|Home | About | Journals | Submit | Contact Us | Français|
Diet may represent a modifiable prostate cancer (CaP) risk factor, but a vegetable-based prostate-healthy diet is a major change for most men. We used a ratio of animal:vegetable proteins (A:V ratio) to evaluate whether a comprehensive dietary change was self-sustaining following completion of 11 weekly dietary and cooking classes that integrated mindfulness training (MT). Thirty-six men with recurring CaP were randomized to the intervention or wait-list control. Assessments were at baseline, three months and six months. Of the 17 men randomized to the intervention, 14 completed the requirements. Nineteen were randomized to control and 17 completed requirements. Compared to controls, a significant post-intervention (3 months) decrease in A:V ratio in the intervention group (p=.01) was self-maintained 3 months post-intervention (p=0.049). At each assessment, the A:V ratio was correlated with lycopene, fiber, saturated fat, and dietary cholesterol; four dietary components linked to clinically relevant outcomes in CaP. Change in A:V ratio was also significantly correlated with changes in fiber, saturated fat and dietary cholesterol intake. Participants reported regular MT practice and there was a significant correlation between MT practice and changes in both initiation and maintenance of the change in the A:V ratio. These pilot results provide encouraging evidence for the feasibility of a dietary program that includes MT in supporting dietary change for men with recurrent CaP and invite further study to explore the possible role of MT as a means of supporting both initiation of dietary changes and maintenance of those changes over time.
Interventions addressing health-related behaviors hold considerable promise for ameliorating adverse sequelae of cancer and its treatment(1). Evidence strongly implicates diet as a factor in the development and progression of prostate cancer (CaP) and in the present debate on the mixed benefits of routine screening for CaP and immediate treatment for all men diagnosed(2–6), diet change may represent an important modifiable risk factor(7–9). As such it may be a valuable self-care cancer survivorship strategy and give some sense of personal control and hope to men(10).
Increased CaP risk appears to be associated with saturated fat from animal sources, milk and other dairy products, and meat (particularly red and processed meat), while intake of whole grain cereals, foods such as tomatoes that contain lycopene, plant foods containing phytoestrogens, vegetables, and fatty fish (with beneficial fatty acids) appear potentially protective(7, 11–20). Unfortunately, this predominantly vegetable-based diet(7) is atypical for American men most at risk for CaP (over age 65) and supporting them in making such a change presents several challenges: Only 28% of CaP survivors adhere to the American Cancer Society recommended five servings of fruit and vegetables/day(21) and are less likely than other cancer patients to initiate or maintain diet and lifestyle changes following diagnosis(22, 23); Changing to a predominantly vegetable-based diet significantly impacts habits of shopping, food preparation and attendant social and familial patterns; Males in this age range are not centrally involved with food purchasing and preparation(24), and are often reliant upon family members or restaurants for their daily food. Recognizing these challenges, most trials of comprehensive dietary change for CaP patients have been feeding trials with prepared food supplied by study personnel(18, 25–31).
We developed and tested the feasibility of a relatively short-term, group-based comprehensive dietary change program (including the men’s partners) that included shopping for and preparing prostate-healthy meals(32). To support the process of change and maintenance, mindfulness training (MT) was integrated into the program. Mindfulness is a learned capacity to attend to and discriminate between physical sensations, perceptions, affective states, thoughts and imagery in a non-reactive way(33, 34). It is associated with increased awareness of everyday activities and increases through formal practice(35). In a dietary change context, the participant learns to approach the apparent seamlessness of reactions to particular foods by recognizing the difference between the actual sensations of taste of the food, and their thoughts and feelings about it. The capacity to distinguish these components of experience and be non-reactive to them also supports the participant in maintaining their changes in the face of impulses to abandon their goal. This increased awareness of taste and other related sensations, feelings and cognitions has been shown supportive of changes in dietary practices(36). If effective, a program such as this would represent an important resource for men either during watchful waiting, or for CaP recurrence.
The aim of the present study was to obtain preliminary data on whether changes to a vegetarian dietary style were self-sustaining following completion of the intervention. The challenge of establishing an adequate summary measure of change in comprehensive dietary interventions has been addressed in a number of ways including the Healthy Eating Index(37) and Alternative Healthy Eating Index(38). It was expected that reducing meat consumption would be the most significant challenge in changing to the prostate-healthy dietary pattern, so our outcome measure was the consumption ratio of land animal protein to vegetable protein (A:V ratio). The A:V ratio was considered preferable as the outcome measure in the present study because it captures the dietary components of interest while leaving further analyses to tease out the various weighted importance of each. This ratio was previously used by Kapiszewska et al., in a large epidemiological study as an indicator of dietary-protection in colon, breast, and prostate cancer incidence(39). A higher incidence of those cancers was found in countries where the ratio was higher and suggested that it was a better method to examine cancer-related dietary patterns than focusing on individual dietary components. To closely reflect the dietary evidence related to CaP and to reflect overall change in our participants’ dietary patterns to the prostate-healthy diet, we adapted the ratio to include fish with protective vegetable proteins in the denominator. To our knowledge, no clinical trial has integrated MT into an intervention to support dietary change, or assessed a prostate-healthy diet using this ratio to reflect change in each of the components important to prostate health.
Participants were 36 men with biopsy-confirmed CaP who had undergone primary treatment (radical prostatectomy, radiation or seed implantation) and had a subsequent rising PSA (two rising PSA levels on at least two sequential tests after achieving a post-treatment nadir). Participants were recruited from within central Massachusetts, including referrals from urologist, oncologists, primary care physicians (30 participants), CaP support groups (2), direct mail campaign (2), Massachusetts Prostate Cancer Symposium (1), and radio public service announcements (1).
Exclusion criteria included undergoing radiation, chemotherapy, or hormonal therapies currently or within six months prior to the intervention; debilitating medical problems such as Alzheimer’s; dietary restrictions limiting participation, such as renal disease, inflammatory bowel disease, or following a specific diet. Participation was voluntary, and did not affect or replace the men’s medical care.
Participants were strongly encouraged to have their spouse or someone with whom they share meals to partner them through the intervention, although they were not excluded if this was not possible. The protocol was approved by University of Massachusetts Medical School (UMMS) Institutional Review Board.
Baseline information was obtained before randomization to the intervention or control arm. The controls received their usual medical care with the option of the intervention program at study completion. Participation lasted 6 months: assessments were at baseline, three months (immediate post-intervention) and six months (three months post-intervention). Baseline visits took place between one and three weeks before beginning the intervention, and post-intervention assessments were within one week of the final class.
The program consisted of eleven weekly 2½-hour classes held in the UMMS Prevention Institute teaching kitchen and the attached conference room. The prostate healthy diet was based upon increasing consumption of plant-based foods and oily fish and included a wide variety of vegetables and fruit (emphasis on cruciferous varieties and tomatoes), whole grains and soy foods, and reducing or eliminating land animal-based protein including dairy(7).
Each class integrated didactic and experiential components. Participants learned to shop for and cook meals conforming to the study diet, including foods that may have been unfamiliar. At each class, participants prepared and cooked a study-compliant meal which they then ate together family-style in the adjacent conference room. Foods containing phytoestrogens, lycopene, other carotenoids and fiber were especially encouraged, as these dietary components appear to be protective in CaP(40–43). Vegetable protein was derived from nuts, seeds, legumes and soy foods, cereal foods, and other vegetables (these foods are nutrient dense, also containing fiber, beneficial fatty acids, carotenoids and other phyto-nutrients). Classes focused on the reduction of saturated fats rather than total fat consumption, since saturated fats are associated with increased CaP risk.
Formal mindfulness training (MT)(44) was incorporated and integrated into each class and included 15 minutes of sitting meditation and mindful body-stretching exercises. Instructor-led discussion supported using the developing mindful awareness to attend to the tastes, flavors, textures and aromas of foods consumed in class and to use their increasing discrimination to distinguish these from any reactions, thoughts and habitual judgments concerning these and other food choices so that their old patterns did not determine the suitability and acceptability of these new foods for integration into their lifestyle. Participants were encouraged to bring this spirit of enquiry to their thoughts and feelings about the dietary change process itself, including any memories and feelings of loss of familiar foods and their associations. Rather than trying to change these reactions, participants were encouraged to accept them as they were, while at the same time not having them occupy their attention and determine their food choices.
Participants kept food records which were reviewed for balance and objectives by study dietitians. They also received a study manual containing nutritional information relevant to the material covered in each class, a cookbook containing recipes for a wide variety of study-compliant meals and snacks, and a CD containing mindfulness practice instructions that they were encouraged to listen to daily at home.
Control participants continued usual care with their physician, and had the option of participating in a post-study series of the intervention classes and/or receiving all the study intervention materials at the completion of study participation.
At each time point, PSA, body weight and quality of life were measured. Results of these are described elsewhere(45).
Dietary Intake was assessed using the 24-Hour Dietary Recall Nutrition Data System for Research (NDS-R version 4.03_31)(46) at each time point through three unannounced phone calls at each of the assessment time points. Calls were made on two week-days and one weekend day on randomly selected days, by a trained registered dietitian (RD) blinded to subjects’ group designations(47). The NDS-R food database includes nutrient ratios (such as animal and vegetable protein intake) and other food components including soluble and insoluble fiber, carotenoids, other antioxidants and phytoestrogens. Since dietary data was collected from three days at each follow-up time point, the A:V ratio was calculated using the average of the three days: land animal protein intake (including dairy) divided by the average of the three days vegetable and fish protein intake.
Mindfulness Practice was represented by the number of minutes intervention participants spent each day in out-of-class formal mindfulness practice during the six-month study period.
The primary outcome was the ratio of protein intake from land animal versus that from vegetable and fish (A:V ratio). Differences at baseline and differences in the changes from baseline at each follow-up time-point were compared between intervention and control using t-tests for comparing means of continuous variables. Daily caloric intake, animal protein, vegetable protein, and A:V ratio were compared between intervention and control using random effects mixed models. The interaction term of group assignment and time was tested using a likelihood ratio test to determine if there was a difference in the change over time in any outcome between intervention and control. Models were examined using all data collected, as well as models that restricted analyses to patients with data at all three time-points, inferences were not different between the two analyses, therefore results are shown for analyses using all data collected.
Pearson correlation coefficients were used to estimate associations of A:V ratio with other nutrients, PSA and other laboratory biomarkers at each time point, and changes in the ratio from baseline to 3 months and 6 months with change in other nutrients and biomarkers at corresponding time periods. Associations between A: V ratio and mindfulness practice reported in the intervention group were similarly examined.
Participants were primarily non-Hispanic White Men (91%). The average age was 69.1 years old (SD=9.0) and average weight 200.8 lbs (SD=32.1) with an average BMI of 30.1 kg/m2 (SD = 4.4). Ninety one percent were married or partnered, and 40% reported a family history of prostate cancer. Average length of time since primary treatment was 5.3 years (SD=3.0).
At baseline, there were no significant differences between the two groups on dietary or physical activity variables, age or BMI. Of the 17 men randomized to the intervention condition, 14 completed the dietary study intervention requirements (attended 7 or more classes, completed baseline, post-intervention and follow-up measurements). Nine (64%) of the 14 intervention participants who came to the classes brought a support person (spouses). Nineteen men were randomized to the control condition and 17 completed the dietary study control requirements (completed the three assessments). Of the 2 men who did not complete, 1 found the study requirements too demanding and 1 objected to his randomization assignment. Eleven (58%) of the control participants chose to attend the program classes after their final assessment and all were accompanied by a support person. Seven men decided to pursue hormone therapy during the course of their participation, 4 in the intervention and 3 in the control arm.
Results from animal protein, vegetable protein, and A:V ratio for each of the study time points are shown in Table 1. Immediately post-intervention, the A:V ratio had decreased significantly in the intervention group, indicating a move away from animal proteins, and dropped steadily from 1.87 at baseline to 0.96 post-intervention (p=.01) and 0.86 six months from baseline (p=.05). The control group’s ratio increased slightly from 2.55 at baseline to 3.00 six months post-baseline (p=0.025). In the intervention group, vegetable protein intake increased significantly (p=.007) and animal protein (which included dairy products) reduced significantly (p=.02). Importantly, these changes were self-maintained three months post-intervention, without further contact or support from study personnel.
At baseline, animal protein and dairy were the largest sources of saturated fat in the men’s diet, and the reductions in those foods were associated with significant reductions in saturated fats. We also found significant correlations of the A:V ratio with fiber (p=0.004 at baseline, p=0.001 at 3 months, and p=0.006 at 6 months). Lycopene was significant at 3 months visit, with the highest intake of lycopene in the intervention group (p=0.022). Saturated fat and cholesterol, both found in dairy and animal proteins, remained significant at both time points.
At six months, sixty-five percent of intervention men reported regular out-of-class mindfulness practice (average 12.3 minutes/day). Pearson correlations found a significant relationship between mindfulness practice and increases in percent vegetable protein intake (p=.009) and animal protein reduction (p=.04). These relationships may reflect a ‘good patient’ phenomenon (i.e.; those who adhered to the dietary changes also adhered to the other intervention demands) but also may indicate the usefulness of MT in supporting dietary change. No significant correlations were found between the A:V ratio and the minutes of reported mindfulness practices.
While a number of trials have studied the effect of minor dietary changes such as nutritional supplements to reduce CaP risk (48) the evidence strongly suggests that it is an overall dietary change that is most protective. Because of the unique and formidable challenges of supporting dietary change in the CaP population outlined in the introduction, studies of radical dietary change on disease risk have not attempted to give responsibility for the change to the men themselves. They have addressed the problem by taking the men to a spa(49), or having prepared food delivered to the men’s homes(25, 50). Those that have required the men to prepare the food at home used numerous individual sessions with dietary counselors over an extended period of time(31, 51). Neither of these strategies is feasible for large numbers of men. And while the utility of mindfulness in health behavior change is receiving considerable attention(52), to our knowledge this is the first intervention to integrate MT to support the process of dietary change and maintenance. Moreover, the average PSA doubling time (a clinical marker of disease progression) of men in the intervention slowed from 21 months at baseline to 58 months six months later. The controls remained unchanged at 18 months at both time points.
Finally, approximately 36% of patients with CaP can be expected to have concurrent cardiovascular disease and/or diabetes(53, 54) and causes of death among CaP patients not dying of the disease are similar to those among non-CaP decedents(55). In this respect, a ‘prostate-healthy’ diet is one likely to also be beneficial for risk factors for cardiovascular diseases and diabetes(56, 57).
The study was limited by the small sample size, and the fact that it was predominantly non-Hispanic White means that it may not be generalizable to other races/cultures, especially African American men who are at increased risk for CaP.
In this intervention which included MT and the support of some partners, men with a CaP recurrence made the significant changes to a vegetable-based diet; changes that were maintained three months later without additional intervention or program support. Given the controversy concerning CaP screening and treatment, the results of this self-care program are promising and warrant further exploration to delineate significant components.
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.
James F. Carmody, Division of Preventive and Behavioral Medicine, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655., 508-856-1205 (phone), 508-856-2022 (fax)
Barbara C. Olendzki, Division of Preventive and Behavioral Medicine, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655., 508-856-5195 (phone), 508-856-2022 (fax)
Philip A. Merriam, Division of Preventive and Behavioral Medicine, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655., 508-856-5848 (phone), 508-856-2022 (fax)
Qin Liu, The Wistar Institute, Philadelphia, PA 19104.
Yongxia Qiao, Division of Preventive and Behavioral Medicine, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655., 508-856-1516 (phone), 508-856-2022 (fax)
Yunsheng Ma, Division of Preventive and Behavioral Medicine, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655., 508-856-1008 (phone), 508-856-2022 (fax)