European Journal of Clinical Nutrition (2014) 68, 64–70 & 2014 Macmillan Publishers Limited All rights reserved 0954-3007/14 www.nature.com/ejcn
ORIGINAL ARTICLE
Changes to dietary intake during a 12-week commercial web-based weight loss program: a randomized controlled trial MJ Hutchesson1, CE Collins1, PJ Morgan2, JF Watson1, M Guest3 and R Callister4 BACKGROUND/OBJECTIVES: The primary aim of this secondary analysis was to compare changes in dietary intake among participants randomized to two versions of a 12-week commercial web-based weight loss program (basic or enhanced) with a waiting-list control. An additional investigation compared changes in dietary intake of successful participants (weight loss X5%) with those not successful. SUBJECTS/METHODS: Dietary intake was assessed at baseline and 12 weeks using a validated 120-item semiquantitative food frequency questionnaire. Adults (n ¼ 268, 60% female participants, body mass index 32.1±3.9) classified as plausible reporters of energy intake were included in the analyses. Analysis of covariance with baseline observations carried forward for drop-outs (n ¼ 38) was used. RESULTS: The basic and enhanced groups significantly increased their percentage of energy contribution from fruits and reduced energy-dense, nutrient-poor foods compared with controls (Po0.001). Successful participants (n ¼ 49) reported superior improvements in dietary intake including greater reductions in the mean daily energy intake (Po0.001), the percentage of energy from energy-dense, nutrient-poor foods ( 12.0% E vs 4.3% E, Po0.001) and greater increases in the energy contribution from fruits (Po0.001), vegetables (P ¼ 0.003) and breads/cereals (P ¼ 0.02). CONCLUSIONS: Use of a commercial web-based weight loss program facilitated some improvements in the dietary intake. The enhanced web-based tools appeared not to have generated greater improvements in reported dietary intake, compared with the basic or control groups. Those who achieved a weight loss of X5% improved their dietary intake in line with the program recommendations and dietary guidelines. Further research to determine web-based components that may improve success and the reasons why programs are successful for some participants is required. European Journal of Clinical Nutrition (2014) 68, 64–70; doi:10.1038/ejcn.2013.194; published online 16 October 2013 Keywords: Internet; weight loss; diet; obesity
INTRODUCTION Dietary change is essential for successful and sustainable weight loss. However, few studies report whether weight loss programs impact on dietary intake or what dietary changes are made by those who are successful in achieving clinically significant weight loss.1 This limits the ability to identify the dietary behaviours that are amenable to change and that contribute to successful weight loss outcomes. Furthermore, it is important to establish whether diet quality is positively or negatively affected during program participation, as a poor diet quality may compromise the health benefits of weight loss.1 As obesity prevalence increases, the demand for weight loss programs with a broad reach escalates.2 Owing to the growth in Internet access during the last decade,3 web-based treatment approaches have the potential to reach a large number of people. Weight loss can be achieved using web-based interventions4 and programs with additional features such as counselling and/or individualized feedback may enhance effectiveness compared with those providing education only.5 Most web-based weight loss intervention studies have inadequately reported changes in dietary intake, predominantly 1
reporting only the change in total energy intake.6–15 Only three web-based weight loss randomized controlled trials (RCTs) have comprehensively evaluated dietary intake, with all reporting small improvements across treatment groups but no between-group differences.16–18 Therefore, there is currently insufficient evidence to confirm the utility of web-based weight loss interventions in improving dietary intake or whether the addition of web-based features (for example, personalized feedback) facilitates greater improvements in dietary intake. Therefore, the primary aim of this secondary analysis was to identify the dietary changes associated with participation in a basic (standard features) or enhanced (basic þ personalized feedback and reminders) version of a 12-week commercial web-based weight loss program compared with a waiting-list control group, and to determine whether there were between-group differences in dietary intake change. An additional investigation, unrelated to the RCT design, compared dietary intake changes between successful (weight loss X5% of baseline weight19) and unsuccessful participants in a cross-sectional design. It was hypothesized that after 12 weeks, improvements in daily intakes of total energy, fibre, the percentage of energy from fat, saturated
Priority Research Centre in Physical Activity and Nutrition, Nutrition and Dietetics, School of Health Sciences, Faculty of Health and Medcine, University of Newcastle, Callaghan, NSW, Australia; 2Priority Research Centre in Physical Activity and Nutrition, School of Education, Faculty of Education and Arts, University of Newcastle, Callaghan, NSW, Australia; 3 Occupational Health and Saftey School of Health Sciences, Faculty of Health and Medicine, University of Newcastle, Callaghan, NSW, Australia and 4Priority Research Centre in Physical Activity and Nutrition, School of Biomedical Sciences and Pharmacy, Faculty of Health, University of Newcastle, Callaghan, NSW, Australia. Correspondence: Professor CE Collins, Priority Research Centre in Physical Activity and Nutrition, Nutrition and Dietetics, School of Health Sciences, Faculty of Health and Medicine, Room 310, Level 3 ATC Building, University of Newcastle, Callaghan 2308, NSW, Australia. E-mail:
[email protected] Received 29 October 2012; revised 17 June 2013; accepted 1 July 2013; published online 16 October 2013
Dietary intake after a web-based program MJ Hutchesson et al
65 fat, fruits, vegetables and energy-dense, nutrient-poor foods would be superior (i) in the basic and enhanced groups compared with the waiting-list control, (ii) in the enhanced group compared with basic and (iii) in successful participants compared with unsuccessful participants. SUBJECTS AND METHODS Study design This is a secondary analysis of a 12-week RCT, and the methods and primary outcomes of the trial are described in detail elsewhere.20,21 Participants were randomized to one of three groups: waiting-list control or a basic or enhanced version of a commercial web-based weight loss program for 12 weeks. The Human Research Ethics Committee of the University of Newcastle, Callaghan, New South Wales (NSW), Australia, approved the study and a written informed consent was obtained from all participants.
Participants Participants were overweight and obese (body mass index (BMI) 25–40 kg/m2) adults (18–60 years) recruited from the Hunter region, NSW, Australia in October to December 2009. They agreed not to participate in other weight loss programs, pass a health screen,22 were available for in-person assessments and had access to a computer with Internet access. Exclusion criteria were as follows: pregnancy or trying to conceive, major medical or orthopaedic problems, weight loss of X4.5 kg in the last 6 months and taking medications affecting weight loss.
Control group The waiting-list control group did not receive access to a weight loss program during the 12 weeks and were requested to maintain their usual eating and physical activity habits.
Intervention groups Both intervention groups received free access to a commercial web-based weight loss program, The Biggest Loser Club (www.biggestloserclub.com.au), provided by SP Health Co. Pty Ltd. The program was based on social cognitive theory23 and targeted key mediators of behaviour change. The basic program included evidence-based features of weight management programs delivered by the Internet,24 including goal setting, selfmonitoring with feedback, education materials and social support. The enhanced program included all features of the basic program, as well as more comprehensive feedback reports at enrolment and weekly thereafter and additional reminders to use the program features. The key features of the basic and enhanced programs are outlined in Table 1 and have been described in detail previously.20,25 The dietary prescription for both groups
Table 1.
was consistent, that is, they were aiming to achieve an individualized daily kilojoule restriction B2600 kJ less than their estimated energy requirements. To promote diet quality, both groups were provided with low-fat meal plans to meet their kilojoule target and received feedback on their nutrient intake compared with the Nutrient Reference Values for Australia- and New Zealand-recommended targets.26 In addition, as part of the feedback reports, the enhanced group received additional individualized feedback on their daily energy, saturated fat, fibre, fruit and vegetable intake, as well as ‘high-risk’ eating behaviours, such as skipping meals.
Measures Height was measured on a Harpenden portable stadiometer (Holtain Limited, Dyfed, UK).Weight was measured on a digital scale (CH-150kp, A&D Mercury Pty Ltd, Adelaide, SA, Australia) at baseline and at 12 weeks and the BMI (kg/m2) was calculated. Dietary intake was assessed using the Australian Eating Survey (AES), a 120-item semiquantitative food frequency questionnaire (FFQ). The AES has been tested for reliability and relative validity and demonstrates acceptable accuracy for ranking nutrient intakes in Australian adults.27 Information about usual frequency of intake of 120 food items over the previous 6 months at baseline and for the previous 12 weeks at follow-up was collected. A standard portion size was used for each food item, determined using ‘natural’ serving size (for example, slice of bread) or those from the 1995 National Nutrition Survey (unpublished data from the Australian Bureau of Statistics). Frequency options for food items ranged from ‘never’ up to ‘X4 times per day’ and up to ‘X7 glasses per day’ for beverages. Individual mean daily intakes were calculated using FFQspecific programs applying the Australian food composition database, AusNut 1999 (All Foods) Revision 14 (Australian Government Publishing Service, Canberra). Total daily energy, sodium, sugar and fibre intakes were calculated, and daily macronutrient intakes were converted to the percentage of total energy to allow for comparison with recommended targets. Foods were categorized into groups according to the Australian Guide to Healthy Eating’s28 ‘core’ foods (that is, breads/cereals/rice/pasta/ noodles; fruits; vegetables; milk/yoghurt/cheese; lean meats/fish/poultry/ eggs/nuts) and ‘non-core’ energy-dense, nutrient-poor foods (that is, sweetened drinks (for example, soft drink), packaged snacks (for example, crisps), confectionery (for example, chocolates and lollies), baked sweet products (for example, cakes, muffins, scones), deep-fried or take-away foods (for example, pies, pizza), fatty meats (for example, salami, devon) and condiments (for example, tomato sauce)). These food groups were further divided into eight food subgroups (that is, fatty meats, packaged savoury snacks, confectionery, baked sweet products, deep-fried or takeaway foods, soft drinks, breakfast cereals and condiments). The energy content of food items from the same food groups and subgroups was combined to calculate the percentage of energy contributed by each group.
Description of the basic and enhanced commercial web-based weight loss programs Basic and enhanced
Participants set weight loss goals, advised to self-monitor their weight, waist and hip girths. Encouraged to self-monitor via weekly e-mail and/or SMS reminders to enter weight on website. Entered data were tracked and displayed graphically and in a body (BMI) silhouette. Individualized daily calorie targets to facilitate 0.5 1 kg weight loss per week (B 2600 kJ less than their estimated energy requirements). Access to weekly low-fat menu plan and grocery lists designed to meet nutrient reference values (41) and daily calorie target. Web-based food and exercise diary to monitor energy intake and energy expenditure. Daily and weekly calculations of energy balance and nutrient summaries compared with recommended targets (41) if food entries made in online diary. Online education in the form of weekly tutorials, fact sheets, meal and exercise plans and weekly challenges. Social support via online discussion forums.
Enhanced only Personalized automated enrolment reports suggesting appropriate weight loss goals and key behaviour changes required for success. Eating behaviours targeted included total energy, saturated fat and fibre intake, daily serves of fruits and vegetables, ‘high-risk eating behaviours’ (e.g., skipping meals, not eating breakfast, drinking soft drinks) and non-hungry eating triggers. Weekly automated personalized feedback for key elements of diet and physical activity based on diary entries; usage patterns for website features; and level of success with weight loss. Eating behaviours targeted were consistent with the enrolment reports. Reminders to use the online diary visit the site and/or weigh-in. The reminder schedule included an initial reminder email; if no response, a text message; if no response, a phone call.
Abbreviations: BMI, body mass index; SMS, short message service.
& 2014 Macmillan Publishers Limited
European Journal of Clinical Nutrition (2014) 64 – 70
Dietary intake after a web-based program MJ Hutchesson et al
66 Misreporting of energy intake Plausible reporters were identified at baseline using the Goldberg cutoffs29 with basal metabolic rate calculated using the Schofield equation (29). A physical activity level of 1.55 was assigned for all participants.30 Using the Goldberg cutoffs, the upper and lower confidence limits were calculated. Participants with an energy intake basal metabolic rate42.75 were classified as over-reporters and those with an energy intake basal metabolic rateo0.87 as under-reporters. Of the 309 participants recruited to the study, 304 completed the AES at baseline assessments and 88% were classified as ‘plausible’ reporters, 34 as under-reporters and two as over-reporters. Only plausible reporters at baseline were included in the primary analysis (n ¼ 268), with the assumption that they would be plausible reporters at 12 weeks.31 There was no significant difference in the number of plausible reporters across the three groups (control n ¼ 92, basic n ¼ 86, enhanced n ¼ 90, P ¼ 0.818).
Statistical analysis Data analysis was undertaken using Stata 11.0 (StataCorp, College Station, TX, USA). Differences between the groups at baseline were tested using analysis of variance for continuous variables and w2 for categorical variables. Analysis of covariance was used to test for differences in weight and dietary intake at 12 weeks between the groups. An intention-to-treat approach was used with baseline observation carried forward for those lost to follow-up at 12 weeks (n ¼ 38). To test for differences in weight, the model was fitted using linear regression with weight at 12 weeks as the outcome variable, treatment group as the predictor variable and weight at baseline included as a covariate. To test for differences in dietary intake variables, the model was fitted using linear regression with the dietary intake variable (for example, energy intake) at 12 weeks as the outcome variable, treatment group as the predictor variable and dietary intake variable (for example, energy intake) at baseline included as a covariate. Post hoc analysis was performed using the Tukey–Kramer method. For participants who completed the 12-week assessments (n ¼ 230), weight loss was calculated from baseline to 12 weeks, and participants were categorized as successful if they achieved a X5% weight loss or unsuccessful (o 5% weight loss). The Wilcoxon rank sum test was used to test for differences in dietary intake between successful and unsuccessful participants at baseline. Analysis of covariance was used to test for differences in dietary intake variables between successful and unsuccessful participants at 12 weeks. The models were fitted using linear regression with dietary intake at 12 weeks as the outcome variable, successful/ unsuccessful as the predictor variable and weight or dietary intake at baseline included as a covariate. Post hoc analysis was performed using the Tukey–Kramer method.
Table 2.
RESULTS Participants Participants (n ¼ 268) had a mean±s.d. age of 42.3±10.1 years, a BMI of 32.1±3.9 kg/m and 59.7% of participants were women. Baseline demographic characteristics did not differ by treatment group (Table 2). Of the 268 participants, 230 completed the 12-week follow-up assessment (85.8%). There were significant differences in retention rates at 12 weeks across the three groups (B: 76.7%, E: 87.8%, C: 92.4%, P ¼ 0.009). Weight loss From intention-to-treat analysis, both the basic and enhanced groups lost significantly more weight compared with the controls, and the enhanced group lost significantly more weight compared with the basic group after 12 weeks (B: 2.3±3.4 kg or 2.5%, E: 3.1±4.0 kg or 3.4%, C: þ 0.5±2.3 kg or þ 0.6%, Po0.001; Table 3). Among study completers, both the basic and enhanced groups lost significantly more weight compared with the controls but there was no significant difference in the weight change in the basic and enhanced groups (B: 3.0±3.6 kg or 3.2%, E: 3.5±4.1 kg or 3.9%, C: þ 0.5±2.5 or þ 0.6%, Po0.001). Twenty-one percent of study completers (n ¼ 49) achieved X5% weight loss and were classified as successful. There were no significant differences in the baseline demographic or anthropometric characteristics between successful and unsuccessful participants. A significantly higher proportion (Po0.001) of participants in the basic (27.3%) and enhanced (36.7%) was defined as successful, than in the control group (2.4%). Baseline dietary intake Participants reported consuming (mean±s.d.) 10 673±2844 kJ/ day, with 31% energy from fat, 46% carbohydrate, 18% protein and 5% alcohol. Energy-dense, nutrient-poor foods contributed 41% to overall energy intake, followed by breads/cereals/rice/ pasta (18%) and lean meats/fish/poultry/eggs/nuts (17%). The dietary intake of treatment groups was similar at baseline for the majority of dietary variables (Table 3). However, the control group reported a significantly lower percentage of energy contributed by dairy foods (P ¼ 0.001) compared with the other groups, the enhanced group reported a significantly higher proportion of energy intake from baked sweet products compared with the basic group (P ¼ 0.047) and the controls reported a significantly higher percentage of energy from fatty meats compared with the basic group (P ¼ 0.01). Successful participants reported a
Sociodemographic characteristics of participants
Gender (male) Age (years) County of birth (Australia) Education level p High school Trade/diploma University degree Higher university degree Weekly household income o$700 $700–$1499 $1500 or more Don’t know/did not answer Weight (kg) BMI (kg/m2)
n (%) Mean (s.d.) n (%) n (%)
n (%)
Mean (s.d.) Mean (s.d.)
Control (n ¼ 92)
Basic (n ¼ 86)
Enhanced (n ¼ 90)
39 (42.4) 42.2 (9.2) 83 (90.2)
33 (38.4) 42.6 (10.7) 78 (90.7)
36 (40.0) 42.1 (10.4) 82 (91.1)
21 (22.8) 35 (38.0) 25 (27.2) 11 (12.0) (n ¼ 87) 8 (9.2) 15 (17.2) 62 (71.3) 2 (2.3) 93.6 (14.1) 32.2 (3.9)
25 (29.1) 28 (32.6) 20 (23.3) 13 (15.1) (n ¼ 82) 7 (8.5) 16 (19.5) 55 (67.1) 4 (4.9) 94.6 (15.7) 32.3 (3.5)
26 (28.9) 35 (38.9) 18 (20) 11 (12.2) (n ¼ 87) 5 (5.7) 12 (13.8) 68 (78.2) 2 (2.3) 92.3 (14.4) 31.9 (4.2)
Abbreviation: BMI, body mass index.
European Journal of Clinical Nutrition (2014) 64 – 70
& 2014 Macmillan Publishers Limited
Dietary intake after a web-based program MJ Hutchesson et al
67 significantly higher percentage of energy contributed by condiments compared with unsuccessful participants (Table 4). Change in energy, macronutrient and food group intakes by treatment group After 12 weeks, there were no significant differences in the mean reduction in daily energy intake (B: 1179±2620, E: 1588±2223, C: 949±2247 kJ/day), macronutrients (%E), alcohol (g), sugar (g), fibre (g), sodium (mg) or the majority of food groups and subgroups (%E) among the groups (Table 3). Both the basic and enhanced groups increased their percentage of energy intake from core food groups and decreased the percentage of energy from energy-dense, nutrient-poor foods compared with controls (Po0.0001), which equated to a reduction in energy intake of 985 kJ/day (B), 1324 kJ/day (E) and 739 kJ/day (C). Both the basic and enhanced groups significantly increased their percentage of energy intake from fruits compared with controls (Po0.001). The basic group reported a significantly greater reduction in the percentage of energy from confectionary compared with controls (P ¼ 0.03), and the enhanced group reported a significantly greater increase in the percentage of energy from breakfast cereals compared with controls (P ¼ 0.03). Changes in energy, macronutrient and food group intakes for successful and unsuccessful completers Successful participants reported significantly greater (Po0.001) reductions in energy intake ( 2789±2893 kJ/day) compared with unsuccessful participants ( 1077±2260 kJ/day; Table 4). Successful participants reported no significant changes in the percentage of energy contributed by fat and carbohydrate but reported significantly greater increases in the proportion of energy from protein (Po0.05), as well as greater reductions in total sugars (Po0.001), and sodium (P ¼ 0.02) compared with unsuccessful participants. Successful participants showed significantly greater reductions in the proportion of energy from energydense, nutrient-poor foods, equating to a 2335 kJ vs 870 kJ reduction. They increased the percentage of energy contributed by fruits (Po0.001), vegetables (P ¼ 0.003), breads/cereals/rice/ pasta/noodles (P ¼ 0.02) and breakfast cereals (P ¼ 0.002) and reported significantly greater reductions in the proportion of energy contributed by soft drinks (P ¼ 0.03), baked sweet products, such as cakes and pastries, (P ¼ 0.0001) and deep-fried take-away foods (P ¼ 0.007; Table 4). DISCUSSION This study provides a comprehensive assessment of the dietary intake changes made by adults participating in a commercial webbased weight loss RCT. Similar patterns of dietary change were observed across the three groups. Although the extent of dietary change tended to be greater in the intervention groups, few significant differences in dietary changes were detected between the intervention and control groups, and no differences were detected between the basic and enhanced groups. An important aspect of our findings was that participation in the commercial web-based weight loss programs did not compromise existing positive eating habits, and some improvements were evident. The magnitude of change in daily energy intake was consistent with other web-based weight loss intervention studies6,7,15 and was largely because of a significant decrease in the energy contributed by energy-dense, nutrient-poor foods. Intervention participants increased the energy contribution from fruits and moved closer to the recommended two serves per day.28 Intakes of most macronutrients at baseline were within Australian Nutrient Reference Values26 and were unchanged after 12 weeks. Baseline sodium intakes exceeded the recommended & 2014 Macmillan Publishers Limited
upper limit (2300 mg/day)26 but both intervention groups reduced their mean intake to this level at 12 weeks. Despite these positive improvements, there are still a number of avenues that can be targeted for further dietary improvement. Although intervention participants reduced their intake of energy-dense, nutrient-poor foods by up to three serves per day, they were still consuming on an average B3300 kJ/day from this category (6 to 7 serves). At baseline, the mean saturated fat intake of all participants was higher than the target level of o7–10% of energy26,32 and remained high at 12 weeks. Therefore, additional behavioural strategies or longer treatment may be required to achieve greater reductions in saturated fat intake and the consumption of energydense, nutrient-poor foods. Research suggests the provision of individualized feedback or counselling as part of a web-based weight management intervention is associated with a greater weight loss success,33 but little is known about its association with dietary intake. In this study, participants in the enhanced group lost significantly more weight compared with the basic group after 12 weeks, but there was no difference in dietary intake change between the intervention groups. However, as the receipt of the weekly personalized feedback on dietary intake was dependent on participants completing the online food diary, the frequency of feedback was likely to be inconsistent among participants. It is known from previous public health interventions delivered via the Internet,34 including an investigation of the current commercial weight loss program,35 that participant engagement with online features is inconsistent and declines over time. This variation in intervention dose may help explain why dietary intake did not differ between the enhanced and basic groups. Further, as the relative macronutrient content did not change substantially, the reduction in energy intake may have been accomplished by a small reduction in the frequency of consumption across a broad range of foods captured in the FFQ. It is also possible that intervention participants reduced the portion size of food items, which we were not able to detect because of the semiquantitative nature of the FFQ. Therefore, further evaluation of enhanced webbased program features, such as personalized feedback, is required to determine what features are necessary to elicit and support individuals to positively change their eating habits. Few studies have reported the dietary changes made by overweight and obese adults who are successful in achieving weight loss1; therefore, this study’s comparison of changes made to dietary intake by successful and unsuccessful participants is novel. Participants were set a daily calorie target to facilitate a 0.5–1 kg weight loss per week (approximately 2600 kJ/day less than their estimated energy requirements) and successful participants reported energy intake reduction consistent with this recommendation ( 2789 kJ/day), whereas unsuccessful participants did not ( 1077 kJ/day). The recent weight loss maintenance RCT found that increasing fruit and vegetable intakes may help achieve weight loss,36 which is consistent with the findings in this study that successful participants achieved greater increases in the energy contributed by fruits and vegetables. Successful participants also reduced their consumption of energy-dense, nutrient-poor foods by B2300 kJ/ day or approximately three serves per day. It appears that successful participants replaced energy-dense, nutrient-poor foods with nutrient-dense, lower energy ‘core’ foods such as fruits and vegetables, thereby decreasing the energy density of their total diet. Overall, participants who were successful in achieving clinically significant weight loss after 12 weeks made changes consistent with dietary guidelines and the key intervention messages. The next step for researchers is to determine why and how successful participants were able to achieve positive improvements to their dietary intake. This knowledge will help progress the design and effectiveness of behavioural weight loss interventions, to improve the likelihood of European Journal of Clinical Nutrition (2014) 64 – 70
European Journal of Clinical Nutrition (2014) 64 – 70 94.1±14.0 9804±2975 19.2±3.4 31.3±4.8 12.9±2.2 3.7±1.0 11.5±2.1 45.1±7.3 5.0±5.7 26.7±8.7 130.1±57.8 2515±864 60.3±12.0 39.7±12.0 8.0±3.7 5.9±3.8 18.9±7.4 8.3±4.9 18.9±7.4 2.5±2.0 5.8±5.6 5.0±3.9 8.9±4.9 2.4±1.7 2.5±2.0 5.7±4.3 4.3±6.4
93.7±14.1 10 753±2752 18.1±3.2 31.0±5.1 13.2±2.7 3.5±0.8 11.3±2.1 46.2±7.0 5.3±6.1 28.8±8.2 145.5±61.8 2741±813 57.3±12.2 42.7±12.2 7.6±3.4 6.4±4.2 18.9±8.2 7.9±4.6 16.5±7.1 2.6±2.1 6.2±5.3 6.3±4.9 9.6±5.6 2.2±1.5 2.5±2.1a 5.2±4.0 4.3±5.8
0.5±2.3 949±2247 1.2±2.5 0.14±3.9 0.3±1.9 0.2±0.8 0.2±1.6 1.1±4.6 0.2±3.4 2.2±6.5 15.4±39.7 226±681 3.0±8.6 3.0±8.6 0.4±2.8 0.4±2.3 0.01±6.5 0.4±4.3 2.4±5.4 1.3±1.8 0.4±4.0 1.3±3.8 0.7±3.7 0.2±1.4 0.01±1.6 0.45±2.9 0.02±4.4
Change 94.6± 15.7 10 581±2970 18.5±3.1 30.9±5.1 13.2±2.7 3.6±0.8 11.3±2.1 45.2±6.3 5.7±6.1 29.6±7.6 139.9±53.2 2649±879 60.6±12.0 39.4±12.0 7.7±3.7 6.9±5.3 18.8±7.0 10.4±6.2a 16.9±5.9 2.9±2.7 6.0±5.6 5.1±4.1a 8.6±6.0 2.4±1.7 1.8±1.5a 5.4±4.6 3.2±4.1
Baseline 92.3±16.0 9402±2702 19.6±3.6 30.5±4.8 12.6±2.2 3.7±1.0 11.3±2.2 44.5±6.0 5.4±5.7 28.7±8.9 122.2±43.8 2354±817 66.0±12.8 34±12.8 8.8±4.7 7.9±5.0 19.8±6.8 10.3±5.6 19.3±7.5 2.3±2.3 4.3±3.9 3.9±3.6 7.6±5.6 2.6±3.0 1.7±1.5 6.6±5.9 2.8±4.1
12 Weeks
Basic (n ¼ 86)
2.3±3.4 1179±2620 1.1±2.7 0.4±4.3 0.6±2.3 0.1±0.7 0.01±1.7 0.7±4.2 0.3±3.3 0.8±7.6 17.6±45.6 295±740.3 5.4±10.8b 5.4±10.8b 1.0±2.9b 1.0±3.7 1.0±6.5 0.2±5.3 2.5±6.0 0.6±2.3 1.6±4.4b 1.1±4.2 1.0±4.8 0.2±2.6 0.1±1.2 1.2±4.6 0.4±2.7
Change
92.3±14.5 10 678±2844 18.5±3.2 32.2±5.3 14.0±2.7 3.7±0.9 11.7±2.2 46.1±7.5 3.6±5.0 29.9±8.5 148.7±67.1 2685±776 58.5±11.8 41.5±11.8 8.1±3.0 6.6±4.9 17.6±5.8 9.8±5.5a 17.0±5.8 3±2.8 6.1±4.8 7.1±5.0a 9.0±5.1 2.8±1.8 2.3±1.8 5.4±4.0 4.3±5.5
Baseline
89.2±15.0 9090±2639 19.8±2.9 30.8±4.8 12.9±2.8 3.6±0.8 11.2±1.8 45.6±6.5 3.9±5.1 27.9±7.7 124.7±58.6 2322±631 65.4±11.1 34.6±11.1 8.9±3.1 8.3±4.6 19.9±5.3 10.0±6.1 19.0±5.5 2.0±2.3 4.8±4.5 4.6±3.7 7.5±3.8 2.6±1.7 2.4±1.8 7.3±4.3 3.3±4.3
12 Weeks
Enhanced (n ¼ 90)
3.1±4.0 1588±2223 1.2±2.8 1.5±4.7 1.1±2.7 0.1±0.8 0.4±1.9 0.5±4.6 0.3±4.3 2.0±5.9 24.0±52.3 363±640 6.9±10.2b 6.9±10.2b 0.8±2.7 1.8±3.7b 2.3±5.5 0.1±6.0 2.0±5.0 1.0±2.5 1.3±3.6 2.6±4.4 1.6±4.0 0.2±4.0 0.1±1.2 1.8±3.7b 1.0±4.0
Change
Data are presented as mean±s.d.. Change: 12-week value–baseline value. aPo0.05 for difference between the groups at baseline. bPo0.05 for difference between the control groups at 12 weeks.
12 Weeks
Baseline
Control (n ¼ 92)
Comparison of control, basic and enhanced participants’ weight loss and dietary intakes from baseline to 12 weeks
Weight (kg) Energy (kJ) Protein (%E) Total fats (%E) Saturated fats (%E) Polyunsaturated fats (%E) Monounsaturated fats (%E) Carbohydrates (%E) Alcohol (%E) Fibre (g) Sugars (g) Sodium (mg) Total core foods (%E) Energy-dense, nutrient-poor foods (%E) Vegetables (% E) Fruits (%E) Breads, cereals, rice, pasta and noodles (%E) Milk, yoghurt and cheese (%E) Lean meats, fish, poultry, eggs and nuts (%E) Packaged snacks (%E) Confectionary (%E) Baked goods (%E) Deep-fried take away (%E) Condiments (%E) Fatty meat (%E) Breakfast cereals (%E) Soft drinks (%E)
Table 3.
Dietary intake after a web-based program MJ Hutchesson et al
68
& 2014 Macmillan Publishers Limited
Dietary intake after a web-based program MJ Hutchesson et al
69 Table 4.
Comparison of successful and unsuccessful participants’ dietary intakes from baseline to 12 weeks o5% Weight loss (n ¼ 181)
5% Or more weight loss (n ¼ 49)
Energy (kJ) Protein (%E) Total fats (%E) Saturated fats (%E) Polyunsaturated fats (%E) Monounsaturated fats (%E) Carbohydrates (%E) Alcohol (%E) Fibre (g) Sugars (g) Sodium (mg) Total core foods (%E) Energy-dense, nutrient-poor foods (%E) Vegetables (% E) Fruits (%E) Breads, cereals, rice, pasta and noodles (%E) Milk, yoghurt and cheese (%E) Lean meats, fish, poultry, eggs and nuts (%E) Packaged snacks (%E) Confectionary (%E) Baked goods (%E) Deep-fried take away (%E) Condiments (%E) Fatty meat (%E) Breakfast cereals (%E) Soft drinks (%E)
Baseline
12 Weeks
Change
Baseline
12 Weeks
Change
11 561±3415 18.2±3.0 31.8±5.3 13.9±2.9 3.6±0.6 11.6±2.1 46.1±7.1 4.4±6.3 31.1±9.5 161.1±70.1 2821.3±836.0 57±11.2 43±11.2 7.4±3.1 6.4±5.0 17.3±6.0 9.8±5.8 16.6±5.4 3.4±3.1 6.4±5.4 7.3±5.6 9.3±6.2 2.7±1.4b 2.2±1.4 5.7±4.4 4.2±5.3
8773±2571 20.5±2.7 30.5±4.5 12.6±2.3 3.7±0.9 11.2±2.1 45.3±5.3 3.7±7.7 28.3±8.2 116.2±38.4 2238.3±651.7 69.0±11.3 31.0±11.3 9.4±4.3 8.8±4.7 20.8±4.8 10.1±5.7 20.3±6.0 1.9±2.0 4.8±5.5 3.2±3.0 6.7±4.0 3.1±3.6 2.0±1.4 8.5±4.8 2.5±3.6
2789±2893a 2.3±3.1a 1.3±5.2 1.3±2.9 0.1±1.0 0.4±2.1 0.9±5.3 0.7±5.2 2.8±8.2 44.9±60.7a 583.0±786.3a 12.0±12.2a 12.0±12.2a 2.0±3.4a 2.4±4.5 a 3.5±6.4a 0.4±6.7 3.8±5.9 1.5±2.8 1.5±4.6 4.1±5.0a 2.7±5.4a 0.4±3.5 0.3±1.6 2.8±4.5a 1.7±3.9a
10 614±2720 18.3±3.2 31.2±5.2 13.3±2.8 3.7±0.9 11.4±2.1 46.1±7.0 4.8±5.5 29.7±7.8 145.4±59.6 2690.0±806.4 59.4±12.1 40.6±12.1 8.0±3.5 6.8±4.8 18.8±7.3 9.3±5.6 16.7±6.6 2.6±2.1 6.3±5.5 6.0±4.6 8.9±5.2 2.4±1.7 2.1±1.5 5.5±4.0 3.8±5.3
9537±2908 19.4±3.5 30.7±4.9 12.7±2.5 3.7±0.9 11.4±2.0 45.2±7.0 4.8±5.5 28.0±8.9 129.4±58.5 2412.1±788.9 63.7±11.7 36.6±11.7 8.5±3.9 7.3±4.5 19.4±6.8 9.4±5.7 19.1±7.2 2.2±1.9 5.1±4.9 4.7±3.8 8.0±4.6 2.4±1.6 2.1±1.6 6.4±4.8 3.6±5.4
1077±2260 1.1±2.7 0.5±4.5 0.6±2.4 0.1±0.8 0.0±1.8 0.9±4.6 0.1±3.6 1.7±6.8 16.0±43.6 277.9±703.2 4.3±9.4 4.3±9.4 0.5±2.8 0.5±3.3 0.6±6.6 0.1±5.3 2.4±5.8 0.4±2.2 1.1±4.3 1.4±4.1 0.9±4.1 0.0±1.5 0.0±1.4 0.9±3.9 0.3±4.1
Data are presented as mean±s.d.. Change: 12-week value–baseline value. aPo0.05 for difference between successful and unsuccessful at 12 weeks. bPo0.05 for difference between successful and unsuccessful at baseline.
participants making changes to key weight-related dietary behaviours. There are several limitations to consider when interpreting the results of this study. Dietary intake was self-reported and is therefore vulnerable to reporting bias. To address this issue we identified misreporters of energy intake using the Goldberg method and only included plausible reporters in the analysis. Dietary data were obtained using a validated semiquantitative FFQ in order to reduce participant burden, as well as the research costs.37,38 However, this form of dietary assessment may not have been sensitive enough to detect changes in dietary intake in the short 12-week time frame, nor could it detect changes in portion sizes consumed, only frequency of consumption. Participants in the intervention groups were asked to self-monitor their dietary intake using an online food diary. Diary completion may have increased their awareness of dietary intake, which may have influenced how the FFQ was completed at 12 weeks. The RCT was powered to detect a significant difference in the primary outcome (BMI change) between the three groups, not dietary intake. Therefore, owing to these limitations, not all changes made to dietary intake or all between-group differences may have been detected. The provision of commercial web-based weight loss programs facilitated some positive dietary changes, including significant reductions in consumption of energy-dense, nutrient-poor foods and increases in fruit intake. Enhanced web-based tools, such as the personalized feedback, did not generate significantly greater improvements in reported dietary intake, compared with the usual web-based weight loss program, but the extent of change in dietary intake tended to be greater in the intervention groups. This suggests that further evaluation of the impact of web-based weight loss interventions on dietary behaviours using larger sample sizes is warranted. Participants who achieved a weight loss of 5% or more after 12 weeks improved their dietary intake in line with the program recommendations and national dietary guidelines. Further research is required to determine the reasons why web-based weight loss programs facilitate positive changes to eating behaviours in some participants only. & 2014 Macmillan Publishers Limited
CONFLICT OF INTEREST CEC has been a nutrition consultant to SP Health Co. MJH received a PhD scholarship supplement from SP Health Co. and Postdoctoral Research Fellowship from the Penn Foundation Australia. The remaining authors declare no potential conflict of interest.
ACKNOWLEDGEMENTS We would like to acknowledge the study subjects and research assistants (Julia Martin, Kate Fletcher, Elroy Aguiar, Ashlee Lucas, Rebecca Collins, Trevor Cripps, James Dower, Sharenjit Gill, Jenna Hannan, Skye Huxley, Hannah Mackay, Bryana Melnick, Justin Nicol, Hannah Lucas, Tom Mitchell, Huiru Teoh, Janine Wright and Mei Yap) who helped with data collection; Scott Penn, Anna Crook, Penelope Jones, Julian Barton, Sandra Mitchell and Laura Welsford from SP Health Pty Ltd. Michelle Palmer for reviewing the manuscript. This trial was funded by an Australian Research Council Linkage Project grant (2009–2012) (LP0990414, G0189752), with SP Health as the Industry Partner Organization (G0189753). CEC is supported by a National Health and Medical Research Council Australian Career Development.
REFERENCES 1 Collins CE. Dietary strategies for successful weight loss and maintenance: more evidence required. J Am Diet Assoc 2011; 111: 1822–1825. 2 Levy RL, Finch EA, Crowell MD, Talley NJ, Jeffery RW. Behavioral intervention for the treatment of obesity: strategies and effectiveness data. Am J Gastroenterol 2007; 102: 2314–2321. 3 Miniwatts Marketing Group. World Internet Users 31st December 2011. 2011, Available from http://www.internetworldstats.com/stats.htm. 4 Kodama S, Saito K, Tanaka S, Horikawa C, Fujiwara K, Hirasawa R et al. Effect of web-based lifestyle modification on weight control: a meta-analysis. Int J Obes 2011; 36: 675–685. 5 Neve M, Morgan P, Jones P, Collins C. Effectiveness of web-based interventions in achieving weight loss and weight loss maintenance in overweight and obese adults: a systematic review with meta-analysis. Obes Rev 2010; 11: 306–321. 6 Gold BC, Burke S, Pintauro S, Buzzell P, Harvey-Berino J.. Weight loss on the web: a pilot study comparing a structured behavioural intervention to a commercial program. Obesity 2007; 15: 155–164. 7 Harvey-Berino J, Pintauro S, Buzzell P, DiGiulio M, Casey Gold B, Moldovan C et al. Does using the Internet facilitate the maintenance of weight loss? Int J Obes Relat Metab Disord 2002; 26: 1254–1260. 8 Harvey-Berino J, Pintauro S, Buzzell P, Gold EC. Effect of internet support on the long-term maintenance of weight loss. Obes Res 2004; 12: 320–329.
European Journal of Clinical Nutrition (2014) 64 – 70
Dietary intake after a web-based program MJ Hutchesson et al
70 9 Harvey-Berino J, Pintauro SJ, Gold EC. The feasibility of using Internet support for the maintenance of weight loss. Behav Modif 2002; 26: 103–116. 10 Micco N, Gold B, Buzzell P, Leonard H, Pintauro S, Harvey-Berino J. Minimal in-person support as an adjunct to internet obesity treatment. Ann Behav Med 2007; 33: 49–56. 11 Polzien KM, Jakicic JM, Tate DF, Otto AD. The efficacy of a technology-based system in a short-term behavioural weight loss intervention. Obesity 2007; 15: 825–830. 12 Svetkey L, Stevens VJ, Brantley PJ, Appel LJ, Hollis JF, Loria CM et al. Comparison of strategies for sustaining weight loss: the weight loss maintenance randomized controlled trial. JAMA 2008; 299: 1139–1148. 13 Tate DF, Jackvony EH, Wing RR. Effects of Internet behavioural counselling on weight loss in adults at risk for type 2 diabetes: a randomised trial. JAMA 2003; 289: 1833–1836. 14 Tate DF, Jackvony EH, Wing RR. Arandomised trial comparing human e-mail counselling, computer-automated tailored counselling, and no counselling in an Internet weight loss program. Arch Intern Med 2006; 166: 1620–1625. 15 Tate DF, Wing RR, Winett RA. Using Internet technology to deliver a behavioural weight loss program. JAMA 2001; 285: 1172–1177. 16 Collins CE, Morgan PJ, Warren JM, Lubans DR, Callister R. Men participating in a weight-loss intervention are able to implement key dietary messages, but not those relating to vegetables or alcohol: the Self-Help, Exercise and Diet using Internet Technology (SHED-IT) study. Public Health Nutr 2011; 14: 168–175. 17 Patrick K, Calfas KJ, Norman GJ, Rosenberg D, Zabinski MF, Sallis JF et al. Outcomes of a 12-month web-based intervention for overweight and obese men. Ann Behav Med 2011; 42: 391–401. 18 Webber KH, Lee E. The diet quality of adult women participating in a behavioural weight-loss programme. J Hum Nutr Diet 2011; 24: 360–369. 19 Thomas P. Weghing the Options: Criteria for Evaluating Weight-Management Programs. National Academy Press: Washington, DC, USA, 1995. 20 Collins CE, Morgan PJ, Jones P, Fletcher K, Martin J, Aguiar EJ et al. Evaluation of a commercial web-based weight loss and weight loss maintenance program in overweight and obese adults: a randomised controlled trial. BMC Public Health 2010; 10: 669. 21 Collins CE, Morgan PJ, Jones P, Fletcher K, Martin J, Aguiar EJ et al. A 12-week commercial web-based weight-loss program for overweight and obese adults: randomized controlled trial comparing basic versus enhanced features. J Med Internet Res 2012; 14: e57. 22 Norton K. Sports Medicine Australia pre-exercise screening system: Australian Department of Health and Ageing 2005, Available from http://sma.org.au/ wp-content/uploads/2009/05/new_pre_screening.pdf.
European Journal of Clinical Nutrition (2014) 64 – 70
23 Bandura A. Social foundations of thought and action: A Social Cognitive Theory. Prentice-Hall: Englewood Cliffs, NJ, USA, 1986. 24 Khaylis A, Yiaslas T, Bergstrom J, Gore-Felton C. A review of efficacious technology-based weight loss interventions: five key components. Telemedicine e-Health 2010; 16: 931–938. 25 Neve M, Morgan P, Collins C. Weight change in a commercial web-based weight loss program and its association with website use: a cohort study. J Med Internet Res 2011; 13: e83. 26 National Health and Medical Research Council. Nutrient Reference Values for Australia and New Zealand including Recommended Dietary Intakes. ACT: Department of Health and Aging: Canberra, 2006. 27 Collins CE, Watson JF, Guest M, Boggess MM, Duncanson K, Pezdirc K, Rollo M, Hatchesson MS, Burrows TL. Reproducibility and comparative validity of a food frequency questionnaire for adults. Accepted 26 September 2013; doi:10.1016/ j.clnu.2013.09.015. 28 Kellet E, Smith A, Schmerlaib Y. The Australian Guide to Healthy Eating. The Department of Health and Aging: Canberra, 1998. 29 Black AE. Critical evaluation of energy intake using the Goldberg cut-off for energy intake:basal metabolic rate. A practical guide to its calculation, use and limitations. Int J Obes Relat Metab Disord 2000; 24: 1119–1130. 30 Schofield WN. Predicting basal metabolic rate, new standards and review of previous work. Human nutrition. Clin Nutr 1985; 39(Suppl 1): 5–41. 31 Black AE, Cole TJ.. Biased over- or under-reporting is characteristic of individuals whether over time or by different assessment methods. J Am Diet Assoc 2001; 101: 70–80. 32 National Heart Foundation of Australia. Position statement: dietary fats and dietary sterols for cardiovascular health. National Heart Foundation of Australia, 2009. 33 Neve M, Morgan PJ, Jones PR, Collins CE. Effectiveness of web-based interventions in achieving weight loss and weight loss maintenance in overweight and obese adults: a systematic review with meta-analysis. Obes Rev 2010; 11: 306–321. 34 Glasgow RE. eHealth evaluation and dissemination research. Am J Prev Med 2007; 32: S119–S126. 35 Neve M, Collins C, Morgan P. Dropout nonusage attrition, and pretreatment predictors of nonusage attrition in a commercial web-based weight loss program. J Med Internet Res 2010; 12: e69. 36 Champagne CM, Broyles ST, Moran LD, Cash KC, Levy EJ, Lin PH et al. Dietary intakes associated with successful weight loss and maintenance during the weight loss maintenance trial. J Am Diet Assoc 2011; 111: 1826–1835. 37 Rutishauser IH. Dietary intake measurements. Public Health Nutr 2005; 8: 1100–1107. 38 Thompson FE, Subar AF. Dietary assessment methodology. In: Coulston A, Boushey C editors. Nutrition in the Prevention and Treatment of Disease. 2nd edn. Elsevier Inc.: Burlington, MA, USA, 2008.
& 2014 Macmillan Publishers Limited
