Development of an individualized WeChat mini program-based intervention to increase adherence to dietary recommendations applying the behaviour change wheel among individuals with metabolic syndrome

Abstract

Background

Metabolic Syndrome (MetS) is a serious public health issue. Dietary changes form the core of MetS treatment. The adherence to dietary recommendations is critical for reducing the severity of MetS components and preventing complications. However, the adherence to dietary recommendations was not adequate among adults with MetS. This study utilizes the Behaviour Change Wheel (BCW) to develop an individualized WeChat mini program-based behavioural change intervention aimed at strengthening adherence to dietary recommendations in people with MetS.

Methods

The BCW theory was used to design an individualized WeChat mini program-based behavioural change intervention. A descriptive qualitative study was conducted to identify the determinants of adherence to dietary recommendations in individuals with MetS. The study was conducted at the health promotion centre of a prominent general university hospital in Zhejiang, China. Subsequently, the intervention functions (IFs) and policy categories were selected following the identified determinants. Afterwards, behaviour change techniques (BCTs) were chosen to translate into potential intervention strategies, and the delivery mode was determined.

Results

Our study identified fifteen barriers to improve the adherence to dietary recommendations in this population. These were linked with six IFs: education, training, persuasion, enablement, modelling, and environmental restructuring. Then, twelve BCTs were linked with the IFs and fifteen barriers. The delivery mode was a WeChat mini program. After these actions, an individualized WeChat mini program-based behavioural change intervention was developed to enhance adherence to dietary recommendations for individuals with MetS.

Conclusions

The BCW theory helped scientifically and systematically develop an individualized WeChat mini program-based behavioural change intervention for individuals with MetS. In the future, our research team will refine and upgrade the WeChat mini program and then test the usability and effectiveness of the individualized WeChat mini program-based behavioural change intervention program.

Key Messages

• This is the first paper to specify the content and active ingredients of an individualized WeChat mini program-based behavioural change intervention using a systematic evidence- and theory-based method for individuals with MetS.

• The findings can be used to provide guidance to improve dietary adherence, and ultimately improve the lives of people with MetS.

• This methodology can serve as a reference for other researchers who are developing behavioural change interventions.

Keywords:

• Metabolic syndrome
• diet
• mobile health
• behaviour change wheel

Introduction

Metabolic syndrome (MetS) is a serious public health problem. It was characterised by increased waist circumference (WC), high systolic blood pressure (SBP) or diastolic blood pressure (DBP), high triglyceride (TG) levels, low high-density lipoprotein cholesterol (HDL-C), and elevated fasting blood glucose (FBG) [1]. Approximately 25% of adults worldwide are estimated to have MetS [2,3]. The general prevalence of MetS is 31.1% among adults ≥20 years old in China [4]. Globally, MetS increases the risk of cardiovascular mortality by 2.5 times, the risk of diabetes by 5 times, the risk of coronary heart and cerebrovascular disease by 2 times, and the risk of all-cause mortality by 1.5 times [5]. Otherwise, MetS costs are trillions and will continue to increase [6]. Therefore, it is critical to develop effective prevention and treatment programs to control MetS.

MetS is a constellation of established risk factors of cardiovascular diseases. The World Health Organization states that almost 80% of cardiovascular diseases can be averted through a healthy diet, increased physical activity levels, and smoking cessation [7]. Therefore, diet plays a pivotal role in the prevention and treatment of MetS, and people with MetS should follow dietary recommendations [8,9]. Although dietary recommendations seem simple, their implementation and maintenance remain unsatisfactory. Chen et al. [10] reported that 61% of the participants (n = 241) exhibited suboptimal dietary adherence. Zheng et al. [11] showed that the diet score was not high among patients with MetS. Suboptimal adherence increases the likelihood of MetS components and the risk of complications [12,13]. A greater understanding of how to improve adherence is needed to enable individuals with MetS to benefit from dietary recommendations. Thus, intervention strategies to overcome inadequate adherence are a research priority.

Numerous interventions aimed at improving adherence to a healthy diet have been identified as having varying degrees of effectiveness. However, most diet interventions were delivered by face to face, booklet, or telephone among adults with MetS [11,14], and necessitated significant dedication from both patients and health care professionals due to the financial implications, time requirements, and absence of immediate outcomes. This greatly limited the implementational scalability of interventions for promoting healthy diets. To overcome the existing drawbacks, mobile health (mHealth)-based interventions have been developed as potential solutions. mHealth involves utilizing mobile devices such as phones, tablets, personal digital assistants, and wireless infrastructure to support medical and public health practices [15]. Two systematic reviews found that mHealth-based interventions had the ability to significantly improve diet behaviours [16,17]. Such findings suggest that the use of advanced technologies could provide a valuable means of improving diet adherence.

However, previous mHealth-based diet interventions offered general ‘one-size-fits-all’ dietary recommendations among people with MetS [18,19], which have achieved limited outcomes [20]. Individualized lifestyle education helped patients with MetS adopt lifestyle interventions as part of their daily routine [21]. The concept of individualization refers to designing intervention content based on an individual’s specific characteristics, including existing behaviours, stages of behaviour change, preferences, and barriers [22]. Some studies that provide individualized recommendations may face limitations in offering fully individualized advice due to the limited functionality of mHealth tools such as email, websites, mobile phones, and mobile texts [18,19]. In this case, a WeChat mini program with diversified functions provides solutions for meeting the diversified individual needs of patients with MetS. Nonetheless, no study has been found that develops an individualized WeChat mini program-based intervention to improve dietary compliance among adults with MetS in China.

Moreover, most diet interventions lack systematic theoretical guidance [23,24], which may lead to suboptimal dietary adherence and limit the overall success of interventions [10]. Using theory to change behaviour enhances the chances of intervention success [25,26]. In this context, it is recommended to utilize a theoretical framework when designing an intervention with the goal of enhancing dietary compliance among people with MetS. Behavioural change has multiple theories and models, but the majority of them ignore the background in which the target behaviour occurs, pay little attention to the reflective process, have a static structure, and fail to clarify how to change behaviour [27]. Michie et al. [28] indicated that existing frameworks of behavioural change interventions also lack comprehensiveness, coherence, and behavioural change models. To address the limitations of the above theories and frameworks, the Behaviour Change Wheel (BCW) framework was created, containing 19 behavioural change frameworks [27]. At present, the BCW has been widely used to design interventions in some contexts. However, we found that the design and development of interventions using the BCW to facilitate adherence to dietary recommendations in people with MetS are scarce.

We aimed to use the BCW to systematically develop an individualized WeChat mini program-based behaviour change intervention to promote adherence to dietary recommendations in adults with MetS in China.

Methods

This study is a component of a comprehensive intervention aimed at promoting a healthy diet and physical activity in individuals with MetS in Mainland China. Our study was performed according to the stages and steps of the BCW guide (see Figure 1) [29]. In short, it consisted of three layers (see Figure 2), with the COM-B (capability, opportunity, motivation, and behaviour) model serving as its core, which synthesises existing behavioural change theories and explores the determinants of the desired behaviour. Moreover, the Theoretical Domains Framework (TDF) was incorporated into the BCW to expand upon the COM-B model and more deeply understand the determinants of target behaviour [30]. The second and outer layers of the BCW are the intervention functions (IFs) and policy categories, respectively, which are designed to promote behavioural change [28]. Moreover, the Behaviour Change Technique Taxonomy v1 (BCTTv1), which is an active ingredient, has been linked to the BCW to help select specific behaviour change techniques (BCTs) and deliver the IFs [31].

Figure 1. Stages involved in an intervention development using the BCW [27] (used with permission from authors).

Figure 2. the Behavior Change Wheel (used with permission from authors) [27].

Stage 1: understand the behaviour

The first stage involved four steps to understand the behaviour.

Step 1: define the problem in behavioural terms

Step 1 involved defining the problem in behavioural terms. Doing so required identifying the problem and specifying the behaviour and target population [29]. To define the problem, we reviewed evidence on dietary adherence among people with MetS. The databases PubMed, Embase, Web of Science, Cumulative Index of Nursing and Allied Health Literature (CINAHL), Cochrane Library, Chinese National Knowledge Infrastructure, Weipu, and Wanfang were systematically searched using the following keywords: ‘metabolic syndrome’, ‘diet’, ‘nutrition’ and ‘lifestyle’. When applicable, manual searches were performed on the cited references in relevant papers. Furthermore, we conducted a cross-sectional study to investigate the diet behaviours of participants with MetS at the health promotion centre of a prominent general university hospital in Zhejiang, China. From January to May 2021, 288 individuals (≥18 years old) diagnosed with MetS were asked to independently complete the diet survey. One dimension of the Health Promoting Lifestyle Profile II, namely, nutrition, with six items, was used to assess diet behaviours [32]. Each item included four options and was measured on a four-point Likert scale from 1 (never) to 4 (routinely). The total score varies between 6 and 24. Two trained researchers contacted the participants and collected self-reported demographic characteristics and diet behaviours from them.

Step 2: select target behaviour

Step 2 aimed to determine which behaviour could address the defined problem. Step 2 involved consideration of all the specific behaviours to potentially target in the intervention design. The BCW suggested starting with 1 or 2 target behaviours and building on these behaviours incrementally [29]. Therefore, not all possible behaviours needed to be included in the intervention. We selected potential target behaviours by reviewing existing literature on diet management measures for individuals with MetS. Furthermore, according to the BCW framework, four criteria were used to select the final target behaviour: (i) how much of an impact changing the behaviour would have on the desired outcome, (ii) how likely it is that the behaviour can be changed, (iii) how likely it is that the behaviour would have a positive or negative impact on other related behaviours, and (iv) how easy it would be to measure the behaviour [29].

Step 3: specify target behaviour

Step 3 involved specifying the context of which the target behaviour occurred, to which there were several components: who would carry out the target behaviour, what they would need to do to achieve change, where and when the behaviour was performed, and how often and with whom. A literature search on diet interventions among people with MetS was undertaken to specify the target behaviour.

Step 4: what needs to change?

Step 4 involved conducting research to explore the barriers and facilitators to the target behaviour. Based on the COM-B model and TDF, a descriptive qualitative method using semi-structured interviews was employed to unravel adherence from the perspective of individuals with MetS. A qualitative descriptive method could provide practitioners with the most direct and essential answers to the questions that concern them; therefore, this method was chosen in our study [33]. This section was performed according to the Standards for Reporting Qualitative Research (SRQR) [34]. The COM-B model includes three domains: capability, opportunity and motivation, which interact with one another to enable behaviour to occur [29]. The TDF includes fourteen domains that can be condensed to fit the three constructions of the COM-B model, as follows: capability (knowledge, cognitive and interpersonal skills, memory, attention and decision processes, behavioural regulation, and physical skills); opportunity (social influences, environmental context, and resources); and motivation (reinforcement, optimism, emotions, social/professional role and identity, beliefs about capabilities, beliefs about consequences, goals, and intentions) [30].

Participants and setting

The present study employed a purposive sampling approach to recruit participants [35].

From May to August 2021, two researchers (CDD and SJ) recruited participants by distributing a recruitment advertisement. Those who agreed to participate in the study were assessed by the researchers to determine the representativeness of their characteristics. Potential participants were provided with information about the study and asked to complete a written informed consent form. The sample size depended on data saturation. The inclusion criteria were as follows: (1) individuals with MetS who met the diagnostic criteria of the 2009 Joint Scientific Statement [1], (2) individuals without food allergies or intolerances, and (3) individuals who were ≥18 years old. Exclusion criteria were the following: (1) people following specific diets; or (2) people who were unable to communicate effectively.

Data collection

According to the existing literature, the three constructs of the COM-B system, and the fourteen domains of the TDF, our research team developed a semi-structured interview guide (Table 1). Through conducting pre-interviews with three patients, the interview guide was further modified and improved. The interview did not strictly follow the present interview guide but was guided by the thoughts and perspectives of the participants. The first author (CDD) conducted and recorded these interviews using a digital recording tool with the permission of the participants. The digital recording was transcribed verbatim within 24 h after the end of the first interview. The study would continue until no new themes emerged, indicating thematic saturation, at which point data collection would stop [36].

Table 1. Interview schedule.

COM–B	TDF	Question
Psychological capability	Knowledge	How much do you know about metabolic syndrome? (e.g. diagnosis)
		What is your understanding of a healthy diet?
	Behavioural regulation	How do you ensure that you are eating healthy foods?
	Memory, attention and decision process	To what extent is eating a diet to relieve metabolic syndrome something you normally do?
Physical capability	Skills	What skills do you think are needed to maintain a healthy diet?
		How much do you know about how to cook healthy foods?
Social opportunity	Social influences	How do your parents or friends or other people help or hinder you eat healthy foods?
Physical opportunity	Environmental context and resources	What factors of work and/or home environment support or hinder your maintenance of a healthy diet?
Reflective motivation	Social/professional role and identity	To what extent do you see maintaining a healthy diet as part of your role?
		To what extent is maintaining a healthy diet accepted by your friends and family?
	Belief about capabilities	How confident do you feel about maintaining a healthy diet?
	Beliefs about consequences	What are the benefits of maintaining a healthy diet?
	Optimism	How confident are you that maintaining a healthy diet will have a positive outcome?
		How confident are you that you can overcome the barriers you face to maintaining a healthy diet?
	Intention	Do you intend to maintain a healthy diet every day? Why?
	Goals	What are your goals when you maintain a healthy diet?
Automatic motivation	Reinforcement	What would be an incentive to maintain healthy diet?
	Emotion	Discuss how you think maintaining healthy diet would make you feel?

Data analysis

The interview transcript was analysed using Braun and Clarke’s six-step framework [37]: (1) becoming familiar with the data; (2) generating initial codes; (3) searching for themes; (4) reviewing themes; (5) defining and naming themes; and (6) writing up the final report. First, CDD listened to and transcribed the audio recordings; checked the transcriptions; and read, listened to again, and reread the final transcripts to become familiar with the data. Simultaneously, CDD took notes for forming hidden codes. Second, the initial codes were developed. Next, the codes were checked and grouped to formulate themes by two researchers (CDD and ZH). These themes were then mapped into the three COM-B model constructs and fourteen TDF domains by two researchers (CDD and SJ). Disagreements and ambiguities were resolved by discussion with a third researcher (YZH).

Rigour

In the process of the data analysis, our research team adhered to the criteria for authenticity, credibility, criticality, and integrity [38]. Authenticity was ensured by two researchers (CDD and YZH) who checked the accuracy of the interview transcriptions. To maintain credibility, our research team discussed any differences related to methodological issues and data analysis. Criticality and integrity were attained through rigorous evaluation of each research decision and conscientious reflection on the researcher’s biases and their potential impact on the research. In addition, the members of our research team are PhD candidates or holders in nursing and have participated several times in qualitative research workshops. These experiences have equipped them with the essential skills of organizing, coding, and analysing textual data. Their experience was instrumental in ensuring the rigor of this research.

Stage 2: identifying intervention options

Steps 5 and 6: identifying IFs and policy categories

According to Michie et al. [29], the relevant IFs and policy categories should be identified based on the results of the COM-B model and TDF diagnosis. The IFs included education, training, restriction, persuasion, incentivization, coercion, modelling, environmental restructuring, and enablement [29]. All the IFs were then evaluated according to the APEASE (affordability, practicability, effectiveness and cost-effectiveness, acceptability, side effects/safety, and equity) criteria [29]. Affordability refers to whether the cost of the proposed intervention fits within the budget. Practicality assesses whether the intervention is delivered as intended to the target population. Effectiveness measures the impact of the intervention in achieving desired objectives in a real-world context. Cost-effectiveness evaluates the ratio of effect to cost. Acceptability gauges the appropriateness of the intervention according to relevant stakeholders (public, professional, and political). Side effects/safety considerations involve assessing potential unwanted effects or unintended consequences of the intervention. Equity examines how the intervention may impact disparities in standard of living, wellbeing, or health among different sectors of society [29]. After the IFs were identified, policy categories—including communication/marketing, guidelines, fiscal measures, regulation, legislation, environmental/social planning, and service provision—would normally be determined to assist in the delivery of the identified IFs in accordance with the BCW framework. However, our study was not related to changing policies on diets; therefore, policy categories were not addressed.

Stage 3: identify content and intervention options

Step 7: identify BCTs

Once the relevant IFs were determined, the intervention designers, who were PhD candidates or holders in nursing and who focused on behavioural change in patients with chronic disease, used the BCTTv1 to identify and link the most frequently used BCTs to each IF [29,31]. Furthermore, to identify any additional BCTs, a matrix was applied to link the fifty-nine BCTs from the BCTTv1 to the twelve TDF domains [39]. Additionally, the APEASE criteria were used to assess and narrow the potentially appropriate BCTs by two researchers [29]. If two researchers gave the same reasons for the inclusion and exclusion of BCTs according to the APEASE criteria, which meant that they had the same choice of BCTs, the BCTs would be finalised. Any disagreements were resolved through constructive discussions within our research team. After the BCTs were identified, our research team translated the BCTs into potential intervention content.

Step 8: identify the model of delivery

The delivery model referred to the method(s) of intervention administration, such as self-help, individual, group, telephone, or other modalities [40]. Before deciding the most appropriate delivery mode, the full range of possible delivery models must be considered, including face-to-face, TV, internet, app, and mobile phone text [29]. The delivery model was determined after considering the target behaviour, target population, and specific setting. Specifically, the delivery model was selected based on previous experience with diet intervention development. Additionally, the APEASE criteria were considered to assess the delivery model [29]. Disagreements were discussed within our research team.

Expert consultation

Following the completion of all stages, we synthesised the key findings from each stage. The intervention materials, including the content and format, were sent via email to a panel of twelve experts with diverse academic backgrounds. This panel consisted of advanced nursing practitioners, behavioural science experts, management scientists, and general physicians. Each expert reviewed the materials individually and shared their feedback and comments. After a period of two weeks, all the feedback and comments were collected. Our research team rigorously scrutinised and deliberated on each feedback point and subsequently refined the intervention content and format based on expert input. The review process aimed to ensure the intervention’s relevance, validity, efficacy, and acceptability across multiple disciplines, perspectives, and contexts.

Translating findings into WeChat mini program features

Our research team and software engineers cooperated to convert the intervention content into WeChat mini program features. This collaboration involves the integration of the research team’s expertise in intervention design and evaluation with the software engineers’ skills in programming, user interface design, and quality assurance. The resulting program should not only deliver the intervention content but also provide interactive features that promote user engagement, monitoring, and feedback. Moreover, the program’s design should also consider factors such as accessibility, usability, security, and scalability to ensure its effectiveness, efficiency, and robustness. The research team participates in the entire process of program development, maintains real-time communication with software engineers, and ensures that the program’s features meet the diet needs of patients with MetS.

Ethical consideration

Our study complied with the Helsinki Declaration and was approved by the Ethics Committee of Sir Run Run Shaw Hospital, Zhejiang University School of Medicine (grant no. 20210220-32). All participants provided voluntary and informed consent by signing consent forms before commencing the research. The participants were informed that their data was confidential. The first author (CDD) stored the obtained electronic and paper data in a password-protected computer file and a locked drawer, respectively.

Results

Step 1: define the problem in behavioural terms

The systematic review found that a healthy diet formed the core of MetS treatment [41]. Therefore, a healthy diet should not be ignored in MetS management. Through a literature search, evidence was found that low adherence to dietary recommendations was a serious problem among people with MetS [42,43]. Additionally, our research team conducted an empirical survey to understand diet behaviour in a general university hospital in Hangzhou, Zhejiang. The empirical study included 241 individuals with MetS. The results showed that the median diet score was 18.0 (interquartile range [IQR] 17.0–20.0), and 61% of the 241 individuals scored no more than 3.0 points in diet, indicating that individuals did not score high [10]. Hence, we identified the behavioural problem as insufficient adherence to dietary recommendations.

Step 2: select target behaviour

Much evidence is available about how to improve dietary compliance among individuals with MetS. The American Heart Association conference recommended a range of target behaviours, such as a low-fat, high-fibre diet to reduce energy intake by 2142–4184 kJ/d (500–1000 kcal/d); a low intake of saturated and trans-fat and cholesterol; an increased intake of fruit, vegetables, and whole grains; and reduced consumption of simple sugars [44,45]. Bach-Faig et al. [46] suggested a low-to-moderate intake of dairy products, fish, and poultry; a high intake of olive oil and plant foods; a low consumption of red meat and sweets; and a moderate consumption of alcohol. Studies included in the systematic review evaluating the relationship between diet patterns and metabolic syndrome in adult subjects proposed an increased intake of fruit, poultry, fish, vegetables, whole grains, legumes, nuts, and low-fat dairy products to maintain healthy diet patterns [47]. In addition, an international panel pointed out that an increased intake of unsaturated fat—primarily from olive oil (range of 20-40 g/d), a variety of legumes, cereals (whole grains), fruits, vegetables, fish, nuts, and dairy products—would be beneficial for preventing and managing MetS [9]. Considering the following four criteria: (i) how much of an impact changing the behaviour would have on the desired outcome, (ii) how likely it is that the behaviour can be changed, (iii) how likely it is that the behaviour would have a positive or negative impact on other related behaviours, and (iv) how easy it would be to measure the behaviour [29], we defined the target behaviours as individuals with MetS eating healthily as specified in public health guidelines.

Step 3: specify target behaviour

The target behaviour needed to be specified by detailing who should perform the behaviour, when, where, how, and with whom, which are presented in Table 2.

Table 2. Specifying the target behavior.

Target behavior	Healthy dietary adherence
Who needs to perform the behavior?	People with metabolic syndrome
When will they do it?	At meal times
Where will they do it?	At home and outside of the home
How often will they do it?	Everyday
With whom will they do it?	Individual or group

Step 4: what needs to change?

Participating in the study were twenty-two adults with MetS. The demographic and clinical characteristics can be found in a supplementary file (see Supplementary File 1). The behavioural analysis revealed the barriers and facilitators of target behaviours categorised into the COM–B model and TDF domains. In our study, 39 themes were identified based on the interview findings, including fifteen barriers, twenty-three facilitators, and one that could be classified as both (see Table 3). Of the 39 extracted themes, 38.5% fell into the reflective motivation category (n = 15), 30.8% fell into the psychological capability category (n = 12), 12.8% fell into the physical opportunity category (n = 5), 7.7% fell into the automatic motivation category (n = 3), 7.7% fell into the social opportunity category (n = 3) and 2.6% fell into the physical capability category (n = 1). Furthermore, the most commonly coded TDF domains were memory, attention, and decision processes (n = 7; 17.9%); environmental context and resources (n = 5; 12.8%); beliefs about consequences (n = 5; 12.8%); and beliefs about capabilities (n = 4; 10.3%). The remaining findings (46.2%) were coded into ten other TDF domains. The barriers that have been identified require modification and adjustment: perceived poor knowledge about a healthy diet, not monitoring food, eating out frequently and eating more, not being accustomed to eating breakfast, eating fast, preferring sweets, preferring red meat, not being accustomed to eating whole grains, having a negative peer influence, having family meals that are dense in oil and fat, lacking time, experiencing hot weather, having work meals that are dense in oil, and lacking self-control. The majority of the barriers were linked to psychological capability and reflective motivation. Table 3 displays the results of the behavioural analysis.

Table 3. Behavioural analysis of the dietary recommendation adherence. The subthemes were used to develop the intervention.

COM–B	TDF	Subtheme	Quote	Barrier or facilitator
Psychological capability	Knowledge	Limited knowledge about the diagnosis of metabolic syndrome	‘I do not know about metabolic syndrome, and it is the first time I have heard of it.’ (P18)	Barrier
			‘When the metabolism is good, bowel movements are regular. When the metabolism is not good, bowel movements are irregular.’ (P3)
		Perceived good knowledge about a healthy diet	‘Eating more vegetables and less meat is a healthy diet. A high-oil, high-salt, and high-fat diet is unhealthy.’ (P4)	Facilitator
			‘I do not like a high-oil and high-fat diet because it is unhealthy. Vegetables and high-protein foods are relatively healthy.’ (P7)
		Perceived poor knowledge about a healthy diet	‘The meals are all healthy. Actually, I do not know what a healthy diet or an unhealthy diet is.’ (P17)	Barrier
			‘Do not know what a healthy diet is. Can you tell me about it?’ (P12)
	Behavioural regulation	Self-devised strategies to maintain a healthy diet	‘We do not eat fried foods.’ (P5)	Facilitator
			‘We do not eat fried foods and do not eat out. We usually eat fresh foods.’ (P20)
		Not monitoring food	‘I do not record the varieties of foods. I do not have that habit.’ (P16)	Barrier
	Memory, attention and decision process	Maintaining a healthy diet as a habit	‘My diet has always been healthy, and I have adhered to it for 5 or 6 years. My mother had diabetes and died in her 70 s because she could not control her mouth. This is also a lesson for me.’ (P5)	Facilitator
		Eating out frequently and eating more	‘Now I go out for dinner 4 times a week. Last year it was 5 times a week. It’s easier to eat more when going out.’ (P19)	Barrier
			‘The food eaten at the dinner party is more than twice the food eaten in the home because you have to drink alcohol at a dinner party. You will eat more food as soon as you drink alcohol.’ (P18)
		Not being accustomed to eating breakfast	‘Basically, I have not eaten breakfast for more than 20 years. We do not have the habit of eating breakfast.’ (P22)	Barrier
		Eating fast	‘I finish my meal in 5 min at noon.’ (P19)	Barrier
		Preferring sweets	‘I love sweets. I usually prepare desserts for my son, and I eat some. Sometimes I drink milk tea with my daughter.’ (P16)	Barrier
		Preferring red meat	‘I like red meat, and I do not like vegetarian diets.’ (P13)	Barrier
		Not being accustomed to eating whole grains	‘I am not accustomed to eating whole grains; for example, I do not eat soba noodles.’ (P4)	Barrier
Physical capability	Skills	Able to cook healthy foods	‘I can cook healthy foods by myself without using many seasonings.’ (P9)	Facilitator
			‘I can cook healthy food, but considering the tastes of the whole family, I still cook a little greasy.’ (P16)
Social opportunity	Social influences	Having a positive family influence	‘The son and daughter-in-law have a low-oil and low-salt diet. We all eat meals together, so we also have a low-oil and low-salt diet.’ (P3)	Facilitator
		Receiving support from family members	‘My father suffered a stroke due to high blood pressure. He paid attention to his diet later and taught me how to maintain good eating habits.’ (P17) ‘I have diabetes. My family does not put sugar in their cooking for me.’ (P9)	Facilitator
		Having a negative peer influence	‘The people around me eat a lot and drink a lot. As soon as they tell me to eat out, I will agree with them.’ (P18)	Barrier
Physical opportunity	Environmental context and resources	Having family meals that are dense in oil and fat	‘My mother-in-law’s cooking is greasy…. All of our family likes eating meat and does not like eating vegetables. Therefore, I eat more meat dishes.’ (P15)	Barrier
		Lacking time	‘My family is in business, and I do not have time to cook at noon, so we usually eat out.’ (P11)	Barrier
		Experiencing hot weather	‘If the weather is hot, we will eat out.’ (P10)	Barrier
		Having work meals that are dense in oil	‘The company’s meals are relatively greasy.’ (P21)	Barrier
		Having a good family dietary environment	‘We (our family) have a low-oil and low-salt diet and do not eat overnight meals. Olive oil is used for cooking; the food ingredients are very fresh.’ (P1)	Facilitator
Reflective motivation	Social/professional role and identity	Prioritising a healthy diet	‘I put healthy eating first. I lost 10 kilograms through a healthy diet, so it is very important.” (P8)	Facilitator
		Being responsible for maintaining a healthy diet	‘My family is supportive of my healthy diet…but in general I still rely on myself.’ (P8)	Facilitator
	Belief about capabilities	Being determined to sustain a healthy diet	‘It is easy to maintain a healthy diet with determination.’ (P16)	Facilitator
		Needing time to sustain healthy changes	‘It is difficult to say whether maintaining a healthy diet is difficult or easy. It takes time to change, and it is impossible to change quickly.’ (P4)	Facilitator
		Ease	‘I have been maintaining a healthy diet for more than 10 years without difficulty. Because I have diabetes, I want to reduce this glucose level.’ (P9)	Facilitator
		Lacking self-control	‘I still have difficulties maintaining a healthy diet. I cannot control my mouth.’ (P15)	Barrier
	Beliefs about consequences	Being in good spirits	‘My spirit has improved.’ (P17)	Facilitator
		Declining metabolic parameters and anthropometric indicators	‘My blood pressure, weight, and triglyceride levels have dropped.’ (P18)	Facilitator
		Living longer	‘I will live longer.’ (P6)	Facilitator
		Being happy	‘I am definitely happy. If my blood glucose level is high, my temper will be slightly grumpier.’ (P22)	Facilitator
		Decreasing stress	‘My psychological pressure will be less.’ (P18)	Facilitator
	Optimism	Having confidence maintaining a healthy diet	‘I will cook less food in the future to avoid eating more due to reluctance to waste. I am confident I can maintain a good diet.’ (P7)	Facilitator
		Reducing the frequency of eating out	‘All of our friends have agreed that we will have fewer dinners together. I will not go to general dinners in the future.’ (P11)	Facilitator
	Intention	Intending and adhering a healthy diet	‘When I eat out in the future, I will eat 6 or 7 points full.’ (P4)	Facilitator
	Goals	Having clear goals that drive intrinsic motivation	‘In order not to cause trouble to the son and daughter, I should eat healthy foods.’ (P2)	Facilitator
Automatic motivation	Reinforcement	Understanding the severity of metabolic syndrome	‘If my metabolic parameters are serious, I will control my mouth.’ (P15)	Facilitator
	Emotion	Being positive	‘When I see a reduction in blood glucose through healthy eating, I am very happy.’ (P15)	Facilitator
		Being emotionally affected	‘It has little effect on mood, and I have been in a better mood.’ (P16)	Neither

Steps 5 and 6: identifying IFs and policy categories

According to the BCW, we linked the behavioural assessment with the IFs (see Table 4). Specifically, six IFs were selected for this intervention design: education, training, persuasion, enablement, environmental restructuring, and modelling. Restriction was not acceptable for individuals with MetS or health care professionals, as rules often require legislative changes to be enforceable and acted upon.

Table 4. Linking the results of the behavioral assessment with the IFs.

COM-B component	TDF	IF	Does the IF meet the APEASE criteria?
Psychological capability	Knowledge	Education	Yes
	Behavior regulation	Education	Yes
		Training	Yes
		Modelling	Yes
		Enablement	Yes
	Memory, attention and decision process	Training	Yes
		Environmental restructuring	Yes
		Enablement	Yes
Social opportunity	Social influences	Restriction	No, it is not acceptable
		Environmental restructuring	Yes
		Modelling	Yes
		Enablement	Yes
Physical opportunity	Environmental context and resources	Training	Yes
		Restriction	No, it is not acceptable
		Environmental restructuring	Yes
		Enablement	Yes
Reflective motivation	Belief about capabilities	Education	Yes
		Persuasion	Yes
		Modelling	Yes
		Enablement	Yes

Step 7: identify BCTs

As shown in Table 5, we identified twelve BCTs that could be used in the intervention. The selected BCTs could serve more than one IF. The identified BCTs included information about health consequences (5.1), prompts/cues (7.1), self-monitoring of behaviour (2.3), demonstration of the behaviour (6.1), instruction on how to perform the behaviour (4.1), social support (practical) (3.2), restructuring the social environment (12.2), focusing on past success (15.3), verbal persuasion about capability (15.1), feedback on behaviour (2.2), goal setting (behaviour) (1.1), and reviewing behaviour goal(s) (1.5).

Table 5. Identification of the possible BCTs that could be used in the intervention.

TDF domains	IF identified	BCTs identified	Does the BCT meet the APEASE criteria?
Knowledge	Education	Information about social and environmental consequences (5.3)	No, it is ineffective or unacceptable
		Information about health consequences (5.1)	Yes
		Feedback on behavior (2.2)	No, it is ineffective or unacceptable
		Feedback on outcome(s) of the behavior (2.7)	No, it is ineffective or unacceptable
		Prompts/cues (7.1)	No, it is ineffective or unacceptable
		Self-monitoring of behavior (2.3)	No, it is ineffective or unacceptable
		Biofeedback (2.6)	No, it is ineffective or unacceptable
		Antecedents (4.2)	No, it is ineffective or unacceptable
Behavior regulation	Education	Information about social and environmental consequences (5.3)	No, it is ineffective or unacceptable
		Information about health consequences (5.1)	Yes
		Feedback on behavior (2.2)	No, it is ineffective or unacceptable
		Feedback on outcome(s) of the behavior (2.7)	No, it is ineffective or unacceptable
		Prompts/cues (7.1)	Yes
		Self-monitoring of behavior (2.3)	No, it is ineffective or unacceptable
	Training	Demonstration of the behavior (6.1)	No, it is ineffective or unacceptable
		Instruction on how to perform a behavior (4.1)	No, it is ineffective or unacceptable
		Feedback on the behavior (2.2)	No, it is ineffective or unacceptable
		Feedback on outcome(s) of behavior (2.7)	No, it is ineffective or unacceptable
		Self-monitoring of behavior (2.3)	No, it is ineffective or unacceptable
		Behavioral practice/rehearsal (8.1)	No, it is ineffective or unacceptable
	Modelling	Demonstration of the behavior (6.1)	No, it is ineffective or unacceptable
	Enablement	Social support (practical) (3.2)	No, it is ineffective or unacceptable
		Goal setting (behavior) (1.1)	No, it is ineffective or unacceptable
		Goal setting (outcome) (1.3)	No, it is ineffective or unacceptable
		Adding objects to the environment (12.5)	No, it is not affordability or acceptable
		Problem solving (1.2)	No, it is ineffective or unacceptable
		Action planning (1.4)	No, it is ineffective or unacceptable
		Self-monitoring of behavior (2.3)	Yes
		Restructuring the physical environment (12.1)	No, it is ineffective or unacceptable
		Review behavior goal(s) (1.5)	No, it is ineffective or unacceptable
		Review outcome goal(s) (1.7)	No, it is ineffective or unacceptable
Memory, attention and decision process	Training	Demonstration of the behavior (6.1)	Yes
		Instruction on how to perform a behavior (4.1)	Yes
		Feedback on the behavior (2.2)	No, it is ineffective or unacceptable
		Feedback on outcome(s) of behavior (2.7)	No, it is ineffective or unacceptable
		Self-monitoring of behavior (2.3)	No, it is ineffective or unacceptable
		Behavioral practice/rehearsal (8.1)	No, it is ineffective or unacceptable
	Environmental restructuring	Adding objects to the environment (12.5)	No, it is not affordability or acceptable
		Prompts/cues (7.1)	No, it is ineffective or unacceptable
		Restructuring the physical environment (12.1)	No, it is ineffective or unacceptable
	Enablement	Social support (practical) (3.2)	Yes
		Goal setting (behavior) (1.1)	No, it is ineffective or unacceptable
		Goal setting (outcome) (1.3)	No, it is ineffective or unacceptable
		Adding objects to the environment (12.5)	No, it is not affordability or acceptable
		Problem solving (1.2)	No, it is ineffective or unacceptable
		Action planning (1.4)	No, it is ineffective or unacceptable
		Self-monitoring of behavior (2.3)	No, it is ineffective or unacceptable
		Restructuring the physical environment (12.1)	No, it is ineffective or unacceptable
		Review behavior goal(s) (1.5)	No, it is ineffective or unacceptable
		Review outcome goal(s) (1.7)	No, it is ineffective or unacceptable
Social influences	Environmental restructuring	Adding objects to the environment (12.5)	No, it is not affordability or acceptable
		Prompts/cues (7.1)	Not effective or acceptable
		Restructuring the physical environment (12.1)	Not effective or acceptable
		Restructuring the social environment (12.2)	Yes
	Modelling	Demonstration of the behavior (6.1)	Yes
	Enablement	Social support (practical) (3.2)	Yes
		Goal setting (behavior) (1.1)	No, it is ineffective or unacceptable
		Goal setting (outcome) (1.3)	No, it is ineffective or unacceptable
		Adding objects to the environment (12.5)	No, it is not affordability or acceptable
		Problem solving (1.2)	No, it is ineffective or unacceptable
		Action planning (1.4)	No, it is ineffective or unacceptable
		Self-monitoring of behavior (2.3)	No, it is ineffective or unacceptable
		Restructuring the physical environment (12.1)	No, it is ineffective or unacceptable
		Review behavior goal(s) (1.5)	No, it is ineffective or unacceptable
		Review outcome goal(s) (1.7)	No, it is ineffective or unacceptable
		Social support (emotional) (3.3)	No, it is ineffective or unacceptable
		Social reward (10.4)	No, it is ineffective or unacceptable
		Social comparison (6.2)	No, it is ineffective or unacceptable
		Social support or encouragement (general) (3.1)	No, it is ineffective or unacceptable
		Information about others’ approval (6.3)	No, it is ineffective or unacceptable
		Vicarious reinforcement (16.3)	No, it is ineffective or unacceptable
		Identification of self as role model (13.1)	No, it is ineffective or unacceptable
Environmental context and resources	Training	Demonstration of the behavior (6.1)	No, it is ineffective or unacceptable
		Instruction on how to perform a behavior (4.1)	No, it is ineffective or unacceptable
		Feedback on the behavior (2.2)	No, it is ineffective or unacceptable
		Feedback on outcome(s) of behavior (2.7)	No, it is ineffective or unacceptable
		Self-monitoring of behavior (2.3)	No, it is ineffective or unacceptable
		Behavioral practice/rehearsal (8.1)	No, it is ineffective or unacceptable
	Environmental restructuring	Adding objects to the environment (12.5)	No, it is not affordability or acceptable
		Prompts/cues (7.1)	No, it is ineffective or unacceptable
		Restructuring the physical environment (12.1)	No, it is ineffective or unacceptable
		Discriminative (learned) cue (7.2)	No, it is ineffective or unacceptable
		Restructuring the social environment (12.2)	Yes
		Avoidance / changing exposure to cues for the behavior (12.3)	No, it is ineffective or unacceptable
	Enablement	Social support (practical) (3.2)	Yes
		Goal setting (behavior) (1.1)	No, it is ineffective or unacceptable
		Goal setting (outcome) (1.3)	No, it is ineffective or unacceptable
		Adding objects to the environment (12.5)	No, it is not affordability or acceptable
		Problem solving (1.2)	No, it is ineffective or unacceptable
		Action planning (1.4)	No, it is ineffective or unacceptable
		Self-monitoring of behavior (2.3)	No, it is ineffective or unacceptable
		Restructuring the physical environment (12.1)	No, it is ineffective or unacceptable
		Review behavior goal(s) (1.5)	No, it is ineffective or unacceptable
		Review outcome goal(s) (1.7)	No, it is ineffective or unacceptable
Belief about capabilities	Education	Information about social and environmental consequences (5.3)	No, it is ineffective or unacceptable
		Information about health consequences (5.1)	Yes
		Feedback on behavior (2.2)	No, it is ineffective or unacceptable
		Feedback on outcome(s) of the behavior (2.7)	No, it is ineffective or unacceptable
		Prompts/cues (7.1)	No, it is ineffective or unacceptable
		Self-monitoring of behavior (2.3)	No, it is ineffective or unacceptable
		Focus on past success (15.3)	Yes
		Verbal persuasion about capability (15.1)	Yes
	Persuasion	Credible source (9.1)	No, it is ineffective or unacceptable
		Information about social and environmental consequences (5.3)	No, it is ineffective or unacceptable
		Information about health consequences (5.1)	No, it is ineffective or unacceptable
		Feedback on behavior (2.2)	Yes
		Feedback on outcome(s) of the behavior (2.7)	No, it is ineffective or unacceptable
	Modelling	Demonstration of the behavior (6.1)	No, it is ineffective or unacceptable
	Enablement	Social support (practical) (3.2)	No, it is ineffective or unacceptable
		Goal setting (behavior) (1.1)	Yes
		Goal setting (outcome) (1.3)	No, it is ineffective or unacceptable
		Problem solving (1.2)	No, it is ineffective or unacceptable
		Action planning (1.4)	No, it is ineffective or unacceptable
		Self-monitoring of behavior (2.3)	No, it is ineffective or unacceptable
		Restructuring the physical environment (12.1)	No, it is ineffective or unacceptable
		Review behavior goal(s) (1.5)	Yes
		Review outcome goal(s) (1.7)	No, it is ineffective or unacceptable

Step 8: identify model of delivery

A systematic review reported that the mHealth-based intervention effectively improved individuals’ eating behaviours [17]. A WeChat mini program is an important component of mHealth. It does not require download or uninstallation, is convenient and inexpensive, and allows users to receive health services without temporal or spatial limitations [48]. The features of the WeChat mini program met the APEASE criteria. As a result, our research selected a WeChat mini program as the delivery method.

Expert consultation

A summary of the final included COM-B, TDF, barriers, IFs, BCTs, and intervention content is presented in Table 6. As shown in step four, most barriers were related to the COM-B model of capability and opportunity, and the TDF domains of knowledge, memory, attention, and decision processes, and environmental context and resources. Specifically, to increase knowledge about MetS, it may be useful to provide adults with MetS with health education about the diagnosis, health consequences, or diet knowledge of MetS via the WeChat mini program. In terms of unhealthy diet habits, instruction on how to perform a behavior, demonstration of the behavior, and social support were also identified as valuable BCTs. Provision of instruction on how to perform a healthy diet, observable samples of individuals who maintain healthy diets, and suggestions on how to eat healthily could improve individuals’ diet habits. Emphasis on restructuring the social environment and providing social support in this intervention, such as advising family members to maintain a healthy diet and advising individuals to seek help from others via the WeChat mini program, may create a good diet atmosphere for individuals with MetS. Furthermore, the other intervention components included providing via the WeChat mini program: (i) suggestions or feedback about the target behaviour; (ii) reminders for adults with MetS to record the food varieties and/or amount they ate; (iii) the ability for individuals with MetS to set their diet goals or plans per day as specified in public health guidelines and review the goals or plans; and (iv) positive messages telling the person that they can maintain a healthy diet. The experts’ feedback on the intervention content and format encompassed the following aspects:

Table 6. Combined link between COM-B model, TDF domains, IFs, BCTs and intervention content.

COM-B	TDF domains	Sub-theme/Barriers	IFs identified	BCTs identified	Intervention content
Psychological capability	Knowledge	Limited knowledge about the diagnosis of metabolic syndrome	Education	Information about health consequences (5.1)	Provide information about the definition, drawbacks, and diet knowledge of metabolic syndrome via the WeChat mini program.
		Perceived poor knowledge about a healthy diet
	Behavior regulation	Not monitoring food	Education	Information about health consequences (5.1)	Provide information about the drawbacks of unhealthy eating behavior via the program.
				Prompts/cues (7.1)	Set up reminders for individuals to record the food varieties and (or) amounts they ate via the program.
			Enablement	Self-monitoring of behavior (2.3)	Record the food varieties and (or) amounts via the program.
	Memory, attention and decision process	Eating out frequently and eating more	Training	Demonstration of the behavior (6.1)	Provide observable examples of individuals who maintain healthy diets via the program.
		Not being accustomed to eating breakfast		Instruction on how to perform a behavior (4.1)	Provide instructions about diet habits to promote healthy eating via the program.
		Eating fast	Enablement	Social support (practical) (3.2)	Advise relatives to support individual healthy eating, such as encouraging individuals to eat breakfast and whole grains via the program.
		Preferring sweets
		Preferring red meat
		Not being accustomed to eating whole grains
Social opportunity	Social influences	Having a negative peer influence	Environmental restructuring	Restructuring the social environment (12.2)	Advise individuals to minimise the time spent with friends who eat more to reduce food intake via the program.
			Modelling	Demonstration of the behavior (6.1)	Provide observable examples of individuals who maintain healthy diets via the program.
			Enablement	Social support (practical) (3.2)	Advise family members to encourage individuals to maintain healthy diets via the program.
Physical opportunity	Environmental context and resources	Having family meals that are dense in oil and fat	Environmental restructuring	Restructuring the social environment (12.2)	Advise family members to maintain a healthy diet via the program.
		Lacking time	Enablement	Social support (practical) (3.2)	Advise individuals to seek for help from others via the program. Provide advice on hot weather, e.g. prepare food when the climate is cool. Provide advice on maintaining a low-oil diet at work, e.g. eat meals brought from home.
		Experiencing hot weather
		Having work meals that are dense in oil
Reflective motivation	Belief about capabilities	Lacking self-control	Education	Information about health consequences (5.1)	Introduce the dangers of an unhealthy diet via the program.
				Focus on past success (15.3)	Advise individuals to think about previous successes in maintaining a healthy diet via the program.
				Verbal persuasion about capability (to boost self-efficacy) (15.1)	Tell the individual that they can maintain a healthy diet via the program.
				Feedback on behavior (2.2)	Provide informative feedback to individuals about their diet plans via the program.
			Enablement	Goal setting (behavior) (1.1)	Set daily diet goals and plans as specified in the public health guidelines via the program.
				Review behavior goal(s) (1.5)	Allow individuals to review their behavioural goals via the program.

‘Advise individuals to minimize the time spent with friends who eat more to reduce food intake via the program.’ should be revised as ‘It is recommended that patients minimize their social activities at night as much as possible, or schedule them at noon if necessary.’

‘Provide information on the definition, drawbacks, and diet knowledge related to metabolic syndrome via the program.’ should be revised as ‘Provide information about the definition, etiology, drawbacks, and individualized diet knowledge of metabolic syndrome via the program.’

‘Share your own diet status with others via the program’ should be added.

‘Provide a healthy diet and cooking assistance.’ should be added.

‘The implementation process of urging a scientific and reasonable diet.’ should be added.

Translating findings into WeChat mini program features

As shown in Figure 3, the program developed by the research team in collaboration with software engineers consists of three main modules: ‘Health Assessment’, ‘My Records’, and ‘My Homepage’, which could be subdivided into ten functional modules.

Figure 3. Functional modules of the WeChat mini program.

The ‘Health Assessment’ module: users need to fill in their personal information, including name, age, gender, height, weight, WC, SBP, DBP, HDL-C, TG, FBG, and other indicators. Based on the collected data, the program provides individualized diet recommendations (Figure 4).

Figure 4. The ‘Health assessment’ interface.

The ‘My Records’ module includes three parts. Firstly, the ‘Daily Records’ section allows users to record their diet type and amount as a check-in each day. The program can intelligently generate individualized daily reports in chart form, including types and amounts of food intake, for patients to review the gap between their own diet and diet goals recommended by related guidelines (Figure 5). The second part is ‘Behaviour Review and Advice’, where users need to compare their diet type and amount against the program’s recommendations and select reasons not to follow diet recommendations. The program can then provide individualized solutions (Figure 6). In the third part, the ‘Recent Seven-Day Report’ section, users can view individualized charts generated by the system, which includes food types and amounts in the last 7 days, for patients to review the gap between their own diet and diet recommendations (Figure 6).

Figure 5. The ‘daily records’ interface.

Figure 6. The ‘behaviour review and advice’ interface and the ‘recent Seven-Day report’ interface.

The ‘My Homepage’ module includes six parts. The ‘Knowledge Encyclopedia’ section provides text and video information about MetS diagnosis, etiology, clinical manifestations, complications, exercise guidance, diet guidance, etc. Users can click on the corresponding content to read and save it (Figure 7). The ‘Health Consultation, Reminders, Help, and Feedback’ module provides individualized feedback according to patients’ questions and provides diet reminders. In addition, if users have suggestions or need help regarding the contents and use of the MetS health management mini program, they can leave a message for feedback. The team members will later reply (Figure 8). In the ‘Questionnaire Survey’ section: users can complete online assessments, including self-assessment questionnaires on health-promoting lifestyles, MetS prevention and treatment knowledge, health behaviour self-efficacy, quality of life, etc. The program calculates the questionnaire score and generates a report that is sent to the user’s phone or email (Figure 9). The ‘Operation Guide’ introduces the usage rules of this program. The ‘Disclaimer’ explains the conditions under which this program is not liable. The ‘Privacy Statement’ introduces the privacy protection of user information.

Figure 7. The ‘Knowledge encyclopedia’ interface.

Figure 8. The ‘Health Consultation, Reminders, Help, and feedback’ interface.

Figure 9. The ‘questionnaire survey’ interface.

In addition, within each module, patients can share their own diet management status with family and friends in order to increase compliance with healthy eating through external incentives or support.

Discussion

The BCW theory helps scientifically and systematically develop an individualized WeChat mini program-based dietary adherence intervention. To our knowledge, this is the first paper to specify the content and active ingredients of an individualized WeChat mini program-based behavioural change intervention using a systematic evidence- and theory-based method for individuals with MetS. According to the BCW, our study identified four components to the COM-B model, six TDF domains, six IFs, and twelve BCTs based on patients’ barriers to a dietary recommendation that could be used to formulate an individualized WeChat mini program-based behavioural change intervention program that aims to strengthen adherence to dietary recommendations among individuals with MetS in China. The findings can be used to provide guidance to improve dietary adherence, and ultimately improve the lives of people with MetS. Additionally, this methodology can serve as a reference for other researchers who are developing behavioural change interventions.

Developing an effective adherence intervention requires an in-depth identification of the key elements affecting the target behaviour. In our study, some barriers were identified based on the COM-B model and the TDF. In terms of psychological capability, the participants reported having poor MetS knowledge about their diagnosis and lacking knowledge about a healthy diet for managing MetS, which was in accordance with Wang et al.’s findings [49]. Moreover, we also found a phenomenon in which many participants ate out frequently and ate more. Most of the participants expressed that the aim of eating out was to establish a working relationship with customers. Additionally, other unhealthy diet habits were identified, such as eating sweets, eating red meat, eating fast, or not being accustomed to eating whole grains or breakfast. Previous research reported similar findings [50] as it aimed to explore the barriers and facilitators to adherence to the diet guidelines among children and unrelated adults and noted that the primary barriers encompassed a lack of cooking skills, poor eating habits, inadequate knowledge of recommendation/portion/health benefits, etc. In our study, nearly none of the patients recorded the varieties of food they consumed. As reported, the self-monitoring of unwanted behaviour could lead to a decrease in such behaviour [51]. Therefore, intervention strategies should encourage self-monitoring of unhealthy foods.

With respect to social opportunity, negative peer influence was a major barrier to a healthy diet. In contrast, positive peer support enhanced adherence to a healthy diet [52]. Additionally, physical opportunities, including family and work dietary atmospheres, needed improvement. Families were a support source through the role model of healthy eating [53]. In our study, some family meals were oil- and fat- dense. For these families, the diet patterns must be adjusted and a balanced diet must be maintained to create a good dietary environment. Moreover, to maintain a healthy diet at work, people did not eat out or in a canteen but ate a lunch they brought to work [54], which provided a reference for a future diet intervention design. In addition, our study reported that the lack of time and hot weather influenced participant adherence to dietary recommendations. Regarding reflective motivation, several of the participants expressed difficulties in adhering to a healthy diet. They stated that the main reason was a lack of self-control when maintaining a healthy diet. This finding fit well with the existing literature [55,56], which indicated that low adherence to dietary recommendations was associated with less self-control.

The most common IFs were education, persuasion, modelling, enablement, and environmental restructuring in a mHealth-based medication adherence intervention [57] and a healthy eating intervention [58,59], which was consistent with our study. Furthermore, our study identified twelve potential BCTs (of 93) aligned to six IFs to address the identified barriers. Similar to the findings of our study, Rohde et al. [59] identified fourteen BCTs (e.g. rewards, graded tasks, or self-monitoring) in the mHealth-based design aimed at improving eating habits. To increase vegetable consumption, Mummah et al. [60] included eighteen BCTs (e.g. social comparison, feedback, goal setting, and prompts/cues) to develop a theory-based mHealth application. Thus, BCT combinations were often used in interventions to improve diets. In future research, researchers should focus on which BCTs or combinations of BCTs are the most effective. The present study was not related to changing policy on diets. However, the evidence gathered from the intervention development process may offer valuable reference information pertaining to policy categories aimed at promoting behaviour changes.

Our study determined the intervention content based on the identified BCTs. To enhance its scientific and practical feasibility, expert consultations were conducted. Through expert consultation, the intervention content was enriched and the intervention model of delivery-a WeChat mini program was affirmed. The mini program is an important component of mHealth and is grounded in WeChat, which is a widely popular social networking app used in China, and it’s also considered to be one of the leading social networks worldwide. The program supports developers to implement individualized functions, and can provide customized contents and experiences according to the needs of patients. Participants may also access a WeChat mini program-based intervention without needing to log onto a particular website or download a new app [48]. Other advantages include cost-effectiveness, automation, and the convenience of accessing healthcare services anytime and anywhere [48].

We invited software engineers to design the WeChat mini program with more individualized elements based on the identified intervention content. The developed program can provide individualized dietary recommendations, behaviour modification solutions, and goal achievement feedback based on data uploaded by patients. Furthermore, health care professionals can offer individualized support to patients with MetS. A randomized controlled trial involving seven European countries reported that an internet-delivered intervention providing individualized nutrition advice yielded greater and more appropriate changes in dietary behaviour compared to non-individualized dietary advice [61]. Moreover, given that our study followed a comprehensive behavioural change intervention framework, the individualized WeChat mini program-based diet intervention will be promising in improving dietary adherence, thereby improving metabolic parameters and quality of life in individuals with MetS. However, whether the intervention works will be tested in the intervention trial.

Strengths

To date, there has been no study conducted to develop a systematic and individualized WeChat mini program-based diet intervention for adults with MetS using the BCW in a Chinese context. Specifically, the present study was performed following a strong theoretical underpinning, which contributed to a better understanding the behavioural change process [62] and uncovering positive effects [63]. The lack of a theoretical foundation is related to the limited effectiveness of interventions. While multiple behavioural models abound, the majority fail to assist in designing interventions or understanding behavioural change processes. The present research aims to address this gap by applying the BCW. An additional strength inherent in the present study is its deliberate adoption of a qualitative research design to explore in depth the determinants influencing the target behaviour. As a result, health care professionals could develop more comprehensive behavioural change intervention strategies. These barriers and facilitators also provided references for other studies exploring dietary adherence. Finally, the current study successfully addresses the limitations observed in previous studies, including limited individualized elements, time-consuming, high costs, and a lack of real-time feedback, by incorporating more individualized elements and utilizing a digital intervention. In the future, our research team will refine and upgrade the WeChat mini program and conduct a controlled intervention trial to demonstrate its impact on dietary adherence. The trial study has been registered with the Chinese Clinical Trial Registry (ChiCTR2100043877). If proven effective, health care professionals could offer the WeChat mini program to individuals with MetS in China, aiming to enhance dietary adherence and ultimately enhance their lives in China.

Limitations

However, this study also presents some weaknesses. First, all the participants were from Zhejiang Province. Although data saturation was attained, the applicability of this study to other contexts is limited. Future research will recruit more participants in different districts. Moreover, when the intervention designers used the BCW, subjectivity existed in the process of identifying the intervention content, such as using APEASE criteria. Therefore, subjectivity is a potential problem. Future research could address this issue by inviting a multidisciplinary team to participate in the process of developing interventions to obtain a wider range of views and suggestions. Additionally, given that some same BCTs were categorized to different IFs, their selection was challenging. Finally, the utilization of the BCW in the development of the WeChat mini program-based behavioural change intervention offers a systematic, theory-driven, and evidence-based approach, but the intervention design process was lengthy and time-consuming, which could potentially limit the efficiency of use.

Conclusions

Adherence to dietary recommendations is important, as it may predict the likelihood of MetS components and the risk of complications. This study outlined how an intervention to increase adherence to dietary recommendations in people with MetS was developed with the use of the BCW. In our study, an individualized WeChat mini program-based behavioural change intervention may increase adherence to dietary recommendations among individuals with MetS. Moving forward, our research team will refine and upgrade the WeChat mini program and then test the usability and effectiveness of the individualized WeChat mini program-based behavioural change intervention program.

Authors contributions

Dandan Chen: Conceptualization, methodology, formal analysis, writing-original draft, writing-review & editing. Jing Shao: Conceptualization, formal analysis, writing-review & editing. Hui Zhang: Conceptualization, methodology, formal analysis. Jingjie Wu: Conceptualization, revising the article. Erxu Xue: Methodology, writing-review & editing. Pingping Guo: Methodology, writing-review & editing. Nianqi Cui and Xiyi Wang: Data curation. Liying Chen and Zhihong Ye: The conception and design of the study, validation, supervision, formal analysis, writing - review & editing. All authors have read and agreed to the published version of the manuscript and Agreed to be accountable for all aspects of the work.

Acknowledgements

We sincerely appreciate the participating hospital and all individuals with MetS who participated in this study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The datasets generated and analysed in this study are not publicly available due to privacy concerns. However, they can be obtained from the corresponding author upon reasonable request.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [72004193].