Patient-centered education in dyslipidemia management: a systematic review

Abstract Background Dyslipidemia management is crucial to reduce mortality and morbidity from cardiovascular diseases (CVDs). Patients must be educated and empowered to enable them to manage their own diseases. Various methods of patient education, such as patient-centered education (PCE) or non-PCE (such as didactic education or any traditional form of education), have been implemented. Objective To review and determine the effectiveness of PCE for dyslipidemia management compared with usual care. The primary outcome chosen was cholesterol level. Other measures, such as psychosocial or cognitive, behavioral, and other relevant outcomes, were also extracted. Additionally, underlying theories and other contributing factors that may have led to the success of the intervention were also reviewed and discussed. Methods We conducted searches in PubMed, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Scopus, and Google Scholar from inception until April 2021. All studies involving randomized controlled trials were included. Study quality was assessed using the Critical Appraisal Skills Program (CASP) checklist specifically for randomized controlled trials. Results The search identified 8,847 records. Of these, 20 studies were eligible for inclusion. Interventions using a PCE approach were largely successful. Contributing factors extracted from the included studies were underlying theories, instant reward system, dietary education, collaborative care, duration of intervention with systematic follow-ups, social support, adherence assessment method, and usage of e-health. Conclusions PCE is successful in achieving the desired outcomes in dyslipidemia management. Future studies may incorporate the elements of PCE to improve the management of dyslipidemia in hospital or community settings where appropriate.

appropriate management of dyslipidemia would significantly impact cardiovascular morbidity and mortality [5]. Due to its asymptomatic nature, the management of dyslipidemia must be strategized and addressed by health-care professionals. Synchronized contributions from all parties, including patients and health-care professionals, have been proposed to improve patients' clinical outcomes and subsequently reduce the burden of CVD [6].
Many articles have highlighted the importance of selfmanagement among patients with chronic conditions such as CVD. For example, in their study, Grady and Gough [7] asserted that self-management is crucial in treating chronic conditions. The authors also mentioned that self-management emphasized the patient's responsibility to manage their disease while guided by their health-care professionals. This stance is supported by findings from Bosworth et al. [8], which highlighted that the self-management strategy is effective and should be integrated as a crucial component in providing highquality care to patients with CVD [8].
For patients with chronic disease to be successful in selfmanagement, they must first be informed and empowered [9]. This approach will allow patients to adopt preventive and curative steps in managing their conditions [10]. The relationship between self-management and patient empowerment has been explored and discussed in many articles. Rappaport [11] defines empowerment as a process by which people gain mastery over their affairs. The empowerment process involves the provision of knowledge, skills, and responsibility to patients, which will lead to behavioral change and potentially improve their overall health [10].
In their review, Bosworth et al. [8] describe the concept of self-management in managing patients with hypertension and heart diseases. The authors have proposed 5 significant factors to ensure the success of self-management, including patient-centered programs [8]. Exploring the concept of patient-centeredness, Stewart et al. [12] proposed 6 dimensions of patient-centered care, which are as follows: (1) exploring both the disease and illness experience; (2) understanding the whole person; (3) finding common ground; (4) incorporating prevention and health promotion; (5) enhancing the patient-doctor relationship; and (6) being realistic. All these dimensions may be incorporated in patient-centered education (PCE). Comparing PCE with didactic education, the latter mainly involves a patient as a passive recipient receiving standardized information from health-care professionals [9]. In this situation, the patient is expected to passively adhere to the instructions and advice given by the health-care professional [9]. By contrast, in PCE, patients are treated with deep respect, listened to, and actively involved in their own plans of treatment, with their wishes being honored throughout their disease management journey [13]. Different types of education may affect patients differently and act as a critical component in determining the success of self-management. Effective patient education will then lead to behavioral change and may improve the patient's health generally. To our knowledge, there is no published review to study the effectiveness of PCE in dyslipidemia management. Therefore, this systematic review aims to analyze the effectiveness of PCE in dyslipidemia management in comparison with usual care. In addition, we also examine the underlying theories and other elements of patient-centered care, which may contribute to the success of the interventions.

Literature search
We conducted a systematic review of the published literature in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) guidelines [14]. We searched the literature in 3 databases, namely PubMed, Cumulative Index to Nursing and Allied Health Literature (CINAHL), and Scopus, from inception to April 2021. Additional articles were sought by reviewing references of eligible studies and conducting a search on Google Scholar. Studies were identified with the following search terms: ("patientcentered education" OR synonyms) AND ("dyslipidemia" OR synonyms). We included both medical subject heading (MeSH) terms and free-text terms. Full search terms are shown in Table 1. A filter was applied to the searches on the databases to restrict retrieval of off-target articles, whereby non-English articles were excluded.

Study selection
By adopting the population, interventions, comparisons, and outcomes (PICO) framework as a template, the inclusion and exclusion criteria presented in Table 2 were applied to set limits to the scope of the review. The population (P) aspect focused on patients with dyslipidemia, which commonly presents with CVD. The intervention (I) was PCE. In determining whether the interventions were patient-centered or not, the definition of patient-centered care by Robinson et al. [15] was adopted. They described the fundamental characteristics of patient-centered care, which are patient involvement in the care and individualization of patient care. In terms of patient involvement, studies were only included when the authors clearly described active participation from the patients. Some studies described that the patients were actively involved in the care plan, while some studies also mentioned specifically the usage of shared decisionmaking (SDM). SDM is a method when the patients and health-care professionals make decisions together [17]. The SDM approach aims to ensure that decisions are made based on the patient's. values and preferences.
Regarding individualization, the education provided in the study must be targeted according to patients' specific needs. Garvey et al. [18] supported this fundamental characteristic of patient-centered care when describing that one of the goals of patient-centered care is the development of an individualized treatment plan. A study was also included in the review when the researchers used other established patient-centered approaches such as motivational interviewing (MI). Elwyn et al. [17] asserted that MI and SDM are patient-centered methods, and both have been associated with significant improvement in patient outcomes.
These interventions were compared against the usual care (C), such as a didactic or traditional form of education. Finally, the outcome (O) measure in this review is the cholesterol level or any other psychosocial, cognitive, or behavioral outcome reported in the studies. Articles retrieved from the database search were then exported to a reference managing software EndNote X9 (Clarivate), and duplicates were omitted. Because our focus was on randomized controlled trials (RCTs), only studies with the keyword "trial" were included using the smart screening tool in EndNote. The eligible studies underwent preliminary screening by 2 authors independently. The authors screened both titles and abstracts for potential relevance in line with the PICO criteria. Articles that met the predefined eligibility criteria were included for full-text screening and were assessed independently by the 2 authors. Any disagreements between the authors regarding study inclusion were resolved through discussion and consensus with the third author. The inclusion and exclusion criteria are listed in Table 2.

Data extraction
The key characteristics and pertinent information from the eligible studies were extracted and documented. The extracted information included author, year of publication, study location, study design, sample size, patient demographic, and study setting, which are summarized in Table 3. Other information collected were methods of intervention such as the health-care professionals involved, intervention duration, elements involved in the intervention group, and the underlying theories or models being used, summarized in Table 4. The key mechanisms were retrieved if no theories or models were presented by the studies. The outcomes of the intervention were also extracted and are summarized in Table 5. These included psychosocial, cognitive, or behavioral outcomes, cholesterol level, and cardiometabolic outcomes, such as blood pressure (BP), weight, body mass index (BMI), and glycated hemoglobin A (HbA1c). Low-density lipoprotein (LDL) reduction was also compared between the intervention and control groups and is reported in Table 5. For studies that did not measure LDL, the difference in total cholesterol (TC) or triglyceride (TG) level Asian Biomed (Res Rev News) 2022; 16(5):214-236  was extracted and reported. Significant information, such as P-value, odds ratio (OR), relative risk (RR), and confidence interval (CI), was also included whenever reported. All values were reported in the intervention group in comparison with the usual care. Other reported outcomes that did not fit any of the criteria mentioned earlier, but were relevant in dyslipidemia management, were also extracted and are described in Table 5.

Study quality assessment
The Critical Appraisal Skills Programme (CASP) [16]) for RCTs was adopted to evaluate the quality of the studies included. The CASP checklist consists of reviewing the RCT in terms of its basic study design (clarity of research question, appropriateness of randomization method), the methodology (blinding method, study groups), the results (effects of intervention being reported comprehensively, precision, benefits that the study brings), and in terms of the impact of the results to our targeted population. A score of 1 was assigned for each criterion if "yes" was the response, whereas a score of 0 was assigned if the response was "no" or "uncertain." The review of quality assessment was conducted independently by 1 reviewer and was further assessed by a second reviewer to avoid the risk of bias. Any disagreements were resolved by consensus before finalizing the articles to be included in the study. The criteria assessments for all the included studies are summarized in Table 6.

Results
Literature selection Figure 1 shows the study selection flowchart. Study selection was based on the PRISMA guidelines [14]. We included 20 studies in this systematic review.

Patient-centered approaches
The details of interventions are described in Table 2. The interventions involved various health-care professionals, including nurses, pharmacists, physicians, dietitians, kinesiologists, nutritionists, psychologists, and internal medicine specialists. Most of the studies involved nurses in the interventions. Some studies involved single health-care professionals only [3,19,20,30,31,33,34], while others used the collaborative care concept, with ≥2 health-care professionals involved in the interventions. The most used theories in the interventions were the chronic care model, social learning theory, social cognitive theory, transtheoretical theory, health belief model, and MI.

Outcome measures
All studies reported cardiometabolic outcomes, such as the BP reading, weight, BMI, or HbA1c, except for 4 studies that measured no cardiometabolic parameters [3,19,23,33]. Most of the studies measured cardiovascular risk factors as one of their outcomes, as either primary or secondary outcomes. Cholesterol indices, namely, the TC, TG, LDL, and high-density lipoprotein (HDL) levels, were reported in 18 studies. Some      Table 6. Continued.
Based on the studies reported, the findings on cholesterol levels comprised both positive and negative outcomes. Other outcomes measured, such as psychosocial or cognitive outcomes showed promising results in most studies. Improvement in cholesterol level was reported in 11 studies. Significant reductions in all cholesterol-related parameters (TC, LDL, and TG levels), but no significant change in HDL, were found by 2 studies [24,31]. However, a few other studies [19,26,[30][31][32][33]35] reported any 1 of the cholesterol levels (TC/LDL/TG) as improving significantly. Only 1 study reported significant improvement in the HDL level [33]. Most studies reported positive outcomes in the aspect of psychosocial, cognitive, or behavioral outcomes. Significant improvements in other cardiometabolic markers, such as BP, weight, BMI, and HbA1c, were also reported [6,20,21,22,24,27,30,31,35].

Discussion
Overall, based on the reported studies, PCE had successfully improved patients' cholesterol level, as well as the psychosocial, cognitive, behavioral, and cardiometabolic outcomes.
There are a few possible contributing factors that may lead to the effectiveness of the PCE. This discussion is further divided into a few subheadings deliberating on each factor in more detail.

Underlying theory
Firstly, the desired outcome achieved might be due to the underlying theory used by the researchers. Studies that reported a significant improvement in cholesterol levels (TC/TC/ LDL) used MI as one of the key concepts [6,21,24,29,30,36]. MI is a communication style that uses specific techniques such as reflective listening, SDM, and eliciting change talk [37]. In addition, the PCE utilizing MI was able to significantly affect the smoking status, which was associated with an improvement in the CVD risk score. Besides smoking status, MI was also beneficial in improving behavioral outcomes, such as physical activity and self-monitoring [21]. Nevertheless, other studies using the MI technique showed no significant improvement in cholesterol levels but achieved significant improvement in psychosocial or cognitive outcomes, such as patient activation [29]. Many confounding factors contribute to the success of the MI intervention, such as health-care professionals' skills, differences in participants' characteristics, and time frame of the study. To conduct MI, health-care professionals need to be highly trained. This hypothesis is supported by a study conducted by Allen et al. [24], which suggested that use of MI coupled with appropriate training of health-care professionals would lead to the successful achievement of the desired outcomes. Their study found that using certified nurse practitioners and MI training was associated with significant improvement in psychosocial or cognitive outcomes, cholesterol levels, and measurements of SBP, DBP, and HbA1c [24]. Although MI is a proven and practical method in patient-centered care, as mentioned by Elwyn et al. [16], it is also important for future studies to emphasize the factors that may contribute to the success of the MI, such as the incorporation of health-care professionals' training.

Rapid reward system
For behavioral outcomes, many studies reported positive findings in terms of physical activity [6,21,25,27,31], consumption of healthy foods [6,20,25,31,33], self-monitoring frequency [23], medication adherence [28,30,31], self-care activities [30], and smoking status [36]. Other studies [6,22,31,32,35] found no significant impact on the smoking status of the participants, except for 1 study by Daumit et al. [36]. This discrepancy may have occurred for a few reasons. A reason for the success of the intervention described by Daumit et al. [36] is the introduction of a point reward system. The investigators successfully educated the patients about smoking cessation, even though encouraging smokers to quit is extremely difficult. In the study, the participants were given points for attendance and behavioral change for smoking cessation. Finally, the points were exchanged with small reward items [36].
This proved that patients might be more motivated by a rapid reward system rather than the promising long-term rewards, such as their health outcomes. This hypothesis is supported by a study by Licthman et al. [38], who described that patients with dyslipidemia are less motivated to improve their cholesterol levels because of the asymptomatic nature of the disease. The motivation of patients with diabetes mellitus is different in that patients can quickly feel the changes in managing their symptoms [38]. Therefore, future studies should incorporate a rapid reward system as an alternative to boost patients' motivation, especially in promoting behavioral change that demands a lot of effort and sacrifice.

Dietary education
Consumption of heart-healthy food may lead to significant improvement in HDL levels. However, none of the studies reported significant improvement in HDL levels, except 1 study [33]. It is important to note that improvement in HDL levels could not be achieved solely by taking a lipid-lowering therapy, unlike LDL levels, which decrease with the help of pharmacological treatment. By contrast, HDL level improvement requires a combination of aggressive behavioral interventions, such as a healthy lifestyle, diet, and exercise. The expected significant increment of HDL achieved by Mok et al. [33] was due to the aggressive dietary intervention of the nurses involved in the study. The nurses emphasized appropriate dietary intake and encouraged the patients to eat hearthealthy food. Consequently, there were significant differences in the consumption of saturated fats and healthy food choices. The importance of achieving desirable levels of HDL, TC, LDL, and TG should be emphasized.

Collaborative care
Collaborative care is a method in which many experts collaborate to improve the quality of health care [39]. Healthcare professionals from various disciplines work together as a single unit with centralized communication and coordination to potentially improve patient outcomes [40]. Of the 20 studies included, 12 used collaborative care methods. Goyer et al. [6] involved the highest number of health-care professionals: nutritionists, psychologists, kinesiologists, nurses, and physicians. This collaborative care approach resulted in significant improvement in the psychosocial or cognitive aspect of mental health status. In addition, the behavioral outcomes of calorie intake and physical activities showed significant improvement, together with TG and TC levels.
Ho et al. [28] conducted a study involving primary care clinicians and pharmacists to improve medication adherence among patients with acute coronary syndrome after hospital discharge. The pharmacists provided education and guided the patients to adhere to their medications. The primary care clinicians were then notified about their patients' medication adherence status using a computerized medical record. The pharmacists' contact number was also included so that they could be reached for any further questions or clarifications. After 12 months of such intervention, significant improvement in medication adherence was observed [28].
In terms of the pharmacists' role, medication adherence is one of the crucial components in dyslipidemia management. Medication adherence is an outcome measured in some studies, but the methods of measuring adherence differ between studies. This difference in the method may result in outcome variability. Thus, it is rather challenging to make a generalized conclusion about medication adherence. Some studies used medication refill data from the pharmacy to measure medication adherence [28]. However, the record taken from the pharmacy refill data may be questionable as patients might not take the medication as prescribed, although they did not miss collecting the medications at the pharmacy [39].
Jarab et al. [30] measured medication adherence via self-reports from the patients, and significant improvement was observed. The authors reported that recall bias and social desirability might have occurred during the selfreporting [30]. Despite the disadvantage of self-reporting, Fortin et al. [21] mentioned that this technique is consistent with a patient-centered approach. Byrne et al. [27] measured medication adherence using a biochemical urine test and found no significant differences between groups. A biochemical urine test identifies the level of statin in the urine and is only applicable to test for atorvastatin [27].
Collectively, various methods of adherence measurement may affect the findings. To date, measuring medication adherence by using a biochemical urine test seems to be one of the alternatives. However, information regarding the cost and time constraints for implementing this method is scarce [27]. Various methods of measurement, such as self-reporting, records taken from pharmacy refill data, or biochemical urine tests, have their advantages and disadvantages. In highlighting the pharmacist's role, future studies in measuring medication adherence should be conducted based on resource availability, drugs involved, and patients' preferences.
The success of collaborative care in the interventions described earlier by 2 studies [6,28] was associated with focused education content based on the health-care professionals' expertise. The collaborative care or multidisciplinary approach should be extended to patients with CVD to optimize their care and risk reduction [6]. Future studies may consider the element of collaborative care to optimize the roles of different health-care professionals who have similar aims in improving patients' overall clinical outcomes.

Duration of intervention and systematic follow-ups
Duration of intervention can contribute to the efficacy of PCE. PCE has been used to reduce CVD risk among patients [36]. Significant improvement in smoking status was observed after 18 months. The effectiveness of the intervention, which leads to behavioral change, is possibly due to its longer duration. An individualized counseling session was conducted weekly or every 2 weeks. Systematic follow-ups played an important role in determining the effectiveness of the intervention. The time frame of the reported studies varies between 2 months and 4 years. In one of the studies reported by Maindal et al. [32], there was no significant difference in the CVD risk levels after 3 years of interventions. Although outcomes were measured after 3 years, there were no systematic follow-ups implemented before measuring the outcomes. Therefore, the authors [32] suggested providing gradual sessions at 1 month, 2 months, and 3 months after the core interventions.
Mok et al. [33] emphasized that systematic follow-ups are important. The authors highlighted that follow-ups conducted via telephone calls were a key component that led to significant positive dietary changes. The telephone calls were made fortnightly for the 2-month duration [33]. This assertion is supported by Jarab et al. [30], who found significant differences during the 6 months of their study duration. However, they were unsure of the impact of the intervention beyond that time [30]. Future studies should be conducted to explore the minimum duration of follow-up needed to sustain the desired outcomes. Systematic follow-ups, either face-to-face or by telephone calls, are crucial to ensure the effectiveness of the intervention.

Social support
The role of family members throughout the intervention period has been highlighted [31]. Recruitment of family members can encourage social support for the patients [41]. If there is no support available through a family, finding supporting individuals or groups is important for the patients to initiate behavioral change [41]. Dunbar et al. [42] asserted that social support is associated with better CVD outcomes. They indicated that family support is linked with the adoption and maintenance of healthy behaviors.
Kitko et al. [43] emphasized the role of caregivers in individuals with heart failure. They highlighted that many individuals with heart failure depend on support from their partners, families, friends, or neighbors to help them manage chronic diseases. Strom and Egede [44] also asserted that social support also significantly improves patients' motivation. Therefore, the active involvement of family members or other individuals to provide social support to the patients throughout the intervention is strongly encouraged.

Usage of e-health
E-health has been discussed in the literature for more than a decade. E-health is an emerging field that refers to health services and information that are conveyed or enhanced through the Internet or related technologies [45]. Snyder et al. [46] mentioned that information technology is able to promote patient-centered care by providing a mechanism for the patient to share their information with health-care professionals. Wong et al. [47] reported that online counseling might serve as an alternative to reach those who remain untreated.
Moessner and Bauer [48] found that an Internet-based service reached a substantial proportion of individuals with eating disorders. More than half of the participants (57.3%) said that the Internet was their first choice for seeking professional assistance. The participants indicated that they would not have shown behavioral changes had it not been for the online services [48]. Many articles have explored online counseling for mental health patients, but limited research has been done on managing CVD patients.
Of the 20 articles included in this review, only 1 study described an e-health intervention [3]. Despite the promising outcomes of this intervention, Eaton et al. [3], who conducted a study on utilizing e-health or electronic devices to improve cholesterol management, did not find any significant improvement in any of the outcomes. The authors considered that the e-health tools might be improved in the future by making them more user-friendly. Before 2011, when the study was conducted, the patients might not have been familiar with technology and might have found it difficult to use, resulting in minimal benefits. Today, people are more familiar with technology, especially after the coronavirus disease-2019 (COVID-19) pandemic. O'Leary [49] found that there was a rapid demand for efforts to use technology to cope with the damage caused by COVID-19. This opportunity should be utilized by the health-care system, whereby online methods should be implemented as much as possible when appropriate. When more people are able to adapt to technology, the delivery of technology-assisted online counseling continues to grow [47]. Therefore, looking at the opportunity of available technology and readiness of patients, education using online platforms may produce a promising result. Online counseling may replace telephonic or face-to-face counseling for systematic follow-ups for patients with CVD and help particularly in managing their dyslipidemia. Online counseling also may enhance the patient-doctor relationship, which is a dimension in the patient-centeredness concept highlighted by Stewart et al. [12]. However, the performance of the utilized software and digitalized platforms should be carefully examined to ensure that they can achieve the objective of patient-centered care.

Conclusion
The use of PCE is beneficial and superior to the usual form of care for managing dyslipidemia. Most studies reported successful findings in terms of either clinical outcomes (cholesterol level) or psychosocial, cognitive, or behavioral outcomes. Underlying theory, the rapid reward systems, dietary education, collaborative care, duration of intervention with systematic follow-ups, social support, and usage of e-health platforms have important roles in achieving desired outcomes.
To our knowledge, this review is the first to report the effectiveness of PCE for patients with dyslipidemia. The studies that were included in the review were methodologically sound and confirmed with CASP. Our review is comprehensive as it reports not only clinical outcomes in terms of cholesterol levels but also includes psychosocial, cognitive, behavioral, and other cardiometabolic parameters. This enabled the reviewers to analyze how the interventions may affect 1 or all of the desired outcomes being measured. This review also extracted patient-centered elements, other than education, which would be helpful in future studies.
However, this review also has a few limitations. First, we integrated research with a variety of interventions, measures, and follow-up times, which made it challenging to synthesize clear conclusions. Study designs and outcomes may have produced a more consistent finding if we had focused on smaller groups. However, this approach would have limited the number of studies we may have included. Some of the research works included in this review were conducted with specific target groups that may not be generalizable to all populations. Therefore, our findings should be interpreted with caution.
Second, there are many confounding factors that may affect the success of the interventions, such as the health-care professionals' training, patient demographics and characteristics, their financial status, and social support that they received. All these will eventually affect the outcomes measured after the interventions. Hence, the findings may not be generalized, and the results should be extrapolated with caution. Third, there is a possibility of medication changes made by the primary physicians in each study, but not being reported. This could potentially affect the final results. This limitation serves as a guide for future studies to include all interventions made throughout the study to enable more conclusive findings. Finally, the outcomes reported were not consistent in all included studies. The majority of the studies reported the LDL levels, whereas other studies reported other cholesterol indices. Complete cholesterol indices comprising TC, TG, LDL, and HDL levels would help to provide clearer conclusions.
Despite these limitations, the present review describes the effects of PCE compared with usual care or didactic/traditional forms of education. It is also hoped that the present review will provide guidance for patient-centered intervention for healthcare professionals in managing patients with dyslipidemia.
Author contributions. All authors made substantial contributions to the conception and design of the study and acquisition of data. FFI and AMR contributed substantially to data analysis. FFI drafted the manuscript, and AMR and CWW revised it critically for important intellectual content. All authors approve the final version submitted for publication and agree to be accountable for all aspects of the work and take responsibility for statements made in the published article.