Rural breast cancer survivors are able to maintain diet quality improvements during a weight loss maintenance intervention

Abstract

Purpose

Obesity and poor diet quality (DQ) are associated with increased risk of morbidity/mortality among breast cancer survivors. This study explored DQ changes during a weight loss maintenance intervention in a cohort of rural female breast cancer survivors (n = 131) who lost ≥ 5% body weight in a weight loss intervention. Previous analyses demonstrated significant DQ improvements during weight loss.

Methods

DQ was calculated using the alternate Healthy Eating Index (aHEI)-2010. Differences in scores across time for the cohort and between those that maintained weight loss within 5% (low regainers) and those that regained > 5% (high regainers) were analyzed by linear mixed models.

Results

Significant improvements in aHEI total score were observed from baseline (M = 52.3 ± 11) to 6 months (M = 60.7 ± 8; p < 0.001); these improvements were sustained from 6 to 18 months (M = 58.4 ± 11; p = 0.16). Total aHEI-2010 score at 18 months was higher in low regainers, compared with high regainers (60.7 vs. 56.0, p = 0.03), with healthier component scores for red meat (p = 0.01) and fruit (p = 0.04), and a trend for a healthier score for sugar-sweetened beverages (p = 0.08).

Conclusions

Overall DQ improvements made during a weight loss intervention for rural breast cancer survivors were sustained during a weight loss maintenance intervention; this intervention was effective in helping low regainers maintain healthier scores in fruit, red meat, and sugar-sweetened beverage components.

Implications for Cancer Survivors

Maintaining higher DQ may help breast cancer survivors maintain weight loss, thereby reducing risk of breast cancer recurrence and premature death from comorbidities.

Introduction

Breast cancer is the most invasive form of cancer among women, with nearly 269,000 new diagnoses and over 42,000 deaths estimated in 2019 [1]. There is a significant association between breast cancer and obesity (BMI ≥ 30 kg/m²), with post-menopausal obese women at a higher risk for breast cancer incidence and both pre- and post-menopausal obese women at a higher risk of breast cancer recurrence [2, 3]. It is estimated that 70% of breast cancer survivors are either overweight or obese, due to weight gain that occurs during treatment and continued weight gain post-treatment [4]. Weight gain of > 10% post-diagnosis is associated with increased risk of death among both pre- and post-menopausal breast cancer survivors [5]. The relationship between obesity and breast cancer may be further compounded in rural areas as rural obesity rates continue to outpace those in urban areas [6] and rates of rural breast cancer diagnosis continue to increase [7–9].

A breast cancer diagnosis can be a life-changing event and may provide an opportunity to motivate patients to make positive lifestyle changes involving diet quality and body weight [4]. Breast cancer survivors who achieve and maintain a healthy weight display improved health outcomes, such as decreased rates of cancer recurrence and death, compared with survivors who are overweight/obese [10–12]. Interventions targeting weight loss and subsequent weight maintenance in breast cancer survivors have generated increasing interest from both researchers and clinicians, as these interventions may help breast cancer survivors reduce risk of premature death from cancer recurrence and comorbidities associated with obesity [13, 14].

The relationship between diet quality (DQ) and breast cancer recurrence is controversial and remains unknown [15–19]; however, some research shows a positive correlation between higher DQ and reduced breast cancer incidence [20–23]. Regardless, higher-quality diets are linked with a reduced risk of death from causes other than cancer in breast cancer survivors [17, 19, 24] and a reduction in other chronic diseases such as cardiovascular disease and obesity among all populations [25, 26]. These findings are key, as breast cancer survivors are more likely to die from comorbidities related to obesity rather than breast cancer [4, 27]. Maintaining high DQ may be an important target for increasing longevity and quality of life [28] among breast cancer survivors.

Many successful weight loss and weight loss maintenance (WLM) interventions include dietary education that promotes a high-quality diet. Previous work within our group and others has shown that higher DQ was associated with greater weight loss in female rural breast cancer survivors [29–31]. However, there are no published evaluations of DQ during maintenance of weight loss in rural breast cancer survivors. The role of DQ in successful WLM is understudied yet has a suggested benefit in some studies in non-cancer survivors [32–35]. The purpose of this ancillary study was to investigate whether DQ could be maintained throughout a WLM intervention in a cohort of rural breast cancer survivors, as measured by the alternate Healthy Eating Index-2010 (aHEI-2010). We also investigated whether changes in DQ scores during this WLM intervention differed between two groups based on level of weight regain: those who regained > 5% body weight during the WLM intervention and those who maintained within 5% body weight or continued to lose weight during the WLM intervention.

Methods

Overview

This ancillary study used data collected as part of the “The Rural Women Connecting for Better Health Trial” at the University of Kansas Medical Center [36, 37], a trial that examined the effectiveness of weight loss/maintenance interventions in a cohort of rural breast cancer survivors from three US Midwestern states. The trial enrolled women with a history of stage 0-III breast cancer into a 6-month behavioral weight loss intervention targeting decreased calories, increased physical activity, and behavior change strategies with intervention sessions delivered via weekly group conference calls. At 6 months, participants were randomized to extended care to address WLM either through continued bi-weekly group conference calls or bi-weekly mailed newsletters.

Study population

Eligible participants for the trial were female breast cancer survivors ≤ 75 years old who had previously been diagnosed with stage 0-IIIc breast cancer within the last 10 years, completed treatment at least 3 months prior to study enrollment, were cleared for physical activity by their primary care provider, lived in rural areas, and had a BMI between 27 and 45 kg/m². The 6-month non-randomized weight loss intervention included 210 participants. If participants lost ≥ 5% of baseline bodyweight, they were then randomized to either the bi-weekly group conference call or mailed newsletter weight maintenance arms (n = 172). This analysis included individuals who were randomized to WLM arms and returned for 18-month data collection (n = 154). Of these, 131 met inclusion criteria of having two valid 24-h food recalls at each timepoint, including baseline, 6 months (end of weight loss intervention), and 18 months (end of WLM intervention). Supplemental Figure 1 (Online Resource 1) summarizes participant flow and inclusion in this analysis. All procedures for this study were approved by the Human Subjects Committee of the University of Kansas Medical Center.

Weight loss and weight loss maintenance interventions

Details of the weight loss and WLM interventions have been published previously [36]. The study was divided into two distinct intervention phases. The entire cohort participated in the first intervention phase, a 6-month structured weight loss intervention. This intervention phase was based in social cognitive theory and aimed for 10% body weight loss during this period. It involved weekly group-based phone counseling sessions, as well as dietary and exercise recommendations. Counseling sessions focused on a variety of topics associated with weight loss, including MyPlate-based diet recommendations and increased exercise.

Participants who lost ≥ 5% of baseline body weight at the end of the weight loss intervention (the 6-month timepoint) were randomized to one of two WLM intervention arms, which constituted the second phase of the trial. As the primary aim of the trial was to evaluate the WLM efficacy of two different information delivery methods, one arm received biweekly group phone counseling sessions, while the other arm received a bi-weekly newsletter via mail. Both intervention arms focused on topics related to WLM, including nutrition and exercise. The WLM intervention lasted 12 months (or to the 18-month timepoint) for both arms.

Measures

Demographics

Participant age, education level, and rurality (based on Rural-Urban Commuting Area codes [38]) were collected at baseline.

Weight, height, and BMI

Study nurses measured participant height using a stadiometer and weight using a calibrated digital scale (± 0.1 kg, Befour, Inc) at all timepoints. Weight/height measurements were taken in duplicate at each timepoint to ensure measurement accuracy. BMI was calculated using height/weight measurements.

Weight maintenance groups

Individuals who regained no more than 5% of their post-weight loss intervention body weight or continued to lose weight during the WLM intervention (6 months to 18 months) were defined as “low regainers.” Individuals who regained > 5% of body weight during the WLM phase were defined as “high regainers.” Greater than 5% body weight regain has been shown to be clinically significant and has been used previously in other studies examining WLM [39]. This also represents an approximate median split for the current sample.

Dietary assessment

Dietary intake measurements included two 24-h recalls conducted by trained staff at each timepoint including baseline, 6 months (end of weight loss intervention), and 18 months (end of WLM intervention). Recalls were conducted using the USDA multiple-pass approach [40] and were collected on one random weekday and one weekend day. One recall was conducted in-person at all timepoints, with trained study personnel using food models, containers, and charts to help participants estimate portion sizes. All other recalls were collected via telephone, with participants using food amount booklets provided by study personnel based on National Health and Nutrition Examination Survey (NHANES) methodology, which uses two-dimensional images to assist participants in estimating food and drink portions consumed [41]. The recalls were entered into the Nutrition Data System for Research (NDS-R, 2010) and analyzed for food group and nutrient intake.

Alternate healthy eating index

The Alternate Healthy Eating Index (aHEI) was developed by McCullough et al. [42] as a modification to the Healthy Eating Index, a measure of diet quality based on the Dietary Guidelines for Americans [43], and was updated in 2010 [25]. Higher aHEI scores have been more closely associated with reduced invasive breast cancer risk in postmenopausal women [44], reduced chronic disease risk [25, 26, 45], and reduced risk of all-cause mortality [19] than similar diet quality indices. The 2010 version of the aHEI includes 11 components, all of which focus on the quality of foods within general food components. All components are scored from 0 to 10, with 10 being the maximum score. The total score is calculated by summing all component scores. aHEI guidelines include six components focusing on adequacy (fruits, vegetables, whole grains, nuts/legumes, omega-3 fatty acids, polyunsaturated fatty acids) and five components focusing on moderation (sugar-sweetened beverages/fruit juice, red/processed meat, trans fats, sodium, and alcohol). Higher consumption of adequacy-focused components yields a higher score; inversely, lower consumption of moderation-focused components yields a higher score.

Statistical analyses

The primary aims of this study were to examine change in DQ after a 1-year weight maintenance intervention and assess whether change in DQ differed in participants that varied in amount of weight regain (≤ 5% regain or > 5% regain). We also report initial change in aHEI from baseline to 6 months to allow for comparison across time and to supplement our prior report examining original HEI scores during weight loss only [31] when aHEI scoring algorithms were more difficult to generate. Linear mixed-effects models (LMM) were constructed to assess whole sample and weight regain group differences in aHEI-2010 scores at each timepoint. These models included participant as a random effect, allowing for each participant to have their own intercept parameter, and all other covariates as fixed effects. LMMs were adjusted for randomization group (phone or mail) by time (baseline, 6 months, and 18 months) interaction and for a priori covariates of age (continuous), baseline BMI (continuous), education (categorical), and rurality (binary categorical). We considered physical activity as a covariate in our models; however, >10% of these data were missing and were not used. The LMM for weight regain status indicated further investigation into diet quality and weight regain at 18 months; thus, we constructed an ordinary least squares regression (OLS) model to assess difference in aHEI-2010 scores between weight regain groups at 18 months adjusted for randomization group, age, baseline BMI, education, and rurality. We also used OLS to assess differences in energy intake at 18 months per kg body mass at 6 months between weight regain groups adjusted for the same covariates. All post hoc pairwise comparisons were adjusted using Tukey’s HSD. Model assumptions were assessed using residual analyses including QQ plots, residual histograms, and scale-location plots. Statistical analyses were performed using R (v. 3.6.1; R Foundation, Vienna, Austria). Statistical significance was set at p < 0.05.

Results

Descriptive statistics of the cohort (n = 131) are displayed in Table 1. Mean ± SD age at baseline was 58.7 (± 7.8) years and mean BMI was 34.1 (± 4.4) kg/m². Mean baseline weight was 91.2 (± 13.7) kg, and mean 6-month weight was 78.0 (± 13.5) kg, an average decrease of 13.2 (± 4.8) kg. We used OLS regression to determine that specific baseline participant characteristics (age, education (continuous variable), rurality, BL BMI, years since treatment, and history of chemotherapy treatment) did not predict weight regain during the weight maintenance phase. Table 2 describes differences in aHEI-2010 total and component scores between the end of the weight loss intervention and the end of the weight loss maintenance intervention. aHEI-2010 total scores improved from baseline (52.3 ± 11.1) to 6 months (60.7 ± 8.5; p < 0.001) and were sustained at a similar level at 18 months (58.4 ± 10.9; p = 0.16). All components, except for long-chain omega-3 fatty acids, changed during the weight loss intervention, with fruit (p < 0.001), vegetable (p < 0.001), red meat (p < 0.001), sugar-sweetened beverage/fruit juice (p < 0.001), trans-fat (p < 0.001), and sodium (p < 0.001) components improving in score. At the end of the WLM intervention, six component scores did not change, while fruit (p < 0.001), vegetable (p = 0.01), red meat (p = 0.04), and trans-fat (p = 0.004) worsened in score, and polyunsaturated fats (p < 0.001), improved in score.

Table 1.

Descriptive characteristics of weight maintenance cohort (n = 131)

	All
Group
Mail	62 (47.3%)
Phone	69 (52.7%)
Age at baseline	58.7 ± 7.8
Education:
Bachelors	30 (22.9%)
Doctorate	1 (0.8%)
High school/GED	29 (22.1%)
Masters	23 (17.6%)
Some college or associate	48 (36.6%)
Race
Non-Hispanic White	129 (98.5%)
Hispanic White	1 (0.7%)
African American	1(0.7%)
Rurality
Largea	58 (44.3%)
Small/isolatedb	73 (55.7%)
Baseline BMI	34.1 ± 4.4
Baseline weight (kg)	91.2 ± 13.7
6-month weight (kg)	78.0 ± 13.5
Weight change during weight loss (kg)	13.2 ± 4.8
Weight change during maintenance (kg)	3.9 ± 5.1
Weight maintenance % change	4.8 ± 6.3

^aRural town with a population between 10,000–49,999 residents

^bRural town 2500–10,000 or < 2500 residents

Table 2.

Changes in aHEI-2010 scores at baseline, 6 months, and 18 months for entire cohort (n = 131)

			Weight loss intervention		Weight loss maintenance intervention
	Baseline	6 months	p (BL-6M)	18 months	p (6M-18M)
aHEI-2010 total score	52.3 ± 11.1	60.7 ± 8.5	<0.001	58.4 ± 10.9	0.16
Foods to increase
Vegetables	4.8 ± 2.7	6.9 ± 2.8	<0.001	6.0 ± 3.0	0.01
Fruit	2.7 ± 2.4	7.4 ± 2.7	<0.001	5.4 ± 3.3	<0.001
Nuts/legumes	4.2 ± 4.2	3.3 ± 3.0	0.02	4.2 ± 3.9	0.16
Long chain omega-3 fatty acids	2.1 ± 2.4	1.6 ± 2.3	0.88	1.8 ± 2.4	0.93
Polyunsaturated fatty acids	7.7 ± 2.3	4.3 ± 2.7	<0.001	5.8 ± 2.7	<0.001
Whole grains	2.4 ± 1.8	1.5 ± 1.5	0.01	2.0 ± 1.7	0.11
Foods to decreasea
Red meat	4.1 ± 3.1	6.7 ± 3.0	<0.001	5.4 ± 3.3	0.04
Sugar-sweetened beverages/fruit juice	6.8 ± 4.2	8.9 ± 2.6	<0.001	8.4 ± 3.1	0.57
Trans-fat	7.4 ± 1.8	9.0 ± 1.4	<0.001	8.4 ± 2.1	0.004
Alcohol	4.1 ± 3.0	3.3 ± 1.9	0.03	3.5 ± 2.3	0.60
Sodium	6.0 ± 2.0	7.7 ± 1.7	<0.001	7.5 ± 1.9	0.52

Differences among timepoints were assessed using linear mixed effects models adjusted for group × time interaction, age, baseline BMI, education, and rurality including participant ID as a random effect. Age was the only significant covariate (p = 0.04), with increased age correlating with higher aHEI-2010 scores

^aHigher aHEI scores are an indicator of a healthier diet and do not necessarily indicate increased consumption of these components

Figure 1 illustrates aHEI-2010 mean scores between high and low regainers. There was no difference in aHEI-2010 scores between high and low regainers at the end of the weight loss intervention (6 months); however, at the end of the WLM intervention, low regainers had higher aHEI-2010 scores than high regainers (p between groups = 0.01). Low regainers maintained similar aHEI-2010 scores during the WLM intervention (p = 0.85); however, aHEI-2010 scores worsened in high regainers during the WLM intervention (p = 0.003). Neither total daily energy consumption nor daily energy consumption per kg differed between high regainers and low regainers at the end of the WLM intervention (1380 ± 380 kcal vs. 1420 ± 350 kcal, respectively, p = 0.47) and kcal/kg (17.6 ± 5.2 kcal vs. 19.1 ± 5.2 kcal, respectively, p = 0.20). Likewise, there was no difference in energy density of food/beverages consumed (measured as kcal/100 g) between low and high regainers at baseline (58.1 ± 22.0 vs. 60.0 ± 18.1, p = 0.60), at the end of the weight loss intervention (41.1 ± 14.2 vs. 43.2 ± 13.0, p = 0.36), or at the end of the WLM intervention (45.7 ± 16.6 vs 47.1 ± 17.9, p = 0.64).

Fig. 1.

Differences in aHEI-2010 score at baseline, end of the weight loss intervention, and end of the weight loss maintenance intervention between high regainers (> 5% regain) and low regainers (≤ 5% regain).

**p = 0.01. Points and error bars represent means and 95% CI

aHEI-2010 total and component scores based on weight regain status are shown in Table 3. Low regainers had no change in component scores during the WLM intervention in all but two components: a decrease in fruit (p = 0.002) and an increase in polyunsaturated fatty acids (p < 0.001). However, high regainers saw worsening in aHEI-2010 scores across multiple components, including red meat (p < 0.001), sugar-sweetened beverages/fruit juice (p = 0.03), vegetables (p = 0.01), fruit (p < 0.001), and trans-fat (p < 0.001). High regainers also saw increases in aHEI-2010 score in whole grains (p = 0.01), nuts/legumes (p = 0.05), and polyunsaturated fatty acids (p = 0.01). Low regainers displayed better scores at the end of the weight loss maintenance intervention compared with high regainers for aHEI-2010 total score (60.7 vs. 56.0, p = 0.01), as well as fruit (6.2 vs 4.6, p = 0.04 ) and red meat (6.1 vs. 4.8, p = 0.01 ) components; there was also a trend for a between-group difference for sugar-sweetened beverage/fruit juice (8.9 vs. 7.9, p = 0.08).

Table 3.

Changes in aHEI-2010 score at the end of the weight loss maintenance intervention within weight regain groups

	> 5% weight regain (high regainers, n = 64)			≤ 5% weight regain (low regainers, n = 67)			Between groups
	6 months	18 months	p value*	6 months	18 months	p value*	p value**
aHEI-2010	60.9 ± 8.0	56.0 ± 11.0	0.003	60.4 ± 8.9	60.7 ± 10.4	0.85	0.03
Foods to increase
Vegetables	7.3 ± 2.6	6.0 ± 3.1	0.01	6.6 ± 2.9	6.1 ± 2.8	0.26	0.91
Fruit	7.0 ± 2.7	4.6 ± 3.5	<0.001	7.8 ± 2.7	6.2 ± 3.0	0.002	0.04
Nuts/legumes	2.9 ± 2.7	4.0 ± 4.0	0.05	3.7 ± 3.1	4.4 ± 3.8	0.28	0.80
Long chain omega-3 fatty acids	1.8 ± 2.6	1.6 ± 2.1	0.52	1.4 ± 1.9	1.9 ± 2.8	0.23	0.55
Polyunsaturated fatty acids	4.8 ± 3.1	6.2 ± 2.6	0.01	3.8 ± 2.1	5.5 ± 2.7	<0.001	0.17
Whole grains	1.3 ± 1.5	2.1 ± 1.7	0.01	1.7 ± 1.6	1.8 ± 1.7	0.55	0.26
Foods to decrease
Red meat	6.7 ± 3.0	4.8 ± 3.2	<0.001	6.6 ± 2.9	6.1 ± 3.2	0.33	0.01
Sugar-sweetened beverages/fruit juice	9.1 ± 2.4	7.9 ± 3.6	0.03	8.7 ± 2.8	8.9 ± 2.4	0.69	0.08
Trans-fat	9.0 ± 1.1	8.0 ± 2.3	<0.001	8.9 ± 1.6	8.8 ± 1.9	0.52	0.11
Alcohol	3.2 ± 1.9	3.4 ± 2.3	0.61	3.4 ± 1.9	3.5 ± 2.3	0.76	0.68
Sodium	7.8 ± 1.9	7.4 ± 1.8	0.20	7.7 ± 1.5	7.6 ± 1.9	0.69	0.55

Differences between timepoints and groups were assessed by ordinary least squares regression models using Tukey’s HSD adjusted for randomization group, age, baseline BMI, education, and rurality

^*p values for change scores within groups

^**p values between groups at the end of the weight loss maintenance intervention

Discussion

This study is the first to show that DQ (as measured by aHEI-2010) can be maintained in BC survivors after a weight loss intervention and that the degree of maintaining DQ varies by the level of weight regain. This suggests the important role DQ plays in both losing weight and maintaining weight loss in BC survivors. Our findings are consistent with previous research examining DQ during a WLM intervention. Ptomey [34] demonstrated that DQ as measured by Healthy Eating Index (HEI)-2010 [46] improved from 46.6 at baseline to 66.6 at 6 months and then decreased during the weight loss maintenance intervention to 57.7; however, it remained significantly above baseline measures. Likewise, Svetkey [47] demonstrated that HEI-2005 [48] scores improved during a weight loss intervention and maintained above baseline after 30 months of a WLM intervention. Both the HEI-2005 and HEI-2010 are based on the Dietary Guidelines for Americans for their respective years [49, 50]. However, the aHEI-2010 is a modification of the HEI-2010 and is based on food components linked to chronic disease prevention/development [42]. This is a key differentiation in our study as it relates to chronic disease risk in a cancer survivor sample.

Mean aHEI-2010 score at the end of the WLM intervention for our sample was 58.4, which correlates with reductions in T2DM risk [25, 51], all-cause mortality risk [25, 52], and CVD incidence [25, 52] risk from a mean aHEI-2010 baseline score of 52.3. Our findings are unique that overall DQ not only increased significantly during the WL intervention, but those increases persisted across the 12 months of the WLM intervention. This is an important finding, as these sustained improvements in DQ may also lead to reduced risk for developing diet-related chronic diseases. It is well-established that BC survivors are at a higher risk of dying from diet-related chronic disease than from BC [4, 27]; if BC survivors can improve DQ and maintain these improvements long term, the chances of developing these diseases may decrease, thus potentially extending life and improving quality of life [28].

In addition, we found that overall aHEI-2010 scores were significantly worse for high regainers at the end of the weight loss maintenance intervention, when compared with low regainers. These data provide compelling evidence towards the important role DQ may play in weight loss maintenance. Weight loss maintenance is a multi-faceted set of behaviors, with components such as dietary intake, physical activity, and behavior change all working together to help a person manage weight, and our findings suggest that DQ plays a role in this process. Therefore, DQ should continue to be reinforced after weight loss in breast cancer survivors, to help individuals maintain hard-fought improvements in weight. Further investigation with longer follow-up and additional timepoints of DQ measurement is needed to uncover the relative contributions of DQ in the ability to maintain weight loss.

Current dietary pattern recommendations from the American Institute for Cancer Research (AICR) stress increased plant consumption, reduced red/processed meat consumption, and decreased empty calorie consumption (characterized as alcohol, added sugars, and solid fats) [53]. Low regainers in this study adopted these patterns into their long-term diets; only the fruit component worsened for low regainers during maintenance as compared with high regainers who showed worsened scores in red meat, fruit, vegetable, trans-fat, and sugar-sweetened beverage/fruit juice. These findings add to the potential role of specific dietary components in weight loss maintenance interventions and may provide a roadmap as to where to focus dietary education and behavioral strategies during weight loss maintenance. Red/processed meat and sugar-sweetened beverage consumption, in particular, have been associated with multiple chronic diseases that affect BC survivors [54–57]; reducing consumption of these foods should be an integral part of any nutrition counseling focused towards breast cancer survivors.

Energy consumption did not differ significantly between high and low regainers. This could be attributed to several possibilities. All aHEI-2010 scores are based on 24-h recalls, which may be more prone to reporting error in amounts than in types of food, i.e., it is possible that participants correctly reported the types of foods consumed but underreported the amounts of foods consumed. It is also possible that underreporting of energy intake was greater among high regainers at the 18-month timepoint, as under-reporting has been associated with obesity status [58]. Additional analyses showed both weight regain groups displayed similar energy density of foods/beverages consumed across all timepoints. As aHEI-2010 scores are calculated per 1000 kcal consumed, it can be assumed that higher aHEI-2010 scores displayed among low regainers are indicative of a more nutrient-dense diet, regardless of total energy consumption. Our findings suggest that DQ measures may better differentiate level of weight regain compared with energy intake.

Strengths of this study include the use of 24-h recall as the ‘gold standard’ for collecting reported dietary intake and for estimating diet quality [59], use of aHEI-2010 which has displayed stronger correlations to chronic disease development than similar DQ indices, and the sufficient sample size from a 18-month randomized controlled trial to examine the impact of degrees of weight regain on change in DQ. Limitations of the study include the lack of racial/ethnic diversity (but with geographic diversity), and missing diet recalls at one or more timepoints among 15% of the sample retained in the trial. In addition, aHEI-2010 components are based on servings per 1000 kcal and do not account for overconsumption; therefore, a person could have an excellent aHEI-2010 score but overconsume total calories, potentially leading to weight gain.

In conclusion, this study showed sustained improvements in DQ during a weight loss maintenance intervention and showed that better maintenance of diet quality was associated with better maintenance of weight loss. These maintained improvements are clinically meaningful for rural BC survivors as they have been linked with decreased all-cause mortality risk, CVD mortality risk, and cancer mortality risk [60]. Future research in this field should focus on further improving component scores during weight loss maintenance related specifically to cancer and chronic disease risk, such as maintaining decreases in red/processed meat consumption.

Supplementary Material

Electronic supplementary material The online version of this article (https://doi.org/10.1007/s11764-020-00951-2) contains supplementary material, which is available to authorized users.