Date of Award
Spring 2024
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Public Health
First Advisor
Irwin, Melinda
Abstract
Background: Obesity, poor diet quality, and low levels of physical activity (PA) are increasingly prevalent public health challenges with important implications for breast cancer treatment adherence and survival. The field of energy balance and breast cancer survivorship has made significant advancements in recognizing the importance of integrating programs that promote positive lifestyle behavior change in oncology care. However, there remain significant gaps in translating research findings into routine clinical practice. These gaps include conflicting evidence surrounding who experiences weight change post breast cancer diagnosis and how post-diagnosis weight patterns impact survival; what aspects need to be addressed when designing and delivering interventions targeting lifestyle behavior change during and after chemotherapy; and how efficacious research interventions can be adapted and disseminated to reach broader populations of women with breast cancer. Methods: We sought to address several research gaps in the field of energy balance and breast cancer survivorship research via four projects that employed a range of study designs and analytical techniques. For the first study (Chapter 2), we assembled a retrospective cohort of women treated for stage I-III breast cancer (n=5,441) within the Smilow Cancer Hospital Network between 2013 and 2019. Individual-level repeated clinical weight measurements were extracted from the electronic health record and survival data from the Yale Tumor Registry and the Connecticut Tumor Registry. We assessed body weight trajectories over 5 years post-diagnosis and if baseline characteristics modified weight trajectories with non-linear multilevel mixed effect models. We also evaluated body mass index (BMI) at diagnosis and weight change 1-year post-diagnosis in relation to all-cause and breast cancer-specific mortality with Cox proportional hazard models. For the second study (Chapter 3), we conducted a secondary analysis of the Lifestyle, Exercise, and Nutrition Early after Diagnosis (LEANer) trial in which newly diagnosed breast cancer patients (n=173) undergoing chemotherapy were randomized to a yearlong nutrition and exercise intervention (n=87) or usual care (UC, n=86). We explored changes in PA and diet quality over the yearlong intervention period and evaluated baseline factors associated with behavior change. Mixed models compared changes in minutes/week of PA, dietary components measured via Food Frequency Questionnaire, and diet quality via the Healthy Eating Index (HEI)-2015 from baseline to one-year by study arm. Among the intervention group, we assessed baseline factors associated with one-year changes in PA and diet quality and participation in strength training at one-year with multivariable linear and logistic regression. For the third study (Chapter 4), we conducted semi-structured qualitative interviews with 29 women who completed the intervention arm of the LEANer trial to identify barriers to and facilitators of adopting healthy eating and exercise behaviors while undergoing active treatment for breast cancer. Stratified, purposeful sampling identified participants meeting both, one, or neither intervention goal at one-year: 1) 150 min/week moderate-to-vigorous intensity exercise via a self-report PA questionnaire; and 2) improving diet quality measured by the HEI-2015. Interviews were audio-recorded and transcribed verbatim. Thematic content analysis identified emerging themes. In the last study (Chapter 5), we conducted an analysis of the Lifestyle, Exercise, and Nutrition (LEAN) Self-Guided randomized controlled trial on the trial’s primary outcome of weight loss. LEAN Self-Guided randomized breast cancer survivors with a BMI ≥ 25 kg/m2 to a 6-month lifestyle intervention (n=102) consisting of only printed and online materials or waitlist group (n=103). Effects of the intervention on self-reported body weight as well as the secondary outcomes of PA, diet quality (via HEI-2010), and quality of life were assessed using mixed model repeated measures analysis. Results: In Chapter 2, we found, on average, women experienced modest body weight change over 5 years post breast cancer diagnosis, (-0.80 kg (standard error [SE]=0.04)), yet post-diagnosis weight change differed significantly across several characteristics. Weight gain was associated with ER/PR-, HER2+ tumors, BMI 18.5 kg/m2, and age 45 years (+4.90 kg (SE=0.59), +3.24 kg (SE=0.34), and +1.75 kg (SE=0.10), respectively). Weight loss was associated with BMI > 35 kg/m2 and age > 70 years (-4.50 kg (SE=0.08) and -4.34 kg (SE=0.08), respectively). Sixty-three percent of women remained weight stable (within 5%) over the first year after diagnosis while 17% gained >5% body weight and 19% lost >5% body weight during this time. Large weight loss (>10%), moderate weight loss (5-10%), and moderate weight gain (5-10%) 1-year after diagnosis were associated with higher all-cause mortality (Hazard Ratio [HR]=2.93, 95% Confidence Interval [CI]=2.28-3.75, HR=1.32, 95% CI=1.02-1.70 and HR=1.39, 95% CI=1.04-1.85, respectively). BMI > 35 kg/m2 or BMI 18.5 kg/m2 at diagnosis were also associated with higher all-cause mortality. In Chapter 3, we found, at one-year, compared to UC, participants randomized to the LEANer intervention increased PA more (mean difference: 136.1 minutes/week; 95% Confidence Interval [CI]=90.2, 182.0; p<0.001), participated in more strength training (56% vs. 15%; p<0.001), and had suggestive improvements in HEI-2015 (mean difference in HEI-2015: 2.5; 95% CI=-0.3, 5.3; p=0.08). Among the intervention group, lower baseline fatigue was associated with greater improvements in PA (p=0.04) and higher education was associated with improved HEI-2015 (p=0.001) at one-year. Higher baseline HEI-2015 (p=0.04) and being married/living with someone (p=0.05) were associated with higher odds of participating in strength training at one-year. In Chapter 4, we identified three themes from our qualitative interviews with women who had completed the LEANer intervention around aspects of the program that facilitated meeting the intervention PA and diet goals: (1) providing a conduit of trustworthy, timely, and personalized support and education; (2) promoting shifts in mindsets and enhanced understanding of the benefits of PA and nutrition during chemotherapy; and (3) fostering a sense of control and an alternative focus. Factors hindering adoption of the intervention goals included: (1) adverse effects of chemotherapy; and (2) family caregiving and other competing priorities. In Chapter 5, we found that the LEAN Self-Guided trial led to significantly greater weight loss at six-months in the intervention arm compared to the waitlist group (mean difference=-1.3 kg, 95% CI=-2.5, -0.13). We also observed improvements in PA (mean difference=18.7 min/week, 95% CI=-24.2, 61.6), diet quality (mean difference in HEI-2010=3.2 points, 95% CI=-0.20, 6.5) and fatigue (mean difference in Functional Assessment of Chronic Illness Therapy-Fatigue scale=1.4 points, 95% CI=-1.1, 3.9), though these differences were not statistically significant. Conclusions: On average, post breast cancer diagnosis weight change was modest and 63% of women remained weight stable (within 5%) over the first year after diagnosis. However, approximately 40% of women still experienced clinically meaningful weight change one-year after diagnosis and moderate weight gain and large and moderate weight loss were associated with worst survival. Characteristics of women at-risk for these weight changes include younger age, ER/PR-, HER2+ tumors, BMI18.5 kg/m2, older age, and BMI>35 kg/m2. Monitoring post-diagnosis weight change among these groups, especially during the first year after diagnosis, is critical. Future studies should examine body composition changes (muscle vs. adiposity) to better understand if weight stability is a maintenance of muscle and adipose tissue or adverse changes that result in weight stability (e.g., increase in adiposity and decrease in muscle mass). We found that most women with breast cancer could increase PA and improve diet quality when enrolled in lifestyle intervention that began at the onset of chemotherapy. Assessing sociodemographic factors, patient reported outcomes, and baseline lifestyle behaviors as well as designing programs that address chemotherapy-related side effects and competing demands, may help future interventions achieve positive behavior change during chemotherapy. The LEAN Self-Guided trial demonstrated that providing tailored printed and online materials may be an effective strategy for promoting weight loss among breast cancer survivors who have completed active treatment. This type of program could be easily implemented across cancer centers and community clinics with minimal staff efforts and costs. Both the LEANer program and the LEAN Self-Guided facilitated lifestyle behavior change during and after breast cancer diagnosis; suggesting potential resources and programs to be tested in real-world effectiveness trials. Overall, the results of this dissertation provide critical evidence to support the integration of weight management, nutrition, and exercise-related supportive care services across the breast cancer care continuum.
Recommended Citation
Puklin, Leah Silverman, "From Efficacy to Implementation: Translating Lifestyle Behavior Interventions into Routine Breast Oncology Care" (2024). Yale Graduate School of Arts and Sciences Dissertations. 1447.
https://elischolar.library.yale.edu/gsas_dissertations/1447
