Abstract

Background. Psychological symptoms such as depression and anxiety are quite common among stroke survivors and have great negative impacts on patients. Objective. To develop a care bundle through reviewing and integrating care strategies for psychological symptoms after stroke and then improve the bundle by Delphi study. Methods. A structured search of the literature was performed to identify studies evaluating interventions for stroke patients with psychological symptoms such as depression and anxiety. Two trained researchers screened papers through the titles, abstracts, and full-texts independently. All studies complying with the eligibility criteria were appraised using quality assessment tools. Related interventions with evaluated evidence levels formed the preliminary bundle. Afterward, the Delphi study was carried out to improve the bundle, and the experts were contacted by e-mail. Ten clinical experts specialized in stroke and psychological rehabilitation were recruited. The reliability of experts was represented by the effective response rate and authority coefficient (Cr). The consensus was deemed to be reached when the mean score for item importance is all equal or above 3.50 and the coefficient of variation (CV) is all equal or below 0.20. The Kendall coefficient W test was adopted to evaluate the consensus on agreement among the experts as well. Data analysis was performed using SPSS V.22.0. Results. Through a systematic evidence summary and two-round Delphi study, the items that were given high scores and got consensus by experts were chosen for the bundle. The iDAME bundle consisted of five interventions eventually: maintaining Interaction, tailored Diet, Acupressure, mindfulness Meditation, and physical Exercise. Conclusion. The development of an evidence-based and consensus-based iDAME bundle which integrated western and traditional Chinese medicine intervention was described. Evidence summary made the bundle become scientific, while the Delphi study made it more maneuverable. Based on these results, the bundle would be potentially implemented in stroke patients for their psychological symptoms.

1. Introduction

Psychological symptoms are quite common among stroke survivors, such as depression, anxiety, apathy, and bipolar disorder [1]. An updated systematic review and meta-analysis found that the prevalence of poststroke depression was 31%∼33% [2], and 55% of patients might have depressive symptoms within 15 years after stroke [3]. The prevalence of anxiety within 10 years after stroke was 17%∼24%, and 57% of stroke patients might experience anxiety symptoms during this period [4]. It indicates that the adverse psychological symptoms of stroke patients occur in the long term and are persistent. Furthermore, psychological symptoms after stroke have great negative impacts on patients, such as self-efficacy reduction, rehabilitation stagnation, and much more heavy economic burden from the extended LOS (i.e., length of hospital stay) [5, 6].

At present, the management of psychological symptoms after stroke can be subdivided into pharmaceutical and nonpharmaceutical management. Common drugs used in clinical treatment include selective serotonin reuptake inhibitor (SSRI), serotonin-norepinephrine reuptake inhibitor (SNRI), a noradrenergic and specific serotonergic antidepressant (NaSSA), tricyclic antidepressants (TCAs), Wuling capsule (traditional Chinese medicine), and Shugan Jieyu capsule (traditional Chinese medicine) [7]. Though pharmaceutical treatments have a significant effect, most of them are accompanied by serious gastrointestinal reactions, liver or neurological damage, and other side effects. Current nonpharmaceutical management mainly includes physical therapy, psychotherapy, and traditional Chinese medicine treatment technology. Repetitive transcranial magnetic stimulation (rTMS) and transcranial direct current stimulation (tDCS) were applied to the treatment of psychological symptoms after stroke. A meta-analysis by Shen et al. analyzed a total of 1764 stroke patients in 22 randomized controlled trials and found that rTMS was effective for poststroke depression [8]. Bucur and Papagno demonstrated the efficacy and safety of tDCS and rTMS in stroke patients with psychological symptoms [9]. Exercise therapy such as muscle strengthening, mobilization exercise, and aerobic treadmill training was also confirmed that could provide positive contributions to mood improving after stroke through some clinical trials [10, 11]. Cognitive behavior therapy (CBT), behavioral activation therapy (BAT), problem-solving therapy (PST), mindfulness-based intervention (MBI), interpersonal therapy (IPT), and motivational interviewing (MI) are effective psychotherapies for mood disorders after stroke, suitable for patients of poor medication compliance and adverse drug reactions [7]. The effectiveness of traditional Chinese medicine treatment technology (e.g., acupuncture, moxibustion, and acupressure) and other adjuvant treatments such as music therapy, meditation, relaxation exercise, and mind-body exercise (e.g., Taichi, Qigong, and yoga) are also demonstrated and have the potential for clinical promotion [7, 12, 13]. However, the contributing factors of psychological symptoms are complex and various, and comprehensive interventions should be worked out. The existing combinations of interventions are empirical and lack of evidence-based. Hence, a synthesis of the existing literature to develop a care bundle for stroke patients with psychological symptoms is required.

A care bundle is a set of evidence-based or self-evident good-practice-based interventions for a defined patient population and care setting [14]. It is a concept proposed by the Institute for Healthcare Improvement (IHI), which refers to combining and integrating evidence-based treatments and nursing measures into a set of intervention program, so as to achieve the goal of treating clinical diseases and improve the quality of care and patients’ outcomes [15]. The combination of multiple intervention measures is often more efficient and beneficial than the single performed alone.

In this study, we present the development of a poststroke psychological symptom care bundle called iDAME. iDAME is the abbreviation of maintaining Interaction, tailored Diet, Acupressure, mindfulness Meditation, and physical Exercise. These five interventions were obtained through evidence summary and two-round Delphi. The formation of the whole bundle is evidence-based; it not only combines the best available evidence but also attaches the importance to the individual clinical experience. The best evidence, practical experience, and patient’s actual condition are comprehensively considered in medical decision making to achieve the transition from evidence to practice effectively.

2. Materials and Methods

2.1. Evidence Summary
2.1.1. Literature Search

According to the 6S evidence resource pyramid model [16], related practical guidelines, evidence summaries, expert consensus, systematic reviews, and clinical trials for the management of poststroke psychological symptoms were retrieved.

To obtain empirical studies on the management of poststroke psychological symptoms, a comprehensive search of multiple databases from January 2010 to May 2019 was conducted, including SinoMed, Wan Fang Database, Chinese National Knowledge Infrastructure (CNKI), PubMed, EMBASE, CINAHL, Web of Science, JBI, and Cochrane (Table 1). In addition, the relevant clinical practice guideline websites were also searched, such as the Scottish Intercollegiate Guidelines Network (SIGN), the National Institute for Health and Care Excellence (NICE), the Registered Nurses’ Association of Ontario, Canada (RNAO), American Heart Association/American Stroke Association (AHA/ASA), Neurocritical Care Society (NCS), European Stroke Organisation (ESO), Royal College of Physicians (RCP), National Stroke Foundation (NSF), and Royal Dutch Society for Physical Therapy (KNGF) using the search terms “stroke,” “cerebrovascular accident,” “psychological disorder,” “psychological symptom,” “mental problem,” “emotional problem,” and “mood stress.” Reference lists of relevant articles were cross-checked and pertinent journals were hand searched for articles. The inclusion criteria were (1) studies written in English and Chinese and (2) studies that involved psychological symptoms in the context of the poststroke period and addressed any aspect of care. The exclusion criteria were (1) duplicates, review articles, and unpublished manuscripts and (2) studies that did not report poststroke psychological symptoms in their findings.

2.1.2. Search Results and Quality Appraisals

Two trained authors (YTC&JP) screened the titles, abstracts, and full-texts independently. A total of 21 of the 11336 articles were included [7, 12, 13, 1734] (Figure 1, Table 2). All studies meeting the eligibility criteria were appraised independently by three authors (YTC, JP, and LJS) through quality assessment tools (Table 3). The evaluation results are shown in Table 4.

2.1.3. Interventions Extraction and the Preliminary Bundle

Evaluated evidence levels of related interventions were extracted by criteria of levels of evidence reported in the Canadian Best Practice Recommendations for Stroke Care [18]. The main interventions from 21 articles were extracted to form the preliminary bundle.

2.2. Delphi Study

A Delphi study was conducted to obtain experts’ suggestions on the practicability and maneuverability of the bundle so as to improve the preliminary bundle further.

2.2.1. Participants

The number of experts can vary according to the size and the scope of the research involved. Generally, it is advisable to be around 8–20 experts [38]. Considering that the scope of the whole research is specific (i.e., poststroke psychological management) and the interventions included in the preliminary bundle through the evidence summary were relatively common, it was proper to recruit ten clinical experts specialized in stroke and psychological rehabilitation.

2.2.2. Delphi Survey Rounds

In the first round, each expert was told about the research question and provided with a three-part structured consulting questionnaire by e-mail. Based information of each expert was collected in part I, including age, gender, primary department, experience of dealing with psychological symptoms after stroke, and potential conflicts of interest. Part II included a list of interventions, and experts were asked to rate scores for the importance of the specific items (i.e., 56 items from the preliminary bundle interventions including “health education,” “emotion relieving,” “sleep improving,” “function promoting,” and “interaction maintaining”) using a Likert scale of 1–5 (1—Not important to 5—Very important). There were text boxes as well in Part II so that experts could write their comments about the interventions being considered. In part III, experts were asked to make an evaluation for the familiarity and judgment basis of the research questions. Two weeks after the e-mail delivery, an electronic reminder was sent by the researcher to those who had not yet completed the first round survey. Before the next round, some items were revised, excluded, or added based on the scores and opinions of experts in the first round. In the next round, updated consulting questionnaires (part II) were sent to experts by e-mail. The process was the same as the first round.

2.3. Data Analysis

All experts’ based information and data in consulting questionnaires were sorted and recorded into Microsoft Excel firstly and then were performed using SPSS V.22.0 to generate descriptive statistics.

The reliability of experts is usually represented by the effective response rate and authority coefficient (Cr) [39]. Cr is determined by familiarity (Cs) and judgment basis (Ca) of the research questions [Cr = (Cs + Ca)/2]. Cs represents experts’ familiarity with the research question, and a Likert scale of 0.1, 0.3, 0.5, 0.7, and 0.9 (0.1—Not familiar to 0.9—Very familiar) was used. Ca represents the judgment criteria the experts are based on. The criteria’s influence degrees and score assignments are theoretical analysis (great-0.5; medium-0.4; small-0.3), working experience (great-0.2, medium-0.2, small-0.1), referring to literature (great-0.2, medium-0.1, small-0.1) and self-intuition (great-0.1, medium-0.1, small-0.1). Ca value is the sum score of these criteria.

The importance of the specific items in the consulting questionnaire was evaluated by experts using a Likert scale of 1–5 (1–Not important to 5–Very important). The average rate score is presented as mean ± standard deviation (SD), and coefficient of variation (CV) defined as the SD divided by the mean is used to describe the relative dispersion degree of the items’ importance evaluation from experts [40]. The lower the CV value, the higher the coordination degree of experts’ opinions. There have been no universally accepted criteria for consensus in a Delphi study [4143]. This study defines that when the mean score for the importance of item is all equal or above 3.50 and CV is all equal or below 0.20, the consensus is reached [44]. In other words, the Delphi survey can finish when all items in the consulting questionnaire are met the two criteria above. The Kendall coefficient W test was adopted to evaluate the consensus on agreement among the experts as well. A two-tailed value of <0.05 was considered statistically significant.

3. Results

3.1. The Reliability of Experts

The age of experts was 43.10 ± 6.28 years old, and the length of working in this field was 17.80 ± 7.38 years. All experts had no conflicts of interest. The effective response rates of the 2 rounds were 100%. The mean value of the expert authority coefficient (Cr) was 0.85 (Table 5). The authority of the experts in this study was relatively high, and the results were thus trustworthy.

3.2. The First Round

Of the 56 items rated by experts in the first round, one item’s mean score and seventeen items’ CVs did not meet the criteria (Table 6). The relative dispersion degree of the items’ importance evaluation from experts was high. Kendall’s coefficient of concordance (W) was calculated to be 0.31 (). What’s more, some experts suggested that the structure of the questionnaire should be adjusted so as to be more clear for clinical professionals and patients to understand and implement in the future. A care bundle is a set of self-evident good-practice-based interventions for a defined patient population and care setting. It was generally a set of three to five evidence-based practices and there was no need to reach every aspect. Therefore, the questionnaire was greatly modified. The main five interventions were changed from “health education,” “emotion relieving,” “sleep improving,” “function promoting,” and “interaction maintaining” to “tailored diet,” “acupressure,” “mindfulness meditation,” “physical exercise,” and “maintaining interaction.” Eighteen existing items were merged according to the experts’ comments (e.g., health education was incorporated into maintaining interaction actually). Ten items such as “medication education” and “color emotion management” were deleted because of great heterogeneity in experts’ rate scores (i.e., CV value). And one item (i.e., home visits per month) was deleted because it is unavailable for professionals to have home visits for patients who lived in different provinces far away from the hospital. A new expert-proposed item was added. Recreation rehabilitation such as playing chess or cards, reading newspapers, and playing games was proposed to be added by some experts. Though there was no high-quality evidence, the experts had concluded that there was a positive effect on improving the bad mood for stroke patients from their own experiences. Therefore, it was included additionally. As a result, the revised consulting questionnaire contained five interventions and 28 items before the second round.

3.3. The Second Round

The results showed that all mean scores for importance on items were all above 3.50 and CVs were all 0.20 or below in the second round (Table 6). Because the structure of the questionnaire changed a lot, Kendall’s coefficient of concordance (W) decreased to 0.28 (). But it was statistically significant. There was no additional comment offered by experts in this round. The consensus was reached, and the Delphi survey was finished. The bundle was improved and finalized. Five interventions (i.e., maintaining Interaction, tailored Diet, Acupressure, mindfulness Meditation, and physical Exercise) made up the iDAME bundle eventually.

4. Discussion

A care bundle integrated western and traditional Chinese medicine interventions together for stroke patients with psychological symptoms were successfully developed. Previous studies showed that each intervention of the iDAME care bundle could improve the negative emotions after strokes, such as depression and anxiety indeed. Through the two-round Delphi, the maneuverability of iDAME care bundle has been approved further.

As western humoralism considering human body was retained to be a container of four liquids (blood, phlegm, black bile, and yellow bile), and health was constituted by the state of equilibrium of these substances, traditional Chinese medicine (TCM) holds that health depends on the balance of Qi, blood, Yin, and Yang in the human body [45, 46]. Different states of Qi, blood, Yin, and Yang make people have different body constitutions (“Tizhi” in Chinese). Common classifications of TCM constitutions are neutral constitution, Qi-deficiency constitution, Yang-deficiency constitution, Yin-deficiency constitution, phlegm-damp constitution, damp-heat constitution, blood-stagnation constitution, Qi-stagnation constitution, and special diathesis constitution [47]. According to TCM constitution theory, all constitutions except neutral constitution (balanced constitution) mean disharmony and can be viewed as an individual’s susceptibility to specific disease or symptoms [48]. TCM believes that the constitution is partly genetically determined and partly acquired, and classifies individuals’ constitution into the above nine types based on Chinese medical theory, multidisciplinary studies, and clinical practice. Hence, the results can be used in disease treatment and rehabilitative care [46, 49, 50]. TCM constitution is a comprehensive and relatively stable trait in the aspects of morphological structure, physiological function, and psychological state, which determines an individual’s susceptibility to a certain pathogenic factor and its tendency of pathological changes [47]. Various types of constitution show various characteristics. For example, patients of stagnant Qi constitution are mostly thin and often feel gloomy or depressed, easy to be nervous, anxious, and sensitive [49]. Patients with different constitutions require tailored care for the prevention and cure of poststroke psychological symptoms. Therefore, the iDAME care bundle includes TCM constitution theory in part of interventions, and patients with different constitutions are given specific care.

Nutrients are essential for the optimal production of neurotransmitters affecting mood [51]. Mediterranean diet pattern and other nutrient elements included in the bundle (e.g., deep-sea fishes rich in n − 3 polyunsaturated fatty acid) have been proved to be beneficial to the improvement of negative emotions [23]. Due to diverse TCM constitutions, the bundle offers choices to different patients. For instance, patients of the Qi-stagnation constitution are encouraged to feed wheat, oranges, radishes, and so on. Sleep disturbances are frequently reported in stroke patients and associated with the mood state after strokes [52]. Hence, insomnia patients are given a sleeping-improving diet to improve sleep quality according to their TCM constitutions. For example, insomniac patients of the Yin-deficiency constitution are suggested to drink some milk and honey water, while patients of the Yang-deficiency constitution are suggested to eat some walnuts and dried longans. An individualized diet helps patients adjust negative emotions better.

Acupressure is one of the Complementary and Alternative Medicine (CAM) modalities, which has the potential for promoting sleep quality and psychological wellbeing [53]. It is a stimulation of acupuncture points using fingers, palms, or devices to balance body energy, maintain good health, and prevent illness [54]. Acupoint stimulation regulates the autonomic nervous system by reducing sympathetic activities and increasing parasympathetic activity, which can reduce the stress response and induce relaxation. Regulating the autonomic nervous system, hormonal factors and neurotransmitters may have biological effects on inducing sleep, calmness, and feelings of psychological wellness [53]. Common acupuncture points for stroke patients with bad mood are Shaohai, Kunlun, Taichong, and so on. The bundle offers additional choices to different TCM constitution patients (e.g., Guanyuan, Qihai, and Shenque for Qi-deficiency patients) as well. Given that music engages in a variety of brain areas involved in emotion, musical interventions have been used to solve mental problems [55]. Moreover, music helps patients get into a relaxed state more easily, so specific pieces of music are supplied to patients during acupressure (e.g., Mozart’s Serenade).

Mindfulness meditation can be defined as a form of spiritual training that aims to improve core psychological capacities, such as attentional and emotional self-regulation [56]. Improvements in emotion regulation associated with mindfulness meditation have been investigated through various approaches [57]. Mindfulness meditation works by strengthening prefrontal cognitive control mechanisms and thus downregulates activity in regions relevant to affect processing, such as the amygdala [56]. Stroke patients with psychological symptoms calm down and focus on what they are doing better.

Physical exercise is indispensable. Studies showed that appropriate exercise could improve mental health status and prevent mechanisms that mediate the association between depression and metabolic syndrome [12, 5860].

Last but not least, maintaining interaction is an important part of this bundle, which makes patients and caregivers solve the problems more easily with help from medical professionals. They can talk with nurses or ward mates to relieve stress and emotions through social software (Wechat texts and videos chatting, group chatting, etc.). Medical professionals could provide information supporting system for patients and caregivers. It is also the guarantee of the entire bundle’s accomplishment after discharge. Patients will be reminded of completing interventions and be helped when they are in trouble. The use of a network platform can improve the compliance of patients [61].

iDAME care bundle is an entirety including physical and psychological interventions for the poor mood of stroke patients. The whole bundle is relatively simple and practical. However, there are still some limitations. Firstly, the quality of evidence included is not pretty high, there is a risk of bias to some extent. Secondly, the whole bundle is content-rich, integrating Chinese and western interventions, so the compliance of the patient is unknown. Thirdly, though each intervention of the bundle has been shown to be effective in improving the psychological symptoms after stroke, the combination effect has yet to be verified.

5. Conclusions

iDAME bundle was developed through a systematic evidence summary and two-round Delphi study. It is an entirety including physical and psychological interventions. The scientificity of iDAME bundle has been approved. Furthermore, an implementation evaluation is required to confirm the feasibility of clinical.

Data Availability

The data used to support the findings of this study are available from the first author upon request.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Authors’ Contributions

All authors contributed to study conception and development. YTC, JP, and LJS piloted the search strategy and contributed to the data extraction and analysis. YTC and ZL contributed to the study design and coordination of Delphi. YTC performed statistical analysis and drafted the paper. All authors provided critical revisions and contributed to the editing of the paper for important intellectual content.

Acknowledgments

Funding was provided by the Fuxing Nursing Research Fund of Fudan University (FNF201905).