The Lifestyle Modification Program (LMP) is carried out at the Centre for Nutritional Studies, The Chinese University of Hong Kong. The LMP design is adopted from the experience of a previous study [2]. As mentioned earlier, lifestyle modification involves both behavioral and cognitive changes [11]. The LMP is developed based on the CBT concept and multiple theories.

Upon enrollment into the program, all individuals undergo a comprehensive health and dietary assessment; each individual completes an initial nutrition assessment with a registered nutritionist. During the initial assessment, the nutritionist carries out a complete behavioral assessment and collects important information, such as the client’s history of health problems, current eating and lifestyle patterns, specific eating-related behaviors, knowledge of risks associated with current eating patterns, concerns and feelings about specific lifestyle changes (perceived susceptibility and perceived severity), past experiences and barriers to change in dietary intake and lifestyle (perceived barriers), reasons for changing current eating patterns (cues to action), and readiness to change (stage of change). The nutritionist also discusses with the client the expected duration and specific dietary and lifestyle advice to achieve a desirable weight status. Thereafter, an exercise instructor designs individualized cardiovascular and resistance exercises to perform at home. On completion of the nutrition assessment and exercise advice, clients are encouraged to attend weekly 15 to 20 minute counseling sessions with the same nutritionist and exercise instructor.

During the follow up counseling, the nutritionists review the client’s diet compliance, exercise adherence, and progression. The nutritionists review compulsory food records to ensure nutritional adequacy of the diet and offer recommendations for controlling caloric intake. The dietary component and portion sizes of the program are based on American Dietetic Association exchange lists, with 20% caloric restriction as appropriate to the individual [16]. A varied balanced diet with an emphasis on fruit and vegetables, and low-fat, low-glycemic index, and low calorific products in appropriate portions are encouraged. Clients are encouraged to increase daily activity, accumulating 30 to 60 minutes of cardiovascular activity one to five days a week. Clients can repeat these exercise consultations as often as they feel necessary. In addition to reviewing the client’s diet compliance and exercise adherence, the nutritionists encourage the clients to share any personal and environmental barriers to lifestyle change. The nutritionists aim to assess the client’s feelings with the dietary advice, provide ongoing support and encouragement, and affirm the client’s efforts and achievements. To minimize the barriers to lifestyle changes, each client is provided with a booklet for portion size exchange and tips for eating out. The nutritionists also encourage the clients to bring new food products and restaurant menus to discuss and explore healthy choices, and to send any enquiries to their nutritionist to exchange case management skills and updated information on nutrition, lifestyle and health.

The LMP training was developed and conducted in several steps. First, a nutritionist manager trained the nutritionists in CBT concept, multiple theories of behavioral change, and the LMP practice. The training consisted of two one-week training sessions. The first week included an introductory training, in which the trainees shadowed the nutritionist manager; the second week consisted of training the nutritionists mostly on knowledge and dietary counseling skills using the CBT concept and multiple behavioral change theories. Basic CBT and behavioral theories were specifically explained to each nutritionist working in the LMP. Each month, the nutritionists further participated in half-day follow up case discussions with the nutritionist manager to discuss initial experiences with the patients and to refresh their CBT knowledge and dietary counseling skills. Finally, they received on-demand feedback and advice on CBT-related issues by the base trainer, supervised until September 2002 when the actual study started. The demographics of the nutritionist participants were female fresh graduates with a Bachelor of Nutrition or Food Science related degree and were members of the British Nutrition Society as registered nutritionists. They had not undertaken any training in counseling and the language of counseling was conducted in Chinese.

1,501 participants were self-referred to the LMP between 2002 and 2006. Research participants were drawn from the Chinese University of Hong Kong, Centre for Nutritional Studies cohort (n = 645) between 2002 and 2006 who met the following inclusion and exclusion criteria. Inclusion criteria were: aged at least 18 years and less than 65 years, BMI ≥ 23 kg/m², and enrollment of at least 12 weeks with completed weight measurements at baseline and at 12 weeks. Based on the World Health Organization’s criteria for Asians, the classifications of BMI of 23–24.9 kg/m² as overweight and BMI ≥ 25 kg/m² as obese were used [17]. Each participant attended one session each week. Participants on any prescription medication were excluded. They were informed of the study when they attended their appointment, and recruited on a voluntary basis. No reward was given to the participants. Informed consent was obtained from all participants. The study was approved by the Ethics Committee of the Chinese University of Hong Kong.

Four nutritionists were based at the centre and subjects were randomly allocated based on first available appointment. The LMP had a structured approach with each subject attending a session once a week with the same nutritionist. The nutritionist manager (nutritionist A) was selected to enable comparison with the other three nutritionists (nutritionists B to D) whom received training as stated above. Upon receiving the client’s consent, the LMP counseling process was audio-taped and transcribed by the researcher. Themes relating to attributes of nutritionists were identified through coding of clinician’s and client’s data and case comparisons.

Individual interviews were conducted by two interviewers with a semi-structured interview guide to collect nutritionist’s views. Topics included weight management skills and techniques for obesity counseling according to CBT [11], and nutritionist’s view on practice and perceptions of what makes a good nutrition counselor. Recall of a recent encounter with an obese patient was used to ground the discussion. The interviews lasted 30 minutes. Each interview was recorded and transcribed verbatim. The transcripts were read twice by each interviewer. Key words and phrases in the transcripts were identified, compared and formed into main themes and subthemes using method described by Ryan and Bernard [18], which were discussed jointly together by the interviewers.

24 subjects (six per nutritionist) were approached to join the study and agreed to allow an observer to watch the nutritionist-client communication in one session. The observation instrument is based on the themes set in the evidence that health professionals need to achieve longer term weight control, and is based on the recommendations for improved training in obesity management for all health professionals [8]. A final sample of 24 subjects were used as no new information emerged from the last few observation sessions of each round indicating that saturation had occurred [19].

Weight, height, BMI, blood pressure, body fat % and waist and hip circumference were measured at baseline, whereas weight was measured again at 12 weeks. Changes from baseline were calculated. After normality was checked and log transformation was applied to baseline weight, six outliers of weight change were excluded from the study. A student’s t test was used to examine gender differences in age and baseline characteristics. Paired t test was used to examine the weight change between baseline and 12 weeks. ANOVA was used to examine the difference in age, baseline characteristics and weight change between the groups that each nutritionist was assigned. ANCOVA was used to examine the effect of nutritionist variation on weight. Men and women were analyzed separately. Weight change was entered as dependent variable. The fixed factor was nutritionist whereas covariates included age and baseline weight. Nutritionist A was used as reference category in the contrast test. SPSS version 13.0 was applied and all tests are two sided with the significance level set at 0.05.

Characteristics of participants are presented in Table 1. Relative normal values of BMI, blood pressure, waist and hip circumference from the Population Health Survey (2003−2004), with a median age of 42 and 43 for females and males respectively [1], and the Heart Health Survey (2004/2005) by four adult age groups [20] were compared with the study participants. All anthropometric data and blood pressure data of the study participants were higher than the relative normal values, supporting that the study sample was overweight and obese.

No significant differences were found in the gender distribution among patients of the four nutritionists. There were significant differences found in some baseline characteristics of female patients, but not male patients. These included weight (p = 0.001), BMI (p = 0.003), % body fat (p = 0.046), systolic blood pressure (p = 0.002) and hip circumference (p = 0.001). ANOVA was used to examine these differences among the four nutritionists; significant differences were found in patient’s weight, BMI, % fat, systolic blood pressure and hip circumference between nutritionists A & B’s and nutritionists A & C’s populations of patients, but no differences were found between nutritionists A & D’s and nutritionists C & D’s populations of patients (details not shown).

The mean weight change of all individuals completing the program is shown in Table 2. There were significant reductions in weight and BMI (p < 0.001) with an average of nine percent loss of initial body weight after 12 weeks on the LMP.

Subject’s age, baseline weight, gender, and staff variation were significantly associated with the weight change. Increasing age was significantly associated with a smaller weight change (r = 0.04, p = 0.001 in women; r = 0.09, p < 0.001 in men). Subjects with greater baseline weight (log transformed) showed greatest weight change (r = −11.14, p < 0.001 in women; r = −18.66, p < 0.001 in men). Women showed smaller weight reduction than men (t = 9.48, p < 0.001). When other variables were held constant, there was nutritionist variation in LMP performance (measured as weight outcome). Table 3 shows the effect of staff variation on weight change during the LMP. Nutritionist C showed significant difference from nutritionist A (p < 0.001) in women, and nutritionist D showed significant differences from A in both women and men (p < 0.001 in women; p < 0.01 in men), but there was no significant difference observed in patient’s weight change between nutritionist A and B .

Based on the data emerging from the interviews, all the themes and subthemes identified are shown in Table 4. Further qualitative analysis was performed to provide an insight explaining such differences through semi-structured interviews of individual nutritionists.

Nutritionist’s vision of what the LMP is was explored. Nutritionists A and B shared the same vision in “helping people achieve a healthy lifestyle including three components: diet, exercise and psychological health.” On the other hand, nutritionists C and D had the vision “to improve health status and lifestyle through provision of dietary advice and lifestyle modification advice” and lacked the importance of psychological aspects of the LMP.

The general view held by nutritionists on their role was to provide information, emotional support, and guidance. Nutritionists C and D understood that their role as a nutritionist was “a provider of information and of emotional support”. A and B also held this view but added that “…nutritionists are also a guidance provider to point out right and wrong and then to provide suggestions.”

Common themes emerged from all four nutritionists on the importance of establishing a good rapport and relationship with the patient. Nutritionists aimed to treat patients as a friend to show genuine empathy:

“…to test their acceptance level and attitude…for different patients we would use a different tone to show we care for them and enhance our relationship with patients. It is important not to interrupt patients, to ask questions, to show respect, and build up a rapport like a friendship” (Nutritionist A).

Professionalism was also seen as important to empathy and rapport as viewed by nutritionists A and B.

“…being firm and professional, to share personal matters to show understanding” (Nutritionist A) and illustrating a “strong tone and strictly pointing out the problem” (Nutritionist B).

There were differences in how to influence weight loss. Nutritionists A and B adopted a staged paradigm for such intervention to be matched to the stage of change of each individual, and that an understanding of this psychological (CBT theory) aspect was essential for sustaining weight loss during the intervention.

“…We need to capture the client’s attitude towards weight loss…use different strategies depending on the patient …discussing client’s motivation and assess readiness to change …and be flexible and use different strategies for different cases and provide ongoing support by follow up calls” (Nutritionist A).

“…need to adjust advice to patient’s current lifestyle…and discuss with clients about the readiness to change and their own responsibility” (Nutritionist B).

In contrast, nutritionists C and D placed more emphasis on empathy and rapport with the patient as their main strategy and lacked change statements.

Nutritionists A and B had a shared understanding of the importance of the nutritionist-patient relationship in helping patients find the underlying issues of their obesity and solutions through meticulous investigation. This was also validated by the independent observer, see Table 5.

“…sometimes we need to change roles and think from patient’s perspective as if we were in their situation…trust is an important issue, as is nutritionist-patient sharing to help tackle the problem together” (Nutritionist A).

“…as a counselor to remind patients to self-reflect on their problem and situation to help find solutions together” (Nutritionist B).

Nutritionists C and D emphasized empathy and rapport as essential skills in the treatment of obesity without extracting the key obstacles and psychological aspects that patients are experiencing with weight loss. Therefore the use of the psychological aspect of the stage of behavior change model was not observed with nutritionist C and D. Table 5 demonstrates comparable client-centered qualities in nutrition counseling.