The Effects of Self-Efficacy and Physical Activity Improving Methods on the Quality of Life in Patients with Diabetes: A Systematic Review

Objective The purpose of this systematic review is to study the impact of self-efficacy-improving strategies on physical activity-related glycemic control of diabetes. Method This systematic review was conducted based on the PRISMA statement. (“Diabetes” OR “glycemic control”) AND (“exercise” OR “physical activity”) AND “self-efficacy” were searched as keywords in databases including PubMed, Google Scholar, Science Direct, Embase, Cochrane, Web of Science, and Scopus between 2000 and 2019 for relesvant articles. Results Two reviewers independently screened articles (n = 400), and those meeting eligibility criteria (n = 47) were selected for data extraction using a predesigned Excel form and critical appraisal using the “Tool for Quantitative Studies.” Different strategies and health promotion programs such as individual or group face-to-face education and multimedia (video conference, video, phone calls, short message service, and Internet-based education) were used in diabetes self-management education programs. The results of different interventions including motivational interviewing (7 studies), exercise (5 studies), multidimensional self-management programs (25 studies), and electronic education (11 studies) had been evaluated. Interventions with more social support, longer duration, combined educative theory-based, and individual education had better outcomes both in postintervention and in follow-up evaluation. Conclusion A combination of traditional and virtual long-lasting self-care promoting (motivating) programs is needed to improve patients' self-efficacy for healthy habits like active lifestyle.


Introduction
Diabetes mellitus refers to a heterogeneous group of metabolic diseases commonly resulting in high blood glucose levels (hyperglycemia). Diabetic patients are at risk of various complications that decrease their quality of life and increase mortality rates [1]. Premature death and long-term disabling complications make diabetes an expensive illness with a significant economic burden, especially in low-and middle-income countries [2]. Chronic hyperglycemia in diabetic patients leads to vascular damage (macro and micro), which is the main factor for the induction of different cardiovascular, nephropathy, retinopathy, neuropathy, and other complications [3,4]. Different kinds of synthetic antidiabetic drugs and herbal remedies with high antihyperglycemic activity [1,5,6] are in use for patients. However, current medications have not been able to slow down the development of the complications of diabetes [7]. Therefore, self-management and self-care strategies are recommended to improve the quality of life and slow down diabetic complications in patients [8]. Self-care behaviors in diabetic patients mean raising the level of knowledge and information about the complex nature of diabetes and taking actions such as controlling blood glucose, healthy eating, physical activity (PA), and foot care [9]. Although evidence suggests the positive effects of PA on diabetes management, studies have shown a low prevalence of PA in people with diabetes [3]. Improvement of patients' exercise self-efficacy might be influenced by behavior related to PA [10]. Improvement in self-efficacy would facilitate controlling the patient challenges of being physically active. Therefore, patient counselors/educators and other practitioners could beneficially construct efficacy-enhancing programs to improve patients' physical activation [11]. Moreover, recent studies suggest that selfefficacy is one of the most influential factors in the self-care of chronic diseases, especially diabetes [12]. Although several educational interventions based on the theory of self-efficacy have been done to improve self-management and glycemic control in diabetic patients, there is not enough literature review and critical appraisal interpreting the results of those studies. Therefore, a systematic review could help better evaluating the effectiveness of self-efficacy-based educational strategies. The purpose of this systematic review was to examine the impact of self-efficacy-improving strategies on PArelated glycemic control of diabetic patients.

Method
2.1. Protocol. This systematic review was conducted based on the Preferred Reporting Items for Systematic Reviews: the PRISMA Statement 15.

Search
Strategy. MESH terms such as ("diabetes" OR "glycemic control") AND ("exercise" OR "physical activity") AND ("Self-efficacy") were searched in various search engines and databases including PubMed, Scopus, Science Direct, Embase, Cochrane, Web of Science, and Google Scholar between 2000 and 2019 for a relevant article. The full text of both the randomized controlled trial and pilot studies written in English was included, and non-English abstracts, original articles, reviews, and grey literature were excluded.

Eligibility Criteria and Study Selection.
Four hundred articles were identified in the initial search. All search results were imported to EndNote X8 citation manager, and duplicate studies were removed. Two reviewers independently screened the titles and abstracts of studies to select relevant ones. Disagreements were resolved by consensus. This process resulted in the selection of forty-seven articles for review ( Figure 1).

Data Extraction and Quality Assessment.
Two reviewers separately collected data from the full texts of the included studies using a predesigned Excel form. Results were compared and double checked by the same reviewers. The data extracted included title, author, year, inclusion/exclusion criteria, design, subjects, strategies of intervention, instruments and measurement, outcome measures, conclusions, and keywords. The methodological quality and validity of each included study were evaluated independently by two reviewers using the "Tool for Quantitative Studies" [13] and Jadad score. Disagreements were resolved by discussion. Studies with no weak rating were defined as strong, with one weak rating as moderate quality, and with more than one weak rating as low (weak) quality. Meta-analysis and outcome measures were not done because of heterogeneity and low quality of the study. Publication bias and statistical analysis were not checked because of low quality and heterogenic studies.

Results
3.1. Self-Efficacy and PA Improvement in Diabetic Patients. Different strategies such as individual or group face-to-face education and multimedia electronic education (education), including video conference, video phone calls, short message service or SMS, and Internet-based education, have been used in diabetes self-management education programs. In addition, motivational interviewing (MI), exercise and education classes, Healthy Eating and Exercise Lifestyle program (HEELP), theory-based group workshops, narrative-based intervention program, peer education program according to health belief model (HBM), home-based exercise program, and other self-management programs had been mentioned in these studies for promoting self-efficacy-related behaviors such as PA in diabetic patients.

3.2.
Role of MI. The effectiveness of MI in both patients and nurses to develop behavioral changes related to diabetes selfefficacy has been reported. Seven studies used MI as a single educational program or in combination with exercise training to improve self-efficacy and behavioral changes regarding to diabetes self-management. The summarized characteristics of the studies and their quality rating are shown in Table 1. Postintervention evaluation showed improvement of diabetes self-efficacy, active lifestyle, and glycemic control in 5 studies [14][15][16][17], and self-efficacy was determined as the main predictor of intention to PA [18]. The motivational intervention was more feasible in women and individuals with a higher educational level [16,17]. The effect of self-efficacy and intention on exercise performance was mediated by planning strategies [18]. Although these studies reported the positive effect of MI on diabetes self-efficacy as the main predictor of PA intention, there are also negative results [19]. In addition, nurses training for MI of diabetic patients had no significant effect on lifestyle behaviors such as healthy diet, physical activity, and self-efficacy in patients [20].

Role of Health Promotion Programs.
Although programs targeting only exercise behavior resulted in patients' active lifestyle behaviors, they did not improve patients' glycemic control. We found five studies using exercise training interventions targeting behavioral changes in diabetes control. Self-monitoring of exercise, home-based resistance training, home-based walking, combination of resistance, and endurance training were the related interventions. Although in some studies, exercise training improved PA self-efficacy [21,2,22], and baseline evaluation showed an association between walking ability and self-efficacy; however, in one study, exercise training did not improve self-efficacyrelated outcomes of diabetic patients [23]. In addition, glycemic control index (HbA1C) and markers of cardiovascular risk changes were nonsignificantly changed in both intervention and control groups [2]. Low self-efficacy for resistance exercise was the most important predictor of patients' dropout [24], and effective interactions between patients and health care professionals are recommended to encourage patients for behavioral changes and overcoming the barrier to PA [21]. Moreover, it seems that the existence of an underlying disease associated with diabetes has a great impact on study outcomes. It was indicated that individuals without MetS had higher exercise self-efficacy than those with MetS and that home-based exercise programs are beneficial for individuals at risk for diabetes [22]. Table 2 shows a summary of studies using exercise interventions.
Health promotion programs also have been used to evaluate the role of self-efficacy and PA in diabetes management. Different long-term interventions, including Mediterranean Lifestyle Program [25], primary care-based walking program (24 weeks), education programs on exercise-related behavioral changes based on the HBM [26], Healthy Eating and Active Living for Diabetes in Primary care networks (HEALD) program [27], and proactive coping [28], have shown to improve the participants' quality of life (healthy diet, exercise, and stress management) and psychosocial factors (self-efficacy and problem-solving). In some studies, the beneficial effects were sustained even at follow-up evaluation [26], and postprogram contact with patients could improve those outcomes [27]. In another study, with a brief lifestyle self-management program, using follow-up phone calls induced effective lifestyle behavior changes; however, selfefficacy was not increased in the intervention group [29].
Moreover, a combination of theory-based group workshops and walking exercise had a better short-term impact on self-regulation/self-efficacy and PA than online education, but these beneficial effects declined at six-month follow-up [30]. The HEELP program also improved patients' exercise adherence and weight loss; however, the results showed that male gender, self-efficacy, time, and depressive symptoms are independent predictors for exercise duration or BMI change. In addition, lack of motivation and time was the most common exercise barriers at baseline, and there was a negative association between lack of motivation and exercise self-efficacy even after 12-month program [31]. Moreover, other baseline factors, including obesity, coronary heart disease, female gender, self-efficacy, and depressive symptoms, need more attention in designing such programs [32]. Familial factors also might influence the patient's adherence to lifestyle changes. The health stress of patients in the form of higher comorbidity number and specific stress of diabetes in both patients and spouse was inversely correlated with patient adherence to exercise and dietary programs. These effects were mediated by diabetes self-efficacy and depressive symptoms reported by couples [33].
The effectiveness of individual or group selfmanagement improvement methods has been evaluated too. Two studies showed that the patient-centered group education and the structured goal-setting method would lead to better patients' self-management, and the effect of time-by-treatment interaction might partially be mediated via the development of self-efficacy 38 [34]. ,In another RCT study, individualized education (IE) had better outcomes compared to group education [35], and long-term evaluation indicated behavioral and psychological improvement in IE; however, this intervention did not show sustained improvement in HbA1c, nutrition, and PA scores 3 Journal of Diabetes Research      Journal of Diabetes Research [36]. Face-to-face education program targeting self-efficacy on self-care skills resulted in better patient glycemic control, diet, medication adherence, and PA improvement [37,38]. Face-to-face and five telephone lifestyle counseling sessions on changing the psychosocial determinants of PA and diet also improved patients' self-efficacy and reduced barriers to active lifestyle such as lack of motivation and energy in the intervention group compared to the control [39]. In a RCT study, a program for improvement of worksite lifestyle in prediabetes employees resulted in better behavioral outcomes such as PA and diet self-efficacy and goal commitment [40]. A self-management coaching program on lifestyle changes had more impact on people with lower selfefficacy [41] and social cognitive (self-efficacy) and selfregulatory (illness beliefs) theory-based intervention programs caused a significant improvement in self-efficacy for exercise [42]. Data showed that illness beliefs play an essential role in patients' quality of life, while self-efficacy had a crucial role in self-management behaviors diabetes care providers [42].
Other interventions such as a narrative-based intervention program [43], Spanish Diabetes Self-Management Program [44,45], nurse-managed health promotion program [46], and prevention program on self-efficacy [47] improved self-management and self-efficacy controlling the disease, although an independent association between social-environmental, problem-solving, and self-efficacy factors with exercise and diet-related behaviors has been reported. However, the development of these psychosocial and socialenvironmental factors could improve diabetes selfmanagement [48]. Improvement of the knowledge about the importance of exercise and self-efficacy in diabetes care providers leads to better performance in patients' exercise learning [49]. Among diabetic patient counselors/educators, factors such as "time allotted for delivering diabetes selfmanagement/support visits" and "inability to engage patients in physical activity" were identified as practice and challenging barriers. To improve physical self-efficacy in patients, educators challenging problems need attention [50]. Table 3 shows a summary of studies using health promotion programs.

Role of Multimedia and Education.
The modulatory effect of self-efficacy on increasing self-care behaviors of diabetic patients was evaluated using different multimedia-and education-based interventions. Education of diabetic patients using a multimedia-(CD-) based health promotion Randomized controlled trial; prediabetic university employees (n = 68) Sixteen-week groupbased diabetes prevention program +3-month follow-up Self-efficacy, behavioral self-regulation, and goal setting determinants were assessed at baseline, postintervention, and 3month follow-up Improvement in behavioral outcomes such as physical activity and diet self-efficacy and goal commitment in the intervention group Strong 9 Journal of Diabetes Research  [52]. Brief proactive telephone "coaching" interventions also increased patient adherence to exercise and a healthy diet and reduced medical complications and depression. Results showed the beneficial impact of awareness of self-care goals, self-efficacy, and reinforcement on foot inspection, psychological symptoms (depression), and PA [53]. Diabetes educators could apply integrative health coaching for the improvement of patient self-efficacy [54]. In a tailored Internet-based intervention, patients with the highest self-efficacy had better outcomes; therefore, self-efficacy may play a moderator role in intervention outcome and should be considered in tailoring educational intervention for diabetes [55]. In addition, online program (algorithm-driven) for diabetes prevention and improvement of diabetes self-management, self-efficacy and satisfaction, can result in promoting PA behavior [56]. However, although online education was shown to improve HbA1C, exercise, patient activation, and self-efficacy, but reinforcement or follow-up had no beneficial effect [57]. Smartphone communication also increased the patients' self-efficacy compared to the control group [58]. However, multimedia education had a better effect compared with short message service (SMS)-based model on patients' self-efficacy and their belief about PA behavior [59].
There are also studies with negative results. Computerbased multimedia program in the waiting administration room of diabetic patients had no significant difference in glycemic control, self-efficacy, and other self-management behavior related to diabetes [60]. Moreover, one-month mobile-based intervention pilot study did not show any significant changes in patients' glycemic control, self-efficacy about food intake, PA, and body mass index [61]. Table 4 shows a summary of studies using multimedia and education.

Discussion
Motivation had been introduced as a pivotal factor for the improvement of lifestyle, especially in behavioral and psychological aspects, because it increases the learner's effort and desire for a certain change and purpose [62]. MI as a single strategy or in combination with other programs (exercise, healthy diet) has been performed in seven studies. The duration of the studies (RCT and pilot) was between 2 and 12 months, and their population size was 12-152. Although four studies reported improved self-efficacy and PA, one study showed no change in PA. Encouragement of the patient in achieving the goals of diabetes self-management shall be considered as a cost-benefit method in education even with no change in HbA1c and PA.
Our search resulted in five studies with a population of 48-145 which evaluated the effect of endurance or resistance exercise behavior on health-related behaviors (e.g., exercise self-efficacy) and/or glycemic control. According to them, low efficacy of exercise has been proposed as a significant predictor of patients dropping out, and just one study reported the improvement in patients' self-efficacy but had no effect on glycemic control or other diabetes complications. The goal of active life is to improve metabolic status and reduce the complications of diabetes. Moreover, most people with diabetes or metabolic disease have low self-efficacy, quality of life, and knowledge/belief about their illness [22]. Therefore, multifaceted health promotion programs should be applied to cover all psychological and behavioral aspects of lifestyle and induce effective changes in patients' beliefs. A systematic review about lifestyle intervention in diabetic patients suggested future interventions targeting health promotion behaviors with emphasize on problemsolving skills and self-efficacy; but there was no recommendation for the best strategies [8].
A multiconceptual basis education strategy (a combination of goal system and social, cognitive, and ecological theory) was associated with better outcomes. In this survey, twenty-five studies with a sample size of 62-550 and a duration of 3 weeks to one year had been assessed, which used multidimensional self-management programs with both individual/group face-to-face sessions and multimedia training. According to the findings, interventions with more social support, longer duration, combined educative theory-based, and individual education had a better outcome after intervention and follow-up evaluation. In addition, the improvement of the knowledge and self-efficacy of diabetes care providers has not resulted in an increase of exercise self-efficacy in patients with diabetes.
Recently, researchers have been interested in educational technologies such as online and virtual training, multimedia, and smartphone health informative applications to provide more effective health promotion interventions. We found eleven trials with sample sizes of 56-760 and duration trials of 1-24 months. Although patients' feedback about participation in e-education was positive, however, as similar as face-to-face methods, two studies with a small study population and short duration showed no change in outcome. It seems that poor baseline motivation, self-efficacy, and depressive symptoms need more attention in designing such programs [32]. In literature review 2, systematic reviews and meta-analysis evaluated the effectiveness of 15 and 16 studies based on "peer support on self-efficacy" and "selfefficacy-focused education." Although peer support did not induce any significant change in self-efficacy and quality of life, however, intervention with long duration (>6 months) had a better effect on patients improvement of quality of life [63] which is in line with findings of the present study. Meta-analysis of 10 selected studies from 16 interventions showed the beneficial impact of "self-efficacy-focused education" on glycemic control and quality of life in a patient with type 2 diabetes, but the lack of high-quality rating studies with good emotion/physiological strategies and complete outcome assessment makes it difficult to choose the best strategies [64].
In this systematic review, different methodological approaches for the development of self-efficacy and physical activity in diabetic patients had been summarized and discussed to facilitate the patients' and researchers' access to available studies and their outcomes. This review tried to show the importance of self-management programs in 11 Journal of Diabetes Research controlling diabetes and emphasized the need for designing most effective methods in improving self-efficacy-focused education. However, this systematic review has several limitations worth mentioning. First, most of the studies were rated as moderate and weak quality with performance bias and detection bias (i.e., lack of double blind, standard randomization, and description of withdrawal). Secondly, the studies were heterogeneous. Study's characteristics, such as population (i.e., number, sex, race, age, education, and concomitant disease), inclusion and exclusion criteria, duration, design (i.e., RCT, prospective observational study, and crosssectional study), and self-management improving methods were heterogeneous. The lack of enough studies with RCT design and limited number of participants in them make meta-analysis impossible. Moreover, the outcomes of the studies, especially with respect to behavioral outcomes, were also heterogonous because different scales had been employed for self-efficacy and self-management assessment, and different primary and secondary outcomes had been reported. Finally, we evaluated the available English reports (full text) of studies; therefore, potentially relevant reports in other language might have been missed. Taken together, the most important limitation of this study was insufficient high-quality RCTs with enough sample size, long-term education, and follow-up periods, which applied physiological/ emotion arousal educational strategies and employed complete outcome assessments with standard scale. Therefore, we could not evaluate the validity and reliability of the instruments and the related outcomes. Regarding those limitations, it is difficult and even impossible to perform a metaanalysis study and combine the findings for achieving descriptive and practical conclusions. Therefore, the impact of self-efficacy-focused education programs including practicing the self-efficacy improvement skills, peer models, goal setting, positive feedback, and health provider persuasion methods on diabetes management is still under question.

Conclusion
A combination of traditional and virtual long-lasting selfcare promoting (motivating) programs with good emotion/ physiological strategies is needed to improve patients' selfefficacy for healthy habits like an active lifestyle. Family and social support play an essential role in establishing healthy behavioral changes in diabetic patients. Future high-quality RCT studies with larger sample size, selfefficacy-focused education-based strategies, long duration and follow-up, and standard outcome assessments are needed to evaluate the effectiveness of self-management strategies.

Data Availability
There is no raw data associated with this review article.

Conflicts of Interest
The authors declare no conflicts of interest in this study.

Authors' Contributions
Sajjad Hamidi was responsible for the database search, screening titles and abstracts, and data collection and extraction. Zahra Gholamnezhad was responsible for the conception and design of the study, screening the titles and abstracts, data extraction and quality assessment, and draft manuscript preparation and approved the final version of the manuscript. Narges Kasraie was responsible for the draft manuscript preparation and English native editing. Amirhossein Sahebkar advised the study and manuscript.