Using Telehealth to Increase Participation in ... - Wiley Online Library

Using Telehealth to Increase Participation in Weight Maintenance Programs Heather A. Haugen,* Zung V. Tran,† Holly R. Wyatt,* Mary J. Barry,* and James O. Hill*

Abstract HAUGEN, HEATHER A., ZUNG V. TRAN, HOLLY R. WYATT, MARY J. BARRY, AND JAMES O. HILL. Using telehealth to increase participation in weight maintenance programs. Obesity. 2007;15:3067–3077. Objective: To compare weight regain, satisfaction, and convenience among three weight maintenance programs: telehealth, traditional classes, and no program. Research Methods and Procedures: This quasi-experimental study compared weight change, satisfaction, and convenience among three program types. The telehealth participants interacted with a registered dietitian (RD) through the web and e-mail, traditional program participants attended a traditional classroom program, and no program participants received no interaction. Eighty-seven subjects (14 men and 73 women) were enrolled in the study: 31 traditional, 31 telehealth, and 25 no program participants. Eligibility included participation in a community-based weight loss program (Colorado Weigh) and minimum 7% weight loss before enrollment. Results: Subject characteristics at baseline were as follows: age, 50 ⫾ 9.3 (standard deviation) years; height, 1.68 ⫾ 0.09 m; weight, 80.5 ⫾ 18.4 kg, with no significant differences between groups. Over 6 months, the traditional group lost 0.5 ⫾ 4.3 kg, the telehealth group lost 0.6 ⫾ 2.5 kg, and the no program group gained 1.7 ⫾ 3.0 kg. Weight change among all three groups was significant (p ⫽ 0.02); no program participants gained significantly more weight than the telehealth and traditional groups. There were no differences in overall satisfaction between the telehealth and traditional groups (p ⫽ 0.43), but individuals in the tele-

Received for review December 7, 2006. Accepted in final form May 1, 2007. The costs of publication of this article were defrayed, in part, by the payment of page charges. This article must, therefore, be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. *Center for Human Nutrition and †Preventive Medicine and Biostatistics, University of Colorado Health Sciences Center, Denver, Colorado. Address correspondence to Heather Haugen, 570 Chadwick Circle, Highlands Ranch, CO 80129. E-mail: [email protected] Copyright © 2007 NAASO

health group rated their program as more convenient compared with the traditional group (p ⫽ 0.0001). Discussion: These results show the usefulness of telehealth programs in long-term weight loss maintenance. They may be a useful alternative for those who successfully lose weight in a structured behavioral program but do not choose to participate in a formal behavioral weight loss maintenance program. Key words: weight management program, weight regain, internet, monitoring

Introduction Programs that are effective in producing weight loss are not necessarily successful in producing long-term maintenance of weight loss (1). Many people who succeed in losing weight have trouble achieving a pattern of diet and physical activity that prevents weight regain (1). Studies have shown that many individuals who complete a comprehensive behavioral weight loss program can lose 6% to 10% of their initial weight (2– 4), but most regain the lost weight within 3 to 5 years (5,6). Helping individuals achieve weight loss is only one component of successful weight management. It is also important to develop better ways to provide support for individuals in keeping the weight off long term. Research has shown that participation in a behavioral support program increases success in weight loss maintenance (7–9), but most weight management programs focus almost exclusively on weight loss and the behaviors associated with losing weight (1). These programs may not be optimum, because losing weight and keeping weight off can be considered as different processes and may require different types of behaviors and behavioral support (1). Few programs offer weight maintenance support, and of the programs offered, most are patterned after successful weight loss programs, using the same level of professional resources and frequent in-person contacts. This requires a continuous high level of commitment in terms of time and resources for participants, and many people who participate OBESITY Vol. 15 No. 12 December 2007

3067

Using Telehealth for Weight Maintenance, Haugen et al.

in formal behavioral weight loss programs opt not to continue in behavioral programs even when they are available. Thus, the dilemma is that, whereas individuals need continued behavioral support for weight loss maintenance, few actually choose to commit long term to group-based programs. Because weight loss maintenance is a very long-term process, there is a need for support programs that do not require the intensive time and resource commitment of behavioral support groups (1). E-mail, computer software, and remote counseling/coaching can be used to deliver and reinforce the knowledge and behavioral skills acquired during weight loss for successful weight maintenance (10 –12). The use of these technologies allows professionals to provide an individualized approach in a cost-effective (13,14) manner and allows nutrition professionals greater reach to a larger target audience.

Research Methods and Procedures Study Design The intent of this quasi-experimental study was to examine whether a telehealth program could be helpful in weight loss maintenance for individuals who succeed in losing weight in a behavioral weight loss program but chose not to participate in a behavioral weight loss maintenance program. Three groups of successful weight losers who lost weight in the same behavioral weight loss program, Colorado Weigh, were studied. Group 1 consisted of individuals who opted to enroll in the Colorado Weigh weight loss maintenance program (called the Colorado Weigh Graduate Program). Group 2 consisted of individuals who chose not to participate in the Colorado Weigh Graduate Program but who agreed to participate in the telehealth program. Group 3 consisted of individuals who chose not to participate in any weight loss maintenance program. The specific aims of this study were to examine differences in weight regain differences in satisfaction and convenience of the program among all three groups. We hypothesized that Group 2 would regain significantly less weight compared with Group 3. Additionally, we compared the weight regain between Groups 1 and 2. Study Population Subjects were men and women (⬎18 years of age) recruited from individuals who completed a 24-week, commercial, behavioral weight loss program, Colorado Weigh. To be eligible, individuals must have lost at least 7% of initial body weight in the program. All subjects were given the opportunity to participate in the Colorado Weigh Graduate Program before receiving information about this study. The traditional program consisted of subjects who chose to enroll in the Colorado Weigh Graduate Program. Only those who elected not to participate in the traditional program 3068

OBESITY Vol. 15 No. 12 December 2007

were given the opportunity to participate in the telehealth program. Those who elected not to participate in either program were recruited into the no program group. This quasi-experimental design, in contrast to a randomized design, was chosen to address a specific question. Our program, like many other behavioral-based interventions, has been successful in helping individuals lose weight, but ⬃40% of our subjects do not participate in the weight maintenance portion of the program because of the barriers to attending classroom sessions. We believe the traditional classroom program is appropriate and effective for many of our participants, but we wished to determine the feasibility of offering an alternative choice to individuals who opted out of the traditional classroom program. A randomized trial would not have allowed us to ask the same question because it would have forced individuals who prefer the classroom into a telehealth program. We were not attempting to determine “the best” program, but rather to determine the effectiveness of a telehealth program for individuals who opt out of the traditional program. Additionally, the no program group served as a comparison group to help us understand the outcomes for participants who leave the classroom and have no alternative program. All participants provided informed written consent. The study protocol was approved by the Colorado Multiple Institutional Review Board. We projected that a final sample size of 21 subjects per group was needed to detect a clinically meaningful difference in weight change between groups (with 80% power). Thus, the goal was to recruit 30 subjects into each group (assuming a 30% dropout rate). There are limited data available from weight maintenance trials to determine mean weight regain and variability. Based on our knowledge of average weight lost during the initial phase, 4 lbs represented ⬃30% of the weight lost for someone who met the 7% criterion. We felt that regaining one third of the weight lost in 6 months was clinically significant from a health outcomes perspective. Therefore, we used a clinically significant difference in weight gain of 4 ⫾ 4.5 (standard deviation) lbs. Colorado Weigh Colorado Weigh is a research-based weight loss program developed by researchers at the Center for Human Nutrition at the University of Colorado School of Medicine. The behavioral curriculum was developed based on information obtained from the National Weight Control Registry (7) and results from the Diabetes Prevention Program (9). The Colorado Weigh program is offered as a fee-for-service program within the metropolitan area of Denver, CO, and currently is not associated with the University of Colorado Health Sciences Center. Colorado Weigh consisted of two 12-week phases of weight loss and the option to enroll in a weight maintenance program (the Colorado Weigh Graduate Program). Partici-


Table 1. Framework for program development: telehealth program Factors influencing weight maintenance Professional support/feedback Transfer of ownership from professional to individual Dietary and activity monitoring Convenient Individualized Limited professional resources/time

How it can be accomplished through telehealth View food/activity logs remotely, feedback through web/e-mail/phone, goal setting and review Software program feedback, individual initiates interaction, “on demand” support Computer software aided, graphs and reports to chart progress/status Participate where and when it works for the individual Professional can customize feedback based on logs/individuals goals Professional can provide feedback anytime and from anywhere

pants met weekly in groups of 15 to 25 people. Each group was led by a registered dietitian with expertise in nutrition, exercise, and behavioral change techniques. The curriculum (created by a team of health professionals) emphasized the behavioral principles of self-monitoring, goal setting, behavioral contracting, time management, prompting and cueing, problem solving, cognitive restructuring, stress management, behavioral chains, and relapse prevention. Individuals in the Colorado Weigh Graduate Program received the standard Colorado Weigh weight maintenance program. The program was aimed at providing help for participants in keeping weight off and in maintaining the behaviors required for success. Participants attended class every other week for 24 weeks and continued to track their calories, fat grams, and activity using paper logs. A registered dietitian led the group classes and offered support and

counseling. Class topics offered over the 24 weeks focused on nutrition, physical activity, and behavior modification, but were oriented to weight loss maintenance rather that weight loss. Dietary Monitoring. Participants tracked their dietary intake using paper logs. A quick reference book was used to record the calories and fat in the foods eaten (15). Individuals were instructed to log daily over the 24 weeks of the study. Dietary Fat Goal. Participants were given a dietary fat goal for weight loss maintenance. They used the reference book (cited under dietary monitoring) to monitor their fat intake throughout the day. Physical Activity Monitoring. Participants were given an individual physical activity goal in steps per day. Each individual’s daily step goal was developed based on an

Figure 1: BalanceLog screen shot of food log.


3069


estimation of the decline in energy requirements that occurred with weight loss. Participants recorded steps per day. Body Weight Monitoring. Participants were weighed by the registered dietitian at each class they attended. Group Leaders. Group leaders were registered dietitians who presented an overview of the week’s topic, led the group discussion, and provided support and feedback to participants based on reported food intake and physical activity during the previous 2 weeks. The group leaders were available before and after class to answer individual questions from the participants. Telehealth Program The Colorado Weigh Graduate Program materials were developed to be delivered in group classes and were not appropriate in their original form for a program delivered remotely. Therefore, the content of the Colorado Weigh Graduate Program materials was modified into a format that was more appropriate for the telehealth program (Table 1). Participants in the telehealth program (called Colorado Weigh High-Tech) received a weight maintenance program using telehealth. The content and scientific basis for the telehealth program was the same as the traditional program; the difference between the two programs was the format and delivery. Participants interacted on-line with the registered dietitian, “Healthy Coach,” every other week for 24 weeks. Individuals used computer software, BalanceLog (HealtheTech, Golden, CO), to record their dietary intake and activity and complete a “Health Report” every other week. The Healthy Coach had access to all of the participants’ logs, and contacted each participant every other week. Using the Health Report and dietary/exercise logs, the Healthy Coach focused on each individual’s goals to provide personalized feedback and encouragement. The Healthy Coach also offered helpful hints on using the computer software to maximize the use of the tools for feedback and decisionmaking. The majority of interaction (⬃95%) was through the computer software, web, and e-mail. Telephone correspondence was used at the discretion of the Healthy Coach. Software Program Description. The BalanceLog software is a comprehensive tool for weight management including goal setting, self-monitoring, and reports/feedback. The database contains ⬎4000 foods and 300 exercises to facilitate daily logging of dietary intake and activity. A screen shot of the BalanceLog food logging screen appears in Figure 1. Simple graphics and summary reports allow the user to view their actual intake vs. their goal intake (calories, macronutrients, and food label nutrients are displayed) and their overall energy balance (energy intake vs. energy expenditure). Dietary Monitoring. Participants tracked their dietary intake using BalanceLog. The software program provided realtime feedback on caloric intake, comparison of dietary goals to intake, and easy-to-understand summary reports. Individuals were instructed to log daily over the 24 weeks of the study. Dietary Fat Goal. Participants were given a dietary fat goal. The BalanceLog software was used to track dietary fat 3070


throughout the day. Each time a food was viewed or logged the dietary fat value was displayed, allowing the individual to choose foods that were compatible with their goals. Physical Activity Monitoring. Individuals in the telehealth group were encouraged to exercise daily. They were asked to log their activity in the BalanceLog program. The activity, intensity, and duration were logged from a comprehensive database of exercises. Throughout the day, they could view their energy balance: calories expended vs. calories consumed. Body Weight Monitoring. Participants self-monitored their body weight every other week. They recorded their body weight in the BalanceLog program for tracking and charting. Healthy Coach. Participants completed the Healthy Report every other week. The Healthy Report requires the individual to review specific information in their BalanceLog reports: calorie intake compared with goal, activity compared with goal, body weight, challenges incurred that week, areas where they could use support/feedback, and goals for the following week. The Healthy Coach used the BalanceLog reports and Healthy Report to provide individualized support and feedback. The Healthy Coach also reminded the individual to review the topic of the week materials in their Colorado Weigh notebook (information that was covered in the traditional program group classes). Participants had access to their Healthy Coach when they needed it; response time was usually within 24 hours. The Healthy Coach sent reminders to individuals who failed to send in weekly reports. If a participant did not respond within 1 week, the Healthy Coach would call the individual. No Program Participants in the no program group were followed for 6 months but received no materials or support. Individuals were asked to return to the clinic at 3 and 6 months for a follow-up assessment. The assessment included measurement of body weight and completion of the survey. Participants were paid for their time at each clinic visit. Outcome Measures Body Weight. Body weight for all groups was measured at baseline and again at 3 and 6 months by study personnel using a Tanita digital scale (TBF-300A; Tanita Corp., Arlington Heights, IL) under fasted conditions. All participants in all programs were measured on the same scale. Weight included subjects wearing light indoor clothing without shoes. Height. Height was measured to the nearest 0.25 in without shoes using a standard stadiometer. Program Satisfaction A survey instrument was developed to assess convenience and satisfaction with the programs and to begin identifying the barriers to weight maintenance program participation. Several resources were used to develop the instrument (16,17). All participants in all groups were sampled. The instrument was self-administered, but it was completed in the


Table 2. Subject anthropometrics at baseline

Age (yrs) Height (m) Weight (kg) Percent weight loss† Absolute weight loss (kg)‡

All participants (n ⴝ 87)

Traditional program (n ⴝ 31)

Telehealth program (n ⴝ 31)

No program (n ⴝ 25)

p*

50.5 (9.3) 1.68 (0.09) 80.5 (18.4) 13.3 (5.2) 12.6 (6.6)

50.0 (7.8) 1.70 (0.07) 85.5 (20.6) 12.9 (4.8) 12.7 (6.5)

49.5 (9.6) 1.67 (0.10) 80.1 (19.1) 13.0 (5.0) 12.5 (6.9)

52.6 (10.7) 1.65 (0.08) 74.9 (12.8) 14.0 (6.1) 12.6 (6.5)

0.10 0.42 0.09 0.99 0.99

Values are means (standard deviation). * Represents the p value comparing all three groups by one-way ANOVA. † Percent weight loss before the weight maintenance phase (Phase I, Phase II). ‡ Absolute weight loss before the weight maintenance phase (Phase I, Phase II).

clinic, and a study coordinator was available to answer questions. The majority of questions were closed-ended, but a few were open-ended questions to study additional responses. The instrument was piloted in a small sample of study participants and volunteers (n ⫽ 8). All participants were asked to complete the instrument. Statistical Analysis Plan The primary aim of the study was to compare weight regain between individuals in the telehealth program, the Colorado Weigh Graduate Program, and no program. Repeated-measures ANOVA was used to compare weights at baseline and 6 months among all groups. Additionally, 95% confidence intervals were calculated for weight regain for each group. The secondary aim of the study was to assess the benefits and barriers of the traditional and telehealth programs by examining the differences in satisfaction with level of support and convenience of participation. The combined convenience score of the traditional group and telehealth group was compared using a one-way ANOVA. The combined satisfaction score of the traditional group and telehealth group was compared using a one-way ANOVA. 95% confidence intervals were also calculated. All data analyses were performed using the Statistical Package for the Social Sciences (version 14.0; SPSS, Inc., Chicago, IL). Missing Data In obesity trials, statistical evaluation of a treatment’s efficacy is usually complicated by missing observations because of dropouts (i.e., subjects who drop out of the clinical trial after some interim follow-up visit and do not return) or by missing observations because of subjects who miss one or more visits (even though they complete the trial) (18). Several techniques have been identified for dealing with missing data in weight management trials. For example, a completers-analysis includes only individuals who complete the study, a last-obser-

vation-carried-forward analysis (LOCF)1 includes all individuals enrolled in the study (the last recorded weight is simply carried forward to 6 months), or using a variety of strategies to impute values (19 –21). The results presented here were analyzed using a LOCF analysis for specific aims related to weight regain for the following reasons: LOCF produced slightly more conservative results than the completers analysis, lack of data from other weight maintenance trials to estimate regain factors, small sample size of each group, small number of dropouts in all groups, and lack of information about why subjects either missed weigh-in appointments or dropped out.

Results Subject Characteristics/Demographics Eighty-seven subjects (14 men and 73 women) were enrolled in the study: 31 in the traditional program, 31 in the telehealth program, and 25 in the no program group. At baseline, the subject characteristics were similar among the groups. No significant differences were found for age, height, weight, or percent of weight loss between groups (Table 2). As a group, the subjects represented a wide range of ages (range, 24 to 73 years), body weights (range, 51 to 150 kg), and BMI (range, 21 to 49 kg/m2). The group was predominantly women (84% women and 16% men). Based only on responders to the surveys, the group was highly educated (87% completing some education beyond high school) and not ethically diverse (93% white; Table 3). Of the participants who completed the survey, a high percentage of the subjects in all three groups had access to the internet at home (91% traditional, 100% telehealth, 91% no program) and access to a computer at home (100% traditional, 100% telehealth, 91% no program). The majority of subjects in

1

Nonstandard abbreviation: LOCF, last-observation-carried-forward.


3071


Table 3. Subject demographics

Sex Male Female Ethnicity White Hispanic Black Asian No response Education (highest completed) ⬍High school GED High school College Postgraduate No response Income (yearly household) ⬍$25,000 $25,000 to 50,000 $50,000 to 75,000 $75,000 to 100,000 ⬎$100,000 No response

All participants (n ⴝ 87)




14 (16%) 73 (84%)

5 (16%) 26 (84%)

6 (19%) 25 (81%)

3 (12%) 22 (88%)

66 (76%) 3 (4%) 1 (1%) 1 (1%) 16 (18%)

23 (74%) 0 0 0 8 (26%)

23 (74%) 1 (3%) 1 (3%) 1 (3%) 5 (17%)

20 (80%) 2 (8%) 0 0 3 (12%)

0 0 9 (10%) 32 (37%) 28 (32%) 18 (21%)

0 0 2 (7%) 7 (23%) 11 (35%) 11 (35%)

0 0 4 (13%) 12 (39%) 10 (32%) 5 (16%)

0 0 3 (12%) 13 (52%) 7 (28%) 2 (8%)

6 (7%) 6 (7%) 18 (21%) 15 (17%) 17 (20%) 25 (28%)

1 (3%) 1 (3%) 4 (13%) 4 (13%) 8 (26%) 13 (42%)

1 (3%) 3 (10%) 9 (29%) 5 (16%) 4 (13%) 9 (29%)

4 (16%) 2 (8%) 5 (20%) 6 (24%) 5 (20%) 3 (12%)

p 0.24

0.44

0.57

0.48

Values represent total frequency of responses (percentage).

all groups rated their computer skills as excellent or good (87% of traditional, 77% of telehealth, 87% of no program). Access to the internet and computers and self-assessed ratings of computer skills were similar among all groups (Table 4). Subject Attrition The recruitment goal for the study was 90 subjects (30 in each group). Thirty-one subjects each enrolled in the traditional and telehealth programs; 25 subjects enrolled in the no program group. Seventy-five individuals completed the study (traditional ⫽ 24, telehealth ⫽ 27, and no program ⫽ 24). Attrition was 23% in the traditional group (seven dropouts), 13% in the telehealth group (four dropouts), and 4% in the no program group (one dropout; p ⫽ 0.14). Details of subject enrollment and attrition are shown in Table 6. Based on the original power calculations, the number of subjects who completed the study in each group exceeded the number required to detect the hypothesized differences. The subject characteristics of the dropouts were similar to the completers (Table 5). 3072


The characteristics of the dropouts by group are listed in Table 6. There do not seem to be any systematic differences between the total dropouts and dropouts in each group. Body Weight Change Over 6 months, the traditional group lost an additional 0.5 ⫾ 4.3 kg, the telehealth group lost an additional 0.6 ⫾ 2.5 kg, and the no program group gained 1.7 ⫾ 3.0 kg. Weight change among all three groups was significant (p ⫽ 0.02), with the no program group gaining more weight than the telehealth and traditional groups (Table 7). Baseline to 6-month weight regain for each group is shown in Figure 2. The mean difference in weight change between the telehealth and the no program groups was 2.3 kg (p ⫽ 0.003; 95% confidence interval ⫽ ⫺8.3 to ⫺1.9). Additionally, weight change among individuals in the telehealth program ranged from a weight loss of 8.6 kg to a weight gain of 3.3 kg, whereas the weight change among individuals in the no program group ranged from a weight loss of 3.5 kg to a weight gain of 8.3 kg. These ranges are shown in Figure 3.


Table 4. Technology: subject access, skills, and interest All participants (n ⴝ 87) Internet access home Yes No No response Computer access home Yes No No response Computer skills Excellent Good Fair Poor No response Interest in telehealth program Very interested Somewhat interested Not interested Need more info No response




p 0.30

68 (78%) 4 (5%) 15 (17%)

21 (68%) 2 (6%) 8 (26%)

26 (84%) 0 5 (16%)

21 (84%) 2 (8%) 2 (8%)

69 (79%) 2 (2%) 16 (17%)

23 (74%) 0 8 (26%)

26 (84%) 0 5 (16%)

20 (80%) 2 (8%) 3 (12%)

35 (40%) 25 (29%) 8 (9%) 4 (5%) 15 (17%)

10 (32%) 10 (32%) 3 (10%) 0 8 (26%)

11 (35%) 9 (29%) 4 (13%) 2 (7%) 5 (16%)

14 (56%) 6 (24%) 1 (4%) 2 (8%) 2 (8%)

0.15

0.49

0.61 19 (34%) 17 (30%) 12 (21%) 1 (2%) 7 (13%)

8 (26%) 8 (26%) 7 (22.5%) 1 (3%) 7 (22.5%)

N/A N/A N/A N/A N/A

11 (44%) 9 (36%) 5 (20%) 0 0

N/A, not applicable. Values represent total frequency of responses (percentage).

The mean difference in weight change between the telehealth and the Colorado Weigh Graduate Program was not significant (0.10 kg; p ⫽ 0.92; 95% confidence interval ⫽ ⫺1.7 to 1.9). Satisfaction and Convenience Individuals in the telehealth and traditional programs rated their satisfaction with the professional support and overall program. Twelve of the survey questions were re-

Table 5. Subject enrollment and attrition

Recruitment goal Enrollment Completers Attrition Percent attrition

Traditional program

Telehealth program

No program

30 31 24 7 23%

30 31 27 4 13%

30 25 24 1 4%

lated to satisfaction. Each question had four possible answers (no opinion, 1 ⫽ excellent, 2 ⫽ fair, 3 ⫽ poor). Therefore, the total possible satisfaction score was 12 to 36. No opinion was given no value and not counted toward the total score. A lower score indicates higher satisfaction. There was no significant difference in total satisfaction score between the telehealth and the Colorado Weigh Graduate Program groups. The satisfaction for the telehealth group was 15.3 ⫾ 3.6 (95% confidence interval ⫽ 13.8 to 16.7) and for the Colorado Weigh Graduate program was 16.1 ⫾ 3.4 (95% confidence interval ⫽ 14.6 to 17.6; p ⫽ 0.43). Mean scores by question and group are listed in Table 8. Individuals in the telehealth and traditional programs rated the convenience of participating in their respective program. Three of the survey questions were related to convenience of participation. Each question had four possible answers (no opinion, 1 ⫽ excellent, 2 ⫽ fair, 3 ⫽ poor). Therefore, the total possible score was 4 to 9. No opinion was given no value and not counted toward the total score. A lower score indicates a higher rating of convenience. Mean scores by question and group are listed in Table 9. OBESITY Vol. 15 No. 12 December 2007

3073


Table 6. Attrition: subject characteristics by group

Age (yrs) Height (m) Percent weight loss† Weight at baseline (kg) Weight at dropout (kg)

Total dropouts (n ⴝ 12)

Traditional dropouts (n ⴝ 7)

Telehealth dropouts (n ⴝ 4)

No program dropouts (n ⴝ 1)*

50.3 (10.1) 1.71 (0.06) 12.7 (4.4) 93.2 (24.6) 92.9 (24.2)

51.1 (10.3) 1.70 (0.06) 10.6 (2.4) 95.4 (18.9) 95.1 (17.7)

47.3 (11.6) 1.73 (0.06) 15.7 (6.0) 95.4 (36.1) 95.0 (36.1)

57.0 1.6 15.0 69.5 68.6

Values are means (standard deviation). * No standard deviation reported (n ⫽ 1). † Percent weight loss before weight maintenance (Phase I, Phase II).

The convenience score rating was significantly lower (better) for the telehealth group (3.3 ⫾ 0.8; 95% confidence interval ⫽ 3.0 to 3.7) vs. the Colorado Weigh Graduate Program (4.8 ⫾ 1.6; 95% confidence interval ⫽ 4.1 to 5.5; p ⫽ 0.0001).

Discussion The results suggest that a telehealth program can be effectively used to help with weight loss maintenance in some people. In this study, weight loss regain was prevented equally effectively with a formal, behavioral, group-based program and a telehealth program. Subjects in both groups were successful in avoiding weight regain, and actually lost a little more weight during weight maintenance. Those who chose not to participate in either program gained weight during the 6-month weight loss maintenance period.

Table 7. Weight change for traditional, telehealth, and no program Traditional program (n ⴝ 31) Baseline weight (kg) 6-month weight (kg) Weight change (kg)



85.5 (20.6)

80.1 (19.1) 74.9 (12.8)

85.0 (19.6)

79.5 (19)

⫺0.5 (4.3)

⫺0.6 (2.5)

76.5 (13.1)* ⫹1.7 (3.0)*

Values are means (standard deviation). * Significant at ⬍0.05, repeated-measures ANOVA with Tukey multiple comparison test to adjust p values.

3074


These results provide further reason to think that support of some type is essential for weight loss maintenance. In this study, all participants had achieved at least a 7% weight loss, and we know that such weight loss results in improved health and reductions in risk factors for diabetes and cardiovascular disease. In most individuals, a 5% to 10% weight loss results in improved health outcomes (2– 4). However, the no program group regained an average of 1.8 kg or 17% of their total weight loss. This weight regain likely diminished the health benefits of their initial weight loss. Furthermore, we know that without support, they are likely to continue to regain all or most of their lost weight (5,6). Thus, this study supports the idea that continued participation in a behavioral program increases success in weight loss maintenance (7–9). There is strong evidence from clinical trials that long-term multidisciplinary interventions are the most effective strategy for successful weight management (22,23). Our results suggest that a telehealth program could be more effective than doing nothing for participants who choose not to participate in a behavioral program, either because of time or resource barriers. The study results are limited in that this was not a randomized study and the participants selected their treatment to a large extent. However, this quasi-experimental study design with self-selection by the participants offers a “real-world” examination of a common problem with long-term participation in behavioral programs. Furthermore, this was a highly educated group with high levels of internet access. A total of 98% of study subjects reported having access to the internet from home. This compares with an average of 59% of all American adults. Thus, we cannot conclude that the telehealth option will be effective for all subjects. However, in this group of subjects, this intervention was very effective. Further work to more carefully identify the type of person who could benefit from this intervention is indicated. The content for the telehealth program was similar to that provided by the group-based, behavioral program, which


3

Weight change (kg)

2

1 Traditional Telehealth No Program

0

-1

-2

-3 Baseline

6 months

*Error bars represent the 95% confidence intervals.

Figure 2: Weight change by group (baseline to 6 months). Error bars represent 95% confidence intervals.

health program rated their satisfaction with the program materials higher. Except for these few differences, both groups rated their level of satisfaction similarly on most questions. In the future, the survey could be simplified by limiting the questions to overall satisfaction with program and overall satisfaction with program materials and still provide an adequate assessment of satisfaction. The inconvenience of attending classes is a common barrier to participation in weight loss programs (26). The telehealth program attempted to overcome the inconvenience associated with class times and locations. Based on all three questions

12

10

Frequency (# of individuals)

was unique in that it was a commercial program offering support for weight loss maintenance rather than weight loss. The difference was in mode of delivery. Given that all subjects had participated in 6 months of weekly groups for weight loss, it is not surprising that some people do not want to continue to participate in groups for weight loss maintenance. Those who chose to participate in 6 months of the Colorado Weigh Graduate Program were very successful in maintaining their weight loss. However, there is a limit to the length of time that people are willing/able to remain in face-to-face behavioral groups. Alternatively, telehealth programs require a less structured time commitment and could be used for longer periods of time. One challenge of a telehealth program is maintaining a high-quality program with adequate professional support and interaction through e-mail and telephone. There were no differences in the two groups when satisfaction scores of the various program components were compared. Other investigators have reported high satisfaction with telehealth programs focused on disease management. Wang et al. (24) used an internet-based system to improve compliance with chronic disease management and found the majority of patients (including the elderly) were satisfied with the telehealth applications. When telehealth programs are well designed and supported, ratings of satisfaction are similar to those found in face-to-face programs (25). Although the total satisfaction scores were similar, some of the questions seemed to differentiate the groups. The traditional program rated their satisfaction with assessments higher compared with the telehealth program, and the tele-

Traditional Telehealth No Program

8

6

4

2

0 > -9

- 8 to -6

-5 to -3

- 2 to 0

1 to 3

4 to 6

7 to 9

> 10

Weight change range (kg)

Figure 3: Weight change ranges by group.


3075


Table 8. Mean satisfaction scores by group Satisfaction questions

Mean score: traditional

Mean score: telehealth

Satisfaction with overall program materials Satisfaction with classes/weekly reviews Satisfaction with computer/paper logging Satisfaction with content covered in class/weekly reviews Satisfaction with assessments (calorie budget/goals) Satisfaction with instructor’s feedback Satisfaction with instructor’s communication of new information Satisfaction with access to instructor Satisfaction with instructor’s response time Satisfaction with program materials supporting goals Overall satisfaction with instructor Overall satisfaction with program Total

1.61 (0.66) 1.35 (0.49) 1.52 (0.79) 1.52 (0.59) 1.30 (0.34) 1.09 (0.29) 1.17 (0.39) 1.30 (0.47) 1.17 (0.39) 1.48 (0.59) 1.17 (0.39) 1.43 (0.59) 16.1 (3.4)

1.27 (0.53) 1.54 (0.65) 1.42 (0.70) 1.42 (0.71) 1.58 (0.70) 1.12 (0.32) 1.12 (0.33) 1.08 (0.27) 1.15 (0.46) 1.27 (0.72) 1.12 (0.33) 1.23 (0.43) 15.3 (3.5)

Values represent the mean (standard deviation).

related to convenience, the telehealth program participants found the program to be more convenient than the traditional participants. One of telehealth’s main strengths is its capacity to help make health care consumer-friendly and adapt to the needs of the individual, rather than demanding the individual adapt to the health care system (27,28). Increasing participation in weight management programs requires a more consumercentric approach. Making participation convenient is an important place to begin and telehealth allows individuals to participant when and where it is most convenient for them. This study did not attempt to compare the costs of the telehealth vs. the traditional program. Our only indication of cost comparison is limited to the professional resources and materials required to run each program. The time required to

run the telehealth program started higher than the traditional program but fell below the traditional program by the 6-month point. One of the benefits of the telehealth program is the personalized program that the professional creates using a software program at the first meeting. This is time consuming in comparison to the first traditional meeting, but allows the professional to do all future monitoring remotely. After the initial enrollment, the time required by the registered dietitian over the 6 months is similar between the two programs (⬃2.4 h/person for the telehealth program, ⬃2.5 h/person for the traditional program). A recent survey of dietitian salaries indicates a median hourly wage of $25. Additional materials cost $33 per person ($3 for notebook/educational materials, $30 for BalanceLog software) for the telehealth program and $18 per

Table 9. Mean convenience scores by group Convenience questions Convenience of class time Convenience of participating any time from home Convenience of class location Convenience of participating from home Overall convenience of participation in program Total Values represent the mean (standard deviation).

3076


Mean score: traditional

Mean score: telehealth

1.65 (0.78) 1.08 (0.27) 1.74 (0.69) 1.43 (0.51) 4.8 (1.6)

1.12 (0.37) 1.12 (0.33) 3.3 (0.8)


person ($3 notebook and educational materials, $8 calorie/fat counter book, $2 paper logs, $5 pedometer) for the traditional program. The total cost of the traditional program is approximately $80.50 per person and the total cost of the telehealth program is approximately $93, a difference of $12.50. In summary, these results showed the potential usefulness of telehealth programs in long-term weight loss maintenance. As a starting point, they might be a useful alternative for those who successfully lose weight in a structured behavioral program but do not have the availability of or do not choose to participate in a formal behavioral weight loss maintenance program. Without continued support, most people who lose weight regain it over time. Participating in a structured behavioral program, such as the Colorado Weigh Graduate Program, clearly helped in weight loss maintenance. However, the telehealth approach offers a more flexible approach to weight loss maintenance that in some people seems to be just as effective. By offering several different program formats for support in weight loss maintenance, we will likely improve long-term weight management.

Acknowledgments This project was supported by an R37 research grant from the NIH (DK042549). We thank Bonnie Jortberg, Susan Ellis, Jennifer Stuht, and Lisa Zucker for their dedication and expertise. References 1. Hill JO, Thompson H, Wyatt H. Weight maintenance: what’s missing? J Am Diet Assoc. 2005;105:63– 6. 2. Wadden TA, Bartlett SJ. Very low calorie diets: an overview and appraisal. In: Wadden TA, Vit B, eds. Treatment of the Seriously Obese Patient. New York: Guilford Press; 1992, pp. 44 –79. 3. The National Heart Lung and Blood Institute. The Practical Guide. Identification, Evaluation, and Treatment of Overweight and Obesity in Adults. Bethesda, MD: National Institutes of Health; 2000. 4. Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med. 2002;346:393– 403. 5. Wadden TA. Treatment of obesity by moderate and severe caloric restriction. Results of clinical research trials. Ann Intern Med. 1993;119:688 –93. 6. Kramer FM, Jeffery RW, Forster JL, Snell MK. Long-term follow-up of behavioral treatment for obesity: patterns of weight regain among men and women. Int J Obes Relat Metab Disord. 1989;13:123–36. 7. Wing RR, Hill JO. Successful weight loss maintenance. Annu Rev Nutr. 2001;21:323– 41. 8. Wing RR, Tate DF, Gorin AA, Raynor HA, Fava JL. A self-regulation program for maintenance of weight loss. N Engl J Med. 2006;355:1563–71.

9. The Diabetes Prevention Program. Description of lifestyle intervention. Diabetes Care. 2002;25:2165–71. 10. Bellazzi R, Arcelloni M, Ferrari P, et al. Management of patients with diabetes through information technology: tools for monitoring and control of the patients’ metabolic behavior. Diabetes Technol Ther. 2004;6:567–78. 11. Wantland DJ, Portillo CJ, Holzemer WL, Slaughter R, McGhee EM. The effectiveness of Web-based vs. non-Webbased interventions: a meta-analysis of behavioral change outcomes. J Med Internet Res. 2004;6:e40. 12. Tate DF, Jackvony EH, Wing RR. Effects of Internet behavioral counseling on weight loss in adults at risk for type 2 diabetes: a randomized trial. JAMA. 2003;289:1833– 6. 13. Noel HC, Vogel DC, Erdos JJ, Cornwall D, Levin F. Home telehealth reduces healthcare costs. Telemed J E Health. 2004; 10:170 – 83. 14. Baker L, Rideout J, Gertler P, Raube K. Effect of an Internetbased system for doctor-patient communication on health care spending. J Am Med Inform Assoc. 2005;12:530 – 6. 15. Borushek A. CalorieKing Calorie, Fat and Carbohydrate Counter. Costa Mesa, CA: Family Health Publications; 2006. 16. Dillman DA. Mail and Internet Surveys: The Tailored Design Method. New York, NY: John Wiley & Sons; 2000. 17. Pascoe GC, Attkisson CC. The evaluation ranking scale: a new methodology for assessing satisfaction. Eval Program Plann. 1983;6:335– 47. 18. Howell DC. Treatment of Missing Data. www.wm.edu/ ~dhowell/statpages/more_stuff/missing_data/missing.html (Accessed January 2005). 19. Gadbury GL, Coffey CS, Allison DB. Modern statistical methods for handling missing repeated measurements in obesity trial data: beyond LOCF. Obes Rev. 2003;4:175– 84. 20. Jain AK, Kaplan RA, Gadde KM, et al. Bupropion SR vs. placebo for weight loss in obese patients with depressive symptoms. Obes Res. 2002;10:1049 –56. 21. Smith IG, Goulder MA. Randomized placebo-controlled trial of long-term treatment with sibutramine in mild to moderate obesity. J Fam Pract. 2001;50:505–12. 22. National Institutes of Health. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults—the evidence report. Obes Res. 1998; 6(Suppl 2):51S–209S. 23. Klem ML. Successful losers. The habits of individuals who have maintained long-term weight loss. Minn Med. 2000;83:43–5. 24. Wang XH, Istepanian RS, Geake T, et al. A feasibility study of a personalized, internet-based compliance system for chronic disease management. Telemed J E Health. 2005;11:559 – 66. 25. Wager K, Lee F, Glaser J. Managing Health Care Information Systems. San Francisco, CA: Jossey-Bass; 2005. 26. Roux L, Ubach C, Donaldson C, Ryan M. Valuing the benefits of weight loss programs: an application of the discrete choice experiment. Obes Res. 2004;12:1342–51. 27. Goldsmith J. Digital Medicine: Implications for Healthcare Leaders. Chicago, IL: Health Administration Press; 2003. 28. Darkins A, Cary M. Telemedicine and Telehealth. New York: Springer Publishing Company; 2000.


3077