Main findings
This study showed that there were significant associations between some socio-demographic characteristics (sex, age, living status, perceived health status, and diabetes duration) and adherence to several diabetes and hypertension self-management behaviors (medication therapy, diet therapy, regular exercise, tobacco and alcohol avoidance, and self-monitoring/self-care). It also showed that self-efficacy in performing self-management significantly mediated the association between age and adherence to diet therapy.
Notably, we found that female patients were less likely to exercise regularly than male patients, consistent with previous observations [
40‐
42]. Possibly, women shoulder more responsibilities for taking care of the family, which leaves them with less time for exercise [
41‐
43]. Women also reported more barriers to exercise (e.g., a lack of safe and appropriate sports facilities, fear of injury during exercise, and a lack of relevant skills and knowledge) than their male counterparts [
40,
41]. Interventions that may help overcome these barriers, such as the provision of safe and easily accessible sports facilities, and the provision of adequate training for women to learn to exercise safely, should be considered. Another reason for the comparatively low exercise level of women may be that women received less social support (e.g., from family and friends) for performing physical activities, which is known to be an important determinant of exercise adherence [
44]. In addition, lack of an exercise companion is also a commonly cited reason for not being physically active, particularly among women [
45]. To help improve adherence to exercise, it has been suggested that interventions (e.g., technology and group-based exercise programs) be designed to enable the connection of companions/caregivers with patients, such that patients are encouraged and motivated to exercise [
46‐
48].
We also found that male patients were less likely to reduce their alcohol consumption and smoking than female patients. This may be because social interactions between men are more likely to involve tobacco and alcohol [
49]. Also, men may perceive smoking and alcohol consumption to be desirable masculine behaviors [
50]. Another explanation may be that men are more likely to use tobacco and alcohol as a (maladaptive) coping mechanism to deal with stress [
51]. If so, this may be ameliorated by interventions that promote the use of adaptive coping strategies (e.g., exercise, and talking to family and friends) to actively cope with stress [
52]. Research has also shown that the perceived risk of negative consequences resulting from smoking and drinking alcohol is much smaller among men [
53], suggesting that male patients need to be further educated about and alerted to the consequences of smoking and drinking alcohol.
The observed associations of younger age with poorer adherence to diet therapy and self-monitoring/self-care and the finding that the former correlation was mediated by self-efficacy in performing self-management may be attributable to the time and effort required to perform self-management. A previous study reported that “not having enough time” may be a major hindrance to the performance of self-management [
54]. Therefore, older patients, who may have fewer work responsibilities and thus more spare time, may be more willing to or may find it less challenging to perform disease-related self-management [
55]. Another study suggested that the daily lives of older people are more ordered and predictable [
56,
57]. Accordingly, older patients may find it easier to integrate regular self-monitoring or other self-care behaviors into their more routine daily lives.
We further observed that patients who lived alone were less likely to perform physical exercise than those who lived with family/others. This may have resulted from the latter group receiving more social support from their family or companions [
58]. A previous study reported that family members can provide patients with reminders and assistance to support self-management behaviors, as well as emotional support [
59]. All of these factors may enhance adherence to self-management behaviors. However, it was found that patients who lived with family/others were less likely to avoid smoking and alcohol than those who lived alone. This may be due to the possible smoking or drinking behaviors of cohabitating family members, which would make it more difficult for the patients to avoid such behaviors [
60]. This is consistent with previous findings that individuals’ health behaviors are strongly influenced by the lifestyles of their close associates (e.g., family members and friends) [
61]. Our findings and those of previous studies suggest that the self-management behaviors of patients are strongly influenced by the people with whom they live.
We also found that patients who reported a longer diabetes duration were more likely to adhere to self-monitoring/self-care practices, possibly because they had more regularly attended clinics for follow-up consultations. This is consistent with previous reports of a positive association between the number of follow-up consultations with a physician and patient adherence [
14,
62]. We infer that more follow-up visits with a physician may enable patients to gain more knowledge about their disease and health condition and thus increase their motivation to perform self-management.
We further observed an association between better perceived health status and better adherence to medication therapy, which is consistent with previous studies [
11,
63‐
65]. This may be attributable to patients with better perceived health status having better levels of physical functioning [
66] and thus being more capable of performing self-management [
67]. Another possible explanation is that patients with better perceived health status may have more positive health attitudes, leading to their exhibiting enhanced adherence to medication therapy [
11]. This explanation is partly supported by our finding that better perceived health status was related to more positive health attitudes, but we did not find a significant association between health attitudes and patient adherence. Further validation is needed of the interrelationships among perceived health status, health attitudes, and patient adherence to self-management.
Our analysis showed that higher self-efficacy in performing self-management was related to better adherence to diet therapy and medication therapy and greater participation in regular exercise. Patients who expressed a higher level of self-efficacy perceived themselves to be more capable of self-management and thus made greater efforts to perform self-management. In contrast, patients who expressed a lower level of self-efficacy perceived themselves to be less capable of self-management, which presumably meant that they were more likely to cease performing self-management at an early stage [
68]. Our findings are therefore consistent with those of previous studies and further emphasize the role of self-efficacy as a major determinant of adherence to a healthy diet and physical exercise [
12,
69‐
71]. We infer that improving patients’ self-efficacy in performing self-management may improve patients’ adherence to the above-described self-management behaviors [
27].
In addition, we found that self-efficacy mediated the association between older age and better adherence to diet therapy, suggesting that a lack of self-efficacy was a reason for poor adherence to healthy diets among younger patients. Therefore, for this patient group, interventions that increase their self-efficacy in maintaining a healthy diet, such as by helping them make informed decisions about diet plans, enhancing their capability to cope with interruptions of diet plans, and eliciting appropriate social support for maintaining a healthy diet, may be effective in improving their adherence to diet therapy [
72]. In a further analysis of the mediating effect of self-efficacy, we found that a higher education level was related to greater self-efficacy, but this did not translate to improved adherence to self-management. This finding is different from that of a previous study where self-efficacy mediated the association between education level and self-management behaviors [
31]. Further research into the role of self-efficacy in shaping patient adherence to self-management behaviors is warranted.
Implications for future research
Type 2 diabetes and hypertension often coexist in patient populations, and there is a considerable overlap between their complications and mechanisms [
73]. However, few studies have examined patients with coexisting type 2 diabetes and hypertension and even fewer have investigated these patients’ self-management strategies. Future research is warranted to investigate the barriers to and facilitators of self-management among these patients.
The self-management of type 2 diabetes and of hypertension have a number of commonalities, as both require long-term medication therapy, the maintenance of a healthy diet, regular exercise, cessation of smoking and alcohol consumption, stress reduction, and self-monitoring [
1], but there may be differences in the ease with which these conditions can be managed. Future research to investigate and compare patient adherence to diabetes-specific and hypertension-specific self-management behaviors is also warranted.
The patient groups exhibiting non-adherence to different self-management behaviors varied greatly, suggesting that the factors shaping these self-management behaviors (medication therapy, diet therapy, exercise, tobacco and alcohol avoidance, and self-monitoring/self-care) may also have varied greatly. It is therefore suggested that future studies conduct in-depth investigations of the underlying reasons for patient non-adherence to each specific self-management behavior. Moreover, further research is warranted to identify the mediators between socio-demographic characteristics and patient adherence, as these would reveal the underlying reasons for non-adherence in certain socio-demographic groups and thus enable the development of bespoke strategies to effectively improve patient self-management in these groups.
As noted above, “not having enough time” may be a major hindrance to the performance of self-management, especially for patients with high levels of work and/or family responsibility and a lack of spare time. This finding suggests the need for more research to identify self-management behaviors that would yield significant health benefits while simultaneously saving time. Few studies have systematically investigated the amount of time needed to perform type 2 diabetes and hypertension self-management behaviors. Therefore, we recommend additional research in this area. We also suggest that future evaluations of self-management interventions should consider and investigate the time-consuming or time-saving nature of the interventions [
54].
Implications for practice
Given the observed associations between self-efficacy and patient adherence, we suggest that interventions intended to improve patient adherence should consider both behavioral and psychological aspects. Previous research has suggested that patient education interventions that merely provide health knowledge may not sufficiently induce the desired behavioral changes [
74]. Accordingly, the incorporation of psychological components into traditional patient-education interventions may yield improvements in patient adherence. Approaches that comprise collaborative care, enhanced patient–physician interaction, and patient empowerment may improve patients’ sense of self-efficacy regarding self-management and thus improve patient adherence [
75,
76]. Moreover, we suggest that these approaches should be applied to younger patients and patients with lower education levels, given the observed associations of these characteristics with poorer self-efficacy.
We examined why patients with a lower education level tended to have lower self-efficacy in performing self-management, and found in the literature that patients with lower education levels tended to have poor health literacy [
77], which is significantly related to poor self-efficacy [
78]. Accordingly, improving the health literacy of patients with lower education levels may be an effective approach to enhance these patients’ self-efficacy in self-management.
Although several self-management interventions have been shown to yield improvements in patient adherence, few have been specifically developed to target patients in certain socio-demographic groups or to improve adherence to certain self-management behaviors. We suggest that the development of bespoke interventions for specific patient groups would more effectively improve patient adherence. In general, the use of networking and mobile technology could be considered for these purposes [
32,
33,
79‐
81]. More specifically, smartphone-based schedule planning and reminder systems could be developed to enable younger patients to fit self-management behaviors (e.g., self-monitoring and exercise) into their busy work schedules, and to remind and motivate them to adhere to health-care behaviors. For female patients, group-based exercise programs may be an effective intervention to promote their participation in physical exercise [
47].
As a lack of social support may explain why patients who lived alone reported poorer adherence to self-management behaviors, interventions that promote social support (e.g., through community care services and patient-support groups) may effectively improve the level of adherence in this population. Analogously, the health behaviors of patients who lived with family/others may have been strongly influenced by the behaviors of their cohabitants, suggesting that interventions that educate a patient’s cohabitants about a patient’s disease, and that encourage their cohabitants to be supportive of a patient’s self-management behaviors, may improve a patient’s adherence to self-management [
82,
83].
In the development and implementation of interventions that aim to promote self-management adherence, such as caregiver-assisted programs or technology-based support, the design, usability, and acceptance of the interventions should be carefully studied and considered. This is necessary to prevent avoid unintended, undesirable effects on the implementation effort or on patient outcomes due to mismatches between the interventions and patients’ needs and characteristics [
33,
48,
79,
84‐
89].