Food intake and blood cholesterol levels of community-based adults with mood disorders

The sampling frame for this study consisted of adult (> 18 years) members of the Mood Disorders Association of British Columbia (MDABC) who resided in BC's lower mainland. Using a computerized random number generation analysis tool, the names of 146 randomly selected MDABC adult members were drawn and sent a letter of invitation to participate in a cross-sectional survey of food intake (73 invited in the summer and 73 in the fall/winter). When an individual declined participation, the interviewer attempted to complete a non-response questionnaire asking about lifestyle habits.

When an individual agreed to participate, the research coordinator explained the study over the phone and trained the respondent to complete a three-day food record (reporting three non-consecutive days, two of which were weekdays and one was a weekend day). Participants were also asked if they would provide a blood sample to measure their cholesterol levels.

At the first appointment, participants read and signed the consent form approved by the University of Calgary's Conjoint Health Research Ethics Board. Then they met with a trained clinical interviewer who administered the Structured Clinical Interview for DSM-IV Axis I Disorders [11], the Global Assessment of Functioning (GAF) Scale (GAF) [12] which measured social, occupational, and psychological functioning (scale of 0 to 100), the Hamilton Depression Scale (Ham-D) [13] and, the Young Mania Rating Scale (YMRS) [14]. Those with conditions associated with psychotic symptoms, dementia, thyroid dysfunction, or neuro-degeneration were excluded. Eligible individuals were subsequently interviewed by a registered dietitian who reviewed the participant's food record, administered a validated food frequency questionnaire [15], and asked selected demographic and health related questions. The procedures for collecting dietary data followed that of the British Columbia Nutrition Survey (BCNS), a study of 1823 British Columbians aged 19-84 years. Detailed reports are available at http://​www.​health.​gov.​bc.​ca/​library/​publications/​year/​2004/​bcnutritionsurve​y.

The software program Food Processor SQL [16], which contains the Canadian Nutrient File [17], was used to analyze the food data. The data were compared with the recommended serving ranges of the national guidelines for Canada, Eating Well with Canada's Food Guide (Food Guide) [18], the North American Dietary Reference Intakes [19], and data from the BCNS [15, 20]. Components of the Dietary Reference Intakes that were used were 1) Adequate Macronutrient Distribution Ranges (AMDR): range of intakes for a particular energy source and expressed as a percentage of total energy intake that is associated with reduced risk of chronic disease while providing adequate intakes of essential nutrients, and 2) The Estimated Average Requirement (EAR): a nutrient intake value that is estimated to meet the requirements of half the healthy individuals in a group. If an EAR was not available, Adequate Intakes (AI) were used instead. The AI is a recommended daily intake level based on observed or experimentally determined approximations by a group (or groups) of healthy people [19].

Statistical comparisons between the sample and BCNS as well as analyses within the study group of socio-demographic variables, type of mood disorder and dietary attributes were carried out using Student's t-tests, Mann-Whitney two-sample statistics, binomial tests of two proportions, Pearson's or Spearman's rho correlations, and ANOVA where appropriate.

The overall response rate was 75% (number of participants/number of possibly eligible participants based on the initial phone screening). Of those who declined participation, 44% (11/25) answered the non-response survey: no differences in lifestyle variables (e.g., smoking) were found when compared to the respondents. Sociodemographic characteristics of the sample with comparisons to the BCNS are outlined in Table 1. Slightly more than half of the sample had bipolar disorder (i.e., 59.8% had bipolar disorder and 40.2% had depressive disorder) with relatively high GAF scores (62.7 ± 14.7) suggesting most were only mildly to moderately impaired. The Ham-D scores indicated that 14 (14.4%) were severely depressed, 14 (14.4%) were experiencing moderate depressive symptoms, 44 (45.4%) were mildly depressed, and the remainder were asymptomatic. Based on YMRS scores, 2.6% (all females) were experiencing symptoms of mania. The sample tended to be overweight or obese (67% had a BMI ≥ 25). The main psychiatric medications taken by the sample included various antidepressants (72.9%) and mood stabilizers (52.9%).

Analysis of nutrient intakes and demographic attributes indicated significantly higher intakes of protein [F (3, 93) = 4.40, p < 0.05] and fibre [F (3, 93) = 3.07, p < 0.05] based on education, and significantly lower intakes for energy (t = -2.19, SE = 195.03, 95% CI = -815.01 to -40.63, p < 0.05) and fibre (t = -2.40, SE = 2.87, 95% CI = -12.60 to -1.19, p < 0.05) for those who were considered single (i.e., widowed, divorced, separated, or never married). All nutrients were also compared based on age (18 to 34, 35 to 49, and 50+ years for food group analysis and 19 to 30, 31 to 50 and 51 to 70 years for nutrient analysis to be consistent with the standards used). None of the study participants exceeded the age of 70; therefore the subsample of 1320 participants aged 19-70 from the full 1823 BCNS sample was used as the comparison group. Only 19% of participants indicated they were following a therapeutic diet; this subsample had significantly lower carbohydrate intake (255 g ± 103 vs 341 g ± 145; t = 2.40, p < 0.05, 95% CI 14.78 to 156.07).

Eighty-nine (92%) participants claimed to have heard of Canada's Food Guide, but less than one-third (n = 30; 31%) actually used it; there were no significant differences in major nutrient intakes between those who reported that they did and did not use the Food Guide. When compared to the 2007 Eating Well with Canada's Food Guide [18] (Table 2), intakes of the study sample within the recommended serving ranges of each of the food groups were significantly lower than the BCNS (p's < 0.05 to 0.0001) with the exception of milk and milk alternatives. Analyses according to age groupings and gender indicated significantly more males between the ages of 18 to 34 and 35 to 49 years ate less than the minimum recommendations for fruits and vegetables (p < 0.05) and grains (p < 0.001) compared to the BCNS. Significantly fewer females between the ages of 35 to 49 years consumed the recommended 5 to 12 servings of fruits and vegetables per day (p < 0.05) compared to the BCNS. There also tended to be more males across all age groups consuming less than 100 g of meat and alternatives per day (range of p < 0.05 to p < 0.0001). In summary, only 6.2% of the sample consumed the minimum recommended level of servings across all food groups. In order to meet the Food Guide standard, at least 45% of the sample needed to add 2-3 more daily servings of vegetables and fruit; about 15% required 2 extra grain servings daily (30% need 1 extra grain serving); and about half would need to add 1 serving/day of milk and alternatives.

Median energy intakes (kilocalories) of the study sample were 2520 (25^th percentile 1920; 75^th percentile 3050); there was slight under-reporting. The study sample had significantly higher total fat intakes (45% vs 19% of the BCNS, p < 0.0001) and lower α-linolenic acid intakes (91% vs 23% of BCNS, p < 0.001) (Table 2). Median α-linolenic intakes did not meet the AI of 1.6. grams/day for males and 1.1 g/day for females and intakes were consistently lower than the BCNS (Figure 1). About 10% of the sample were below the EAR (0.66 g/kg/day) for protein, which is considered the minimum continuing intake of dietary protein required for body nitrogen equilibrium [19].

Age and gender analysis revealed that a significantly lower proportion of males (12%) between 31 to 50 years had fat intakes that were 30 to 35% of total calories (41% for BCNS). In addition, 2% (all females 51 to 70 years) were consuming less than the EAR for carbohydrates. Median carbohydrate intakes (grams) of females 19 to 30 years (488 vs 257, p < 0.05) and 31 to 50 years (326 vs 219, p < 0.05) were significantly higher than the BCNS. Only females 19 to 30 years met the AI for fibre. Fibre intakes may be underestimated by about 5 g/day, as inulin and fructo-oligosaccharides were not included [19].

Compared to the BCNS, the study sample had significantly greater intakes of less healthy foods such as high-fat whole grain products (e.g., granola cereals), processed meats, and higher sugar, fat or salt foods (range of p's < 0.01 to 0.00001) (Table 3). When comparing those who were and were not taking psychiatric medications, there were no differences found for food group, fat and sugar intakes. Participants were also asked about the frequency of meal and snack food intake from sources outside the home based on the three day food record. Of the 1746 total meals and snacks consumed, 11% were from fast food sources, 10% from vending machines or snack bars, 18% from restaurants, take-out or deli. When asked who prepared most of the foods they ate that day, males reported fewer food preparation days (n = 25/84; 29.8%) than females (199/207; 96.1%) (p < 0.0001).

Of the 48 participants who agreed to blood tests, 44% had mild hypercholesterolemia (> 5.2 and ≤ 6.2 mmol/L) and 21% had hypercholesterolemia (> 6.2 mmol/L). The prevalence of total cholesterol levels exceeding 5.2 mmol/L in the Canadian population is 41% [21]. Blood cholesterol levels did not differ based on psychiatric medication use, type of mood disorder or BMI status.