Dietary and behavior changes following RDN-led corporate wellness counseling: A secondary analysis

Abstract

BACKGROUND:

Worksite wellness programs have the ability to activate health promotion and stimulate behavior change.

OBJECTIVE:

To measure longitudinal associations between visits with a Registered Dietitian Nutritionist (RDN), as part of worksite wellness programs, on dietary and lifestyle behavior changes.

METHODS:

The study sample included 1,123 employees with 77 different worksite wellness programs across the United States from March to December 2017. Hierarchical linear modeling was used to evaluate the associations of RDN visits with behavior changes.

RESULTS:

The mean BMI at baseline was 33.48, indicating over half of all employees are considered obese. Employees who attended more than one visit showed an increase in whole grain consumption and corresponding weight loss (t-ratio = 2.41, p = 0.02). Age played a significant factor in the rise of systolic blood pressure; employees who attended more visits showed an increase in whole grain consumption and corresponding blood pressure (t-ratio = –2.11, p = 0.04).

CONCLUSIONS:

RDNs as part of worksite wellness programs, can contribute to improvements in lifestyle behavior changes. These data highlight the need for nutrition intervention at the workplace. Research on nutrition-focused worksite wellness programs is needed to assess the long-term health outcomes related to dietary and lifestyle behavior changes.

Keywords

Worksite safety and health behavior change nutrition health promotion prevention

1 Background

Obesity is a major and costly public health epidemic, which affects more than 78 million American adults [1, 2]. Obesity results in a significant increase in medical disbursements and truancy among full-time employees [3]. U.S. companies lose an estimated $4.3 billion each year due to obesity-related absenteeism and loss of productivity [4, 5]. Obesity increased work productivity impairment in twelve different occupations between 7.3%to 19.3%when using the Work Productivity and Activity Impairment-General Health questionnaire. As such, indirect costs due to obesity were between $860 to $5,622 per year [6]. It is estimated that 16%to 18%of total U.S. healthcare costs will be directly ascribed to the overweight and obese by the year 2030, for a value of $860.7 to $956.9 billion [7]. Cawley et al. reported that adult obesity raised the annual medical care costs by $3,508 per obese individual, an estimated cost to the United States of $315.8 billion [4]. Furthermore, for every one unit increase in BMI medical and pharmaceutical costs rise by 4%and 7%[8]. Due to these extreme costs, employers are finding it imperative to their business success to improve the health of their employees.

Employer-sponsored wellness programs are one strategy for addressing employees’ chronic disease risk factors, promoting healthy lifestyles, and reducing employers’ healthcare costs. In 2013, 16%of employees reported having mandatory participation in employer wellness programs [9]. These worksite wellness programs are expected to improve employee health, therefore curb employer costs. However, the efficacy of said programs is still up for debate. A systematic review by Lerner et al., found that the economic impact of worksite wellness programs to be limited and inconsistent [10]. Similarly, van Dongen and colleagues examined net benefits, benefit-cost-ratio and return on investments in worksite wellness programs and reported that evidence on profitability to be inconclusive. The authors, however, noted that financial savings in terms of reduced absenteeism and medical costs were evident in the review of non-randomized control studies [11]. It’s been reported that lower average medical claim costs among participants enrolled in a worksite wellness program compared to non- participants [12]. Despite the inconsistency in the effectiveness of these programs, the prevalence of employer-sponsored wellness programs that target weight or BMI is expected to increase [13, 14].

The approach of the worksite wellness program has a great impact on the success of the program. There are various types of worksite wellness programs. A comprehensive worksite wellness program, focused on nutrition, physical activity and stress has been shown to have significant long-term impact on weight loss and weight maintenance in overweight and obese individuals [15]. Delivery of worksite wellness programs by trained professionals, including Registered Dietitian Nutritionists (RDN) has strong implications on the success of the program. For example, worksite wellness programs that included a dietitian, resulted in significant changes in biochemical, anthropometrics, fruit and vegetable intakes and improvements in other cardiometabolic risk factors [16 –18]. RDNs are trained professionals in evidence-based nutrition interventions. Most worksite wellness initiatives include the use of nutrition to improve health. As such, RDNs provide a needed resource for many worksite wellness programs [19, 20]. Medical Nutrition Therapy delivered by RDNs is an effective tool for the management, prevention, or reversal of common metabolic conditions including prediabetes and type II diabetes, obesity, and other metabolic conditions associated with weight [21 –23]. Although shown as effective treatment, few worksite wellness programs, and only in localized one-facility programs, utilized nutrition counseling from RDNs in their approach.

This study contributes to the literature by using hierarchical linear modeling to discover the associations between visits with an RDN with dietary and lifestyle behavior modification. This study is the first of its kind to analyze data from multiple companies across the country to describe the valuable impact from RDN-led worksite wellness programs.

2 Methods

2.1 Aim and objective

The primary aim of this study was to determine changes in dietary and lifestyle related factors after face-to-face counseling intervention(s) by a Registered Dietitian Nutritionist (RDN) as part of multi-site worksite wellness programs. It was hypothesized that the participants who returned for more than one counseling session would exhibit a statistically significant improvement in anthropometrics and dietary intake compared to baseline measurements.

2.2 Data

This study examined the retrospective association between dietitian visits, as part of worksite wellness programs, and self-reported behaviors from March 2017 to December 2017. De-identified individual-level employee data was obtained, and then cleaned by an investigator on the study team. The study population was limited to a cohort of 2,705 employees, of which only 1,123 had follow up visits, who were employed by one of the participating worksite wellness locations. Each company had a varied composition of their respective worksite wellness programs, some including a range of opportunities beyond consultations with a registered dietitian nutritionist (e.g. body composition testing, annual physical, smoking cessation programs, online health databases). Participants were unequally distributed across 77 employers from various industries and locations across the United States. Participation was voluntary and participants could engage in one or more sessions, depending on their needs and/or goals. Incentives for participation in worksite wellness varied per company, of which all visits required no out-of-pocket expenses or monetary compensation for the employee.

Each visit documented was an individual face-to-face interaction between an employee and a Registered Dietitian Nutritionist (RDN). RDN’s have been trained on basic motivational interviewing and counseling skills which were used during each employee session. Nutrition coaching and Medical Nutrition Therapy was based on personalized instruction, involving participants’ health goals and associated nutrition and lifestyle activity. Each participant received a counseling session ranging from 15–60 minutes, depending on the needs of the participant. Data were collected from March to December 2017. Analyses were conducted in 2018.

2.3 Measures

Participation in worksite wellness was defined as participating in at least one counseling session with an RDN at the work location. For each session, the RDN recorded assessment and treatment notes via an internal medical charting software. No details on individual nutrition assessment were provided to the researcher team nor analyzed in this study.

Data collected from each employee at each visit included:

Self-reported anthropometrics (height, weight, calculated BMI)

RDN measured anthropometrics (systolic blood pressure, diastolic blood pressure, waist circumference, neck circumference)

Self-reported dietary intake (daily servings of whole grains, vegetables, fruit, dairy, fish and water)

Self-reported physical activity measures

Nutrition, physical activity and social goals (set between RDN and employee and monitored progress at each session)

This study was deemed exempt by the Texas A&M University Institutional Review Board. This study was deemed exempt under federal regulation 45 46.101 (IRB2016–0674).

2.4 Statistical analysis

SPSS syntax was created to create two databases, one for level-1 (appointment-level) data and one for level-2 (patient-level) data. Level-1 data contained 2,180 appointments with patient ID numbers repeated for each appointment reported by that particular patient (i.e., each patient will have multiple rows of data). Level-2 data included only one entry per patient and contained 1,123 patients. This type of multilevel data analysis allows for analysis of variance across levels of data, controlling for the variances explained by level-2 group membership. Significant relationships were identified if p < 0.05. Descriptive statistics were are presented in Table 1. Two separate HLM models were conducted; one with weight as the dependent variable and one with systolic blood pressure as the dependent variable.

Table 1
Descriptive statistics

Appointment-level

Baseline^a Last appointment Overall^b

Variable M (n) SD (%) M (n) SD (%) M (n) SD (%)

Weight (lbs) 204.03 51.52 207.43 51.25 206.63 51.45

Systolic (mmHg) 130.05 17.93 129.17 18.38 129.43 18.28

Diastolic (mmHg) 84.58 12.56 84.7 12.09 84.44 12.33

BMI 33.48 7.83 34.28 7.91 34.04 7.90

Whole grain servings 1.65 1.09 1.69 1.07 1.66 1.06

Dairy servings 1.24 0.88 1.23 0.85 1.23 0.86

Combined fruit and vegetable servings 3.64 1.70 3.68 1.55 3.67 1.58

Physical activity intensity 1.22 0.48 1.14 0.40 1.17 0.42

Physical days/week 2.01 2.05 2.03 1.97 2.04 2.02

Waist:neck ratio 2.48 1.00 2.60 1.32 2.52 1.20

Patient-level

Variable M (n) SD (%)

Male gender (230) (20.5)

Apple body type (725) (64.6)

Diabetes (100) (8.9)

Number of appointments 2.20 1.45

Age 46.95 12.19

	Appointment-level
Weight (lbs)	204.03	51.52	207.43	51.25	206.63	51.45
Systolic (mmHg)	130.05	17.93	129.17	18.38	129.43	18.28
Diastolic (mmHg)	84.58	12.56	84.7	12.09	84.44	12.33
BMI	33.48	7.83	34.28	7.91	34.04	7.90
Whole grain servings	1.65	1.09	1.69	1.07	1.66	1.06
Dairy servings	1.24	0.88	1.23	0.85	1.23	0.86
Combined fruit and vegetable servings	3.64	1.70	3.68	1.55	3.67	1.58
Physical activity intensity	1.22	0.48	1.14	0.40	1.17	0.42
Physical days/week	2.01	2.05	2.03	1.97	2.04	2.02
Waist:neck ratio	2.48	1.00	2.60	1.32	2.52	1.20
Patient-level
Variable	M (n)	SD (%)
Male gender	(230)	(20.5)
Apple body type	(725)	(64.6)
Diabetes	(100)	(8.9)
Number of appointments	2.20	1.45
Age	46.95	12.19

Note = ^aIncludes data from patients who only had one appointment, ^bincludes all appointments from all patients.

Separate predictive equations were created to account for the multiple levels. The outcome variable is identified as Y in models with linear outcomes [24]. The level-1 equation was:

Y =β₀ +β₁1X₁ +β₂X₂ + R

Where Y = the dependent variable, β₀ is the constant (i.e., intercept), β₁ is the regression coefficient of X₁ (i.e., whole grain intake), β₂ is the effect of X₂ (i.e., physical activity intensity), and R is error. In multilevel analysis, each X becomes an outcome for level-2 equations. The predicted level-2 equations were therefore:

β₀₀ = γ ₀₀ + γ ₀₁W₁ + γ ₀₂W₂ + γ ₀₃W₃ + γ ₀₄W₄ + u₀

β₁₀ = γ ₁₀ + γ ₁₁W₁ + γ ₁₂W₂ + γ ₁₃W₃ + γ ₁₄W₄ + u₁

β₂₀ = γ ₂₀ + γ ₂₁W₁ + γ ₂₂W₂ + γ ₂₃W₃ + γ ₂₄W₄ + u₂

where γ₀₀ is the overall mean intercept (i.e., intercept of the intercepts), γ₀₁ is the regression coefficient of W₁ (i.e., gender) relative to its intercept, γ₀₂ is the regression coefficient of W₂ (i.e., body type) relative to its intercept, γ₀₃ is the regression coefficient of W₃ (i.e., number of visits) relative to its intercept, γ₀₄ is the regression coefficient of W₄ (i.e., age) relative to its intercept, γ₁₀ is overall mean of β₁ adjusted for all W, γ₂₀ is overall mean of β₂ adjusted for all W, γ₁₁ is the regression coefficient for the effect of W₁ on β₁, γ₁₂ is the regression coefficient for the effect of W₂ on β₁, γ₁₃ is the regression coefficient for the effect of W₃ on β₁, γ₁₄ is the regression coefficient for the effect of W₄ on β₁, γ₂₁ is the regression coefficient for the effect of W₁ on β₂, γ₂₂ is the regression coefficient for the effect of W₂ on β₂, γ₂₃ is the regression coefficient for the effect of W₂ on β₃, γ₂₄ is the regression coefficient for the effect of W₄ on β₃, u₀ is the random effects on the intercept, u₁ is the random effects on the slope of β₁, and u₂ is the random effects on the slope of β₂. This model will test both direct and moderating effects while controlling for all variables in the model. The hypothesized models are shown in Fig. 1

Fig. 1

This figure shows the hypothesized model being tested, including all 28 individual hypotheses being tested.

3 Results

The first models analyzed were the null models for both dependent variables. The null models indicated that approximately 98.7%of the variance in weight was at the patient-level (rather than at the appointment-level) and approximately 84.0%of the variance in systolic blood pressure was at the patient-level. The results of the full HLM modeling for weight and systolic blood pressure are presented in Tables 2 3 respectively.

Table 2
Results of the HLM modeling for weight

Fixed effects Coefficient Standard error

Intercept ( β₀₀)

Intercept ( γ ₀₀) 311.55^* 17.58

Gender ( γ ₀₁) –22.08^ 8.36

Body type( γ ₀₂) –26.36^*** 5.86

Number of visits ( γ ₀₃) –0.57 1.79

Age ( γ ₀₄) –0.43 0.30

Whole grain servings ( β₀₁)

Intercept ( γ ₁₀) 2.78 6.36

Gender ( γ ₁₁) –3.95^* 1.83

Body type( γ ₁₂) –0.35 1.87

Number of visits ( γ ₁₃) 1.27^* 0.53

Age ( γ ₁₄) –0.02 0.07

Physical activity intensity ( β₀₂)

Intercept ( γ ₂₀) –19.92 13.06

Gender ( γ ₂₁) 3.76 6.38

Body type( γ ₂₂) –0.64 3.20

Number of visits ( γ ₂₃) –0.24 1.08

Age ( γ ₂₄) 0.23 0.22

Fixed effects	Coefficient	Standard error
Intercept ( β₀₀)
Intercept ( γ ₀₀)	311.55^***	17.58
Gender ( γ ₀₁)	–22.08^**	8.36
Body type( γ ₀₂)	–26.36^***	5.86
Number of visits ( γ ₀₃)	–0.57	1.79
Age ( γ ₀₄)	–0.43	0.30
Whole grain servings ( β₀₁)
Intercept ( γ ₁₀)	2.78	6.36
Gender ( γ ₁₁)	–3.95^*	1.83
Body type( γ ₁₂)	–0.35	1.87
Number of visits ( γ ₁₃)	1.27^*	0.53
Age ( γ ₁₄)	–0.02	0.07
Physical activity intensity ( β₀₂)
Intercept ( γ ₂₀)	–19.92	13.06
Gender ( γ ₂₁)	3.76	6.38
Body type( γ ₂₂)	–0.64	3.20
Number of visits ( γ ₂₃)	–0.24	1.08
Age ( γ ₂₄)	0.23	0.22

Note: # = p≤0.10, ^* = p≤0.05, ^** = p≤0.01, ^*** = p≤0.0

Table 3

Results of the HLM modeling for systolic blood pressure

Fixed effects	Coefficient	Standard error
Intercept ( β₀₀)
Intercept ( γ ₀₀)	117.79^***	15.63
Gender ( γ ₀₁)	–9.07#	5.28
Body type( γ ₀₂)	–1.09	4.94
Number of visits ( γ ₀₃)	2.15	1.66
Age ( γ ₀₄)	0.47^*	0.24
Whole grain servings ( β₀₁)
Intercept ( γ ₁₀)	1.05	4.19
Gender ( γ ₁₁)	–0.73	1.50
Body type( γ ₁₂)	4.56^**	1.58
Number of visits ( γ ₁₃)	–0.082^*	0.38
Age ( γ ₁₄)	–0.06	0.06
Physical activity intensity ( β₀₂)
Intercept ( γ ₂₀)	–0.14	11.33
Gender ( γ ₂₁)	6.22#	3.58
Body type ( γ ₂₂)	–4.46	3.15
Number of visits ( γ ₂₃)	–1.68	1.26
Age ( γ ₂₄)	–0.01	0.19

Note: # = p≤0.10, ^* = p≤0.05, ^** = p≤0.01, ^*** = p≤0.001.

The HLM model presented in Table 2 indicates that patient gender (t-ratio = –2.64, p = 0.008) and patient body type (t-ratio = –4.50, p < 0.001) have statistically significant direct effects on weight. Females, on average, were 22.08 pounds less than males. Also, patients with a pear body shape were, on average, 26.36 pounds lighter than patients with an apple body shape independent of gender. This HLM model also identified two statistically significant cross-level interaction (moderating) effects. The first such effect was gender’s moderation (t-ratio = –2.16, p = 0.03) of whole grain servings’ effect on weight. The second such effect was number of visit’s moderation (t-ratio = 2.41, p = 0.02) of whole grain servings’ effect on weight.

The HLM model presented in Table 3 indicates that patient age (t-ratio = 1.98, p = 0.05) has a statistically significant direct effect on systolic blood pressure. For every additional year in age, systolic blood pressures increase.468 units on average. This HLM model also identified two statistically significant cross-level interaction (moderating) effects. The first such effect was body type moderation (t-ratio = 2.88, p = 0.004) of whole grain servings’ effect on systolic blood pressure. The second such effect was number of visit’s moderation (t-ratio = –2.11, p = 0.04) of whole grain servings’ effect on systolic blood pressure.

4 Conclusion

These results demonstrate the current need for nutrition intervention within worksite wellness, as many employees have room for improvement in lifestyle choices for optimal health. These data indicate that regardless of nutrition counseling, there are improvements and regressions in diet and lifestyle behaviors. Moreover, these impacts were seen at multiple worksite wellness programs, across multiple companies, across multiple states. Positively, it was seen that as clients increased the number of visits with an RDN, their whole grain servings improved and associated systolic blood pressure. These findings support the national recommendation from the Dietary Guidelines for Americans 2015–2020 to increase whole grain intake as part of a healthy eating pattern [25]. Moreover, there was slight improvement in the consumption of daily fruits and vegetables and weekly physical activity, all important characteristics of a healthy lifestyle.

The research has shown that implementing a worksite wellness program and utilizing behavior change counseling, such as motivational interviewing, can be key to promote behavior change and reducing risk of chronic disease [26, 27]. While no largely significant differences were observed in this study, there were slight improvements in some dietary and lifestyle behaviors. Other single-site programs have shown similar results in improvements of nutrition and lifestyle behaviors, but no significant changes in associated weight [15 , 28–30]. This suggests that there are barriers and challenges with nutrition counseling in the workplace. Long-term behavior change in lifestyle habits requires more frequent engagement with RDNs to monitor progress and encourage further behavior change. If diet and lifestyle are at the root of this obesity epidemic, then integration of a more personalized RDN-led worksite wellness program may be the first step into prevention and management of chronic disease. There is sufficient evidence to support that an improvement in dietary and lifestyle behaviors has profound benefits on work productivity and healthcare costs to our systems [6, 7].

4.1 Strengths and limitations

This study had several areas of strengths. This is a large data set from companies across the United States, providing generalizability to the results in the worksite setting. This study is one of the first of its kind to document the impact of multiple site, RDN-led worksite wellness in its relationship to behavior change. There are several study limitations that need to be addressed. First, structured sessions with specific outcomes was not a focus during these counseling sessions, rather it was personalized nutrition and lifestyle counseling. As this may result in individual benefits, it is difficult to quantify the group benefits because each person may seek different behavior changes. Employees who participated in this program were self-selected and have a similar profile (female, Caucasian), thus limiting the generalizability of the results to diverse, multicultural populations. Engagement in more than one session was optional by the employees, suggesting there may be some participation bias due to motivation. Lastly, the measures that were used in this study were self-reported data and thus may not accurately reflect true dietary and lifestyle behaviors.

Future research needs to measure individual impacts and disease-specific impacts of RDN-led medical nutrition therapy in the workplace setting, rather than grouping all participants together. In addition, measuring corresponding behavior change impacts on medical care costs and medical claims per individual. These future studies may yield additional information on how worksite wellness programs can be improved and implemented within more companies.

Conflict of interest

KH: Previous employment with Family Food LLC as a corporate wellness RDN. Family Food LLC has paid this author to run statistical analyses on individual company data and provided funding to present research results at conferences. MF: Employment with Nutrimedy, which provides RDN-led technology-based corporate wellness to employers. Nutrimedy has not contributed any money to this study.

Informed consent

This study was deemed exempt by the Texas A&M University Institutional Review Board. This study was deemed exempt under federal regulation 45 46.101 (IRB2016-0674).

References

, Fleming

, Robinson

, Thomson

, Graetz

, Margono

, et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 1980-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet (London, England) [Internet]. 2014 Aug 30 [cited 2019 May 15]; 384(9945):766–81.Available from: http://www.ncbi.nlm.nih.gov/pubmed/24880830.

Ogden

, Carroll

, Kit

, Flegal

. Prevalence of Childhood and Adult Obesity in the United States, 2011-2012. JAMA [Internet]. 2014 Feb 26 [cited 2019 May 15];311(8):806. Available from: http://jama.jamanetwork.com/article.aspx?doi=10.1001/jama.2014.732

Finkelstein

, Fiebelkorn

, Wang

. The Costs of Obesity among Full-Time Employees. Am J Heal Promot [Internet]. 2005 Sep 25 [cited 2019 May 15];20(1):45–51. Available from: https://journals-sagepub-com.web.bisu.edu.cn/doi/10.4278/0890-1171-20.1.45

Cawley

, Rizzo

, Haas

. Occupation-specific absenteeism costs associated with obesity and morbid obesity. J Occup Environ Med [Internet]. 2007 Dec 1 [cited 2019 May 15];49(12):1317–24. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18231079

Wilson

, Kyle

, Nadglowski

, Stanford

. Obesity coverage gap: Consumers perceive low coverage for obesity treatments even when workplace wellness programs target BMI. Obesity [Internet]. 2017 Feb 1 [cited 2019 May 15];25(2):370–7. Available from: http://doi.wiley.com/10.1002/oby.21746

Kudel

, Huang

, Ganguly

. Impact of Obesity on Work Productivity in Different US Occupations. J Occup Environ Med. 2017;60(1):6–11.

Wang

, Beydoun

, Liang

, Caballero

, Kumanyika

. Will All Americans Become Overweight or Obese? Estimating the Progression and Cost of the US Obesity Epidemic. Obesity [Internet]. 2008 Oct 1 [cited 2019 May 15];16(10):2323–30. Available from: http://doi.wiley.com/10.1038/oby.2008.351

Wang

, McDonald

, Bender

, Reffitt

, Miller

, Edington

. Association of healthcare costs with per unit body mass index increase. J Occup Environ Med. 2006;48(7):668–74.

Theodore Kyle

, Joseph Nadglowski

, Mechelle Claridy

, Fatima Cody Stanford. US Employee Wellness Programs and Access to Obesity Treatment in Employer Sponsored Health Insurance. Am J Manag Care [Internet]. 2015 [cited 2019 May 15];21(15):513–5. Available from: https://www.ajmc.com/journals/evidence-based-diabetes-management/2015/november-2015/us-employee-wellness-programs-and-access-to-obesity-treatment-in-employer-sponsored-health-insurance

10.

Lerner

, Rodday

, Cohen

, Rogers

. A systematic review of the evidence concerning the economic impact of employee-focused health promotion and wellness programs. J Occup Environ Med [Internet]. 2013 Feb 1 [cited 2019 May 15];55(2):209–22. Available from: http://www.ncbi.nlm.nih.gov/pubmed/23287723

11.

van Dongen

, Proper

, van Wier

, van der Beek

, Bongers

, van Mechelen

, et al. Systematic review on the financial return of worksite health promotion programmes aimed at improving nutrition and/or increasing physical activity. Obes Rev [Internet]. 2011 Dec 1 [cited 2019 May 15];12(12):1031–49. Available from: http://doi.wiley.com/10.1111/j.1467-789X.2011.00 25.x

12.

Merrill

, LeCheminant

. Medical cost analysis of a school district worksite wellness program. Prev Med Reports. 2016;3:159–65.

13.

Barleen

, Marzec

, Boerger

, Moloney

, Zimmerman

, Dobro

. Outcome-based and Participation-based Wellness Incentives: Impacts on Program Participation and Achievement of Health Improvement Targets. J Occup Environ Med. 2017;59(3):304–12.

14.

The Business Value of a Healthy Workforce. 2013.

15.

Ovbiosa-Akinbosoye

, Long

. Factors associated with long-term weight loss and weight maintenance: Analysis of a comprehensive workplace wellness program. J Occup Environ Med. 2011;53(11):1236–42.

16.

Radler

, Marcus

, Griehs

, Touger-Decker

. Improvements in Cardiometabolic Risk Factors Among Overweight and Obese Employees Participating in a University Worksite Wellness Program. Health Promot Pract. 2015;16(6):805–13.

17.

Van De Weert

, McIsaac

, Callas

, Nickerson

. Impact of Hospital Worksite Wellness Initiative on Fruit and Vegetable Consumption. J Acad Nutr Diet. 2016;116(9):A21.

18.

Kanauss

, Shupe

. Academia and Industry Collaboration: A Nutrition-Related Worksite Wellness Program. Int J Heal Wellness, Soc. 2016;6(3):1–7.

19.

Mincher

, Leson

. Worksite Wellness: An Ideal Career Option for Nutrition and Dietetics Practitioners. J Acad Nutr Diet. 2014;114(12):1895–901.

20.

Andersen

, Baird

, Bates

, Chapel

, Cline

, Ganesh

, et al. Academy of Nutrition and Dietetics: Revised Scope of Practice for the Nutrition and Dietetics Technician, Registered. J Acad Nutr Diet. 2018;118(2):327–42.

21.

Briggs Early

, Stanley

. Position of the Academy of Nutrition and Dietetics: The Role of Medical Nutrition Therapy and Registered Dietitian Nutritionists in the Prevention and Treatment of Prediabetes and Type 2 Diabetes. J Acad Nutr Diet [Internet]. 2018 Feb 1 [cited 2019 May 15];118(2):343–53. Available from: http://www.ncbi.nlm.nih.gov/pubmed/29389511

22.

Plodkowski

, St. Jeor

. Medical nutrition therapy for the treatment of obesity. Endocrinol Metab Clin North Am [Internet]. 2003 Dec 1 [cited 2019 May 15];32(4):935–65. Available from: https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/pii/S088985290300077X?via%3Dihub

23.

Hoelscher

, Kirk

, Ritchie

, Cunningham-Sabo

. Position of the Academy of Nutrition and Dietetics: Interventions for the Prevention and Treatment of Pediatric Overweight and Obesity. J Acad Nutr Diet [Internet]. 2013 Oct 1 [cited 2019 May 15];113(10):1375–94. Available from: https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/pii/S2212267213012926?via%3Dihub

24.

Raudenbush

, Byrk

. Hierarchial Linear Models. Thousand Oaks, CA: SAGE Publications; 2002.

25.

US Department of Health and Human Services, US Department of Agriculture. 2015-2020 Dietary Guide- lines for Americans [Internet]. 2015. Available from: http://health.gov/dietaryguidelines/2015/guidelines/.

26.

Yarborough

, Brethauer

, Burton

, Fabius

, Hymel

, Kothari

, et al. Obesity in the Workplace. J Occup Environ Med [Internet]. 2018;60(1):97–107. Available from: http://insights.ovid.com/crossref?an=00043764-201801000-00015

27.

White

, Hartley

, Ozminkowski

. Association between corporate wellness program participation and changes in health risks. J Occup Environ Med. 2015;57(10):1119–26.

28.

Lang

, Cluff

, Payne

, Matson-Koffman

, Hampton

. The Centers for Disease Control and Prevention. J Occup Environ Med. 2017;59(7):631–41.

29.

Martinez

, Prada

. What are some of the features of worksite wellness interventions in the United States? J Occup Environ Med. 2015;57(3):e26–9.

30.

Macaluso

, Marcus

, Rigassio-Radler

, Byham-Gray

, Touger-Decker

. Increased physical activity leads to improved health-related quality of life among employees enrolled in a 12-week worksite wellness program. J Occup Environ Med. 2015;57(11):1214–21.

	Appointment-level
	Baseline^a		Last appointment		Overall^b
Variable	M (n)	SD (%)	M (n)	SD (%)	M (n)	SD (%)
Weight (lbs)	204.03	51.52	207.43	51.25	206.63	51.45
Systolic (mmHg)	130.05	17.93	129.17	18.38	129.43	18.28
Diastolic (mmHg)	84.58	12.56	84.7	12.09	84.44	12.33
BMI	33.48	7.83	34.28	7.91	34.04	7.90
Whole grain servings	1.65	1.09	1.69	1.07	1.66	1.06
Dairy servings	1.24	0.88	1.23	0.85	1.23	0.86
Combined fruit and vegetable servings	3.64	1.70	3.68	1.55	3.67	1.58
Physical activity intensity	1.22	0.48	1.14	0.40	1.17	0.42
Physical days/week	2.01	2.05	2.03	1.97	2.04	2.02
Waist:neck ratio	2.48	1.00	2.60	1.32	2.52	1.20
Patient-level
Variable	M (n)			SD (%)
Male gender	(230)			(20.5)
Apple body type	(725)			(64.6)
Diabetes	(100)			(8.9)
Number of appointments	2.20			1.45
Age	46.95			12.19