Development and validation of the treatment satisfaction questionnaire for GERD (TSQ-G)

Development and validation of the treatment satisfaction questionnaire for GERD (TSQ-G)

'P1295 sleeping (38%), drinking alcohol (31%), social activities (21%), sports/exercise (16%), work (17%), hobbies (15%) and sexual activity (9%, exd...

489KB Sizes 2 Downloads 78 Views


sleeping (38%), drinking alcohol (31%), social activities (21%), sports/exercise (16%), work (17%), hobbies (15%) and sexual activity (9%, exdudiug Japan, where this question was omitted). Almost one-quarter of patients used >1 prescription drug (22%) or OTC drugs in addition to their prescription drugs (21%); however, PCPs reported that only f4% of their patients used additional OTC medication. The most common reason given by patients for asmg additional OTC drugs was that prescription medication alone was not dfective (41%). Conclusion: Although must patients t}d that prescription GERD therapy improves their quality of life, many report that GERD continues to have a substantial impact on their activities of daily life. The level of patients' dissatisfaction with therapy is reflected by the proportion of patients using > 1 prescription medication or using OTC drugs in addition to prescription therapy'. The use of OTC therapy was underestimated by physicians, suggesting that physicians are unaware that patients do not achieve adequate relief from prescription therapy alone

Reliability and Validity of a Modified Questionnaire of Upper GI Symptoms Utilized in a Disease Management Program of Patients on Chronic Acid Suppression Braden Kuo, Timothy Ferris Background: An efficient practical upper GI symptom assessment questionnaire was needed to guide management recommendations for a population based effort to systematically improve the care of patients on chronic acid suppression medication. The questionnaire also served for baseline and stud), endpoim evaluation Aims: To describe the psychometric properties of a brief questionnaire of upper GI symptom~ that was modified, from existing upper GI symptom assessment instmmems. Methods: Study population included patients on chronic acid suppression identified from pharmacy billing. 7 questions about upper GI symptom frequentT were modified from previously' validated questionnaires. Symptoms included the frequencies of heartburn(l), supine heartburn(2), use of antacids(3), stomach pare(4), bloating(5), pain alter eating(6), and nansen/vomiting(7). Symptom domains of GERD (symproms 1-3) and Dyspepsia (symptoms 3-7) were computed as scales (GSCALE, DSCALE) from the sum of disease specific s)anptom frequency scores Validity was tested through correlation with standard qoality of life questions including impact of symptoms on daily activities and days missed from work. Results: 812 patients completed the questionnacre (53% of aligtble patients) The internal consistency reliability within the GERD and Dyspepsia symptom dotnains was reasonable (Cronhach's alpha = 0.41-0.70). The Pearson cra~'clatiou coefficients between the symptom questions and the quality of life questions were modest to moderate (0.11-0.55, p<0.0001 tbr aN. The weakest correlations were between hea~bttrn symptoms and days missed t}om work. The strongest correlations were with heartburn and dyspepsia symptoms and interference of normal activitias The GSCALE and DSCALE were both associated with unpact on daily actb/tttes (pearson=0.50 and 0 5 5 respectively, pC0,0001 [br both). ]'he GSCALE and DSCALE showed more modest correlation with days missed from work (pearson=0.I7 and 0.37 respectively, p

Impact Of Gastroesophageal Reflux Disease On Worker Productivity In An Employed U.S. Population Bonnie B. Dean, Daniel Agnflar, Joseph A. Crawley, Robert W. Dubois Purpose: Our objective was to assess the impact of gastroesopbageal reflux disease (GERD) in comparison to other health conditions on work-minted productivity in an employed population. Methods: A survey to evaluate the impact of GERD on work pmduetiviry was administered to employees at a major service industry employer The survey was avMable to employees either on-line or by paper and pencil. The diagnostic classification of GERD was determined using the Reflux Disease Diagnostic Questionnaire (RDQ). Data on other health conditions were also collected. Subjective work productivity was obtained using the Work Productivity and Activity Impairment (WPAI) questionnaire. Mean productivity scores were calculated fbr participants with each condition. The Work Producnvity Score (WPS) combining work time absent and impairment while working (presenteeism) as a percent of potential total work productivity was reported. Results: 1186 employees (14.8% of current employees) responded to the survey, with complete data available on 956 employees. 273 (28.5%) respondents met the criteria for GERD. In addition to GERD, the most prevalent conditious reported were allergies, chronic headaches, high blood pressure, arthritis, and back pain. The mean number of conditions anmng individuals was 1.2 ( +/-3.1 ). The percent of total reduced work productivity ranged fi'om 9,7% for high blood pressure to 163% for headaches and back pain. For a Ihll-time 4Ghour per week employee, this translates to over 4.5 hours per week of lost productivity Participants with GERD reported total work productivity losses of 15.8%, just below those reported for headaches and back pain (16.3% each). Productivity losses of this magnitude represent, over 6.5 hours of lost productivity per week for employees with back pain, headaches, and GERD. Conclusions: Chronic conditions affecting employees can impact not only their absenteeism but their actual productivity while working, When measured, presenteeism impacted the employer morn than absenteeism and suggests that lost productivity due to chronic diseases such as GERD are a significant cost to employers.


Primary Care Physicians Perceptions and Practices on Xhe Management of GERD: Results of a National Survey W. D. Chey, j M. Inadomi, A. K Booher, A M. Fendrick, V. K Shanua, C W. Howden Purpose: Gastroesophageal reflux disease (GERD) is commonly managed by primary care physicians (PCPs) There is currently little information addressing how PCPs treat GERD We sought to better understand PCPs perceptions on GERD treatment. Methods: From 12/ 01 to 6/02, a 30 question multiple choice survey was administered to a geographically diverse US sample of PCPs prior to a series of regional conferences on GERD. Data was analyzed tbr the overall sample, using multivariable logistic regression to observe the effect of PCP age and geog)'aphic region. Results: Data from 1046 respondents are reported. The mean age of the study cohort was 48 yrs, 85% were male, and the mean # of yrs in practice was 17.43% were internists & 46% practiced family medicine. 13% were from the Southeast, 20% the South, 32% file Northeast, 20% the Midwest, and 15% the West. 15% reported that a trial with an H2RA was required prior to using a PPI 32% reported the need for pro-authorization for chronic PPt therapy; PCPs in Western and Southeastern states required pro-authorization more often than elsewhere (pC0.05) 36% of PCPs instructed pts to take a PPI with or after a meal or did not specit}, the timing of dosing. When asked about the critical hctors influencing selection of a specific PPL 39% chose efficacy while 56% chose formulary availability or insurance coverage. Price, satety, and FDA indications were less :mportam 75% of PCPs relerred pts tor surgical anti-reflux therapy and 20% referred pts directly to a surgeon without GI consuhation. Younger PCPs, and those in the Midwest and Southeast, were more likely to refer pts to surgery (pC0.001). When asked about the types of pts referred for surgery, 83% of PCPs preferentially" referred pts who bad failed medical therapy. Condnsion: Formulary availability & insurance coverage are the principal [actors influencing PPl selecuon by PCPs. 36% of PCPs do not properly counsel pta regarding the optimal timing of PPI dosing which likely impacts therapy" efficacy'. Direct referral of pts to surgeons and improper pt selection tbr anti-re flux surgery may help explain the disappointing results reported from community practices. Further education initiatives for PCPs to optimize GERD management are warranted.

WPS by Condition Productivity Measure WPS - Percent total reduced wo~ p~uc~ ,,~

GERD (n=273)

Allergies (n=292)



Headache HighBlood Back Pain Pressure {n=129) ( n = l ~ (n=125) 16.3



A~hriU$ tn='124) 12,8


Development and Validation of the Treatment Satisfaction Questionnaire for GERD (1,SQ-G) Karin Coyme, fngela Wiklund, Dennis Revicki, Jordana Schmier, Katanna Hailing, Alessio Degl' Iunoceuti Purpose: Treatment satisfaction (TxS) is a patiem-reported outcome that is bel:e-eed to influence adherence with treatment and treatment outcomes. Currently, no condkion-spedfic, patient-based treatment satisfaction instruments related to gastroesophageal reflux disease (GERD) exist; thus the purpose of this study was to develop and validate a TxS questionnaire in GERD. Methods: Six focus groups of GERD patients (n = 36) were conducted to ascertain patient perceptions of Tx$. Complete and immediate symptom relief was the most important aspect of TxS with side efl?cts, provider relationships, and cost bring secondary. The resulting 64-item questionnaire which used a 6-poim Liken scale ranging from very strongly agree to very strongly disagree was piloted in 64 patients and reduced to 41 items. The 41-item questionnaire was then vahdated in treated GERD patients; 25% were randomized to a 2-week test-retest visit Ancillary measures in the validation study were: SF-36, QOLRAD, GSRS, SDRS, PSQ-18, and physician and patient global measures of symptoms and satisfaction. Exploratory factor analysis and Rasch analysis were used to examine the items and subscale structure and assist in item reduction, Internal consistency, test-retest reliability, and construct validity were assessed using Cronbach alpha, intra-class correlations, ANOVA, and t-tests. Results: A total of 198 GERD patients participated in the validation study with a mean age = 50.7 yrs; 68% female, and 84% Caucasian. Clinicianrated severity of GERD was: mild (32%); moderate (50%); and severe (18%) and 83.3% were on a PPI. The final [email protected] consisted of 28 items with 7 subscales (symptoms, satistaction, MD relationship, cost, PRN, expectations, bother) with each subscale score ranging from 1 to 6. Crunbach alphas ranged from 0.58 (MD) to 0.94 (symptoms). Correlations were moderate to strong within the expected subscales of the SF-36, QOLRAD, GSRS, and PSQ18. No significant correlations were present with the SDRS indicting no acquiescence bias. Age and gender did not affect TSQ-G responses. TSQ-G subscales, particularly symptoms and satisfaction, significantly discriminated among levels of clinician-rated disease seventy,

T1297 A Multinational Survey of Activities of Daily Living and GERD in Clinical l~actiee: ls Prescription Therapy Adequate? Roger ~ones, David AL-msuong, Peter Malferth.einer Philippe Ducrotte, Raymond Colin Purpose: Clinical trial data suggest that therapy for gastroesophageal reflux disease (GERD) achieves healing and symptom relief in a very high propra'tion of patients; however, there are few data on treatment efficacy in teru~s of patients' perspective. The purpose of the survey was to investigate percepti~ms of GERD and its therapy' m patients being treated for GERD (regardless of diagnostic criteria) and their physicians and, specifically, to assess the effects of GERD on the activities o[ daily living and the use of ove>the-com:ter (OTC) medication. Methods: Randomly selected primary care physicians (PCPs: USA, UK, Prance, German),', Japan) answered questions about GERD and, with relerence to patient records, about the last 4 patients Who had consulted for GERD symptums. The 4 patients were then invited to participate in interviews using a standardized questionnaire, and for each patient who declined, an additional patient record was selected by the relevant physician~ Results: It: total, 261 PCPs (USA: 50, UK: 58, France: 50, Germaw: 53, Japan: 50) and 927 patients were interviewed, Overall, 91% of patients felt that GERD treatment had improved their quality of life (UK: 85% to Germany: 97%). Despite this, patients reported that GERD had a substantial impact, during the previous 3 months, on eating habits (59%),




and ROME II critena for IBS. The second phase survey included a modified versran of the validated Work Productivity" and Activity impairment (WPAI) Questionnaire, designed to measure work time absent (absenteeism) and productivW loss while working (presenteeism) due to GI symptoms (abdominal pairb/discomfort, bloating, and constipation). The Work Productivity Score (WPS), combimng absenteeism and presenteeism, enumerates reduced productivity due to GI syznptoms as a percent of potential total work productivity. Nonparametric methods were used to compare the mean difference in percent WPS scores between respondents with and without IBS. Results: 2615 (22%) of 11,806 employees returned the phase I survey. 2,276 employees (19%) returned completed surveys and consented to the second phase. A total of 1,776 employees responded to both surveys and were included in the final analysis. A total of 720 (41%) respondents met the Rome II criteria for IBS. Respondents with and without IBS wmv similar with respect to sociodemographic variables, except those with IBS were more hkely female and Caucasian (p<0.05). Employees with IBS reported significantly higher total work productivity loss compared to employees without iBS. Participants with IBS reported total work productivity losses of 19.8% due to GI symptoms compared with only"5.6% among those without IBS, producing a mean difference of 14.2% (95% CI: 12.5-15s Conclusions: Among employees with IBS, reduced work productivity" of almost 20% was reported. This is equivalent to working less than 4 days out of a total 5-day work-week. Reduced productivity of this magnitude may have a substantial financial impact on employers.

number of symptom days, and patient and clinician ratings of satisfaction. Test-retest correlations ranged fi-onl 0 6 5 to 0.92, Conclusions: The TSQ-G has shown excellent reliability and concurrent and discriminatiue validity It appears to be a useful tool for evaluating treatment sat*sfaction among GERD patients

T1300 Immediate Esophagectomy or Surveillance for Barrett's Patients with High Grade Dysplasia Dawn Provenzale Purpose: The cancer risk in Barrett's patients vath high grade dysplasia (HGD) is uncertain, with reported r~sks of 16%-59% in a 5-7 year period The purpose of this analysis was to evaluate immediate esophagectomy vs continued survedlance of HGD patients tbllowed over a 5 year period Methods: We modified our presqonsty published decision model to consider HGD patients only We considered immediate esophagectomy and surveillance every 3 months and compared this to no surveillance. We incorporated published literature on the risks of developing cancer, and complications of endoscopy and surgery We fbllowed a hypothetical group of individuals with HGD [br a 5 year period, and varied the cancer risk over the full range of reported values, to examine the effect on the preferred management strategy Results: Immediate esophagectorny was the prefen'ed strategy for the full range of reported cancer risks (16%-59%) The results suggest that immediate esophagecmmy will increase fill.'expectancy, on average, by up to 3 nmnths compared to surveillance. For those who choose surveillance as their management strategy, surveillance every 3 months will increase life expectancy by up to 7 months compared to no surveillance, immediate esophagectomy is the prderred strategy' unless the 5 year cancer risk in those who undergo surveillance tails below 11%. For a cancer risk ranging t)'om 9-11% surveillance every 3 months will provide the greatest gain in life expectancy'. If the 5 year cancer risk is less than 9%, no snrvedlance is the preterred strategy, because the risks of endoscopy and surgery outweigh the benefits of cancer prevention. When qnality of lite is considered, no surveillance is preferred unless the quafi~y of hfe afier esophagectomy is at least 0.8 (0 = dead, 1 = perfect health), and the 5 year cancer risk fnr those undergoing surveillance is at least 47% Surveillance is prdenzd when the cancer risk is 47-57% and immediate esophagectomy is preferred when the cancer risk exceeds 57% Conclusions: 1) For HGD patients, immediate esopbagectomy will prosade the greatest gain in 5 year hff' expectancy over the full range of reported cancer nsks. 2) Surveillance every 3 months also increases life expectancy. 3) The carmer risk in HGD patients is the critical parameter in decisimls for esophagectomy and surveillance Larger studies with longer periods of follow-up are needed 4) Quality" of file afier esophagectorW is another important component of decisions for immediate esophagectomy vs continued surveillance


Predictors of Health and Activity Limitation Index (HALex) Scores in Individuals with Self-reported Irritable Bowel Syndrome (IBS) and IBS-like Symptoms Patrick D. Meek Objectives: To evaluate demographic and socioeconomic characteristics alone and with severity characteristics of persons ~*nth self-reported irritable bowel syndrome as predictors of health and activity limitation index scores. Methods: The Health and Activity [imitation lndex (HALex) scoring system has recently been suggested as a way to classify individuals into health states with an assigned utility score. HALex scoring is based on two domains, activiry limitation and perceived health that are combined in a multiplicative model to form a generic quality of life index. Using correlation and ordinary least squares regression in a two-step process to evaluate predictors of HALex scores of individuals identified from cross sectional data t):om the 1989 National Health Interview Survey, a stratified random probability sample representative of the U.S. civilian noninstitutionalized population Survey respondents aged 18 years and older who answered questions on digestive conditions and self reported irritable bowel syndrome were asked questions about self-perceived health status and activities limitation. Health Results: As previously reported, weighted estimates indicate that over 6 million Americans seffreport that they have irritable bowel syndrome, and 3 million have lBSdike symptoms matching Rome II diagnostic criteria. Tire nrean health and activities limitation index scores of these groups were 0.74 (0.24 S.D.), and 0.75 (0.25 S.D.) for respondents with serf-reported and IBS-like synlptoms, respectively. Conclusion: Health and activity limitation index scores were affected by demographic characteristics (age, education, and family" income) of the involved population When severity of illness variable were considered, the denrographic variables remained significant predictors in the model, however to a lower degree. In the population of patients with self-reported IBS, the addition of severity variables did not appear to improve the model for predicting HAbex scores. Conversely, in respondents with IBS-fike symptoms, who did not self-report IBS, severity variables (physician visit for diarrhea, number of days with pain and activity limitation due to pain) appeared to [email protected] improve tire predictive model


Heahh-related Quality of Life in Patients Attending a Gastroenterology" Outpatient Clinic: Functional Disorders Versus Organic Diseases Magmls Simren Jan Svedlund, Iris Posserud less[ca Fem,amder, Einar S Bjornsson, Hasse Abrahamsson Health-related quality of life (HRQOL) is wldely perceived to be a useful concept in clinical reseanh that enlphasizes ~he patient's perspective of the impact of the illness. Several gastrointestinal (Gi) disorders have profound eifects on the HRQO[., but direct comparisons between functional GI disorders and organic GI diseases are lacking METHODS We invited 600 consecutive patients attending a uniw.'rsity hospitabbased GI outpatient clinic to participate in the study They completed validated HRQOL instruments (SF-36 and Psychohigioal General Well-Being index (PGWB)) and rated thair Gi synlptoms using the Gastrointestinal Symptom Pating Scale (GSRS) 434 patients agreed to participat< 35 were later excluded due tu severe non-GI diseases. For the analyses we divided the matenal into 2 groups: patients vnth functional GI disorders (FG[D) and organic Gi diseases (OGID) (mainly inilamnlato U bowel disease) RESUI'{S: Completed qnestionnaires were evaluated in 112 patients w t h FG1D (86 females mean age 41 (20-73) years) and 287 w7th OGID (159 temales mean age 46 (20-83) years) Compared w~th norm values on SF-36 and PGWB both patients w i h FGiD and OGID demonstrated profound reductions in tfRQOL The patients w~th FGiD demonstrated sigoihcant/y lower scores than patients with OG1D on 6/ 8 of the domains of 5F-36 role physical (p<0.05), bodily pain (p<0 0001), general health (p<005), vitality (p<0 01), racial iunctioning (p<0.05) and role emotional (p<0 01) They also had lowe~ scores on afi domains of PGWB: anxiety (p<0 0001), depressed nmod (p<0 01), positive well-being (p<0 05), serf-control (p<0 0001), general health (p<0.001) and vitality (p<0.05) According tu GSRS the patients with FGID had more severe reflux ( p < 0 01) abdominal pain ( p < 0 . 0 0 0 l ) , constipation ( p < 0 0001) and indigestion (p<0 000it, but the severity of diarrhea did not difler between the goups. All domains on the HRQOL instruments were reduced with increasing seventy of gastrointestinal symptoms. Of special interest patients vwth nncroscopio colitis (n= 22) demonstrated reductions in HRQOL to a similar degree as patients with FGID CONCLUSION: Gastrointestinal disorders fiave profimnd effects on ttRQOL and the impact seeras to be greater in patients with hinctional G[ disorders as compared with organic GI chseases. Tile reduction in HRQOL is strongly correlated with the severity of GI symptoms.


Tegaserod Treatment for IBS: A Model of Indirect Costs Dean Smith, Joshua J. Ofman, Victoria Barghout, Andrew D. Feld irritable Bowel Syndrome (iBS) has been associated with substantial time lost fi'om work (absenteeism) and reduced productivity at work (presenteeism), the indirect costs of illness of 1BS The objective of the present study was to moctel the redirect costs of IBS and the possible reduction in such costs for persons treated with tegasemd, a 5-HT4 receptor partial agonist approved for treatment of women with 1BS whose primary bowel symptom is constipauon. An economic model was developed based upon wages, prevalence of iBS, work loss and treatment efficacy Wages came from the US Depamnem of Labor. Prevalence and absenteeism associated with IBS were based on data from tile US Householder Survey (Drossman et ak, 1993). Presenteeism was not included in the US Honsehol&r Survey, but inlbrmation on this endpomt was collected in a survey of 1BS patients and estimated to be 44% of absenteeism (Hahn et al, 1999) A Phase III chilioal trial of tegaserod with Subjects Global Assessment of Relief as fhe primary endpoint found that 38% of female patients with IBS experienced relief with tegaserod (Mueller-Lissner, et al. 2001). it was assumed, that patients whose 1BS symptoms were relieved with tegaserod regained 75% of the productivity lost due to iBS The model extends the 12 week trial [or a year. The model uses these data for base-case results and permits user nrput of alternative parameter values and assumptions to present case-specific results and sensitivity analyses. For an employed population of 10,000 persons with the US average age/gender distribution, lost productivity due to IBS would total $375,903. In the base-case, the model pred*cts that 7.9% of these work day's lost due to IBS would be regained following successi\d treatment with tegaserod. Such therapy would cost $27,190, but return $29,543 of the lost wages for a net cost-savings of $2,353. The threshold for cost-savings were 23% for efficacy"and 70% for return to work. Under a series of assumptions on wages, epidemiology, therapy and costs, results suggest that treatment of IBS with tegaserod may" be cost-effective in reducing redirect cost under a variety of scenarios. However, further clinical validation of this model is warranted.


Impact of Irritable Bowel Syndrome on Worker Productivity" in an Employed U.S. Population Bonnie B. Dean, Daniel Agnilar Micbelle Dylan Victoria Barghout, Fende Frech David Groves, Joshua J Otman Purpose: Our obiective was to assess the impact of irritable bowd syndrome (IBS) on workrelated productivity in a U S employed population Methods: A [email protected], voluntary sureey was administered to [ha workforce of a US bank with employees in 4 states (MI, CA, TX, and FL) An initial survey collected sociodemographics, gastrointestinal (Gi) symptom profile,

AGA Abstracts