INTRODUCTION
Psychological attitudes are significantly related to health outcomes, and a growing body of evidence has demonstrated the salutary role of positive psychological factors on health.1 These include factors such as subjective wellbeing,2–4 positive affect,5,6 emotional vitality,7,8 and dispositional optimism,9,10 all of which have been associated with favorable biological markers and lower incidence and severity of coronary heart disease3,11,12 in diverse populations.
Among myriad psychological factors associated with good health, dispositional optimism (or positive expectancies for the future)13 has emerged as a key trait for healthy behavior. Optimistic people have been shown to engage and persist in health behaviors such as regular physical activity, moderate alcohol consumption, and smoking avoidance,12,14 and are more likely to adhere to medical advice.15,16
In the largest study to date on dispositional optimism and health, optimism prospectively and independently predicted lower incident coronary heart disease risk and all-cause mortality in 100,000 postmenopausal women participating in the Women’s Health Initiative (WHI).12 In this study, health behaviors such as not smoking and being physically active explained part of this observed relationship. Dietary information was not included in that study, but a more recent study of middle-aged US men and women found that optimists had higher serum concentrations of antioxidants and that this was explained in part by their healthier diet.17 The findings raise intriguing questions about whether optimism can be used as a marker to predict which individuals may make healthy behavior changes, or whether optimism can enhance the effects of behavior change programs designed to impact chronic disease risk factors. In another WHI study, optimism was as a novel predictor of adherence to a low-fat eating pattern following a one-year dietary intervention.16 This intervention focused primarily on decreasing fat intake, although participants were also advised to consider other changes in diet including increasing intake of vegetables, fruits, and fiber.18 It remains unclear how optimism relates to changes in dietary behavior over time.
Optimists exhibit adaptive coping styles and robust self-regulation, both of which are critical to achieving successful behavior change.19 An improved understanding of the relationships between dispositional optimism, dietary intake and changes in diet can help explain how this trait may be used to prospectively influence nutrition and health.
The purpose of this study was to investigate the potential role of optimism as a predictor of diet quality and changes in diet quality in postmenopausal women, ages 50 to 79, enrolled in the WHI Observational Study (OS) and the WHI Clinical Trial (CT). Study objectives were threefold: first, to characterize relationships between optimism, diet quality, and lifestyle and demographic factors in women enrolled in the WHI OS and CTs; second, to evaluate whether optimism was associated with change in diet quality over one year in women enrolled in the WHI Diet Modification Clinical Trial (DM-CT); and, finally, to investigate potential interactions between baseline optimism and lifestyle and demographic factors on diet quality change in intervention (DM-CT) participants.
METHODS
Study Design and Sample
The WHI is a large clinical investigation of strategies for the prevention and control of common causes of morbidity and mortality, including cancer, cardiovascular disease (CVD), and osteoporotic fractures among postmenopausal women, between 50 and 79 years old. Briefly, the WHI consisted of a set of overlapping, randomized CTs (hormone therapy, diet modification, and calcium/vitamin D, which collectively enrolled 68,132 women, and an Observational Study (OS) that enrolled 93,676 women into a prospective epidemiological cohort. Eligibility criteria also included ability and willingness to provide written informed consent and planning to reside in the area for a minimum three years after enrollment. Recruitment methods have been described in more detail elsewhere.20 Eligible women could enroll in one, two, or all three of the CTs; women ineligible or unwilling to join the CTs were invited to participate in the OS. Written informed consent was obtained from all study participants prior to study enrollment, and each of the trials were approved by the Institutional Review Boards of the 40 participating institutions.
The present analysis includes women with reliable dietary data (energy intake between 600 and 5000 kcal/d) and available optimism data. To characterize relationships between optimism, diet quality, and lifestyle and demographic factors in women enrolled in the WHI OS and CTs, the cross-sectional analysis of diet quality and optimism included 87,630 OS participants and 65,360 CT participants (Study Objective 1). To evaluate whether optimism was associated with change in diet quality over one year in women enrolled in the DM-CT, the analysis testing the association between optimism and diet quality score at baseline and year 1 included 13,645 DM-CT participants randomized to the diet intervention arm with available diet data at year 1 (Objective 2). Finally, to examine potential interactions between baseline optimism and lifestyle and demographic factors on diet quality change in DM-CT participants, the test for interaction between optimism and trial arm, smoking status, age, and race/ethnicity included women randomized to the DM-CT intervention arm (n=13,645) and the control arm (n=20,242) (Objective 3).
Measurement of Optimism
Optimism was assessed at baseline using the 6-item Life Orientation Test-Revised (e.g., “I’m always hopeful about my future” and “In unclear times, I usually expect the best”), with a Cronbach’s alpha of 0.78, test-retest reliability of 0.68 and adequate predictive and discriminant validity.13 A summary score was calculated from the 6 components, coded from 1=strongly disagree to 5=strongly agree. Scores ranged from 6 to 30 where a higher score indicated greater optimism.13
Diet Quality Assessment
Dietary intake was assessed using a food frequency questionnaire (FFQ) developed and validated by the Fred Hutchinson Cancer Research Center (Seattle, WA).21 This 122-item FFQ was designed for the WHI to include foods representative of the national food supply as well as capture regional and ethnic food selections. Respondents were asked to indicate the portion consumed and frequency of consumption over the previous 3-month period. Nutrient intake was derived using software developed at Fred Hutchinson and the Nutrition Data System for Research (NDS-R) food and nutrient database (University of Minnesota Nutrition Coordinating Center).22 All WHI participants completed the FFQ during the screening period, and DM-CT participants completed a second survey one year later.
The Alternate Healthy Eating Index (AHEI)23 is a tool that when applied to dietary datasets, can be used to evaluate multiple and related aspects of dietary intake important to chronic disease prevention. The AHEI is a composite numerical measure24,25 previously used in this population to assess diet quality26 and a robust predictor of chronic disease risk in older men and women including CVD24,26 and Type 2 diabetes.27 It is comprised of 11 components, each scored from 0 (non-adherence) to 10 points (perfect adherence): vegetables, fruit, whole grains, sugar-sweetened beverages and fruit juice, nuts and legumes, red/processed meat, trans fat, long-chain (omega-3) fats, polyunsaturated fatty acids, sodium, and alcohol. Established scoring procedures were followed to obtain a total dietary score ranging from 0 to 110.23,24 Mean ± standard deviation AHEI score for all WHI participants in this study was 44.9 ± 10.2.
Lifestyle and Demographic Data
Demographic data collected from participants at baseline included age, self-reported race/ethnicity, education (≤ high school, any post-secondary, ≥ college), family income (< $20k, $20 to < 35k, $35 to < 50k, $50 to < 75k, ≥ $75k), and weekly attendance at religious services (yes, no). Medical data included history of diabetes treatment (yes, no), current hypertension treatment (yes, no), high cholesterol-requiring medication (ever, never), current smoking (yes, no), physical activity (< 2.5, ≥ 2.5 MET-hr/wk), hormone therapy use (never, former, current), waist circumference (< 88, ≥ 88 cm), and body mass index (BMI) (< 30, ≥ 30 kg/m2).
Statistical Analyses
Baseline characteristics were compared across quintiles of AHEI scores or tertiles of optimism scores using linear (continuous variables), logistic (binary variables), or multinomial logistic (categorical variables) regression models with each variable of interest as a function of AHEI score quintiles or optimism tertiles (study objective 1). The association between optimism score (tertiles as a categorical variable) and baseline AHEI score (continuous variable), and change in AHEI score between baseline and year 1 in the DM-CT (continuous variable), were tested using multivariate linear regression, adjusted for potential confounders (see footnotes, Tables 1, 2). Optimism tertiles were subsequently modeled as an ordinal variable to test for trends (objective 2). The potential interaction between optimism and trial arm on AHEI change was tested using a likelihood ratio test. Likewise, potential interactions between optimism and race/ethnicity, age, and smoking on AHEI change for women in the intervention arm were tested using likelihood ratio tests (objective 3). All analyses were conducted using Stata 12.1 (StataCorp, College Station, TX).
Table 1.
Baseline characteristics of Women’s Health Initiative (WHI) participants, by Alternate Healthy Eating Index (AHEI)-2010 quintiles1
| Characteristic | Quintile
|
P trend2 | ||||
|---|---|---|---|---|---|---|
| 1 12.7–36.1 (n = 30,598) |
2 36.1–41.7 (n = 30,598) |
3 41.7–46.9 (n = 30,598) |
4 46.9–53.4 (n = 30,598) |
5 53.4–92.0 (n = 30,598) |
||
| AHEI, median (interquartile range) | 31.4 (29.2– 34.3) | 39.0 (37.6–40.4) | 44.3 (43.0–45.6) | 50.0 (48.3–51.5) | 59.8 (55.6–62.6) | |
| Age (y), mean ± SD | 62.3 ± 7.2 | 63.1 ± 7.2 | 63.2 ± 7.2 | 63.6 ± 7.2 | 63.7 ± 7.2 | < 0.001 |
| Waist circumference (cm), mean ± SD | 90.3 ± 14.6 | 88.2 ± 14.0 | 86.5 ± 13.3 | 85.0 ± 13.3 | 82.3 ± 12.4 | < 0.001 |
| BMI (kg/m2), mean ± SD | 29.6 ± 6.4 | 28.7 ± 6.1 | 28.0 ± 5.8 | 27.4 ± 5.6 | 26.2 ± 5.2 | < 0.001 |
| Energy intake (kcal), mean ± SD | 1764 ± 654 | 1646 ± 647 | 1631 ± 658 | 1607 ± 640 | 1571 ± 595 | < 0.001 |
| Physical activity (MET-hr/wk), mean ±SD | 8.4 ± 11.3 | 10.4 ± 12.2 | 11.9 ± 12.7 | 14.0 ± 14.0 | 17.7 ± 15.9 | < 0.001 |
| Race/ethnicity, % | < 0.001 | |||||
| American Indian or Alaskan Native | 0.57 | 0.47 | 0.45 | 0.33 | 0.26 | |
| Asian or Pacific Islander | 1.09 | 2.02 | 2.64 | 3.18 | 3.73 | |
| Black or African American | 12.5 | 9.28 | 7.61 | 6.53 | 5.71 | |
| Hispanic/Latina | 5.40 | 4.22 | 3.67 | 2.83 | 2.07 | |
| Non-Hispanic white | 79.4 | 83.0 | 84.5 | 86.0 | 87.0 | |
| Other | 0.98 | 1.05 | 1.15 | 1.13 | 1.22 | |
| Education, % | < 0.001 | |||||
| ≤ High school diploma | 31.3 | 25.8 | 21.8 | 17.7 | 12.7 | |
| Any post-secondary education | 40.6 | 40.1 | 38.2 | 37.4 | 33.4 | |
| ≥ College graduate | 28.2 | 34.1 | 40.0 | 45.0 | 53.9 | |
| Marital status, % | < 0.001 | |||||
| Never married | 4.20 | 4.09 | 4.43 | 4.42 | 4.96 | |
| Divorced or separated | 16.5 | 15.1 | 14.8 | 15.6 | 16.9 | |
| Widowed | 17.9 | 17.7 | 17.2 | 16.5 | 15.5 | |
| Married | 60.1 | 61.7 | 62.1 | 61.7 | 60.6 | |
| Marriage-like relationship | 1.38 | 1.48 | 1.50 | 1.75 | 2.11 | |
| Family income, % | < 0.001 | |||||
| < $20,000 | 21.9 | 18.7 | 15.5 | 13.4 | 10.9 | |
| $20,000 to < $35,000 | 27.8 | 26.0 | 24.9 | 22.8 | 20.0 | |
| $35,000 to < $50,000 | 21.0 | 21.0 | 20.9 | 20.9 | 19.5 | |
| $50,000 to < $75,000 | 17.4 | 19.4 | 20.0 | 21.3 | 22.5 | |
| ≥ $75,000 | 11.9 | 15.1 | 18.7 | 21.6 | 27.1 | |
| Smoking, % | < 0.001 | |||||
| Never | 54.0 | 53.3 | 51.5 | 49.0 | 46.3 | |
| Past | 35.8 | 38.8 | 41.8 | 45.4 | 49.7 | |
| Current | 10.2 | 7.95 | 6.72 | 5.63 | 4.03 | |
| Religious service/church attendance, % | < 0.001 | |||||
| Not at all in the past month | 30.6 | 30.7 | 32.2 | 34.6 | 40.5 | |
| Once in the past month | 8.24 | 8.64 | 8.77 | 9.40 | 10.2 | |
| 2–3 times in the past month | 12.1 | 12.1 | 12.5 | 12.4 | 12.4 | |
| Once a week | 32.2 | 32.4 | 31.2 | 29.8 | 25.4 | |
| 2–6 times a week | 15.6 | 14.9 | 14.1 | 12.6 | 10.4 | |
| Every day | 1.21 | 1.24 | 1.29 | 1.29 | 1.12 | |
| History of diabetes mellitus treatment, % | 4.37 | 4.63 | 4.48 | 4.18 | 3.69 | < 0.001 |
| Hormone therapy use, % | < 0.001 | |||||
| Never | 37.0 | 34.7 | 32.8 | 31.3 | 29.6 | |
| Past | 24.7 | 23.4 | 23.3 | 22.5 | 21.7 | |
| Current | 38.3 | 41.9 | 43.9 | 46.2 | 48.6 | |
| Hypertension, % | < 0.001 | |||||
| Never hypertensive | 61.4 | 63.6 | 65.8 | 68.3 | 72.3 | |
| Currently untreated hypertensive | 8.40 | 8.26 | 8.24 | 7.81 | 7.74 | |
| Currently treated hypertensive | 30.2 | 28.2 | 26.0 | 23.9 | 20.0 | |
| High cholesterol requiring pills, % | 13.4 | 14.1 | 14.3 | 14.4 | 13.3 | 0.913 |
| Depression score (short CES-D/DIS) | < 0.001 | |||||
| Low (< 0.009) | 71.4 | 74.3 | 75.3 | 76.6 | 77.9 | |
| Medium (0.009–0.06) | 14.6 | 14.3 | 14.2 | 13.8 | 13.5 | |
| High (> 0.06) | 14.0 | 11.5 | 10.5 | 9.64 | 8.61 | |
Quintile 1 represents the least-healthy scores, whereas quintile 5 represents the most-healthy scores
P values calculated from linear (continuous variables), logistic (binary variables), or multinomial logistic (categorical variables) regression models with the variable of interest as a function of AHEI quintile
Table 2.
Baseline characteristics of Women’s Health Initiative (WHI) participants, by optimism score tertiles1
| Characteristic | Optimism score tertiles
|
P trend2 | ||
|---|---|---|---|---|
| 1: 6–22 (n = 57,998) | 2: 23–25 (n = 55,557) | 3: 26–30 (n = 39,435) | ||
| Age (y), mean ± SD | 63.3 ± 7.3 | 63.3 ± 7.1 | 62.8 ± 7.1 | < 0.001 |
| Waist circumference (cm), mean ± SD | 87.6 ± 14.3 | 86.1 ± 13.6 | 85.2 ± 13.4 | < 0.001 |
| BMI (kg/m2), mean ± SD | 28.4 ± 6.2 | 27.8 ± 5.8 | 27.5 ± 5.7 | < 0.001 |
| Energy intake (kcal), mean ± SD | 1644 ± 673 | 1641 ± 624 | 1649 ± 623 | 0.310 |
| Physical activity (MET-hr/wk), mean ±SD | 11.1 ± 13.1 | 12.7 ± 13.6 | 14.2 ± 14.6 | < 0.001 |
| Race/ethnicity, % | < 0.001 | |||
| American Indian or Alaskan Native | 0.54 | 0.33 | 0.35 | |
| Asian or Pacific Islander | 3.61 | 2.18 | 1.44 | |
| Black or African American | 8.88 | 7.64 | 8.49 | |
| Hispanic/Latina | 5.21 | 2.82 | 2.49 | |
| Non-Hispanic white | 80.4 | 86.0 | 86.4 | |
| Other | 1.35 | 1.00 | 0.89 | |
| Education, % | < 0.001 | |||
| ≤ High school diploma | 29.3 | 19.8 | 13.9 | |
| Any post-secondary education | 38.5 | 38.8 | 35.7 | |
| ≥ College graduate | 32.2 | 41.3 | 50.4 | |
| Marital status, % | < 0.001 | |||
| Never married | 4.79 | 4.25 | 4.12 | |
| Divorced or separated | 16.8 | 14.6 | 16.0 | |
| Widowed | 18.8 | 16.4 | 15.0 | |
| Married | 58.1 | 63.2 | 63.1 | |
| Marriage-like relationship | 1.59 | 1.57 | 1.83 | |
| Family income, % | < 0.001 | |||
| < $20,000 | 22.2 | 13.5 | 10.7 | |
| $20,000 to < $35,000 | 26.8 | 24.1 | 20.9 | |
| $35,000 to < $50,000 | 19.7 | 21.8 | 20.4 | |
| $50,000 to < $75,000 | 17.5 | 21.0 | 22.7 | |
| ≥ $75,000 | 13.8 | 19.6 | 25.3 | |
| Smoking, % | < 0.001 | |||
| Never | 50.2 | 50.6 | 51.9 | |
| Past | 41.4 | 43.2 | 42.3 | |
| Current | 8.35 | 6.17 | 5.79 | |
| Religious service/church attendance, % | < 0.001 | |||
| Not at all in the past month | 35.9 | 32.3 | 32.5 | |
| Once in the past month | 9.48 | 8.82 | 8.75 | |
| 2–3 times in the past month | 12.5 | 12.4 | 11.9 | |
| Once a week | 29.5 | 31.1 | 30.0 | |
| 2–6 times a week | 11.6 | 14.2 | 15.5 | |
| Every day | 1.10 | 1.22 | 1.42 | |
| History of diabetes mellitus treatment, % | 5.57 | 3.74 | 3.12 | < 0.001 |
| Hormone therapy use, % | < 0.001 | |||
| Never | 34.7 | 32.4 | 31.6 | |
| Past | 24.5 | 22.7 | 21.7 | |
| Current | 40.7 | 44.9 | 46.8 | |
| Hypertension, % | < 0.001 | |||
| Never hypertensive | 62.9 | 66.7 | 70.6 | |
| Currently untreated hypertensive | 9.02 | 7.72 | 7.24 | |
| Currently treated hypertensive | 28.0 | 25.5 | 22.1 | |
| High cholesterol requiring pills, % | 15.5 | 13.6 | 12.0 | < 0.001 |
| Depression score (short CES-D/DIS) | < 0.001 | |||
| Low (< 0.009) | 63.6 | 79.7 | 85.3 | |
| Medium (0.009–0.06) | 17.3 | 13.2 | 10.7 | |
| High (> 0.06) | 19.2 | 7.07 | 3.98 | |
Tertile 1 represents the least-optimistic scores, whereas tertile 3 represents the most-optimistic scores
P values calculated from linear (continuous variables), logistic (binary variables), or multinomial logistic (categorical variables) regression models with the variable of interest as a function of optimism tertile
The proportion of missing data for each baseline characteristic was as follows: age (0%), waist circumference (0.4%), BMI (0.9%), energy intake (0%), physical activity (4.6%), race/ethnicity (0.2%), education (0.7%), marital status (0.4%), family income (6.4%), smoking (1.1%), religious service/church attendance (0.2%), history of diabetes mellitus treatment (0.1%), hormone therapy use (2.9%), hypertension (5.4%), and high cholesterol requiring pills (5.8%), depression score (2.3%).
RESULTS AND DISCUSSION
Associations Between Optimism, Diet Quality, and Lifestyle and Demographic Factors
Among WHI OS and CT participants, higher baseline AHEI scores were associated with older age, non-Hispanic White race, higher education, higher income, being married or in a marriage-like relationship, use of hormone therapy, higher physical activity, being a past or never-smoker, and lower energy intake (Table 1). AHEI scores were inversely associated with BMI, waist circumference, history of diabetes or hypertension, depression, and attendance at religious services. The inverse relationship between religious services attendance and diet quality was a unique finding in this analysis. The role of religiosity in health has been investigated in many studies with mostly positive associations;28 however, little is known about diet quality and religiosity. A review of twenty studies that examined religion and diet found that 60% of the studies reported a better diet in those who were more religious/spiritual.28
When examining baseline characteristics by baseline optimism scores, high optimism was associated with younger age, lower waist circumference and BMI, and increased physical activity (Table 2). Further, women in the highest tertile of optimism were more likely to have higher income and be non-Hispanic White, college-educated, married or in a marriage-like relationship, a past or never smoker, current hormone therapy users, and without history of depression, hypertension, high cholesterol, or diabetes.
Association of Optimism with Diet Quality and Diet Quality Change by Intervention Trial Arm
Women in the DM-CT intervention and control arms both showed significant positive associations between optimism and AHEI at baseline, and between optimism and change in AHEI between baseline and 1 year (Table 3). Magnitude of change (mean ± SD change of 1.5 ± 8.4 points, data not shown) was similar to other dietary intervention studies using this index to assess change.29 Although the absolute magnitude of AHEI score improvement was greater among intervention arm women (mean AHEI scores of 42.0, 43.3, and 44.3 for optimism tertiles 1–3, respectively, P-trend < 0.001) (Table 3), the association between optimism and diet quality was stronger in the control arm (P interaction = 0.014). Among control participants, the most optimistic women demonstrated a three-fold increase in AHEI score compared with least optimistic women over one year. This finding is consistent with results from a clinical trial of treatment for depression among men and women who had undergone coronary artery bypass surgery, in which the effect of optimism in promoting recovery from depression was seen only among control participants.15 This suggests optimistic attitudes may be particularly important to promote healthy change in the absence of formal behavioral intervention. Progovac and colleagues found that optimism and cynical hostility predicted longitudinal smoking behavior among WHI women who were smoking at baseline.14 As in the case of smoking behavior, understanding the extent to which optimism is associated with longitudinal dietary intake is important when designing tailored behavioral approaches to optimize nutrient intake in older populations.
Table 3.
Association between optimism and change in AHEI-2010 score between baseline and year 1 in the WHI Diet Modification Clinical Trial by trial arm, race/ethnicity, age, and smoking
| Subgroup | AHEI | Optimism score at baseline
|
P trend1,2 | ||
|---|---|---|---|---|---|
| Tertile 1 (low) scores 6–22 | Tertile 2 (middle) scores 23–25 | Tertile 3 (high) scores 26–30 | |||
|
Stratified by Trial Arm
| |||||
| Intervention (n = 13,645) | |||||
| Baseline (mean ± SD) | 42.1 ± 9.4 | 43.3 ± 9.4 | 44.3 ± 9.8 | < 0.001 | |
| Year 1 (mean ± SD) | 43.8 ± 9.1 | 44.8 ± 9.0 | 45.7 ± 9.2 | ||
| Change (95% CI)2 | 1.36 (1.15–1.57) | 1.52 (1.32–1.72) | 1.77 (0.53–2.01) | 0.014 | |
| Control (n = 20,242) | |||||
| Baseline (mean ± SD) | 42.2 ± 9.5 | 43.3 ± 9.5 | 44.1 ± 9.7 | < 0.001 | |
| Year 1 (mean ± SD) | 42.7 ± 9.4 | 44.1 ± 9.6 | 44.9 ± 9.7 | ||
| Change (95% CI)2 | 0.32 (0.15–0.49) | 0.82 (0.66–0.98) | 1.01 (0.81–1.20) | < 0.001 | |
| Test for interaction3, P = 0.014 | |||||
| Stratified by race/ethnicity (Intervention arm only) | |||||
| NHW (n = 11,380) | |||||
| Baseline (mean ± SD) | 42.2 ± 9.4 | 43.5 ± 9.4 | 44.7 ± 9.8 | < 0.001 | |
| Year 1 (mean ± SD) | 43.8 ± 9.1 | 44.9 ± 9.0 | 45.9 ± 9.2 | ||
| Change (95% CI)2 | 1.26 (1.02–1.49) | 1.38 (1.17–1.60) | 1.62 (1.36–1.88) | 0.045 | |
| Black (n = 1,311) | |||||
| Baseline (mean ± SD) | 40.1 ± 9.1 | 40.7 ± 9.1 | 41.8 ± 9.4 | 0.146 | |
| Year 1 (mean ± SD) | 42.4 ± 8.9 | 43.4 ± 9.1 | 44.4 ± 9.3 | ||
| Change (95% CI)2 | 2.10 (1.42–2.79) | 2.58 (1.87–3.29) | 2.98 (2.20–3.75) | 0.105 | |
| Hispanic (n = 402) | |||||
| Baseline (mean ± SD) | 39.5 ± 8.7 | 42.6 ± 8.9 | 42.0 ± 9.5 | 0.142 | |
| Year 1 (mean ± SD) | 42.2 ± 8.1 | 45.0 ± 8.3 | 44.7 ± 7.5 | ||
| Change (95% CI)2 | 2.11 (1.06–3.16) | 3.13 (1.85–4.40) | 2.95 (1.33–4.58) | 0.326 | |
| Asian (n = 345) | |||||
| Baseline (mean ± SD) | 47.6 ± 8.9 | 46.5 ± 10.1 | 43.0 ± 7.9 | 0.005 | |
| Year 1 (mean ± SD) | 48.8 ± 8.8 | 47.0 ± 8.8 | 45.2 ± 8.7 | ||
| Change (95% CI)2 | 1.65 (0.54–2.76) | 0.31 (–0.99–1.62) | 1.07 (–0.98–3.12) | 0.357 | |
| Stratified by age (Intervention arm only) | |||||
| Age 50–59 y (n = 4,835) | |||||
| Baseline (mean ± SD) | 41.5 ± 9.6 | 42.7 ± 9.5 | 43.6 ± 9.8 | 0.001 | |
| Year 1 (mean ± SD) | 43.7 ± 9.2 | 44.7 ± 9.2 | 45.5 ± 9.3 | ||
| Change (95% CI)2 | 2.04 (1.68–2.39) | 1.99 (1.64–2.35) | 2.13 (1.73–2.53) | 0.747 | |
| Age 60–65 y (n = 4,898) | |||||
| Baseline (mean ± SD) | 42.2 ± 9.3 | 43.1 ± 9.2 | 44.3 ± 9.8 | 0.002 | |
| Year 1 (mean ± SD) | 43.6 ± 9.0 | 44.8 ± 8.9 | 45.8 ± 9.0 | ||
| Change (95% CI)2 | 1.08 (0.73–1.43) | 1.65 (1.31–1.98) | 1.92 (1.52–2.32) | 0.002 | |
| Age 66–79 y (n = 3,912) | |||||
| Baseline (mean ± SD) | 42.6 ± 9.3 | 44.4 ± 9.4 | 45.3 ± 9.8 | < 0.001 | |
| Year 1 (mean ± SD) | 44.0 ± 9.0 | 45.0 ± 8.9 | 46.0 ± 9.3 | ||
| Change (95% CI)2 | 0.85 (0.47–1.24) | 0.80 (0.43–1.17) | 1.15 (0.69–1.61) | 0.382 | |
| Stratified by smoking (Intervention arm only) | |||||
| Never smoker (n = 7,082) | |||||
| Baseline (mean ± SD) | 41.5 ± 9.2 | 42.9 ± 9.4 | 43.4 ± 9.6 | < 0.001 | |
| Year 1 (mean ± SD) | 43.2 ± 8.9 | 44.5 ± 8.9 | 45.0 ± 9.0 | ||
| Change (95% CI)2 | 1.43 (1.14–1.72) | 1.75 (1.47–2.02) | 1.82 (1.50–2.15) | 0.076 | |
| Past smoker (n = 5,686) | |||||
| Baseline (mean ± SD) | 43.1 ± 9.5 | 44.2 ± 9.4 | 45.5 ± 10.0 | < 0.001 | |
| Year 1 (mean ± SD) | 44.8 ± 9.2 | 45.4 ± 9.1 | 46.7 ± 9.3 | ||
| Change (95% CI)2 | 1.28 (0.95–1.62) | 1.25 (0.93–1.56) | 1.71 (1.34–2.09) | 0.115 | |
| Current smoker (n = 877) | |||||
| Baseline (mean ± SD) | 40.4 ± 9.1 | 41.4 ± 8.9 | 44.3 ± 9.7 | 0.001 | |
| Year 1 (mean ± SD) | 42.3 ± 9.1 | 42.9 ± 8.7 | 45.0 ± 9.5 | ||
| Change (95% CI)2 | 1.34 (0.56–2.12) | 1.43 (0.55–2.31) | 1.80 (0.72–2.88) | 0.527 | |
Categories of optimism modeled as an ordinal variable
Adjusted predictions calculated from linear regression models adjusted for baseline AHEI, age (continuous), race/ethnicity (American Indian or Alaskan Native, Asian or Pacific Islander, black or African-American, Hispanic/Latina, non-Hispanic white, other), education (≤ high school, any post-secondary, ≥ college), family income (< $20k, $20 to < 35k, $35 to < 50k, $50 to < 75k, ≥ $75k), attended 1+ religious service per week (yes, no), history of diabetes treatment (yes, no), current hypertension treatment (yes, no), high cholesterol requiring pills (ever, never), current smoking (yes, no), physical activity (< 2.5, ≥ 2.5 MET-hr/wk), hormone replacement therapy use (never, former, current), waist circumference (< 88, ≥ 88 cm), BMI (< 30, ≥ 30 kg/m2), and depression score (< 0.009, 0.009–0.06, > 0.06).
Likelihood ratio test for interaction between optimism (tertile) and trial arm on change in AHEI
Association of Optimism with Diet Quality Change by Race/Ethnicity, Age, and Smoking Status
Overall, intervention arm participants demonstrated a positive change in AHEI score from baseline to year 1 (mean ± SD change of 1.5 ± 8.4 points). This trend appeared to hold across categories of age, smoking status, and certain racial/ethnic groups. While Asian women showed an inverse relationship between optimism and baseline AHEI scores (mean AHEI scores of 47.6, 46.5, and 43.0 for optimism tertiles 1–3, respectively, P-trend = 0.005), positive change in AHEI score from baseline to year 1 was similar to other groups. Black and Hispanic women experienced larger improvements in AHEI scores than other racial/ethnic groups. Similar positive changes were observed for younger (age 50–59) versus older women (age 70–79) and never versus current smokers (Table 3). There were no statistically significant interactions between optimism and race/ethnicity, age, or smoking status on change in AHEI.
Strengths of this study include a large sample of ethnically diverse, postmenopausal women who are well characterized in relation to demographic, clinical, and lifestyle variables, validated (and repeated) measures of dietary intake and optimism assessment. The inclusion of a large intervention cohort provides an opportunity to evaluate the role of optimism in change in dietary quality among women participating in a dietary intervention. The inclusion of an established predictor of chronic disease risk in older men and women (AHEI) is another strength.24,26
This study also has several limitations. Although there is currently no gold standard diet quality assessment approach, we applied published methods,23,24 previously used and validated in the WHI population.26 We were not able to report the precise mechanisms by which dispositional optimism prospectively influences dietary change. It is known that optimistic (vs. pessimistic) women of the WHI reported greater adherence to dietary advice, so adherence is likely to have played a role, but the WHI did not collect longitudinal information on adherence. Other factors such as coping styles may also explain the observed results. For example, optimists are known to cope better than pessimists in response to stress and, thus, theoretically, may not be as vulnerable to stress eating and other unhealthy eating behaviors leading to improved diet quality over time. The WHI did not collect data that would shed light on the biological mechanisms underlying differential stress responses among individual women. We were also not able to explain the observed differences among racial/ethnic subgroups with regard to the relationship between optimism and diet quality; lack of association among Asian women may relate to cultural differences that were not assessed during the WHI.
An unresolved question is whether a threshold exists at which a particular psychological characteristic should be addressed in order for dietary or other health behavior changes to occur (e.g., tailoring of intervention for participants with high vs. low optimism). Future work should determine whether interventions that incorporate psychosocial components enhance dietary change and increase intervention adherence. Future studies might also test whether interventions can modify specific traits or whether modification of traits should be considered during recruitment (e.g., inclusion/exclusion criteria) or during an intervention (e.g., stratification based on psychological traits). For example, among school-aged children, an intervention to modify dispositional optimism and pessimism was shown to lower the incidence of depression over three years.30 In adult populations, interventions to enhance and sustain greater psychological wellbeing and optimistic attitudes have shown promise.31 This area requires further study to understand the extent to which attitudes may be modified in adults and under different circumstances.
CONCLUSIONS
These data support a relationship between optimism and dietary quality score in postmenopausal women at baseline and greater positive diet quality score changes after a one-year dietary intervention. These associations should be tested more robustly as an a priori hypothesis before conclusions can be made regarding a women’s level of optimism and capacity to improve diet quality within the context of a dietary intervention.
Footnotes
Funding/Support Disclosure
The WHI program is funded by the National Heart, Lung, and Blood Institute; National Institutes of Health; and U.S. Department of Health and Human Services through contracts HHSN268201100046C, HHSN268201100001C, HHSN268201100002C, HHSN268201100003C, HHSN268201100004C, and HHSN271201100004C.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Contributor Information
Melanie D. Hingle, Email: hinglem@u.arizona.edu, Assistant Research Professor, Department of Nutritional Sciences, College of Agriculture & Life Sciences, University of Arizona, 1177 E 4th Street, Shantz Bldg, Room 328, Tucson, AZ, 85721, 520.621.9446 (fax), 520.621.3087 (office)
Betsy C. Wertheim, Email: bwertheim@uacc.arizona.edu, Assistant Scientific Investigator, University of Arizona Cancer Center, 1501 N Campbell, Tucson, AZ, 85719
Hilary A. Tindle, Email: tindleha@upmc.edu, Assistant Professor of Medicine, Center for Research on Healthcare, 230 McKee Place, Ste 600, University of Pittsburgh, Pittsburgh, PA 15213
Lesley Tinker, Email: ltinker@whi.org, Nutrition Scientist, Fred Hutchinson Cancer Research Center, 1100 Fairview Avenue North Seattle, WA 98109
Rebecca A. Seguin, Email: rs946@cornell.edu, Assistant Professor, Division of Nutritional Sciences, College of Human Ecology, Savage Hall, Cornell University, Ithaca, NY 14850
Milagros C. Rosal, Email: milagros.rosal@umassmed.edu, Associate Professor, Preventive and Behavioral Medicine, University of Massachusetts Medical School, 55 Lake Ave North, S7-746 Worcester, MA 01655, Phone: 508-856-4685
Cynthia A. Thomson, Email: cthomson@email.arizona.edu, Professor, Mel & Enid Zuckerman College of Public Health, 295 N. Martin, Campus PO Box 245209, Drachman Hall A260, Tucson, AZ 85724
References
- 1.Pressman SD, Cohen S. Does positive affect influence health? Psychol Bull. 2005 Nov;131(6):925–971. doi: 10.1037/0033-2909.131.6.925. [DOI] [PubMed] [Google Scholar]
- 2.Boehm JK, Kubzansky LD. The heart’s content: the association between positive psychological well-being and cardiovascular health. Psychol Bull. 2012 Jul;138(4):655–691. doi: 10.1037/a0027448. [DOI] [PubMed] [Google Scholar]
- 3.Chida Y, Steptoe A. Positive psychological well-being and mortality: a quantitative review of prospective observational studies. Psychosom Med. 2008 Sep;70(7):741–756. doi: 10.1097/PSY.0b013e31818105ba. [DOI] [PubMed] [Google Scholar]
- 4.Steptoe A, Demakakos P, de Oliveira C, Wardle J. Distinctive biological correlates of positive psychological well-being in older men and women. Psychosom Med. 2012 Jun;74(5):501–508. doi: 10.1097/PSY.0b013e31824f82c8. [DOI] [PubMed] [Google Scholar]
- 5.Segerstrom SC, Sephton SE. Optimistic expectancies and cell-mediated immunity: the role of positive affect. Psychol Sci. 2010 Mar;21(3):448–455. doi: 10.1177/0956797610362061. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Steptoe A, Dockray S, Wardle J. Positive affect and psychobiological processes relevant to health. J Pers. 2009 Dec;77(6):1747–1776. doi: 10.1111/j.1467-6494.2009.00599.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kubzansky LD, Thurston RC. Emotional vitality and incident coronary heart disease: benefits of healthy psychological functioning. Arch Gen Psychiatry. 2007 Dec;64(12):1393–1401. doi: 10.1001/archpsyc.64.12.1393. [DOI] [PubMed] [Google Scholar]
- 8.Walter-Ginzburg A, Shmotkin D, Blumstein T, Shorek A. A gender-based dynamic multidimensional longitudinal analysis of resilience and mortality in the old-old in Israel: the cross-sectional and longitudinal aging study (CALAS) Soc Sci Med. 2005 Apr;60(8):1705–1715. doi: 10.1016/j.socscimed.2004.08.023. [DOI] [PubMed] [Google Scholar]
- 9.Ikeda A, Schwartz J, Peters JL, et al. Optimism in relation to inflammation and endothelial dysfunction in older men: the VA Normative Aging Study. Psychosom Med. 2011 Oct;73(8):664–671. doi: 10.1097/PSY.0b013e3182312497. [DOI] [PubMed] [Google Scholar]
- 10.Roy B, Diez-Roux AV, Seeman T, Ranjit N, Shea S, Cushman M. Association of optimism and pessimism with inflammation and hemostasis in the Multi-Ethnic Study of Atherosclerosis (MESA) Psychosom Med. 2010 Feb;72(2):134–140. doi: 10.1097/PSY.0b013e3181cb981b. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Giltay EJ, Kamphuis MH, Kalmijn S, Zitman FG, Kromhout D. Dispositional optimism and the risk of cardiovascular death: the Zutphen Elderly Study. Arch Intern Med. 2006 Feb 27;166(4):431–436. doi: 10.1001/archinte.166.4.431. [DOI] [PubMed] [Google Scholar]
- 12.Tindle HA, Chang YF, Kuller LH, et al. Optimism, cynical hostility, and incident coronary heart disease and mortality in the Women’s Health Initiative. Circulation. 2009 Aug 25;120(8):656–662. doi: 10.1161/CIRCULATIONAHA.108.827642. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Scheier MF, Carver CS, Bridges MW. Distinguishing optimism from neuroticism (and trait anxiety, self-mastery, and self-esteem): a reevaluation of the Life Orientation Test. J Pers Soc Psychol. 1994 Dec;67(6):1063–1078. doi: 10.1037//0022-3514.67.6.1063. [DOI] [PubMed] [Google Scholar]
- 14.Progovac A, Chang YF, Matthews K, et al. Psychological attitudes predict smoking cessation over time in the Women’s Health Initiative (WHI). Abstract #POS4-16. Presented at Society for Research on Nicotine and Tobacco International Meeting; Boston, MA. 2013. [Google Scholar]
- 15.Tindle H, Belnap BH, Houck PR, et al. Optimism, response to treatment of depression, and rehospitalization after coronary artery bypass graft surgery. Psychosom Med. 2012 Feb-Mar;74(2):200–207. doi: 10.1097/PSY.0b013e318244903f. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Tinker LF, Rosal MC, Young AF, et al. Predictors of dietary change and maintenance in the Women’s Health Initiative Dietary Modification Trial. J Am Diet Assoc. 2007 Jul;107(7):1155–1166. doi: 10.1016/j.jada.2007.04.010. [DOI] [PubMed] [Google Scholar]
- 17.Boehm JK, Williams DR, Rimm EB, Ryff C, Kubzansky LD. Association between optimism and serum antioxidants in the midlife in the United States study. Psychosom Med. 2013 Jan;75(1):2–10. doi: 10.1097/PSY.0b013e31827c08a9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ritenbaugh C, Patterson RE, Chlebowski RT, et al. The Women’s Health Initiative Dietary Modification trial: overview and baseline characteristics of participants. Ann Epidemiol. 2003 Oct;13(9 Suppl):S87–97. doi: 10.1016/s1047-2797(03)00044-9. [DOI] [PubMed] [Google Scholar]
- 19.Rasmussen HN, Wrosch C, Scheier MF, Carver CS. Self-regulation processes and health: the importance of optimism and goal adjustment. J Pers. 2006 Dec;74(6):1721–1747. doi: 10.1111/j.1467-6494.2006.00426.x. [DOI] [PubMed] [Google Scholar]
- 20.Hays J, Hunt JR, Hubbell FA, et al. The Women’s Health Initiative recruitment methods and results. Ann Epidemiol. 2003 Oct;13(9 Suppl):S18–77. doi: 10.1016/s1047-2797(03)00042-5. [DOI] [PubMed] [Google Scholar]
- 21.Patterson RE, Kristal AR, Tinker LF, Carter RA, Bolton MP, Agurs-Collins T. Measurement characteristics of the Women’s Health Initiative food frequency questionnaire. Ann Epidemiol. 1999 Apr;9(3):178–187. doi: 10.1016/s1047-2797(98)00055-6. [DOI] [PubMed] [Google Scholar]
- 22.Schakel SF, Sievert YA, Buzzard IM. Sources of data for developing and maintaining a nutrient database. J Am Diet Assoc. 1988 Oct;88(10):1268–1271. [PubMed] [Google Scholar]
- 23.Chiuve SE, Fung TT, Rimm EB, et al. Alternative dietary indices both strongly predict risk of chronic disease. J Nutr. 2012 Jun;142(6):1009–1018. doi: 10.3945/jn.111.157222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.McCullough ML, Feskanich D, Stampfer MJ, et al. Diet quality and major chronic disease risk in men and women: moving toward improved dietary guidance. Am J Clin Nutr. 2002 Dec;76(6):1261–1271. doi: 10.1093/ajcn/76.6.1261. [DOI] [PubMed] [Google Scholar]
- 25.McCullough ML, Willett WC. Evaluating adherence to recommended diets in adults: the Alternate Healthy Eating Index. Public Health Nutr. 2006 Feb;9(1A):152–157. doi: 10.1079/phn2005938. [DOI] [PubMed] [Google Scholar]
- 26.Belin RJ, Greenland P, Allison M, et al. Diet quality and the risk of cardiovascular disease: the Women’s Health Initiative (WHI) Am J Clin Nutr. 2011 Jul;94(1):49–57. doi: 10.3945/ajcn.110.011221. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Fung TT, McCullough M, van Dam RM, Hu FB. A prospective study of overall diet quality and risk of type 2 diabetes in women. Diabetes Care. 2007 Jul;30(7):1753–1757. doi: 10.2337/dc06-2581. [DOI] [PubMed] [Google Scholar]
- 28.Koenig H, King D, Carson VB. Handbook of Religion and Health. Oxford University Press; 2012. pp. 538–539. [Google Scholar]
- 29.Turner-McGrievy GM, Barnard ND, Cohen J, Jenkins DJ, Gloede L, Green AA. Changes in nutrient intake and dietary quality among participants with type 2 diabetes following a low-fat vegan diet or a conventional diabetes diet for 22 weeks. J Am Diet Assoc. 2008 Oct;108(10):1636–1645. doi: 10.1016/j.jada.2008.07.015. [DOI] [PubMed] [Google Scholar]
- 30.Gillham JE, Reivich KJ, Freres DR, et al. School-based prevention of depressive symptoms: A randomized controlled study of the effectiveness and specificity of the Penn Resiliency Program. J Consult Clin Psychol. 2007 Feb;75(1):9–19. doi: 10.1037/0022-006X.75.1.9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Boehm JK, Lyubomirsky S, Sheldon KM. A longitudinal experimental study comparing the effectiveness of happiness-enhancing strategies in Anglo Americans and Asian Americans. Cogn Emot. 2011 Nov;25(7):1263–1272. doi: 10.1080/02699931.2010.541227. [DOI] [PMC free article] [PubMed] [Google Scholar]