A Community-Based Feasibility Study of Weight-Loss in Rural, Older Adults with Obesity

Abstract

This study was a 12-week feasibility weight-loss intervention consisting of caloric restriction and aerobic/resistance exercise in older adults with obesity (body mass index ≥ 30kg/m²) in a geographically isolated area. Primary outcomes assessed weight and physical function. Mean age was 71.0±5.1years (67% female). Individuals completed 100% of all assessments, attended 88% of the physical therapy classes and 89% of the nutrition sessions. Level of satisfaction (5-point Likert) was high (5.0, 1 – low; 5 - high). Weight decreased from 93.7±9.7 to 89.4±4.0kg (p<0.001). Mean BMI and waist circumference decreased, respectively, from 35.4±3.4 to 33.6±3.7 (p<0.001), and 116.3±7.5 to 108.7±9.2cm (p=0.002). Grip strength, gait speed, and 5-times sit-to-stand time all improved from 29.2±7.5 to 35.2±6.7kg (p=0.006), 1.16±0.21 to 1.35±0.23 m/s (p=0.004), and 12.5±4.0 to 9.6±1.7s (p=0.02). The intervention was feasible and acceptable, and holds promise in promoting weight loss with a concomitant improvement in physical function in older adults.

INTRODUCTION

Obesity in older adults aged 65+ is a growing concern with over 40% of the US population fulfilling diagnostic criteria using body mass index (BMI).¹ Excess fat is strongly associated with an increased risk of falls, mobility impairment, institutionalization and mortality, in part due to intramuscular fat deposition that contributes to adverse outcomes.² Efficacy-based weight-loss interventions in older adults consisting of caloric restriction and exercise lead to weight-loss and enhance physical function with preservation of lean mass³. However, the shortcomings of current academic center-based trials in older adults is the challenge in translating them to community-based settings making them less externally-valid, and thus difficult to implement.⁴ For instance, to our knowledge, only two successful community-based trials exist: Rejeski’s YMCA study⁵ focused on calorie restriction with aerobic or resistance training which showed improvements in 400m walk time without any weight regain; and the CROSSROADS diet-exercise trial which found reductions in body fat⁶.

Health promotion programs in rural areas are negatively impacted by the geographic barriers, access to care, built environment, workforce shortages of qualified personnel, and availability of programs.^7,8. Rates of obesity in rural areas are 35.6–37.7%⁹ in older adults making a need for such programs even greater. Older adults have limited mobility and transportation options, need to drive long distances, and face restrictions in navigating the built environment. Further, there are no studies predominantly focusing on older adults residing in rural areas. While MOVE-UP’s¹⁰ use of community health workers in delivering a healthy aging, diet/exercise intervention may provide insights into rural geriatric obesity care, developing interventions unique to rural needs that could potentially be disseminated and scaled to regions without adequate expertise or staff is a key first step in reaching such distant populations. Our formative work suggested that rural older adults with obesity desired opportunities to engage in multi-component interventions, even if it required individuals to overcome distance and transportation hurdles.¹¹

This feasibility study was the first of a series of developing a research program in rural health promotion among an at-risk population for obesity-related comorbidities residing in rural areas. The feasibility, acceptability and preliminary effectiveness of a multi-component weight-loss intervention delivered at a local community-aging center was ascertained. Data from this feasibility study will permit refinement of measures and provide a foundation for further studies this population.

MATERIALS AND METHODS

Design & Setting

A 12-week weight-loss intervention was conducted in the Northeastern United States (USA), in Lebanon, New Hampshire, a small rural community of 13,522 persons, at a community-based aging center affiliated with Dartmouth-Hitchcock, a rural, academic medical center. We based our definition of rural using the urban-rural classification from the US Census on the basis of population density.¹² The Committee for the Protection of Human Subjects at Dartmouth College and the Dartmouth-Hitchcock Institutional Review Board approved the study. The trial was registered on Clinicaltrials.gov (NCT03104192).

Recruitment

Participants were recruited from the Dartmouth-Hitchcock primary care clinic using posters, tear-off cards, and presentations to faculty. Selection criteria were reviewed by the research assistant through an electronic medical record review. Inclusion criteria consisted of English-speaking, community-dwelling participants, age ≥65 years, a BMI greater than 30kg/m². Participants were excluded if they had cognitive impairment, uncontrolled psychiatric illness; a history of bariatric surgery, life-threatening illness or receiving palliative/hospice services; participated in another weight-loss program; on medications promoting obesity (steroids, anti-psychotics); or had advanced cardiac, renal or liver dysfunction subject to the obesity paradox.¹³ Additional exclusion criteria consisted of weight-loss ≥5% in the past six months, a Callahan cognitive¹⁴ screen of <3, or a Functional Status Questionnaire¹⁵ score of less than 71.2 and 56.4 for basic and instrumental activities of daily living, respectively. Baseline demographics were abstracted from the medical record and obtained from questionnaires, including social support.¹⁶ Participants were compensated with a $25 gas card for each assessment.

Intervention:

A licensed, registered dietitian supervised and delivered the diet and behavioral intervention by meeting participants individually for 30 minutes weekly to go over education materials and food diaries (Appendix 1 for curriculum). Sessions focused on intensive behavioral therapy using motivational interviewing, on specific, measurable, attainable, relevant and timely goals using evidence-based materials. Individualized meal plans were created at baseline using data from the 24-hour automated self-administered 24-hour dietary assessment tool.¹⁷ A calorie restricted diet of 500–750 kCal/day (minimum intake of 1200 kCal/day) with sufficient dietary vitamin D (1000 IU/day) and protein (1–1.2 g/kg/day or 20% of intake) intake was advised. All participants were provided a digital scale for home self-monitoring of weekly weights.

An initial physical therapy-based assessment of strength, flexibility, balance and aerobic capacity permitted the creation of personalized exercise plans aimed at gradually increasing physical activity level. All participants subsequently participated in a group-based exercise session twice weekly at the center led and guided by the physical therapist, aimed at performing moderate intensity targeting major muscle groups using resistance bands and adjustable cuff weights. The participants were encouraged to perform the same strength, flexibility and balance exercises one day per week outside of the structured sessions, spaced 24 hours apart and/or focusing on different muscle groups using the same bands and weights. Aerobic exercise was performed independently and progressed weekly with guidance. These exercises were guided and tracked using structured, weekly diaries and encouraged for 150 minutes of moderate-vigorous intensity activity weekly, reviewed by the therapist at the twice weekly visits.

Outcome measures

Feasibility was defined as achieving a target enrollment of eight participants to ensure our intervention and its content could be delivered to a larger population with limited access to such programs. The proportion of enrollees relative to those screened and the proportion of those eligible wishing to participate of those screened positively was assessed. A dropout of <20% was considered successful retention. Acceptability measures entailed attending >75% of sessions and completing >80% of study measures, and were assessed using a Likert scale (range 1–5, low-high) of the following questions: “how would you rate your level of satisfaction with the intervention; how helpful was the overall intervention in assisting you in achieving your goals. A question (yes/no) was asked as to whether the participant would recommend the intervention to a family member.

Weight and height were measured using an A+D digital scale and a Seca 216 stadiometer, respectively. BMI was calculated as weight in kilograms divided by height in meters squared. Waist circumference was measured at the level of the iliac crest using a standard tape measure. Hip circumference was measured at the widest part of the buttocks with the tape parallel to the floor. Waist-hip ratio was calculated as the ratio of the two circumferences. A trained research assistant assessed grip strength, gait speed, 5-times sit-to-stand, and 6-minute walk tests at the center. We measured handgrip strength using a JAMAR dynamometer. Participants squeezed the device as tightly as possible for 15 seconds. The maximum value of three trials of each hand was used in the analysis.Gait speed was timed over a distance of 5m; there was 1.67 meters on either side of timed markings that represented an acceleration and deceleration phase. The sit-to-stand test was performed with participants seated at the edge of a chair, arms folded, buttocks hitting the chair on each of the five repetitions. Six-minute walk test was conducted to evaluate aerobic function in a long 70 foot corridor in accordance with guidelines from the American Thoracic Society¹⁸, with cones spaced every 10 feet. Subjective measures of Patient Reported Outcome Measurement Information System¹⁹ (PROMIS) Global function (physical and mental health), the Function component of the Late-life Function and Disability Instrument,²⁰ and the Patient Activation Measure²¹ were collected using a tablet-based version of RedCAP. Adverse events were reported by participants to the intervention staff throughout the study, and logged by the research assistant per institutional guidelines.

Statistical Analysis

Baseline measures were evaluated using descriptive statistics. The primary outcomes were change in weight and physical function. All pre-post continuous variables were assessed using paired t-tests and categorical variables using chi-square, or their non-parametric equivalents. All data was analyzed using R (www.R-project.org). A p-value of <0.05 was considered statistically significant.

RESULTS

Table 1 represents the baseline characteristics. Mean distance to the center was 15.0±11.3miles, mean age was 71.0 years (66.7% female, 100% Caucasian). Socioeconomic heterogeneity was high as evidence by low education (n=2) and low income (n=4). Participant ADL and IADL scores were high. Participants had low scores of social support with exercise in family and friend participation (20±10.7 and 14.9±6.0).

Table #1:

Participant Characteristics

	Total (N=9)
Age, years	71.0±5.1
Female Sex, %	6 (66.7)
White race, %	9 (100)
Socioeconomic Factors
Marital status, %
Single	1 (11.1%)
Married	6 (66.7%)
Divorced	2 (22.2%)
Widowed	0 (0.0%)
Insurance#, %
Medicare	9 (100)
Medicaid	0 (0)
Private insurance	7 (77.8)
Smoking status, %
Non-smoker	2 (22.2)
Former smoker	7 (77.8)
Education, %
High school	2 (25.0)
Some College	---
College Degree	3 (37.5)
Post-College Degree	3 (37.5)
Income, %
Less than $25,000	4 (44.4)
$25,000 to $49,999	1 (11.1)
$50,000 to $74,999	2 (22.2)
$75,000 to $99,999	---
$100,000 to $199,999	2 (22.2)
Anthropometric Measures
Weight, kg	93.7±9.7
BMI, kg/m2)
Waist circumference, cm	116.3±7.5
Co-morbidities
Anxiety, %	2 (22.2)
Coronary artery disease, %	1 (11.1)
COPD, %	1 (11.1)
Depression, %	3 (33.3)
Diabetes, %	2 (22.2)
Fibromyalgia, %	1 (11.1)
High cholesterol, %	4 (44.4)
Hypertension, %	5 (55.6)
Non skin cancer, %	1 (11.1)
Osteoarthritis, %	2 922.2)
Rheumatologic disease, %	---
Sleep apnea, %	2 (22.2)
ADL score	93.8±9.8
IADL score	82.1±15.4

All variables indicated are mean ± standard deviations, or counts (%)

Abbreviations: ADL – activities of daily living; BMI – body mass index; COPD – chronic obstructive pulmonary disease; CAD – coronary artery disease; IADL – instrumental activities of daily living; NAFLD – non-alcoholic fatty liver disease; OSA – obstructive sleep apnea.

^#-participants had both types of insurances

A CONSORT diagram is presented in Figure 1. Of 14 screened, 10 (71.4%) enrolled; one was unable to participate due to a scheduling conflict. There were no dropouts (100% completion rate). Of the 24 physical therapy sessions, participants attended a mean of 23 (median 21, IQR 20–23), with 88% attending all sessions. Similar results were observed for nutrition where 89% attended all classes, and mean number of classes attended was 11.6 of 12 (median 12, IQR 11–13). All 9 participants noted their level of satisfaction with this intervention as high (5.0±0.0) and that the intervention permitted achievement of goals (5.0±0.0). Participants would have recommended this intervention to a family member with a weight-problems (89%).

Figure 1:

CONSORT flow diagram presented

Participants achieved a mean weight-loss of 4.3±3.2kg (4.9±3.7%) at 12-weeks (p<0.001). Waist circumference dropped significantly (Δ=7.6±4.9; p=0.002), as did waist-hip-ratio (Δ=−0.03±0.04; p=0.04). Figure 2 represents the per-person change in weight over the intervention. There were marked improvements in objective functional measures (all p<0.05). We found significant improvements in patient activation scores (Δ=9.22±11.3; p=0.04) and in the function from the late-life function and disability instrument (Δ=3.7±4.8;p=0.05). Only significant changes were observed in the PROMIS mental health scores (Table 2). There was a total of nine adverse events. Expected study-related events were predominantly musculoskeletal (n=4). One participant sustained a transient ischemic attack believed to be unrelated to the study after conferring with the safety monitor. Another was diagnosed with diverticulitis. Three participants developed headaches that resolved (n=3).

Figure 2:

Individual changes in the n=9 participants in this multicomponent weight loss study at baseline and after 12-weeks of intervention. Horizontal bars and corresponding P values indicate comparisons between these two time points. *P <0.05 compared with baseline.

Table 2:

Preliminary Objective Effectiveness Measures

	Baseline (N=9)	Week 12 (N=9)	Difference (N=9)	Percent Change	p value
Anthropometric
Weight, kg	93.7±9.7	89.4±12.1	−4.3±3.2	−4.9±3.7	<0.001
BMI, kg/m2	35.4±3.4	33.6±3.7	−1.7±1.2	−4.9±3.7	0.002
Waist circumference, cm	116.3±7.5	108.7±9.2	−7.6±4.9	−6.5±4.4	0.002
Hip circumference, cm	121.0±5.3	116.9±7.3	−4.1±4.0	−3.4±3.4	0.02
Waist to hip ratio	0.96±0.05	0.93±0.03	−0.03±0.04	−3.2±4.1	0.04
Objective Measures
6 min walk test, m	400.6±81.0	448.1±122.0	47.5±83.0	11.5±18.93	0.12
Max Gait Speed, s	1.16±0.21	1.35±0.23	0.19±0.15	17.7±13.3	0.004
Mean Grip Strength, kg	29.1±7.5	35.2±6.7	6.2±5.1	---	0.02
5-Times Sit To Stand, s	12.5±4.0	9.6±5.2	−2.9±2.9	---	0.02
Subjective Measures
Late-Life Functionality
Total	58.9±11.5	62.6±11.1	3.7±4.8	6.9±7.3	0.05
Upper extremity	77.1±15.4	83.9±15.9	6.8±12.0	10.0±14.2	0.13
Basic lower extremity	71.3±16.5	80.8±17.3	9.5±12.4	14.5±14.6	0.05
Patient Activation Measure	68.1±17.5	77.3±19.3	9.22±11.3	15.0±15.6	0.04
PROMIS, total
Mental Health T score	52.0±3.8	55.7±5.9	3.60±3.5	6.8±6.5	0.02
Physical Health T score	46.6±6.5	48.6±6.8	2.95±4.4	6.9±11.2	0.10

All values listed are mean (SD), or count (%)

Abbreviations: BMI: body mass index; PROMIS – patient reported outcomes measurement information systems.

DISCUSSION

This small feasibility study provides foundational data to suggest our approach is feasible and acceptable to older adults with obesity in developing a community-based, rural weight-loss intervention. Participants lost weight, improved their function, and felt the program had promise in improving health.

Nutrition and weight are key priorities in Rural Healthy People 2020,⁸ and a recent review suggested that significant weight-loss can be achieved in rural areas.²² A number of different strategies and follow-up exist within rural areas including group phone calls (vs. in-person),²³ hybrid phone/in-person (vs. low/high intensity visits)²⁴, or community-health workers.²⁵ Yet, such trials mainly focus on behavior or nutrition, rather than diet and exercise and do not target seniors. Our results have potential implications for improving rural health as rural areas lag in offering health-promotion programs and few studies focus specifically on weight-loss.^5,6,10

The intervention delivery was feasible as was the ability to recruit despite geographic distances suggesting that the trial can be considered elsewhere. It is unknown if longer distances would be acceptable to travel. The intervention’s content held promise; while many rural regions may not necessarily have the staffing resources to deliver such a program, our approach provides a basis for remote-based interventions using telehealth that can overcome many of these rural health disparities. Further, participants were compensated for each assessment that may have removed barriers to transportation. This approach that enhances attendance parallels and supports the currently sponsorship programs related to meals, nutrition education and exercise classes in rural areas provided by the local area agencies on aging.²⁶

Weight-loss in older adults is not highly endorsed by clinicians, partly due to the potential belief of the protective effect of fat on mortality (e.g., obesity paradox).¹³ However, intentional weight-loss is associated with lower mortality²⁷ and the paradox is fraught with challenges²⁸. Weight-loss interventions are even safe and efficacious in frail, older adults with obesity that lead to improved physical performance.²⁹ Our approach was intended to mitigate weight-loss induced sarcopenia; we demonstrated improved function, despite weight-loss. Future studies should evaluate changes in body composition during such efforts.

While we acknowledge that the intervention lasted only 12-weeks and that longer-term strategies are needed, participants achieved a significant degree of weight-loss of 5%. While there was variability, as evidenced by the standard deviation of the group’s weight loss, we recognize that not all interventions are suited for everyone. Importantly, there were improvements in various functional parameters suggesting that further investigation should evaluate whether our approach mitigates weight-loss induced sarcopenia. It is unclear why we failed to observe changes in subjective measures; whether their improvement lags or the instruments are not sensitive to change require additional research. The recruitment area lacked racial diversity with over 95% identifying themselves as Caucasian making generalizability a concern. Rural areas differ extensively; whether our intervention can be delivered elsewhere is unknown. Cost implications are real and only a larger and longer study can validate our results. Remote-based telehealth strategies can also address rural health barriers to care delivery to limited resource availability. Ensuring the content of the intervention is acceptable is a necessary first step to delivering an intervention remotely. While our results are promising, we do acknowledge they should be interpreted with caution due to a lack of a control group that would provide more robust results, and to a small sample size that prohibited subgroup analyses.

This study provided invaluable information to permit modification of the existing intervention. First, the intervention materials were adjusted to add additional behavioral strategies. Second, we recognized the importance of parsimony, both for objective measures and subjective questionnaire data. Third, body composition is an important factor in proving that this intervention did not lead to loss of muscle mass. Fourth, standardized monitoring of activity and data is needed to enhance motivation and track progress. Last, we recognized that our inclusion and exclusion criteria may have been too limited, thus modifying them to permit a more pragmatic approach should be considered in the design of a larger, randomized controlled trial.

TAKE AWAY POINTS.

• A weight management intervention is feasible and acceptable in older adults

• Physical function can improve in populations at high risk for disparities

• Evidence-based programs can be adapted for rural populations