Introduction

Population ageing has been accompanied by a shift in disease profile to non-communicable diseases (NCD), increased levels of disability, and an increasing loss of physical and cognitive functioning. Most high-income countries have accepted the chronological age of 65 years as a definition of 'elderly' or older person, but like many westernized concepts, this does not adapt well to the situation in Africa. As a result of the changing in legislation via the Social Assistance Amendment Act, 6 of 2008, South Africa aligned itself with the United Nations definition of the ‘older persons’ as all persons over the age of 60 years. [1–2] South Africa has one of the most rapidly ageing populations in Africa despite the demographic impact of the HIV/AIDS epidemic. The population aged 60 years and older numbered over 4.5 million in 2016, thereby accounting for just over 8% of the South African population [3]. It is expected that, by 2025, the proportion of the older population will increase by 10.5% to reach 5.23 million [4–5]. The latest estimates are that three-and-a-half million South Africans (about 6% of the population) have diabetes, and there are many more who are undiagnosed. This number is anticipated to grow by 30% by the year 2030. [6–7] According to the WHO Global Report on diabetes in 2016, the combination of increasing prevalence of diabetes and increasing life expectancy in many populations with diabetes may be leading to a shift in the types of morbidity that accompany diabetes, such as cancers and cognitive disability. [8] Inevitably, this will place further strain on both healthcare resources and health providers.

In South Africa, 80% of the population receive their health care through the government funded public sector. [9] In general, community health centres (CHC) and smaller primary care clinics are the older persons’ first point of contact with the healthcare services. These are staffed according to their size and location and provide a comprehensive package of care. The CHCs are overcrowded and poorly resourced due to the multiple disease burden, leaving limited time for the front-line health workers to deal with the management of patients with diseases such as diabetes. [9] A qualitative study found that patients with diabetes in this setting were ill-equipped to play an active role in self-care due to their limited opportunities for education and counselling. [10] In this setting too, poor control of glycaemia and hypertension together with high levels of multimorbidity are commonly encountered. [11]

Diabetes self-management practices (DSMP) form the foundation of diabetes care. These involve knowledge, skills and motivation as it requires, amongst others, adjustment of the diet, monitoring of blood glucose levels, where appropriate and an increase in physical exercise. [12,13] Sprague et al. found that the decreased priority given to patient education among older individuals, their support systems, and the healthcare community is a factor that negatively impacts their learning diabetes self-management. [14]

A number of barriers associated with ageing reduce the older persons’ potential for engaging with traditional self-management education programmes such as lectures/group sessions; for example, hearing and visual deterioration. Further self-management practice may be affected by reduced manual dexterity due to osteoarthritis which is common in this group. [15–16] Therefore, older people might have fewer resources available to manage their condition than younger people and will then have a higher need for self-management support. The loss of friends and family makes them more vulnerable to loneliness and social isolation. [17]

Poor social support is associated not only with an increase in mortality, morbidity and psychological distress but a decrease in overall general health and well-being. [18] Several studies have found social support vital to SMP in people with chronic diseases. [18–19] Social support can be either emotional or physical. Emotional, social support is defined as the degree to which interpersonal relationships serve the purpose of providing emotional, informational or influential quality of life for the individual. [20] Physical support is defined as the forms and numbers of social relationships (marital status, the number of friends) and the degree of connection between these relationships (social network).

Most frequently social support for persons with diabetes covers aspects of active support and emotional encouragement with taking medications, monitoring blood glucose, foot and eye care, following diabetic meal plans and increasing physical activity. [21] This study was undertaken in older people with diabetes in South Africa to examine their knowledge about living with and managing their diabetes; and to determine the relationship of social support, especially that of family and friends with their self-management.

Methods

Results

Descriptive and bivariate analysis

Patients’ demographic profile and clinical characteristics.

Socio-demographic and clinical characteristics of the study sample are presented in Table 1. Of the 406 participants, 68.5% were females, 60.5% were living with a family member, and almost half were married. Two hundred thirty-three (57.4%) had less than 7 years of education. Unsurprisingly most of the participants were pensioners 374 (92%) and 348 (84%) reported family income ≤R 1500 (US$107) per month.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Socio-demographic and clinical characteristics of the study participants.

https://doi.org/10.1371/journal.pone.0230173.t001

The mean duration of diabetes from diagnosis was eight years. Sixty-two percent were using oral therapy for glycaemic control, 31% combined insulin and oral agents and 7% insulin alone. The majority were taking medication for other conditions: 306 (75.4%) for hypertension, 138 (34%) for other chronic diseases and 61 (15%) for heart problems. The HbA1c was higher than 8.5% in nearly 185 (47.7%) of the participants, <7.5% in 112 (28.9%) and 7.5–8.5% in 91 (23.4%).). Women were more likely than men to self-report experiencing low and high blood sugar (21% vs 11%, p = 0.016) and (53% vs 37%, p = 0.004) respectively.

Knowledge, diabetes self-management practices, social support scores.

A ‘good knowledge’ score was for those whose score for correct answers was ≥50%. Overall the level of knowledge was poor in 204 (50.2%) (Table 2). The deficiencies were particularly noticeable in relation to symptoms of diabetes and complications of diabetes and hypertension. There was a better level of knowledge about aspects of self-management e.g., a healthy diet and foot care; with 61.3% and 64.8% of respondents correctly answering questions about walking barefoot and daily foot inspection respectively (Fig 1A and 1B).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1.

a. Distribution of correct answers to questions on knowledge on symptoms and complications of diabetes and hypertension (n = 406). b Distribution of correct answers to questions on diabetes self-management knowledge (n = 406).

https://doi.org/10.1371/journal.pone.0230173.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Knowledge, diabetes self-management practices, social support scores.

https://doi.org/10.1371/journal.pone.0230173.t002

Similarly, 233 (57.5%) were assessed as having a good level of physical exercise and almost two-thirds (262 (64.5%)) of the participants were following a diabetic eating plan (Table 3). Over three-quarters of participants (312 (76.8%)) had good social support (Table 2). The majority of participants agreed that their family supported them to follow all aspects of self-care included in the questionnaire and encouraged them in managing their diabetes.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Self-management practice of lifestyle risk factor.

https://doi.org/10.1371/journal.pone.0230173.t003

Only 74 (18.4%) reported that their family nagged them about their diabetes and 71 (18%) reported that their family felt uncomfortable about them because of their diabetes (Table 4). The mean diabetes knowledge, self-management practice and social support scores of the participants by socio-demographic and clinical characteristics are presented in S1 and S2 Tables. The mean knowledge score was significantly lower for a single (41.2 (SD = 2.9)) compared to married respondents (46.7 (SD = 12.5)) or those in a companionship (49.4 (SD = 12.7)). The mean knowledge score was greater for participants with higher education level and those who had experienced high or low blood sugar (p = 0.005 and p = 0.001 respectively), and those receiving medications for chronic diseases other than hypertension (p = 0.038) than their counterparts.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Social support assessment of the study participants (n = 406).

https://doi.org/10.1371/journal.pone.0230173.t004

The social support score means also differed significantly with living arrangements; it was lowest in those living with a spouse (67.7 SD = 21.0), intermediate in those living alone (75.0 SD = 21.5) and highest in those living in multiple household members (82.3 SD = 18.0 p = 0.001). On the other hand, there were no significant associations between diabetic knowledge, self-management practice and social support score with duration of diabetes, the type of medication used to treat diabetes, or receiving treatment for hypertension.

The multivariable regression models.

The ordinal logistic regression models of knowledge, self-management practice and the components of social support scale are given in Table 5. The table shows the effect of the K and SMP indices on the 12 individual SS components, which were measured on a Likert scale. Social support was positively associated with the self-management practice score for following a diabetic meal plan, taking care of feet, physical activity, handling participants’ feelings about being diabetic and testing blood sugar, but not for taking medication. Family and/or friend emotional support (nagging, encouraging /reassuring and handling feelings about being diabetic) were positively associated with knowledge score. Multivariable linear regression analysis (S3 and S4 Tables) showed older age was negatively associated with knowledge (®: -1.893, 95% CI-3.754; -0.031) and higher income was positively associated with self-management practice (®: 3.434, 95% CI 0.797; 6.070). There were no significant associations between socio-demographic variables, HbA1c and social support with knowledge or self-management practice.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. The ordinal logistic regression models of knowledge, self-management practice with the components of social support scale.

https://doi.org/10.1371/journal.pone.0230173.t005

Discussion

In this study, half of the participants had poor knowledge about diabetes and its complications. Just under two-thirds were assessed as having a good level of physical exercise and two-third of the participants were following a diabetic eating plan. Three quarters perceived that their family supported them to follow all aspects of self-care management. Being in the high-income group was associated with good level of self-management practice. Finally, social support was positively associated with both knowledge and a number of self-care aspects. The deficiencies noted in the participant's knowledge relating to the complications of diabetes and hypertension are alarming.

Education of patients including those with diabetes is essential for self-management. It suggests in this study that whatever diabetes educational opportunities participants, particularly the older group, have been exposed to, have not been effective. There are many potential reasons for this. For instance, the high patient numbers and multiple disease burden in primary care clinics, are likely to negatively impact on the time available for patient education by health promoters, nurses or doctors [32]. Other factors to be taken into consideration include lack of attendance at educational sessions when they take place, communication barriers, such as poor hearing, lack of concentration, inability to engage with the material presented and use of didactic modes of communication [33]. However, the participants seemed to have a better knowledge of self-management practice such as foot care and healthy eating. Whether this is because these messages are practical and easier to convey or that the information comes from multiple sources and not only health care workers is uncertain.

To date, there is a scarcity of evidence regarding diabetes self-management education and support in older adults [34]. Some studies that included older persons suggest that this group needs diabetes self-management education that stresses problem-solving skills rather than “rules” to follow [35]. For example, The Diabetes Education and Self-Management Ongoing and Newly Diagnosed (DESMOND) educators observed that older persons contributed to the group and brought valuable experience, but that they may have required a different approach at times [36]. However, no specific examples of such approaches were given in their study.

Sinclair et al. reported that older people benefitted more than middle-aged people from a highly structured group diabetes self-management education intervention with embedded cognitive behavioural strategies compared to standard group education or individual sessions with dietitians and nurse educators [37]. As older persons may have difficulties concentrating and understanding abstract concepts, there is a need for educational material to be provided in the form of simple messages, delivered in a style that engages the person with diabetes and is personalized to their needs, with the emphasis on what they need to know, rather than all there is to know about diabetes [38,39]. Notably, these concepts to enhance knowledge and self-management practices are not unique to the older person with diabetes and are relevant in all societies. Income and financial issues are possible barriers to optimum self-management for many older diabetic patients because of the costs of blood glucose testing, medication and following diet recommendations [39].

Management of chronic disease require support from a patient’s network. We found that almost 75% of participants perceived that their family supported them to follow all aspects of self-care management. Earlier studies have shown that social support and social networks influence health behaviours and health outcomes [40–41]. For example, a study of family behaviours and relationships to adherence and metabolic control, individuals with diabetes negative perceptions of support from family were associated with lower adherence to diabetes management areas (i.e. glucose testing, diet adherence, and insulin injections). On the other hand, positive impacts have been shown to affect an individual’s management of the disease. For instance, Sinclair et al. found that adherence to self-care regimens (i.e. insulin treatments, monitoring blood glucose, exercise, and self-care away from home) was associated with emotional and instrumental support from friends and family [42]. This suggests that the perceived availability and knowledge of friends and family as being present positively impacts self-management efforts of individuals with diabetes.

Furthermore, Connell et al found that social support had only a positive association with general morale among women, while there was a direct correlation between social support and adherence to treatment among older men with diabetes [43]. Among older persons, it has been found that women tend to exhibit better self-care behaviour, are less likely to be married, and are more likely to discuss personal issues with friends than men are [43]. This is in line with our study that showed that women tested their glucose levels more frequently than men.

In contrast, men are more likely to have a family member who assists with various aspects of their self-care regimen. The self-care behaviour of older women with diabetes is also influenced by social role obligations, and this is especially true of certain communities like the South African community, where women often bear a bigger responsibility as the caregiver for the whole family [43]. Unsurprisingly, such women also report a lower quality of life as well as encountering more barriers to the self-management of their diabetes [43]. Weaver et.al reported that both symptoms of diabetes and difficulties achieving socially important roles contribute to poor mental health among Indian diabetic women.[44]

Diabetes self-management education (DSME) is an integral part of diabetes care “for all people with diabetes who want to attain successful health-related outcomes,” irrespective of age [44]. It is important to be aware of current DSME guidelines for older persons and how these guidelines can be implemented in a clinical setting. However, older persons are under-represented in DSME research studies, so evidence-based guidelines specifically targeted toward older individuals are challenging to formulate [45].

The American Association of Diabetes Educators (AADE) and the American Geriatric Society (AGS) have formulated guidelines for DSME in the older adults mainly based on expert consensus [46,47]. For example, Older persons who are experiencing difficulties with daily tasks such as hearing loss, vision problems, decreased mobility and falls, will need individual rather than group DSME. Indeed, learning new skills will take longer and may require referral to a visiting nurse/CHW to make sure the task is fully integrated into the person's self-care regimen. If needed, family members or other caregivers should be included in DSME [47].

Additional information about the influence of social support on chronic illness self-management has been supported by research. A systematic review reported evidence for a modest positive relationship between social support and chronic illness self-management, particularly for diabetes [48]. The finding was that a large information network is beneficial for self-management capabilities, especially in low education populations. This may be of an advantage in many cultures such as in South Africa where strong family relationships and family caring are important and highly valued [49–51].

A cohesive and supportive family may provide older diabetic patients with an opportunity to express feelings and fears. When DSMP is reviewed as a shared responsibility with the whole family, older persons may adopt DSMP activities more easily and feel more self-confident in managing diabetes [52]. As family-focused interventions may be more effective in improving DSMP performance than individual-focused interventions, including family members or friends in education programmes should be considered [53–56].

However, the shortage of professional health care workers in South Africa highlights the need to develop alternative delivery models for education and self-management for people with diabetes who attend primary care services. These include using the services of community health workers (CHWs) and peer supporters and should draw on previous lessons learnt [56]. For example, while a pragmatic trial of a group diabetes education programme led by health promoters in Cape Town improved blood pressure, but not self‐efficacy, locus of control or glycaemia control; process evaluation suggested numerous problems. These included, finding suitable space for group education, with patient attendance and with full adoption of a guiding style by the health promoters. Thus, groups held outside of primary care clinics in the community and led by well-trained CHWs or peers may be a better option, so too may the active participation by family members in these groups [57]. In addition, the emphasis on diabetes prevention programs in middle-aged people must be highlighted, because it will enable the next generation of older persons to live with a reduced diabetes burden. For these reasons, South Africa's health care system needs to transform its services offered to older persons to reduce health care costs and ensure quality of life [57].

The various initiatives currently underway to re-engineer the primary healthcare system in SA to more effectively deal with NCDs, will go some way to meeting the identified needs of older diabetic patients and to addressing their barriers to care [58–60].However, as part of this re-modelling exercise, it is perhaps opportune for the health department to consult older chronic care patients and involve them in decision making and the planning of services. This study alerts policymakers and clinicians to some of the specific issues considered to be pertinent and important in the care and management of older persons with diabetes.

As our results show that weak social support is a predictor of both poor knowledge and poor SMP, consideration should be given to health practitioners assessing social support when people with diabetes are reviewed clinically. However, interventions need to be put in place to enhance the level of support. This could include recommending that the person be open to accepting support from family and friends, suggest that the main carer attend some clinic visits with the person or referral of the person to a support group.

Conclusions

Consideration needs to be given to developing and evaluating education programmes that focus on the needs of older people with diabetes mellitus and emphases the role of family and friends. However, it is imperative to introduce programmes at a younger age so that diabetes self-management strategies are embedded as a life course perspective to enhance positive outcomes for persons living with diabetes.

Supporting information

Acknowledgments

The authors acknowledge the Chronic Diseases Initiative for Africa (CDIA) administrative Team, all field assistants, and the study participants, who provided useful information for this study. They gratefully acknowledge support Ms. Susan Terblanche from OLRAC SPS south Africa, Tawanda Chavies, Wisdom Basera from the Department of Medicine UCT who provided technical support and assisted in the data management.