Design

We used an exploratory qualitative design: open-ended, semi-structured interviews were conducted and recorded. Responses to each question were qualitatively analysed to determine themes. This study was approved by the medical ethics committee of the Vrije Universiteit Brussel and University Hospital (VUB &UZ-Brussel) and took place within the diabetes clinic of University Hospital (UZ) Brussel during the period from 6 April 2021 to 21 April 2021. Written informed consent to participate in the study and publish the results was obtained from all participants, whilst the study protocol was positively reviewed by the medical ethical institutional review board (approval number B.U.N. 1432021000375)).

Recruitment of participants

Participants were recruited via purposive sampling. Inclusion criteria were: 1. having type 2 diabetes (as defined by ADA criteria) [10], 2. aged ≥ 18 years, 3. registered at the diabetes clinic for at least 6 months, and 4. being of Belgian or Moroccan descent. The recruitment of participants took place within the diabetes clinic of UZ Brussel, Belgium. Medical specialists responsible for consultations were informed about the study and referred participants based on the inclusion criteria to the researcher who performed the interview (DI). During sample selection we aimed to include a similar number of people of Belgian and Moroccan origin, as well as aiming for a gender balance.

Semi-structured interviews

An interview guide was compiled, inspired by the themes found by Peeters et al. (2015) who conducted a similar study with a different immigrant population [5]. The final interview guide was reviewed by SS and AK. Prior to data collection, the interview guide was tested by DI, SS and AK. In an experiential learning setting, DI was mentored by SS, a medical anthropologist, in the acquisition and application of cross-cultural communication skills. This was achieved through a role-play exercise, wherein AK, a diabetes specialist, took on the role of a typical patient. This scenario emulates the consultations that AK frequently conducts within his clinical practice. The relevance of this prior training is underlined by the fact that Brussels, the location of AK’s practice, is the second most culturally diverse city globally. Statbel (2022), the Belgian statistical office, estimates that approximately 75% of the population in Belgium’s capital comprises individuals of non-Belgian origin, with the top three countries of origin being Morocco, France, and Italy [11]. DI conducted all seventeen interviews; six were conducted in Dutch, with the others conducted in French (DI is fluent in both languages). The interviews took place in a designated room within the diabetes clinic and were audio recorded.

Analysis

All interviews were recorded and transcribed verbatim. French-spoken interviews were transcribed and then translated into Dutch. We used thematic analysis as described by Clarke & Braun [12]. All transcripts were read closely for surface and underlying meaning and codes were developed to represent units of meaning. Primary coding was conducted by DI, reviewed by SS and discrepancies were resolved to consensus. Themes were developed by identifying patterns of meaning within and across transcripts and compared with the themes found by Peeters et al. [5]. Although the interview guide was inspired by the themes found by Peeters et al., we used primarily an inductive approach to coding and analysis, from the point of view of checking whether the same themes were found, as well as to allow new findings specific to the patient group of interest. Memo-writing was used to (further) develop themes and verify relationships between themes and data elements. These memos were discussed at team meetings to facilitate an integrated reflection on our relationship to the data, study participants and analysis. The final structure of relationship between themes and sub-themes was decided by consensus of all investigators involved. Representative quotes were selected to illustrate themes. NVivo software package was used for coding and managing the data [13].

Beliefs about diabetes

All participants, except for one Moroccan patient and one Belgian patient, indicated that they received sufficient information about their illness and diagnosis. Two participants were convinced that diabetes is an acute illness. Three Belgian and four Moroccan participants stated that although they were well informed, they would have preferred to receive more preventive information. Moroccan participants highlighted that preventive information should be adapted to the Moroccan target group. Most participants, from both target groups, showed awareness that an unhealthy lifestyle can lead to diabetes.

Yes, it is a question of education! What are they going to give Moroccans as good advice? They hardly know anything about it! But it is not their fault! There are many who cannot read or write.

(Male, Moroccan)

Among Moroccan participants, a healthy lifestyle is more linked to avoiding sweets rather than having sufficient physical activity. Whereas among the Belgian group, healthy food in combination with physical activity is always seen as the ideal lifestyle. Five Moroccan participants did not deem it necessary to modify their eating habits on account of their disease.

I think that doesn’t bother me—I still continue to appreciate [unadjusted] Moroccan cuisine which is very rich.

(Female, Moroccan)

Some Moroccan patients said they compensated for their unhealthy eating habits by adding extra vegetables to dishes. Two Moroccan participants mentioned that younger generations are better able to adapt their eating habits, while the older ones tend to be more conservative regarding nutritional changes.

There does seem to be a difference in lifestyle between generations. In general, the older generations are more used to traditional Moroccan cuisine. And the younger ones are more flexible and eat a more European cuisine

(Female, Moroccan).

Just over half of Moroccan participants assumed that diabetes has a hereditary factor. For five Moroccan participants, the onset of diabetes was linked to a stressful or emotional event. Among Belgian participants, religion had little influence on the perception of the disease, while three Moroccan participants were convinced that it was their fate, according to their religious beliefs, that they were diagnosed with diabetes. Some Moroccan participants indicated that they had no fear of the disease (T2DM) since several people close to them have grown old despite this condition. It is therefore no longer seen as something exceptional within the Moroccan community according to the participants.

It is also just assumed that the disease diabetes is something normal within our community!

(Female, Moroccan)

Proper compliance with prescribed therapy was experienced to be intensive by most participants. Two Moroccan participants expressed reluctance towards blood glucose self-monitoring (via finger prick testing). They would like to replace this by flash or continuous glucose monitoring (via a sensor worn on the body) but cited the financial aspect as a barrier to this switch. (…) If there is a way to get a sensor or something that measures blood without needles. They said it exists, but you must pay for it and it’s expensive. I can’t afford it. And I regret that. (Male, Moroccan) (As of July 1^st 2023 continuous glucose sensor (GCS) monitoring use is being reimbursed in Belgium for type 2 diabetes on an intensive insulin regimen (i.e. 3 or more injections per day); at the time of this study there was no reimbursement for CGS.)

Patient—Doctor relationship

Participants indicated high confidence in the expertise of both the general practitioner and the diabetologist, noting that they did not expect their general practitioner to be an expert on diabetes. In this sense, patients indicated more confidence in the diabetologist’s recommendations. This trust led patients to passively accept treatment recommendations, i.e., the majority of patients did not intend to check the validity of the recommendations made in any way. Only a proportion of participants indicated that they actively sought additional information about the proposed treatment plan. Patients indicated that the overall positive relationship with the general practitioner and diabetologist was partly due to the facilitation of communication in Dutch, French or, if necessary, in the patient’s original language with the support of an interpreter.

He is French speaking but can speak Dutch well and for medical matters you like to have information in your own language. It depends from hospital to hospital. You do like to have your own language when it comes to medical matters. Language does a lot because you understand the difficult terms better.

(Female, Belgian)

With regard to the transfer of information between the relevant physician and the patient, none of the patients indicated that they experienced problems in the transfer of information. Moroccan patients indicated that although they have confidence in the professional’s expertise, and therefore feel obliged to adhere to treatment as much as possible, this in itself is not sufficient to increase compliance. This mainly concerned nutrition and the proper use of medication.

He has an influence: yes and no. He asks me to do a lot of things and then sometimes I think when I want to go and eat a burger, I am not going to do it anyway to satisfy the doctor. But sometimes I also think the opposite and really don’t care what the doctor is going to say. And then I eat less healthy

(Female, Moroccan).

Furthermore, two Belgian participants indicated that the information offered in and by the hospital could be more comprehensive. In addition, a Moroccan participant reported a lack of in-depth and detailed information, as only the general causes of diabetes were discussed with the diabetes specialist, without addressing the specific causes and details related to the individual diagnosis. Finally, one Moroccan participant reported withholding information about eating habits to maintain a false image of correctly following the treatment plan for fear of disappointing the physician.

Barriers and facilitators

From the group of Belgian participants, only two persons indicated that they completely and correctly followed the proposed recommendations. This was not the case in the Moroccan group where none of the participants stated to be fully compliant.

For both groups, maintaining a balanced and healthy intake of food seemed to be challenging, albeit for different reasons. The Belgian participants, indicated to experience difficulties in social settings whereas the Moroccan group, indicated to experience difficulties given that the required changes are difficult to combine with the traditional Moroccan kitchen. The latter may be related to the dominant male culture, in which Moroccan women often conform to their husbands’ wishes, which do not always correspond to the advised nutritional plan.

Women are usually submissive; the man is usually the boss in the house. The woman usually must listen to the man’s wishes, their life is more difficult. (…) And this can then have an effect on bad living habits.

(Female, Moroccan)

From a religious point of view the vast majority indicated that their religion and more specifically strict fasting during Ramadan, does not pose a barrier for adhering to treatment as their religion in principle exonerates ill people of the requirement to fast.

Since I am diabetic, I am considered someone who is sick. So, I have the right to eat. So, I do not participate in Ramadan. So, I don’t adjust my medication either.

(Female, Moroccan)

Belgian participants who, as discussed, experience difficulties in social settings also indicate to experience difficulties because of both practical considerations and lack of support from the environment.

You have to leave a lot behind. You also have to have the character and support of the people around who help you to do that. Having support helps a lot. That you are not allowed to snack anymore. (…) My husband likes to eat biscuits for breakfast. I dare to eat two pastries then.

(Female, Belgian)

Note that the latter has been indicated to be both a barrier as well as a main facilitator depending on the participant. In addition, mainly the Belgian participants indicated to be able to adhere better to treatment as a direct result of their retirement.

In both groups, a lack of intrinsic motivation appears to be at the basis of limited physical activity. In the group of Belgian, things such as a lack of time, being physically unable to exercise and having done enough sport in the past are mentioned as reasons for not exercising enough. Moroccan participants considered adequate physical activity to be less important in their treatment plan.

Taking treatment less seriously also appears to lead to less consistent compliance with medication intake. Participants who took the disease too lightly showed little interest in taking their medication correctly.

In the beginning, I was a bit sceptical, I didn’t believe in it very much. I thought it was a simple disease like a cold with no consequences. (…) That’s why I didn’t have much interest in taking my medicines. When I forgot to take them, I didn’t think it was serious.

(Female, Belgian)

Some Moroccan participants described adjusting medication on their own based on their meals. When participants ate a healthy meal, doses of medication were skipped. Forgetfulness due to fatigue, the amount of medication or breaking a certain routine were also cited by participants for not taking their medication correctly.

At 22h15, the long-term insulin was then injected. And I did have problems with that. Sometimes I was asleep before I got there

(Male, Belgian).

Among Moroccan participants, scepticism towards medication, expected outcomes and actual results showed reduced compliance.

One day I do it, two days I don’t and that’s it. I tell myself it’s always the same. I see the results. So, there’s no point.

(Male, Moroccan).

This in combination with the fact that they indicated to freely adjust the medication and treatment plan in function of their lifestyle at a current point in time as they saw fit. A last frequently mentioned barrier within the Moroccan group was ’stress’. Five participants reported that they found it very challenging to follow the proposed therapy during times of stress or emotional difficulties. This applied to both medication and lifestyle.

As soon as I have a moment of mental difficulty and go into a slump, I spend less attention on the disease.

(Female, Moroccan)

In both groups, patients reported that adhering to therapy is supported by having structure and routine.

You get used to that (…). That becomes a routine in the morning, I get up 15 minutes early for that. I put them in a bag from the supermarket here. Yes, every morning I just take those. The ones from the evening are on my night table and before going to bed I take them. I actually have little problem taking that. I rarely skip that.

(Male, Belgian)

Additionally, all participants expressed a desire for improved health and acknowledged that they adhered to treatment out of fear of the consequences of non-compliance. However, some patients did not consider the disease to be serious, and as a result, they gave less importance to treatment compliance.

Strengths and limitations

This study has notable strengths, as it represents the first investigation of its kind in a Belgian context. Specifically, this research examines underlying factors, beliefs, and experiences that shape therapy compliance among an important ethnic minority group in Belgium which has not been studied previously. The application of semi-structured interviews as a qualitative research method enabled patients to provide a comprehensive account of their experiences.

This study also has limitations. Firstly, the potential for generalization of these findings is limited due to the small sample size. It is unclear whether the findings can be applied to other contexts. However, the results could potentially be transferred to other Moroccan communities with similar characteristics, such as level of education, generation of immigrants, and location. Secondly, recruitment of participants through the doctors present in the diabetes clinic may potentially have introduced selection bias. The doctors made the decision to refer participants to the researcher for inclusion in the study, which could have influenced characteristics of the study population. Thirdly, in this study we did not collect socioeconomic data or variables related to possible systemic inequities, thereby limiting our ability to assess the potential role of non-culture related factors, such as e.g. healthcare access, or morbidity differences between the migrant and host population [22].

Despite these limitations, our findings show that cultural factors play a significant role in treatment compliance in this patient group and that it is necessary to include these elements to improve patient care. More research is needed in other health care settings with participants from the Moroccan community to further explore how care is better tailored to this group.