- Email: [email protected]
Cross-cultural variation in disease-related concerns among patients with inflammatory bowel disease
THE AMERICAN JOURNAL OF GASTROENTEROLOGY © 2001 by Am. Coll. of Gastroenterology Published by Elsevier Science Inc.
Vol. 96, No. 6, 2001 ISSN 0002-9270/01/$20.00 PII S0002-9270(01)02441-8
Cross-Cultural Variation in Disease-Related Concerns Among Patients With Inflammatory Bowel Disease Susan Levenstein, M.D., Zhiming Li, M.D., Sven Almer, M.D., Antonio Barbosa, M.D., Ph.D., Patrick Marquis, M.D., Gabriele Moser, M.D., Ami Sperber, M.D., Brenda Toner, Ph.D., and Douglas A. Drossman, Ph.D., F.A.C.P. Gastroenterology Division, San Camillo-Forlanini Hospital, Rome, Italy; University of North Carolina, Chapel Hill, North Carolina; Division of Gastroenterology and Hepatology, IHM, Linko¨ping University, Linko¨ping, Sweden; Hospital de Santa Maria, Lisbon, Portugal; MAPI, Lyons, France; Clinic of Internal Medicine IV, University Hospital of Vienna, Vienna, Austria; Ben Gurion University of the Negev, Beer Sheva, Israel; and Clarke Institute of Psychiatry, University of Toronto, Toronto, Canada
OBJECTIVE: The aim of this work was to study cross-cultural variations in the impact of inflammatory bowel disease (IBD) on health-related quality of life by an international comparison of disease-related concerns. METHODS: Item and factor scores on the Rating Form of Inflammatory Bowel Disease Patient Concerns and overall mean concern levels were compared by analysis of variance among 2002 IBD patients in eight countries. RESULTS: The overall level of concern varied from 51 out of 100 in Portugal to 19 in Sweden, with intermediate scores for Italy (43), Canada (40), United States (39), France (39), Austria (33), and Israel (25). Having surgery, an ostomy, the uncertain nature of the disease, and medication side effects were each rated among the first five in importance in six countries. Other items varied considerably. For example, concern regarding pain and suffering was high in Israel and low in Portugal, whereas concern over developing cancer was low in Italy. Concern over financial issues and access to high-quality health care were inversely associated with measures of national economic prosperity. CONCLUSIONS: 1) Cross-cultural comparisons of patient concerns related to IBD are feasible using translated scales. 2) Reporting tendencies vary greatly; within Europe, patients from southern countries report greater overall concern. 3) The complications and the variable evolution of disease elicit general concern, but the importance of specific issues varies among countries. 4) The reasons for national differences may have social, cultural, and/or economic determinants with relevance to the patient-physician relationship, patient education, and therapeutic decision making. (Am J Gastroenterol 2001;96:1822–1830. © 2001 by Am. Coll. of Gastroenterology)
Zhiming Li is currently affiliated with Pharmacia and Upjohn, Kalamazoo, Michigan.
INTRODUCTION There has been increasing interest during recent years in the impact of chronic medical conditions on patients’ global well-being or health-related quality of life (1), with recognition that symptoms and objective measures of pathophysiology are inadequate to describe the many-faceted nature of the disease experience. Under the assumption that one of the most important aims of medical treatment in chronic disease should be to optimize the overall quality of the life of individuals who must live with illness, this trend has led to questioning of traditional measures of disease severity and to development of a new body of quality of life instruments. Health-related quality of life may vary greatly between patients with a given level of objective pathology because patients differ in social context, life stage, values, resources, priorities, and cognitive/emotional reactivity. Just as these characteristics vary among individuals, they may also vary among populations according to differences in cultural heritage, value systems, family structure, medical systems, values regarding the communication of illness, or other factors. The impact of a given disease may, therefore, vary systematically between one country and another, even if its biological behavior is uniform. Cross-cultural studies of health-related quality of life have the potential to shed light in two directions: on disease processes and on cultural phenomena. The development of broadly applicable quality of life instruments such as the Sickness Impact Profile (2) and the SF-36 (3) has made it feasible to make comparisons across disease states as well as across cultural settings. Another general category of instruments, disease-specific measures, may provide a clearer, more clinically relevant, and more subtly nuanced understanding of the patient’s condition, though without the capacity to compare patients with different diseases. Several disease-specific instruments have been developed for inflammatory bowel disease (IBD); a prominent example is the Inflammatory Bowel Disease Questionnaire (4), a
AJG – June, 2001
32-item scale that covers symptoms, physical functioning, and psychological state. Another disease-specific instrument, the Rating Form of Inflammatory Bowel Disease Patient Concerns (RFIPC), represents an unusual approach to assessing health status. Because it examines diseaserelated concerns and worries, it is explicitly sensitive to the needs, perceptions, and priorities of individual patients (5). The RFIPC, even more than most other health-related quality of life instruments, operates in a realm where psychological and cultural factors may be as important as pathological anatomy and symptoms. As such, it is a particularly interesting tool to explore cross-cultural differences in disease experience. Though it was originally developed in the United States, the RFIPC has been translated into several languages and administered internationally. The present analyses were designed to compare RFIPC results obtained among patients in different countries with the aim of identifying cross-cultural patterns in the impact of IBD on patients.
MATERIALS AND METHODS Measures and Patients The RFIPC covers 25 potential issues related to having IBD (5). The patient is told, “Patients often develop concerns or worries related to their disease,” and is then asked, “Because of your condition, how concerned are you with . . .?.” Items are rated from 0 to 100 on a visual analog scale from “not at all” to “a great deal,” specifying the time scale of “how you feel today.” The development and validation of the RFIPC have been described in detail (5, 6). Briefly, a 21-item scale based on issues gleaned from concerns expressed by patients spontaneously to their physician and in focused videotaped interviews was validated on 150 patients with ulcerative colitis and Crohn’s disease. It proved reliable and internally consistent, and correlated modestly with physician estimates of current disease activity (6). The resultant scale, slightly changed in wording and with four additional items, was then studied among a large national sample of individuals in an IBD patient organization (7). High RFIPC scores were associated with low functional status as measured by the Sickness Impact Profile (5), poorer psychological function, and lower socioeconomic status (7); all these associations were again moderate, as is desirable for a measure designed to tap the patient’s perceptions. Later studies have shown that the measure is responsive to changes related to medical and surgical treatment (8 –10). During the early 1990s, the questionnaire was translated into German, French, Hebrew, Italian, Portuguese, and Swedish using translation-back-translation methodology: items were translated from English by a native speaker of the other language, and the resulting items were then backtranslated into English by a native speaker of English unfamiliar with the original questionnaire. The original and the back-translated English items were compared to detect is-
Cross-Cultural Variation in IBD
1823
sues requiring resolution (11), with repetition of the procedure on any new item versions that were generated. Formal assessment of psychometric characteristics and convergent/ divergent validity was carried out in most countries, and has been published in several cases (12–14). The present analyses compare RFIPC results obtained between 1991 and 1996 on IBD patients in Austria, Canada, France, Israel, Italy, Portugal, and Sweden; some of the national series have been published previously (12–15). Consecutive consenting patients, from sources that varied with local needs and resources, were studied in each country. Subjects’ age, sex, diagnosis (ulcerative colitis or Crohn’s disease), hospitalization status, disease duration, and educational achievement were recorded in all cases. To put the results in context, several socioeconomic parameters that were publicly available in comparable form were obtained for each country: adult literacy rates, Gross National Product in U.S. dollars per capita, health care expenditure as a percentage of Gross National Product, all as of the early 1990s (16), and life expectancy at birth and unemployment rates as of 1996 –1997 (17). The health care system in each country was categorized as providing or not providing universal health care that is free at the point of delivery. Statistical Methods Individual item scores and overall mean concern level or sum score (the mean of all item scores), each on a scale of 0 –100, were compared across countries by analysis of variance (ANOVA) using SAS software (SAS Institute, Cary, NC). To adjust for any differences in response tendencies, comparisons were also performed on item scores standardized by the overall concerns level in each country of administration (T scores), a procedure which results in an identical adjusted mean concern level of 50 for each country. The value of F in the ANOVA was used as a measure of variability, with F ⬍ 2 (p ⬎ 0.1) taken to indicate similarity of scores in all countries and F ⬎ 9 (p ⬍ 0.0001) taken to indicate a wide range of variability. Factor analysis of the U.S. national sample had yielded four factors for the RFIPC (7): disease impact (based on items 1, 2, 3, 4, 7, 8, 9, 10, 19, 20, 23, 24, 25), intimacy (items 12, 21, 22), complications (items 5, 6, 16, 17), and stigmata (items 11 and 18); items 13, 14, and 15 did not enter any factor. For the present study, factor scores were derived for every country’s patients by averaging mean scores on each of the four item sets. These scores were then converted into T scores, to adjust for response tendencies, before being compared among countries by ANOVA. Associations between the RFIPC and national economic and health measures were evaluated using Spearman correlation coefficients. Comparability of patient demographics was assessed using ANOVA (for age) and 2 test (for gender). No adjustment was made for multiple comparisons
1824
Levenstein et al.
AJG – Vol. 96, No. 6, 2001
Table 1. Demographic and Disease Characteristics of Subjects With IBD Who Completed The Rating Form of IBD Patient Concerns in Eight Countries Crohn’s Disease
Number* Austria Canada France Israel Italy Portugal Sweden United States Total
105 167 189 140 102 78 230 991 2002
72 75 93 0 60 27 37 671 57%
Ulcerative Colitis 33 77 96 140 41 51 193 320 43%
Female Gender
Age (yr)
Duration (yr)
% Inpatients
Education
62% 57% 47% 54% 48% 54% 47% 60% 56%
36 35 37 39 36 40 40 43 41
6.0 7.8 6.5 11.9 9.9 8.0 11.0 15.2 12.3
0% 0% 23% 0% 67%‡ 0% 10% 0% 7%
22% grads† 14 yr 37% grads† 12.5 yr 70% grads† 6.6 yr 11.4 yr 15.0 yr
* Total subject number includes some patients with indeterminate colitis, not included in either of the following two columns. † Graduated from secondary school (varies in meaning among countries). ‡ Most of the inpatients in the Italian sample were hospitalized for routine checkups, not for active disease.
because the purpose of this study was descriptive rather than hypothesis-testing.
RESULTS Comparability of Populations A total of 2002 IBD patients who had completed the RFIPC in eight countries were included in the analyses (Table 1). When demographic data were compared across samples, mean age proved to range from 34.5 to 45.2 yr (p ⬍ 0.0001), and the proportion of women from 47.0 to 61.9% (p ⬍ 0.001). The statistical significance of the differences in age and gender was not unexpected because of the large size of the total sample. However, the relatively narrow ranges suggest that the samples were still comparable. Long-standing disease predominated in all countries, though mean disease duration varied. All eight samples included patients with ulcerative colitis, and all except Israel included some Crohn’s disease patients as well. Education differed more among countries than did other demographic characteristics, with a particularly low level in Portugal, though structural differences in the educational systems make direct comparisons problematic. The subject population consisted chiefly of patients who were in remission or mildly ill; the apparent severity of the Italian sample, with its high proportion of inpatients, is misleading, because this team hospitalized
their IBD patients routinely for periodic checkups— 48% of the Italian inpatients had severity ratings of inactive or mild according to the classic British classification (18). Selected national characteristics are summarized in Table 2. Israel and Portugal have lower per capita Gross National Product and lower levels of health care spending than the other countries. The United States has particularly high health care expenditure and is the one country without universal health care, and infant mortality rates vary by more than 2-fold. Variation in Sum Scores The overall absolute level of patient concern as reflected in the raw all-item mean or “sum score” ranged from a high of 51 out of 100 in Portugal to a low of 19 in Sweden, with intermediate scores for Italy (43), Canada (40), United States (39), France (39), Austria (33), and Israel (25). To make it easier to compare the perception of specific disease-related concerns in different countries, item means were standardized to T scores to adjust for the large intercultural differences in overall sum scores. Examination of adjusted item scores (Fig. 1) shows that concerns about intimacy, loss of sexual drive, feeling alone, ability to achieve one’s full potential, producing unpleasant odors, and dying early were felt to a similar degree by patients in all countries relative to their overall level of reported con-
Table 2. National Demographic Characteristics of Countries With Participating Subjects
Austria Canada France Israel Italy Portugal Sweden United States Year of data
GNP ($) per Capita
Literacy Rate of Adults
Health Expenditure as % of GNP
Unemployment (%)
Universal, Free Health Care?
Life Expectancy
Infant Mortality/1000
22,380 20,710 22,260 13,220 20,460 7,450 27,010 23,240 1992
99.0 99.0 99.0 95.0 97.4 86.2 99.0 99.0 1991
8.5 9.1 9.9 4.2 8.3 4.6 8.8 12.7 1992*
5.4 8.1 9.0 9.8 11.0 5.5 1.6 5.5 1996–97
Yes Yes Yes Yes Yes Yes Yes No 1990–99
77.2 79.0 78.4 78.2 78.3 75.4 79.1 76.0 1996–97
5.2 5.7 5.8 8.3 6.5 7.0 3.9 6.6 1991
GNP ⫽ Gross National Product. * 1989 –90 for United States and Canada.
AJG – June, 2001
Cross-Cultural Variation in IBD
1825
Figure 1. Rating Form of IBD Patient Concerns: adjusted (T) scores on each item for IBD patients in eight countries.
cern (F ⬍ 2, p ⬎ 0.10). In contrast, there was wide variation among countries (F ⬎ 9, p ⬍ 0.0001) in concerns regarding energy level, access to quality medical care, passing the disease onto others, pain or suffering, feeling dirty or smelly, and the uncertain nature of the disease. Factor Scores Raw scores on the four RFIPC factors followed patterns similar to those of the overall mean item scores. When the factor scores were adjusted for reporting tendencies by conversion into T scores (Fig. 2), patients in all countries proved to have similar and moderate levels of concern regarding the impact of disease (mean T score ⫽ 51.5, F ⫽ 0.52, p ⫽ 0.82 by ANOVA), and similar low levels of concern regarding intimacy (mean T score ⫽ 46.3, F ⫽ 0.72, p ⫽ 0.65). There was more diversity regarding the complications of disease, which was rated the highest of the four factors overall (mean T score ⫽ 54.1, F ⫽ 2.29, p ⫽ 0.03): the French and Austrian patients showed relatively high and the Portuguese relatively low concern. Concern over “stigmata” was even more varied (mean T score ⫽ 48.4, F ⫽ 3.59, p ⬍ 0.001) with Portuguese patients particularly concerned and French patients unconcerned. Patterns of Concerns A. BY RANK ORDER. When the items were ranked among all 2002 patients, the highest level of concern was
found to be evoked by the uncertain course of IBD, followed by adverse effects of medication, having an ostomy bag, energy level, and the need for surgery. The five items that elicited the lowest levels of concern were the ability to have children, being treated as different, transmitting the disease to others, feeling smelly or dirty, and sexual performance. Whereas the rank order was broadly comparable among countries—for example, uncertain disease course, adverse effects of medications, and ostomies were ranked among the first nine concerns by patients in every national sample— several items (especially pain and suffering, cancer, and access to medical care) were ranked quite differently in different countries. B. BY T SCORES. When item scores were adjusted to T scores (Fig. 1), the French sample proved to have a distinctive pattern marked by an extreme range of responses. As compared with patients in other countries, they reported particularly high concern regarding pain and suffering, incontinence, needing surgery, energy level, body feelings, and the uncertain course of the disease, and particularly low concern regarding feeling dirty or smelly, sexual performance, having children, being treated as different, and access to quality medical care. Portuguese patients also had a relatively broad range of responses. They felt particularly high concern regarding
1826
Levenstein et al.
AJG – Vol. 96, No. 6, 2001
Figure 2. Rationg Form of IBD Patient Concerns: adjusted (T) scores on each factor for IBD patients in eight countries.
feeling dirty or smelly, passing the disease onto others, bad smells, being treated as different, and access to quality medical care, and particularly low concern regarding the financial impact of the disease, pain and suffering, having surgery, the uncertain course of the disease, and effects of medication. Italian patients reported a particularly low level of concern over developing cancer, whereas Austrians were particularly concerned about needing an ostomy bag and medication effects and little concerned about financial implications. When compared only with other ulcerative colitis patients, the Israeli sample appeared to be particularly concerned about pain and suffering, and particularly unconcerned about needing an ostomy. The American, Canadian, and Swedish samples did not stand out for high or low responses on any item. National Sociodemographic Characteristics Overall mean concern level (sum score) and the individual items regarding financial worries and access to quality med-
ical care were examined with regard to sociodemographic characteristics of the various countries (Table 3). No associations observed reached levels of statistical significance. In the strongest associations (0.05 ⬍ p ⬍ 0.10), concern over access to care appeared to be inversely associated with health care expenditure and with literacy rates. The sum score was modestly associated with Gross National Product per capita, life expectancy, and literacy rate, and financial concern with unemployment rate and life expectancy. All countries but one offer free universal health care to patients with chronic disease, so this variable was not analyzed. Specific Diagnoses Among the seven countries that had administered the RFIPC to both ulcerative colitis and Crohn’s disease patients, the unadjusted mean concerns sum score was higher in Crohn’s disease (40.1) than in ulcerative colitis (34.4). Ulcerative colitis patients expressed greater concern than Crohn’s disease patients about developing cancer and about dying pre-
Table 3. Association of Rating Form of IBD Patient Concerns With National Economic and Health Parameters Correlation With Aspect of Concern
Per Capita GNP
Health Spending
Adult Literacy
Unemployment
Life Expectancy
Infant Mortality
Overall concern level Financial Access to care
⫺0.57 ⫺0.17 ⫺0.40
⫺0.05 ⫺0.07 ⫺0.72*
⫺0.41 ⫺0.06 ⫺0.63*
0.35 0.50 ⫺0.14
⫺0.43 0.55 ⫺0.40
0.38 0.12 0.11
GNP ⫽ Gross National Product. * 0.05 ⬍ p ⬍ 0.10 by Spearman rank correlation coefficient, adjusted for ties.
AJG – June, 2001
Cross-Cultural Variation in IBD
Table 4. Raw Scores on the Rating Form of IBD Patient Concerns: Ulcerative Colitis and Crohn’s Disease Patients From Eight Countries Mean Raw Score Item 1: Financial difficulties 2: Pain or suffering 3: Ability to achieve full potential 4: Loss of bowel control 5: Developing cancer 6: Dying early 7: Being a burden (or depending) on others 8: Attractiveness 9: Feeling alone 10: Feeling out of control 11: Feeling “dirty” or “smelly” 12: Ability to perform sexually 13: Ability to have children 14: Passing the disease onto others 15: Being treated as different 16: Having surgery 17: Having an ostomy bag 18: Producing unpleasant odors 19: Energy level 20: Feelings about your body 21: Intimacy 22: Loss of sexual drive 23: Access to quality medical care 24: Uncertain nature of disease 25: Effects of medication
All Ulcerative Crohn’s Patients Colitis Disease 28.3 38.4 41.9
23.4 33.6 37.2
33.6 44.9 47.9
44.2 41.9 32.0 43.1
44.3 48.1 33.2 38.8
47.2 37.3 31.8 47.1
31.0 27.9 34.1 25.0 26.0 18.7 23.0
27.4 24.6 32.6 24.0 24.1 17.5 19.4
35.6 31.8 37.4 27.7 29.8 23.4 26.2
20.5 49.7 51.1 39.4 50.1 38.1 30.7 29.2 32.6 53.5 51.6
17.9 46.9 50.6 35.5 43.2 34.7 28.3 26.7 31.1 49.4 49.1
24.5 52.9 55.2 44.8 59.2 43.5 35.0 33.7 35.2 60.0 55.2
maturely, but less concern related to all other specific areas of concern (Table 4). The patterns by country of specific adjusted item scores among ulcerative colitis patients was similar (Fig. 3) to the pattern among the entire IBD population. Among Crohn’s disease patients (Fig. 4), the Portuguese appeared to have a strong distinctive pattern, but the size of the Portuguese Crohn’s disease sample was small (N ⫽ 27), and the SD for specific items was correspondingly large.
DISCUSSION Perhaps the most striking finding of this cross-cultural study of the RFIPC was that the overall level of concern reported by IBD patients regarding the impact of their disease varied by more than 2-fold among countries. Artifactual explanations for this wide variability, notably translation difficulties, must be taken into consideration. Careful employment of the translation-back-translation method does not preclude subtle differences between one language and another, especially of the key phrases in the instructions: “concerns or worries related to their disease,” and “how concerned are you with . . .?.” Another possible explanation for the range in concern
1827
levels could be sampling differences in patient populations. Educational level did vary among the countries but did not seem to have an impact: concern levels were high not only among the Portuguese, who had the lowest educational attainment, but also in the very well-educated Canadian and U.S. samples. Different anatomical distributions of disease among samples could theoretically have influenced the results, but the similar raw levels of concern in Italy and France, which were known to have markedly different patterns of disease extent (24% vs 67% of ulcerative colitis patients had left-sided disease, 38% vs 14% of Crohn’s disease patients had isolated ileitis), suggest such an influence is limited. Differences in age and gender could also have affected patients’ concerns, though these two variables were relatively comparable among countries. Unfortunately, none of these factors could be controlled in our analysis because of the limitations of the data; the analyses in this report were based on findings previously obtained independently by researchers in the various countries, precluding adjustment of intercountry differences for patient or disease characteristics. The present findings must therefore be considered exploratory, aimed at generating hypotheses that could open the door to future investigation. Cultural characteristics may play a role in the high intercountry variability in overall concerns level. Examination of the sum score among countries reveals a geographic trend with scores roughly aligned, within Europe, along a northsouth gradient: Italy and Portugal, the two most “Mediterranean” countries, have the highest scores and Sweden, the most “nordic,” has the lowest. To some extent, this northsouth gradient corresponds to trends in national prosperity, educational level, and health expenditures, but it also is likely to reflect some more elusive cultural trends. Israel, whose sum score was relatively low, is a Mediterranean country by geographic location and in some cultural respects, and thus might be expected to have results similar to those of southern European countries. However, 32.4% of the Israeli patients were born in Northern Europe, Eastern Europe, or America, as were the parents of most of the 46.8% native-born Israelis in the sample, making it difficult to characterize this unusual population group definitively. National characteristics of the physician-patient relationship could be relevant, such as the relative prominence of a paternalistic style versus emphasis on patient education and self-management. A high degree of doctor-patient continuity and ease of physician access, likely to vary among institutions as well as among countries, could also allay concern, contributing for example to the particularly low mean score in Sweden. Previous studies have found (13) that IBD patients who perceive themselves as better informed about their disease have lower levels of concerns, and that there is generally an inverse correlation between degress of concern and educational level (5). Our U.S. population is likely to be even better informed than the average American patient with IBD, however, by virtue of having been drawn from an IBD
1828
Levenstein et al.
AJG – Vol. 96, No. 6, 2001
Figure 3. Rating Form of IBD Patient Concerns: adjusted (T) scores on each item for ulcerative colitis patients in seven countries.
self-help group, so that their relatively high concern level with respect to other countries suggests that knowledge may not always decrease concern. Although overall levels of reported concern varied sharply from country to country, the patterns of relative importance assigned to specific concerns and factors were relatively uniform, suggesting that IBD patients from Western countries tend to share common concerns about their disease regardless of their specific country of residence. In particular, there was high concern regarding the uncertain course of the disease, the possibility of ostomy surgery, the adverse effects of medications, and about the complications of disease in general. These IBD patients also broadly shared a relative lack of concern over a series of interpersonal issues related to their disease, such as difficulty having children (not unexpected in a population with a mean age of 41) or being treated as “different.” There were, however, several items that proved to be of considerable concern to patients in some countries but not in others. It is tempting to speculate on possible reasons for some of these differences among countries, though caution is obviously required. For example, Israeli ulcerative colitis patients were particularly concerned about pain and suffering, consistent with older studies of pain behavior among Jewish patients in the United States (19). Other findings are also open to possible cultural interpretations, such as the low
concern over impairment of sexual performance among the French, the low concern of Portuguese subjects over the adverse effects of medication (and their high concern with social stigma), and the high concern over ostomy surgery in Austria. Differences among medical approaches may account for some of the observed variation: concern over ostomy surgery could be related to the local rates of this surgical procedure, and concern over developing cancer may be lower in countries (which, in the experience of the authors, includes Italy) where many physicians downplay cancer risks in their explanations to patients. Finally, in some countries, patients may find responding to questions on sexual performance and intimacy more difficult than the less “personal” questions, which could affect the validity of the responses. Concern over access to quality medical care, which was strikingly high in Portugal and low in France, seems to vary less with the universal availability of medical care or with the direct costs of treatment than with general economic conditions in the various countries and, possibly, with the quality of those resources that are available. The absence of universal free health care in the United States seemed to have no impact on the level or pattern of concerns. Relative concern about the financial repercussions of disease was highest in Israel and Canada and lowest in Austria and Portugal, showing a surprising absence of association
AJG – June, 2001
Cross-Cultural Variation in IBD
1829
Figure 4. Rating Form of IBD Patient Concerns: adjusted (T) scores on each item for Crohn’s disease patients in six countries.
either with out-of-pocket health care costs or with the overall state of the economy. Perceived difficulties in making a living because of disability could be another relevant factor. Considering the substantial differences between ulcerative colitis and Crohn’s disease in their symptoms and course, there was remarkably little difference in the pattern of disease-related concerns among patients with the two diagnoses. The ability of the RFIPC to reveal cross-cultural differences of potential clinical and theoretical importance confirms the value of this innovative approach to quality of life assessment. Previous studies have similarly demonstrated differences in the pattern of disease-related concerns between IBD patients who seek psychological counseling and those who do not (9), and between African-American and other American Crohn’s disease patients (20). In combination with the studies of validity and of sensitivity to changes in disease status described under Materials and Methods, these findings suggest that the RFIPC is an effective and responsive measure of quality of life or health status. The notion of assigning group differences in behaviors, diseases, or questionnaire responses to cultural characteristics is more complex than might be at first apparent. Culture may be viewed as a combination of many factors, some obvious and others obscure, which form the unique characteristics of a specific ethnic population. Such factors may
include—among others— genetics, education, values, religion, lifestyle, and habits. The more a study attributes crosscultural differences to specific factors, the smaller the residual effect of “culture” in explaining the differences. In other words, identification of specific components of importance in a given setting will help the “black box” of culture to become less black. From this viewpoint, the concept of culture may be seen as a tool to help us pin down the specific relevant factors. In conclusion, international variations in health-related quality of life among patients with a common disease have been little examined. The present results suggest that crosscultural comparisons of patient concerns related to IBD are feasible using translated scales, and support the validity of the RFIPC for this purpose. There are large differences in reporting tendencies among countries, and within Europe patients from southern countries tend to report greater overall concern than those from the north. Patients agree across cultures on the relative importance of some concerns but diverge widely on others, with the variable course and outcome of IBD and the prospect of living with an ostomy of particularly high concern to patients in many countries. Semantic or methodological causes may contribute to these differences, but it is likely that the reasons for differences in specific concerns items or in overall concern level also have social, cultural, and/or economic determinants that are open
1830
Levenstein et al.
to further study. These findings may have bearing on the patient-physician relationship, patient education, and therapeutic decision-making.
ACKNOWLEDGMENTS We thank Vilma Varvo, Henrik Hjortswang, Anton Steinberg, Wolfgang Tillinger, and Elizabeth de Rooy for their vital contributions to this project. This study was supported in part by the “Stiftelsen for psykosomatisk och klinisk forskning,” grant 1105/1991, the County Council of Ostergotland, grants 92/017 and 95/042, and the Swedish Society of Medicine, grant 1995/22.0.
AJG – Vol. 96, No. 6, 2001
8.
9. 10.
11. 12.
Reprint requests and correspondence: Susan Levenstein, M.D., Via del Tempio 1A, 00186 Rome, Italy. Received Aug. 21, 2000; accepted Jan. 25, 2001. 13.
REFERENCES 1. WHOQOL Group. The World Health Organization quality of life assessment (WHOQOL): Position paper from the World Health Organization. Soc Sci Med 1995;10:1403–9. 2. Bergner M, Bobbitt R, Carter W, et al. The Sickness Impact Profile: Development and final revision of a health status measure. Med Care 1981;19:787– 805. 3. Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36) I. Conceptual framework and item selection. Med Care 1992;30:473– 83. 4. Irvine EJ, Feagan B, Rochon J, et al. Quality of life: A valid and reliable measure of therapeutic efficacy in the treatment of inflammatory bowel disease. Gastroenterology 1994;106:287– 96. 5. Drossman DA, Leserman J, Li Z, et al. The rating form of IBD patient concerns: A new measure of health status. Psychosom Med 1991;13:701–12. 6. Drossman DA, Patrick DL, Mitchell CM, et al. Health-related quality of life in inflammatory bowel disease: Functional status and patient worries and concerns. Dig Dis Sci 1989;34: 1379 – 86. 7. Drossman DA, Leserman J, Mitchell CM, et al. Health status
14.
15.
16. 17. 18. 19. 20.
and health care use in persons with inflammatory bowel disease: A national sample. Dig Dis Sci 1991;36:1746 –55. Provenzale D, Shearin M, Phillips-Bute B, et al. Health-related quality of life after ileoanal pull-through: Evaluation and assessment of new health status measures. Gastroenterology 1997;113:7–14. Yazdanpanah Y, Klein O, Gambiez L, et al. Impact of surgery on quality of life in Crohn’s disease. Am J Gastroenterol 1997;92:1897–900. Tillinger W, Mittermaier C, Lochs H, Moser G. Health-related quality of life in patients with Crohn‘s disease: Influence of surgical operation — a prospective trial. Dig Dis Sci 1999; 44:932– 8. Sperber AD, Devellis RF, Boehlecke B. Cross-cultural translation: Methodology and validation. J Cross-Cultural Psych 1994;25:501–24. Colombel J-F, Yazdanpanah Y, Laurent F, et al. Qualite´ de vie dans les maladies inflammatoires chroniques de l’intestin: Validation d’un questionnaire et premie`res donne´es franc¸aises. Gastroenterol Clin Biol 1996;20:1071–7 (in French). Moser G, Tillinger W, Sachs G, et al. Disease-related worries and concerns: A study on out-patients with inflammatory bowel disease. Eur J Gastroenterol Hepatol 1995;7:853– 8. Hjortswang H, Stro¨m M, Almeida RT, Almer S. Evaluation of the RFIPC, a disease-specific health-related quality of life questionnaire, in Swedish patients with ulcerative colitis. Scan J Gastroenterol 1997;32:1235– 40. Blondel-Kucharski F, Chircop C, Marquis P, et al. Quality of life in Crohn’s disease: A prospective, longitudinal study in 231 patients. Presented at Journe´es Francophones de Pathologie Digestive; March 22–24, 1999; Nantes, France. World Health Organization. Global health-for-all indicators. Available at: http://www.who.int/hst/hsp/a/country.htm. Accessed Feb. 17, 1999. Central Intelligence Agency. CIA Factbook. Available at: http://www.odci.gov/cia/publications/factbook. Accessed Feb. 17, 1999. Truelove SC, Witts LJ. Cortisone in ulcerative colitis: Preliminary report on a therapeutic trial. Br Med J 1954;2:375– 8. Zborowski M. Cultural components in responses to pain. J Soc Issues 1952;8:16 –30. Eisen G, Strauss W, Sandler R, et al. Health status in Crohn’s disease: Comparison of the Rating Form of IBD Patient Concerns in African and non-African Americans. Gastroenterology 1996;110:A15 (abstract 1157).
Vol. 96, No. 6, 2001 ISSN 0002-9270/01/$20.00 PII S0002-9270(01)02441-8
Cross-Cultural Variation in Disease-Related Concerns Among Patients With Inflammatory Bowel Disease Susan Levenstein, M.D., Zhiming Li, M.D., Sven Almer, M.D., Antonio Barbosa, M.D., Ph.D., Patrick Marquis, M.D., Gabriele Moser, M.D., Ami Sperber, M.D., Brenda Toner, Ph.D., and Douglas A. Drossman, Ph.D., F.A.C.P. Gastroenterology Division, San Camillo-Forlanini Hospital, Rome, Italy; University of North Carolina, Chapel Hill, North Carolina; Division of Gastroenterology and Hepatology, IHM, Linko¨ping University, Linko¨ping, Sweden; Hospital de Santa Maria, Lisbon, Portugal; MAPI, Lyons, France; Clinic of Internal Medicine IV, University Hospital of Vienna, Vienna, Austria; Ben Gurion University of the Negev, Beer Sheva, Israel; and Clarke Institute of Psychiatry, University of Toronto, Toronto, Canada
OBJECTIVE: The aim of this work was to study cross-cultural variations in the impact of inflammatory bowel disease (IBD) on health-related quality of life by an international comparison of disease-related concerns. METHODS: Item and factor scores on the Rating Form of Inflammatory Bowel Disease Patient Concerns and overall mean concern levels were compared by analysis of variance among 2002 IBD patients in eight countries. RESULTS: The overall level of concern varied from 51 out of 100 in Portugal to 19 in Sweden, with intermediate scores for Italy (43), Canada (40), United States (39), France (39), Austria (33), and Israel (25). Having surgery, an ostomy, the uncertain nature of the disease, and medication side effects were each rated among the first five in importance in six countries. Other items varied considerably. For example, concern regarding pain and suffering was high in Israel and low in Portugal, whereas concern over developing cancer was low in Italy. Concern over financial issues and access to high-quality health care were inversely associated with measures of national economic prosperity. CONCLUSIONS: 1) Cross-cultural comparisons of patient concerns related to IBD are feasible using translated scales. 2) Reporting tendencies vary greatly; within Europe, patients from southern countries report greater overall concern. 3) The complications and the variable evolution of disease elicit general concern, but the importance of specific issues varies among countries. 4) The reasons for national differences may have social, cultural, and/or economic determinants with relevance to the patient-physician relationship, patient education, and therapeutic decision making. (Am J Gastroenterol 2001;96:1822–1830. © 2001 by Am. Coll. of Gastroenterology)
Zhiming Li is currently affiliated with Pharmacia and Upjohn, Kalamazoo, Michigan.
INTRODUCTION There has been increasing interest during recent years in the impact of chronic medical conditions on patients’ global well-being or health-related quality of life (1), with recognition that symptoms and objective measures of pathophysiology are inadequate to describe the many-faceted nature of the disease experience. Under the assumption that one of the most important aims of medical treatment in chronic disease should be to optimize the overall quality of the life of individuals who must live with illness, this trend has led to questioning of traditional measures of disease severity and to development of a new body of quality of life instruments. Health-related quality of life may vary greatly between patients with a given level of objective pathology because patients differ in social context, life stage, values, resources, priorities, and cognitive/emotional reactivity. Just as these characteristics vary among individuals, they may also vary among populations according to differences in cultural heritage, value systems, family structure, medical systems, values regarding the communication of illness, or other factors. The impact of a given disease may, therefore, vary systematically between one country and another, even if its biological behavior is uniform. Cross-cultural studies of health-related quality of life have the potential to shed light in two directions: on disease processes and on cultural phenomena. The development of broadly applicable quality of life instruments such as the Sickness Impact Profile (2) and the SF-36 (3) has made it feasible to make comparisons across disease states as well as across cultural settings. Another general category of instruments, disease-specific measures, may provide a clearer, more clinically relevant, and more subtly nuanced understanding of the patient’s condition, though without the capacity to compare patients with different diseases. Several disease-specific instruments have been developed for inflammatory bowel disease (IBD); a prominent example is the Inflammatory Bowel Disease Questionnaire (4), a
AJG – June, 2001
32-item scale that covers symptoms, physical functioning, and psychological state. Another disease-specific instrument, the Rating Form of Inflammatory Bowel Disease Patient Concerns (RFIPC), represents an unusual approach to assessing health status. Because it examines diseaserelated concerns and worries, it is explicitly sensitive to the needs, perceptions, and priorities of individual patients (5). The RFIPC, even more than most other health-related quality of life instruments, operates in a realm where psychological and cultural factors may be as important as pathological anatomy and symptoms. As such, it is a particularly interesting tool to explore cross-cultural differences in disease experience. Though it was originally developed in the United States, the RFIPC has been translated into several languages and administered internationally. The present analyses were designed to compare RFIPC results obtained among patients in different countries with the aim of identifying cross-cultural patterns in the impact of IBD on patients.
MATERIALS AND METHODS Measures and Patients The RFIPC covers 25 potential issues related to having IBD (5). The patient is told, “Patients often develop concerns or worries related to their disease,” and is then asked, “Because of your condition, how concerned are you with . . .?.” Items are rated from 0 to 100 on a visual analog scale from “not at all” to “a great deal,” specifying the time scale of “how you feel today.” The development and validation of the RFIPC have been described in detail (5, 6). Briefly, a 21-item scale based on issues gleaned from concerns expressed by patients spontaneously to their physician and in focused videotaped interviews was validated on 150 patients with ulcerative colitis and Crohn’s disease. It proved reliable and internally consistent, and correlated modestly with physician estimates of current disease activity (6). The resultant scale, slightly changed in wording and with four additional items, was then studied among a large national sample of individuals in an IBD patient organization (7). High RFIPC scores were associated with low functional status as measured by the Sickness Impact Profile (5), poorer psychological function, and lower socioeconomic status (7); all these associations were again moderate, as is desirable for a measure designed to tap the patient’s perceptions. Later studies have shown that the measure is responsive to changes related to medical and surgical treatment (8 –10). During the early 1990s, the questionnaire was translated into German, French, Hebrew, Italian, Portuguese, and Swedish using translation-back-translation methodology: items were translated from English by a native speaker of the other language, and the resulting items were then backtranslated into English by a native speaker of English unfamiliar with the original questionnaire. The original and the back-translated English items were compared to detect is-
Cross-Cultural Variation in IBD
1823
sues requiring resolution (11), with repetition of the procedure on any new item versions that were generated. Formal assessment of psychometric characteristics and convergent/ divergent validity was carried out in most countries, and has been published in several cases (12–14). The present analyses compare RFIPC results obtained between 1991 and 1996 on IBD patients in Austria, Canada, France, Israel, Italy, Portugal, and Sweden; some of the national series have been published previously (12–15). Consecutive consenting patients, from sources that varied with local needs and resources, were studied in each country. Subjects’ age, sex, diagnosis (ulcerative colitis or Crohn’s disease), hospitalization status, disease duration, and educational achievement were recorded in all cases. To put the results in context, several socioeconomic parameters that were publicly available in comparable form were obtained for each country: adult literacy rates, Gross National Product in U.S. dollars per capita, health care expenditure as a percentage of Gross National Product, all as of the early 1990s (16), and life expectancy at birth and unemployment rates as of 1996 –1997 (17). The health care system in each country was categorized as providing or not providing universal health care that is free at the point of delivery. Statistical Methods Individual item scores and overall mean concern level or sum score (the mean of all item scores), each on a scale of 0 –100, were compared across countries by analysis of variance (ANOVA) using SAS software (SAS Institute, Cary, NC). To adjust for any differences in response tendencies, comparisons were also performed on item scores standardized by the overall concerns level in each country of administration (T scores), a procedure which results in an identical adjusted mean concern level of 50 for each country. The value of F in the ANOVA was used as a measure of variability, with F ⬍ 2 (p ⬎ 0.1) taken to indicate similarity of scores in all countries and F ⬎ 9 (p ⬍ 0.0001) taken to indicate a wide range of variability. Factor analysis of the U.S. national sample had yielded four factors for the RFIPC (7): disease impact (based on items 1, 2, 3, 4, 7, 8, 9, 10, 19, 20, 23, 24, 25), intimacy (items 12, 21, 22), complications (items 5, 6, 16, 17), and stigmata (items 11 and 18); items 13, 14, and 15 did not enter any factor. For the present study, factor scores were derived for every country’s patients by averaging mean scores on each of the four item sets. These scores were then converted into T scores, to adjust for response tendencies, before being compared among countries by ANOVA. Associations between the RFIPC and national economic and health measures were evaluated using Spearman correlation coefficients. Comparability of patient demographics was assessed using ANOVA (for age) and 2 test (for gender). No adjustment was made for multiple comparisons
1824
Levenstein et al.
AJG – Vol. 96, No. 6, 2001
Table 1. Demographic and Disease Characteristics of Subjects With IBD Who Completed The Rating Form of IBD Patient Concerns in Eight Countries Crohn’s Disease
Number* Austria Canada France Israel Italy Portugal Sweden United States Total
105 167 189 140 102 78 230 991 2002
72 75 93 0 60 27 37 671 57%
Ulcerative Colitis 33 77 96 140 41 51 193 320 43%
Female Gender
Age (yr)
Duration (yr)
% Inpatients
Education
62% 57% 47% 54% 48% 54% 47% 60% 56%
36 35 37 39 36 40 40 43 41
6.0 7.8 6.5 11.9 9.9 8.0 11.0 15.2 12.3
0% 0% 23% 0% 67%‡ 0% 10% 0% 7%
22% grads† 14 yr 37% grads† 12.5 yr 70% grads† 6.6 yr 11.4 yr 15.0 yr
* Total subject number includes some patients with indeterminate colitis, not included in either of the following two columns. † Graduated from secondary school (varies in meaning among countries). ‡ Most of the inpatients in the Italian sample were hospitalized for routine checkups, not for active disease.
because the purpose of this study was descriptive rather than hypothesis-testing.
RESULTS Comparability of Populations A total of 2002 IBD patients who had completed the RFIPC in eight countries were included in the analyses (Table 1). When demographic data were compared across samples, mean age proved to range from 34.5 to 45.2 yr (p ⬍ 0.0001), and the proportion of women from 47.0 to 61.9% (p ⬍ 0.001). The statistical significance of the differences in age and gender was not unexpected because of the large size of the total sample. However, the relatively narrow ranges suggest that the samples were still comparable. Long-standing disease predominated in all countries, though mean disease duration varied. All eight samples included patients with ulcerative colitis, and all except Israel included some Crohn’s disease patients as well. Education differed more among countries than did other demographic characteristics, with a particularly low level in Portugal, though structural differences in the educational systems make direct comparisons problematic. The subject population consisted chiefly of patients who were in remission or mildly ill; the apparent severity of the Italian sample, with its high proportion of inpatients, is misleading, because this team hospitalized
their IBD patients routinely for periodic checkups— 48% of the Italian inpatients had severity ratings of inactive or mild according to the classic British classification (18). Selected national characteristics are summarized in Table 2. Israel and Portugal have lower per capita Gross National Product and lower levels of health care spending than the other countries. The United States has particularly high health care expenditure and is the one country without universal health care, and infant mortality rates vary by more than 2-fold. Variation in Sum Scores The overall absolute level of patient concern as reflected in the raw all-item mean or “sum score” ranged from a high of 51 out of 100 in Portugal to a low of 19 in Sweden, with intermediate scores for Italy (43), Canada (40), United States (39), France (39), Austria (33), and Israel (25). To make it easier to compare the perception of specific disease-related concerns in different countries, item means were standardized to T scores to adjust for the large intercultural differences in overall sum scores. Examination of adjusted item scores (Fig. 1) shows that concerns about intimacy, loss of sexual drive, feeling alone, ability to achieve one’s full potential, producing unpleasant odors, and dying early were felt to a similar degree by patients in all countries relative to their overall level of reported con-
Table 2. National Demographic Characteristics of Countries With Participating Subjects
Austria Canada France Israel Italy Portugal Sweden United States Year of data
GNP ($) per Capita
Literacy Rate of Adults
Health Expenditure as % of GNP
Unemployment (%)
Universal, Free Health Care?
Life Expectancy
Infant Mortality/1000
22,380 20,710 22,260 13,220 20,460 7,450 27,010 23,240 1992
99.0 99.0 99.0 95.0 97.4 86.2 99.0 99.0 1991
8.5 9.1 9.9 4.2 8.3 4.6 8.8 12.7 1992*
5.4 8.1 9.0 9.8 11.0 5.5 1.6 5.5 1996–97
Yes Yes Yes Yes Yes Yes Yes No 1990–99
77.2 79.0 78.4 78.2 78.3 75.4 79.1 76.0 1996–97
5.2 5.7 5.8 8.3 6.5 7.0 3.9 6.6 1991
GNP ⫽ Gross National Product. * 1989 –90 for United States and Canada.
AJG – June, 2001
Cross-Cultural Variation in IBD
1825
Figure 1. Rating Form of IBD Patient Concerns: adjusted (T) scores on each item for IBD patients in eight countries.
cern (F ⬍ 2, p ⬎ 0.10). In contrast, there was wide variation among countries (F ⬎ 9, p ⬍ 0.0001) in concerns regarding energy level, access to quality medical care, passing the disease onto others, pain or suffering, feeling dirty or smelly, and the uncertain nature of the disease. Factor Scores Raw scores on the four RFIPC factors followed patterns similar to those of the overall mean item scores. When the factor scores were adjusted for reporting tendencies by conversion into T scores (Fig. 2), patients in all countries proved to have similar and moderate levels of concern regarding the impact of disease (mean T score ⫽ 51.5, F ⫽ 0.52, p ⫽ 0.82 by ANOVA), and similar low levels of concern regarding intimacy (mean T score ⫽ 46.3, F ⫽ 0.72, p ⫽ 0.65). There was more diversity regarding the complications of disease, which was rated the highest of the four factors overall (mean T score ⫽ 54.1, F ⫽ 2.29, p ⫽ 0.03): the French and Austrian patients showed relatively high and the Portuguese relatively low concern. Concern over “stigmata” was even more varied (mean T score ⫽ 48.4, F ⫽ 3.59, p ⬍ 0.001) with Portuguese patients particularly concerned and French patients unconcerned. Patterns of Concerns A. BY RANK ORDER. When the items were ranked among all 2002 patients, the highest level of concern was
found to be evoked by the uncertain course of IBD, followed by adverse effects of medication, having an ostomy bag, energy level, and the need for surgery. The five items that elicited the lowest levels of concern were the ability to have children, being treated as different, transmitting the disease to others, feeling smelly or dirty, and sexual performance. Whereas the rank order was broadly comparable among countries—for example, uncertain disease course, adverse effects of medications, and ostomies were ranked among the first nine concerns by patients in every national sample— several items (especially pain and suffering, cancer, and access to medical care) were ranked quite differently in different countries. B. BY T SCORES. When item scores were adjusted to T scores (Fig. 1), the French sample proved to have a distinctive pattern marked by an extreme range of responses. As compared with patients in other countries, they reported particularly high concern regarding pain and suffering, incontinence, needing surgery, energy level, body feelings, and the uncertain course of the disease, and particularly low concern regarding feeling dirty or smelly, sexual performance, having children, being treated as different, and access to quality medical care. Portuguese patients also had a relatively broad range of responses. They felt particularly high concern regarding
1826
Levenstein et al.
AJG – Vol. 96, No. 6, 2001
Figure 2. Rationg Form of IBD Patient Concerns: adjusted (T) scores on each factor for IBD patients in eight countries.
feeling dirty or smelly, passing the disease onto others, bad smells, being treated as different, and access to quality medical care, and particularly low concern regarding the financial impact of the disease, pain and suffering, having surgery, the uncertain course of the disease, and effects of medication. Italian patients reported a particularly low level of concern over developing cancer, whereas Austrians were particularly concerned about needing an ostomy bag and medication effects and little concerned about financial implications. When compared only with other ulcerative colitis patients, the Israeli sample appeared to be particularly concerned about pain and suffering, and particularly unconcerned about needing an ostomy. The American, Canadian, and Swedish samples did not stand out for high or low responses on any item. National Sociodemographic Characteristics Overall mean concern level (sum score) and the individual items regarding financial worries and access to quality med-
ical care were examined with regard to sociodemographic characteristics of the various countries (Table 3). No associations observed reached levels of statistical significance. In the strongest associations (0.05 ⬍ p ⬍ 0.10), concern over access to care appeared to be inversely associated with health care expenditure and with literacy rates. The sum score was modestly associated with Gross National Product per capita, life expectancy, and literacy rate, and financial concern with unemployment rate and life expectancy. All countries but one offer free universal health care to patients with chronic disease, so this variable was not analyzed. Specific Diagnoses Among the seven countries that had administered the RFIPC to both ulcerative colitis and Crohn’s disease patients, the unadjusted mean concerns sum score was higher in Crohn’s disease (40.1) than in ulcerative colitis (34.4). Ulcerative colitis patients expressed greater concern than Crohn’s disease patients about developing cancer and about dying pre-
Table 3. Association of Rating Form of IBD Patient Concerns With National Economic and Health Parameters Correlation With Aspect of Concern
Per Capita GNP
Health Spending
Adult Literacy
Unemployment
Life Expectancy
Infant Mortality
Overall concern level Financial Access to care
⫺0.57 ⫺0.17 ⫺0.40
⫺0.05 ⫺0.07 ⫺0.72*
⫺0.41 ⫺0.06 ⫺0.63*
0.35 0.50 ⫺0.14
⫺0.43 0.55 ⫺0.40
0.38 0.12 0.11
GNP ⫽ Gross National Product. * 0.05 ⬍ p ⬍ 0.10 by Spearman rank correlation coefficient, adjusted for ties.
AJG – June, 2001
Cross-Cultural Variation in IBD
Table 4. Raw Scores on the Rating Form of IBD Patient Concerns: Ulcerative Colitis and Crohn’s Disease Patients From Eight Countries Mean Raw Score Item 1: Financial difficulties 2: Pain or suffering 3: Ability to achieve full potential 4: Loss of bowel control 5: Developing cancer 6: Dying early 7: Being a burden (or depending) on others 8: Attractiveness 9: Feeling alone 10: Feeling out of control 11: Feeling “dirty” or “smelly” 12: Ability to perform sexually 13: Ability to have children 14: Passing the disease onto others 15: Being treated as different 16: Having surgery 17: Having an ostomy bag 18: Producing unpleasant odors 19: Energy level 20: Feelings about your body 21: Intimacy 22: Loss of sexual drive 23: Access to quality medical care 24: Uncertain nature of disease 25: Effects of medication
All Ulcerative Crohn’s Patients Colitis Disease 28.3 38.4 41.9
23.4 33.6 37.2
33.6 44.9 47.9
44.2 41.9 32.0 43.1
44.3 48.1 33.2 38.8
47.2 37.3 31.8 47.1
31.0 27.9 34.1 25.0 26.0 18.7 23.0
27.4 24.6 32.6 24.0 24.1 17.5 19.4
35.6 31.8 37.4 27.7 29.8 23.4 26.2
20.5 49.7 51.1 39.4 50.1 38.1 30.7 29.2 32.6 53.5 51.6
17.9 46.9 50.6 35.5 43.2 34.7 28.3 26.7 31.1 49.4 49.1
24.5 52.9 55.2 44.8 59.2 43.5 35.0 33.7 35.2 60.0 55.2
maturely, but less concern related to all other specific areas of concern (Table 4). The patterns by country of specific adjusted item scores among ulcerative colitis patients was similar (Fig. 3) to the pattern among the entire IBD population. Among Crohn’s disease patients (Fig. 4), the Portuguese appeared to have a strong distinctive pattern, but the size of the Portuguese Crohn’s disease sample was small (N ⫽ 27), and the SD for specific items was correspondingly large.
DISCUSSION Perhaps the most striking finding of this cross-cultural study of the RFIPC was that the overall level of concern reported by IBD patients regarding the impact of their disease varied by more than 2-fold among countries. Artifactual explanations for this wide variability, notably translation difficulties, must be taken into consideration. Careful employment of the translation-back-translation method does not preclude subtle differences between one language and another, especially of the key phrases in the instructions: “concerns or worries related to their disease,” and “how concerned are you with . . .?.” Another possible explanation for the range in concern
1827
levels could be sampling differences in patient populations. Educational level did vary among the countries but did not seem to have an impact: concern levels were high not only among the Portuguese, who had the lowest educational attainment, but also in the very well-educated Canadian and U.S. samples. Different anatomical distributions of disease among samples could theoretically have influenced the results, but the similar raw levels of concern in Italy and France, which were known to have markedly different patterns of disease extent (24% vs 67% of ulcerative colitis patients had left-sided disease, 38% vs 14% of Crohn’s disease patients had isolated ileitis), suggest such an influence is limited. Differences in age and gender could also have affected patients’ concerns, though these two variables were relatively comparable among countries. Unfortunately, none of these factors could be controlled in our analysis because of the limitations of the data; the analyses in this report were based on findings previously obtained independently by researchers in the various countries, precluding adjustment of intercountry differences for patient or disease characteristics. The present findings must therefore be considered exploratory, aimed at generating hypotheses that could open the door to future investigation. Cultural characteristics may play a role in the high intercountry variability in overall concerns level. Examination of the sum score among countries reveals a geographic trend with scores roughly aligned, within Europe, along a northsouth gradient: Italy and Portugal, the two most “Mediterranean” countries, have the highest scores and Sweden, the most “nordic,” has the lowest. To some extent, this northsouth gradient corresponds to trends in national prosperity, educational level, and health expenditures, but it also is likely to reflect some more elusive cultural trends. Israel, whose sum score was relatively low, is a Mediterranean country by geographic location and in some cultural respects, and thus might be expected to have results similar to those of southern European countries. However, 32.4% of the Israeli patients were born in Northern Europe, Eastern Europe, or America, as were the parents of most of the 46.8% native-born Israelis in the sample, making it difficult to characterize this unusual population group definitively. National characteristics of the physician-patient relationship could be relevant, such as the relative prominence of a paternalistic style versus emphasis on patient education and self-management. A high degree of doctor-patient continuity and ease of physician access, likely to vary among institutions as well as among countries, could also allay concern, contributing for example to the particularly low mean score in Sweden. Previous studies have found (13) that IBD patients who perceive themselves as better informed about their disease have lower levels of concerns, and that there is generally an inverse correlation between degress of concern and educational level (5). Our U.S. population is likely to be even better informed than the average American patient with IBD, however, by virtue of having been drawn from an IBD
1828
Levenstein et al.
AJG – Vol. 96, No. 6, 2001
Figure 3. Rating Form of IBD Patient Concerns: adjusted (T) scores on each item for ulcerative colitis patients in seven countries.
self-help group, so that their relatively high concern level with respect to other countries suggests that knowledge may not always decrease concern. Although overall levels of reported concern varied sharply from country to country, the patterns of relative importance assigned to specific concerns and factors were relatively uniform, suggesting that IBD patients from Western countries tend to share common concerns about their disease regardless of their specific country of residence. In particular, there was high concern regarding the uncertain course of the disease, the possibility of ostomy surgery, the adverse effects of medications, and about the complications of disease in general. These IBD patients also broadly shared a relative lack of concern over a series of interpersonal issues related to their disease, such as difficulty having children (not unexpected in a population with a mean age of 41) or being treated as “different.” There were, however, several items that proved to be of considerable concern to patients in some countries but not in others. It is tempting to speculate on possible reasons for some of these differences among countries, though caution is obviously required. For example, Israeli ulcerative colitis patients were particularly concerned about pain and suffering, consistent with older studies of pain behavior among Jewish patients in the United States (19). Other findings are also open to possible cultural interpretations, such as the low
concern over impairment of sexual performance among the French, the low concern of Portuguese subjects over the adverse effects of medication (and their high concern with social stigma), and the high concern over ostomy surgery in Austria. Differences among medical approaches may account for some of the observed variation: concern over ostomy surgery could be related to the local rates of this surgical procedure, and concern over developing cancer may be lower in countries (which, in the experience of the authors, includes Italy) where many physicians downplay cancer risks in their explanations to patients. Finally, in some countries, patients may find responding to questions on sexual performance and intimacy more difficult than the less “personal” questions, which could affect the validity of the responses. Concern over access to quality medical care, which was strikingly high in Portugal and low in France, seems to vary less with the universal availability of medical care or with the direct costs of treatment than with general economic conditions in the various countries and, possibly, with the quality of those resources that are available. The absence of universal free health care in the United States seemed to have no impact on the level or pattern of concerns. Relative concern about the financial repercussions of disease was highest in Israel and Canada and lowest in Austria and Portugal, showing a surprising absence of association
AJG – June, 2001
Cross-Cultural Variation in IBD
1829
Figure 4. Rating Form of IBD Patient Concerns: adjusted (T) scores on each item for Crohn’s disease patients in six countries.
either with out-of-pocket health care costs or with the overall state of the economy. Perceived difficulties in making a living because of disability could be another relevant factor. Considering the substantial differences between ulcerative colitis and Crohn’s disease in their symptoms and course, there was remarkably little difference in the pattern of disease-related concerns among patients with the two diagnoses. The ability of the RFIPC to reveal cross-cultural differences of potential clinical and theoretical importance confirms the value of this innovative approach to quality of life assessment. Previous studies have similarly demonstrated differences in the pattern of disease-related concerns between IBD patients who seek psychological counseling and those who do not (9), and between African-American and other American Crohn’s disease patients (20). In combination with the studies of validity and of sensitivity to changes in disease status described under Materials and Methods, these findings suggest that the RFIPC is an effective and responsive measure of quality of life or health status. The notion of assigning group differences in behaviors, diseases, or questionnaire responses to cultural characteristics is more complex than might be at first apparent. Culture may be viewed as a combination of many factors, some obvious and others obscure, which form the unique characteristics of a specific ethnic population. Such factors may
include—among others— genetics, education, values, religion, lifestyle, and habits. The more a study attributes crosscultural differences to specific factors, the smaller the residual effect of “culture” in explaining the differences. In other words, identification of specific components of importance in a given setting will help the “black box” of culture to become less black. From this viewpoint, the concept of culture may be seen as a tool to help us pin down the specific relevant factors. In conclusion, international variations in health-related quality of life among patients with a common disease have been little examined. The present results suggest that crosscultural comparisons of patient concerns related to IBD are feasible using translated scales, and support the validity of the RFIPC for this purpose. There are large differences in reporting tendencies among countries, and within Europe patients from southern countries tend to report greater overall concern than those from the north. Patients agree across cultures on the relative importance of some concerns but diverge widely on others, with the variable course and outcome of IBD and the prospect of living with an ostomy of particularly high concern to patients in many countries. Semantic or methodological causes may contribute to these differences, but it is likely that the reasons for differences in specific concerns items or in overall concern level also have social, cultural, and/or economic determinants that are open
1830
Levenstein et al.
to further study. These findings may have bearing on the patient-physician relationship, patient education, and therapeutic decision-making.
ACKNOWLEDGMENTS We thank Vilma Varvo, Henrik Hjortswang, Anton Steinberg, Wolfgang Tillinger, and Elizabeth de Rooy for their vital contributions to this project. This study was supported in part by the “Stiftelsen for psykosomatisk och klinisk forskning,” grant 1105/1991, the County Council of Ostergotland, grants 92/017 and 95/042, and the Swedish Society of Medicine, grant 1995/22.0.
AJG – Vol. 96, No. 6, 2001
8.
9. 10.
11. 12.
Reprint requests and correspondence: Susan Levenstein, M.D., Via del Tempio 1A, 00186 Rome, Italy. Received Aug. 21, 2000; accepted Jan. 25, 2001. 13.
REFERENCES 1. WHOQOL Group. The World Health Organization quality of life assessment (WHOQOL): Position paper from the World Health Organization. Soc Sci Med 1995;10:1403–9. 2. Bergner M, Bobbitt R, Carter W, et al. The Sickness Impact Profile: Development and final revision of a health status measure. Med Care 1981;19:787– 805. 3. Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36) I. Conceptual framework and item selection. Med Care 1992;30:473– 83. 4. Irvine EJ, Feagan B, Rochon J, et al. Quality of life: A valid and reliable measure of therapeutic efficacy in the treatment of inflammatory bowel disease. Gastroenterology 1994;106:287– 96. 5. Drossman DA, Leserman J, Li Z, et al. The rating form of IBD patient concerns: A new measure of health status. Psychosom Med 1991;13:701–12. 6. Drossman DA, Patrick DL, Mitchell CM, et al. Health-related quality of life in inflammatory bowel disease: Functional status and patient worries and concerns. Dig Dis Sci 1989;34: 1379 – 86. 7. Drossman DA, Leserman J, Mitchell CM, et al. Health status
14.
15.
16. 17. 18. 19. 20.
and health care use in persons with inflammatory bowel disease: A national sample. Dig Dis Sci 1991;36:1746 –55. Provenzale D, Shearin M, Phillips-Bute B, et al. Health-related quality of life after ileoanal pull-through: Evaluation and assessment of new health status measures. Gastroenterology 1997;113:7–14. Yazdanpanah Y, Klein O, Gambiez L, et al. Impact of surgery on quality of life in Crohn’s disease. Am J Gastroenterol 1997;92:1897–900. Tillinger W, Mittermaier C, Lochs H, Moser G. Health-related quality of life in patients with Crohn‘s disease: Influence of surgical operation — a prospective trial. Dig Dis Sci 1999; 44:932– 8. Sperber AD, Devellis RF, Boehlecke B. Cross-cultural translation: Methodology and validation. J Cross-Cultural Psych 1994;25:501–24. Colombel J-F, Yazdanpanah Y, Laurent F, et al. Qualite´ de vie dans les maladies inflammatoires chroniques de l’intestin: Validation d’un questionnaire et premie`res donne´es franc¸aises. Gastroenterol Clin Biol 1996;20:1071–7 (in French). Moser G, Tillinger W, Sachs G, et al. Disease-related worries and concerns: A study on out-patients with inflammatory bowel disease. Eur J Gastroenterol Hepatol 1995;7:853– 8. Hjortswang H, Stro¨m M, Almeida RT, Almer S. Evaluation of the RFIPC, a disease-specific health-related quality of life questionnaire, in Swedish patients with ulcerative colitis. Scan J Gastroenterol 1997;32:1235– 40. Blondel-Kucharski F, Chircop C, Marquis P, et al. Quality of life in Crohn’s disease: A prospective, longitudinal study in 231 patients. Presented at Journe´es Francophones de Pathologie Digestive; March 22–24, 1999; Nantes, France. World Health Organization. Global health-for-all indicators. Available at: http://www.who.int/hst/hsp/a/country.htm. Accessed Feb. 17, 1999. Central Intelligence Agency. CIA Factbook. Available at: http://www.odci.gov/cia/publications/factbook. Accessed Feb. 17, 1999. Truelove SC, Witts LJ. Cortisone in ulcerative colitis: Preliminary report on a therapeutic trial. Br Med J 1954;2:375– 8. Zborowski M. Cultural components in responses to pain. J Soc Issues 1952;8:16 –30. Eisen G, Strauss W, Sandler R, et al. Health status in Crohn’s disease: Comparison of the Rating Form of IBD Patient Concerns in African and non-African Americans. Gastroenterology 1996;110:A15 (abstract 1157).