Abstract
Purpose:
The purpose of the study was to explore the barriers to and facilitators of self-management among older adults with type 2 diabetes mellitus (T2DM).
Methods:
This study followed a qualitative descriptive methodology. Older adults with T2DM living in Jinan, Shandong Province, China were recruited using purposive sampling. Information saturation was used to gauge the sample size. Semistructured interviews were conducted with 23 participants. The data analysis was guided using a thematic approach. Themes were inducted from the interview data undergirded by the cumulative complexity model.
Results:
The key findings of this study are presented in terms of 2 themes: facilitators of T2DM self-management and barriers to T2DM self-management. Each theme has subthemes, including that having family members with diabetes, having family members who are health care professionals, and visual cues were factors for good self-management practices by older adults with T2DM. Conversely, poor health status (ie, multimorbidity and lower-limb dysfunction and pain) and intergenerational care responsibilities were identified as barriers to effective self-management. Moreover, the use of media resources, especially traditional media, was found to both assist and hinder participants in their self-management practices.
Conclusion:
The findings from this study can inform new research to build on existing self-management promotion programs and restructure existing services to improve the self-management of older adults with T2DM. With the increase in the number and types of media outlets, our finding implies that researchers or clinical practitioners may develop strategies to leverage media resources to enhance the self-management of diabetes among older adults with T2DM.
In 2021, 536.6 million people globally were living with type 2 diabetes mellitus (T2DM), and of those, nearly 50% were older adults. 1 China has the largest number of patients with T2DM in the world, and 74.22 million of those patients are older adults. 2 T2DM often leads to serious complications, such as cardiac dysfunction, kidney damage, and so on, and these problems are linked to high mortality, poor functionality, and increased use of health care services. 3 Older adults have a greater risk of developing T2DM complications than other age groups. 4
Self-management is the main approach that people with diabetes use to prevent T2DM complications. Self-management includes strategies such as exercising, maintaining a healthy diet, taking medications correctly and on time, and monitoring blood glucose regularly. 5 However, diabetes self-management presents specific challenges, 6 and older adults often demonstrate unsatisfactory practices. One of the reasons for poor self-management is that individuals have changing needs across the care continuum, as proposed in the “timing it right” theory, 7 and prescriptions tailored to patients during their clinical visits may not well fit their needs for practice in daily life. Multiple studies have been undertaken to address this issue and to understand patients’ needs during their self-management period and thereafter. As a result of these studies, several self-management promotion programs have been developed and tested. 8 These programs typically were developed by health care professionals and are based on empirical evidence in addition to the core components recommended by the World Health Organization, for example, a healthy lifestyle, medication for blood glucose control, and so on. 9 However, effective outcomes for these programs are mixed. For example, a review of self-management promotion programs for patients with multimorbidity (including T2DM) reported that existing programs often attempted to promote self-management through case management or enhanced multidisciplinary teamwork, but it reported no significant improvement in the patients’ self-management practices. 10 A more recent trial reported significant improvement from a self-management promotion program 11 by using a customized self-management program and home visits from a registered nurse or registered dietitian. In addition, self-management promotion programs should be tailored to individuals’ characteristics (eg, health condition) and their everyday life and work, and these variables differ substantially across age cohorts, but self-management programs that are tailored specifically to older adults with T2DM are limited in current literature. To date, there has been limited tailoring of these self-management support programs for older adults. All these issues thwart the effectiveness of health care providers in recommending self-management promotion programs for older adults with T2DM. As a result, studies continue to report poor glycemic control by older adults with T2DM. For example, a study in China showed that the control rate of T2DM in individuals age 65 years and older was only 41.33%. 12 Therefore, research is needed to further explore the potential factors that affect the self-management of older adults with T2DM at multiple levels, from the environment to the individual.
Cumulative complexity model proposed that factors affecting patients’ adherence to management programs can be explored from the overall perspective of patient burden and resource balance. 13 This model provided the theoretical underpinning for this study. This study aims to understand the life experiences reported by older adults with T2DM, with a focus on the barriers to and facilitators of their self-management practices. The findings may provide information for health care professionals to restructure the content of programs and resources by building on existing self-management promotion programs, promote the patients in daily life to perform the self-management process, and provide guidance and facilitate good self-management practices among older adults with T2DM.
Methods
Design
This study used a qualitative descriptive methodology. Characteristics of qualitative descriptive methodology include having preliminary assumptions, drawing from a naturalistic perspective, and examining a phenomenon in its natural state; the interpretation of the content is of low inference. 14 The common themes of this study inducted from the interview data are undergirded by the cumulative complexity model. The model describes factors that influence patients’ self-management adherence at a holistic level from the perspective of patient burden and resource balance, and this study categorized patients’ self-management experiences into two main themes: barriers and facilitators. This study aimed to explore the factors that influence the self-management practices of older adults with T2DM in a naturalistic state of life as comprehensively as possible. Therefore, the qualitative descriptive methodology fits well with the objectives of this study. The data of this study were collected via semistructured, in-depth interviews and analyzed via thematic content analysis.
Participants
The use of maximum variation sampling has been noted for its effectiveness in obtaining comprehensive insights and in-depth information.14,15 Accordingly, in this study, purposive sampling was employed to obtain a maximum variation sample that would offer rich insights into the research focus: identifying the barriers to and facilitators of self-management practices in older adults with T2DM. Participants were recruited from a single community located in Jinan, Shandong Province, China. Potential eligible participants, who were documented in the community health care information system, were referred by the community doctor. The study considered heterogeneity in the following aspects of participants during recruitment: gender, age, living status, duration of diabetes, comorbidities status, and glycemic control. Participants who met the criteria and differed in terms of the duration of their diabetes and self-management practices were invited to participate. The inclusion criteria were that candidates (1) were age 60 years or older, (2) had been diagnosed with T2DM by a physician for 1 year or more, and (3) were able to understand Mandarin. The exclusion criteria were that candidates had (1) a diagnosed mental disease and/or (2) moderate to severe cognitive impairment (ie, a Mini-Mental State Examination score of 20 or less). Theoretical saturation was used as the stopping criterion. No incentives were provided to any participants.
Data Collection
A semistructured interview guide informed by the literature was drafted and then later revised based on consultations with a professional with expertise in qualitative study design and a nursing professional who was working in an endocrinology department. The author RYX, who is a graduate student, conducted all interviews. She was not known to the participants of this research before undertaking the study. Depending on the participant’s preference, the interviews were conducted either in person at the community health center or via mobile phone calls. Prior to initiating the main study, the interviewer conducted pilot interviews with 3 participants. The study used the pilot interview to verify the appropriateness of the interviewer’s skills, based on which, the qualitative expert would provide refinements to the interviewer’s skills. These 3 interviews were also included in the final data analysis. Participant consent was obtained prior to each formal interview, followed by collection of basic information. Probes specific to participants’ comments were used to complement the interview guide when necessary, and any requests by the participants to clarify confusing words or sentences were documented. Each interview lasted approximately 30 minutes and was recorded using a Philips VTR9000AI smart recorder. The questions in the interview guide are as follows:
How do you manage your diabetes?
Do you encounter any problems during your self-management practice?
Do you have any successful diabetes management experiences to share with us?
Do you need additional assistance to facilitate your diabetes self-management?
Is there anything else you would like to add?
Data Analysis
The recordings were transcribed verbatim by the recorder and checked for accuracy by RYX and CYW. Transcripts were further validated via member checking. Member checking consisted of returning the transcribed interview transcript to participants via phone retelling and asking them to provide feedback regarding whether there was further insight or interpretation. RYX and CYW made additions to the transcripts based on participants’ feedback to ensure the accuracy and trustworthiness of the data. After member checking, transcripts were uploaded to Nvivo 12.
A thematic analytic approach was used for data analysis. This study started with a coding framework developed from the cumulative complexity model and involved the process of applying the coding framework to the data and adding and modifying codes based on the data.
Familiarizing with data
Data analysis began with CW and RYX reading all the transcripts several times to achieve immersion and to obtain a sense of each participant’s story.
Generating initial codes
CW and RYX independently coded 5 interview transcripts along with the audit trail and developed a codebook that was finalized based on discussion with all the team members. Guided by the codebook, CW and RYX completed coding of the transcripts. Any new codes that emerged were discussed with all the team members until agreement was reached.
Searching for themes
Codes were collated into potential themes, all data relevant to each potential theme were gathered.
Defining and naming themes
Ongoing analysis was conducted to refine the specifics of each theme and the overall story that the analysis tells and to generate clear definitions and names for each theme.
Reviewing themes
Checking was conducted to determine if the themes work in relation to the coded extracts and the entire data set, generating a thematic map. The process of grouping codes into similar themes likewise was discussed with all team members until agreement was reached. The generated themes were augmented using illustrative quotes from the participants.
Rigor
This study adopted the recommended guidelines by Lincoln and Guba 16 to ensure the trustworthiness of this study. Dependability was assured by having the same key researcher conduct all the interviews. The description of participant demographics and verbatims facilitated transferability. The data analysis was performed jointly by the first two authors, CW and RYX and the results were then discussed by all members of the research group to ensure confirmability. Illustrative quotes linked to the theme and each category in the results enhanced the study’s credibility. This article was written in accordance to the Standards for Reporting Qualitative Research. 17 This study has ongoing reflection on the potential impact of personal biases and experiences on the research process to ensure rigor, and this reflexivity also contributes to enhancing the study’s credibility.
Ethical Issues
The ethical oversight of the qualitative study was obtained from the Institutional Review Board of the researchers’ university. Prior to each interview, participants were provided with verbal descriptions of the study, and informed consent was obtained. Interviews were anonymized, and each participant was provided with a code number.
Findings
A total of 23 participants with a median age of 74 years old (interquartile range 68-78) were enrolled in this qualitative study, and more than half (15/23, 65.2%) were female. Most participants (21/23, 91.3%) were married, and of those, 17 participants lived with their spouses only; 4 lived with their spouses together with their grandchildren. Most participants had a high school diploma and above (21/23, 91.3%), a prescription for medication (19/23, 82.6%), multimorbidity (17/23, 73.9%), and diabetes-related complications (19/23, 82.6%). The median duration of their diabetes since diagnosis was 15 years (interquartile range 6-20). Nearly 40% of the participants had poor glycemic control. More than 70% (17/23) of the participants were interviewed by phone instead of in person. Demographic descriptions of all the participants are shown in Table 1.
Demographic Data of Participants (n = 23)
The key findings of this study are presented in terms of 2 themes: facilitators of T2DM self-management and barriers to T2DM self-management. Each theme has subthemes that are complemented here by statements of the study participants (Table 2). Note that the translations from Chinese to English of the participants’ own words are intended to be as true as possible to their expressions and responses as recorded in the transcripts.
Themes and Subthemes
Abbreviation: T2DM, type 2 diabetes mellitus.
Theme 1: Facilitators of T2DM Self-Management
Four facilitators were found to be associated with T2DM self-management in older adults : family members with T2DM, family members who were health care professionals, visual cues (reminders to take medications), and helpful media resources (see Figure 1).
Factors that influence the self-management of older adults with type 2 diabetes mellitus.
Subtheme 1: having family members with T2DM
T2DM is hereditary, and T2DM patients may have 1 or more family members with diabetes mellitus (DM). The negative and positive self-management strategies and experiences of the participants’ family member(s) may model sets of behaviors that could motivate older adults with T2DM to either (1) practice good self-management to avoid the poor experiences or mistakes that their relatives with DM have made and thus avoid complications or (2) follow similar self-management strategies that have been successful for their relatives with DM and live longer lives.
My brother and sister are both people with diabetes. My brother couldn’t control himself on diet and got complications; he died in his thirties because of acidosis. In contrast, my sister ate a healthy diet and took regular walks along the moat, in that way, she got her diabetes well managed. Through their experiences, I pay more attention to regular glucose monitoring and other management practices of diabetes. (P9, female, age 77)
Subtheme 2: having family members who are health care professionals
Among the participants, it was noted that a few individuals were fortunate to have a medical professional within their family network, such as their children and siblings. This proved to be a valuable asset because patients can easily get in touch with medical professionals. These medical professionals are aware of patients’ daily life behavior and being well-versed in medical knowledge, could offer long-term and continuous personalized guidance. This included dietary recommendations and explanations of the latest research and medical findings, ensuring that the participants made informed dietary choices.
I can strictly follow the prudent diet of diabetes because my daughter is a doctor and she always helps me create healthy-eating plans. (P4, female, age 78) My brother is a doctor in my hometown, and he often instructs me to watch my diet. He told me to eat something like millet and coarse bread, to drink smashed bean beverage, and to avoid sweet foods and beverages, and said that would be good for my glycemic control . . . that is why I could well follow a healthy diet in my daily life. (P6, female, age 66)
Subtheme 3: visual cues (reminders to take medications)
Remembering to take prescribed medication on time posed challenges for older adults with T2DM, particularly due to common factors such as forgetfulness or distractions that can disrupt a daily medication schedule. In response to these challenges, patients proactively attempted to seek out strategies. The common strategy they used was pairing a specific action with frequently encountered visual cues, including placing their daily medication in a conspicuous spot at home or setting a glass of water on the tea table as a constant reminder to take the prescribed medication in the right amount and at the designated time.
It’s very easy to forget to take medications, I missed often at the very beginning. But gradually, I found a way to avoid missing. I place my daily medication in a conspicuous location at home, and in this way, I won’t forget to take the medicine on time. (P12, female, age 72) Forgetting to take the medicine is very common at the beginning, and I know that is detrimental to my glycemic control. To avoid this issue, I tried multiple ways. Nowadays, I always put a glass of water on the tea table in the living room. When the glass is empty, it means that I have taken the medicine, and vice versa. That’s a very useful way for me to take medicine on time. (P19, male, age 91)
Subtheme 4: helpful media resources
Endocrinologists typically provide patients with recommended guidelines for a healthy diet and exercise to help manage T2DM. However, older adults with T2DM often look for other resources to supplement their prescribed diet and exercise plan. Resources found through traditional media outlets (TV, radio, books, newspapers, magazines, etc) were reported by older adults with T2DM as being useful for enhancing their self-management practices.
I have a long medical history of diabetes, and I keep learning information on diet and exercise from TV programs, the radio, and community presentations. (P10, male, age 80) I subscribed the Health News [the authoritative and influential national health newspaper] and learnt a lot of information on diet and exercises from it to facilitate the management of my diabetes. (P11, male, age 78) I pay more attention to the dietary management of T2DM, and get a lot information from books. (P14, male, age 78)
Theme 2: Barriers to T2DM Self-Management
Three barriers to T2DM self-management were found to impede older adults with T2DM self-management practice: poor health conditions, intergenerational care responsibilities, and misleading media resources (see Figure 1).
Subtheme 1: poor health conditions
Multimorbidity, lower-limb dysfunction, and pain were prevalent among older adults with T2DM. Older adults with T2DM often experience conflicting results and complexities with regard to their various treatments and prescriptions (ie, polypharmacy) for their multiple health conditions, including conflicting treatment options and the unclear sequence of multimorbidity management. Such issues placed significant self-management burdens on older adults with T2DM.
I have diabetes, hypertension, and coronary heart disease. I think hypertension and coronary heart disease are more important than diabetes. . . . I do not dare to engage in physical exercises because of them, and that means I cannot well follow the exercise prescription of diabetes. (P2, female, age 77) I have diabetes and hypertension, and there are too many medications to take. Hypertension control is my priority; therefore, I only take medications prescribed for it. (P3, female, age 76)
Lower-limb dysfunction and pain also can compromise the capacity of older adults to engage in regular exercise, which impeded their ability to adhere to the prescribed exercise recommendations from physicians.
The physician told me that eating well and being physically active were required to well manage my diabetes, but it was hard for me to do regular exercises because I had weak legs. (P4, female, age 78) I have knee pain and cannot walk 30 minutes at a time. I feel also inconvenient to engage in other physical activities. (P6, female, age 66) I have severe leg pain now and could not engage in regular exercises as prescribed. (P15, male, age 78)
Subtheme 2: intergenerational care responsibilities
In China, older adults often take on the responsibility of caring for the next generation, and caregiving consumes much of their time and energy. Older adults with T2DM who live with grandchildren or periodically are visited by grandchildren often reported that intergenerational care responsibilities came at the expense of their own health and put the self-management of their diabetes at risk.
Life is changing all the time, you know . . . my grandchildren periodically visit my family and live with us for quite a while, and I cannot spare time to do exercises as I usually do to take good care of them. (P5, male, age 65) I spend most of my time looking after my grandson and cannot walk outside as I usually do . . . I don’t have time to go to clinics or hospitals to have the glucose test, either. (P21, female, age 68) When taking care of my grandson, I often forget to take my medicine on time. (P22, male, age 74)
Subtheme 3: misleading media resources
Medications should be prescribed by an endocrinologist and in accordance with standard guidelines. However, older adults with T2DM continue to seek new or alternative medication from media resources that they believe will treat or even cure their diabetes. Misleading information (eg, medications that are advertised as being a cure for diabetes) entices patients to place an order for such “wonder drugs.” However, such fraudulent medications fail to help them improve their glycemic control.
I often check commercial ads on new drugs for T2DM that airs on TV but notice that most of them are not reliable. They often provided information that T2DM would be cured under the treatment of the drugs, which would entice the users to place the order. But you will find that the drug does not work at all, and you cannot return the purchase and get the money back. I learnt that from my own experience. (P13, male, age 64) Jinan radio station [a local radio station] broadcasts lots of commercial ads for T2DM, and I am its loyal listener. The propaganda of these ads was fascinating, and that enticed me to buy Tang Yi-Kang [a drug for T2DM]. I bought a lot to get the bigger discount, and I could use them for more than 1 year. But soon I felt deceived because it did not help control my blood sugar at all . . . I must eat them up because I cannot return the purchase and get the money back . . . they cost me a lot. (P18, female, age 83)
Discussion
Older adults with T2DM who have lived with T2DM for many years continue to learn about diabetes management strategies that they can adapt into their life and practice. Based on the self-reported experiences of living with T2DM, the participants in this study reported multiple factors as being relevant to their T2DM self-management: Facilitators to T2DM self-management include having family members with T2DM, having a family member who is a health care professional, and utilizing visual cues to help them remember to take their medications on time; barriers to T2DM self-management include poor health conditions, intergenerational care responsibilities, and media resources.
T2DM is hereditary, and this study found that older adults with T2DM were motivated to practice good self-management by witnessing both the positive and negative self-management experiences of family members with diabetes. This finding is consistent with the empirical evidence and theoretical proposition. Cunningham et al 18 reported that people with diabetes construct their illness representations (ie, their beliefs about and expectations of their illness) through their exposure to family experiences. Also, as demonstrated in the commonsense model of the self-regulation of health and illness, illness representations can lead to coping responses. 19 In addition, this study found that intrafamily health care professionals can be a source of informational social support (ie, they can provide their older diabetic relatives with information about healthy eating plans and exercise plans that can lead to improved self-management). These findings indicate that health care professionals should evaluate the communication and dialogue between patients and intrafamily stakeholders (ie, family members with diabetes and family members who are health care professionals) during clinical visits or home visits and mobilize their interaction when necessary. Such efforts may benefit self-management in older adults with T2DM.
Despite the recommendation of using reminder cues and memory aids as effective strategies for improving self-management for people with diabetes, 20 the older adults with T2DM who were participants in this study had devised ways on their own to use visual cues as reminders to take their medication on time. One possibility underlying this phenomenon is that health care professionals may have failed to provide patients with enough options about various reminder cues and necessary information, either because they did not feel the need to do so or because they found it challenging to devote time to this effort. In addition, health care professionals may have failed to provide patients with recommendations for appropriate and affordable memory aids. For example, smartphone apps,21,22 pill organizers, and pill poppers (a memory aid for medication dose tracking) are available options, but they can be costly and are not reimbursable through China’s basic public health insurance scheme. Therefore, the costs associated with such memory aids may lead older adults with T2DM to explore other feasible options rather than purchasing the physician-recommended ones. Future studies should try to identify the true reasons that underlie this phenomenon and implement strategies to eliminate potential barriers to gaining access to feasible reminder cues and memory strategies for older adults with T2DM.
Multimorbidity is a pressing health issue among people with T2DM. This study found that 73.9% of the participants had multimorbidity, which presented challenges for their self-management due to conflicting treatment regimens and issues of polypharmacy. Although current evidence favors a multidisciplinary team approach to provide self-management education to patients with T2DM, 23 the considerable time, effort, and financial resources required for health care professionals to implement multidisciplinary team meetings limit the accessibility of such services. 24 In addition, this study found that older adults with T2DM often struggle with lower-limb function decline and pain, which limits their capacity to follow exercise programs prescribed for glycemic control. 25 Future studies should address multimorbidity and lower-limb dysfunction and pain among older adults with T2DM to develop feasible interventions at scale.
Rooted in the Confucian cultural tradition of familism and influenced by government initiatives to reap the second demographic dividend, more and more older adults in China are providing support and care for their grandchildren. 26 In this study, 17.4% (4/23) of older adults with T2DM were the leading family members who provided care for their grandchildren. Older adults with T2DM who provide custodial care or occasional care to their grandchildren often compromise their diabetes self-management practices. This finding is in contrast with findings from other studies that find that grandparents who provide care for grandchildren often experience both mental and physical health advantages. 27 To address the issue of providing care for grandchildren interfering with the self-management in older adults with T2DM, future research could consider employing Knafl and Deatrick’s 28 family management style framework to analyze the patient’s family’s management styles. It helps understand how tasks and activities are organized and coordinated within the family and will enhance our ability to develop and customize interventions for promoting optimal family adaptation to illness. It is also important to consider empowering grandchildren to provide self-management reminders for their grandparents so that they can play a meaningful role in the family dynamic. Future research should continue to explore strategies to reverse the adverse effects of taking care of grandchildren on self-management practices in older adults with T2DM.
Older adults with T2DM have ongoing learning needs regarding the self-management of T2DM, and they often seek answers from media resources rather than from health care professionals to satisfy those needs. Or they may seek answers from media resources to supplement the advice they receive from health care professionals. Information found in traditional media outlets, such as television, newspapers, magazines, books, and radio, can help older adults with T2DM devise their own dietary and exercise plans. However, false commercial advertising can be detrimental to their self-management, which is consistent with the finding in a recent study that misleading information in the media can influence viewers’ decisions and behavior. 29 The positive and negative effects of media resources have implications for health care professionals and government agencies. Health care professionals should continue to recommend only validated programs to their patients, and given that sometimes older adults are consumers of fake news, health care professionals also should remind their patients to be wary of pharmaceutical advertisements and implement strategies to improve the digital literacy of their patients. 30 Meanwhile, government agencies should take measures to crack down on drug and media companies for false and misleading advertising as an essential complementary strategy. In addition to fraudulent products that can be obtained from traditional media, rich and useful resources with acceptable quality are also available through social media. 31 Although the participants in this study were found not to use social media, the use of social media is reported to have increased dramatically among older adults.32,33 Therefore, studies are warranted that investigate social media exposure to diabetes-related information in people with diabetes and its impacts on their self-management. Such evidence may help in including relevant and useful media resources in programs and policies to facilitate older adults’ self-management practices.
Limitations
This study has 2 main limitations. First, the barriers to and facilitators of self-management may be reported by a single participant, but the combined effect of the factors on individual self-management could not be discerned in this qualitative study. Second, some of the findings of this study were based on the assumption that diabetic patients spend time seeking facilitators for their self-management, such as learning from family role models and obtaining medical advice from family members who are health care professionals; however, this assumption was not empirically tested. Third, we do admit the importance of taking observation into description, but the study was launched during the pandemic, and most participants preferred interviews via mobile phone call, which made observation impossible. Some participants accepted field interview, but epidemic prevention and control measures (ie, wearing a mask and social distancing) limited observing nonverbal cues.
Implications
The findings from this study have clinical implications. The findings may help health care providers better understand the factors that need to be assessed for their older patients with diabetes and the resources that need to be provided to their patients so that they can better self-manage their diabetes at home. This study’s findings also may help community health care workers who conduct home visits develop interventions that complement traditional prescribed education programs. In addition, the findings have 2 innovative implications for academic studies. First, future research should explore the moderators that buffer the adverse effects of intergenerational care on the self-management of older adults with T2DM. Second, future research should lead to recommendations and warnings that are relevant to traditional media in order to complement traditional education for older adults with T2DM and explore the feasibility of expanding such practice to social media.
Conclusions
Having family members with T2DM, having family members who are health care professionals, and visual cues (reminders) are essential factors for facilitating good self-management practices of older adults with T2DM. In contrast, poor health conditions and intergenerational care responsibilities are barriers to good self-management practices. Resources found in traditional media outlets can both facilitate and hinder good self-management practices of older adults with T2DM. The findings from this study can provide useful information to optimize both clinical and academic practice to improve the self-management of older adults with T2DM in their daily lives.
Footnotes
Acknowledgements
We want to thank all older adults who contributed to this study’s data collection. We also want to thank Marry Brown for providing language help.
Authors’ Note
Chen Wu and Ruiyang Xu contributed equally to this article.
Declaration of Conflicting Interests
The authors declare no conflicts of interest.
Funding
This work was supported by the National Key R&D Program of China (No.2020YFC2008802).
