Health education for patients with acute coronary syndrome and type 2 diabetes mellitus: an umbrella review of systematic reviews and meta-analyses

Objectives This umbrella review aimed to identify the current evidence on health education-related interventions for patients with acute coronary syndrome (ACS) or type two diabetes mellitus (T2DM); identify the educational content, delivery methods, intensity, duration and setting required. The purpose was to provide recommendations for educational interventions for high-risk patients with both ACS and T2DM. Design Umbrella review of systematic reviews and meta-analyses. Setting Inpatient and postdischarge settings. Participants Patients with ACS and T2DM. Data sources CINAHL, Cochrane Library, Joanna Briggs Institute, Journals@Ovid, EMBase, Medline, PubMed and Web of Science databases from January 2000 through May 2016. Outcomes measures Clinical outcomes (such as glycated haemoglobin), behavioural outcomes (such as smoking), psychosocial outcomes (such as anxiety) and medical service use. Results Fifty-one eligible reviews (15 for ACS and 36 for T2DM) consisting of 1324 relevant studies involving 2 88 057 patients (15 papers did not provide the total sample); 30 (58.8%) reviews were rated as high quality. Nurses only and multidisciplinary teams were the most frequent professionals to provide education, and most educational interventions were delivered postdischarge. Face-to-face sessions were the most common delivery formats, and many education sessions were also delivered by telephone or via web contact. The frequency of educational sessions was weekly or monthly, and an average of 3.7 topics was covered per education session. Psychoeducational interventions were generally effective at reducing smoking and admissions for patients with ACS. Culturally appropriate health education, self-management educational interventions, group medical visits and psychoeducational interventions were generally effective for patients with T2DM. Conclusions Results indicate that there is a body of current evidence about the efficacy of health education, its content and delivery methods for patients with ACS or T2DM. These results provide recommendations about the content for, and approach to, health education intervention for these high-risk patients.

Acute coronary syndrome (ACS) is the leading cause of death worldwide. The risk of high mortality rates relating to ACS is markedly increased after an initial cardiac ischemic event. 1 Globally, 7.2 million (13%) deaths are caused by coronary artery disease (CAD), 2 and it is estimated that more than 780,000 persons will experience ACS each year in the United States. 3 Moreover, about 20-25% of patients with ACS reportedly also have diabetes mellitus (DM); predominantly type 2 diabetes mellitus [T2DM]). 4 5 Patients with ACS and DM have an increased risk of adverse outcomes such as death, recurrent myocardial infarction (MI), readmission, or heart failure during follow up. 6 Longer median delay times from symptom onset to hospital presentation, have been reported among patients with ACS and DM than patients with ACS alone. 7 DM is now considered to confer a risk equivalent to that of CAD for patients for future MI and cardiovascular mortality. 8 Mortality was significantly higher among patients with ACS and DM than among patients with ACS only following either ST segment elevation myocardial infarction (STEMI) (8.5% [ACS and DM] versus 5.4% [ACS]) or unstable angina/non-STEMI (NSTEMI) (2.1% [ACS and DM] vs. 1.1% [ACS]). 9 ACS and T2DM are often associated with high-risk factors such as low levels of physical exercise, obesity, smoking, and unhealthy diet. 10 Some of these and other risk factors, specifically glycemia, high blood pressure (BP), lipidemia, and obesity, are frequently addressed by health education interventions. 10 Health education interventions are comprehensive programs that healthcare providers deliver to patients aimed at improving patients' clinical outcomes through the increase and maintenance of health behaviours. 11 Along with education about, for example, medication adherence, these programs seek to increase behaviors such as physical exercise and a healthy diet thus reducing  6 patient morbidity or mortality. 11 Although there are numerous systematic reviews of educational interventions relating to ACS, or T2DM, an umbrella review providing direction on educational interventions for high risk patients with both ACS and T2DM is not available, indicating a need to gather the current evidence and develop an optimal protocol for health education programs for patients with ACS and T2DM. This umbrella review will examine the best available evidence on health education-related interventions for patients with ACS or T2DM. We will synthesize these findings to provide direction for health education-related interventions for high risk patients with both ACS and T2DM.
An umbrella review is a new method to summarize and synthesize the evidence from multiple systematic reviews/meta-analyses into one accessible publication. 12 Our aim is to systematically gather, evaluate and organize the current evidence relating the health education interventions for patients with ACS or T2DM, and proffer recommendations for the scope of educational content and delivery methods that would be suitable for patients with ACS and T2DM.

Data Sources
This umbrella review performed a literature search to identify systematic reviews and metaanalyses examining health education-related interventions for patients with ACS or T2DM. The

Eligibility Assessment
The title and abstract of all of the retrieved articles were assessed independently by two reviewers (XL-L, YS) based on the inclusion criteria. All duplicate articles were identified within EndNote version X7 15 and subsequently excluded. If the information from the titles and abstract was not clear, the full articles were retrieved. The decision to include an article was based on an appraisal of the full text of all retrieved articles. Any disagreements during this process were settled by discussion and, if necessary, consensus was sought with a third reviewer. We developed an assessment form in which specific reasons for exclusion were detailed.

Assessment of Methodological Quality
The methodological quality and risk of bias were assessed for each of the included publications using the Assessment of Multiple Systematic Reviews (AMSTAR) , 16 independently by the same two reviewers (see Table 1). The AMSTAR is an 11-item tool, with each item provided a score of 1 (specific criterion is met) or 0 (specific criterion is not met, unclear, or not applicable). 16 17 An overall score for the review methodological quality is then calculated as the sum of the individual item scores: high quality, 8-11; medium quality, 4-7; or low quality, 0-3. 17 If the required data were not available in the article, the original authors were contacted for more information.

Data Extraction
Data were independently extracted by two reviewers using a predefined data extraction form. For missing or unclear information, the primary authors were contacted for clarification.

Statistical Presentation of Results from Reviews
All of the results were extracted for each included systematic review or meta-analysis, and the overall effect estimates are presented in a tabular form. The number of systematic reviews or meta-analyses that reported the outcome, total sample (from included publications), and information of health education interventions is also presented in the Tables (Table 2, 3). 18 A final 'summary of evidence' was developed to present the intervention, included study synthesis, and indication of the findings from the included papers (Table 4). 18

Synthesizing the Results and Rating the Evidence for Effectiveness
The statements of evidence were based on a rating scheme to gather and rate the evidence across the included publications. 19 The statements of evidence were based on the following rating scheme: sufficient evidence, sufficient data to support decisions about the effect of the health education-related interventions. 19 A rating of sufficient evidence in this review is obtained when systematic reviews or meta-analyses with a large number of included articles or participants produce a statistically significant result between the health education group and the control group. 19 Some evidence, is a less conclusive finding about the effects of the health educationrelated interventions 19 with statistically significant findings found in only a few included reviews or studies. Insufficient evidence, refers to not enough evidence to make decisions about the effects of the health education-related interventions, such as non-significant results between the health education group and the control group in the included systematic reviews or metaanalyses. 19 Insufficient evidence to determine, refers to not enough pooled data to be able to determine whether of the health education-related interventions are effective or not based on the included reviews. 19

Characteristics of Included Trials
The selection process and number of studies at each step was illustrated as presented in Figure 1.  10 The database search yielded 692 publications, with removal of 197 duplicates and 372 articles that did not meet the inclusion criteria, 124 full-text articles were retrieved after applying the methodological quality rating (AMSTAR scale), and three studies [20][21][22] were removed due to low scores ≤3 on the AMSTAR scale. Fifty-one systematic reviews or meta-analyses  conducted between 2001 and 2016 and published in English were included ( Figure 1; Tables 1 -3); 15 relating to ACS and 36 relating to T2DM. None of the articles included patients with both ACS and T2DM. The umbrella review involved a total of 277,493 patients, including 225,034 patients with coronary heart disease or ACS (one article did not report the total sample) and 52,459 patients with T2DM (16 papers did not report the total sample). The average sample size was 8,161 (range, 536-68,556) participants. Of the included systematic reviews or meta-analyses, eleven were published in The Cochrane Library. Nine of the articles described meta-analyses, 29 articles described systematic reviews, and the remaining 13 articles were described as systematic reviews and meta-analyses or meta-regressions or narrative reviews. Electronic database searches were conducted for all systematic reviews or meta-analyses, with an average of 6 databases searched (range, [2][3][4][5][6][7][8][9][10][11][12][13][14][15][16]. The dates searched ranged widely, from inception of the database through December 2014. Most of the included reviews were randomized controlled trials (RCTs), and an average of 25.6 (range, 7-132) studies was included per systematic review or meta-analyses. Of the total 1,308 studies that were included in all of the reviews or meta-analyses, 406 included patients with ACS, and 902 included patients with

Methodological Quality of Included Systematic Reviews and Meta-Analyses
The methodological quality of the included publications is presented in Table 1. Thirty (58.8%) publications were classified as high quality (scores [8][9][10][11], and 21 (41.2%) publications were classified as medium quality (scores [4][5][6][7]. Twenty five (49%) reviews specifically provided an a priori design, while the use of such a design was unclear for 26 (51%) publications. The inclusion of other forms of literature (such as grey literature) was described in 18 (35%) reviews.
Only 14 out of 51 (27%) reviews included a table of included and excluded studies. Only 2 (4%) reviews did not provide a characteristics table of the included papers. The scientific quality of the included papers was evaluated and documented in 47 (92%) reviews. The scientific quality of the included studies was used appropriately to formulate conclusions in 47 (92%) reviews. The methods to combine the results of the included studies were appropriate in 43 (86%) reviews.
Publication bias was assessed in only 19 (37%) reviews. Finally, conflicts of interest were reported in 47 (92%) reviews.   17 and durations varied, and there was limited information about the intensity of health education for patients provided. The frequency of educational sessions was weekly or monthly, and an average of 3.7 topics was covered per education session. Nurses and multidisciplinary teams were the most frequent educators, and most education programs were delivered post-discharge.

Effect of Interventions
The outcomes of the included systematic reviews and meta-analyses were summarized in Table  F 19 depression. The findings are based on synthesis of results from six publications. 24 28 29 38 49 73 There is sufficient evidence that psychoeducational programs are effective at decreasing smoking, achieving smoking abstinence, and reducing depression. One review reported no effect on smoking cessation 24 or total mortality 49 .

Secondary Prevention Educational Interventions
The following statements are based on our synthesis of results from three papers. 27 34 37 There is some evidence that secondary prevention educational interventions reduce MI readmission rates and improve quality of life, but the intervention was ineffective in reducing revascularization, cholesterol levels, and improving smoking cessation rates. The results are mixed for mortality and re-infarction rates; two reviews 27 34 found positive effects on mortality, while one review 37 did not.

Patients with Type 2 Diabetes Mellitus
Ten types of health education-related interventions were used for patients with T2DM: culturally appropriate health education (tailored to the religious beliefs, culture, literacy, and linguistics of the geographical area), dietary advice, foot health education, group medical visits (a group education component taught by health professionals), general health education (only included general health information), improving adherence to medical treatment recommendations (e.g., promoting oral hypoglycemic adherence), lifestyle interventions (specific focus on dietary changes and increased physical activity, or stress management), psychoeducational interventions and self-management educational interventions (based on self-management skills), and therapeutic education (collaborative process needed to modify behavior and more effectively manage risk factors).

Culturally Appropriate Health Education
Findings are based on our synthesis of results from 8 publications. 26 35 45-47 51 55 64 Overall, there was some evidence of the effects of culturally appropriate health education on clinical outcomes for T2DM. There was sufficient evidence that culturally appropriate health education improves HbA1c control and knowledge scores. There is some evidence that physical activity and clinical outcomes (blood glucose, HbA1c, BP) were improved. There was no data relating to adverse events during the intervention and follow-up (such as hypoglycemic events and mortality), and there was insufficient evidence about improvements in quality of life.

General Health Education
The statements are based on our synthesis of results from five papers. 33 39 53 65 71 Overall, there were mixed effects of general health education programs on clinical outcomes for T2DM, including HbA1C, cholesterol level, and triglyceride level. There was some evidence of the effectiveness of general health education on glycemic control, weight reduction, and some diabetic control outcomes (HbA1C, diabetes complications). There were no data supporting the effectiveness of general health education on reduced health service utilization, diabetic complications, improved knowledge, psychosocial outcomes, or smoking habits.

Lifestyle Interventions
The following statements are based on our synthesis of results from 6 reviews. 32 42 48 62 63 67 Overall, there were mixed effects of the lifestyle interventions on cholesterol level, HbA1C level, and weight control. There is some evidence that lifestyle interventions or behavioral programs are effective for blood glucose and BP management, but they appear ineffective for HbA1c control.

Improving Adherence to Medical Treatment Recommendations
The statements are based on our synthesis of results from three publications. 50 68 70 There is some evidence of the effectiveness of improving adherence to medical treatment recommendations for HbA1C control and oral hypoglycemics.

Self-Management Educational Interventions
The statements are based on our synthesis of results from 9 reviews. 36 40 54 57 58 60 61 66 69 Overall, there was sufficient evidence of the effects of self-management education interventions on HbA1c control, knowledge, lifestyle outcomes, and main psychosocial outcomes. However, there was insufficient evidence of the benefits of this education intervention on depression, quality of life, and weight control.

Other Health Education-Related Interventions
Other health education-related interventions for patients with T2DM included therapeutic education, foot health education, group medical visits, psychoeducational interventions, and dietary advice. Statements for all of these interventions are based on our synthesis of results from one review.
There is some evidence that foot health education is effective in reducing the incidence of lower extremity amputation. 25 There is some evidence that group medical visits are effective for improving HbA1c control and systolic BP management 31 . There is also some evidence that psychoeducational programs are effective for improving HbA1c control and psychological status. 23 Finally, there is insufficient evidence that dietary advice improves glycemic and weight control or reduces diabetic micro-and macrovascular diseases. 56 There is also insufficient evidence for the

Discussion
This umbrella review identified 51 systematic reviews or meta-analyses (15 for ACS and 36 for T2DM) that assessed the outcomes of various aspects (such as the duration, contact hours, educational content, delivery mode) of the delivery of health education-related interventions relevant to high risk patients with ACS and T2DM. Health education has become an integral part of the management for people with ACS and T2DM. The most appropriate focus of the education provided to ACS patients with T2DM remains largely undefined in the literature. For example, it remains unknown if the focus should be primarily on cardiovascular risk factors, blood glucose monitoring, or all educational components for patients with both conditions. 66 72 In addition, should cardiovascular risk factors be the focus during the acute inpatient stay with other educational needs such as the smoking cessation occurring within the primary care or outpatient settings. 24 72 73 It remains challenging to determine the specific strategy or format that is the most effective delivery mode for patients with ACS or T2DM. There is very limited evidence to guide clinicians on the duration, contact hours, educational content, delivery mode, total length, and setting of health education programs for cardiac patients. 43 For patients with DM, one study reported that more successful programs were longer than 6 months (longer duration), consisted of greater than 10 contact sessions (high intensity), and were one-on-one sessions with individualized assessment. 75 Use of theory when designing behavioral change interventions may also influence effectiveness. 75 Health education using a cognitive behavioral strategy is most consistently effective in changing maladaptive illness beliefs, 44 and studies using more than 2 behavioral change strategies reported significant differences between the intervention and control groups. 24 In one review, a significant change in smoking cessation was not observed in subgroup analyses between studies that did or did not report using a theory in intervention planning; 49 however, the authors did not suggest that using a theory in program planning should be disregarded but reported that examining actual theories or mechanisms underlying health education programs is required. 49 Owing to the considerable overlap between different theories and the detailed description of the theoretical approach in only approximately 40% of the included papers, it is difficult to determine the most effective theoretical approach, but many models can be used with success, such as the health belief model (HBM), social cognitive theory (SCT) and

Use of theoretical orientation to develop educational intervention
transtheoretical model (TTM). 49 60 69 73 Educational content Most reviews reported that the educational content of the interventions was comprehensive. The most common topics, of the average 3.7 topics per education session, were behavioral change, cardiovascular risk factors management, exercise, psychosocial issues, and smoking cessation.
An underlying principle of health education for patients with ACS is that knowledge is necessary, but not enough to develop health behaviors and change risk factors. 24 43 Age, cognitive factors, environmental factors, and social and economic background are also important considerations. 43 While interventions using a behavioral program, telephone-based content, or self-care are effective for smoking cessation, there was insufficient evidence to support that any type of educational program was more efficacious than the others. 73 Psychoeducation, which is  24 defined as multimodal, educationally based, self-management interventions, 24 led to enhanced physical activity levels within 6-12 months when added to cardiac rehabilitation (CR) and was more effective than an exercise program or health education alone. 24 49 Moreover, psychoeducational interventions were more effective for patients with ACS than other types of health education. 24 49 Teaching strategies and outcomes Most reviews reported that the education was provided using multiple teaching methods and in multiple settings. Nurses and multidisciplinary teams were the most frequent people providing education, and most education programs were delivered post-discharge. Although face-to-face sessions were the most common delivery format, many education sessions were also delivered by telephone or through individualized counseling. Telephone-based health education appeared to be effective for reducing hospitalizations, systolic BP, smoking rates, depression, and anxiety. 52 The educational interventions for patients with ACS focused primarily on clinical outcomes (hospitalization and mortality), modifiable risk factors (BP, low-density lipoprotein levels, and smoking cessation), and other psychological outcomes (anxiety and depression).

Delivery, timing and follow up
Most educational sessions were delivered weekly. Few reviews provided information regarding the duration of education interventions; when the duration was reported, it varied from 4 weeks to 48 months. These findings suggest that there is a significant gap in the evidence in relation to the duration, contact hours, educational content, optimal delivery mode, total length, and setting of health education programs for cardiac patients. 43 For patients with ACS, one systematic review that included 7 studies with a total of 536 participants reported that studies with education lasting at least 6 months resulted in the most significant changes in the primary outcomes (such as behavioral change, smoking cessation) 24 and that at least 12 months of follow-up is needed to evaluate the impact of telephone-based education. 52 Another review reported that the intensity of education programs is important for efficacy regarding smoking cessation: interventions with a very low intensity and brief interventions, do not have a significant effect, 73 and programs for smoking cessation among patients with coronary heart disease should last >1 month. 73 Most of the reviews were provided for patients with ACS in inpatient settings and then within post discharge settings, five reviews 24 29 38 41 52 did not explicitly state the settings in which the health education related interventions were provided.

Use of theoretical orientation to develop educational intervention
Although the theoretical approach underpinning the health education program was not always described, there is evidence that health education interventions based on a theoretical model are likely to be effective. 36 Vugt et al suggested that self-care education programs should be based on theories and that theory-based self-care interventions are more effective than non-theory-based programs. 69 76 Theories could help to specify the key target health behaviors and behavioral change techniques required to generate the desired outcomes. 69 The decision regarding the theory should be based on the aim of the program and factor for intervention. 63 Only one review reported that a theoretical approach underpinning the health education program is not necessary for better outcomes. 66

Educational content
The educational content for patients with T2DM focused more on behavioral change, diet, exercise, glycemic control, medication, and self-management. Health education that was selfmanagement was more effective for patients with T2DM. 33 63 The educational interventions for patients with T2DM focused primarily on HbA1C, lipid levels, quality of life, and body weight. HBM and SCT were the most common theories used in the included reviews.

Teaching strategies and outcomes
Mixed health education programs generally included group sessions combined with educatorfacilitated individual sessions, covering basic knowledge and problem-solving skills. These programs produced greater benefits and larger effect sizes for blood glucose control and knowledge level in patients with T2DM. 40 In contrast, individual education programs have been reported as more effective in achieving outcomes than group-based education. This may be because education programs might be more efficient at addressing personal needs, with greater participant engagement. 64 Although face-to-face sessions were the most common delivery format, many education sessions were also delivered by telephone or individualized counseling.
Face-to-face health education programs were most effective for enhancing blood glucose control and knowledge levels, while mixed delivery models (face-to-face, phone contact, online or webbased, or video) produced a moderate effect for knowledge levels. 40 Another review reported that face-to-face health education programs generated a greater benefit for metabolic control than those delivered using electronic communication technology. 64 Community workers, dieticians, multidisciplinary teams, and nurses were the most frequent educators, and most of the education programs were delivered post-discharge. Some reviews indicated that health education programs delivered by a group of different educators, with some degree of education reinforcement at additional points of contact, may provide the best results. 53  66 However, based on two studies that reported HbA1C at 12 months, it was showed that the outcomes in studies with only a diabetes nurse as the educator also tended to do better than the outcomes in studies with a multidisciplinary team , while the biggest effect was seen when a dietician was the only educator. 66 Health education programs delivered by one person may focus more on the patient's ability than the educational content or quality of the health education programs. 66 However, no clear conclusion can be drawn whether having one educator delivering the intervention is best due to few information. 53

Delivery, timing and follow up
Education sessions were delivered weekly or monthly. Longer health education programs for T2DM (>6 months) produced larger effects for all primary outcomes (such as HbA1C). 40 Health education lasting >3 months resulted in the largest effect size compared with health education of a shorter duration (<3 months). 26 For HbA1C, the effect size at 6 months seemed to be significantly greater than at 3 and 12 months; in other words, the effect size peaked at 6 months. 55 In general, health education of a greater intensity (longer duration and more sessions) was more effective for blood glucose control and knowledge level among patients with T2DM. 40 65 Compared with health education programs covering only one topic, programs that included multiple or mixed educational topics yielded consistently greater benefits in blood glucose control and knowledge levels. 40 In addition, health education programs combined with specific behavioral change strategies (such as self-care strategies) seemed more effective than other programs. 40 Health education related interventions were mainly delivered in hospital settings, primary care settings, diabetes centers or community based settings; although six reviews 25 32 48 51 60 67 did not explicitly state the delivery settings. These results from included systematic reviews and meta-analyses help to provide recommendations about the content of a health education intervention for patients with ACS and T2DM, requiring further evaluation. Future development of educational programs for patients with ACS and T2DM by health care professionals should consider the needs of people with these diseases. 30 33 35 72 Based on the results and findings from this umbrella review, recommendations are made in Table 5.

Overall Completeness and Applicability of Evidence
This overview potentially provides an estimate with the lowest level of bias for the impact of health education-related interventions for patients with ACS or T2DM and could be regarded as an all-inclusive summary of the current evidence base for health education for these patients.
While this umbrella review identified evidence for each of the types of health education, there was only a small number of reviews within some categories (such as psychoeducational intervention 23 and dietary advice 56 ), and these studies were not very informative. This umbrella review also found no reviews that systematically analyzed varying doses of health education; therefore, could not examine the dose-response effects. There was insufficient information about the evaluated doses (total contact hours and duration of education) to enable comparison of the benefits of differences in the magnitude of the doses across the different research. This umbrella review found no reviews focused on patients with ACS and T2DM; instead, all of the systematic reviews and meta-analyses focused on only one of these diseases.

Quality of the Evidence
The methodological quality of the included systematic reviews and meta-analyses varied. All of the included reviews or meta-analyses were of moderate to high methodological quality, as assessed using AMSTAR. However, only 30 (58.8%) systematic reviews or meta-analyses were rated as high quality and only three (5.9%) systematic reviews or meta-analyses 36 46 73 adequately met all 11 AMSTAR criteria. This indicates that some of the reviews included in this umbrella review may have limitations in their design, conduct, and/or reporting that could have influenced the findings when considered both individually and collectively. 25 58 The quality of the primary studies in the included systematic reviews or meta-analyses also  30 varied. The main sources of bias were inadequate reporting of allocation concealment and randomization processes, as well as lack of outcome blinding. 26 35 72 73 This bias in the methodological quality led to lower quality assessments, which varied by results within each included review. Other reasons for lower methodological quality included heterogeneity in, or inconsistency of, the effect and imprecise findings. Heterogeneity between studies in this umbrella review was described in terms of the intervention, participant characteristics and length of follow-up. Heterogeneity was an important factor indicating the complexity of the health education interventions. 49 The variability in the approaches, tools or scales used to measure outcomes between the included studies are likely to introduce some heterogeneity. 23 The heterogeneity of the educational interventions seen in the reviews included in this umbrella review may reflect the uncertainty about the optimal strategy for providing health education to patients. 30 This umbrella review is the first synthesis of systematic reviews or meta-analyses to take a broad perspective on health education-related interventions for patients with ACS or T2DM. Given that health education is complex, the biggest challenge for systematic reviews or meta-analyses of health education is accounting for the potential clinical heterogeneity in health education-related interventions (content and delivery approaches) and the population of patients who receive health education. To facilitate comparisons across systematic reviews of health education and the efficient future update of this umbrella review, future reviews or meta-analyses need high-quality research and to standardize their design and reporting, including the reporting of included study characteristics, assessment criteria for risk of bias, outcomes, and methods to synthesize evidence synthesis. For clinicians providing educational interventions to individuals with ACS and T2DM, the results from this review provide a contemporaneous perspective on current evidence on effectiveness of health education (its content, and delivery methods) for this high risk patient group. The current evidence compiled by this umbrella review supports current international clinical guidelines, that theoretically-based education interventions lasting 6 months, delivered in multiple modes (face-to-face, phone contact, online or web-based, or video), and with individualized education delivered weekly are more likely to generate positive outcomes. This review also supports health education-related interventions provided by health professionals, including nurses only and multidisciplinary teams, and the delivery content included specific clinical factors for ACS and T2DM (BP, glycemic control, and medication), modifiable risk factors (unhealthy diet, inactivity, and smoking), and other psychological factors (anxiety and depression). These health education interventions could be delivered post-discharge, such as rehabilitation centers, primary care centers, and the community and should be at least 6 months in duration. The effectiveness of these programs was based on HbA1C levels, knowledge, psychosocial outcomes, readmission rates, and smoking status rather than clear evidence of reduced mortality and MI.

Conclusions
In addition, psychoeducational interventions were more effective for patients with ACS, and health education that was culturally appropriate or taught self-management was more effective for patients with T2DM. We also found that longer durations and high-intensity health education provided in an individualized format were more helpful for patients with ACS or T2DM.
The fact that none of the included reviews included patients with both ACS and T2DM indicates a clear need for further rigorous experimental studies with patients with both diseases. Future research that includes these aspects of education are likely determine the effectiveness of

Types of studies:
RCTs;

Number of studies:
Sixty-two independent studies.

Types of studies:
RCTs;

Number of studies:
A total of 14 studies were included.
Diabetes education intervention.
Unclear, did not describe.

Number of session:
Unclear; Total contact hours: Unclear, did not describe.

Number of session:
Monthly;

Total contact hours:
Unclear;

Number of studies:
10;
An educational intervention using SMS.
Unclear, did not describe.

Number of session:
Weekly; or 2 messages daily or unclear;

Total contact hours:
Unclear.

Number of studies:
Seventeen studies;

Types of studies:
RCTs; Total sample: Unclear. nutritionists or unclear.
Nurse or other healthcare professionals.

Number of session:
Unclear; Total contact hours:

Number of studies:
Seven studies;

Types of studies:
RCTs; Total sample: Unclear.
Educational interventions. Nurses; pharmacists; other skilled healthcare professionals.

Number of studies:
Forty-seven studies; Types of studies:

Number of studies:
Nine studies;

Types of studies:
RCTs;

Types of studies:
RCTs and RCT was followed by a beforeand-after study;

Number of studies:
Twenty-one published trials;

Types of studies:
RCTs and CCTs; Diabetes educator; nurse or did not describe.

Number of session:
Unclear;
Strategies: Face-to-face; telephone; Format: face-to-face; group based and telemedicine;

Number of studies:
A total of 11 trials;

Types of studies:
RCTs;

Number of studies:
Sixty-three RCTs;

Number of session:
Unclear;

7
Information sources 7 Describe all information sources (e.g., databases with dates of coverage, contact with study authors to identify additional studies) in the search and date last searched.

6-7
Search 8 Present full electronic search strategy for at least one database, including any limits used, such that it could be repeated.

6-7
Study selection 9 State the process for selecting studies (i.e., screening, eligibility, included in systematic review, and, if applicable, included in the meta-analysis).

8
Data collection process 10 Describe method of data extraction from reports (e.g., piloted forms, independently, in duplicate) and any processes for obtaining and confirming data from investigators.

8-9
Data items 11 List and define all variables for which data were sought (e.g., PICOS, funding sources) and any assumptions and simplifications made.

8-9
Risk of bias in individual studies 12 Describe methods used for assessing risk of bias of individual studies (including specification of whether this was done at the study or outcome level), and how this information is to be used in any data synthesis.

Study selection
17 Give numbers of studies screened, assessed for eligibility, and included in the review, with reasons for exclusions at each stage, ideally with a flow diagram.

9-11
Study characteristics 18 For each study, present characteristics for which data were extracted (e.g., study size, PICOS, follow-up period) and provide the citations.

9-12
Risk of bias within studies 19 Present data on risk of bias of each study and, if available, any outcome level assessment (see item 12).

12-13
Results of individual studies 20 For all outcomes considered (benefits or harms), present, for each study: (a) simple summary data for each intervention group (b) effect estimates and confidence intervals, ideally with a forest plot.

16-21
Synthesis of results 21 Present results of each meta-analysis done, including confidence intervals and measures of consistency.

16-21
Risk of bias across studies 22 Present results of any assessment of risk of bias across studies (see Item 15).

DISCUSSION
Summary of evidence 24 Summarize the main findings including the strength of evidence for each main outcome; consider their relevance to key groups (e.g., healthcare providers, users, and policy makers).

22,29-30
Limitations 25 Discuss limitations at study and outcome level (e.g., risk of bias), and at review-level (e.g., incomplete retrieval of identified research, reporting bias).

4,29-30
Conclusions 26 Provide a general interpretation of the results in the context of other evidence, and implications for future research.

Outcomes measures:
Clinical outcomes (such as HbA1C), behavioral outcomes (such as smoking), psychosocial outcomes (such as anxiety) and medical service use.
Results: Fifty-one eligible reviews (15 for ACS and 36 for T2DM) consisting of 1324 relevant studies involving 288,057 patients (15 papers did not provide the total sample); 30 (58.8%) reviews were rated as high quality. Nurses only and multidisciplinary teams were the most frequent professionals to provide education, and most educational interventions were delivered post-discharge. Face-to-face sessions were the most common delivery formats, and many education sessions were also delivered by telephone or via web contact.

Strengths and limitations of this review:
• This umbrella review is the first synthesis of systematic reviews or meta-analyses to consider health education-related interventions for patients with ACS or T2DM.
• These results provide recommendations about the content of a health education intervention for patients with ACS and T2DM.
• The diversity of the educational interventions seen in the reviews included in this umbrella review may reflect the uncertainty about the optimal strategy for providing health education to patients.
• This umbrella review found no reviews focused on patients with ACS and T2DM the intended target group; instead, all of the systematic reviews and meta-analyses focused on only one of these two diseases.

Introduction
Acute coronary syndrome (ACS) is the leading cause of death worldwide. The risk of high mortality rates relating to ACS is markedly increased after an initial cardiac ischemic event. [1] Globally, 7.2 million (13%) deaths are caused by coronary artery disease (CAD), [2] and it is estimated that more than 780,000 persons will experience ACS each year in the United States. [3] Moreover, about 20-25% of patients with ACS reportedly also have diabetes mellitus (DM); predominantly type 2 diabetes mellitus [T2DM]). [4 5] Patients with ACS and DM have an increased risk of adverse outcomes such as death, recurrent myocardial infarction (MI), readmission, or heart failure during follow up. [6] Longer median delay times from symptom onset to hospital presentation, have been reported among patients with ACS and DM than patients with ACS alone. [7] DM is now considered to confer a risk equivalent to that of CAD for patients for future MI and cardiovascular mortality. [8]  ). [9] ACS and T2DM are often associated with high-risk factors such as low levels of physical exercise, obesity, smoking, and unhealthy diet. [10] Some of these and other risk factors, specifically glycemia, high blood pressure (BP), lipidemia, and obesity, are frequently addressed by health education interventions. [10] Health education interventions are comprehensive programs that healthcare providers deliver to patients aimed at improving patients' clinical outcomes through the increase and maintenance of health behaviours. [11] Along with education about, for example, medication adherence, these programs seek to increase behaviors such as physical exercise and a healthy diet thus reducing patient morbidity or mortality. [11] Most diabetes education is provided through programs within outpatient services or physicians' practices. [12] Many recent education programs have been designed to meet national or international education standards [13][14][15] with diabetes education being individualized to consider patients' existing needs and health conditions. [16] Patients with T2DM have reported feelings of hopelessness and fatigue with low levels of self-efficacy, after experiencing an acute coronary episode. [17] Although there are numerous systematic reviews of educational interventions relating to ACS, or T2DM, an umbrella review providing direction on educational interventions for high risk patients with both ACS and T2DM is not available, indicating a need to gather the current evidence and develop an optimal protocol for health education programs for patients with ACS and T2DM. This umbrella review will examine the best available evidence on health educationrelated interventions for patients with ACS or T2DM. We will synthesize these findings to provide direction for health education-related interventions for high risk patients with both ACS and T2DM.
An umbrella review is a new method to summarize and synthesize the evidence from multiple systematic reviews/meta-analyses into one accessible publication. [18] Our aim is to systematically gather, evaluate and organize the current evidence relating the health education interventions for patients with ACS or T2DM, and proffer recommendations for the scope of educational content and delivery methods that would be suitable for patients with ACS and T2DM.

Participants
All participants were diagnosed with ACS or T2DM using valid, established diagnostic criteria.
The diagnostic standards included those described by the American College of Cardiology or American Heart Association, [3] National Heart Foundation of Australia & Cardiac Society of Australia and New Zealand, [19] World Health Organization, [20] or other associations.

Intervention Types
For this umbrella review, health education-related interventions refer to any planned activities or programs that include behavior modification, counseling, and teaching interventions. Results considered for this review included changes in clinical outcomes (including blood pressure levels, body weight, diabetes complications, HbA1C, lipid levels, mortality rate and physical activity levels), behavioral outcomes (such as diet, knowledge, self-management skills, selfefficacy, and smoking), and psychosocial outcomes (such as anxiety, depression, quality of life, and stress) and medical service use (such as medication use, health care utilization, and cost- delivered in any format, including face-to-face, telephone and group-based or one-on-one, and the settings include community, hospital and home. The interventions were delivered by nurses (including diabetes nurse educators), physicians, community health care workers, dietitians, lay people, rehabilitation therapists, or multidisciplinary teams.

Study Types
Only systematic reviews and meta-analyses were included in this review.

Eligibility Assessment
The title and abstract of all of the retrieved articles were assessed independently by two reviewers (XL-L, YS) based on the inclusion criteria. All duplicate articles were identified within EndNote version X7 [21] and subsequently excluded. If the information from the titles and abstract was not clear, the full articles were retrieved. The decision to include an article was based on an appraisal of the full text of all retrieved articles. Any disagreements during this process were settled by discussion and, if necessary, consensus was sought with a third reviewer.
We developed an assessment form in which specific reasons for exclusion were detailed.

Assessment of Methodological Quality
The methodological quality and risk of bias were assessed for each of the included publications using the Assessment of Multiple Systematic Reviews (AMSTAR), [22] independently by the same two reviewers (see Table 1). The AMSTAR is an 11-item tool, with each item provided a score of 1 (specific criterion is met) or 0 (specific criterion is not met, unclear, or not applicable). [22 23] An overall score for the review methodological quality is then calculated as the sum of the individual item scores: high quality, 8-11; medium quality, 4-7; or low quality, 0-3. [23] If the required data were not available in the article, the original authors were contacted for more information. The low quality reviews (AMSTAR scale: 0-3) were excluded in this umbrella review.

Data Extraction
Data were independently extracted by two reviewers using a predefined data extraction form. For missing or unclear information, the primary authors were contacted for clarification.

Statistical Presentation of Results from Reviews
All of the results were extracted for each included systematic review or meta-analysis, and the overall effect estimates are presented in a tabular form. The number of systematic reviews or meta-analyses that reported the outcome, total sample (from included publications), and information of health education interventions is also presented in the Tables (Table 2, 3). [24] A final 'summary of evidence' was developed to present the intervention, included study synthesis, and indication of the findings from the included papers (Table 4). [24] This umbrella review calculated the Corrected Covered Area (CCA) (Appendix 2, 3). The CCA statistic is a measure of overlap of trials (the repeated inclusion of the same trial in subsequent systematic reviews included in an umbrella systematic review). A detailed description of the calculation is provided by the authors who note slight CCA as 0-5%, moderate CCA as 6-10%, high CCA as 11-15% and very high CCA is more than 15%. [25] The lower the CCA the lower the likelihood of overlap of trials included in the umbrella review.

Synthesizing the Results and Rating the Evidence for Effectiveness
The statements of evidence were based on a rating scheme to gather and rate the evidence across the included publications. 26 The statements of evidence were based on the following rating scheme: sufficient evidence, sufficient data to support decisions about the effect of the health education-related interventions. [26] A rating of sufficient evidence in this review is obtained when systematic reviews or meta-analyses with a large number of included articles or participants produce a statistically significant result between the health education group and the control group. [26] Some evidence, is a less conclusive finding about the effects of the health educationrelated interventions [26] with statistically significant findings found in only a few included reviews or studies. Insufficient evidence, refers to not enough evidence to make decisions about the effects of the health education-related interventions, such as non-significant results between the health education group and the control group in the included systematic reviews or metaanalyses. [26] Insufficient evidence to determine, refers to not enough pooled data to be able to determine whether of the health education-related interventions are effective or not based on the included reviews. [26] Results

Characteristics of Included reviews
The selection process and number of studies at each step was illustrated as presented in Figure 1.
The database search yielded 692 publications, with removal of 197 duplicates and 371 articles that did not meet the inclusion criteria, 124 full-text articles were retrieved after applying the methodological quality rating (AMSTAR scale), and three studies [27][28][29] were removed due to low scores ≤3 on the AMSTAR scale. Fifty-one systematic reviews or meta-analyses  conducted between 2001 and 2016 and published in English were included ( Figure 1; Tables 1 -3); 15 relating to ACS. The overlap of the trials included in the 15 reviews and meta-analyses related to ACS was slight (CCA = 2.6%). For the 36 systematic reviews relating to T2DM, the overlap of trials within these 35 reviews and meta-analyses (one review [47] did not report the included and 177 studies related to T2DM, were included in more than one systematic review or metaanalysis (see Appendix 2 and 3 and CCA statistics). The sample of these studies would therefore be included more than once. Of the included systematic reviews or meta-analyses, eleven were published in The Cochrane Library. Nine of the articles described meta-analyses, 29 articles described systematic reviews, and the remaining 13 articles were described as systematic reviews and meta-analyses or meta-regressions or narrative reviews.  The methodological quality of the included publications is presented in Table 1. Thirty (58.8%) publications were classified as high quality (scores [8][9][10][11], and 21 (41.2%) publications were classified as medium quality (scores 4-7). Twenty five (49%) reviews specifically provided an a priori design, while the use of such a design was unclear for 26 (51%) publications. The inclusion of other forms of literature (such as grey literature) was described in 18 (35%) reviews.

Methodological Quality of Included Systematic Reviews and Meta-Analyses
Only 14 out of 51 (27%) reviews included a table of included and excluded studies. Only 2 (4%) reviews did not provide a characteristics table of the included papers. The scientific quality of the included papers was evaluated and documented in 47 (92%) reviews. The scientific quality of the included studies was used appropriately to formulate conclusions in 47 (92%) reviews. The methods to combine the results of the included studies were appropriate in 43 (86%) reviews.
Publication bias was assessed in only 19 (37%) reviews. Finally, conflicts of interest were reported in 47 (92%) reviews.

Characteristics of Health Educational Interventions
The description of the health educational interventions followed the Workgroup for Intervention Development and Evaluation Research (WIDER) reporting guidelines for behavior change interventions. [81] The characteristics of the recipients, setting, delivery methods, intensity, duration, and educational content of health educational interventions for patients with ACS or T2DM are summarized in Tables 2 and 3. The delivery strategies for health education included face-to-face, internet-based, phone-based, videotape, written educational materials, or mixed.
The format included one-on-one (individualized), group, or both. Face-to-face sessions were the most common delivery formats, and many education sessions were also delivered by telephone/web contact or individualized counseling. The number of sessions, total contact hours, and durations varied, and there was limited information about the intensity of health education for patients provided. The frequency of educational sessions was weekly or monthly, and an average of 3.7 topics was covered per education session. Nurses and multidisciplinary teams were the most frequent educators, and most education programs were delivered post-discharge.

Effect of Interventions
The outcomes of the included systematic reviews and meta-analyses were summarized in Table  F

Patients with Acute Coronary Syndrome
Three major types of health education-related interventions were used for patients with ACS: general health education (only included general health information), psychoeducational interventions, and secondary prevention educational interventions (including strategies to promote a healthy lifestyle, manage medications, and reduce cardiovascular complications) as well as internet-based interventions.

General Health Education
The findings are based on our synthesis of the findings from 6 systematic reviews. [37 48 50 51 59 70] Overall, there were mixed effects of general health education on behavioral change or clinical outcomes in patients with ACS. There was some evidence of a positive effect of general health education on knowledge, behavior, psychosocial indicators, beliefs, and risk factor modification, but no effects for key clinical outcomes, such as cholesterol level, hospitalization, mortality, MI, and revascularization. The results for health-related quality of life, healthcare utilization, and costs were mixed; several reviews reported a significant change, and other reviews reported no significant change for these outcomes. Only one review focused on telephone-based health education. There is some evidence that telephone-based health education during cardiac rehabilitation might improve all-cause hospitalization, anxiety, depression, smoking cessation, and systolic BP but there is no evidence for improvements in all-cause mortality and reductions in low-density lipoprotein cholesterol. [59] Psychoeducational Interventions Strategies for psychoeducational interventions have a specific focus on smoking cessation, and  56 69] There is sufficient evidence that psychoeducational programs are effective at decreasing smoking, achieving smoking abstinence, and reducing depression. One review reported no effect on smoking cessation [31] or total mortality [56] .

Secondary Prevention Educational Interventions
The following statements are based on our synthesis of results from three papers. [34 41 44] There is some evidence that secondary prevention educational interventions reduce MI readmission rates and improve quality of life, but the intervention was ineffective in reducing revascularization, cholesterol levels, and improving smoking cessation rates. The results are mixed for mortality and re-infarction rates; two reviews [34 41] found positive effects on mortality, while one review [44] did not.

Patients with Type 2 Diabetes Mellitus
Ten types of health education-related interventions were used for patients with T2DM: culturally appropriate health education (tailored to the religious beliefs, culture, literacy, and linguistics of the geographical area), dietary advice, foot health education, group medical visits (a group education component taught by health professionals), general health education (only included general health information), improving compliance with taking medications (e.g., promoting oral hypoglycemic adherence), lifestyle interventions (specific focus on dietary changes and increased physical activity, or stress management), psychoeducational interventions and selfmanagement educational interventions (activities that promote or maintain the behaviors to manage T2DM often based on the National Standards for Diabetes Self-Management Education [13] ), and therapeutic education (collaborative process needed to modify behavior and more effectively manage risk factors).

Culturally Appropriate Health Education
Findings are based on our synthesis of results from 8 publications. [33 42 52-54 58 62 72] Overall, there was some evidence of the effects of culturally appropriate health education on clinical outcomes for T2DM. There was sufficient evidence that culturally appropriate health education improves HbA1c control and knowledge scores. There is some evidence that physical activity and clinical outcomes (blood glucose, HbA1c, BP) were improved. There were no data relating to adverse events during the intervention and follow-up (such as hypoglycemic events and mortality), and there was insufficient evidence about improvements in quality of life.

General Health Education
The statements are based on our synthesis of results from five papers. [40 46 60 74 79] Overall, there were mixed effects of general health education programs on clinical outcomes for T2DM, including HbA1C, cholesterol level, and triglyceride level. There was some evidence of the effectiveness of general health education on glycemic control, weight reduction, and some diabetes control outcomes (HbA1C, diabetes complications). There were no data supporting the effectiveness of general health education on reduced health service utilization, diabetes complications, improved knowledge, psychosocial outcomes, or smoking habits.

Lifestyle Interventions
The following statements are based on our synthesis of results from 6 reviews. [39 49 55 71 72 77] Overall, there were mixed effects of the lifestyle interventions on cholesterol level, HbA1C level, and weight control. There is some evidence that lifestyle interventions or behavioral programs are effective for blood glucose and BP management, but they were ineffective for reductions in HbA1c scores [71 72] .

Improving Adherence to Medication Regimes
The statements are based on our synthesis of results from three publications. [57 78 80] There is some evidence of the effectiveness of improving adherence to taking medications for HbA1C control including oral hypoglycemic agents.

Self-Management Educational Interventions
The statements are based on our synthesis of results from 9 reviews. [43 47 61 64 65 67 68 75 76] Overall, there was sufficient evidence of the effects of self-management education interventions on HbA1c control, knowledge, lifestyle outcomes, and main psychosocial outcomes. However, there was insufficient evidence of the benefits of this education intervention on depression, quality of life, and weight control.

Other Health Education-Related Interventions
Other health education-related interventions for patients with T2DM included therapeutic education, foot health education, group medical visits, psychoeducational interventions, and dietary advice. Statements for all of these interventions are based on our synthesis of results from one review.
There is some evidence that foot health education is effective in reducing the incidence of lower extremity amputation. [32] There is some evidence that group medical visits are effective for improving HbA1c control and systolic BP management [38] . There is also some evidence that psychoeducational programs are effective for improving HbA1c control and psychological status. [30] Finally, there is insufficient evidence that dietary advice improves glycemic and weight control or reduces micro-and macrovascular diseases. [63] There is also insufficient evidence for the cost-

Discussion
This umbrella review identified 51 systematic reviews or meta-analyses ( 69 70] It remains challenging to determine the specific strategy or format that is the most effective delivery mode for patients with ACS or T2DM. There is very limited evidence to guide clinicians on the duration, contact hours, educational content, delivery mode, total length, and setting of health education programs for cardiac patients. [50] For patients with DM, one study reported that more successful programs were longer than 6 months (longer duration), consisted of greater than 10 contact sessions (high intensity), and were one-on-one sessions with individualized assessment. [82]

Use of Theoretical Orientation to Develop Educational Intervention
For patients with Acute Coronary Syndrome Use of theory when designing behavioral change interventions may also influence effectiveness. [75] Health education using a cognitive behavioral strategy is most consistently effective in changing maladaptive illness beliefs, [51] and studies using more than 2 behavioral change strategies reported significant differences between the intervention and control groups. [31] In one review, a significant change in smoking cessation was not observed in subgroup analyses between studies that did or did not report using a theory in intervention planning; [56] however, the authors did not suggest that using a theory in program planning should be disregarded but reported that examining actual theories or mechanisms underlying health education programs is required. [56] Owing to the considerable overlap between different theories and the detailed description of the theoretical approach in only approximately 40% of the included papers, it is difficult to determine the most effective theoretical approach, but many models can be used with success, such as the health belief model (HBM), social cognitive theory (SCT) and transtheoretical model (TTM). [56 67 69 75] Three reviews [31 41 44] noted that some included studies used behavioural strategies such as goal setting. These strategies were found to be beneficial for patients with coronary heart disease.

For patients with Type 2 Diabetes Mellitus
Although the theoretical approach underpinning the health education program was not always described, 13 of the 36 reviews (36.11%) related to T2DM reported the theoretical approach used in their included studies. The most common theories were SCT (including self-efficacy), empowerment theories (eg., Empowerment Behavior Change Model, Self-determination and Autonomy Motivation Theory, Middle-range Theory of Community Empowerment) and TTM.
There is evidence that health education interventions based on a theoretical model are likely to be effective. [43] Vugt et al suggested that self-care education programs should be based on theories and that theory-based self-care interventions are more effective than non-theory-based programs. [75 83] Theories could help to specify the key target health behaviors and behavioral change techniques required to generate the desired outcomes. [75] The decision regarding the theory should be based on the aim of the program and factor for intervention. [77] Only one review reported that a theoretical approach underpinning the health education program is not necessary for better outcomes. [76] Fourteen reviews [30 33 40 46 52 57 60 63 64 67 68 73 75 77] reported that goal setting was conducted in the included studies. Goal setting by patients, health professionals or mutually-agreed goals were linked to improved patient outcomes.

For Patients with Acute Coronary Syndrome
Most reviews reported that the educational content of the interventions was comprehensive. The most common topics, of the average 3.7 topics per education session, were behavioral change, cardiovascular risk factors management, exercise, psychosocial issues, and smoking cessation.
An underlying principle of health education for patients with ACS is that knowledge is necessary, but not enough to develop health behaviors and change risk factors. [31 50] Age, cognitive factors, environmental factors, and social and economic background are also important considerations. [50] While interventions using a behavioral program, telephone-based content, or self-care are effective for smoking cessation, there was insufficient evidence to support that any type of educational program was more efficacious than the others. [69] Psychoeducation, which is defined as multimodal, educationally based, self-management interventions, 31 led to enhanced physical activity levels within 6-12 months when added to cardiac rehabilitation (CR) and was more effective than an exercise program or health education alone. [

For Patients with Type 2 Diabetes Mellitus
The educational content for patients with T2DM focused more on behavioral change, diet, exercise, glycemic control, medication, and self-management. Health education that was selfmanagement was more effective for patients with T2DM. [40 47] In addition, based on the current evidence, the educational content should be culturally sensitive, especially for patients with T2DM; [33 42 54] culturally appropriate diabetes health education may have a greater impact on glycemic control and reduce diabetes complications. [77] The educational interventions for patients with T2DM focused primarily on HbA1C, lipid levels, quality of life, and body weight.
HBM and SCT were the most common theories used in the included reviews.

For Patients with Acute Coronary Syndrome
Most reviews reported that the education was provided using multiple teaching methods and in multiple settings. Nurses and multidisciplinary teams were the most frequent people providing education, and most education programs were delivered post-discharge. Although face-to-face sessions were the most common delivery format, many education sessions were also delivered by telephone or through individualized counseling. Telephone-based health education appeared to be effective for reducing hospitalizations, systolic BP, smoking rates, depression, and anxiety. [59] The educational interventions for patients with ACS focused primarily on clinical outcomes (hospitalization and mortality), modifiable risk factors (BP, low-density lipoprotein levels, and smoking cessation), and other psychological outcomes (anxiety and depression). Mixed health education programs generally included group sessions combined with educatorfacilitated individual sessions, covering basic knowledge and problem-solving skills. These programs produced greater benefits and larger effect sizes for blood glucose control and knowledge level in patients with T2DM. [47] In contrast, individual education programs have been reported as more effective in achieving outcomes than group-based education. This may be because education programs might be more efficient at addressing personal needs, with greater participant engagement. [73] However, one systematic review reported that individual and group patient education, demonstrated similar outcomes among T2DM patients. [46] Although face-to-face sessions were the most common delivery format, many education sessions were also delivered by telephone or individualized counseling. Face-to-face health education programs were most effective for enhancing blood glucose control and knowledge levels, while mixed delivery models (face-to-face, phone contact, online or web-based, or video) produced a moderate effect for knowledge levels. [47] Another review reported that face-to-face health education programs generated a greater benefit for metabolic control than those delivered using electronic communication technology. [73] Nurses (including diabetes nurses educators), community workers, dieticians and multidisciplinary teams were the most frequent educators, and most of the education programs were delivered post-discharge. Some reviews indicated that health education programs delivered by a group of different educators, with some degree of education reinforcement at additional points of contact, may provide the best results. [60 76] However, based on two studies that reported HbA1C at 12 months, it is indicated that the outcomes in studies with only a diabetes nurse as the educator also tended to do better than the outcomes in studies with a multidisciplinary team , while the biggest effect was seen when a dietician was the only educator. [76] Health education programs delivered by one person may focus more on the patient's ability than the educational content or quality of the health education programs. [76] However, no clear conclusion can be drawn whether having one educator delivering the intervention is best due to few information. [60] Delivery, Timing and Follow up

For Patients with Acute Coronary Syndrome
Most educational sessions were delivered weekly. Few reviews provided information regarding the duration of education interventions; when the duration was reported, it varied from 4 weeks to 48 months. These findings suggest that there is a significant gap in the evidence in relation to the duration, contact hours, educational content, optimal delivery mode, total length, and setting of health education programs for cardiac patients. [50] For patients with ACS, one systematic review that included 7 studies with a total of 536 participants reported that studies with education lasting at least 6 months resulted in the most significant changes in the primary outcomes (such as behavioral change, smoking cessation) [31] and that at least 12 months of follow-up is needed to evaluate the impact of telephone-based education. [59] Another review reported that the intensity of education programs is important for efficacy regarding smoking cessation: interventions with a very low intensity and brief interventions, do not have a significant effect, [69] and programs for smoking cessation among patients with coronary heart disease should last >1 month. [69] Most of the reviews were provided for patients with ACS in inpatient settings and then within post discharge settings, five reviews [31 36 45 48 59] did not explicitly state the settings in which the health education related interventions were provided.

For Patients with Type 2 Diabetes Mellitus
Education sessions were delivered weekly or monthly. Longer health education programs for T2DM (>6 months) produced larger effects for all primary outcomes (such as HbA1C). [47] F o r p e e r r e v i e w o n l y 24 Health education lasting >3 months resulted in the largest effect size compared with health education of a shorter duration (<3 months). [33] For HbA1C, the effect size at 6 months seemed to be significantly greater than at 3 and 12 months; in other words, the effect size peaked at 6 months. [62] In general, health education of a greater intensity (longer duration and more sessions) was more effective for blood glucose control and knowledge level among patients with T2DM. [47 74] Compared with health education programs covering only one topic, programs that included multiple or mixed educational topics yielded consistently greater benefits in blood glucose control and knowledge levels. [47] In addition, health education programs combined with specific behavioral change strategies (such as self-care strategies) seemed more effective than other programs. [47] Health education related interventions were mainly delivered in hospital settings, primary care settings, diabetes centers or community based settings; although six reviews [32 39 55 58 67 72] did not explicitly state the delivery settings.

Recommendations about Health Education Interventions for Patients with ACS and T2DM
These results from included systematic reviews and meta-analyses help to provide recommendations about the content of a health education intervention for patients with ACS and T2DM, requiring further evaluation. Future development of educational programs for patients with ACS and T2DM by health care professionals should consider the needs of people with these diseases. [37 40 42 70] Based on the results and findings from this umbrella review, recommendations are made in Table 5. The acute life-threatening nature of ACS requires that increased emphasis should be placed on cardiovascular risk factors in any combined education program. Both ACS and T2DM have common lifestyle factors such as inactivity and high fat diet requiring modifications. This overview potentially provides an estimate with the lowest level of bias for the impact of health education-related interventions for patients with ACS or T2DM and could be regarded as an all-inclusive summary of the current evidence base for health education for these patients.

Overall Completeness and Applicability of Evidence
While this umbrella review identified evidence for each of the types of health education, there was only a small number of reviews within some categories (such as psychoeducational intervention [30] and dietary advice [63] ), and these studies were not very informative. This umbrella review also found no reviews that systematically analyzed varying doses of health education; therefore, could not examine the dose-response effects. There was insufficient information about the evaluated doses (total contact hours and duration of education) to enable comparison of the benefits of differences in the magnitude of the doses across the different research. This umbrella review found no reviews focused on patients with ACS and T2DM; instead, all of the systematic reviews and meta-analyses focused on only one of these diseases.

Quality of the Evidence
The methodological quality of the included systematic reviews and meta-analyses varied. All of the included reviews or meta-analyses were of moderate to high methodological quality, as assessed using AMSTAR. However, only 30 (58.8%) systematic reviews or meta-analyses were rated as high quality and only three (5.9%) systematic reviews or meta-analyses [43 53 69] adequately met all 11 AMSTAR criteria. This indicates that some of the reviews included in this umbrella review may have limitations in their design, conduct, and/or reporting that could have influenced the findings when considered both individually and collectively. [32 65] The quality of the primary studies in the included systematic reviews or meta-analyses also varied. The main sources of bias were inadequate reporting of allocation concealment and randomization processes, as well as lack of outcome blinding. [33 42 69 70] This bias in the methodological quality led to lower quality assessments, which varied by results within each included review. Other reasons for lower methodological quality included heterogeneity in, or inconsistency of, the effect and imprecise findings. Heterogeneity between studies in this umbrella review was described in terms of the intervention, participant characteristics and length of follow-up. Heterogeneity was an important factor indicating the complexity of the health education interventions. [56] The variability in the approaches, tools or scales used to measure outcomes between the included studies are likely to introduce some heterogeneity. [30] The heterogeneity of the educational interventions seen in the reviews included in this umbrella review may reflect the uncertainty about the optimal strategy for providing health education to patients. [37] In addition, 240 studies were included more than once in the included reviews and meta-analyses. However, the overall overlap of studies among reviews and meta-analyses related ACS and T2DM was slight, CCA of 2.6% and 2.1%, respectively. [25] This umbrella review is the first synthesis of systematic reviews or meta-analyses to take a broad perspective on health education-related interventions for patients with ACS or T2DM. Given that health education is complex, the biggest challenge for systematic reviews or meta-analyses of health education is accounting for the potential clinical heterogeneity in health education-related interventions (content and delivery approaches) and the population of patients who receive health education. To facilitate comparisons across systematic reviews of health education and the efficient future update of this umbrella review, future reviews or meta-analyses need high-quality research and to standardize their design and reporting, including the reporting of included study characteristics, assessment criteria for risk of bias, outcomes, and methods to synthesize evidence synthesis. For clinicians providing educational interventions to individuals with ACS and T2DM, the results from this review provide a contemporaneous perspective on current evidence on the effectiveness of health education (its content, and delivery methods) for this high risk patient group. The current evidence compiled by this umbrella review supports current international clinical guidelines, that theoretically-based education interventions lasting 6 months, delivered in multiple modes (face-to-face, phone contact, online or web-based, or video), and with individualized education delivered weekly, are more likely to generate positive outcomes. This review also supports health education-related interventions provided by health professionals, including nurses and multidisciplinary teams, delivering content including specific clinical factors for ACS and T2DM (BP, glycemic control, and medication), modifiable risk factors (unhealthy diet, inactivity, and smoking), and other psychological factors (anxiety and depression). These health education interventions could be delivered post-discharge, such as rehabilitation centers, primary care centers, and the community and should be at least 6 months in duration. The effectiveness of these programs was based on HbA1C levels, knowledge, psychosocial outcomes, readmission rates, and smoking status rather than clear evidence of reduced mortality, MI, or short and long-term complications. In addition, psychoeducational interventions were more effective for patients with ACS, and health education that was culturally appropriate or taught self-management was more effective for patients with T2DM. We also found that longer durations and high-intensity health education provided in an individualized format were more helpful for patients with ACS or T2DM.

Conclusions
The fact that none of the included reviews included patients with both ACS and T2DM indicates a clear need for further rigorous experimental studies with patients with both diseases. Future research that includes these aspects of education are likely determine the effectiveness of
All Internet based interventions.

Number of session:
Weekly or monthly or unclear;

Total contact hours:
Unclear.

Number of session:
Weekly or 2-3 times per week;

Total contact hours:
Unclear. outcomes.

Number of studies:
26 studies;

Types of studies:
RCTs;

Number of studies:
Twenty-four papers reporting on 13 RCTs;

Number of studies:
Sixty-two independent studies.

Total contact hours:
Varying from 10 to 240 minutes.

Duration: Unclear.
Strategies: Face-to-face sessions, telephone contact or unclear; Format: Group or unclear;

Number of session:
Unclear;
Strategies: Did not describe the strategies; Format: Group setting, combination of group and one-on-one education and one-on-one format only;

Number of studies:
24 papers reporting on 13 studies.

Number of session:
Unclear;

Total contact hours:
Unclear; Duration: 4 days to two weeks or unclear.

Number of studies:
A total of 14 studies were included.

Types of studies:
RCTs;

Forty trials;
Types of studies:
Cardiologist, nurse, physician or study nurse.

Number of studies:
Nineteen trials;

Types of studies:
RCTs; Total sample: 2,548 patients.
Unclear, did not describe.

Number of session:
Unclear;

Types of studies:
RCTs; Total sample: Unclear.
Diabetes education intervention.
Unclear, did not describe.

Number of session:
Unclear; Total contact hours: Pharmacist or unclear.

Number of session:
Unclear; Unclear, did not describe.

Number of session:
Monthly;
Unclear, did not describe.

Number of session:
Unclear;

Number of studies:
10;

Types of studies:
RCTs; Total sample: 960 patients.
An educational intervention using SMS.
Unclear, did not describe.

Number of session:
Weekly; or 2 messages daily or unclear;

Total contact hours:
Unclear.

Number of studies:
Seventeen studies;

Types of studies:
RCTs; Total sample: Unclear. nutritionists or unclear.
Strategies: Face-to-face or unclear; Format: individual and group lessons;

Number of studies:
Thirteen studies;

Types of studies:
RCTs and comparative studies; Total sample: Unclear.

Number of session:
Unclear;

Total contact hours:
Unclear; Duration: One session - Nurses (

Number of studies:
Seven studies;

Types of studies:
RCTs; Total sample: Unclear.
Educational interventions. Nurses; pharmacists; other skilled healthcare professionals.

Number of studies:
Forty-seven studies; Types of studies:

Number of session:
Unclear;

Total contact hours:
Unclear; Duration: one session to 12 months; Strategies: Face to face; visual aids, leaflets and teaching materials; Format: group approach, one-to-one interviews and a mixed approach;

Number of studies:
Nine studies;

Types of studies:
RCTs;

Types of studies:
RCTs and RCT was followed by a beforeand-after study;

Number of studies:
Twenty-one published trials;

Types of studies:
RCTs and CCTs; Total sample: Unclear.
Diabetes educator; nurse or did not describe.

Number of session:
Unclear;
Strategies: Face-to-face; telephone; Format: face-to-face; group based and telemedicine;

Number of studies:
A total of 11 trials;

Types of studies:
RCTs;

Number of studies:
Sixty-three RCTs;

Types of studies:
RCTs; Total sample: 2,720 patients.

Number of session:
Unclear;

Total contact hours:
Unclear.
Author-year ( Trial reference) Barth   x Beckie T (1989) A supportive-educative telephone program: Impact on knowledge and anxiety after coronary artery bypass graft surgery. Heart and Lung: Journal of Critical Care 18 (1) (2011) Six-year follow-up of a randomised controlled trial examining hospital versus home-based exercise training after coronary artery bypass graft surgery. Heart 97 (14): 1169-1174.  (2002) Coaching patients with coronary heart disease to achieve the target cholesterol: A method to bridge the gap between evidence-based medicine and the "real world"randomized controlled trial. Journal of Clinical Epidemiology. 55 (3) (2007) Agurs-Collins TD, Kumanyika SK, Ten Have TR, Adams-Campbell LL. A randomized controlled trial of weight reduction and exercise for diabetes management in        x Glasgow     x Perry

Abstract
Objectives: This umbrella review aimed to identify the current evidence on health educationrelated interventions for patients with acute coronary syndrome or type 2 diabetes mellitus; identify the educational content, delivery methods, intensity, duration and setting required. The purpose was to provide recommendations for educational interventions for high risk patients with both acute coronary syndrome and type 2 diabetes mellitus.
Design: Umbrella review of Systematic Reviews and Meta-Analyses.
Setting: Inpatient and post discharge settings.

Participants: Patients with ACS and T2DM.
Data sources: CINAHL, Cochrane Library, Joanna Briggs Institute, Journals@Ovid, EMBase, Medline, PubMed, and Web of Science databases from January 2000 through May 2016.

Outcomes measures:
Clinical outcomes (such as HbA1C), behavioral outcomes (such as smoking), psychosocial outcomes (such as anxiety) and medical service use.
Results: Fifty-one eligible reviews (15 for ACS and 36 for T2DM) consisting of 1324 relevant studies involving 288,057 patients (15 papers did not provide the total sample); 30 (58.8%) reviews were rated as high quality. Nurses only and multidisciplinary teams were the most frequent professionals to provide education, and most educational interventions were delivered post-discharge. Face-to-face sessions were the most common delivery formats, and many education sessions were also delivered by telephone or via web contact. The frequency of educational sessions was weekly or monthly, and an average of 3.7 topics was covered per education session. Psychoeducational interventions were generally effective at reducing smoking and admissions for patients with ACS. Culturally appropriate health education, self-management educational interventions, group medical visits and psychoeducational interventions were Conclusions: Results indicate that there is a body of current evidence about the efficacy of health education, its content, and delivery methods for patients with ACS or T2DM. These results provide recommendations about the content for, and approach to, health education intervention for these high risk patients.

KEY WORDS: Health Education; Acute Coronary Syndrome; Type 2 Diabetes Mellitus;
Umbrella Review

Strengths and limitations of this review:
• This umbrella review is the first synthesis of systematic reviews or meta-analyses to consider health education-related interventions for patients with ACS or T2DM.
• These results provide recommendations about the content of a health education intervention for patients with ACS and T2DM.
• The diversity of the educational interventions seen in the reviews included in this umbrella review may reflect the uncertainty about the optimal strategy for providing health education to patients.
• This umbrella review found no reviews focused on patients with ACS and T2DM the intended target group; instead, all of the systematic reviews and meta-analyses focused on only one of these two diseases.

Introduction
Acute coronary syndrome (ACS) is the leading cause of death worldwide. The risk of high mortality rates relating to ACS is markedly increased after an initial cardiac ischemic event. [1] Globally, 7.2 million (13%) deaths are caused by coronary artery disease (CAD), [2] and it is estimated that more than 780,000 persons will experience ACS each year in the United States. [3] Moreover, about 20-25% of patients with ACS reportedly also have diabetes mellitus (DM); predominantly type 2 diabetes mellitus [T2DM]). [4 5] Patients with ACS and DM have an increased risk of adverse outcomes such as death, recurrent myocardial infarction (MI), readmission, or heart failure during follow up. [6] Longer median delay times from symptom onset to hospital presentation, have been reported among patients with ACS and DM than patients with ACS alone. [7] DM is now considered to confer a risk equivalent to that of CAD for patients for future MI and cardiovascular mortality. [8]  ). [9] ACS and T2DM are often associated with high-risk factors such as low levels of physical exercise, obesity, smoking, and unhealthy diet. [10] Some of these and other risk factors, specifically glycemia, high blood pressure (BP), lipidemia, and obesity, are frequently addressed by health education interventions. [10] Health education interventions are comprehensive programs that healthcare providers deliver to patients aimed at improving patients' clinical outcomes through the increase and maintenance of health behaviours. [11] Along with education about, for example, medication taking, these programs seek to increase behaviors such as physical exercise and a healthy diet thus reducing patient morbidity or mortality. [11] Most diabetes education is provided through programs within outpatient services or physicians' practices. [12] Many recent education programs have been designed to meet national or international education standards [13][14][15] with diabetes education being individualized to consider patients' existing needs and health conditions. [16] Patients with T2DM have reported feelings of hopelessness and fatigue with low levels of self-efficacy, after experiencing an acute coronary episode. [17] Although there are numerous systematic reviews of educational interventions relating to ACS, or T2DM, an umbrella review providing direction on educational interventions for high risk patients with both ACS and T2DM is not available, indicating a need to gather the current evidence and develop an optimal protocol for health education programs for patients with ACS and T2DM. This umbrella review will examine the best available evidence on health educationrelated interventions for patients with ACS or T2DM. We will synthesize these findings to provide direction for health education-related interventions for high risk patients with both ACS and T2DM.
An umbrella review is a new method to summarize and synthesize the evidence from multiple systematic reviews/meta-analyses into one accessible publication. [18] Our aim is to systematically gather, evaluate and organize the current evidence relating the health education interventions for patients with ACS or T2DM, and proffer recommendations for the scope of educational content and delivery methods that would be suitable for patients with ACS and T2DM.

Participants
All participants were diagnosed with ACS or T2DM using valid, established diagnostic criteria.
The diagnostic standards included those described by the American College of Cardiology or American Heart Association, [3] National Heart Foundation of Australia & Cardiac Society of Australia and New Zealand, [19] World Health Organization, [20] or other associations.

Intervention Types
For this umbrella review, health education-related interventions refer to any planned activities or programs that include behavior modification, counseling, and teaching interventions. Results considered for this review included changes in clinical outcomes (including blood pressure levels, body weight, diabetes complications, HbA1C, lipid levels, mortality rate and physical activity levels), behavioral outcomes (such as diet, knowledge, self-management skills, selfefficacy, and smoking), and psychosocial outcomes (such as anxiety, depression, quality of life, and stress) and medical service use (such as medication use, health care utilization, and cost-

Study Types
Only systematic reviews and meta-analyses were included in this review.

Eligibility Assessment
The title and abstract of all of the retrieved articles were assessed independently by two reviewers (XL-L, YS) based on the inclusion criteria. All duplicate articles were identified within EndNote version X7 [21] and subsequently excluded. If the information from the titles and abstract was not clear, the full articles were retrieved. The decision to include an article was based on an appraisal of the full text of all retrieved articles. Any disagreements during this process were settled by discussion and, if necessary, consensus was sought with a third reviewer.
We developed an assessment form in which specific reasons for exclusion were detailed.

Assessment of Methodological Quality
The methodological quality and risk of bias were assessed for each of the included publications using the Assessment of Multiple Systematic Reviews (AMSTAR), [22] independently by the same two reviewers (see Table 1). The AMSTAR is an 11-item tool, with each item provided a score of 1 (specific criterion is met) or 0 (specific criterion is not met, unclear, or not applicable). [22 23] An overall score for the review methodological quality is then calculated as the sum of the individual item scores: high quality, 8-11; medium quality, 4-7; or low quality, 0-3. [23] If the required data were not available in the article, the original authors were contacted for more information. The low quality reviews (AMSTAR scale: 0-3) were excluded in this umbrella review.

Data Extraction
Data were independently extracted by two reviewers using a predefined data extraction form. For missing or unclear information, the primary authors were contacted for clarification.

Statistical Presentation of Results from Reviews
All of the results were extracted for each included systematic review or meta-analysis, and the overall effect estimates are presented in a tabular form. The number of systematic reviews or meta-analyses that reported the outcome, total sample (from included publications), and information of health education interventions is also presented in the Tables (Table 2, 3). [24] A final 'summary of evidence' was developed to present the intervention, included study synthesis, and indication of the findings from the included papers (Table 4). [24] This umbrella review calculated the Corrected Covered Area (CCA) (Appendix 2, 3). The CCA statistic is a measure of overlap of trials (the repeated inclusion of the same trial in subsequent systematic reviews included in an umbrella systematic review). A detailed description of the calculation is provided by the authors who note slight CCA as 0-5%, moderate CCA as 6-10%, high CCA as 11-15% and very high CCA is more than 15%. [25] The lower the CCA the lower the likelihood of overlap of trials included in the umbrella review.

Synthesizing the Results and Rating the Evidence for Effectiveness
The statements of evidence were based on a rating scheme to gather and rate the evidence across the included publications. 26 The statements of evidence were based on the following rating  [26] A rating of sufficient evidence in this review is obtained when systematic reviews or meta-analyses with a large number of included articles or participants produce a statistically significant result between the health education group and the control group. [26] Some evidence, is a less conclusive finding about the effects of the health educationrelated interventions [26] with statistically significant findings found in only a few included reviews or studies. Insufficient evidence, refers to not enough evidence to make decisions about the effects of the health education-related interventions, such as non-significant results between the health education group and the control group in the included systematic reviews or metaanalyses. [26] Insufficient evidence to determine, refers to not enough pooled data to be able to determine whether of the health education-related interventions are effective or not based on the included reviews. [26] Results

Characteristics of Included reviews
The selection process and number of studies at each step was illustrated as presented in Figure 1.
The database search yielded 692 publications, with removal of 197 duplicates and 371 articles that did not meet the inclusion criteria, 124 full-text articles were retrieved after applying the methodological quality rating (AMSTAR scale), and three studies [27][28][29] were removed due to low scores ≤3 on the AMSTAR scale. Fifty-one systematic reviews or meta-analyses  conducted between 2001 and 2016 and published in English were included ( Figure 1; Tables 1 -3); 15 relating to ACS. The overlap of the trials included in the 15 reviews and meta-analyses related to ACS was slight (CCA = 2.6%). For the 36 systematic reviews relating to T2DM, the overlap of trials within these 35 reviews and meta-analyses (one review [47] did not report the included patients with T2DM (16 papers did not report the total sample). The average sample size of included articles was 8,161 (range, 536-68,556) participants, however, 63 studies related to ACS and 177 studies related to T2DM, were included in more than one systematic review or metaanalysis (see Appendix 2 and 3 and CCA statistics). The sample of these studies would therefore be included more than once. Of the included systematic reviews or meta-analyses, eleven were published in The Cochrane Library. Nine of the articles described meta-analyses, 29 articles described systematic reviews, and the remaining 13 articles were described as systematic reviews and meta-analyses or meta-regressions or narrative reviews.  The methodological quality of the included publications is presented in Table 1. Thirty (58.8%) publications were classified as high quality (scores [8][9][10][11], and 21 (41.2%) publications were classified as medium quality (scores [4][5][6][7]. Twenty five (49%) reviews specifically provided an a priori design, while the use of such a design was unclear for 26 (51%) publications. The inclusion of other forms of literature (such as grey literature) was described in 18 (35%) reviews.

Methodological Quality of Included Systematic Reviews and Meta-Analyses
Only 14 out of 51 (27%) reviews included a table of included and excluded studies. Only 2 (4%) reviews did not provide a characteristics table of the included papers. The scientific quality of the included papers was evaluated and documented in 47 (92%) reviews. The scientific quality of the included studies was used appropriately to formulate conclusions in 47 (92%) reviews. The methods to combine the results of the included studies were appropriate in 43 (86%) reviews.
Publication bias was assessed in only 19 (37%) reviews. Finally, conflicts of interest were reported in 47 (92%) reviews.

Characteristics of Health Educational Interventions
The description of the health educational interventions followed the Workgroup for Intervention Development and Evaluation Research (WIDER) reporting guidelines for behavior change interventions. [81] The characteristics of the recipients, setting, delivery methods, intensity, duration, and educational content of health educational interventions for patients with ACS or T2DM are summarized in Tables 2 and 3. The delivery strategies for health education included face-to-face, internet-based, phone-based, videotape, written educational materials, or mixed.
The format included one-on-one (individualized), group, or both. Face-to-face sessions were the most common delivery formats, and many education sessions were also delivered by telephone/web contact or individualized counseling. The number of sessions, total contact hours, and durations varied, and there was limited information about the intensity of health education for patients provided. The frequency of educational sessions was weekly or monthly, and an average of 3.7 topics was covered per education session. Nurses and multidisciplinary teams were the most frequent educators, and most education programs were delivered post-discharge.

Effect of Interventions
The outcomes of the included systematic reviews and meta-analyses were summarized in Table

Patients with Acute Coronary Syndrome
Three major types of health education-related interventions were used for patients with ACS: general health education (only included general health information), psychoeducational interventions, and secondary prevention educational interventions (including strategies to promote a healthy lifestyle, manage medications, and reduce cardiovascular complications) as well as internet-based interventions.

General Health Education
The findings are based on our synthesis of the findings from 6 systematic reviews. [37 48 50 51 59 70] Overall, there were mixed effects of general health education on behavioral change or clinical outcomes in patients with ACS. There was some evidence of a positive effect of general health education on knowledge, behavior, psychosocial indicators, beliefs, and risk factor modification, but no effects for key clinical outcomes, such as cholesterol level, hospitalization, mortality, MI, and revascularization. The results for health-related quality of life, healthcare utilization, and costs were mixed; several reviews reported a significant change, and other reviews reported no significant change for these outcomes. Only one review focused on telephone-based health education. There is some evidence that telephone-based health education during cardiac rehabilitation might improve all-cause hospitalization, anxiety, depression, smoking cessation, and systolic BP but there is no evidence for improvements in all-cause mortality and reductions in low-density lipoprotein cholesterol. [59] Psychoeducational Interventions Strategies for psychoeducational interventions have a specific focus on smoking cessation, and  56 69] There is sufficient evidence that psychoeducational programs are effective at decreasing smoking, achieving smoking abstinence, and reducing depression. One review reported no effect on smoking cessation [31] or total mortality [56] .

Secondary Prevention Educational Interventions
The following statements are based on our synthesis of results from three papers. [34 41 44] There is some evidence that secondary prevention educational interventions reduce MI readmission rates and improve quality of life, but the intervention was ineffective in reducing revascularization, cholesterol levels, and improving smoking cessation rates. The results are mixed for mortality and re-infarction rates; two reviews [34 41] found positive effects on mortality, while one review [44] did not.

Patients with Type 2 Diabetes Mellitus
Ten types of health education-related interventions were used for patients with T2DM: culturally appropriate health education (tailored to the religious beliefs, culture, literacy, and linguistics of the geographical area), dietary advice, foot health education, group medical visits (a group education component taught by health professionals), general health education (only included general health information), improving the uptake and maintenance of medication regimes (e.g., promoting the use of oral hypoglycemic medications), lifestyle interventions (specific focus on dietary changes and increased physical activity, or stress management), psychoeducational interventions and self-management educational interventions (activities that promote or maintain the behaviors to manage T2DM often based on the National Standards for Diabetes Self-Management Education [13] ), and therapeutic education (collaborative process needed to modify behavior and more effectively manage risk factors).

Culturally Appropriate Health Education
Findings are based on our synthesis of results from 8 publications. [33 42 52-54 58 62 72] Overall, there was some evidence of the effects of culturally appropriate health education on clinical outcomes for T2DM. There was sufficient evidence that culturally appropriate health education improves HbA1c reduction and knowledge scores. There is some evidence that physical activity and clinical outcomes (blood glucose, HbA1c, BP) were improved. There were no data relating to adverse events during the intervention and follow-up (such as hypoglycemic events and mortality), and there was insufficient evidence about improvements in quality of life.

General Health Education
The statements are based on our synthesis of results from five papers. [40 46 60 74 79] Overall, there were mixed effects of general health education programs on clinical outcomes for T2DM, including HbA1C, cholesterol level, and triglyceride level. There was some evidence of the effectiveness of general health education on the management of glycemia, weight reduction, and some diabetes management outcomes (HbA1C, diabetes complications). There were no data supporting the effectiveness of general health education on reduced health service utilization, diabetes complications, improved knowledge, psychosocial outcomes, or smoking habits.

Lifestyle Interventions
The following statements are based on our synthesis of results from 6 reviews. [39 49 55 71 72 77] Overall, there were mixed effects of the lifestyle interventions on cholesterol level, HbA1C level, and body weight. There is some evidence that lifestyle interventions or behavioral programs are effective for blood glucose and BP management, but they were ineffective for reductions in HbA1c scores [71 72] .

Uptake and maintenance of medication regimes
The statements are based on our synthesis of results from three publications. [57 78 80] There is some evidence of the effectiveness of increased uptake and maintenance of medication regimes for taking medications for HbA1C regulation including oral hypoglycemic agents.

Self-Management Educational Interventions
The statements are based on our synthesis of results from 9 reviews. [43 47 61 64 65 67 68 75 76] Overall, there was sufficient evidence of the effects of self-management education interventions on HbA1c level, knowledge, lifestyle outcomes, and main psychosocial outcomes. However, there was insufficient evidence of the benefits of this education intervention on depression, quality of life, and body weight.

Other Health Education-Related Interventions
Other health education-related interventions for patients with T2DM included therapeutic education, foot health education, group medical visits, psychoeducational interventions, and dietary advice. Statements for all of these interventions are based on our synthesis of results from one review.
There is some evidence that foot health education is effective in reducing the incidence of lower extremity amputation. [32] There is some evidence that group medical visits are effective for improving HbA1c and systolic BP management [38] . There is also some evidence that psychoeducational programs are effective for improving HbA1c regulation and psychological status. [30] Finally, there is insufficient evidence that dietary advice improves glycemic and weight management or reduces micro-and macrovascular diseases. [63] There is also insufficient evidence for the cost-effectiveness of therapeutic education for patients with T2DM. [66]

Discussion
This umbrella review identified 51 systematic reviews or meta-analyses ( 69 70] It remains challenging to determine the specific strategy or format that is the most effective delivery mode for patients with ACS or T2DM. There is very limited evidence to guide clinicians on the duration, contact hours, educational content, delivery mode, total length, and setting of health education programs for cardiac patients. [50] For patients with DM, one study reported that more successful programs were longer than 6 months (longer duration), consisted of greater than 10 contact sessions (high intensity), and were one-on-one sessions with individualized assessment. [82]  Use of theory when designing behavioral change interventions may also influence effectiveness. [75] Health education using a cognitive behavioral strategy is most consistently effective in changing maladaptive illness beliefs, [51] and studies using more than 2 behavioral change strategies reported significant differences between the intervention and control groups. [31] In one review, a significant change in smoking cessation was not observed in subgroup analyses between studies that did or did not report using a theory in intervention planning; [56] however, the authors did not suggest that using a theory in program planning should be disregarded but reported that examining actual theories or mechanisms underlying health education programs is required. [56] Owing to the considerable overlap between different theories and the detailed description of the theoretical approach in only approximately 40% of the included papers, it is difficult to determine the most effective theoretical approach, but many models can be used with success, such as the health belief model (HBM), social cognitive theory (SCT) and

For patients with Acute Coronary Syndrome
transtheoretical model (TTM). [56 67 69 75] Three reviews [31 41 44] noted that some included studies used behavioural strategies such as goal setting. These strategies were found to be beneficial for patients with coronary heart disease.

For patients with Type 2 Diabetes Mellitus
Although the theoretical approach underpinning the health education program was not always described, 13 of the 36 reviews (36.11%) related to T2DM reported the theoretical approach used in their included studies. The most common theories were SCT (including self-efficacy), empowerment theories (eg., Empowerment Behavior Change Model, Self-determination and Autonomy Motivation Theory, Middle-range Theory of Community Empowerment) and TTM.
There is evidence that health education interventions based on a theoretical model are likely to be effective. [43] Vugt et al suggested that self-care education programs should be based on theories and that theory-based self-care interventions are more effective than non-theory-based programs. [75 83] Theories could help to specify the key target health behaviors and behavioral change techniques required to generate the desired outcomes. [75] The decision regarding the theory should be based on the aim of the program and factor for intervention. [77] Only one review reported that a theoretical approach underpinning the health education program is not necessary for better outcomes. [76] Fourteen reviews [30 33 40 46 52 57 60 63 64 67 68 73 75 77] reported that goal setting was conducted in the included studies. Goal setting by patients, health professionals or mutually-agreed goals were linked to improved patient outcomes.

For Patients with Acute Coronary Syndrome
Most reviews reported that the educational content of the interventions was comprehensive. The most common topics, of the average 3.7 topics per education session, were behavioral change, cardiovascular risk factors management, exercise, psychosocial issues, and smoking cessation.
An underlying principle of health education for patients with ACS is that knowledge is necessary, but not enough to develop health behaviors and change risk factors. [31 50] Age, cognitive factors, environmental factors, and social and economic background are also important considerations. [50] While interventions using a behavioral program, telephone-based content, or self-care are effective for smoking cessation, there was insufficient evidence to support that any type of educational program was more efficacious than the others. [69] Psychoeducation, which is defined as multimodal, educationally based, self-management interventions, 31 led to enhanced physical activity levels within 6-12 months when added to cardiac rehabilitation (CR) and was more effective than an exercise program or health education alone. [31 56] Moreover, psychoeducational interventions were more effective for patients with ACS than other types of

For Patients with Type 2 Diabetes Mellitus
The educational content for patients with T2DM focused more on behavioral change, diet, exercise, glycemic regulation, medication, and self-management. Health education that was selfmanagement was more effective for patients with T2DM. [40 47] In addition, based on the current evidence, the educational content should be culturally sensitive, especially for patients with T2DM; [33 42 54] culturally appropriate diabetes health education may have a greater impact on the management of glycemia and reduce diabetes complications. [77] The educational interventions for patients with T2DM focused primarily on HbA1C, lipid levels, quality of life, and body weight. HBM and SCT were the most common theories used in the included reviews.

For Patients with Acute Coronary Syndrome
Most reviews reported that the education was provided using multiple teaching methods and in multiple settings. Nurses and multidisciplinary teams were the most frequent people providing education, and most education programs were delivered post-discharge. Although face-to-face sessions were the most common delivery format, many education sessions were also delivered by telephone or through individualized counseling. Telephone-based health education appeared to be effective for reducing hospitalizations, systolic BP, smoking rates, depression, and anxiety. [59] The educational interventions for patients with ACS focused primarily on clinical outcomes (hospitalization and mortality), modifiable risk factors (BP, low-density lipoprotein levels, and smoking cessation), and other psychological outcomes (anxiety and depression). Mixed health education programs generally included group sessions combined with educatorfacilitated individual sessions, covering basic knowledge and problem-solving skills. These programs produced greater benefits and larger effect sizes for blood glucose reduction and knowledge levels in patients with T2DM. [47] In contrast, individual education programs have been reported as more effective in achieving outcomes than group-based education. This may be because education programs might be more efficient at addressing personal needs, with greater participant engagement. [73] However, one systematic review reported that individual and group patient education, demonstrated similar outcomes among T2DM patients. [46] Although face-to-face sessions were the most common delivery format, many education sessions were also delivered by telephone or individualized counseling. Face-to-face health education programs were most effective for enhancing blood glucose regulation and knowledge levels, while mixed delivery models (face-to-face, phone contact, online or web-based, or video) produced a moderate effect for knowledge levels. [47] Another review reported that face-to-face health education programs generated a greater benefit for metabolic management than those delivered using electronic communication technology. [73] Nurses (including diabetes nurses educators), community workers, dieticians and multidisciplinary teams were the most frequent educators, and most of the education programs were delivered post-discharge. Some reviews indicated that health education programs delivered by a group of different educators, with some degree of education reinforcement at additional points of contact, may provide the best results. [60 76] However, based on two studies that reported HbA1C at 12 months, it is indicated that the outcomes in studies with only a diabetes nurse as the educator also tended to do better than the outcomes in studies with a multidisciplinary team , while the biggest effect was seen when a dietician was the only educator. [76] Health education programs delivered by one person may focus more on the patient's ability than the educational content or quality of the health education programs. [76] However, no clear conclusion can be drawn whether having one educator delivering the intervention is best due to few information. [60] Delivery, Timing and Follow up

For Patients with Acute Coronary Syndrome
Most educational sessions were delivered weekly. Few reviews provided information regarding the duration of education interventions; when the duration was reported, it varied from 4 weeks to 48 months. These findings suggest that there is a significant gap in the evidence in relation to the duration, contact hours, educational content, optimal delivery mode, total length, and setting of health education programs for cardiac patients. [50] For patients with ACS, one systematic review that included 7 studies with a total of 536 participants reported that studies with education lasting at least 6 months resulted in the most significant changes in the primary outcomes (such as behavioral change, smoking cessation) [31] and that at least 12 months of follow-up is needed to evaluate the impact of telephone-based education. [59] Another review reported that the intensity of education programs is important for efficacy regarding smoking cessation: interventions with a very low intensity and brief interventions, do not have a significant effect, [69] and programs for smoking cessation among patients with coronary heart disease should last >1 month. [69] Most of the reviews were provided for patients with ACS in inpatient settings and then within post discharge settings, five reviews [31 36 45 48 59] did not explicitly state the settings in which the health education related interventions were provided.

For Patients with Type 2 Diabetes Mellitus
Education sessions were delivered weekly or monthly. Longer health education programs for T2DM (>6 months) produced larger effects for all primary outcomes (such as HbA1C). [47]  Health education lasting >3 months resulted in the largest effect size compared with health education of a shorter duration (<3 months). [33] For HbA1C, the effect size at 6 months seemed to be significantly greater than at 3 and 12 months; in other words, the effect size peaked at 6 months. [62] In general, health education of a greater intensity (longer duration and more sessions) was more effective for blood glucose reduction and knowledge levels among patients with T2DM. [47 74] Compared with health education programs covering only one topic, programs that included multiple or mixed educational topics yielded consistently greater benefits in blood glucose reduction and knowledge levels. [47] In addition, health education programs combined with specific behavioral change strategies (such as self-care strategies) seemed more effective than other programs. [47] Health education related interventions were mainly delivered in hospital settings, primary care settings, diabetes centers or community based settings; although six reviews [32 39 55 58 67 72] did not explicitly state the delivery settings.

Recommendations about Health Education Interventions for Patients with ACS and T2DM
These results from included systematic reviews and meta-analyses help to provide recommendations about the content of a health education intervention for patients with ACS and T2DM, requiring further evaluation. Future development of educational programs for patients with ACS and T2DM by health care professionals should consider the needs of people with these diseases. [37 40 42 70] Based on the results and findings from this umbrella review, recommendations are made in Table 5. The acute life-threatening nature of ACS requires that increased emphasis should be placed on cardiovascular risk factors in any combined education program. Both ACS and T2DM have common lifestyle factors such as inactivity and high fat diet requiring modifications. This overview potentially provides an estimate with the lowest level of bias for the impact of health education-related interventions for patients with ACS or T2DM and could be regarded as an all-inclusive summary of the current evidence base for health education for these patients.

Overall Completeness and Applicability of Evidence
While this umbrella review identified evidence for each of the types of health education, there was only a small number of reviews within some categories (such as psychoeducational intervention [30] and dietary advice [63] ), and these studies were not very informative. This umbrella review also found no reviews that systematically analyzed varying doses of health education; therefore, could not examine the dose-response effects. There was insufficient information about the evaluated doses (total contact hours and duration of education) to enable comparison of the benefits of differences in the magnitude of the doses across the different research. This umbrella review found no reviews focused on patients with ACS and T2DM; instead, all of the systematic reviews and meta-analyses focused on only one of these diseases.

Quality of the Evidence
The methodological quality of the included systematic reviews and meta-analyses varied. All of the included reviews or meta-analyses were of moderate to high methodological quality, as assessed using AMSTAR. However, only 30 (58.8%) systematic reviews or meta-analyses were rated as high quality and only three (5.9%) systematic reviews or meta-analyses [43 53 69] adequately met all 11 AMSTAR criteria. This indicates that some of the reviews included in this umbrella review may have limitations in their design, conduct, and/or reporting that could have influenced the findings when considered both individually and collectively. [32 65] The quality of the primary studies in the included systematic reviews or meta-analyses also varied. The main sources of bias were inadequate reporting of allocation concealment and randomization processes, as well as lack of outcome blinding. [33 42 69 70] This bias in the methodological quality led to lower quality assessments, which varied by results within each included review. Other reasons for lower methodological quality included heterogeneity in, or inconsistency of, the effect and imprecise findings. Heterogeneity between studies in this umbrella review was described in terms of the intervention, participant characteristics and length of follow-up. Heterogeneity was an important factor indicating the complexity of the health education interventions. [56] The variability in the approaches, tools or scales used to measure outcomes between the included studies are likely to introduce some heterogeneity. [30] The heterogeneity of the educational interventions seen in the reviews included in this umbrella review may reflect the uncertainty about the optimal strategy for providing health education to patients. [37] In addition, 240 studies were included more than once in the included reviews and meta-analyses. However, the overall overlap of studies among reviews and meta-analyses related ACS and T2DM was slight, CCA of 2.6% and 2.1%, respectively. [25] This umbrella review is the first synthesis of systematic reviews or meta-analyses to take a broad perspective on health education-related interventions for patients with ACS or T2DM. Given that health education is complex, the biggest challenge for systematic reviews or meta-analyses of health education is accounting for the potential clinical heterogeneity in health education-related interventions (content and delivery approaches) and the population of patients who receive health education. To facilitate comparisons across systematic reviews of health education and the efficient future update of this umbrella review, future reviews or meta-analyses need high-quality research and to standardize their design and reporting, including the reporting of included study characteristics, assessment criteria for risk of bias, outcomes, and methods to synthesize evidence synthesis.  psychosocial outcomes, readmission rates, and smoking status rather than clear evidence of reduced mortality, MI, or short and long-term complications. In addition, psychoeducational interventions were more effective for patients with ACS, and health education that was culturally appropriate or taught self-management was more effective for patients with T2DM. We also found that longer durations and high-intensity health education provided in an individualized format were more helpful for patients with ACS or T2DM.

Conclusions
The fact that none of the included reviews included patients with both ACS and T2DM indicates a clear need for further rigorous experimental studies with patients with both diseases. Future research that includes these aspects of education are likely determine the effectiveness of    [43] 48 Vermeire et al., 2005 [80] Yes     (Devi et al., 2015) [44] ; The

Number of session:
Weekly or monthly or unclear;

Total contact hours:
Unclear.

Number of studies:
Twenty-four papers reporting on 13 RCTs;

Number of studies:
Sixty-two independent studies.

Types of studies:
RCTs;

Number of studies:
A total of 14 studies were included.

Types of studies:
RCTs;

Number of studies:
Nineteen trials;

Types of studies:
RCTs; Total sample: Unclear. Unclear, did not describe.

Number of session:
Unclear; Total contact hours: Unclear, did not describe.

Number of session:
Monthly;

Total contact hours:
Unclear;

Number of studies:
10;

Number of session:
Weekly; or 2 messages daily or unclear;

Total contact hours:
Unclear.

Number of studies:
Seventeen studies;

Types of studies:
RCTs; Total sample: Unclear. nutritionists or unclear.
Strategies: Face-to-face or unclear; Format: individual and group lessons;

Number of studies:
Thirteen studies;

Types of studies:
RCTs and comparative studies; Total sample: Unclear.

Number of session:
Unclear;

Total contact hours:
Unclear; Duration: One session -

Number of studies:
Seven studies;

Types of studies:
RCTs; Total sample: Unclear.
Educational interventions. Nurses; pharmacists; other skilled healthcare professionals.

Number of studies:
Forty-seven studies; Types of studies:

Number of studies:
Nine studies;

Types of studies:
RCTs;

Types of studies:
RCTs and RCT was followed by a beforeand-after study;

Number of studies:
Twenty-one published trials;

Types of studies:
RCTs and CCTs; Total sample: Unclear.

Number of session:
Unclear;
Strategies: Face-to-face; telephone; Format: face-to-face; group based and telemedicine;

Number of studies:
A total of 11 trials;

Types of studies:
RCTs;

Number of session:
Unclear;

Types of studies:
RCTs;

Number of studies:
Sixty-three RCTs;

Number of session:
Unclear;

Total contact hours:
Unclear.

Continuous smoking cessation
Total mortality Barth et al., 2008 [36] Abstinence by self-report or

Some evidence
Vermeire et al., 2005 [80] HbA1c Dietary advice One/1,467 patients Nield et al., 2007 [63] Glycaemic control ( (+) ：Intervention Group is Significantly Better than Control Group; For example: "91% studies (+)" means 91% studies reported a significant better compared with control group;       (2011) Six-year follow-up of a randomised controlled trial examining hospital versus home-based exercise training after coronary artery bypass graft surgery. Heart 97 (14): 1169-1174.  (2002) Coaching patients with coronary heart disease to achieve the target cholesterol: A method to bridge the gap between evidence-based medicine and the "real world"randomized controlled trial. Journal of Clinical Epidemiology. 55 (3)