This study is a pragmatic evaluation of an ongoing weight management programme delivered in real life supported by Ningbo government. In total, 977 patients from July 2015 to January 2020, who were invited and willing to participate in this 36-month programme during outpatient counselling visits at Department of the Endocrinology of Ningbo First Hospital, Zhejiang Province, China. All adults aged 18–75 years with body mass index (BMI) ≥24 kg/m² were included in this programme (16). The exclusion criteria of this programme included: (1) age <18 years at baseline; (2) secondary obesity caused by other medical conditions or diseases; (3) diagnosis of any types of cancer or severe coronary heart disease; (4) receiving any related non-pharmaceutical/pharmaceutical intervention; (5) pregnancy or lactation; (6) mental illness; (7) cognitive impairment. The target goal of this programme is to achieve 5–15% weight loss in the overweight and obese patients.

The ongoing research programme is approved by the Ethics Committee of Ningbo First Hospital, China (No. 2019-R049). Written informed consents were obtained from all the participants.

All the eligible participants were asked to complete dietary behaviour and lifestyle questionnaires including socio-demography, dietary intakes for the past week, physical activity for the past week, medication record and medical history. Besides that, participants were asked to fill out questionnaire of Yale Food Addiction Scale (YFAS) to identify the addiction of their usual dietary intakes (17).

Health status including anthropometry, blood pressure and biomarkers was examined by experienced nurses. Anthropometry, including height, weight, waist circumference, hip circumference, basal metabolic rate, body fat percentage and muscle mass percentage, was examined by body composition detector (GAIA KIKA, Korea) (18). Blood pressure was measured using an electronic sphygmomanometer on the right or left arm after a 5-min rest. Biomarkers containing blood lipids (total cholesterol, triglyceride, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol), fasting plasma glucose, fasting plasma insulin, fasting c-peptide, glycated haemoglobin A1c, uric acid, alanine aminotransferase, aspartate aminotransferase, and free fatty acids were examined by biochemical automatic analyzer.

The body weight and height were measured at 3, 6, 12, 24, and 36-month follow-ups, after completing the same questionnaires in baseline.

The weight management programme including nutrition, physical activity, psychology and endocrinology delivered by a multidisciplinary team consisting of three clinicians, two dieticians, two exercise specialists, and one psychologist (Table 1). All the participants were divided into 10-person groups and assigned to the intervention programme on group basis. One group leader was selected by the group members, who was a key factor to support our multidisciplinary team. The function of the group leaders was to motivate his/her members participating the rest of follow-ups and reminded them to record their daily dietary intake, physical activities and changes in their lifestyle. At the baseline visit, each individual participated in face-to-face counselling sessions for personal guidance according to the medical reports (Table 1). The purpose of nutritional advice was to reduce energy-dense foods by controlling diet portion size and regulating dietary intakes rationally. During the nutritional counselling session, each participant was advised to adjust their dietary patterns based on the distribution of daily energy intake from macronutrients and change dietary patterns appropriately (19). During the counselling session of physical activity, participants were motivated to join outdoor and indoor physical activity at conveniences, guided the appropriate types of physical activity and encouraged to exercise at least 150 min per week. The advice on physical activity followed the American College of Sports Medicine (ACSM)'s guideline for exercise testing and prescription tenth edition for overweight and obese adults (20). Participants at psychological counselling session were asked to report their recent mood status (happiness, satisfaction, self-esteem, depression stress, anxiety, and sleep quality) and their behaviour changes in their recent daily life were discussed with clinical psychologist. Participants were suggested to record their behaviour changes and mood status, and document the progress. During the endocrinology session, advice on medication and was delivered based on their baseline health examination. After face-to-face counselling sessions, all the participants as group unit gathered together for education programmes delivered by one general physican.

All the participants recorded their daily dietary intakes and physical activity via Bohe health APP on mobile. Bohe health APP was created for personal management to evaluate dietary patterns, nutrition status and anthropometric status (e.g., body weight) (21). Besides APP online programme, participant group basis joined WeChat group (22), organised by the multidisciplinary team. All adults participated in a 21-day intensive intervention programme (phase 2). During phase 2, participants were compulsory to report their anthropometry (e.g., body weight), daily dietary intakes and physical activity to the multidisciplinary team through WeChat. Self-measured body weight was suggested to participants reporting on WeChat online programme in order to motivate participants to self-regulate and self-control their body weight. At the WeChat online counselling session, participants got their personal feedbacks on personal advice, their doubts of weight management, performance, and progress.

During the follow-ups, advice on nutrition and physical activity was delivered face to face to provide their personal solution (phase 3). At the psychological counselling session, participants gained professional advice if they had questions and desired to have a chat with clinical psychologists. Moreover, participants on group basis followed synchronous online (Bohe health APP on mobile) and WeChat for further advice as well when they had doubts and questions. Each patient got personal feedback from our professional multidisciplinary team.

Descriptive analysis was presented as number and percentage for category variables and mean and standard deviation (SD) or median for continuous variables. Statistical differences in age and biomarker parameters were compared between men and women by Student's t-test and Mann-Whitney U test. A sensitivity analysis was undertaken for comparison between participants' baseline weight and baseline weight among those dropout participants at 36 months aiming to detect whether missing values of dropouts affect the outcomes. Mean changes in weight across 36-month study period controlling for age, education levels and marriage status were examined with Bonferroni correction by repeated measures analysis of covariance. Mean weight loss at 36 month between men and women was tested by non-parametric test.

Results were considered statistically significant at a two-tailed level of 0.05. Statistical analysis were conducted using the STATA statistical software package version 15 (2017).

In total, 692 out of 977 adults (40.8% men) fulfilling all inclusion criteria and willing to participate in this 36-month programme were included in the baseline (Table 2). Among all the enrolled patients, 579, 465, 290, 204, and 122 patients participated in the follow-ups at 3, 6, 12, 24, and 36 months, respectively.

The baseline characteristics of participants was described in Table 3. The mean age of all the participants was 30.9 years (men: 29.4 years, women: 31.9 years). Mean weight and BMI were significant higher in men than women. Around 87.6% and 95.7% of participants were defined to have general obesity and abdominal obesity, respectively. The result of sensitivity analysis showed that no significant difference of baseline weight between all participants and dropout participants at 36 months (P = 0.678).

Weight loss was evaluated across the entire 36-month programme (Table 4). Weight significantly reduced after 36 months in both sex with and without Bonferroni correction (adjusted difference, male: −7.0, P < 0.001; female: −7.4, P < 0.001). Participants had a significant initial weight loss of adjusted weight at 3-month follow-up in both sex. Mean weight at each follow-up was significant from mean baseline weight. However, adjusted weights were slightly regained in both sex (adjusted weight: men, −7.4 kg at 24 months and −7.0 kg at 36 months; women, −7.4 kg at 36 months).

A sizable proportion of participants sustained clinically significant weight loss at 3, 12, and 36-month follow-ups (Figure 1). Overall, at 36-month follow-up, 8.7% of all baseline participants (6.7% of men and 10.0% of women) reached the target goal of weight loss 5–15%. Moreover, 2.3% of all baseline participants (2 men and 14 women) reached ≥20% weight loss at 36 months, exceeding the originally target goal.

Our study is the first comprehensive lifestyle counselling intervention programme to evaluate the effect of our strategies on weight loss and maintenance of weight loss. Our progrmme strategies included individual counselling and group education, provided personal feedback through WeChat APP during the entire programme. However, this programme still has some limitations. First, the complicated intervention design with long-term follow-up resulted in high dropouts at 36 months indicating participants lost interests in this long-term programme. Relocation of our participants and switching hospitals caused high dropouts as well. Second, self-reported body weight used for their personal feedbacks via WeChat online sessions might not reflect the real body weight of overweight and obese patients due to psychological factor. The under-reported body weight cannot affect the accurate weight and patients could not gain appropriate feedbacks and advice resulting in the influence in weight control and weight loss maintenance. Third, due to the structure of our questionnaires, some basic information was not well-collected, thus, baseline study could not reflect more details on daily energy intake and physical activity level. The last but not least is that no control group can be compared to the intervention group because all the outpatients followed the same standard intervention programme in this ongoing real life study. In the future study, energy intake and physical activity should be well-collected to estimate the levels for total energy intake and physical activity, which may affect weight loss.