Cultural adaptation of the mental health first aid guidelines for depression used in English-speaking countries for China: a Delphi expert consensus study

We used the Delphi method to elicit consensus on potential statements to be included in the mental health first aid guidelines for depression for mainland China. The Delphi method is an iterative multistage process, designed to transform the opinions of individual experts into group consensus [36]. By using this method to develop an evidence base to guide decisions, policymakers and practitioners can move beyond relying on their own experience and draw on the accumulated experience of a larger, expert group [37]. This method has been widely used in disparate fields, including in mental health research, for example in the development of mental health first aid guidelines for English-speaking countries and some LMICs [26, 28, 30, 38].

In this study, the Delphi expert consensus survey involved four stages: (1) questionnaire development for Round 1 of the Delphi survey; (2) panel identification and recruitment; (3) data collection over 3 rounds of survey; and (4) guidelines development.

The questionnaire for Round 1 was developed as follows: Firstly, the original statements endorsed from the English-language questionnaire were translated into Mandarin, with some of the translated statements modified to better reflect the Chinese context. For example, ‘to see a GP’ was replaced with ‘to see a mental health specialist’, as in China the number of GPs is limited and they typically lack the skills to manage people with mental health problems [39]; rather, mental health specialists are responsible for providing the majority of mental health services. In addition, all phrases referring to contacting a ‘mental health crisis team’ in the guidelines for English-speaking countries were revised to ‘contact the police’ as this is the common practice in China. The modification process involved one professional translator and four of the authors who are skilled in both English and Chinese (SL, WL, YH and WY).

There were 175 original statements included in 8 sections of the questionnaire for Round 1, which can be viewed in Table 1 of the Additional file. All statements in this questionnaire were rated for importance of inclusion in the mental health first aid guidelines for a member of the public in China to help someone who is experiencing depression.

We recruited two expert panels: one comprising mental health professionals and the other comprising consumers and carers. Mental health professionals were eligible to participate if they were psychiatrists, psychiatric nurses, psychotherapists or social workers, and had been involved in the clinical treatment/management of depression in a specialised mental health institute for at least 2 years. Evidence suggests that a panel size of 23 participants is necessary for stability of response characteristics in Delphi surveys [40]. Allowing for attrition, we therefore aimed to recruit 30 participants for each panel.

A purposive snowball sampling method was used to select participants. An initial recruitment advertisement sent out via email to personal contacts of researchers to potential professionals in the two specialised mental health institutions in Shanghai (Shanghai Mental Health Centre) and Suzhou (Suzhou Guangji Hospital). Participants were encouraged to send the email on to other eligible mental health professionals they knew.

Potential consumer and carer panellists were recruited from clinic sessions or public health lectures provided for people with affective disorders and their carers/families in the two specialised mental health institutions mentioned above. Consumers and carers were eligible to participate if they met the following criteria:

Panellists were told that their participation was voluntary and that their responses would only be reported at the aggregate level. Panellists were reimbursed a gift card valued at RMBҰ100 for completing at least the Round 1 survey, which took 55 min on average.

Recruited panel members were sent a link and a Quick Response (QR) code, both of which led them to an online questionnaire hosted by Questionnaire Star (Wen Juan Xing, https://​www.​wjx.​cn/​) via a computer or mobile phone (by WeChat – a commonly used mobile application for social interaction in China). Participants were instructed to rate how important the helping statements were to be included in the guidelines for providing mental health first aid to a person experiencing depression. In Round 1, panellists were also encouraged to provide comments on existing statements or to suggest new helping actions that were not covered in the questionnaire.

All comments collected were sorted, translated into English and then reviewed by the working group (SL, WL, YH, NR, WY). Suggestions that contained novel ideas were used to create new statements to be included in the questionnaire of the subsequent survey round. Statements from Round 1 that met the criteria to be re-rated (i.e. being rated as essential or important by 70–79% of either panel) were also included in the Round 2 questionnaire. The Round 3 questionnaire comprised statements presented in Round 2 but requiring re-rating in a further round. Statements that still did not achieve consensus after three rounds were not included in the guidelines.

Following the first two rounds, panellists were sent a report containing a summary of the overall ratings for the statements, as well as their ratings for each statement. This allowed the panellists to compare their ratings with the level of endorsement given by the group as a whole and to inform their future ratings for those statements that needed to be re-rated.

The correlation relationship between the statement endorsement rates from the two panels was measured by Spearman’s correlation coefficient using the STATA software (version 15).

Endorsed statements (i.e. those being rated as either essential or important by ≥80% of both panels) from all three rounds were compiled. Author WL drafted the guidelines by writing the list of endorsed statements into sections of connected text. Where possible, statements were combined and repetition deleted. Statements that received comments suggesting ambiguity in the interpretation of their meaning were re-worded to make them clear and easy to understand. The draft was then circulated to members of the working group (SL, WL, YH and WY) who were native Mandarin-speakers to finalise structure and wording, creating a set of guidelines that were written in plain Mandarin and could easily be followed by members of the public in China. A number of iterations were circulated and completed before the group agreed on the final text for the guidelines.

A total of 67 expert panellists (31% male) representing the two panels of mental health professionals (n = 37) and consumers and carers (n = 30) completed Round 1 in this Delphi study. The sociodemographic characteristics of participants are shown by panel in Table 1.

Of the 37 experts in the mental health professional panel, 25 were psychiatrists, 10 psychotherapists and 2 psychiatric nurses. Most of these professional panellists also conducted research and some of them had more than one role. For example, some psychiatrists were also qualified to work as a psychotherapist in their institutions. Of the 30 members of the consumer and carer panel, 16 had lived experience as a patient with depression, 5 as a carer and another 9 had lived experience of both.

The retention rates across rounds are shown by panel in Table 2. Overall, 70% of participants in Round 1 (n = 47) completed Round 2 and 48% (n = 32) completed Round 3. The health professional panel had higher retention rates than the consumer and carer panel in both rounds 2 and 3.

An overview of the 3 rounds of survey is provided in Fig. 1, while the full lists of statements in the questionnaires of the three rounds can be viewed in Tables 1, 2 and 3 of the Additional file. We started with 175 statements in the questionnaire in Round 1 and developed a further 12 statements from panellists’ comments, resulting in a total of 187 statements being rated across the three rounds. After 3 rounds, 14 statements from the Round 1 questionnaire were excluded (5 from Round 1 and 2, respectively, and 4 from Round 3), while all of the 12 newly-developed statements were endorsed, leading to a total of 173 statements being included in the adapted guidelines (see Table 4 of the Additional file for the full list of these statements).

The 173 statements in the adapted guidelines covered eight sections: (1) How do I know if someone is experiencing depression; (2) How should I approach someone who may be experiencing depression; (3) How can I be supportive; (4) Communicating effectively; (5) Difficulties the first aider may encounter; (6) Help-seeking; (7) What to do if the person doesn’t want help; and (8) Concerns for safety.

The consumer and carer panel had a higher endorsement rate than the professional panel (95% vs. 89%, χ² = 4.856, P = 0.028), and the Spearman’s correlation coefficient between statement endorsement rates of the two panels was 0.53 in Round 1 (P < 0.001). The correlation coefficients for Round 2 and 3 were not calculated due to the unequal drop-out of the two panels.

There were 14 statements from the guidelines for English-speaking countries that were excluded and 12 new ones that were added in the adapted guidelines. As summarised in Table 3, changes in statements happened across the eight sections of the questionnaire for Round 1. However, most changes emerged in the section ‘How can I be supportive’ (with 6 original statements excluded and 2 new ones endorsed), followed by the section ‘How should I approach someone who may be experiencing depression’ (with 4 originals excluded and 2 new ones endorsed).

Many similarities between the English-language guidelines and the Chinese guidelines were found. The endorsement rate of initial statements included in the Round 1 questionnaire was high (92%, 161 out of 175 statements being endorsed), suggesting a wide agreement on providing mental health first aid to people with depression between China and English-speaking countries.

Nonetheless, there were also a number of important differences, which were best reflected by exclusion of the 14 statements from the guidelines for English-speaking countries and inclusion of the 12 new statements developed specifically for the Chinese context (see Table 3). A prominent issue illustrated by these differences related to the autonomy of the person with depression. For example, experts from China proposed two new statements of ‘The first aider should not push the person too much to talk about their feelings and experiences’, ‘If the person is not willing to seek professional help, the first aider can leave information about this with them’ (i.e., respect the person’s choice not to seek for professional help immediately), and both of these statements were highly endorsed in the subsequent round (endorsement rates were 89.4% and 97.7%, respectively). In contrast, some statements in the guidelines for English-speaking countries, such as ‘The first aider should be open to any opportunity that presents itself to talk about their concerns with the person’ and ‘The first aider should know that often just taking the time to talk to or be with the person lets them know that someone cares’ (i.e., to provide support without taking into account the person’s feeling), were consistently rejected by both panels. Given that mental illness is often highly stigmatised in Chinese society [15], it is often considered a very ‘private’ issue for the person (and sometimes even for their families), so that helping actions without agreement of the person could be considered humiliating and intrusive, rather than being perceived as supportive or caring. This very important cultural difference in community attitudes towards mental illness suggests that a person providing mental health first aid in China should pay significant attention to the person’s autonomy and their willingness to talk about a very personal issue such as mental illness. Also, they should give greater consideration to the issues of potential shame and stigma [42].

For a long time in China, involuntary admission and treatment for mental health problems have generally been accepted as a necessary measure to protect patients, others, and society [43], whilst the rights of people with mental illness related to admission and treatment procedures have been largely overlooked [44]. However, in recent decades, with the national reforms in the field of mental health [3, 10, 11] and rapid development of research on mental health literacy [14], respect for the autonomy of people with mental illness has attracted more attention. This Delphi study shows that the core ideas of MHFA training, such as respect, non-judgement, sympathy and understanding to people with mental health problems [19, 20], were widely endorsed by Chinese experts, with extra statements on the autonomy of the person being included. Therefore, it is proposed that the adaptation of the guidelines and their dissemination may further contribute to the issue of respecting the autonomy of people with mental illness in Chinese society.

However, it is important to keep in mind of the gap between ‘knowing’ and ‘changing’. In many LMICs, such as China, attitudes of the public towards people with mental disorders are often associated with prejudice and discrimination and they are influenced by strong traditional values related to opinions on mental illness [15]; hence, we should not assume that respecting the autonomy of the person and the core ideas of MHFA training (e.g., respect, non-judgement, sympathy and understanding) would be incorporated into Chinese first aiders’ value system or reflected in their helping actions simply by inclusion of relevant statements in the guidelines. However, it is likely that MHFA training based on the guidelines can play a role in achieving change.

A further difference between guidelines for China and English-speaking countries relates to the role of families in the process of providing mental health first aid. Panellists agreed that families should be involved and contacted if the person ‘refuses to seek or accept professional help’, ‘is at risk of harming themselves or others’ or simply ‘if needed‘, although the term ‘if needed’ is ambiguous and is likely to vary from person to person. By contrast, the participants in the development of the guidelines for English-speaking countries advised the involvement of public services (e.g., GP, the police or mental health crisis teams) in similar circumstances. In Chinese society, mental illness is considered not only a personal problem but also a family issue, so it is common for families to assume primary responsibility for the care of a mentally ill member [35]. Another reason for such differences could be the lack of community mental health services and social support systems for people with mental disorders, particularly for non-psychotic conditions like depression [34]. People with mental illness have no option but to largely depend on their families, particularly in a crisis situation. Additionally, in a ‘collectivist culture’ like China’s, people tend to believe that the role of the local community as a whole and the family is more important than that of individuals; therefore, it is not surprising for Chinese panellists to agree that families should be contacted ‘if needed’ (possibly subjectively judged by the first aider), rather than letting the person make the decision. However, it is possible that the role of families in caring for a person with mental illness might be changing in line with other traditional family functions (e.g., education, physical and emotional support), due to smaller family sizes caused by China’s ‘One Child Policy’ and the rapid urbanization process happening in Chinese society [45].

Interestingly, Chinese panellists agreed that they should ‘not push the person too much to talk’ and should respect the person’s choice of ‘not willing to seek professional help’, but they failed to reach consensus on the statement ‘If the first aider is worried about someone who may be depressed, they should let the person decide when to open up’, even after 3 rounds of the survey. This hesitation may be related to the argument on ‘who should have the right to decide if a person with mental illness should seek help or not’, which has long been controversial in China, as in many other cultures [46]. Influenced by opinions about ‘respecting the autonomy of the person’ and ‘the traditional role of families’, Chinese panellists endorsed statements supporting both sides (e.g., ‘If the person does not incline to discuss how they are feeling, the first aider should not put pressure on them to do so’ vs. ‘If the person refuses to seek or accept professional help, the first aider should tell their family members about any precautions to take’). Accordingly, it is possible for Chinese first aiders to be caught in a dilemma when providing mental health first aid in practice when the person refuses to seek professional help: ‘leave the person to decide’ or ‘tell their families’. This raises another important issue for MHFA training in the Chinese context.

Some emerging opinions associated with recent reforms in the field of mental health in China are also reflected in the adapted guidelines. A good example of this is the inclusion of the new statement ‘The first aider should have some knowledge of the Mental Health Law in China’ (overall endorsement rate, 94%). Also, increased availability of some novel therapies for mental health problems (e.g., art or play therapy) also underlies the following new statements ‘If the person does not want to talk, the first aider should consider encouraging them to write or draw’ and ‘The first aider should encourage the person to do more leisure activities that they enjoy’.

Overall, both the professional panel and the consumer and carer panel had high endorsement rates (89% and 95%, respectively), and the results of the correlation analysis suggest a statistically significant correlation relationship between statement endorsement rates of the two panels (Spearman’s correlation coefficient = 0.53 in Round 1, P < 0.001). However, the correlation coefficient observed in this study is much smaller compared to those reported in similar studies in English-speaking countries. For example, Bond et al. reported a correlation coefficient of 0.95 between panels of professionals and consumers in a Delphi study to re-develop the mental health first aid guidelines for depression [26]. Such difference is likely to be due to the truncated range, as the statements with low endorsement in the English-language questionnaire were not included in the Chinese questionnaire.

Differences in opinions between the two panels (difference in the endorsement rates > ±10%), mainly in terms of professionals’ underestimating the capacity of patients with depression, were also observed. For example, professionals did not think the person’s depression would just go away without proper treatment (all professional panellists endorsed the statement ‘The first aider should not assume the person’s depression will just go away’) nor did they think that it was necessary to seek advice from people who have recovered from depression (more than one quarter of professional panellists rejected the statement ‘The first aider should learn more about depression by seeking advice from people who have experienced and recovered from depression’). Instead, they expressed concern about giving too much information as this could be overwhelming for the person (81% of professional panellists endorsed the statement ‘The first aider should not overwhelm the person with too much information or too many resources’). Consumer and carer panellists had the opposite opinion on these statements. Furthermore, the consumer and carer panel gave much higher endorsement rates to statements related to positive attitudes and respectful behaviours towards people with mental illness (e.g., ‘…the first aider should tell the person about the specific changes that they have noticed in a supportive and sensitive manner’), as well as their key role in leading the recovery from illness (e.g., ‘The first aider should know that recovery, for the most part, must be led by the person’).

These differences suggest significant divergence of views between mental health professionals and consumers and carers, which may be partly explained by the lack of mutual understanding, sometimes even opposing attitudes, between medical professionals and patients in Chinese society [47]. Therefore, to help the adapted guidelines to better reflect the needs of future users, it is likely to be important to include consumers and carers and value their voice equally to that of health professionals in future research, something that is not yet common in China.

This study aimed to harness the expertise of Chinese mental health professionals and consumers and carers to inform the actions that could be undertaken by a person providing help to someone with depression in China. The adapted guidelines will be available as a stand-alone document and also used to inform the development of a MHFA training manual and curriculum content. However, before using these guidelines to inform the public, it is important to consider the following issues: Firstly, the statements in the guidelines should be interpreted as a whole, with relevant information across sections being considered in a systematic way, rather than individually. The interpretation of the guidelines should also take the health systems and cultural understanding of mental health into consideration. Secondly, the adapted guidelines provide a new framework for mental health first aid intervention, particularly for depression - a common but inadequately addressed mental health problem in China. Lastly, with future improvements in mental health services and the public’s mental health literacy and attitudes towards people with mental illness in China, the guidelines will need to be updated.

There are many methods that may be used as part of a cultural adaptation process for behavioural health interventions [48]. The Delphi method offers a systematic way of doing this and is particularly appropriate for this study, because it parallels the process used to inform the English-language guidelines [26]. There is now a need for further exploration of how the adapted guidelines, and the associated training, might be implemented in the Chinese context, including the mental health care system, existing workforce and cultural values.