The prevalence of overweight and obesity among adults is increasing worldwide [1]. Therefore, effective treatment programs with a large reach are required. Web-based weight-loss programs have emerged in response to the exponential growth in Internet access, as well as increasing use of the Internet as part of daily life and improved accessibility. In the United States, 66% of households have access to broadband Internet [2], while 72% of Australian households have access to the Internet [3]. Furthermore, many adults (61% in the United States) use the Internet to access information about health, nutrition, physical activity, and weight loss [4]. The multimedia capabilities of the Internet allow up-to-date, interactive, and individualized lifestyle programs to be provided, which endeavor to emulate traditional face-to-face consultations [5]. These programs also overcome several barriers of attending face-to-face consultations, such as poor accessibility [6], lack of anonymity [7], and participant burden associated with attendance.

However, Web-based weight-loss programs are an underevaluated treatment medium. A recent systematic review and meta-analysis examined the efficacy of 12 randomized controlled trials (RCTs) of Web-based weight-loss programs [8]. The results suggest that, in general, participants in these programs achieve similar weight-loss outcomes to control or minimal intervention groups. In addition, it has been suggested that Web-based programs with enhanced features (eg, tailored information and counseling) achieve greater weight loss than those that focus on information alone, although these studies are highly heterogeneous [8]. Further studies are required, as it has not yet been possible to establish the overall efficacy of Web-based interventions or the superiority of those with more enhanced features, due to the heterogeneity of study designs and therefore the small number of comparable studies.

Of the small number of Web-based programs that have been evaluated using an RCT, remarkably few are available to the public. Commercial Web-based weight-loss programs are likely to be the most accessible to consumers [9] but have rarely been independently evaluated [10]. Of the two RCTs of eDiets, a commercial Web-based weight-loss program in the United States, one found that after 12 months those using eDiets lost significantly less weight than those following a self-help manual (–1.1% vs –4.0%) [11], while the second compared eDiets with a structured behavioral Web-based program [12] and found the behavioral program achieved significantly greater weight loss after 12 months (–2.8% vs –5.5%). Overall, limited evidence exists for the efficacy of commercial Web-based programs as a viable obesity treatment option. Therefore, examining the efficacy of commercially available Web-based weight-loss programs on weight-related outcomes is warranted to increase the treatment options for people seeking to engage in commercial treatment programs, especially those who have limited options in their region.

The primary aim of this study was to determine whether there was a significant difference in body mass index (BMI) posttreatment among participants randomly assigned to a standard (basic) 12-week commercial Web-based weight-loss program versus a version of this program with additional online features and personalized e-feedback and reminder calls (enhanced) versus a wait-list control (control). We hypothesized that the reduction in BMI would be greater in the basic and enhanced groups than in the control group, with the BMI reduction greater in the enhanced than in the basic group.

The present study was an assessor-blinded RCT with a 12-week follow-up. The methods have been published in detail [13]. Briefly, overweight and obese adults were recruited offline and enrolled by research assistants at the University of Newcastle from October to December 2009 from the Hunter community in New South Wales, Australia. Eligibility criteria were age 18 to 60 years, BMI 25 to 40 kg/m², not participating in other weight-loss programs, passing a health screen [14], being available for in-person assessments, and having access to a computer with email and Internet services, although neither computer nor eHealth literacy was assessed. Exclusion criteria were pregnancy or trying to conceive, major medical problems, orthopedic problems, recent weight loss of ≥4.5 kg, and medications affecting or affected by weight loss. Trial sample size calculations have been previously published [13].

The primary aim of this study was to determine whether using a commercial Web-based weight-loss program, with or without enhanced online features including personalized e-feedback and reminder calls, would lead to a greater reduction in BMI compared with each other or with a wait-list control group. This study demonstrated that participation in either version of the 12-week commercial Web-based weight-loss program (standard or enhanced with feedback) resulted in statistically significant and clinically important objectively measured weight loss. Many desirable improvements in secondary risk markers for chronic disease were achieved, irrespective of the program version used. This suggests that the fundamental elements of the Web-based program were the drivers of the behavior change. However, the enhanced program version achieved a greater retention rate, which in the longer term is critical for maximizing program reach and opportunity to facilitate maintenance of lost weight. There were advantages for those in the enhanced program group related to the extent of improvements in several secondary outcomes compared with the controls, including decreases in waist circumference, plasma total cholesterol, blood pressure, energy intake, and steps per day. Overall, both program versions provided important reductions in several risk factors for cardiovascular disease, as well as improvements in domains of quality of life and eating behavior.

In the only previous RCT of a commercial Web-based weight-loss program [11,12], the commercial Web-based program was found to be less effective than a self-help manual [11] and a more structured Web-based behavioral weight-loss program [12]. The eDiets commercial Web-based program had some similar program features to The Biggest Loser Club program, such as weekly self-monitoring of weight, meal and exercise plans, educational materials, and social support via a discussion board. Although eDiets also offered alternative sources of social support such as Web-based meetings and chat rooms, it did not provide a tool for participants to monitor and receive feedback on their dietary intake and physical activity, nor were participants reminded to use the program, which differs from the program in the current study. Therefore, we hypothesized that the mean weight losses for the basic and enhanced versions after 12 weeks in this study would be greater than those from the 2004 RCT using eDiets (–0.9% after 16 weeks) [11]. Interestingly, the 12-week weight loss in the current study is similar to the results of a more recent RCT using eDiets, which reported a mean weight loss of –3.6% after 6 months [12], which may indicate that developments in online capabilities facilitate program effectiveness.

This is the first RCT to examine the efficacy of a commercial program that specifically sought to include enhanced features (provision of additional personalized e-feedback, reminders, and phone calls) compared with the basic commercial program version without these features. We have shown that participants using the enhanced program did not lose significantly more weight after 12 weeks than did those in the basic program. This finding is not consistent with previous RCTs comparing basic versus enhanced versions of Web-based programs, as these have typically demonstrated greater weight loss in the enhanced study arm [12,23-27]. However, the basic programs in these previous studies had fewer program features and typically did not include all of the key components of Web-based interventions reported to be essential, such as being based on a theoretical framework, providing diet and physical activity feedback to participants, and having interactive features [28,29]. For example, Rothert et al’s basic Web-based program provided information only [27], while the enhanced features in other studies commonly used more human interventions, such as individualized human e-feedback generated by a therapist [24-26] or Web-based chat sessions conducted by a trained therapist [12]. The lack of difference between the basic and enhanced program in the current study, along with the magnitude of the weight lost after 12 weeks, suggests that either format of the program can produce clinically important initial weight loss. However, it also suggests that the enhanced program may require a greater number or intensity of enhanced features to be more effective than the basic version, or that the enhanced features provided are not necessary for many participants.

Although no significant difference in weight loss between basic and enhanced groups was observed, the enhanced group achieved significantly greater improvements in waist circumference than the basic group. They also demonstrated greater improvements in blood pressure, plasma total cholesterol, steps per day, measures of general health, and reduction in total daily energy intake than the control group, whereas the basic group did not. This suggests that the enhanced program offers additional benefit to participants in terms of risk factor reduction and in achieving behavior change that may assist with long-term maintenance of lost weight. Furthermore, the basic group were more likely to drop out of the study (25%) than the enhanced group (15%). The impact of this should not be overlooked because weight-loss success is associated with greater adherence to the prescribed treatment plan, and retention within treatment is the primary component of program adherence. Further follow-up of these participants will determine whether greater initial improvements in indicators of health status, quality of life, dietary intake, and physical activity, as well as higher initial retention, affect weight-loss outcomes in the long term and whether improvements are sustained over time. This will be important in evaluating the long-term efficacy of the basic program compared with the enhanced program.

This is one of the first RCTs of a Web-based weight-loss program to comprehensively assess secondary outcomes and to capture the potential of the program to have significant health benefits, irrespective of the weight loss achieved. To our knowledge, no other RCT evaluating Web-based weight-loss studies have assessed changes in lipids or insulin levels, and only one has evaluated changes in glucose levels [26]. The current study also demonstrated significantly greater reductions in blood pressure (systolic and diastolic) among the enhanced program users. Only two other RCTs evaluating Web-based weight-loss programs have assessed changes in blood pressure [30-32], with one of these finding greater reductions in systolic blood pressure among male participants only. Also, these improvements were demonstrated after 12 months [31], but not after 3 or 6 months [32]. Therefore, this is the first Web-based weight-loss RCT to demonstrate significant improvements in plasma total cholesterol and blood pressure after 12 weeks via participation in a Web-based weight-loss program that provides comprehensive personalized feedback and reminders to engage with the program.

Only a few Web-based weight-loss RCTs have reported changes in dietary intake and physical activity, and assessed difference in change across study arms [23-26,32-35]. To date, no study has demonstrated significant differences in dietary intake or physical activity change among participants of Web-based weight-loss programs compared with control groups, other treatment formats (eg, face-to-face), or enhanced Web-based programs. The current study found significantly greater reductions in mean total daily energy intake and increases in steps per day in the enhanced program users, which suggests that additional personalized feedback and reminders to use the program had positive influences on food and physical activity behaviors. However, further detailed investigation is needed to examine how participants change their food and physical activity behaviors in response to Web-based intervention. Further, no previous Web-based weight-loss studies have measured components of appetite and hunger using the TFEQ-R18. We have demonstrated improvements in both intervention groups compared with a control group in the domains of cognitive restraint and uncontrolled eating, but no significant improvements in emotional eating scores. Previous research has highlighted the association between appropriate eating patterns, such as higher dietary restraint and less emotional eating, and long-term weight-loss maintenance [36], and therefore this is an area where the current program could be refined. Longer-term follow-up of participants will demonstrate whether these initial improvements in eating patterns can be sustained or improved, and whether this affects weight-loss maintenance.

One other Web-based weight-loss study has measured quality of life. McConnon and colleagues found significant improvements in quality of life in Internet and usual-care groups with no significant difference between groups [33]. Therefore, the finding in the current study of significant improvements in quality of life, namely the general health domain of the SF-36, after 12 weeks’ participation in the enhanced version of the program is important.

A limitation of the current study is that all participants did receive human contact when they attended the clinical research center for clinical assessments. However, they were assessed by blinded assessors. Further, we gave them no advice on how to log in or engage with the program, other than giving them their login details. This was to simulate real-world engagement and use of the program, making the results generalizable to the overweight population of adults using commercial weight-loss programs. Due to the inclusion of a control group, the study also did not consider the potential differential impact of the Web-based programs as a result of participants’ varying levels of website usage. Study strengths include the use of a control group compared with two versions of the Web-based program, as well as the robust study design and the use of predominantly objective measures.

The results of this study demonstrate that Web-based weight loss can be efficacious across a range of weight-related outcomes and lifestyle behaviors, and that commercial providers can deliver effective Web-based programs targeting this important public health issue. Further study is needed to examine longer-term outcomes and whether Web-based programs with enhanced program features, including provision of personalized feedback, can retain people in the long term and lead to a greater treatment impact over time.