Systematic review: Perceptions of type 2 diabetes of people of African descent living in high‐income countries

Abstract Aims To describe how people of African descent perceive and understand type 2 diabetes, and to examine the impact of their perceptions and beliefs on the uptake of diet, exercise, weight control and adherence to medication recommendations. Design Systematic literature review of quantitative and qualitative studies. Data sources We searched MEDLINE, CINAHL Complete, Psych INFO, Academic Search Premier, Education Research Complete, Web of Science and Scopus, World Health Organization (WHO), Diabetes UK and American Diabetes Association for articles published from January 1999 to December 2019. Review methods Informed by the PRISMA guidelines, we independently reviewed titles and abstracts, identified articles for full‐text review that met inclusion criteria, conducted a quality assessment and extracted data. Findings were synthesized using a thematic approach. Results Twenty‐six studies met the inclusion criteria. Knowledge and understanding of diabetes were poor. Beliefs and behaviours about diet, exercise, weight and health care were erroneous. Most diabetic participants could not recognize diabetes symptoms, failed to take their diagnosis seriously and did not adhere to medication recommendations. The resultant effect was an increased risk of complications with undesirable outcomes. Conclusion Poor diabetes perceptions are linked to negative consequences and may be responsible for poorer outcomes among people of African descent. This review highlights the need to consider this population's beliefs and practices in structuring culturally sensitive programmes for diabetes management. Impact This systematic literature review is the first to exclusively explore perceptions of people of African descent in relation to diabetes. It is important to consider people of African descents' diabetes perceptions and practices before formulating interventions for their diabetes management.


| INTRODUC TI ON
The incidence of diabetes continues to rise worldwide, especially among people of African descent. Despite various educational programmes to enhance prevention and self-management, people of African descent suffer worse outcomes than other ethnicities (Assari et al., 2017). Perceptions of health and illness influence behaviour and attitudes towards conditions like diabetes (Diefenbach & Leventhal, 1996;Langston et al., 2017). Accurate perceptions of diabetes increase individual adherence to educational programmes (Attridge et al., 2014;Seear et al., 2019). Hence, a clear understanding of how health and illness are individually or collectively perceived may enhance development of interventions to improve diabetes self-care and overall quality of life for this population.

| BACKG ROU N D
Diabetes is a chronic, metabolic disorder characterized by excessive accumulation of blood glucose leading to serious damage to the heart, blood vessels, eyes, kidneys and nerves over time. Type 2 diabetes is the most common, usually occurring in adults when the body either does not respond to, or does not make enough insulin (WHO, 2021). Reasons for developing diabetes are multifactorial, including genetic susceptibility, metabolic disorder and environmental factors such as diet, physical activity and obesity (Golden et al., 2019). People of African descent (sub-Saharan African, African Caribbean and African American) living in highincome countries, irrespective of their migration periods or history, show high incidence of this condition and suffer worse outcomes (Assari et al., 2017). Lifestyle changes can prevent and/or delay the onset of type 2 diabetes (Glechner et al., 2018;Kriska et al., 2018;Siddiqui et al., 2018;Uusitupa et al., 2019). However, willingness and readiness to engage in necessary lifestyle changes depend on an individual's or group's perspectives about diabetes. Individual's beliefs and perceptions are influenced by their cultural and social backgrounds and experiences which, in turn, influence selfmanagement behaviour (Jakub et al., 2018;Shiyanbola et al., 2018).
Policies based on studies exploring perceptions have been shown to be more effective in empowering high-risk populations to adopt relevant diabetes management life changes (Attridge et al., 2014;Seear et al., 2019).
Exploring perceptions of diabetes, causes and effective management interventions of people of African descent, therefore, is fundamental to develop effective preventive and self-management interventions required for a chronic condition like diabetes.
Nurses, the largest group of healthcare professionals worldwide, play varied roles in health promotion, illness prevention, needs assessment, planning, implementing and evaluating interventions to improve the safety and quality of care in diverse settings. Hence, they are likely to be the first group of providers to come in contact with individuals at risk of or diagnosed with diabetes. Awareness of perceptions and beliefs about diabetes among people of African descent may facilitate high-quality care and improved health outcomes for this population.
We found no published literature summarizing the current state of knowledge in this area. Our overall purpose, therefore, was to summarize perceptions and understanding of type 2 diabetes among people of African descent to inform efforts to develop healthcare interventions that will improve outcomes in this population.

| AIMS
Specific study aims were: 1. To describe how people of African descent perceive and understand type 2 diabetes; and 2. To examine the impact of their perceptions and beliefs on the uptake of diet, exercise, weight control and adherence to medication recommendations.

| RE VIE W QUE S TION
Our specific research questions were: 1. What are the perceptions and beliefs of people of African descent in relation to diabetes, causes and type 2 diabetes effective management interventions? 2. What impact do their perceptions and beliefs about diabetes have on the uptake of diet, exercise, weight-management and use of healthcare? 5 | ME THODS

| Search strategy
This review was informed by the Cochrane Handbook version 6.1 (Higgins et al., 2019). The search strategy was developed using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines (Moher et al., 2015). Due to the scarcity of literature on this topic, all original research studies published in the English language that included people of African descent over 18 years of age, with or without type 2 diabetes, were included.
Systematic, scoping and literature reviews, commentaries, opinions, editorials and studies focused on diabetes other than type 2 were excluded ( Figure 1). Mixed-method systematic reviews preserve the originality of the findings (Hemingway & Brereton, 2009) without loss of scientific rigour (Harden & Thomas, 2005). Therefore, due to the scarcity of literature, we included all qualitative and quantitative designs to provide a broad perspective on the topic.

| Search methods
Studies initially underwent title and abstract screening, then a full-text review. Results were independently cross-checked by the three authors. The authors discussed and resolved any differences concerning inclusion or exclusion of studies. Further screening of the studies was conducted by three other independent reviewers and one more article was retrieved. The search strategy, inclusion and exclusion criteria, and data extraction elements are presented in Appendix S1.

F I G U R E 1
The artwork demonstrates the process of the study selection guided by the preferred reporting items for systematic reviews and meta-analysis (PRISMA) statement guideline (Moher et al., 2015) Records Commentaries (n = 4) Systematic reviews (n = 3).
Various type 2 diabetes topics other than the topic of interest (n= 5).
Studies included in studies (n = 26)

| Quality appraisal
Methodological quality of the included articles was assessed by LO and independently reviewed by PD and JST. Two quality assessment tools, the Critical Review Form version 2 for qualitative studies (Letts et al., 2007a;Letts et al., 2007b) and the Guidelines for Critical Review Forms for quantitative Studies (Law et al., 1998) were adapted for this review. Both tools are well-defined with clear instructions on how each question should be answered. We merged the components of both tools which were the same except for 'outcome', 'intervention' and 'results' applicable only to the quantitative studies.
Sets of questions for each component of the quality assessment tools were rated: (1) 'very likely'; (2) 'somewhat likely'; (3) 'not likely'; and (4) 'can't tell'. For example, critical questions to determine the sampling strength were: 'whether the study described the purposive sampling process'; 'whether sampling continued until saturation'; and 'whether informed consent was obtained'.
Where each of these questions within 'sampling strength' was scored 1, sampling for that article was scored 'strong'. The overall quality of each study was determined by the number of 'strong' ratings assigned. Four 'strong' ratings without a 'weak' rating were scored 'strong'. Moderate scores were assigned to studies with less than four strong ratings plus 1 weak rating. Two or more weak ratings in a study indicated a weak score. However, data from all the studies were included.

| Data extraction
The data extraction form was derived from the aims of this review (see Appendix S1). The appropriateness of the extraction form was tested on two of the studies prior to use (Higgins et al., 2011), and was satisfactory.

| Data synthesis
An inductive thematic approach to data synthesis was used (Braun & Clarke, 2006). Each of the studies was read and key words of the findings were recorded on a spreadsheet by LO. Words with similar meanings were grouped to form codes. Codes were then organized into themes and subthemes. The process was independently crosschecked by PD and JST.
Diabetes was perceived to be present only when the blood sugar level was high, and absent when it was low (Mann et al., 2009) Okop et al., 2016;Skelly et al., 2006). Liburd et al. (1999) assessed body size and shape of black womens' body silhouettes.
Participants showed preferences for small to middle-sized silhouettes, while perceiving middle-sized silhouettes as healthiest. Some considered smaller body size next to their current weight as ideal (Okop et al., 2016). Others felt good or bad about their current weight based on their levels of satisfaction. Some obese and optimal weight participants with low satisfaction levels viewed obesity as threat to their health (Okop et al., 2016). Many overweight and obese participants were satisfied with their body size (Baptiste- Roberts et al., 2006;Cameron et al., 2018;Okop et al., 2016). Some argued that no image is ideal and rejected the use of BMI in measuring health or unhealthiness (Cameron et al., 2018). Thinness was associated with health conditions like HIV/AIDS, cancer or psychological issues like depression (Okop et al., 2016).
Some perceived diabetes as causing weight loss (Skelly et al., 2006).
These views were reflected in participants' preferences for heavysized bodies (Skelly et al., 2006). Both genders in Baptiste-Robert et al.'s study (Baptiste-Roberts et al., 2006) believed that persons of the same gender should become heavier as their ages increased.

| Exercise perception and diabetes
Lack of understanding of the link between exercise and diabetes was also apparent (Foster et al., 2016;Skelly et al., 2006), even among those who believed in the importance of exercise (Cooper & Lemonde, 2016;Foster et al., 2016). In one study, only a small fraction of participants actually exercised regularly (Foster et al., 2016).

| The role of diet
The importance of diet was discussed in some studies (Foster et al., 2016;Skelly et al., 2006), but there was limited knowledge of what a healthy diet entailed (Foster et al., 2016;Skelly et al., 2006).
Understanding the effect of diet modification on blood glucose was lacking and home remedies were commonly used to make up for the limited dietary knowledge (Breland et al., 2013). Participants' ideas of healthy eating were influenced by their culture and experiences.
For example, some considered dietary restrictions for diabetics harmful, arguing that the hunger and craving for sugar that is associated with diabetes should be met to avoid emotional instability (Breland et al., 2013). Eating small portions were believed to be responsible for causing other health issues including hypoglycaemic symptoms that were perceived as worse than hyperglycaemia (Breland et al., 2013;Brown et al., 2007). Participants often relied on subjective feelings of blood glucose to control their diabetes (Brown et al., 2007;Mann et al., 2009). Others deliberately maintained a high blood glucose level for convenience and many ate frequently to avoid feelings of weakness (Brown et al., 2007;Mann et al., 2009).
Dietary recommendations were abandoned when there was no immediate improvement in expected outcomes (Breland et al., 2013).
Dietary advice given by healthcare workers was seen as suboptimal because they did not understand patients' needs and dietary preferences (Breland et al., 2013;Brown et al., 2007). Some participants believed that clinicians' prescriptions and dietary recommendations were generic, unfamiliar or unrealistic and, so, rejected them. The use of intuition to substitute for unclear food advice was reported; for example, participants would skip meals to substitute for eating less (Breland et al., 2013). Some argued that the diet they were asked to give up did not make them sick in the past, while others believed that their blood glucose went up regardless of what they ate (Breland et al., 2013). Participants were more willing to discuss treatment with medication than with diet change (Wenzel et al., 2006).
The impact of families on food choice was also identified. Some participants stated that they chose to eat the same unhealthy food as their families because their families were unwilling to adapt to the healthy diabetic diet. Eating different meals from the family menu was isolating and costly (Breland et al., 2013). Others expressed that the location of healthy food stores/shops made it difficult to access them due to transportation, cost and time, forcing reliance on the unhealthy foods that were readily accessible (Breland et al., 2013).

| Risk perception and symptom recognition
Risk perception and symptom recognition were shown to be deficient in most of the studies (Balls-Berry et al., 2015;Brown et al., 2007;Calvin et al., 2011;Foster et al., 2016;Mann et al., 2009;Skelly et al., 2006). Most participants stated that it was difficult to take the condition seriously even after they had been diagnosed and they only realized the need for treatment after experiencing serious diabetes-related illnesses (Wenzel et al., 2006). Some perceived that diabetes has few symptoms with minimal consequences, while most had not heard about HbA1c (Mann et al., 2009). Nephropathy, amputation and retinopathy/blindness were the least perceived risks to their health (Calvin et al., 2011;Foster et al., 2016). Sixty percent of participants in Foster et al.'s (2016) study were unaware of retinopathy screening at the time of diabetes diagnosis, half did not know they required annual screening and most of the male participants thought that they could only see the ophthalmologist when their vision was impaired.

| Diagnosis
Many of the diabetic participants stated that they had experienced symptoms for several years prior to being diagnosed after suffering major symptoms such as loss of consciousness (Wallin & Ahlström, 2010;Wenzel et al., 2006). Some were diagnosed through symptom manifestations such as excessive thirst and urination (Wenzel et al., 2006) which were sometimes detected after several visits to the doctor (Wallin & Ahlström, 2010). A few found out the diagnosis during routine check-ups (Wallin & Ahlström, 2010;Wenzel et al., 2006) or through treatment for other conditions. Some individuals found out through the use of their friend's or family member's glucometer (Wenzel et al., 2006). The news of diagnosis was received with varied emotions. Most were not surprised because of diabetes prevalence among their family and friends (Wenzel et al., 2006). Others described their reaction of shock, fear, anxiety and the tendency to doubt the diagnosis where no symptoms existed (Wallin & Ahlström, 2010). The

TA B L E 1 Summary of included articles
Authors/year Aim of study Abubakari et al. 2011 To investigate the relationship between diabetes-specific knowledge, illness perceptions and self-management behaviours and the relationship between self-management behaviours and metabolic control To explore self-management practices, use of diabetes information and heath service for diabetes care Issaka et al., 2016 To explore perception of type 2 diabetes Liburd et al., 1999 To explore perception of body size and shape of black women Mann et al., 2009 To identify suboptimal diabetes knowledge and beliefs, medication and monitoring that may be hindrances to effective self-management Moise et al., 2017 To generate culturally informed insight into diabetes knowledge, management and prevention Okop et al., 2016 To explore perception of body size, obesity risk and willingness to lose weight Peek et al., 2010 To explore perception of the influence of race on shared decision-making (SDM) Skelly et al., 2006 To understand the views and prevention of diabetes Stover et al., 2001 To explore health perception and their relationship to symptoms Wallhagen and Lacson, 1999 To explore perception of sense of control of type 2 diabetes Wallin and Ahlström, 2010 To investigate experience of diabetes diagnosis and to investigate how they describe their health beliefs Wenzel et al., 2006 To explore the experience of being diagnosed with diabetes Yeoh and Furler, 2011 To explore diabetes perception within the context of broader health circumstances attitudes of denial and refusal to acquire diabetes knowledge were adopted to forget the diagnosis (Wallin & Ahlström, 2010). Most immigrant participants considered the positive consequences as favourable; for example, giving up sweets was considered advantageous.
They also took solace from the belief that normal life could still be maintained as a diabetic in their host countries with stable economies, a wide knowledge of the disease and the availability of medication (Wallin & Ahlström, 2010;Wenzel et al., 2006). Positive comparisons of type 2 with type 1 diabetes and other conditions that were thought to be more dangerous were used as techniques to cope with the diagnosis (Wallin & Ahlström, 2010).

| Medication perception
Views about the use of diabetes medications differed. The severity of the disease was measured by the type of treatment that was recommended (Wenzel et al., 2006). Glycaemic control with diet and medication were understood as a less serious case of type 2 diabetes (Bockwoldt et al., 2016;Brown et al., 2007;Skelly et al., 2006).
Insulin use was perceived as being for more serious cases (Bockwoldt et al., 2016;Brown et al., 2007), and was, therefore, surprising and harder to accept (Bockwoldt et al., 2016). Disappointment, shame, anxiety, frustration, despondency and self-blame were attributed  to insulin initiation, perceived as indicating failure to take care of self (Bockwoldt et al., 2016). Some believed that it was restrictive, caused weight gain and indicated that their condition was getting worse with imminent complications, others feared the injection (Brown et al., 2007).

Country of study
Two studies (Brown et al., 2007;Mann et al., 2009) explored non-compliance with diabetes medications. Participants were non-compliant with treatment advice that did not conform to their health beliefs (Brown et al., 2007). Worries about side effects of diabetes medications (Brown et al., 2007;Mann et al., 2009) and possible addictions were expressed along with complaints about the medication being hard to swallow (Mann et al., 2009). Some participants believed that medication could be skipped when blood glucose levels were 'normal' (Mann et al., 2009). Caribbean participants believed that the traditional medicines they used in their homeland were more effective than western medicine (Brown et al., 2007). There were beliefs that medications could cure diabetes. Physical activity and weight loss were misunderstood by some as only necessary when diagnosed with diabetes, but not also for prevention (Skelly et al., 2006). Those who believed that insulin was an ideal treatment for the disease and those who feared the risk of complications accepted insulin while others did not accept that they could develop complications (Brown et al., 2007).

| Perception of spiritual influence
Perceptions of spiritual influence were linked to experiences with the diagnosis, acceptance of the condition and their trust in God's help in managing it (Brown et al., 2007;Cooper & Lemonde, 2016;Issaka et al., 2016;Skelly et al., 2006;Wallin & Ahlström, 2010;Wenzel et al., 2006). Some participants expressed a sense of powerlessness (fatalism) over their condition which they saw as the will of God and not within their control (Issaka et al., 2016;Mann et al., 2009) while most African immigrants believed it was due to unseen forces or sheer bad luck (Issaka et al., 2016). Many trusted God for the outcome of their diabetes although, for some, this

Consequences of Poor Perceptions & Knowledge of Diabetes
Increased diabetes symptoms and risk of complications Liburd et al., 1999;Wallhagen & Lacson, 1999;Stover et al., 2001;Ford et al., 2002;Skelly et al., 2006;Baptiste-Roberts et al., 2006;Mann et al., 2009;Calvin et al., 2011;Hyman et al., 2014Okop et al., 2016Issaka et al., 2016;Foster et al., 2016 Beliefs about the Causes of Diabetes Cultural & environmental effect Brown et al., 2007;Yeoh & Furler, 2011;Hyman et al., 2014;Issaka et al., 2016;Moise et al., 2017 Racial discrimination & stress Brown et al., 2007;Yeoh & Furler, 2011;Breland et al., 2013;Issaka et al., 2016;Cooper & Lemonde, 2016 Communication & information sharing Yeoh & Furler, 2011;Skelly et al., 2006;Brown et al., 2007;Peek et al., 2010;Breland et al., 2013;Hyman et al., 2014;Balls-Berry et al., 2015;Cameron et al., 2018 understanding did not hinder them from seeking medical treatment (Brown et al., 2007;Wenzel et al., 2006). Older adults expressed the belief that God could cure their diabetes through behaviour change and by the work of the doctors (Skelly et al., 2006). Some stated that illness came from God and that asking why could imply ingratitude, lack of faith and sinfulness, and may result in God's punishment. Hence, they accepted the condition because life belongs to God and they must be thankful no matter what hssed (Wallin & Ahlström, 2010). Health was perceived as a gift from God and individuals only shared in the responsibility of maintaining the good life while God could also take it back as he is in ultimate control, both to protect individuals from getting ill and to decide who gets cured (Wallin & Ahlström, 2010). Depending on their faith background, most non-diabetic participants stated that they would engage in prayer and believe in miracles for their healing if diagnosed (Issaka et al., 2016).
Some participants expressed a sense of loss from reduced mobility resulting from amputation, decreased or loss of sexual drive (Ford et al., 2002) and vision (Ford et al., 2002;Foster et al., 2016). Diabetic retinopathy increased with the duration of diabetes.

| Cultural and environmental effect
Culture and weather change were seen as the causative agents of diabetes by some Caribbean immigrants (Brown et al., 2007). Some argued that their lives before migration were healthy because of the living conditions in their homeland (Brown et al., 2007;Hyman et al., 2014;Issaka et al., 2016;Yeoh & Furler, 2011). They believed that their active and strenuous lifestyle in their home country (Brown et al., 2007;Issaka et al., 2016;Yeoh & Furler, 2011), coupled with the hot weather, helped to burn off excess calories (Brown et al., 2007). However, ease of transportation (Hyman et al., 2014;Issaka et al., 2016;Yeoh & Furler, 2011), unemployment (Yeoh & Furler, 2011) and long hours watching television led to social isolation in the host-country (Brown et al., 2007;Issaka et al., 2016;Yeoh & Furler, 2011). This, they believed, brought about decreased activity that facilitated their developing diabetes. Most African participants perceived diabetes as a disease of wealthy urban people in Africa because of their luxurious easy lifestyle (Issaka et al., 2016).

| Racial discrimination and stress
Most of the studies discussed participants' beliefs about the role of stress in developing diabetes and poor glycaemic control (Breland et al., 2013;Brown et al., 2007;Cooper & Lemonde, 2016;Issaka et al., 2016;Yeoh & Furler, 2011). Sources of stress included displacement from their homeland (Yeoh & Furler, 2011), migration, acculturation and adaptation to new unfamiliar environments (Brown et al., 2007;Issaka et al., 2016;Yeoh & Furler, 2011), poor living conditions and racism (Brown et al., 2007). Stress associated with racism and poverty was perceived as the main cause of poor eating habits and diabetes; for example, less nutritious foods that could not be sold in white or more affluent neighbourhoods were sent to the minority neighbourhoods (Breland et al., 2013). Other sources of stress included work (Cooper & Lemonde, 2016) and unpleasant experiences with the healthcare system (Breland et al., 2013;Brown et al., 2007;Yeoh & Furler, 2011). Such unpleasant experiences were perceived as due to race, ethnicity, language differences, their poor economic status and suboptimal healthcare (Breland et al., 2013).
Physicians, patients and socioeconomic factors like health insurance were identified as barriers to effective communication and shared decisions between physicians and patients (Peek et al., 2010).
Healthcare professionals were perceived as not providing enough information; for example, on medical conditions, not explaining test results, not paying attention to patients due to conflicting cultural views and stereotypes, dominating the conversation with little regards to patients, not reviewing treatment options with them, and making treatment decisions without considering patients' preferences. Conversely, patients were viewed by some participants as being unable to share information with their physicians especially about health concerns and medication use due to poor communication resulting from limited education, negative attitudes, beliefs and behaviours and poor participation in practices that foster good health. An erroneous belief of less self-worth grounded in internalized racism, which makes them unable to voice their opinions and question the providers' recommended treatments, were most likely to lead to non-compliance with the recommended treatment (Peek et al., 2010).
Some obtained information about symptoms, complications, medication and diet adjustment by observing affected individuals (Balls-Berry et al., 2015;Skelly et al., 2006). Some relied on friends, colleagues, families and their community leaders for diabetes information (Balls-Berry et al., 2015). Where illness conditions were not openly discussed by the affected individuals with relatives, as was noted in some studies (Balls-Berry et al., 2015;Cameron et al., 2018;Skelly et al., 2006), awareness of the implications of the condition could be hindered as opposed to if they shared their experiences (Balls-Berry et al., 2015).
The desire to learn more about the causes, prevention and management of the disease was expressed by some. They suggested that there was a need to consider the culture and age of people when structuring culturally targeted appropriate messaging systems that would capture the attention of the intended audience as well as the importance of healthcare providers visiting communities for better understanding of the people (Balls-Berry et al., 2015). Our results revealed that diabetes perceptions significantly influenced beliefs and adherence to medication as was also identified by Urata et al. (2019). Insulin prescription was perceived as worse diabetes while management with diet was understood to be a lesser condition.

| DISCUSS ION
We identified fatalistic beliefs that emanated from illness and Other factors influencing development of type 2 diabetes were acculturation, perceived racial discrimination (Bilal et al., 2020), poverty and the associated stress (Breland et al., 2013). Racism, poverty and stress were perceived by most African Americans as leading to poor eating habits of low quality foods available in ethnic minority neighbourhoods (Breland et al., 2013). Analysis of the primary research studies included in this systematic review revealed that nine different topics concerning type 2 diabetes were covered. Most of the primary studies were descriptive/exploratory in design, with only one randomized controlled trial (Table 1), leaving a large gap in the science, impeding adequate understanding of diabetes perceptions and behaviours of Africans who live in high-income countries. There is an urgent need, therefore, for more randomized controlled trials examining this topic with both homogenous and broad samples of African migrants to reduce bias and to close the gap in the science. The results generated by such studies must be translatable into evidence-based clinical practice (Hariton & Locascio, 2018;Spieth et al., 2016) and have a positive impact on this populations' diabetes' management.
Programmes are needed to empower people of African descent to be more involved in their health management, to ask questions about their conditions and available treatment options when they visit their healthcare providers and to seek second opinions when they feel the need to do so. Equally, healthcare providers should be ready to address any concerns that patients from this population might have and be able to explain things in ways that they will most likely understand.
Ways of improving social support for delivering accurate information about diabetes and its management are worth exploring because of the high dependency of people of African descent on families, friends and colleagues for diabetes negotiations. Such interventions may help to bridge the diabetes health inequalities that currently exist between this population and their counterparts.

| LI M ITATI O N S
Most of the studies in this review were conducted in the USA, making African Americans the largest percentage of participants.
Therefore, most of the findings from this review are from the perspective of the African American population whose sociocultural backgrounds and experiences might be different from other people of African descent. Most participants were female and we limited our search to studies with participants 18 years of age and older.
Males and younger participants may perceive type 2 diabetes differently. The literature search for the included studies for this review followed a systematic and rigorous process, but we did not include grey literature, so some relevant studies may have been missed, impacting the findings. Given that eight of 26 (about 30%) of included studies were of weak to moderate quality, this may have implications for data analysis and findings. Future systematic literature review on this topic may include only articles of high quality, and/or consider only participants from African American or African Caribbean or African immigrant communities.

| CON CLUS ION
This review explored evidence from primary studies of diverse designs about perceptions, beliefs and management of type 2 diabetes of people of African descent living in high-income countries. The findings demonstrate that the incidence and rate of complications of type 2 diabetes among this population could be due to poor perception and management. This has important implications for policymakers and healthcare providers, nurses in particular. Basing their diabetes interventions for people of African descent on understanding of their diabetes perceptions may be more effective with successful outcomes.

ACK N OWLED G EM ENT
The authors would like to acknowledge the University of Hull Library Skills Team for their assistance.

CO N FLI C T O F I NTE R E S T
There are no conflict of interest in this study.

PEER R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/jan.15266.