Intensive Nutrition Counseling as Part of a Multi-Component Weight Loss Intervention Improves Diet Quality and Anthropometrics in Older Adults with Obesity

Abstract

Background and Aims:

Obesity significantly impacts older adults. Intensive nutrition counseling can aid in weight reduction and improve diet quality, but data are sparse in this population. The objective of this intervention is to determine how intensive nutrition counseling affects diet quality and anthropometric measures during a multi-component weight loss intervention in rural older adults with obesity.

Methods:

A series of 12-week, single-arm feasibility pilots were conducted in fall 2017 and winter/spring 2018 in a community aging center in rural Northern New England. Adults were eligible if ≥65 years old with a Body Mass Index (BMI) ≥30kg/m². Exclusion criteria included dementia/cognitive impairment, uncontrolled psychiatric illness, weight-loss surgery, weight loss >5% in previous 6-months, life-threatening illness, palliative/hospice services, current participation in another weight-loss study/program, obesogenic medications, or presence of major chronic conditions. Participants received once-weekly nutrition counseling by a registered dietitian nutritionist (RDN), and twice-weekly exercise sessions by a physical therapist (PT). Primary outcomes were diet quality changes measured by total Rapid Eating and Activity Assessment for Patients-Short Version (REAP-S) and Automated Self-Administered 24-hour dietary recall (ASA-24). Secondary outcome measures were changes in weight (kilograms) and waist circumference (centimeters). McNemar test was conducted for all paired categorical data while paired t-tests were conducted for all paired continuous data. All analyses were conducted in R; p-value<0.05 was significant.

Results.

Total n=23. Mean age was 72.2 (5.8) years (73.9% female); mean BMI was 35.9 (±5.0) kg/m². At 12 weeks, diet quality significantly improved. REAP-S scores increased by 3.53±3.13 points (p<0.001). Kilocalories, grams fat, grams saturated fat, milligrams sodium, grams added sugar, and grams alcohol via ASA-24 significantly decreased (all p<0.05). Significant reductions in weight (−5.22±3.13kg) and waist circumference (−6.88±5.67cm) were observed (both p<0.001).

Conclusion.

Intensive nutrition counseling significantly enhances diet quality and reduces weight and waist circumference in rural older adults with obesity.

Introduction

Obesity impacts nearly 40 % of adults over the age of 60 in the United States(1, 2), and poses unique challenges to their overall health, physical function, and quality of life(3, 4). Multi-component interventions focusing on diet and scheduled physical activity (both resistance and aerobic training) lead to weight loss which improves these outcomes(5, 6). Intensive nutrition counseling, defined as several and recurrent sessions of nutrition therapy provided by a registered dietitian nutritionist (RDN), enhances weight loss efforts(7–9), improves primary care outcomes, and can attenuate metabolic parameters(10). While there is evidence to support that intensive lifestyle counseling can lead to weight loss in older adults, and adherence may be higher in this population(11, 12), data remain sparse on the effects of intensive nutrition counseling in older adults with obesity. This is in part because of the lack of interventions(1), and that clinical trials often have upper age limitations, excluding adults with co-morbid conditions (13).

Improved diet quality is a core component of weight loss interventions, and has been independently linked with a reduction of mortality and reduced risk of certain chronic diseases (14–18). While reduced caloric intake is clearly essential(1, 5, 19), enhancing protein above current recommendations is likely important(20), suggesting diet quality and composition should be considered. Some groups have addressed specific nutrient intake and quality during weight loss interventions using partial meal replacements (21–23) or laboratory parameter measurement after diet intervention in older adults (24). However, there is limited evidence of systematic diet quality evaluation during weight loss interventions in older adults, as measured by changes in food records, 24-hour recalls, or diet surveys.

Validated methods for nutritional assessment is important during weight loss management and can permit the development of an individualized, patient-centered approach to care(25). Previous research demonstrated misreporting, and particularly under-reporting, of energy intake on food frequency questionnaires among individuals with obesity(26). While 24-h dietary recalls are both timely and accurate, many require in-person administration by trained interviewers(27), and shorter assessments are often sought in clinical and research settings. However, such assessments are often not evaluated in older adults. The purpose of this paper is to determine how intensive nutrition counseling affects changes in diet quality (as measured by changes in Rapid Eating and Activity Assessment for Patients-Short version(28) and the Automated Self-Administered 24-hour dietary recall(27)) and anthropometric measures during a multi-component, 12-week intervention for weight loss in rural older adults with obesity.

Materials and Methods

Study Setting & Design

A series of single-arm, feasibility pilot interventions focusing on diet and physical activity, each 12 weeks long, were conducted in a small, rural community in Northern New England in the fall of 2017 and winter/spring of 2018. After a review of electronic medical records (EMR), community-dwelling adults were eligible if they were at least 65 years of age or older and had a Body Mass Index (BMI) ≥30kg/m². Participants were excluded if the initial EMR review showed the diagnosis of dementia or cognitive impairment; uncontrolled psychiatric illness; weight-loss surgery; weight loss >5% in the previous 6-months; life-threatening illness or use of palliative/hospice services; current participation in another weight-loss study or program; obesogenic medications(1); or the presence of major chronic conditions such as advanced congestive heart failure, or renal or liver insufficiency. From the EMR, we abstracted demographic information, BMI, co-morbidities, and smoking status. Phone screening of all subjects then followed, with minimum score of ≥ 3 on the Callahan Scale(29) in addition to scores of ≥ 71.2 on the Functional Status Questionnaire and > 56.4 for instrumental activities of daily living(30–32) needed for enrollment. The study was approved by both the Committee for the Protection of Human Subjects at Dartmouth College and the Dartmouth-Hitchcock Institutional Review Board (#28905). After obtaining informed consent by the research assistant, baseline subjective and objective assessments were completed, with gas cards provided for their completion. Participation was free of charge.

Intervention Description

The intervention was delivered at a community-based aging center. Each participant received an RDN-delivered nutrition counseling intervention, and an exercise component provided by a geriatric physical therapist (PT). The nutrition component consisted of 12 intensive, 30-minute weekly, one-on-one counseling sessions guided by motivational interviewing techniques focusing on specific, measurable, attainable, relevant and timely (SMART) goals(33). Study staff underwent motivational interviewing training via an interactive in-person course, and motivational interviewing techniques were overseen by the PI and team. Individualized meal plans were created by the RDN for each participant using evidence-based guidelines for health, with a focus on plants, whole grains, fiber, low fat dairy, lean protein, and decreased saturated fat, salt, and added sugar(11, 12, 34–37).

Participants were prescribed a meal plan with a 500 – 750 kcal deficit from baseline energy expenditure measures to help promote weight loss, and a minimum of 1200 kcal per day to ensure nutritional sufficiency(8, 19, 38). Meal plans contained about 20 % of kcal from protein, or about 1.2 grams protein/kg body weight/day(20). Each participant was encouraged to consume 1000 international units (IU) of vitamin D from food or supplements daily; need for additional supplementation was determined on an individual basis(39, 40). Education materials focused on evidence-based guidelines(11, 12, 34–37) were placed in a binder for each participant. These education materials, based on the results of the LOOK-Ahead(11) and the Diabetes Prevention Program(12) were formulated from either institutional materials (Dartmouth-Hitchcock Medical Center Weight and Wellness Center) or from government and other organizations focused on health education (such as the National Institutes of Aging(41) and The American Heart Association(42)). Each nutrition session focused on a different topic. The topics addressed were as follows: Meal planning (food and fluids), protein, carbohydrates, fat, serving sizes and portion control, mindful eating, sodium and micronutrients, choosing healthy snacks, vegetables, fruits, eating outside the home and special occasion eating, and finally, weight loss maintenance.

The exercise component consisted of twice-weekly, on-site group exercise sessions led by the PT. A baseline assessment by the PT permitted development of an exercise plan aimed at gradually increasing activity at moderate intensity, using resistance bands and adjustable cuff weights. A combination of resistance, flexibility, balance, and aerobic exercises were included. All participants were encouraged to perform resistance exercises one day per week on their own, at least 24 hours apart, and/or focusing on different muscle groups. Exercise education materials were provided and were based on the LIFE study(43). Daily aerobic activity was also encouraged and was tracked using weekly diaries. Intervention details have been discussed previously(44).

Dietary Measures

The Automated Self-Administered 24-hour dietary recall (ASA-24) is based on the validated Multiple Pass Method(45) and has been shown to estimate mean total energy and protein intakes to proximate recovery biomarkers(27). There is close agreement between the ASA24 method and standardized interviewer-administered 24-hour recalls according to validation and evaluation studies(46). The validated Automated Self-Administered 24-hour Dietary Assessment Tool 2016 (ASA-24 2016)(27) was conducted at two separate dates during baseline assessment, in order to determine dietary intake at study entry, and this helped guide meal plan formation. Indirect calorimetry using REEVUE (Korr Technologies, Salt Lake City, UT) also helped guide participant caloric prescription. The ASA-24 was conducted twice at the end of the intervention to potentially capture any dietary changes.

The Rapid Eating and Activity Assessment for Patients (REAP) was developed for primary care providers to quickly and easily assess the diet quality and physical activity of their patients(47). The REAP-S is a shortened version of this tool, consisting of 16 questions focused on food intake. It is designed as a triage tool to quickly assess overall diet quality based on US Dietary Guidelines(48), with an additional question asking individuals to rate their willingness to change their current eating behaviors(28). The REAP-S was validated(49) and found to be a useful triage screening tool for weight loss(28), with higher scores indicating improved diet quality(47). The (REAP-S)(28) was conducted once at baseline and again at the completion of the intervention. By also providing the participants’ rating of their willingness to change, the REAP-S offered additional guidance to focus the motivational interviewing technique of the dietitian.

Anthropometric Measures

Weight was measured weekly in kilograms using an A+D scale, and height in meters was measured at baseline using a Seca 216 stadiometer. Waist circumference in centimeters was measured monthly at the iliac crest using a standard tape measure. All anthropometric measures were conducted by trained research staff (overseen by the principal investigator and exercise physiologist). Measurements (weight, height) were not taken in duplicate.

Statistical Analysis

All data was combined in a single dataset for analysis. Continuous variables are represented as means (standard deviation) and categorical as counts (percent). Our primary outcomes were changes in diet quality as measured by the total REAP-S score and ASA-24. Secondary outcome measures included changes in anthropometric measures of weight and waist circumference. Change in continuous variables are represented as +/−.

The McNemar test was conducted for all paired categorical data while paired t-tests were conducted for all paired continuous data. All analyses were conducted in R(50). A p-value <0.05 was considered statistically significant. This pilot was an adjunctive analysis of the primary study intervention, and as a secondary analysis, a sample size calculation was not conducted.

Results

25 participants were enrolled, and 23 subjects (92%) completed this study. 2 subjects withdrew due to chest pain, one of whom had undiagnosed coronary artery disease, and one due to gastroesophageal reflux disease. Table 1 details participants’ baseline characteristics. As there were no observed differences between the cohorts, all data was combined (not shown). Mean age was 72.2±5.8 years, the majority were female (73.9%), with a mean BMI of 35.9±5.0 kg/m². Seventeen participants had a BMI < 40 kg/m² and six had a BMI ≥ 40 kg/m²; the differences in baseline characteristics and outcome measures were not significant (supplemental tables 1 and 2). At onset, mean Activities of Daily Living score was 96.1±7.9 and Instrumental Activities of Daily Living score was 83.3±13.5. There was a high attendance rate for the nutrition (93.8%) and exercise sessions (88.3%) among the study participants.

Table 1:

Baseline Characteristics of Included Participants

	2A (N=9)	2B (N=14)	Total (N=23)	p value
Age	71.0 (5.1)	73.0 (6.3)	72.2 (5.8)	0.44
Female Sex	6 (66.7%)	11 (78.6%)	17 (73.9%)	0.53
White Race	9 (100.0%)	14 (100.0%)	23 (100.0%)	1.00
Marital status				0.35
Single	1 (11.1%)	0 (0.0%)	1 (4.3%)
Married	6 (66.7%)	10 (71.4%)	16 (69.6%)
Divorced	2 (22.2%)	3 (21.4%)	5 (21.7%)
Widowed	0 (0.0%)	1 (7.1%)	1 (4.3%)
Insurance
Medicare	9 (100.0%)	14 (100.0%)	23 (100.0%)
Medicaid	0 (0.0%)	0 (0.0%)	0 (0.0%)
Private insurance	7 (77.8%)	10 (71.4%)	17 (73.9%)	0.74
Smoking status				<0.001
Non-smoker	2 (22.2%)	13 (92.9%)	15 (65.2%)
Former smoker	7 (77.8%)	1 (7.1%)	8 (34.8%)
Education				0.43
N-miss	1	0	1
High school	2 (25.0%)	0 (0.0%)	2 (9.1%)
Some college	0 (0.0%)	3 (21.4%)	3 (13.6%)
College degree	3 (37.5%)	5 (35.7%)	8 (36.4%)
Post-college degree	3 (37.5%)	6 (42.9%)	9 (40.9%)
Drinks per week				0.57
None	4 (44.4%)	7 (50.0%)	11 (47.8%)
1 to 5	3 (33.3%)	6 (42.9%)	9 (39.1%)
6 to 10	2 (22.2%)	1 (7.1%)	3 (13.0%)
11 to 15	0 (0.0%)	0 (0.0%)	0 (0.0%)
Income				0.02
Less than $25,000	4 (44.4%)	0 (0.0%)	4 (17.4%)
$25,000 to $49,999	1 (11.1%)	9 (64.3%)	10 (43.5%)
$50,000 to $74,999	2 (22.2%)	3 (21.4%)	5 (21.7%)
$75,000 to $99,999	0 (0.0%)	1 (7.1%)	1 (4.3%)
$100,000 to $199,999	2 (22.2%)	1 (7.1%)	3 (13.0%)
$200,000 or more	0 (0.0%)	0 (0.0%)	0 (0.0%)
Weight kg	93.7 (9.7)	98.7 (16.4)	96.8 (14.1)	0.42
BMI, kg/m2	34.8 (4.0)	36.6 (5.5)	35.9 (5.0)	0.40
Waist circumference cm	116.3 (7.5)	115.8 (10.4)	116.0 (9.2)	0.89
Comorbidities
Anxiety	2 (22.2%)	1 (7.1%)	3 (13.0%)	0.30
Coronary artery disease	1 (11.1%)	2 (14.3%)	3 (13.0%)	0.83
COPD	1 (11.1%)	0 (0.0%)	1 (4.3%)	0.21
Depression	3 (33.3%)	3 (21.4%)	6 (26.1%)	0.54
Diabetes	2 (22.2%)	2 (14.3%)	4 (17.4%)	0.63
Fibromyalgia	1 (11.1%)	1 (7.1%)	2 (8.7%)	0.75
High cholesterol	4 (44.4%)	4 (28.6%)	8 (34.8%)	0.45
Hypertension	5 (55.6%)	7 (50.0%)	12 (52.2%)	0.80
Non skin cancer	1 (11.1%)	0 (0.0%)	1 (4.3%)	0.21
Osteoarthritis	2 (22.2%)	6 (42.9%)	8 (34.8%)	0.32
Rheumatologic disease	0 (0.0%)	1 (7.1%)	1 (4.3%)	0.42
Sleep apnea	2 (22.2%)	4 (28.6%)	6 (26.1%)	0.74
Stroke	1 (11.1%)	0 (0.0%)	1 (4.3%)	0.21
ADL score	93.8 (9.8)	97.6 (6.4)	96.1 (7.9)	0.27
IADL score	82.1 (15.4)	84.1 (12.7)	83.3 (13.5)	0.73
Miles to aging resource center	15.0 (11.3)	16.6 (10.9)	16.0 (10.8)	0.75

Continuous variables are represented as mean (standard deviation), and categorical as counts [percent]. COPD is chronic Obstructive Pulmonary Disease, ADL is Activities of Daily Living, IADL is Instrumental Activities of Daily Living

Table 2 lists the study outcome measures. Diet quality as measured by REAP-S significantly improved at the conclusion of the intervention, with REAP-S scores increasing by 3.53±3.13 points (p < 0.001). Twenty-two out of 23 participants completed all four sets of the ASA-24; one participant did not complete a baseline ASA-24. In terms of individual nutrients, as measured by the ASA-24, diet quality significantly improved at the completion of 12 weeks. Intake of total kilocalories, grams fat, grams saturated fat, milligrams sodium, grams of added sugar, and grams of alcohol significantly decreased at the completion of the interventions (all p <0.05) (Table 2). A significant weight loss of 5.22±3.13 kg (5.39% weight loss; p <0.001) was observed at the completion of the interventions, at 12 weeks, with a reduction in waist circumference of 6.88±5.67 cm (p<0.001) also observed (Table 2).

Table 2:

Outcomes - REAP-S, ASA-24, and Anthropometrics

	Baseline	Week 12	Difference	p-value
REAP-S n=23
Cumulative Score	32.35 (3.42)	35.87 (3.46)	3.52 (3.13)	<0.001
Willing to Change	1.35 (0.65)	1.17	−0.17 (0.39)	0.04
Feels well to Cook, shop	23 (100)	23 (100)	0	N/A
ASA-24 Scores n = 22
Total kcal	2096.24 (649.95)	1544.16 (455.84)	−552.08 (761.84)	0.003
Protein (g)	76.95 (18.46)	76.75 (24.60)	−0.20 (30.70)	0.98
% kcal from Protein	15.34 (3.53)	20.20 (4.37)	4.86 (5.10)	<0.001
Carbohydrate (g)	234.42 (88.83)	176.06 (52.64)	−58.36 (91.87)	0.007
% kcal from carbohydrate	44.84 (9.15)	46.24 (8.48)	1.40 (6.68)	0.34
Fat (g)	90.70 (38.86)	62.08 (30.91)	−28.62 (40.30)	0.003
% kcal from fat	38.34 (10.06)	35.19 (9.92)	−3.14 (8.28)	0.09
Fiber (g)	20.35 (8.45)	22.60 (9.38)	2.25 (12.97)	0.43
Saturated fat (g)	29.80 (13.65)	17.28 (9.09)	−12.52 (13.30)	<0.001
Sodium (mg)	3297.14 (1024.55)	2664.90 (913.91)	−632.24 (1180.61)	0.02
Calcium (mg)	1026.58 (358.82)	872.81(260.69)	−153.77 (347.85)	0.05
Vitamin D (mcg)	4.92 (3.82)	3.99 (3.15)	−0.93 (5.03)	0.38
Sugars intake (g)	60.02 (41.26)	25.09 (31.80)	−34.93 (40.38)	<0.001
Other (alcohol, g)	4.92 (21.44)	−3.69 (8.22)	−8.60 (17.85)	0.04
Anthropometrics (n= 23)
Weight, kg	96.76 (14.08)	91.54 (14.99)	−5.22 (3.13)	<0.001
Waist Circumference, cm	115.97 (9.15)	109.08 (9.66)	−6.88 (5.67)	<0.001

Continuous variables are represented as mean (standard deviation). Kcal is kilocalories, g is grams, mg in milligrams, mcg is micrograms, kg is kilograms, kg/m² is kilograms per meters squared, REAP-S is Rapid Eating and Activity Assessment and ASA-24 is Automated Self-Administered 24-Hour Dietary Assessment Tool, ASA-24 values are from food intake.

Discussion

To our knowledge, this is the first evaluation of overall diet quality measuring several nutrition parameters using survey questionnaires and 24-hour recalls in older adults with obesity enrolled in a weight loss intervention. Frequent, once-weekly nutrition counseling sessions combined with physical activity resulted in significantly improved diet quality with reductions in both body weight and waist circumference. This multi-component intervention to enhance health is more effective than interventions that focus on one component of health counseling (51), and can be adapted in many clinical and research areas to improve health parameters in this population.

The improvements in REAP-S values can be interpreted as an indicator of improved nutrient intake and dietary diversity, which is linked to nutrient density and obesity prevention(52). REAP-S also reflects participants’ willingness to accept diet advice and to change food habits. The corresponding weight loss and reductions in waist circumference suggest the REAP-S should be considered for guiding nutrition interventions among large populations. In addition, participant’s willingness to change habits significantly increased - suggesting the success of the intervention on participant attitudes towards behaviors consistent with health. Its use in a community-based study is unique as the questionnaire was originally developed for physicians to increase nutrition interventions in the primary care setting(48).

Changes in the intake of individual nutrients can impact health(53). For example, intakes of sodium and saturated fat are linked with obesity(15, 54). Reducing energy dense foods with low levels of key nutrients may be linked with weight loss(55). These foods contain high levels of total fat, processed carbohydrates, added sugar, saturated fat, and sodium(56). The reduction these nutrients as measured by ASA-24 results may reflect a reduced intake of empty calories, which in turn supports weight loss. A reduction of total calories by more than 500 kcal per day was also observed, further indicating a possible drop in “empty calorie” intake, and aligning with current recommendations(8). Americans are advised to take no more than 10 percent of their total daily calories from saturated fat(53).

Participants in these interventions were encouraged to eat a plant-based diet which is low in saturated fats, and high in both mono- and polyunsaturated fats(57). Some groups have shown that decreasing saturated fat can lead to weight loss, despite total fat intake; at either a total dietary fat daily intake of 20% or 40%, diets that are lower in saturated fat are successful in promoting weight loss(58). Saturated fat intake was significantly reduced by the end of the intervention and this may in turn be related to the significant weight loss observed. Diets that are high in added sugar may increase insulin resistance and are thought to be linked with inflammation and obesity(59–61). Total added sugar intake was significantly lowered by the end of the intervention. Fiber intake has been shown to aid in weight loss(62). Fiber intake did not change significantly at week 12; perhaps this was due to the fact that mean fiber intake at baseline (> 20 g per day) was higher in study participants than the national average in US adults(63). It follows that, in both the clinical and research setting, a focus of nutrition counseling on the reduction of specific nutrients (fat, saturated fat, added sugar, and sodium) with a goal to increase fiber intake may be indicated in this population to best enhance health outcomes.

Heavier alcohol intake may be linked with increasing waist circumference and obesity, particularly in men(64, 65). A serving of alcohol can contribute several hundred calories to a person’s daily energy intake, increasing the chance of weight gain over time(65). Individuals who consume more alcohol generally consume less variety from other foods(64), which may indicate poorer diet quality and diversity. In moderate alcohol drinkers, calorie and nutrient intake from food differ on drinking and nondrinking days(66), with an increase in total calories, total fat, saturated fat, and sodium observed on drinking days(67). A significant decrease in alcohol intake at follow-up indicates that reducing alcohol may improve chances of weight loss and could be another focus of nutrition counseling in this population.

Despite efforts to increase participant protein intake, our findings did not demonstrate differences in protein intake at follow-up. Perhaps this is because of a slightly elevated intake of protein at baseline in our population (>75 g daily on average). However, the percent of total calories from protein increased significantly from 15.3 to 20.2%. Fat intake concurrently dropped, and total carbohydrate intake also decreased from baseline, but both macronutrient decreases were not significant. increasing percent kcal from protein may be beneficial in older adults who are attempting weight loss, as protein intake at around 20 % of calories may be protective and may modulate muscle loss and strength in this population(20).

Limitations

We acknowledge a number of limitations in this study. First, this pilot had a small sample size of 23 subjects. Secondly, the study participants and setting lacked diversity (predominantly white, female participants in a rural area of northern New England). Another possible limitation is the total number of diet recalls performed. Some groups have shown that a minimum of 5-10 days of 24-hour diet-recalls for men, and 12-15 days for women, are needed to accurately assess nutrient intake in adults with obesity(68). While having multiple longitudinal measurements for a lengthier time would enhance the ability to capture dietary quality, conducting the ASA-24 as outlined in our protocol maximized our ability to capture dietary data while minimizing participant burden. Since this was a community-based intervention, and to minimize the total number of staff that the older adult participants would have to interact with, behavioral psychologists were not part of study staff, though the protocol was overseen and reviewed by a behavioral psychologist. As an exploratory analysis, while we demonstrated that baseline characteristics between those with a BMI >40 did not significantly differ from that of <40kg/m², future studies should account for initial weight as a predictor of outcomes. Finally, study length could be considered a limitation. While 12 weeks may lead to observed changes, longer interventions may be needed to ascertain long-term changes(51).

Conclusion

Intensive nutrition counseling as part of a multi-disciplinary, multi-component weight loss intervention significantly enhances overall diet quality in rural older adults with obesity, and, in combination with regular scheduled physical activity, results in significant weight loss and reductions in waist circumference in this population. These feasibility results may serve as a basis for larger, lengthier trials, and RDNs may use the results of these pilots to design clinical or research programs in order to further the health of older adults with obesity.

Highlights:

• Obesity is a serious problem in older adults.

• Intensive nutrition counseling as part of a multi-component intervention improves diet quality in older adults with obesity

• Weight and waist circumference significantly improve with intensive nutrition counseling in older adults

• Much-needed clinical and community programs for older adults with obesity can be designed successfully