549254
research-article2014
PENXXX10.1177/0148607114549254Journal of Parenteral and Enteral NutritionVagianos et al
Original Communication
What Are Adults With Inflammatory Bowel Disease (IBD) Eating? A Closer Look at the Dietary Habits of a Population-Based Canadian IBD Cohort
Journal of Parenteral and Enteral Nutrition Volume XX Number X Month 201X 1–8 © 2014 American Society for Parenteral and Enteral Nutrition DOI: 10.1177/0148607114549254 jpen.sagepub.com hosted at online.sagepub.com
Kathy Vagianos, RD, MSc1; Ian Clara, PhD2; Rachel Carr, MSc3; Leslie A. Graff, CPsych, PhD4; John R. Walker, CPsych, PhD5; Laura E. Targownik, MD6; Lisa M. Lix, PhD, PStat7; Linda Rogala, BN8; Norine Miller, RN8; and Charles N. Bernstein, MD9 Abstract Background: A comprehensive study of what individuals with inflammatory bowel disease (IBD) are eating that encompasses food avoidance, dietary sugar consumption, and a comparison with the non-IBD Canadian population has not been documented. The aim was to analyze these interrelated dietary components. Methods: Food avoidance and sugar intake data were collected from 319 patients with IBD enrolled in the University of Manitoba IBD Cohort Study. Diets of those with IBD (n = 256) were compared with a matched, non-IBD Canadian cohort using the nutrition questions obtained from the Canadian Health Measures Survey (CHMS). Results: Food avoidance among IBD is prevalent for alcohol, popcorn, legumes, nuts, seeds, deep-fried food, and processed deli meat, with a higher prevalence among those with active IBD. Patients with active IBD also consumed significantly more portions of sports drinks and sweetened beverages compared with those with inactive disease. Compared with the non-IBD Canadian population, patients with IBD consume significantly less iron-rich food but more milk. Conclusions: Food avoidance is common among those with IBD but may be due more to personal preferences, while sugar-laden beverages may be displacing other foods higher in nutrients. The overall diet of patients with IBD differed from that of the non-IBD Canadian population, but deficiencies were observed in both groups. Considering malnutrition among persons living with IBD, nutrition education by trained dietitians as part of the IBD team is imperative to address food avoidance and overall balance nutrition as part of treating and preventing nutrition deficiencies. (JPEN J Parenter Enteral Nutr. XXXX;xx:xx-xx)
Keywords nutrition assessment; nutrition; gastroenterology; research and diseases; vitamins
Clinical Relevancy Statement This study compares the diet of a population-based sample with inflammatory bowel disease (IBD) compared with a nonIBD Canadian population. We confirm that dietary deficiencies exist in both the IBD and non-IBD populations. We also report that among all patients with IBD, there is a high prevalence of food avoidance, oftentimes of food with high nutrition value, while an increased intake of sugar-laden beverages exists among patients with active IBD. Deficiencies in the overall diet of patients with IBD, compounded with food avoidance and increased sugar intake, are interrelated dietary concerns that must be determined since these patients are at risk for malnutrition. A comprehensive nutrition assessment and education by trained dietitians as part of an IBD team is imperative.
Introduction Inflammatory bowel disease (IBD) is a chronic inflammatory disease that exclusively involves the colon in ulcerative colitis (UC) and may involve any part of the gastrointestinal tract (GI) in Crohn’s disease (CD). The disruption of the GI tract by inflammation and the associated symptoms of IBD, including anorexia, abdominal pain, nausea, and diarrhea, can result in changes in dietary habits as a means of controlling symptoms.
From the 1Departments of Nutrition and Food Services, Health Sciences Centre and the University of Manitoba IBD Clinical and Research Centre, Winnipeg, Manitoba; 2Departments of Community Health Sciences, University of Manitoba and the University of Manitoba IBD Clinical and Research Centre, Winnipeg, Manitoba; 3Alberta Health Services and the University of Manitoba IBD Clinical and Research Centre, Winnipeg, Manitoba; 4Department of Clinical Health Psychology, Faculty of Medicine and the University of Manitoba IBD Clinical and Research Centre, Winnipeg, Manitoba; 5Departments of Clinical Health Psychology, University of Manitoba and the University of Manitoba IBD Clinical and Research Centre, Winnipeg, Manitoba; 6University of Manitoba Department of Internal Medicine, Section of Gastroenterology and the University of Manitoba IBD Clinical and Research Centre, Winnipeg, Manitoba; 7 Departments of Community Health Sciences and the University of Manitoba IBD Clinical and Research Centre, Winnipeg, Manitoba; 8University of Manitoba IBD Clinical and Research Centre, Winnipeg, Manitoba; and 9 University of Manitoba Departments of Medicine and the University of Manitoba IBD and Clinical Research Centre, Winnipeg, Manitoba. Financial disclosure: The study was funded by the Canadian Institutes of Health Research (CIHR). Received for publication April 7, 2014; accepted for publication August 2, 2014. Corresponding Author: Kathy Vagianos, RD, MSc, Health Sciences Centre, Nutrition and Food Services, MS789A, 820 Sherbrook St, Winnipeg, Manitoba R3A 1R9, Canada. Email: [email protected]
Downloaded from pen.sagepub.com at GEORGIAN COURT UNIV on April 16, 2015
2
Journal of Parenteral and Enteral Nutrition XX(X)
Patients with IBD report that certain foods can induce or exacerbate IBD symptoms.1 These food types are commonly referred to in the IBD literature as “trigger foods.” It is important to consider the assessment of dietary intake in the clinical care of patients with IBD as malnutrition has been reported in up to 85% of these patients and can affect them, particularly those with CD, even when their disease is inactive.2-5 Recent work by our group showed that many persons with IBD have an inadequate intake of vitamins A, C, D, and E, as well as calcium, folate, and iron.4 Patients’ beliefs about food avoidance and ultimately their dietary restrictions may further compromise their nutrition status, particularly if foods avoided are of high nutrient density. While many patients believe their diet directly affects their disease course, the impact of diet on the onset of IBD, or on the exacerbation of IBD symptoms, is less clear. During the 1970s and 1980s, there was interest in the role of simple sugars and carbohydrates in IBD following an observation that adolescents with CD had eaten large amounts of sweets prior to their diagnosis,6,7 with results suggesting that carbohydrate intake may play a role in the onset of IBD or the flare-up of symptoms. Subsequent research has investigated the role of various macronutrients in the diets of IBD, but these have been consistently criticized for their recall bias, limiting the extent of the reliability of the results.8,9 Interestingly, the dietary intake of patients with IBD compared with non-IBD individuals has not been well documented. Nutrition deficiencies cited as specific to IBD may reflect more common eating habits in the general population and, as such, need to be better delineated. Hence, there are distinctive yet interrelated “gaps” in the understanding of diet for patients with IBD. Using a population-based cohort of participants with IBD and a matched community comparison group, this study aimed to determine (1) if patients with IBD avoid particular foods and their reasons for avoidance, (2) if sugar intake specifically is different between patients with active IBD vs inactive IBD, and (3) other differences in dietary intake among patients with IBD compared with Canadians who do not have IBD.
Methods Participants The Manitoba IBD Cohort Study was initiated in 2002, with participating individuals 18 years or older and diagnosed with IBD within the previous 7 years (median, 4.3 years). They were recruited from the population-based University of Manitoba IBD Research Registry.10,11 We have previously reported that the IBD Cohort Study participants were similar demographically and thus representative of the general IBD population in Manitoba.12 Participants completed surveys semiannually and underwent annual in-person interviews. At study initiation, there were 388 IBD Cohort Study participants.13 Data regarding
food avoidance and sugar intake were collected at 48 months (2006–2007) after entry into the longitudinal study. At this point, there were 319 individuals with complete data for analysis in this study, and they had a median of 8 years’ disease duration. A second wave of data regarding intake of specific dietary components using the Canadian Health Measures Survey (CHMS) food questionnaire was collected at 72 months (2009– 2010) after study entry, at a median of 10 years’ disease duration. The final sample included 256 individuals who had complete data across these time points for analysis in this study. The Manitoba IBD Cohort Study was approved by the University of Manitoba Health Research Ethics Board, and participants provided written informed consent.
Disease Activity To assess disease status, participants completed the Manitoba IBD Index (MIBDI). The MIBIDI is a single-level, 6-level response rating scale validated to characterize symptomatic disease activity, based on symptom frequency over the previous 6 months, and is particularly applicable for longitudinal studies involving multiple measurement periods.11 Those who rated their symptoms from levels 1 to 4 (constantly active, often active, sometimes active, or occasionally active 1 or 2 days per month) were categorized as having active disease. Those reporting their symptoms as occurring rarely to never over the previous 6 months (levels 5 or 6) were categorized as having inactive disease.
Nutrition Intake Measures Food avoidance. A food avoidance questionnaire was developed by the research dietitian in collaboration with the research team based on clinical interviews (n = 126) in a previous IBD study.4 At that time, participants were asked to list all foods that are routinely avoided. The most frequently identified foods that were avoided were included in the food avoidance measure used in the current study, with an additional question related to reasons for avoiding these foods. These foods included nuts and seeds; salad or raw vegetables of any type; legumes; tomato products (sauce, juice, or ketchup); raw fruit; deep-fried/higher fat foods; milk and milk products; processed, luncheon, or deli meat; tea or coffee; alcohol; popcorn; and red meat (unpublished data). On the food avoidance questionnaire, for each food item, patients were asked to indicate if they normally eat that food (yes or no), if they consumed the food when their IBD was active (yes or no), and if they consumed the food when their IBD was inactive (yes or no). If the respondent avoided the food, he or she was asked to identify the primary reason why the food item was avoided. Response choices were as follows: (1) “Eating this food causes me to have GI upset with symptoms that last up to 24 hours,” (2) “Eating this food causes me to have GI upset with symptoms that last days to weeks,” (3) “I have read/heard that people with IBD should avoid this
Downloaded from pen.sagepub.com at GEORGIAN COURT UNIV on April 16, 2015
Vagianos et al
3
food,” (4) “A health professional (doctor, nurse, dietitian, other) has advised me to avoid this food,” (5) “I do not like this food (personal preference),” or (6) other. The food avoidance questionnaire was administered by the research nurse at the 48-month point of the IBD cohort study (2006–2007) during face-to-face interviews with the participants. Sugar consumption. A questionnaire was developed by the research dietitian in collaboration with the research team to assess sugar intake. Foods commonly consumed that are determined to be high in simple sugar were included in the survey. The questionnaire asked about amounts of foods high in simple sugars that were consumed during the prior 6 months. Food types included sugar, candy, chocolate, pastries, jams or jelly, regular soft drinks, diet soft drinks, sports drinks, fruit juices, and sweetened drinks. Participants were asked to identify if they normally eat/drink this food (yes or no) and then reported the number of portions consumed per week over the past 6 months. The sugar intake questionnaire was administered by the research nurse at the 48-month point of the IBD cohort study (2006– 2007) during face-to-face interviews with the participants.
Specific Dietary Components The nutrition questions regarding specific dietary components were taken directly from the CHMS and administered as part of the 72-month data collection phase for the IBD Cohort Study (2009–2010). This time point was chosen to better match the time during which the CHMS was administered to the nonIBD Canadian population and the anticipated availability of the CHMS data results for comparison with our IBD data. The primary outcome of interest was to determine how the food intake of IBD cohort participants compared with a matched sample from the Canadian population. The CHMS nutrition questions were administered by a research nurse via face-toface interviews with the participants.
Comparison of the IBD Diet vs the Non-IBD Canadian Diet The CHMS is a nationally representative survey that includes the Canadian population aged 6–79 years living in private households at the time of the survey.14-17 Data were collected at 15 sites across Canada from March 2007 through February 2009. Measures included intake frequency of certain food groups and associated nutrients. To develop the matched comparison group with the IBD cohort, a subsample of the CHMS respondents who were proportionally matched to the IBD sample on sex and age (collapsed into 5-year groupings) distributions was used. This resulted in a sample of 300 individuals from the CHMS.
Data Analysis Descriptive statistics were used to provide means, frequencies, and proportions as appropriate. For the comparisons of the IBD
Table 1. Demographics and Characteristics of Manitoba IBD Cohort Participants at Months 48 and 72. Demographic/Characteristic Age, mean (SD), y Body mass index, mean (SD) Sex, No. (%) Male Female Disease activity, No. (%) Inactive Active Disease type, No. (%) Crohn’s disease Ulcerative colitis
Month 48 (n = 319)
Month 72 (n = 256)
44.6 (14.6) 27.2 (7.2)
46.4 (14.5) 27.0 (7.5)
122 (39) 195 (61)
96 (38) 159 (62)
152 (48) 163 (52)
137 (51) 132 (49)
166 (52) 151 (48)
135 (53) 120 (47)
and community samples, the data from the 2 samples were merged. The CHMS data were weighted using the master weight provided by Statistics Canada, and a weight was created for each participant with IBD that was proportional to his or her probability of being selected from the IBD Research Registry. Mean frequencies of food intake across several types of food were compared between the IBD and CHMS samples using independent-sample t tests, with a significance level of P < .05.
Results Table 1 illustrates the demographic and disease characteristics of the IBD cohort at month 48 and month 72. Similar demographics were observed at both time points. At month 48, 52% had CD and 48% had UC, and overall 48% had active disease. The mean age was 44.6 years, and 61% were female.
Food Avoidance Table 2 illustrates the types of foods those with IBD regularly avoided or avoided selectively when their disease was active. A substantial proportion always avoided nuts and seeds (27%), legumes (30%), deep-fried/higher fat food (25%), processed deli meat (25%), alcohol (31%), and popcorn (30%). However, a larger proportion tended to eat these foods when they were feeling well but avoided them during active disease (ie, nuts and seeds, 35%). The most commonly avoided foods during active disease were salad and raw vegetables, deep-fried foods, and alcohol. A primary reason for avoidance of foods such as legumes, processed deli meat, and tea/coffee was personal preference (ie, “I do not like this food”; Table 3). A large percentage of participants with IBD reported that they avoided food items because they caused GI upset that lasted up to 24 hours (eg, 71% for salad/raw vegetables; 51% for milk and milk products; Table 3). Items most commonly avoided due to GI upset included nuts and seeds, salad and raw vegetables, tomato products, raw fruit, deep-fried/higher fat foods, milk
Downloaded from pen.sagepub.com at GEORGIAN COURT UNIV on April 16, 2015
4
Journal of Parenteral and Enteral Nutrition XX(X)
Table 2. Proportion of Participants With IBD Who Always Avoid Particular Food Items or Avoid When Disease Is Active.a Food Item Alcohol Popcorn Legumes (beans, chickpeas, lentils) Nuts and seeds (peanuts, almonds, walnuts, sunflower seeds, pumpkinseeds) Deep-fried higher fat (food purchased at fast-food restaurants, fried potatoes, or burgers) Processed deli meat (bologna, corned beef, or salami) Tea or coffee Milk/milk products (milk, cheese, yogurt, ice cream) Salad or raw vegetables, any type Tomato products (tomato sauce, tomato juice, or ketchup) Red meat (ground beef, steak, or pork) Raw fruit
Always Avoid, No. (%)
Normally Eat but Avoid When Disease Is Active, No. (%)
104 (31) 100 (30) 103 (30) 92 (27)
142 (42) 129 (38) 96 (28) 119 (35)
85 (25)
143 (42)
85 (25) 43 (13) 42 (12) 35 (10) 30 (9) 26 (8) 22 (6.5)
65 (19) 86 (25) 97 (29) 156 (46) 69 (20) 91 (27) 100 (29)
a
n = 319 participants with inflammatory bowel disease (IBD) participants; responses collected between 2006 and 2007.
Table 3. Proportion of Participants With IBD Reporting Particular Reasons for Avoiding Food Item.a Food Avoided (No. of Responses)
GI Upset (24 h), %
GI Upset (Days to Weeks), %
Heard “Should Avoid,” %
Professional Advice, %
Do Not Like, %
Other, %
Alcohol (n = 96) Popcorn (n = 93) Legumes (n = 97) Nuts and seeds (n = 88) Deep-fried/higher fat food (n = 78) Processed deli meat (n = 73) Tea or coffee (n = 36) Milk/milk products (n = 41) Salad and raw vegetables (n = 35) Tomato products (n = 29) Red meat (n = 25) Raw fruit (n = 20)
14 24 16 38 40 15 8 51 71 55 44 40
8 14 7 10 8 5 3 15 17 10 4 15
1 3 4 2 1 1 3 0 0 3 4 5
3 10 0 3 9 3 3 15 6 3 4 5
70 45 69 32 32 67 81 12 6 24 36 25
4 4 3 15 10 8 3 7 0 3 8 10
a n = 319 participants with inflammatory bowel disease (IBD) participants; responses collected between 2006 and 2007. Response options for food avoidance: (1) eating this food causes me to have gastrointestinal (GI) upset with symptoms that last up to 24 hours, (2) eating this food causes me to have GI upset with symptoms that last days to weeks, (3) I have read/heard that people with IBD should avoid this food, (4) a health professional has advised me to avoid this food, (5) I do not like this food, and (6) other.
and milk products, and red meat. Up to 15% of the patients avoided foods on the basis of professional advice, with milk and milk products being the item most typically avoided on health professional’s advice.
2.6 cups of sweetened drinks per week; P < .05). There were no differences in consumption or intake of sugar-laden food or drink in CD vs UC. There was a trend toward a higher intake of regular soft drinks among those with CD vs UC (P = .08).
Sugar Consumption
Specific Dietary Components: Comparison of the IBD Diet vs the Non-IBD Canadian Diet
The mean weekly intake of sugar-laden foods and beverages is reported in Table 4, comparing those with active and inactive disease, as well as those with CD and UC. Participants with active disease consumed significantly more portions of sports drinks and sweetened beverages compared with those with inactive disease (mean, 2.3 bottles vs 1 bottle of sports drinks; 5.1 vs
The general diet, using food categorized according to the CHMS questionnaire, was compared between the IBD cohort and the non-IBD community controls (Table 5). Participants with IBD consumed significantly less red meat, saltwater fish, shellfish,
Downloaded from pen.sagepub.com at GEORGIAN COURT UNIV on April 16, 2015
Vagianos et al
5
Table 4. Mean Number of Weekly Portions of Sugar-Laden Food and Drink Among Participants With IBD Comparing Those With Inactive and Active Disease and Comparing IBD Subtypes.a Food (Portion Size) Sugar (tsp) Candy (1 candy) Chocolate (bar = 70 g) Pastries (1 piece) Jam or jelly (1 tsp) Regular soft drinks (can = 355 mL) Diet soft drinks (can = 355 mL) Sports drinks (bottle = 591 mL) Fruit juice (250 mL) Sweetened drinks (1 cup)
Active IBD
Inactive IBD
CD
UC
10.1 4.5 1.5
11.2 5.9 1.3
11.7 5.7 1.2
9.7 4.7 1.6
3.3 2.8 3.4
3.1 2.7 3.1
3.3 2.5 3.8
3.1 3 2.8
4.4
4.0
4.4
4.2
2.3b
1.0
2.03
1.4
6.3 5.1b
6.6 2.6
7.1 4.0
5.9 3.5
Table 5. Frequency of Consumption of Food Items (Times per Week) Comparing Persons With IBD and a Matched Community Sample (CHMS) Without IBD and Comparing Those With Active and Inactive IBD.a IBD
a n = 319 participants with inflammatory bowel disease (IBD) participants; responses collected between 2006 and 2007. b P = .05 was for mean comparisons between active and inactive disease and between ulcerative colitis (UC) and Crohn’s disease (CD).
eggs, dried beans, nuts, pasta, fruit, tomatoes, lettuce, fried potatoes, regular soft drinks, and fruit juices compared with those without IBD. Participants with IBD consumed significantly more sausages/bacon, freshwater fish, milk, cottage cheese, spinach, other potatoes, diet soft drinks, and water compared with the nonIBD group. Differences in weekly food intake were also observed in the IBD cohort between those with active disease and inactive disease. Participants with active disease reported significantly less frequent intake of nuts, cereal, fruit, lettuce, and spinach compared with those with inactive disease, as well as more frequent consumption of red meat and regular soft drinks.
Discussion To our knowledge, this is the first study that compares the dietary intake of a population-based sample with IBD with that of controls sampled from the population at large who do not have IBD. Garriguet18 recently analyzed the Canadian diet based on the 2004 Canadian Community Health Survey. Canadians, on average, are within acceptable ranges for the number of servings from the 4 major food groups. However, these averages mask the substantial proportions of adults who do not have a balanced diet. That is, between 40% and 60% of Canadian adults are not consuming the minimum amount of recommended fruits and vegetables per day, and 46%–84% of adults are not consuming the minimum number of milk products as recommended.18 Others have reported that despite fortification and supplement use, the calcium intake of Canadians needs to be improved.19
Food Meat, fish, eggs, and legumes Red meat Sausage/bacon Hot dogs Saltwater fish Freshwater fish Shellfish Eggs Dried beans Nuts Milk products Milk Cottage cheese Yogurt Ice cream Grains Hot/cold cereal Pasta Fruits and vegetables Fruit Tomatoes/tomato sauce Lettuce/green leafy salad Spinach/mustard greens Fries, hashbrowns Other potatoes Other vegetables Beverages Regular soft drinks Diet soft drinks Sports drinks Fruit juices Fruit-flavored drinks Vegetable juices Water
CHMS
Total IBD
Active
Inactive
3.2 0.89 0.37 1.17 0.22 0.41 2.36 1.06 2.06
2.8b 1.17b 0.37 1.03b 0.34b 0.28b 2.06b 0.84b 1.74b
3.02c 1.25b 0.37 0.98b 0.32b 0.30b 1.99b 0.85b 1.56bc
2.59b 1.09b 0.36 1.07 0.35b 0.25b 2.13 0.83b 1.91c
6.8 0.35 3.02 0.83
7.8b 0.56b 2.99 0.82
7.46 0.64b 3.01 0.79
8.16b 0.49b 2.9 0.85
3.16 1.9
3.29 1.64b
2.99 1.65b
3.59bc 1.62b
10 3.2
7.99b 2.55b
7.2b 2.47b
8.7bc 2.64b
3.13
1.88b
1.65b
2.11bc
0.9
1.6b
1.29b
1.9bc
1.08 1.69 6.6
0.82b 2.11b 6.28
0.76b 1.99b 6.1
0.87b 2.22b 6.45
1.98
1.93
2.59c
1.31b
0.76 0.39 4.58 0.89
1.18b 0.38 3.94b 1.03
1.0 0.45 4.03b 1.33
1.36b 0.31 3.85b 0.73
0.97 29
0.88 32.9b
0.98 34
0.78 31.8b
a
n = 256 participants with inflammatory bowel disease (IBD) participants; responses collected between 2009 and 2010. b Significantly different compared with CHMS, P < .05. c Significant difference between active and inactive IBD, P < .05.
Persons with IBD in our study consumed milk 7.8 times per week, which equates to approximately once daily. This was the main source of calcium considered in this survey and would equate to approximately 300 mg of calcium per day (if
Downloaded from pen.sagepub.com at GEORGIAN COURT UNIV on April 16, 2015
6
Journal of Parenteral and Enteral Nutrition XX(X)
using a 1-cup serving size). These findings would suggest that adequate calcium is not being consumed to facilitate prevention of osteoporosis, a potential comorbidity in IBD. However, other sources of calcium may have been consumed that were not specifically quantified in the CHMS nutrition questionnaire (ie, fortified orange juice, almonds, hard cheese). Interestingly, participants with IBD consumed more milk than did the community sample, despite the fact that 15% were specifically avoiding milk on the advice of their health professionals. Persons with IBD were found to consume significantly less iron-rich and protein foods, such as red meat, fish, eggs, dried beans, and nuts, compared with the non-IBD community sample. This is a concern considering that iron deficiency, protein malnutrition, and malabsorption are common nutrition complications in IBD. In our previous work, we reported that 39% of participants with IBD had iron deficiency.4 Although our current study did not analyze for micronutrient deficiencies, it could be assumed that the decreased intake of iron-rich foods among the IBD cohort is a contributor to the development of iron deficiency anemia. Our data confirm that a significant proportion of persons with IBD avoid various types of foods and that the avoidance is more widespread when disease is active. The foods avoided were similar to those reported elsewhere.1 Some food types (nuts and seeds, salad and raw vegetables, tomato products, raw fruit, deep-fried foods, milk, and red meat) may not necessarily exacerbate IBD-related symptoms as participants reported symptoms that lasted only 24 hours upon ingesting those food types. Of particular interest is the large number of foods that are avoided simply due to personal preference by many respondents (legumes, processed meat, tea or coffee, alcohol, popcorn). In previous reports, these foods have been identified as “trigger foods” for IBD exacerbation by patients.20 Our study suggests that many of these foods are excluded from the diet due to personal preferences and may not be related to their IBD. Persons with IBD have strong beliefs about the role diet may play on the exacerbation of symptom.21 Our data underscore the importance of considering a person’s belief system on food and symptoms despite the lack of evidence from prospective controlled trials to support a direct link to dietary beliefs and exacerbation of IBD symptoms. A primary approach to providing advice on food avoidance for persons with IBD should incorporate food diaries to track symptoms (whereby patients then become empowered to control their own symptoms), while dietitians/healthcare providers have a tool to identify dietary nutrition deficiencies and make recommendations appropriately. Interestingly, some of the foods that persons with IBD in our present study reported as “always avoiding” (legumes, milk, fruit) are considered high FODMAP foods (fermentable, oligo-, di-, mono-saccharides and polyols). High-carbohydrate diets and diets high in FODMAPs have been suggested to be associated with abdominal symptoms in functional GI disorders, implying that a low FODMAP diet may be warranted in patients with IBD who report functional gut symptoms.22,23
Ballegaard et al24 identified that 65% of persons with IBD report being intolerant to 1 or more food items compared with only 14% of healthy controls. The foods most often blamed for causing diarrhea and abdominal pain include onions, cabbage, apples, strawberries, citrus fruit, beef, and smoked meat. Among a Canadian IBD cohort, 34 foods have been cited as never eaten by 40% or more of persons with IBD.25 These foods include milk (whole and chocolate), artificial sweetener, diet soft drinks, beer, decaffeinated coffee, liquor, wine, Mexican food, fried fish, liver, bran cereal, blackberries, and apricots. More systematic evaluation is needed to determine to what degree food types trigger symptoms in the absence of inflammation in persons with IBD vs triggering definite inflammation, as well as to what degree food is avoided as a preference or taste preference vs unpleasant effects of the ingestion. Of interest, 15% of persons with IBD in our study who avoided milk responded that this was secondary to advice received from a health professional. While lactose intolerance may occur as a result of small intestinal inflammation, causing loss of brush-border lactase activity, we did not gather information on why lactose avoidance was recommended or which type of health professional was providing this guidance. Avoidance of nutrient-dense foods is a particular concern for the management of IBD. Although our study did not analyze for micronutrient deficiencies, avoidance of salad and raw vegetables in 10% of all participants with IBD and in 46% of those with active disease suggests some risk of dietary deficiency of vitamins C and A, folate, and fiber for many patients, as these foods are key sources of these nutrients. Avoidance of milk products in 12% of all IBD and 29% of active IBD may lead to a potential deficiency in calcium and vitamin D among a group known to be at risk for osteoporosis.2 Micronutrient deficiency of vitamin D was not analyzed in our present study; however, in our previous work, 44% of patients with IBD had vitamin D deficiency.4 Considering that persons with IBD are at risk or may already be malnourished, care providers should ensure these nutrients are provided through other food or supplemental sources. We report that there is a significantly greater intake of sports drinks and sweetened drinks for those with active IBD compared with those with inactive disease. Whether this was a cause or by-product of the disease activity itself cannot be determined from these data. Studies have postulated that an increased intake of sugar may be caused by the disease itself as a result of a decreased ability to taste sweet substances from zinc deficiency.26 We were unable to report on zinc deficiency, but in our previous work, we reported a trend to a higher prevalence of zinc deficiency in active vs inactive IBD.4 Alternatively, a diet high in sugar may simply be a marker for another dietary excess or deficiency. Large percentages of participants with active IBD avoided nuts and seeds, fruits, vegetables, and milk products; hence, a situation may emerge where many nutrientdense food groups are avoided and sugar is increased. Highsugar products may also be a source of easy-to-ingest calories
Downloaded from pen.sagepub.com at GEORGIAN COURT UNIV on April 16, 2015
Vagianos et al
7
for individuals who have reduced their intake but need to maintain adequate hydration and calories. The high intake of soft drinks is of interest since carbonated beverages are a modern phenomenon, and if these were ingested to excess before the onset of disease, it may have implications for disease etiology. Mayberry et al26 reported that patients with IBD added significantly more teaspoons of sugar per day to their diet compared with controls. Silkoff et al27 reported that patients with CD consumed significantly more grams of carbohydrate per day compared with controls. In 2004, Sakamoto et al8 used a validated semiquantitative food frequency questionnaire (FFQ) to identify that in the preillness diet of patients with IBD, there was an increased consumption of sugars, sweeteners, and sweets and that these were positively associated with an increased risk of CD and UC. Recall bias limits the interpretation and validity of the results.28 Ideally, prospective dietary intervention studies are more reliable. Interestingly, in only 1 intervention study were patients with IBD and controls provided either a high-fiber/low-sucrose diet or a low-fiber/unrestricted sucrose diet.29 In this study, there were no significant differences between the dietary groups in terms of surgery, hospital admissions, or symptom exacerbation. We, too, could not determine as to when persons with IBD increased their sugar intake in relation to their disease becoming active. Further exploration of sugar intake habits (particularly prior to disease onset) is warranted, since the relative balance of nutrients ingested can affect the gut microbiome and could be considered a trigger of IBD.30
Conclusions This study is one of the few to document what patients with IBD typically eat. Previous work from our group has shown that although the vast majority (80%–89%) of persons diagnosed with IBD view nutrition information as being very important, only 8%–16% of patients report receiving adequate information on nutrition in the early period following diagnosis, despite a strong interest in obtaining this knowledge.31 Our study shows that persons with IBD will avoid particular foods, oftentimes nutrient-rich foods, based on symptoms and personal preference, while sugar-laden beverages may be displacing other foods that are higher in nutrients. The overall diet of our IBD sample differed to some extent from a general non-IBD community sample. However, deficiencies exist in both groups, flagging the need to assess the diet and address deficiencies, particularly for the chronic disease group given other health risks. Considering the risk of malnutrition among people with IBD, nutrition education by trained dietitians as part of the IBD team is imperative to address food avoidance and overall balanced nutrition as part of treating and preventing nutrition deficiencies among this group.
References 1. Prince A, Whelan K, Moosa A., Lomer MCE, Reidlinger D. Nutritional problems in inflammatory bowel disease: the patient perspective. J Crohns Colitis. 2011;5:443-450. 2. Lomer MCE. Symposium 7: Nutrition in inflammatory bowel disease. Dietary and nutritional considerations for inflammatory bowel disease. Proc Nutr Soc. 2011;70:329-335. 3. Husain A, Korzenik JR. Nutritional issues in inflammatory bowel disease. Semin Gastrointest Dis. 1998;9:21-30. 4. Vagianos K, Bector S, McConnell J, Bernstein CN. Nutrition assessment of patients with inflammatory bowel disease. JPEN J Parenter Enteral Nutr. 2007;31:311-319. 5. O’Sullivan M, O’Morain C. Nutrition in inflammatory bowel disease. Best Pract Res Clin Gastroenterol. 2006;20:561-573. 6. Miller B, Fervers F, Rohbeck R, Strohmeyer G. Sugar consumption in patients with Crohn’s disease [in German]. Verh Dtsch Ges Inn Med. 1976;82:922-924. 7. Martini GA, Brandes JW. Increased consumption of refined carbohydrates in patients with Crohn’s disease. Klin Wschr. 1976;54:367-371. 8. Sakamoto N, Kono S, Wakai K, et al. Dietary risk factors for inflammatory bowel disease: a multicentre case-control study in Japan. Inflamm Bowel Dis. 2004;11:154-162. 9. Chapman-Kiddell CA, Davies PSW, Gillen L, Radford-Smith GL. Role of diet in the development of inflammatory bowel disease. Inflamm Bowel Dis. 2012;16:137-151. 10. Bernstein CN, Blanchard JF, Rawsthorne P, et al. The epidemiology of Crohn’s disease and ulcerative colitis in a central Canadian province: a population-based study. Am J Epidemiol. 1999;149:916-924. 11. Clara I, Lix LM, Walker JR, et al. The Manitoba IBD index: evidence for a new and simple indicator of IBD activity. Am J Gastroenterol. 2009;104:1754-1763. 12. Longobardi T, Walker JR, Graff LA, Bernstein CN. Health service utilization in IBD: comparison of self-report and administrative data. BMC Health Serv Res. 2011;11:137. 13. Graff LA, Walker JR, Lix L, et al. The relationship of inflammatory bowel disease type and activity to psychological functioning and quality of life. Clin Gastroenterol Hepatol. 2006;4:1491-1501. 14. Bryan SN, St-Denis M, Wojitas D. Canadian Health Measures Survey: clinic operations and logistics. Health Rep. 2007;18(suppl):53-69. 15. Day B, Langlois R, Tremblay M, Knoppers BM. Canadian Health Measures Survey: ethical, legal and social issues. Health Rep. 2007;18(suppl): 37-52. 16. Giroux S. Canadian Health Measures Survey: Sampling strategy overview. Health Rep. 2007;18(suppl):31–36. 17. Tremblay M, Wolfson M, Conner Gorbe S. Canadian Health Measures Survey: rationale, background and overview. Health Rep. 2007;18(suppl):7-20. 18. Garriguet D. Canadians’ eating habits. Health Rep. 2007;18:17-32. 19. Vatanparast H, Dolega-Cieszkowski JH, Whiting SJ. Many adult Canadians are not meeting current calcium recommendations from food and supplement intake. Appl Physiol Nutr Metab. 2009;34:191-196. 20. Zallot C, Quilliot D, Chevaux J, et al. Dietary beliefs and behavior among inflammatory bowel disease patients. Inflamm Bowel Dis. 2013;19:66-72. 21. Hou JK, Lee D, Lewis J. Diet and inflammatory bowel disease: review of patient-targeted recommendations [published online October 6, 2013]. Clin Gastroenterol Hepatol. 22. Gibson PR, Shepherd SJ. Evidence-based dietary management of functional gastrointestinal symptoms: the FODMAP approach. J Gastroenterol Hepatol. 2010;25:252-258. 23. Gearry RB, Irving PM, Barrett JS, Nathan DM, Shepherd SJ, Gibson PR. Reduction of dietary poorly absorbed short-chain carbohydrates (FODMAPs) improves abdominal symptoms in patients with inflammatory bowel disease—a pilot study. J Crohns Colitis. 2009;3:8-14.
Downloaded from pen.sagepub.com at GEORGIAN COURT UNIV on April 16, 2015
8
Journal of Parenteral and Enteral Nutrition XX(X)
24. Ballegaard M, Bjerstrom A, Brondum S, Hylander E, Jensen L, Ladefoged K. Self reported food intolerance in chronic inflammatory bowel disease. Scand J Gastroenterol. 1997;32:569-571. 25. Joachim G. Responses of people with inflammatory bowel disease to foods consumed. Gastroenterol Nurs. 2000;23:160-167. 26. Mayberry JF, Rhodes J, Newcombe RG. Increased sugar consumption in Crohn’s disease. Digestion. 1980;20:323-326. 27. Silkoff K, Hallak A, Yegena L, et al. Consumption of refined carbohydrate by patients with Crohn’s disease in Tel-Aviv-Yafo. Postgrad Med J. 1980;56:842-846.
28. Krishnan A, Korzenik JR. Inflammatory bowel disease and environmental influences. Gastroenterol Clin. 2002;31:21-39. 29. Ritchie JK, Wadsworth J, Lennard-Jones JE, Rogers E. Controlled multicentre therapeutic trial of an unrefined carbohydrate, fibre rich diet in Crohn’s disease. BMJ. 1987;295:517-520. 30. Ananthakrishnan A. Environmental risk factors for inflammatory bowel disease. Gastroenterol Hepatol. 2013;9:367-374. 31. Wong S, Walker JR, Carr R, et al. The information needs and preferences of persons with longstanding inflammatory bowel disease. Can J Gastroenterol. 2012;26:525-531.
Downloaded from pen.sagepub.com at GEORGIAN COURT UNIV on April 16, 2015
research-article2014
PENXXX10.1177/0148607114549254Journal of Parenteral and Enteral NutritionVagianos et al
Original Communication
What Are Adults With Inflammatory Bowel Disease (IBD) Eating? A Closer Look at the Dietary Habits of a Population-Based Canadian IBD Cohort
Journal of Parenteral and Enteral Nutrition Volume XX Number X Month 201X 1–8 © 2014 American Society for Parenteral and Enteral Nutrition DOI: 10.1177/0148607114549254 jpen.sagepub.com hosted at online.sagepub.com
Kathy Vagianos, RD, MSc1; Ian Clara, PhD2; Rachel Carr, MSc3; Leslie A. Graff, CPsych, PhD4; John R. Walker, CPsych, PhD5; Laura E. Targownik, MD6; Lisa M. Lix, PhD, PStat7; Linda Rogala, BN8; Norine Miller, RN8; and Charles N. Bernstein, MD9 Abstract Background: A comprehensive study of what individuals with inflammatory bowel disease (IBD) are eating that encompasses food avoidance, dietary sugar consumption, and a comparison with the non-IBD Canadian population has not been documented. The aim was to analyze these interrelated dietary components. Methods: Food avoidance and sugar intake data were collected from 319 patients with IBD enrolled in the University of Manitoba IBD Cohort Study. Diets of those with IBD (n = 256) were compared with a matched, non-IBD Canadian cohort using the nutrition questions obtained from the Canadian Health Measures Survey (CHMS). Results: Food avoidance among IBD is prevalent for alcohol, popcorn, legumes, nuts, seeds, deep-fried food, and processed deli meat, with a higher prevalence among those with active IBD. Patients with active IBD also consumed significantly more portions of sports drinks and sweetened beverages compared with those with inactive disease. Compared with the non-IBD Canadian population, patients with IBD consume significantly less iron-rich food but more milk. Conclusions: Food avoidance is common among those with IBD but may be due more to personal preferences, while sugar-laden beverages may be displacing other foods higher in nutrients. The overall diet of patients with IBD differed from that of the non-IBD Canadian population, but deficiencies were observed in both groups. Considering malnutrition among persons living with IBD, nutrition education by trained dietitians as part of the IBD team is imperative to address food avoidance and overall balance nutrition as part of treating and preventing nutrition deficiencies. (JPEN J Parenter Enteral Nutr. XXXX;xx:xx-xx)
Keywords nutrition assessment; nutrition; gastroenterology; research and diseases; vitamins
Clinical Relevancy Statement This study compares the diet of a population-based sample with inflammatory bowel disease (IBD) compared with a nonIBD Canadian population. We confirm that dietary deficiencies exist in both the IBD and non-IBD populations. We also report that among all patients with IBD, there is a high prevalence of food avoidance, oftentimes of food with high nutrition value, while an increased intake of sugar-laden beverages exists among patients with active IBD. Deficiencies in the overall diet of patients with IBD, compounded with food avoidance and increased sugar intake, are interrelated dietary concerns that must be determined since these patients are at risk for malnutrition. A comprehensive nutrition assessment and education by trained dietitians as part of an IBD team is imperative.
Introduction Inflammatory bowel disease (IBD) is a chronic inflammatory disease that exclusively involves the colon in ulcerative colitis (UC) and may involve any part of the gastrointestinal tract (GI) in Crohn’s disease (CD). The disruption of the GI tract by inflammation and the associated symptoms of IBD, including anorexia, abdominal pain, nausea, and diarrhea, can result in changes in dietary habits as a means of controlling symptoms.
From the 1Departments of Nutrition and Food Services, Health Sciences Centre and the University of Manitoba IBD Clinical and Research Centre, Winnipeg, Manitoba; 2Departments of Community Health Sciences, University of Manitoba and the University of Manitoba IBD Clinical and Research Centre, Winnipeg, Manitoba; 3Alberta Health Services and the University of Manitoba IBD Clinical and Research Centre, Winnipeg, Manitoba; 4Department of Clinical Health Psychology, Faculty of Medicine and the University of Manitoba IBD Clinical and Research Centre, Winnipeg, Manitoba; 5Departments of Clinical Health Psychology, University of Manitoba and the University of Manitoba IBD Clinical and Research Centre, Winnipeg, Manitoba; 6University of Manitoba Department of Internal Medicine, Section of Gastroenterology and the University of Manitoba IBD Clinical and Research Centre, Winnipeg, Manitoba; 7 Departments of Community Health Sciences and the University of Manitoba IBD Clinical and Research Centre, Winnipeg, Manitoba; 8University of Manitoba IBD Clinical and Research Centre, Winnipeg, Manitoba; and 9 University of Manitoba Departments of Medicine and the University of Manitoba IBD and Clinical Research Centre, Winnipeg, Manitoba. Financial disclosure: The study was funded by the Canadian Institutes of Health Research (CIHR). Received for publication April 7, 2014; accepted for publication August 2, 2014. Corresponding Author: Kathy Vagianos, RD, MSc, Health Sciences Centre, Nutrition and Food Services, MS789A, 820 Sherbrook St, Winnipeg, Manitoba R3A 1R9, Canada. Email: [email protected]
Downloaded from pen.sagepub.com at GEORGIAN COURT UNIV on April 16, 2015
2
Journal of Parenteral and Enteral Nutrition XX(X)
Patients with IBD report that certain foods can induce or exacerbate IBD symptoms.1 These food types are commonly referred to in the IBD literature as “trigger foods.” It is important to consider the assessment of dietary intake in the clinical care of patients with IBD as malnutrition has been reported in up to 85% of these patients and can affect them, particularly those with CD, even when their disease is inactive.2-5 Recent work by our group showed that many persons with IBD have an inadequate intake of vitamins A, C, D, and E, as well as calcium, folate, and iron.4 Patients’ beliefs about food avoidance and ultimately their dietary restrictions may further compromise their nutrition status, particularly if foods avoided are of high nutrient density. While many patients believe their diet directly affects their disease course, the impact of diet on the onset of IBD, or on the exacerbation of IBD symptoms, is less clear. During the 1970s and 1980s, there was interest in the role of simple sugars and carbohydrates in IBD following an observation that adolescents with CD had eaten large amounts of sweets prior to their diagnosis,6,7 with results suggesting that carbohydrate intake may play a role in the onset of IBD or the flare-up of symptoms. Subsequent research has investigated the role of various macronutrients in the diets of IBD, but these have been consistently criticized for their recall bias, limiting the extent of the reliability of the results.8,9 Interestingly, the dietary intake of patients with IBD compared with non-IBD individuals has not been well documented. Nutrition deficiencies cited as specific to IBD may reflect more common eating habits in the general population and, as such, need to be better delineated. Hence, there are distinctive yet interrelated “gaps” in the understanding of diet for patients with IBD. Using a population-based cohort of participants with IBD and a matched community comparison group, this study aimed to determine (1) if patients with IBD avoid particular foods and their reasons for avoidance, (2) if sugar intake specifically is different between patients with active IBD vs inactive IBD, and (3) other differences in dietary intake among patients with IBD compared with Canadians who do not have IBD.
Methods Participants The Manitoba IBD Cohort Study was initiated in 2002, with participating individuals 18 years or older and diagnosed with IBD within the previous 7 years (median, 4.3 years). They were recruited from the population-based University of Manitoba IBD Research Registry.10,11 We have previously reported that the IBD Cohort Study participants were similar demographically and thus representative of the general IBD population in Manitoba.12 Participants completed surveys semiannually and underwent annual in-person interviews. At study initiation, there were 388 IBD Cohort Study participants.13 Data regarding
food avoidance and sugar intake were collected at 48 months (2006–2007) after entry into the longitudinal study. At this point, there were 319 individuals with complete data for analysis in this study, and they had a median of 8 years’ disease duration. A second wave of data regarding intake of specific dietary components using the Canadian Health Measures Survey (CHMS) food questionnaire was collected at 72 months (2009– 2010) after study entry, at a median of 10 years’ disease duration. The final sample included 256 individuals who had complete data across these time points for analysis in this study. The Manitoba IBD Cohort Study was approved by the University of Manitoba Health Research Ethics Board, and participants provided written informed consent.
Disease Activity To assess disease status, participants completed the Manitoba IBD Index (MIBDI). The MIBIDI is a single-level, 6-level response rating scale validated to characterize symptomatic disease activity, based on symptom frequency over the previous 6 months, and is particularly applicable for longitudinal studies involving multiple measurement periods.11 Those who rated their symptoms from levels 1 to 4 (constantly active, often active, sometimes active, or occasionally active 1 or 2 days per month) were categorized as having active disease. Those reporting their symptoms as occurring rarely to never over the previous 6 months (levels 5 or 6) were categorized as having inactive disease.
Nutrition Intake Measures Food avoidance. A food avoidance questionnaire was developed by the research dietitian in collaboration with the research team based on clinical interviews (n = 126) in a previous IBD study.4 At that time, participants were asked to list all foods that are routinely avoided. The most frequently identified foods that were avoided were included in the food avoidance measure used in the current study, with an additional question related to reasons for avoiding these foods. These foods included nuts and seeds; salad or raw vegetables of any type; legumes; tomato products (sauce, juice, or ketchup); raw fruit; deep-fried/higher fat foods; milk and milk products; processed, luncheon, or deli meat; tea or coffee; alcohol; popcorn; and red meat (unpublished data). On the food avoidance questionnaire, for each food item, patients were asked to indicate if they normally eat that food (yes or no), if they consumed the food when their IBD was active (yes or no), and if they consumed the food when their IBD was inactive (yes or no). If the respondent avoided the food, he or she was asked to identify the primary reason why the food item was avoided. Response choices were as follows: (1) “Eating this food causes me to have GI upset with symptoms that last up to 24 hours,” (2) “Eating this food causes me to have GI upset with symptoms that last days to weeks,” (3) “I have read/heard that people with IBD should avoid this
Downloaded from pen.sagepub.com at GEORGIAN COURT UNIV on April 16, 2015
Vagianos et al
3
food,” (4) “A health professional (doctor, nurse, dietitian, other) has advised me to avoid this food,” (5) “I do not like this food (personal preference),” or (6) other. The food avoidance questionnaire was administered by the research nurse at the 48-month point of the IBD cohort study (2006–2007) during face-to-face interviews with the participants. Sugar consumption. A questionnaire was developed by the research dietitian in collaboration with the research team to assess sugar intake. Foods commonly consumed that are determined to be high in simple sugar were included in the survey. The questionnaire asked about amounts of foods high in simple sugars that were consumed during the prior 6 months. Food types included sugar, candy, chocolate, pastries, jams or jelly, regular soft drinks, diet soft drinks, sports drinks, fruit juices, and sweetened drinks. Participants were asked to identify if they normally eat/drink this food (yes or no) and then reported the number of portions consumed per week over the past 6 months. The sugar intake questionnaire was administered by the research nurse at the 48-month point of the IBD cohort study (2006– 2007) during face-to-face interviews with the participants.
Specific Dietary Components The nutrition questions regarding specific dietary components were taken directly from the CHMS and administered as part of the 72-month data collection phase for the IBD Cohort Study (2009–2010). This time point was chosen to better match the time during which the CHMS was administered to the nonIBD Canadian population and the anticipated availability of the CHMS data results for comparison with our IBD data. The primary outcome of interest was to determine how the food intake of IBD cohort participants compared with a matched sample from the Canadian population. The CHMS nutrition questions were administered by a research nurse via face-toface interviews with the participants.
Comparison of the IBD Diet vs the Non-IBD Canadian Diet The CHMS is a nationally representative survey that includes the Canadian population aged 6–79 years living in private households at the time of the survey.14-17 Data were collected at 15 sites across Canada from March 2007 through February 2009. Measures included intake frequency of certain food groups and associated nutrients. To develop the matched comparison group with the IBD cohort, a subsample of the CHMS respondents who were proportionally matched to the IBD sample on sex and age (collapsed into 5-year groupings) distributions was used. This resulted in a sample of 300 individuals from the CHMS.
Data Analysis Descriptive statistics were used to provide means, frequencies, and proportions as appropriate. For the comparisons of the IBD
Table 1. Demographics and Characteristics of Manitoba IBD Cohort Participants at Months 48 and 72. Demographic/Characteristic Age, mean (SD), y Body mass index, mean (SD) Sex, No. (%) Male Female Disease activity, No. (%) Inactive Active Disease type, No. (%) Crohn’s disease Ulcerative colitis
Month 48 (n = 319)
Month 72 (n = 256)
44.6 (14.6) 27.2 (7.2)
46.4 (14.5) 27.0 (7.5)
122 (39) 195 (61)
96 (38) 159 (62)
152 (48) 163 (52)
137 (51) 132 (49)
166 (52) 151 (48)
135 (53) 120 (47)
and community samples, the data from the 2 samples were merged. The CHMS data were weighted using the master weight provided by Statistics Canada, and a weight was created for each participant with IBD that was proportional to his or her probability of being selected from the IBD Research Registry. Mean frequencies of food intake across several types of food were compared between the IBD and CHMS samples using independent-sample t tests, with a significance level of P < .05.
Results Table 1 illustrates the demographic and disease characteristics of the IBD cohort at month 48 and month 72. Similar demographics were observed at both time points. At month 48, 52% had CD and 48% had UC, and overall 48% had active disease. The mean age was 44.6 years, and 61% were female.
Food Avoidance Table 2 illustrates the types of foods those with IBD regularly avoided or avoided selectively when their disease was active. A substantial proportion always avoided nuts and seeds (27%), legumes (30%), deep-fried/higher fat food (25%), processed deli meat (25%), alcohol (31%), and popcorn (30%). However, a larger proportion tended to eat these foods when they were feeling well but avoided them during active disease (ie, nuts and seeds, 35%). The most commonly avoided foods during active disease were salad and raw vegetables, deep-fried foods, and alcohol. A primary reason for avoidance of foods such as legumes, processed deli meat, and tea/coffee was personal preference (ie, “I do not like this food”; Table 3). A large percentage of participants with IBD reported that they avoided food items because they caused GI upset that lasted up to 24 hours (eg, 71% for salad/raw vegetables; 51% for milk and milk products; Table 3). Items most commonly avoided due to GI upset included nuts and seeds, salad and raw vegetables, tomato products, raw fruit, deep-fried/higher fat foods, milk
Downloaded from pen.sagepub.com at GEORGIAN COURT UNIV on April 16, 2015
4
Journal of Parenteral and Enteral Nutrition XX(X)
Table 2. Proportion of Participants With IBD Who Always Avoid Particular Food Items or Avoid When Disease Is Active.a Food Item Alcohol Popcorn Legumes (beans, chickpeas, lentils) Nuts and seeds (peanuts, almonds, walnuts, sunflower seeds, pumpkinseeds) Deep-fried higher fat (food purchased at fast-food restaurants, fried potatoes, or burgers) Processed deli meat (bologna, corned beef, or salami) Tea or coffee Milk/milk products (milk, cheese, yogurt, ice cream) Salad or raw vegetables, any type Tomato products (tomato sauce, tomato juice, or ketchup) Red meat (ground beef, steak, or pork) Raw fruit
Always Avoid, No. (%)
Normally Eat but Avoid When Disease Is Active, No. (%)
104 (31) 100 (30) 103 (30) 92 (27)
142 (42) 129 (38) 96 (28) 119 (35)
85 (25)
143 (42)
85 (25) 43 (13) 42 (12) 35 (10) 30 (9) 26 (8) 22 (6.5)
65 (19) 86 (25) 97 (29) 156 (46) 69 (20) 91 (27) 100 (29)
a
n = 319 participants with inflammatory bowel disease (IBD) participants; responses collected between 2006 and 2007.
Table 3. Proportion of Participants With IBD Reporting Particular Reasons for Avoiding Food Item.a Food Avoided (No. of Responses)
GI Upset (24 h), %
GI Upset (Days to Weeks), %
Heard “Should Avoid,” %
Professional Advice, %
Do Not Like, %
Other, %
Alcohol (n = 96) Popcorn (n = 93) Legumes (n = 97) Nuts and seeds (n = 88) Deep-fried/higher fat food (n = 78) Processed deli meat (n = 73) Tea or coffee (n = 36) Milk/milk products (n = 41) Salad and raw vegetables (n = 35) Tomato products (n = 29) Red meat (n = 25) Raw fruit (n = 20)
14 24 16 38 40 15 8 51 71 55 44 40
8 14 7 10 8 5 3 15 17 10 4 15
1 3 4 2 1 1 3 0 0 3 4 5
3 10 0 3 9 3 3 15 6 3 4 5
70 45 69 32 32 67 81 12 6 24 36 25
4 4 3 15 10 8 3 7 0 3 8 10
a n = 319 participants with inflammatory bowel disease (IBD) participants; responses collected between 2006 and 2007. Response options for food avoidance: (1) eating this food causes me to have gastrointestinal (GI) upset with symptoms that last up to 24 hours, (2) eating this food causes me to have GI upset with symptoms that last days to weeks, (3) I have read/heard that people with IBD should avoid this food, (4) a health professional has advised me to avoid this food, (5) I do not like this food, and (6) other.
and milk products, and red meat. Up to 15% of the patients avoided foods on the basis of professional advice, with milk and milk products being the item most typically avoided on health professional’s advice.
2.6 cups of sweetened drinks per week; P < .05). There were no differences in consumption or intake of sugar-laden food or drink in CD vs UC. There was a trend toward a higher intake of regular soft drinks among those with CD vs UC (P = .08).
Sugar Consumption
Specific Dietary Components: Comparison of the IBD Diet vs the Non-IBD Canadian Diet
The mean weekly intake of sugar-laden foods and beverages is reported in Table 4, comparing those with active and inactive disease, as well as those with CD and UC. Participants with active disease consumed significantly more portions of sports drinks and sweetened beverages compared with those with inactive disease (mean, 2.3 bottles vs 1 bottle of sports drinks; 5.1 vs
The general diet, using food categorized according to the CHMS questionnaire, was compared between the IBD cohort and the non-IBD community controls (Table 5). Participants with IBD consumed significantly less red meat, saltwater fish, shellfish,
Downloaded from pen.sagepub.com at GEORGIAN COURT UNIV on April 16, 2015
Vagianos et al
5
Table 4. Mean Number of Weekly Portions of Sugar-Laden Food and Drink Among Participants With IBD Comparing Those With Inactive and Active Disease and Comparing IBD Subtypes.a Food (Portion Size) Sugar (tsp) Candy (1 candy) Chocolate (bar = 70 g) Pastries (1 piece) Jam or jelly (1 tsp) Regular soft drinks (can = 355 mL) Diet soft drinks (can = 355 mL) Sports drinks (bottle = 591 mL) Fruit juice (250 mL) Sweetened drinks (1 cup)
Active IBD
Inactive IBD
CD
UC
10.1 4.5 1.5
11.2 5.9 1.3
11.7 5.7 1.2
9.7 4.7 1.6
3.3 2.8 3.4
3.1 2.7 3.1
3.3 2.5 3.8
3.1 3 2.8
4.4
4.0
4.4
4.2
2.3b
1.0
2.03
1.4
6.3 5.1b
6.6 2.6
7.1 4.0
5.9 3.5
Table 5. Frequency of Consumption of Food Items (Times per Week) Comparing Persons With IBD and a Matched Community Sample (CHMS) Without IBD and Comparing Those With Active and Inactive IBD.a IBD
a n = 319 participants with inflammatory bowel disease (IBD) participants; responses collected between 2006 and 2007. b P = .05 was for mean comparisons between active and inactive disease and between ulcerative colitis (UC) and Crohn’s disease (CD).
eggs, dried beans, nuts, pasta, fruit, tomatoes, lettuce, fried potatoes, regular soft drinks, and fruit juices compared with those without IBD. Participants with IBD consumed significantly more sausages/bacon, freshwater fish, milk, cottage cheese, spinach, other potatoes, diet soft drinks, and water compared with the nonIBD group. Differences in weekly food intake were also observed in the IBD cohort between those with active disease and inactive disease. Participants with active disease reported significantly less frequent intake of nuts, cereal, fruit, lettuce, and spinach compared with those with inactive disease, as well as more frequent consumption of red meat and regular soft drinks.
Discussion To our knowledge, this is the first study that compares the dietary intake of a population-based sample with IBD with that of controls sampled from the population at large who do not have IBD. Garriguet18 recently analyzed the Canadian diet based on the 2004 Canadian Community Health Survey. Canadians, on average, are within acceptable ranges for the number of servings from the 4 major food groups. However, these averages mask the substantial proportions of adults who do not have a balanced diet. That is, between 40% and 60% of Canadian adults are not consuming the minimum amount of recommended fruits and vegetables per day, and 46%–84% of adults are not consuming the minimum number of milk products as recommended.18 Others have reported that despite fortification and supplement use, the calcium intake of Canadians needs to be improved.19
Food Meat, fish, eggs, and legumes Red meat Sausage/bacon Hot dogs Saltwater fish Freshwater fish Shellfish Eggs Dried beans Nuts Milk products Milk Cottage cheese Yogurt Ice cream Grains Hot/cold cereal Pasta Fruits and vegetables Fruit Tomatoes/tomato sauce Lettuce/green leafy salad Spinach/mustard greens Fries, hashbrowns Other potatoes Other vegetables Beverages Regular soft drinks Diet soft drinks Sports drinks Fruit juices Fruit-flavored drinks Vegetable juices Water
CHMS
Total IBD
Active
Inactive
3.2 0.89 0.37 1.17 0.22 0.41 2.36 1.06 2.06
2.8b 1.17b 0.37 1.03b 0.34b 0.28b 2.06b 0.84b 1.74b
3.02c 1.25b 0.37 0.98b 0.32b 0.30b 1.99b 0.85b 1.56bc
2.59b 1.09b 0.36 1.07 0.35b 0.25b 2.13 0.83b 1.91c
6.8 0.35 3.02 0.83
7.8b 0.56b 2.99 0.82
7.46 0.64b 3.01 0.79
8.16b 0.49b 2.9 0.85
3.16 1.9
3.29 1.64b
2.99 1.65b
3.59bc 1.62b
10 3.2
7.99b 2.55b
7.2b 2.47b
8.7bc 2.64b
3.13
1.88b
1.65b
2.11bc
0.9
1.6b
1.29b
1.9bc
1.08 1.69 6.6
0.82b 2.11b 6.28
0.76b 1.99b 6.1
0.87b 2.22b 6.45
1.98
1.93
2.59c
1.31b
0.76 0.39 4.58 0.89
1.18b 0.38 3.94b 1.03
1.0 0.45 4.03b 1.33
1.36b 0.31 3.85b 0.73
0.97 29
0.88 32.9b
0.98 34
0.78 31.8b
a
n = 256 participants with inflammatory bowel disease (IBD) participants; responses collected between 2009 and 2010. b Significantly different compared with CHMS, P < .05. c Significant difference between active and inactive IBD, P < .05.
Persons with IBD in our study consumed milk 7.8 times per week, which equates to approximately once daily. This was the main source of calcium considered in this survey and would equate to approximately 300 mg of calcium per day (if
Downloaded from pen.sagepub.com at GEORGIAN COURT UNIV on April 16, 2015
6
Journal of Parenteral and Enteral Nutrition XX(X)
using a 1-cup serving size). These findings would suggest that adequate calcium is not being consumed to facilitate prevention of osteoporosis, a potential comorbidity in IBD. However, other sources of calcium may have been consumed that were not specifically quantified in the CHMS nutrition questionnaire (ie, fortified orange juice, almonds, hard cheese). Interestingly, participants with IBD consumed more milk than did the community sample, despite the fact that 15% were specifically avoiding milk on the advice of their health professionals. Persons with IBD were found to consume significantly less iron-rich and protein foods, such as red meat, fish, eggs, dried beans, and nuts, compared with the non-IBD community sample. This is a concern considering that iron deficiency, protein malnutrition, and malabsorption are common nutrition complications in IBD. In our previous work, we reported that 39% of participants with IBD had iron deficiency.4 Although our current study did not analyze for micronutrient deficiencies, it could be assumed that the decreased intake of iron-rich foods among the IBD cohort is a contributor to the development of iron deficiency anemia. Our data confirm that a significant proportion of persons with IBD avoid various types of foods and that the avoidance is more widespread when disease is active. The foods avoided were similar to those reported elsewhere.1 Some food types (nuts and seeds, salad and raw vegetables, tomato products, raw fruit, deep-fried foods, milk, and red meat) may not necessarily exacerbate IBD-related symptoms as participants reported symptoms that lasted only 24 hours upon ingesting those food types. Of particular interest is the large number of foods that are avoided simply due to personal preference by many respondents (legumes, processed meat, tea or coffee, alcohol, popcorn). In previous reports, these foods have been identified as “trigger foods” for IBD exacerbation by patients.20 Our study suggests that many of these foods are excluded from the diet due to personal preferences and may not be related to their IBD. Persons with IBD have strong beliefs about the role diet may play on the exacerbation of symptom.21 Our data underscore the importance of considering a person’s belief system on food and symptoms despite the lack of evidence from prospective controlled trials to support a direct link to dietary beliefs and exacerbation of IBD symptoms. A primary approach to providing advice on food avoidance for persons with IBD should incorporate food diaries to track symptoms (whereby patients then become empowered to control their own symptoms), while dietitians/healthcare providers have a tool to identify dietary nutrition deficiencies and make recommendations appropriately. Interestingly, some of the foods that persons with IBD in our present study reported as “always avoiding” (legumes, milk, fruit) are considered high FODMAP foods (fermentable, oligo-, di-, mono-saccharides and polyols). High-carbohydrate diets and diets high in FODMAPs have been suggested to be associated with abdominal symptoms in functional GI disorders, implying that a low FODMAP diet may be warranted in patients with IBD who report functional gut symptoms.22,23
Ballegaard et al24 identified that 65% of persons with IBD report being intolerant to 1 or more food items compared with only 14% of healthy controls. The foods most often blamed for causing diarrhea and abdominal pain include onions, cabbage, apples, strawberries, citrus fruit, beef, and smoked meat. Among a Canadian IBD cohort, 34 foods have been cited as never eaten by 40% or more of persons with IBD.25 These foods include milk (whole and chocolate), artificial sweetener, diet soft drinks, beer, decaffeinated coffee, liquor, wine, Mexican food, fried fish, liver, bran cereal, blackberries, and apricots. More systematic evaluation is needed to determine to what degree food types trigger symptoms in the absence of inflammation in persons with IBD vs triggering definite inflammation, as well as to what degree food is avoided as a preference or taste preference vs unpleasant effects of the ingestion. Of interest, 15% of persons with IBD in our study who avoided milk responded that this was secondary to advice received from a health professional. While lactose intolerance may occur as a result of small intestinal inflammation, causing loss of brush-border lactase activity, we did not gather information on why lactose avoidance was recommended or which type of health professional was providing this guidance. Avoidance of nutrient-dense foods is a particular concern for the management of IBD. Although our study did not analyze for micronutrient deficiencies, avoidance of salad and raw vegetables in 10% of all participants with IBD and in 46% of those with active disease suggests some risk of dietary deficiency of vitamins C and A, folate, and fiber for many patients, as these foods are key sources of these nutrients. Avoidance of milk products in 12% of all IBD and 29% of active IBD may lead to a potential deficiency in calcium and vitamin D among a group known to be at risk for osteoporosis.2 Micronutrient deficiency of vitamin D was not analyzed in our present study; however, in our previous work, 44% of patients with IBD had vitamin D deficiency.4 Considering that persons with IBD are at risk or may already be malnourished, care providers should ensure these nutrients are provided through other food or supplemental sources. We report that there is a significantly greater intake of sports drinks and sweetened drinks for those with active IBD compared with those with inactive disease. Whether this was a cause or by-product of the disease activity itself cannot be determined from these data. Studies have postulated that an increased intake of sugar may be caused by the disease itself as a result of a decreased ability to taste sweet substances from zinc deficiency.26 We were unable to report on zinc deficiency, but in our previous work, we reported a trend to a higher prevalence of zinc deficiency in active vs inactive IBD.4 Alternatively, a diet high in sugar may simply be a marker for another dietary excess or deficiency. Large percentages of participants with active IBD avoided nuts and seeds, fruits, vegetables, and milk products; hence, a situation may emerge where many nutrientdense food groups are avoided and sugar is increased. Highsugar products may also be a source of easy-to-ingest calories
Downloaded from pen.sagepub.com at GEORGIAN COURT UNIV on April 16, 2015
Vagianos et al
7
for individuals who have reduced their intake but need to maintain adequate hydration and calories. The high intake of soft drinks is of interest since carbonated beverages are a modern phenomenon, and if these were ingested to excess before the onset of disease, it may have implications for disease etiology. Mayberry et al26 reported that patients with IBD added significantly more teaspoons of sugar per day to their diet compared with controls. Silkoff et al27 reported that patients with CD consumed significantly more grams of carbohydrate per day compared with controls. In 2004, Sakamoto et al8 used a validated semiquantitative food frequency questionnaire (FFQ) to identify that in the preillness diet of patients with IBD, there was an increased consumption of sugars, sweeteners, and sweets and that these were positively associated with an increased risk of CD and UC. Recall bias limits the interpretation and validity of the results.28 Ideally, prospective dietary intervention studies are more reliable. Interestingly, in only 1 intervention study were patients with IBD and controls provided either a high-fiber/low-sucrose diet or a low-fiber/unrestricted sucrose diet.29 In this study, there were no significant differences between the dietary groups in terms of surgery, hospital admissions, or symptom exacerbation. We, too, could not determine as to when persons with IBD increased their sugar intake in relation to their disease becoming active. Further exploration of sugar intake habits (particularly prior to disease onset) is warranted, since the relative balance of nutrients ingested can affect the gut microbiome and could be considered a trigger of IBD.30
Conclusions This study is one of the few to document what patients with IBD typically eat. Previous work from our group has shown that although the vast majority (80%–89%) of persons diagnosed with IBD view nutrition information as being very important, only 8%–16% of patients report receiving adequate information on nutrition in the early period following diagnosis, despite a strong interest in obtaining this knowledge.31 Our study shows that persons with IBD will avoid particular foods, oftentimes nutrient-rich foods, based on symptoms and personal preference, while sugar-laden beverages may be displacing other foods that are higher in nutrients. The overall diet of our IBD sample differed to some extent from a general non-IBD community sample. However, deficiencies exist in both groups, flagging the need to assess the diet and address deficiencies, particularly for the chronic disease group given other health risks. Considering the risk of malnutrition among people with IBD, nutrition education by trained dietitians as part of the IBD team is imperative to address food avoidance and overall balanced nutrition as part of treating and preventing nutrition deficiencies among this group.
References 1. Prince A, Whelan K, Moosa A., Lomer MCE, Reidlinger D. Nutritional problems in inflammatory bowel disease: the patient perspective. J Crohns Colitis. 2011;5:443-450. 2. Lomer MCE. Symposium 7: Nutrition in inflammatory bowel disease. Dietary and nutritional considerations for inflammatory bowel disease. Proc Nutr Soc. 2011;70:329-335. 3. Husain A, Korzenik JR. Nutritional issues in inflammatory bowel disease. Semin Gastrointest Dis. 1998;9:21-30. 4. Vagianos K, Bector S, McConnell J, Bernstein CN. Nutrition assessment of patients with inflammatory bowel disease. JPEN J Parenter Enteral Nutr. 2007;31:311-319. 5. O’Sullivan M, O’Morain C. Nutrition in inflammatory bowel disease. Best Pract Res Clin Gastroenterol. 2006;20:561-573. 6. Miller B, Fervers F, Rohbeck R, Strohmeyer G. Sugar consumption in patients with Crohn’s disease [in German]. Verh Dtsch Ges Inn Med. 1976;82:922-924. 7. Martini GA, Brandes JW. Increased consumption of refined carbohydrates in patients with Crohn’s disease. Klin Wschr. 1976;54:367-371. 8. Sakamoto N, Kono S, Wakai K, et al. Dietary risk factors for inflammatory bowel disease: a multicentre case-control study in Japan. Inflamm Bowel Dis. 2004;11:154-162. 9. Chapman-Kiddell CA, Davies PSW, Gillen L, Radford-Smith GL. Role of diet in the development of inflammatory bowel disease. Inflamm Bowel Dis. 2012;16:137-151. 10. Bernstein CN, Blanchard JF, Rawsthorne P, et al. The epidemiology of Crohn’s disease and ulcerative colitis in a central Canadian province: a population-based study. Am J Epidemiol. 1999;149:916-924. 11. Clara I, Lix LM, Walker JR, et al. The Manitoba IBD index: evidence for a new and simple indicator of IBD activity. Am J Gastroenterol. 2009;104:1754-1763. 12. Longobardi T, Walker JR, Graff LA, Bernstein CN. Health service utilization in IBD: comparison of self-report and administrative data. BMC Health Serv Res. 2011;11:137. 13. Graff LA, Walker JR, Lix L, et al. The relationship of inflammatory bowel disease type and activity to psychological functioning and quality of life. Clin Gastroenterol Hepatol. 2006;4:1491-1501. 14. Bryan SN, St-Denis M, Wojitas D. Canadian Health Measures Survey: clinic operations and logistics. Health Rep. 2007;18(suppl):53-69. 15. Day B, Langlois R, Tremblay M, Knoppers BM. Canadian Health Measures Survey: ethical, legal and social issues. Health Rep. 2007;18(suppl): 37-52. 16. Giroux S. Canadian Health Measures Survey: Sampling strategy overview. Health Rep. 2007;18(suppl):31–36. 17. Tremblay M, Wolfson M, Conner Gorbe S. Canadian Health Measures Survey: rationale, background and overview. Health Rep. 2007;18(suppl):7-20. 18. Garriguet D. Canadians’ eating habits. Health Rep. 2007;18:17-32. 19. Vatanparast H, Dolega-Cieszkowski JH, Whiting SJ. Many adult Canadians are not meeting current calcium recommendations from food and supplement intake. Appl Physiol Nutr Metab. 2009;34:191-196. 20. Zallot C, Quilliot D, Chevaux J, et al. Dietary beliefs and behavior among inflammatory bowel disease patients. Inflamm Bowel Dis. 2013;19:66-72. 21. Hou JK, Lee D, Lewis J. Diet and inflammatory bowel disease: review of patient-targeted recommendations [published online October 6, 2013]. Clin Gastroenterol Hepatol. 22. Gibson PR, Shepherd SJ. Evidence-based dietary management of functional gastrointestinal symptoms: the FODMAP approach. J Gastroenterol Hepatol. 2010;25:252-258. 23. Gearry RB, Irving PM, Barrett JS, Nathan DM, Shepherd SJ, Gibson PR. Reduction of dietary poorly absorbed short-chain carbohydrates (FODMAPs) improves abdominal symptoms in patients with inflammatory bowel disease—a pilot study. J Crohns Colitis. 2009;3:8-14.
Downloaded from pen.sagepub.com at GEORGIAN COURT UNIV on April 16, 2015
8
Journal of Parenteral and Enteral Nutrition XX(X)
24. Ballegaard M, Bjerstrom A, Brondum S, Hylander E, Jensen L, Ladefoged K. Self reported food intolerance in chronic inflammatory bowel disease. Scand J Gastroenterol. 1997;32:569-571. 25. Joachim G. Responses of people with inflammatory bowel disease to foods consumed. Gastroenterol Nurs. 2000;23:160-167. 26. Mayberry JF, Rhodes J, Newcombe RG. Increased sugar consumption in Crohn’s disease. Digestion. 1980;20:323-326. 27. Silkoff K, Hallak A, Yegena L, et al. Consumption of refined carbohydrate by patients with Crohn’s disease in Tel-Aviv-Yafo. Postgrad Med J. 1980;56:842-846.
28. Krishnan A, Korzenik JR. Inflammatory bowel disease and environmental influences. Gastroenterol Clin. 2002;31:21-39. 29. Ritchie JK, Wadsworth J, Lennard-Jones JE, Rogers E. Controlled multicentre therapeutic trial of an unrefined carbohydrate, fibre rich diet in Crohn’s disease. BMJ. 1987;295:517-520. 30. Ananthakrishnan A. Environmental risk factors for inflammatory bowel disease. Gastroenterol Hepatol. 2013;9:367-374. 31. Wong S, Walker JR, Carr R, et al. The information needs and preferences of persons with longstanding inflammatory bowel disease. Can J Gastroenterol. 2012;26:525-531.
Downloaded from pen.sagepub.com at GEORGIAN COURT UNIV on April 16, 2015
