI, Work Productivity and Activity Impairment questionnaire.

  2. * Sample size may vary due to missing information. Two subjects were not employed and are excluded from work variables.

  3. † Pearson product moment correlations.

  4. ‡ Two outliers excluded from percentage work time missed. When included r = 0.52 (0.382–0.638).

Agreement between first and second administrations of the WPAIr = 0.99 (0.99–0.1.0) 95%r = 0.98 (0.97–0.99) 95%r = 0.98 (0.97–0.99) 91%r = 0.97 (0.96–0.98) 95%
Agreement between WPAI score and average of day-by-day scores‡r = 0.84 (0.78–0.89) 71%r = 0.75 (0.66–0.82) 18%r = 0.75 (0.66–0.82) 15%r = 0.74 (0.65–0.81) 8%
WPAI score vs. average of day-by-day score4.3% vs. 2.9%32.4% vs. 22.0%34.4% vs. 22.8%41.1% vs. 23.8%
Agreement between WPAI score and day-by-day scores (days with highest agreement)NADays 4, 3, 1 and 2NADays 3, 7 and 4
Agreement between WPAI score and the most severe daily scoreNAr = 0.71 (0.62–0.79)NAr = 0.74 (0.65–0.80)
Contradictory data0.5% (n = 96.8 h)7.0% (n = 115)NA9.5% (n = 126)

Correlation coefficients between WPAI:IBS and diary scores ranged from r = 0.52 for work time missed to 0.74–0.75 for the other measures. For comparison, the same correlation for IBS symptom severity VAS was r = 0.72. When two outliers were removed from the work time missed analysis (i.e. one patient reporting 30 h missed on the WPAI:IBS and 0 on the diary, and another reporting 0 h missed on the WPAI:IBS and 4 on the diary) the correlation coefficient was r = 0.84 for missed work time. Exact agreement between WPAI:IBS scores and the average (or sum) of diary scores was 71% for work time missed, and much lower for the other measures, ranging from 8 to 18%. Compared with the WPAI:IBS, diary scores were lower for worked time missed, absenteeism (4.3% vs. 2.9%); impairment at work, preseenteism (32.4% vs. 22.0%); overall work productivity loss, absenteeism plus preseenteism (34.4% vs. 22.8%) and activity impairment (41.1% vs. 23.8%).

In the test for the effect of recency on scores, the score for the most recent day (day 1) or the second most recent day (day 2) was never the score most highly correlated with the WPAI:IBS score or IBS severity measure VAS. In the test for the effect of saliency on scores, the WPAI:IBS work impairment score had a slightly higher correlation with the average of the 7 days than with the most severe day's score (r = 0.75 vs. 0.71), while for activity impairment, the two correlation coefficients were equivalent (r = 0.74 vs. 0.74). For the severity of IBS symptoms, the correlation coefficient between the average score and the 7-day recall score was slightly higher than the coefficient with the most severe day (r = 0.72 vs. 0.70).

A total of 36 patients (27.1%) reported missing 96.8 h of time during the past 7 days, as measured by the WPAI:IBS. Only 1 h was not supported in the debriefing; hence 99.5% of missed work time was supported. Among the 115 patients (86.5%) who reported productivity loss while working, seven indicated in the debriefing that there was no productivity loss, and another subject stated that a missed work day was the only reason for productivity loss, making a total of eight patients with contradictory data (7.0%). Of the 126 (93.3%) who reported activity impairment, 12 (9.5%) did not indicate there was activity impairment in the debriefing.

Order of administration of the questionnaire had little or no effect on accuracy except for the correlations between the WPAI:IBS scores and the diary scores. When the WLQ was administered first, the correlation coefficient between WPAI:IBS and diary scores was higher for work time missed (r = 0.86 vs. 0.15) and lower for work impairment (r = 0.69 vs. 0.80) and activity impairment (r = 0.69 vs. 0.79). A separate investigation of the very large differences between the groups in the reporting of work time missed on the WPAI:IBS vs. the diary indicated that the relatively low rate of patients with missed time and the erratic responses of a few patients account for this difference. When the two outliers were removed from group A for the analysis of work time missed, the two groups were comparable (r = 0.77 vs. 0.86). For both groups, agreement between WPAI:IBS and diary item responses (as opposed to scores) was much higher for hours missed (79%) than for hours worked (28%).

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Subject enrolment and study design
  6. Questionnaire
  7. Data handling and statistical methods
  8. Results
  9. Characteristics of the population
  10. Discriminative validity of the WPAI:IBS
  11. Accuracy of the WPAI:IBS
  12. Discussion
  13. Acknowledgement
  14. References

Developing a valid and accurate measure to quantify the effects of the disorder on productivity loss is an important step in evaluating treatment outcomes for IBS patients.

Since there is no clinical test that can confirm either the diagnosis of IBS or its severity, we used three patient-reported symptom measures as criteria for validating the WPAI:IBS and found that these measures were significant predictors of the WPAI:IBS scores. Furthermore, the average of daily impairment scores for the past 7 days obtained from retrospective diary was a significant predictor of the corresponding WPAI:IBS impairment measure. In addition, the WLQ index and the DDAI were significant predictors of the WPAI:IBS work and activity impairment measures, respectively. These results demonstrate the discriminate validity of the WPAI:IBS, and are consistent with the WPAI validation results for other diseases.

While a valid instrument is useful for measuring the relative differences between groups, imputing an economic valuation to these differences requires that the accuracy of the measures be considered. Some researchers have found that retrospective questionnaires measuring work productivity may understate time missed from work relative to employment35 and overstate productivity loss on the job relative to diary entries,36, 37 while others have demonstrated that patients tend to underestimate work absence and over-report performances.38 In this study, we measured WPAI:IBS accuracy relative to supportive and contradictory data obtained with debriefing questions, and by comparing the WPAI:IBS 7-day recall to the average or sum of day-by-day recall obtained from a retrospective daily diary.

We found excellent accuracy for percentage work time missed, with the mean difference between the WPAI:IBS and diary results equivalent to 34 min per 40-h workweek. This difference was largely attributed to the responses of one patient, and either estimate is a more accurate assessment of absenteeism than the standard measure of days missed. For impairment at work (presenteeism), while the exact agreement between the WPAI:IBS score and the diary was low, the difference in mean scores was equivalent to only 4 h in a 40-h workweek.

As both the WPAI:IBS and the retrospective diary rely on recall, we cannot determine which represents a more accurate assessment of impairment during the past 7 days. Indeed, the validity of the impairment scores obtained by retrospective diary is untested, as is the validity of almost all scores obtained from prospective diaries. Patient diaries are often deemed the gold standard, but when validated against independent measures, have been found to have some limitations.39, 40 While the retrospective diary used in this study is theoretically more accurate than the WPAI:IBS (given that patients were first debriefed about the productivity loss reported on the WPAI:IBS and then prompted to recall the details of each day separately prior to completing the diary), subject fatigue completing repetitive sections of the questionnaire may have affected its accuracy. However, the fact that the debriefing, diary and the WPAI:IBS responses converge is strong evidence to support the relative accuracy of WPAI:IBS responses.

Users of the WPAI:IBS for economic analyses should consider the potential effect of over-reporting productivity impact by providing ranges for results, instead of just point statistics, e.g. mean values. For example, in this study, using the retrospective diary as the low end of the range and the WPAI:IBS scores as the high end, the estimated percentage work time missed because of IBS in the past 7 days was approximately 2.9–4.3% (absenteeism). The range for impairment at work (presenteeism) was 22–32%, while the range for activity impairment was 24–41%.

A limitation of this study is that no independent employment measure was used as a validation criterion. Severens et al. demonstrated that reported days of sick leave matched registered data perfectly when the recall period was limited to 2 weeks,41 but no one has investigated the relationship between employment records and self-reported work hours missed. Much of the missed work hours in our population, an hour or 2 at a time, would have gone undetected by employment records except where the rigorous punching of time clocks is required. However, even clocked work hours are suspect because of recent allegations of widespread manipulation by managers to show fewer employee work hours,42 which suggests that self-reports are the only reliable option for obtaining hours missed and worked from most employees. The high convergence of the WPAI:IBS, debriefing questionnaire, and retrospective diary suggests that self-reports of percentage work hours missed during the past week are accurate.

Investigations of the relationship between self-report and objective measures of productivity on the job have been difficult to execute and are limited in their applicability across the occupational spectrum. We included the WLQ as a validation measure in this study since Lerner et al. established the relationship between the WLQ and objective employment measures in two populations.33 Although we found the WLQ was a significant predictor of WPAI:IBS work impairment as a result of IBS, the amount of productivity loss reported with the WLQ was much lower. It is not possible to determine whether the WPAI approach (a single VAS) or the WLQ (25 work activities grouped in four categories) is a more accurate measure of work productivity loss, as the relationship between the instruments and the objective loss in the quality and quantity of work in a highly educated, white-collar population working in diverse settings and industries, is not known. There is no reason to suspect that the WPAI VAS approach is any less precise than the WLQ multidimensional approach, as global assessments have been repeatedly found to be excellent predictors of clinical and quality of life outcomes, and are among the most responsive to clinical change.43–45 The investigation of the relationship between self-reports and objective measures of productivity loss continues to represent an area for future research.

The results of the validation of the WPAI:IBS have important implications for other work productivity measures and patient-reported outcomes in general. Since researchers have demonstrated the increased fall-off in accuracy of patient reports when the recall interval lengthens,41 we speculate that any lack of precision in a 7-day recall, such as used in the WPAI:IBS, will increase in proportion to the recall interval, when work hours, not work days, is the measurement unit and when productivity on the job is assessed. Although we found no indication that the most recent days unduly affected results, there was evidence that the most severe day in the recall period did affect results, not only for WPAI:IBS scores, but symptom scores as well, suggesting that patient-reported outcomes over an interval may be unduly influenced by the most severe day in that interval, and not be a true average assessment. This is the well-documented phenomenon of recall bias as a function of the salience of the event.46 Shorter recall periods, even as brief as 24 h, may be warranted to reduce the fall-off in recall and the effect of saliency when precision, not only relative differences, is required. However, any loss in precision with 7-day recall needs to be balanced against the increase in cost and patient burden with daily assessments. A retrospective or prospective diary might be useful in some situations, provided its validity had been established.

Discriminative validity of the WPAI:IBS was established making the WPAI:IBS the only validated tool for measuring the relative differences between IBS severity groups and quantifying work productivity loss and activity impairment in IBS patients. The WPAI:IBS is a useful tool for measuring productivity outcomes in clinical practice, but for greater accuracy in economic analyses, results should be presented as ranges.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Subject enrolment and study design
  6. Questionnaire
  7. Data handling and statistical methods
  8. Results
  9. Characteristics of the population
  10. Discriminative validity of the WPAI:IBS
  11. Accuracy of the WPAI:IBS
  12. Discussion
  13. Acknowledgement
  14. References
  • 1
    Badia X, Mearin F, Balboa A, et al. Burden of illness in irritable bowel syndrome comparing Rome I and Rome II criteria. Pharmacoeconomics 2002; 20: 74958.
  • 2
    Camilleri M, Choi MG. Review article: Irritable bowel syndrome. Aliment Pharmacol Ther 1997; 11: 315.
  • 3
    Talley NJ. Pharmacologic therapy for the irritable bowel syndrome. Am J Gastroenterol 2003; 98: 7508.
    Direct Link:
  • 4
    Hungin AP, Whorwell PJ, Tack J, Mearin F. The prevalence, patterns and impact of irritable bowel syndrome: an international survey of 40,000 subjects. Aliment Pharmacol Ther 2003; 17: 64350.
  • 5
    Olden KW. Approach to the patient with severe, refractory irritable bowel syndrome. Curr Treat Options Gastroenterol 2003; 6: 3317.
  • 6
    Pace F, Molteni P, Bollani S, et al. Inflammatory bowel disease versus irritable bowel syndrome: a hospital-based, case-control study of disease impact on quality of life. Scand J Gastroenterol 2003; 38: 10318.
  • 7
    Gralnek IM, Hays RD, Kilbourne A, Naliboff B, Mayer EA. The impact of irritable bowel syndrome on health-related quality of life. Gastroenterology 2000; 119: 65460.
  • 8
    Silk DB. Impact of irritable bowel syndrome on personal relationships and working practices. Eur J Gastroenterol Hepatol 2001; 13: 132732.
  • 9
    Wells NE, Hahn BA, Whorwell PJ. Clinical economics review: irritable bowel syndrome. Aliment Pharmacol Ther 1997; 11: 101930 (Review).
  • 10
    Levy RL, Von Korff M, Whitehead WE, et al. Costs of care for irritable bowel syndrome patients in a health maintenance organization. Am J Gastroenterol 2001; 96: 31229.
    Direct Link:
  • 11
    Martin R, Barron JJ, Zacker C. Irritable bowel syndrome: toward a cost-effective management approach. Am J Manage Care 2001; 7(Suppl. 8): S26875.
  • 12
    American Gastroenterological Association. The burden of gastrointestinal diseases, 2001. Bethesda, MD: 136.
  • 13
    Leong SA, Barghout V, Birnbaum HG, et al. The economic consequences of irritable bowel syndrome: a US employer perspective. Arch Intern Med 2003; 163: 92935.
  • 14
    Hahn BA, Yan S, Strassel S. Impact of irritable bowel syndrome on quality of life and resource use in the United States and United Kingdom. Digestion 1999; 60: 7781.
  • 15
    Jhingran P, Decker C, Watson M, Northcutt A, Ricci J-F. Alosetron reduces time lost from work in women with diarrhoea-predominant irritable bowel syndrome: data from phase III placebo-controlled clinical trials. Clin Drug Invest 2001; 21: 84352.
  • 16
    Sandler RS, Everhart JE, Donowitz M, et al. The burden of selected digestive diseases in the United States. Gastroenterology 2002; 122: 150011.
  • 17
    Murray JJ, Nathan RA, Bronsky EA, Olufade AO, Chapman D, Kramer B. Comprehensive evaluation of cetirizine in the management of seasonal allergic rhinitis: impact on symptoms, quality of life, productivity, and activity impairment. Allergy Asthma Proc 2002; 23: 3918.
  • 18
    Wahlqvist P. Symptoms of gastroesophageal reflux disease, perceived productivity, and health-related quality of life. Am J Gastroenterol 2001; 96(Suppl. 8): S5761.
  • 19
    Reilly MC, Zbrozek AS, Dukes EM. The validity and reproducibility of a work productivity and activity impairment instrument. Pharmacoeconomics 1993; 4: 35365.
  • 20
    Reilly MC, Tanner A, Meltzer EO. Work, classroom and activity impairment instruments: validation studies in allergic rhinitis. Clin Drug Invest 1996; 11: 27888.
  • 21
    Reilly MC, Lavin PT, Kahler KH, Pariser DM. Validation of the dermatology life quality index (DLQI) and the work productivity and activity impairment: chronic hand dermatitis questionnaire (WPAI-ChHD) in chronic hand dermatitis (ChHD). J Am Acad Dermatol 2003; 48: 12830.
  • 22
    Wahlqvist P, Carlsson J, Stålhammar N-O, Wiklund I. Validity of a work productivity and activity impairment questionnaire for patients with symptoms of gastro-esophageal reflux disease (WPAI-GERD) – results from a cross-sectional study. Value Health 2002; 5: 10613.
  • 23
    Kobelt G, Borgstrom F, Mattiasson A. Productivity, vitality and utility in a group of healthy professionally active individuals with nocturia. Br J Urol Int 2003; 9: 1905.
  • 24
    Andréasson E, Svensson K, Berggren F. The validity of the work productivity and activity impairment questionnaire for patients with asthma (WPAI-asthma): results from a web-based study. Value Health 2003; 6: 780.
  • 25
    Jacobs RJ, Davidson JR, Gupta S, Meyerhoff AS. The effects of clonazepam on quality of life and work productivity in panic disorder. Am J Manage Care 1997; 3: 118796.
  • 26
    Meltzer EO, Casale TB, Nathan RA, Thompson AK. Once daily fexofenadine HCI improves quality of life and reduces work and activity impairment in patients with season allergic rhinitis. Ann Allergy Asthma Immunol 1999; 83: 3117.
  • 27
    Tanner LA, Reilly M, Meltzer EO, Bradford JE, Mason J. Effect of fexofenadine HCI on quality of life and work, classroom and daily activity impairment in patients with seasonal allergic rhinitis. Am J Manage Care 1999; 5(Suppl. ): S23547.
  • 28
    Wahlqvist P, Bergenheim K, Långström G, Næsdal J. Resolution of upper gastrointestinal symptoms associated with chronic non-steroidal anti-inflammatory drug therapy has a positive impact on patient-reported productivity. Gut 2003; 52(Suppl. VI): A226.
  • 29
    Perrillo RP, Rothstein KD, Imperial JC, et al. Improved Work Productivity, Safety, and Quality of Life with Pegylated (40 kDa) Interferon ALFA-2a (PEGASYS®) Therapy in the Treatment of Chronic Hepatitis C (Poster presentations). Roche Laboratories, Nutley, NJ, 2000.
  • 30
    Dean B, Crawley J, Schmitt CM, Wong J, Ofman JJ. The burden of illness of gastro-oesophageal reflux disease: impact on work productivity. Aliment Pharmacol Ther 2003; 17: 130917.
  • 31
    Wittchen HU, Beloch E. The impact of social phobia on quality of life. Int Clin Psycopharmacol 1996; II: 1523.
  • 32
    Drossman DA. Rome II the Functional Gastrointestinal Disorders, 2nd edn. McLean, VA: Degnon Associates, 2000.
  • 33
    Lerner D, Amick BC III, Rogers WH, Malspeis S, Bungay K, Cynn D. The work limitations questionnaire. Med Care 2001; 39: 7285.
  • 34
    Lerner D, Rogers W, Chang H. Technical Report: Confidential Scoring the Work Limitations Questionnaire (WLQ©) Scales and the WLQ Index© for Estimating Work Productivity Loss. Revised April 22, 2003.
  • 35
    van Popppel MNM, de Ver HCW, Koes BW, Smid T, Bouter LM. Measuring sick leave: a comparison of self-reported data on sick leave and data from company records. Occup Med 2002; 52: 48590.
  • 36
    John G. Absentee estimates by employees and managers: divergent perspectives and self-serving behavior. J Appl Psychol 1994; 79: 22939.
  • 37
    Stewart WF, Lipton RB, Kolodner KB, Sawyer J, Lee C, Liberman JN. Validity of the migraine disability assessment (MIDAS) score in comparison to a diary-based measure in a population sample of migraine sufferers. Pain 2000; 88: 4152.
  • 38
    Kessler R, Barber C, Beck A, et al. The World Health Organization health and work performance questionnaire (HPQ). JOEM 2003; 45: 15674.
  • 39
    Day N, McKeown N, Wong M, Welch A, Bingham S. Epidemiological assessment of diet: a comparison of a 7-day diary with a food frequency questionnaire using urinary markers of nitrogen, potassium and sodium. Int J Epidemiol 2001; 30: 30917.
  • 40
    Santanello NC, Demuro-Mercon C, Davies G, et al. Validation of a pediatric asthma caregiver diary. J Allergy Clin Immunol 2000; 106: 8616.
  • 41
    Severens JL, Mulder J, Laheij R, Verbeek A. Precision in measuring absence from work as a basis for calculating productivity costs in the Netherlands. Soc Sci Med 2000; 51: 2439.
  • 42
    Greenhouse S. Altering of worker time cards spurs growing number of suits. New York Times, April 4, 2004.
  • 43
    Cappelleri JC, Siegel RL, Osterloh IH, Rosen RC. Relationship between patient self-assessment of erectile function and the erectile function domain of the international index of erectile function. Urology 2000; 56: 47781.
  • 44
    Buchbinder R, Bombardier C, Yeung M, Tugwell P. Which outcome measures should be used in rheumatoid arthritis clinical trials? Clinical and quality-of-life measures’ responsiveness to treatment in a randomized controlled trial. Arthritis Rheum 1995; 38: 156880.
  • 45
    Ward MM. Clinical measures in rheumatoid arthritis: which are most useful in assessing patients? J Rheumatol 1994; 21: 1727
  • 46
    Evans C, Crawford B. Patient self-reports in pharmacoeconomic studies. Their use and impact on study validity. Pharmacoeconomics 1999; 15: 24156 (Review).