Variables that have been considered to be correlates of various adherence behaviours in diabetes can be organized into four clusters:
- treatment and disease characteristics;
- intra-personal factors;
- inter-personal factors; and
- environmental factors.
A. Treatment and disease characteristics
Three elements of treatment and of the disease itself have been associated with adherence: complexity of treatment, duration of disease and delivery of care (see also Table 4).
In general, the more complex the treatment regimen, the less likely the patient will be to follow it. Indicators of treatment complexity include frequency of the self-care behaviour - i.e. the number of times per day a behaviour needs to be performed by the patient. Adherence to oral hypoglycaemic agents has been associated with frequency of dosing. Higher adherence levels were reported by patients required to take less frequent doses (a once-daily dose), compared to those prescribed more frequent doses (three times daily) (48). Dailey et al. (50) showed that patients prescribed a single medication had better short-term and long-term adherence rates than patients prescribed two or more medications.
Duration of disease appears to have a negative relationship with adherence: the longer a patient has had diabetes, the less likely he or she is to be adherent to treatment. Glasgow et al. (21) studied a sample of patients with type 1 diabetes (mean age = 28 years), and found that level of physical activity was associated with duration of disease. Patients who had had diabetes for 10 years or less reported greater energy expenditure in recreational physical activities, and exercising on more days per week, than those with a longer history of diabetes. Patients with a longer history of diabetes also reported eating more inappropriate foods, consuming a greater proportion of saturated fats and following their diets plans less well. More recently, in a study conducted in both Polish and American children with type 1 diabetes (41), duration of disease was also associated with adherence to insulin administration, as children with a longer history of diabetes were more likely to forget their insulin injections than children who had been diagnosed more recently.
Delivery of care for diabetes can vary from intensive treatment delivered by a multidisciplinary diabetes team, to outpatient care delivered by a primary care provider. Yawn et al. (57) observed interactions between patients and providers in a family practice setting and reported that patients with diabetes seen specifically for their diabetes received more counselling on diet and adherence than patients with diabetes seen for an acute illness. Kern & Mainous (58) found that although physicians preferred to follow a planned, systematic strategy for treating diabetes, acute illness and failure of patients to adhere forced them to spend less time on diabetes care.
Adherence can also be affected by the setting in which care is received. Piette (59) examined the problems experienced by patients in accessing care in two public health settings in the United States and found that the cost of care was a major barrier to access, especially for patients in a community treatment setting. Perceived barriers to access to care were also associated with poor metabolic control.
B. Intra-personal factors
Seven important variables have been associated with adherence: age, gender, self-esteem, self-efficacy, stress, depression and alcohol abuse.
Age of the patient has been associated with adherence to physical activity regimens in a sample of patients with type 1 diabetes (21). Compared to younger participants, patients over 25 years of age reported exercising on fewer days per week, and spending less time (and expending fewer calories) in recreational physical activities. There were no associations reported between age and adherence to other self-care measures.
Age has also been associated with adherence to insulin administration in a study of adolescents with type 1 diabetes. The investigators found that older adolescents were more likely to mismanage their insulin (missing injections) than their younger counterparts (42). In a study assessing adherence to self-monitoring of blood glucose, younger adolescents reported monitoring their blood glucose concentrations more frequently than did the older ones (60). Older adults may also practice better self-management than younger adults (61).
Gender has also been associated with adherence. The men in a sample of patients with type 1 diabetes (21) were found to be more physically active than the women, but they also consumed more calories, ate more inappropriate foods and had lower levels of adherence as assessed using a composite measure of diet.
Self-esteem has been associated with adherence to self-management of diabetes among patients with type 1 diabetes. High levels of self-esteem were related to high levels of adherence to physical activity regimens, adjustment of insulin doses and dental self-care (62).Murphy-Bennett, Thompson & Morris (63) found that lower self-esteem in adolescents with type 1 diabetes was associated with less frequent testing of blood glucose.
Self-efficacy has been studied in relation to adherence to prescribed treatments for diabetes. In a combined sample of patients with type 1 and type 2 diabetes in Canada (64), a measure of diabetes-specific self-efficacy beliefs was found to be the strongest predictor of energy expenditure suggesting a positive relationship between self-efficacy and adherence to prescribed physical activity. Senecal, Nouwen & White (65) reported that beliefs in self-efficacy were a strong predictor of adherence and that both self-efficacy and autonomy predicted life satisfaction. Ott et al. (66) found that self-efficacy was a predictor of adherence to diabetes care behaviours in adolescents with type 1 diabetes. Aljasem et al. (67) showed that self-efficacy beliefs predicted adherence to a prescribed regimen in 309 adults with type 2 diabetes after controlling for health beliefs and perceptions of barriers.
Stress and emotional problems are also correlated with adherence. Fewer minor stressors were associated with higher levels of adherence to insulin administration and diet in women with gestational diabetes (56,68). In a study using a diabetes-specific stress scale in a combined sample of adults with type 1 and type 2 diabetes (69), stress was found to be significantly associated with two aspects of the diet regimen (diet amount and diet type). However, no associations were found between stress and adherence to physical activity regimens or glucose testing in this sample. Peyrot et al. (70) reported that psychosocial stress was associated with poor adherence to a prescribed regimen and poor metabolic control in a mixed group of patients with type 1 and type 2 diabetes. Mollema et al. (71) reported that patients who had an extreme fear of insulin injections or self-monitoring of blood glucose had lower levels of adherence and higher levels of emotional distress. Schlundt, Stetson & Plant (72) grouped patients with type 1 diabetes according to the problems they encountered in adhering to prescribed diets and found that two of the groups of patients - emotional eaters and diet-bingers - had adherence problems related to negative emotions such as stress and depression.
Depression. The incidence of depression has been observed to be twice as high among persons with diabetes than in the general population (73). Patients with depression are more likely to experience complications of diabetes (74), have worse glycaemic control (75), and be less adherent to self-care behaviours than patients who are not depressed. Depression is also associated with higher costs of medical care in patients with diabetes (76).
Alcohol abuse. Patterns of alcohol use have been related to the quality of diabetes self-management. Johnson, Bazargan & Bing (77), studied 392 patients with type 2 diabetes from ethnic minority groups in Los Angeles, CA, and found that alcohol consumption within the previous 30 days was associated with poor adherence to diet, self-monitoring of blood glucose, oral medications and appointment-keeping. Cox et al. (78) examined alcohol use in 154 older men with diabetes and found that greater alcohol use was associated with poorer adherence to insulin injections.
C. Inter-personal factors
Two important inter-personal factors: the quality of the relationship between patients and providers of care, and social support, have been found to correlate with adherence. Good communication between patient and provider has been related to improved adherence. Among patients with type 2 diabetes, adherence to administration of oral hypoglycaemic agents and glucose monitoring were significantly worse in patients who rated their communication with their care provider as poor (79).
Social support has been the subject of much research. Greater social support was found to be associated with better levels of adherence to dietary recommendations and insulin administration in women with gestational diabetes (68). Parental involvement, as a measure of social support, has also been associated with adherence to blood glucose monitoring. Adolescents and children with type 1 diabetes, who experienced greater parental involvement with their blood glucose monitoring, reported higher levels of daily checks of blood sugar concentrations (60).McCaul et al. (21) followed a sample of adolescents and adults with type 1 diabetes. For both adults and adolescents disease-specific social support was associated with better adherence to insulin administration and glucose testing. For the adolescent group only, general family support was associated with adherence to insulin administration and glucose testing. The study found no association between any of the social support measures and adherence to diet and physical activity regimens. Other studies have shown a relationship between poor social support and inadequate self-management of diabetes (80 - 84).
D. Environmental factors
Two environmental factors - high-risk situations and environmental systems - have been linked to poor adherence in patients with diabetes. Self-care behaviours occur in the context of a continually changing series of environmental situations at home, at work, in public, etc., which are associated with different demands and priorities. As their circumstances change, patients are challenged to adjust and maintain their self-care behaviours. Patients are frequently called upon to choose between giving attention to diabetes self-management or to some other life priority. Situations associated with poor adherence have been called "high-risk" situations (85).
Schlundt, Stetson & Plant (72) created a taxonomy of high-risk situations that posed difficulties for patients following diet prescriptions. The situations included: overeating in response to people, place and emotions; situations associated with under-eating, and difficulty in integrating food intake according to social context, time of day and place. Schlundt et al. (82) described 10 high-risk situations for poor dietary adherence that included social pressure to eat; being alone and feeling bored; interpersonal conflicts, and eating at school, social events or holidays. Schlundt et al. (83) identified 12 categories of high-risk dietary situations in adults with type 1 and type 2 diabetes: these included resisting temptation, eating out, time pressure, competing priorities and social events. Other studies have also shown that environmental barriers are predictive of adherence to various aspects of diabetes self-care (34,67,86).
Many environmental factors that influence behaviour operate on a larger scale than the immediate situation confronting a person (87). These environmental systems include economic, agricultural, political, health care, geographical, ecological and cultural systems (88). The large-scale environmental changes that occurred in the twentieth century created the current epidemics of obesity and type 2 diabetes (89 - 91). These changes included increased availability of inexpensive fast foods high in fat, salt and calories (92), and the mechanization of transport systems (93,94). Changes in economic and political systems have allowed women to move into the workforce, but these same changes have altered the composition of families and the way in which families deal with food selection and preparation (95,96). Large corporations spend billions of dollars each year on marketing foods high in fat and calories (97). Increasing segments of the population spend many hours per day in sedentary activities. These activities have been linked to obesity in both children and adults (98 - 101) and to the risk of developing type 2 diabetes (102).
Some authors have described the current environment as "toxic" to healthy lifestyles (103,104). The incidences of both obesity and diabetes are rapidly increasing in developing nations and are likely to be associated with urbanization, mechanized transportation and widespread changes in food supply. The same factors that encourage sedentary lifestyles and the over-consumption of food, and lead to obesity and diabetes, probably also make it difficult for people who do develop diabetes to adhere to best-practice protocols.
Many people in developed nations, including the poor and members of ethnic minority groups, have to some degree been bypassed by the economic prosperity of the twentieth century. It is these groups that have been most adversely affected by the environmental changes that lead to disparities in health status (105,106). Even living in a poor community can contribute to poor health outcomes (107).
Given the powerful influence of these larger social factors, it is important to avoid over-attributing the responsibility for adherence to patient-related factors or to health care providers (108). A patient's ability to manage his or her behaviour, achieve tight metabolic control and prevent the long-term complications of diabetes is determined by a host of intra-personal, inter-personal and environmental factors that interact in ways that are not yet understood (27,109).