Toward enhanced e-health literacy: A qualitative exploration of facilitators,barriers,and behavioral triggers among Chinese older adults with diabetes

Abstract

Objective

This qualitative study explores the facilitators, barriers, and behavioral triggers that shape e-health literacy (e-HL) among Chinese older adults with diabetes.

Methods

In this descriptive qualitative study guided by the Health Belief Model (HBM), we purposively recruited 19 older adults with diabetes. Semistructured interviews using an HBM-aligned topic guide were analyzed with reflexive thematic analysis to examine facilitators, barriers, and behavioral triggers of e-HL.

Results

Thematic analysis identified three themes-facilitators, barriers, and behavioral triggers, six subthemes and twenty-one categories. Facilitators comprised (a) perceived benefits (e.g. ease and convenience of digital health tools, improved self-management and quality of life, trust in institutional/governmental platforms) and (b) self-efficacy (e.g. previous successful digital experience, skill improvement methods, family or peer support). Barriers included (a) perceived barriers (e.g. operational difficulties with digital health tools, lack of time for digital health tools, reliance and preference for traditional methods) and (b) self-efficacy (e.g. lack of confidence and digital skills, and low perceived competence). Behavioral triggers reflected perceived severity/susceptibility (e.g. lifestyle impact post-diagnosis, and anxiety and worries about long-term complications) and cues to action (e.g. recommendations and encouragement from doctors or others, peer modeling and informal learning, health deterioration or new symptoms).

Conclusion

The HBM provides a robust explanation of e-HL in older Chinese adults with diabetes, with the model's constructs mapping to the relevant facilitators, barriers, and behavioral triggers. A scalable, three-component strategy targeting the HBM, focusing on the needs of older users and integrating at a system level can sustain e-health adoption and help bridge digital health divides in ageing populations.

Keywords

e-Health literacy diabetes facilitators barriers behavioral triggers older adults qualitative research

Introduction

Diabetes is one of the most prevalent and fastest-growing chronic diseases worldwide.¹ Its prolonged course can result in multiorgan dysfunction and failure, especially affecting the eyes, kidneys, peripheral nerves, heart, and blood vessels.² According to the International Diabetes Federation, around 537 million people were living with diabetes in 2021, a figure projected to rise to 784 million by 2045.³ China, which now has the world's largest diabetic population, is facing significant public health and socioeconomic challenges as a result.⁴ Globally, the prevalence of diabetes exceeds 20% among individuals aged 65 to 95 years, while remaining below 1% in those under 20 years of age.⁵ Consequently, the urgency of developing effective and targeted diabetes management strategies for older adults has been underscored by these age-related disparities.

Achieving optimal diabetes control in older adults hinges on sustained engagement in evidence-based self-care behaviors, such as regular blood glucose monitoring, adherence to medical nutrition therapy, participation in physical activity, and compliance with prescribed medication.⁶ In response to the evolving digital landscape, there has been an increase in digital health interventions for diabetes care in recent years, including mobile health (mHealth) applications, continuous glucose monitoring systems, decision support tools, online education platforms and telehealth services.⁷ Recognizing the potential of these innovations, the Chinese government has actively integrated digital health strategies-most notably through the “Internet + Medical Health” initiative-into the Healthy China 2030 agenda. The aim is to transform chronic disease management from a treatment-centered to a health-centered model.⁸

Central to the successful use of digital health resources is e-health literacy (e-HL), which is defined as the ability to seek, find, understand and apply health information from electronic sources in order to solve or manage health-related problems.⁹ Adequate e-HL can lead to improved behavioral and cognitive outcomes among older adults.¹⁰ Specifically, higher levels of e-HL have been associated with improved diabetes-related information-seeking, enhanced self-care practices, and more consistent self-management behaviors in individuals with diabetes.^11,12 However, uptake of digital health resources remains low among older diabetic populations in practice, with many exhibiting limited e-HL skills such as difficulty locating, interpreting and utilizing electronic health information.^13,14 These challenges are exacerbated by age-related cognitive decline, limited digital skills, and cultural barriers.^15,16 In China, older adults exhibit particularly low levels of e-HL, with disparities being further magnified by regional inequality, digital exclusion in rural communities, and local sociocultural norms.¹⁷

Moreover, socioeconomic disparities, urban–rural divides and uneven access to digital infrastructure exacerbate these inequalities in e-health.¹⁷ Despite the growing relevance of this issue, existing research has primarily employed quantitative methods to focus on demographic predictors or evaluate intervention efficacy. This approach offers limited insight into the subjective experiences and sociocultural interpretations of older adults regarding e-HL.^18,19 Notably, few studies have examined the cognitive mechanisms or behavioral decision-making processes underlying e-HL practices using robust theoretical frameworks.

In this context, qualitative research offers a significant methodological advantage in enabling an in-depth exploration of the factors that facilitate or impede behavior. The Health Belief Model (HBM), a widely validated framework in health behavior research, provides a comprehensive structure for investigating health-related decision-making processes. Originally developed to explain preventive health actions, the HBM posits that such actions are influenced by six constructs: perceived susceptibility, perceived severity, perceived benefits, perceived barriers, cues to action and self-efficacy.²⁰ The model has been extensively utilized in studies on diabetes management and digital health adoption, offering insights into behavioral facilitators and inhibitors.^21–23 Applying the HBM makes it possible to systematically identify the cognitive and motivational factors that facilitate or hinder the adoption of e-health technologies. Health care and digital health evolve within a complex adaptive system where multilevel interactions, couplings, and feedback loops shape individual perceptions and behaviors.²⁴ In parallel, the capabilities of artificial intelligence continue to expand and are increasingly integrated into care pathways in a complementary fashion. However, real-world effectiveness still hinges on users’ e-HL and contextual fit.²⁵ Building on this systems-aware view, we adopt the HBM to examine how e-HL emerges and changes among older Chinese adults with diabetes, thereby adding empirical insight into the joint dynamics of people, technologies, and institutions.

This qualitative study aims to explore the facilitators, barriers and behavioral triggers that enhance e-HL among Chinese older adults with diabetes, guided by the HBM. By focusing on the lived experiences, cultural values and belief systems of this population, the study will generate contextually grounded insights to inform the design of more responsive and effective e-HL interventions. The findings will be highly relevant to healthcare policymakers, practitioners and technology developers who are seeking to promote digital inclusion, improve outcomes for people with chronic diseases and optimize health service delivery for ageing populations in China.

Method

Study design

This study adopted a descriptive qualitative research²⁶ approach to analyze the facilitators, barriers and behavioral triggers for enhancing e-HL among Chinese older adults with diabetes. Semistructured one-to-one interviews were employed for data collection,²⁷ with adherence to the 32-item Consolidated Criteria for Reporting Qualitative Research checklist²⁸ ensured (Supplemental Table S1). The principles of informed consent and confidentiality were strictly followed during the interviews. The study adhered to the principles of the 1975 Declaration of Helsinki²⁹ and medical research ethics, and was approved by the Ethics Committees of Fuyong People's Hospital, Bao'an District, Shenzhen (KY-2024-110).

Participants

Participants were recruited through purposive sampling from the Endocrinology Department of a Grade A tertiary hospital in Shenzhen between December 2024 and January 2025. Participants were recruited based on the following criteria: patients aged 60 years or older who meet the 1999 WHO diagnostic criteria for diabetes,³⁰ are able to self-care and can read, write and speak Chinese. They must also be currently using or have previously used digital health tools related to diabetes (e.g. blood glucose management applications, smart blood glucose meters or online consultations, etc.). Exclusion criteria: not able to provide informed consent, cognitive impairment, unclear thinking and mental illness. During interviews, participants could refuse or withdraw if irresistible factors arose, such as an older person not understanding the interviewer and deciding to stop midway through the interview.

The sample size was determined based on data saturation, whereby no new themes emerged from the participants’ responses.³¹ A total of 19 older diabetic patients took part in the study. Three eligible patients declined to participate due to treatment conflicts or physical discomfort. The cohort had a mean age of 68.58 ± 5.92 years and consisted of 11 males (57.9%) and eight females (42.1%). Among the participants, 12 resided in urban areas (63.2%) and 7 (36.8%) in rural areas. The educational background was diverse: two participants had primary school education or below (10.5%), six had junior high school (31.6%), four had senior high school (21.1%), two had technical secondary school/junior college (10.5%), four had a bachelor's degree (21.1%), and one had a master's degree or above (5.3%). The average duration of diabetes was 6.53 ± 5.91 years. Their demographics and clinical characteristics are shown in Table 1.

Table 1.

Demographic and clinical characteristics of participants (n = 19).

Identifier	Gender	Age	Residence	Education level	Years of illness
A1	Male	64	Urban	Master's degree and above	3
A2	Male	63	Rural	Senior high school	7
A3	Male	65	Urban	Senior high school	5
A4	Male	66	Urban	Bachelor degree	1
A5	Female	72	Rural	Junior high school	3
A6	Female	75	Urban	Senior high school	4
A7	Male	66	Urban	Technical secondary school/junior college	1
A8	Female	82	Urban	Primary school and below	23
A9	Female	78	Rural	Junior high school	1
A10	Male	66	Urban	Bachelor degree	3
A11	Male	66	Rural	Junior high school	2
A12	Male	75	Urban	Bachelor degree	5
A13	Female	64	Rural	Junior high school	9
A14	Female	75	Rural	Junior high school	6
A15	Female	65	Rural	Technical secondary school/junior college	1
A16	Male	66	Urban	Junior high school	11
A17	Male	61	Urban	Senior high school	14
A18	Male	63	Urban	Bachelor degree	10
A19	Female	71	Urban	Primary school and below	15
		68.58 ± 5.92			6.53 ± 5.91

Note: The data present a heterogeneous single-center sample of older adults. Key parameters for comparison include gender, age, residence, education level, disease duration.

Theoretical framework

The HBM was employed to guide this study,²⁰ with the facilitators, barriers and behavioral triggers for enhancing e-HL among Chinese older adults with diabetes being analyzed. Key constructs encompass perceived susceptibility, perceived severity, perceived benefits, perceived barriers, cues to action and self-efficacy, which influence diabetes management and digital health adoption.^21–23

Tools

The research team developed and tested the interview topic guide based on extensive literature reviews and consultations with coauthors. Prior to the formal interviews, a free medical consultation was conducted in Jiedong District, Jieyang City, during which six patients who had been diagnosed with diabetes by clinical doctors were preinterviewed. Based on feedback from the participants and expert opinions in the preinterview stage, and guided by the HBM structure,²⁰ the final interview topic guide was created (Supplemental Table S2).

Data collection

The interviewers were a PhD student in public health, a Doctor of Pharmacy, and a nurse manager. All three had completed standardized training in qualitative methodology, including interview techniques and ethical communication. The research team consists of one Doctor of Public Health, one Doctor of Nursing, two Master of Nursing, two Health Education researchers and two endocrinology nurses, etc. There were three men and nine women, all of whom had extensive experience in clinical and digital health research, as well as teamwork.

The interviews were conducted using a semistructured format and an interview topic guide. The interviews took place in a room on the participants’ ward. Treatment times were agreed with nurses to avoid scheduling conflicts, and no one else was present except the participants and researchers. The interviewers provided comprehensive written and verbal information about the study to ensure informed consent. Basic participant information (gender, age, residence, educational level and years of illness) was collected. With permission, the interviews were recorded and lasted between 25 and 40 min. The interviewers maintained eye contact and used body language to encourage participants to recount their experiences of e-health. They restated and clarified responses to prevent awkward silences and promptly checked unclear statements with participants to ensure accuracy. After each interview, participants received health education manuals as a thank you. All recorded interviews were transcribed verbatim and anonymized.

Data analysis

Within 24 h of the interview, the researchers (SYC, ZKF and MC) transcribed the audio recordings verbatim and assigned each participant a unique number (A1-A19). Nonverbal cues were annotated in the transcripts to aid analysis. All recordings and data were promptly uploaded to the shared research database and backed up. The data were analyzed using NVivo software (version 12).

A thematic analysis was conducted using Braun and Clarke's approach,³² alongside line-by-line coding and other grounded theory techniques. The six phases of thematic analysis were employed: (1) the authors familiarized themselves with the data, (2) the data were coded and collated into potential themes, (3) searching for themes, (4) reviewing themes, (5) defining and naming themes, and (6) writing a report. The initial open coding was conducted directly on Chinese transcripts by a multilingual research team who were fluent in Chinese, English and Cantonese, in order to ground the thematic analysis in the original linguistic and cultural context. Thematic segments and illustrative quotes underwent a rigorous translation process, prioritizing conceptual accuracy over literal equivalence. This was followed by a review by the team. For culturally embedded expressions such as idioms, a foreignization strategy was often employed, retaining the original notion and adding brief clarifications rather than substituting them with potentially inaccurate Western concepts. To validate the accuracy of the analysis, summaries of the findings, including key translated quotes, were shared with the participants for confirmation to ensure that their experiences were authentically represented.

The HBM served as the guiding theoretical framework for the coding process, enabling relevant codes to be identified.²⁰ Themes and subthemes were refined by comparing external heterogeneity between themes and internal homogeneity within them. This process took into account both the coded extracts and the full dataset. In the event of coder disagreements, discussions were held first. If these did not resolve the issue, other researchers (JL and BL) were invited to act as arbiters. Throughout the analysis, SYC and ZKF reviewed and discussed themes in several meetings. Any discrepancies were addressed in team meetings with all coauthors. Field notes were also used to assist with the analysis. Finally, a full description of the observed phenomena was produced.

Rigor

The interviews and primary analysis were conducted by SYC, ZKF and MC. Both researchers had received formal training in qualitative methods and had five to 10 years’ experience in relevant clinical and fieldwork roles. All interviewers were highly proficient in using recording tools for data collection and drew on their experience of professional and personal interactions to minimize technical disruptions and create a more natural interview setting. Crucially, there was no prior treatment or professional relationship between the researchers and the participant cohort. All interviews were conducted outside of active clinical encounters, in a private ward room, to minimize power differentials and encourage open dialog. The research team comprised members with diverse backgrounds, all of whom were trained in qualitative research and geriatric care. This enabled the team to collaborate on developing and refining the topic guide through brainstorming sessions. To safeguard voluntary participation based on altruism, the study did not offer financial compensation. All recruitment communications highlighted the value of participants’ contributions to health knowledge and potential social benefits.

Thematic analysis was conducted through iterative team discussions until consensus was reached. Reflexivity was also maintained throughout the collection and analysis of data. This enabled the critical examination of potential researcher influences via several strategies, including the use of reflexive memos, bracketing, peer debriefing, dual independent coding of initial cases (with consensus and third-researcher adjudication when necessary), constant comparison and negative case analysis. A predefined thematic saturation criterion was applied. The text translation was handled by a bilingual team who were fluent in Chinese, English and Cantonese. Together, these measures enhanced the credibility and transparency of the research process and its findings.

Results

A thematic analysis identified three themes: facilitators, barriers, and behavioral triggers for e-HL literacy enhancement. These themes and their respective subthemes were then mapped onto the theoretical domains of the HBM, reflecting the underlying cognitive and behavioral orientations of the participants. All the themes and subthemes are summarized in Table 2. The dynamic HBM shows how contextual and system conditions, facilitators, barriers, and behavioral triggers jointly shape e-HL in older Chinese adults with diabetes (Supplemental Figure S1).

Table 2.

Overview of themes and subthemes and their mapping to HBM constructs.

Theme	Subthemes	Categories
Facilitators	Perceived benefits	Ease and convenience of digital health tools
		Improved self-management and life quality
		Trust in institutional/governmental platforms
		Comparative preference for digital health tools
	Self-efficacy	Previous successful digital experience
		Skill improvement methods
		Family or peer support
		Confidence-building factors
Barriers	Perceived barriers	Operational difficulties with digital health tools
		Lack of time for digital health tools
		Reliance and preference for traditional methods
		Excessive dependency on others for help
	Self-efficacy	Lack of confidence and digital skills
	Self-efficacy	Low perceived competence
Behavioral triggers	Perceived severity and susceptibility	Lifestyle impact postdiagnosis
	Perceived severity and susceptibility	Anxiety and worries about long-term complications
	Cues to action	Recommendations and encouragement from doctors or others
		Peer modeling and informal learning
		Health deterioration or new symptoms
		Demand for user-friendly e-health tool features
		Suggestions for improving elderly patients’ e-health literacy

Three overarching themes (facilitators, barriers, and cues to behavioral triggers), with subordinate subthemes, explain how e-health literacy emerges and changes in this population.

Theme 1: Facilitators for e-HL enhancement

Perceived benefits

Ease and convenience of digital health tools

Ease and convenience as the dominant perceived benefit driving e-health adoption among older Chinese adults with diabetes (reported spontaneously by 89.5% of participants), digital health tools were overwhelmingly valued for simplifying diabetes management routines, saving time, and reducing physical burden compared to traditional methods.

Before, checking blood glucose required my children to take leave to accompany me to the hospital—2 h in line for a 5-min test. Now with the ‘Yitangtong’ app, I do it myself at home in 3 min, and the data goes straight to the doctor. The saved time lets me play with my grandson. (A7)

My arthritic fingers shake when writing. After syncing my glucometer to the app … I just say ‘Record glucose 6.2'—no more pens or notebooks. (A17)

Improved self-management and life quality

Electronic health (e-health) tools fundamentally enhanced diabetes self-management efficacy and holistic well-being of participants, manifesting as clinical control, psychological relief, and social reintegration.

HbA1c used to swing between 7–10% like a rollercoaster … The app's trend graph exposed my snack mistakes—now I maintain 6.5% for 18 months, no hypoglycemia shocks. (A6)

Seeing ‘stable glucose’ alerts calmed my heart racing. I stopped checking my feet for ulcers hourly—finally slept through the night. (A17)

This app freed me from diabetic isolation, restoring my place at banquets and dignity as a teacher … Now when it buzzes during mahjong, friends call it my ‘guardian angel'—we laugh together, no longer hiding. (A18)

Trust in institutional/governmental platforms

Participants exclusively preferred government-affiliated e-health platforms, perceiving them as more credible, secure, and medically authoritative than commercial alternatives.

Private apps’ ‘gold-medal doctor’ titles look gimmicky! But government platforms list chief physicians from Peking Union Hospital and CMA—credentials my son verified … Impossible to fake…. (A1)

…The county hospital's official recipe plan had the director's photo on it! My HbA1c dropped from 9% to 7% in 3 months … After trying a viral diet from ‘X Health’ app? Glucose spiked to hospitalization levels…. (A2)

Comparative preference for digital health tools

Compared with traditional methods, some participants prefer electronic health care tools, and the preference of rural users is significantly stronger than that of urban users. Key drivers diverged: rural users prioritized geographic barrier reduction, whereas urban users valued time efficiency.

Algorithms detected hidden postprandial spikes—manual logs missed it… (A19)

Township clinics can't test HbA1c … Reaching city hospitals costs ￥200 via three buses. Now Shenzhen experts review reports by video—savings cover three months of meds! (A13)

Solo city trips felt like drifting … video consults avoid begging kids for leave. (A5)

Self-efficacy

Previous successful digital experience

Prior digital proficiency amplified e-health self-efficacy, yet rural users exhibited greater gains than urban counterparts. Scarcity in rural areas cultivated precision skill targeting—transferred voice-command mastery to digital health tools versus. Conversely, urban users’ broader digital exposure induced feature overload anxiety, undermining efficacy despite higher baseline literacy.

Mastering WeChat voice made glucometer dialect reports intuitive. My son said ‘Mom can manage diabetes if she sends 60 s voice notes’—he was right! (A11)

I use Taobao, but health apps have pop-ups like a maze—less reliable than my old glucometer with one button! (A18)

Skill improvement methods

Structured skill-training interventions increased e-health self-efficacy with rural participants exhibiting greater efficacy gains than urban counterparts due to context-adaptive learning designs. It demonstrates that resource constraints can catalyze pedagogical innovation. Context-adaptive methods outperform one-size-fits-all training.

City experts’ ‘standard training videos’ were useless! Then our village doctor led level-up drills: first taught glucose testing via WeChat voice, then used dialect recordings for app alerts—dared to operate solo in three days … No Wi-Fi? Printed ‘glucose recording rhymes’ on the stove wall—memorized while cooking! (A5)

Foolproof training’ at the senior center was genius! Turned app steps into fridge magnet flowcharts—muscle memory built in three days … But initial ‘expert webinars’? Feature overload made me want to flee! (A12)

Family or peer support

Family and peer support significantly enhanced e-health self-efficacy among older adults with diabetes. Crucially, support-source divergence emerged along urban–rural lines: Rural users relied more on family support, while urban counterparts leveraged more peer networks.

My son rhymes: “Test-Snap-Share” - even someone with a low level of knowledge like me knows it! (A9)

The dance team's WeChat group shared an ‘APP pitfalls guide’. Aunt Li taught me the ‘shake-to-check-glucose’ method. She's ten times more patient than my kids…. (A17)

Confidence-building factors

Context-adaptive confidence-building strategies increased e-health self-efficacy, participants achieving sustained behavioral engagement through three key mechanisms: micro-success scaffolding, social affirmation, and error normalization.

Nurse split ‘glucose check’ into 5 steps: power on → insert strip → blood sample → read → record. Practicing one step/day, mastered in 3 days! (A8)

Daughter posted my ‘7-day perfect logs’ in family chat—uncle praised ‘Big sis is so tech-savvy'—dared to try new features despite clumsy fingers! (A10)

Need built-in ‘glitch lab': intentionally crash interface to teach rescue—practice until fearless like bike-falling! (A1)

Theme 2: Barriers to e-HL enhancement

Perceived barriers

Operational difficulties with digital health tools

Participants reported significant operational barriers to e-health tool adoption, with interface complexity, input/output failures, and system opacity reducing self-efficacy and increasing abandonment risk.

Homepage crams 30 + icons—'glucose log’ hides like an ant! Eyes strained, fingers slipped, still needed grandson's help. (A10)

Good digital health tools should be like reading glasses—see clearly when picked up, no worries when put down … Current apps? Like threading needles blindfolded! (A12)

Glucose test on App A, reports on Platform B, consults on Mini-Program C—three phones can't cope, just handed all to daughter. (A15)

Lack of time for digital health tools

Participants generally reported that time-consuming factors like inflexible work hours, rigid childcare duties, and workplace pressures reduced the time available to focus on their health online.

I don't have a fixed time to get to work. It's late by the time I've had a shower and dinner, and then it's bedtime. (A11)

I don't have time for all that … my work is particularly busy at the moment. Every night, I have to go home and look for information and read books. I also have to do training for them and create some slides. (A10)

Sometimes I need to socialize, but after I get home and the kids start their homework, I'm pretty busy. Even if I do something for myself in the evening, I don't have much time or energy. (A1)

Reliance and preference for traditional methods

Some patients demonstrated strong preference for traditional diabetes management despite digital resource availability, prioritizing clinician-mediated epistemic trust, tactile reassurance, and cognitive congruence over digital health tools—significantly correlating with digital abandonment rates.

Now that we're still in the hospital, we need to follow the doctor's treatment plan … When we're discharged, we'll still need to know how to protect ourselves … There are doctors and nurses here now, which is an extra layer of protection. (A12)

There's so much on the internet that I'd rather go to a professional … That way, I won't have to search for so many answers. (A6)

Excessive dependency on others for help

Participants reported reliance on external assistance as a critical barrier to e-health adoption, triggering autonomy erosion, timeliness deficits, and relational strain. This dependency correlated with higher digital tool abandonment and lower self-efficacy.

Must wait for daughter to test glucose—if she works late, I skip 3 days… (A6)

Glucose 7.8—should I add meds? Won't act until son replies—pills sweated in hand uneaten… (A13)

Self-efficacy

Lack of confidence and digital skills

Digital skill gaps as the primary barrier to e-health adoption, reducing self-efficacy and triggering avoidance behaviors more frequently.

The app homepage is a maze—finding ‘glucose log’ needs 5 clicks, always hitting ads! Smashed phone, returned to paper. (A10)

Learning digital health tools should be like eating tofu … Current apps? Like forcing elders to swallow iron walnuts! (A6)

Low perceived competence

Low perceived competence in using digital health tools reduces adoption intent and increases avoidance behaviors. This deficit in perceived competence manifested as preemptive defeatism, exaggerating the importance of errors, and skepticism or mistrust of web-based information.

Seeing granny next door master apps—felt I had terminal dementia! Later learned she was a computer teacher…. (A16)

Icons like alien code—cloud symbol seemed like weather forecast, turned out to be ‘cloud sync’…. (A2)

I think there are too many unregulated adverts in videos or science videos on the internet. What is said is not true, so it is becoming less and less credible and can only be skimmed. (A3)

Theme 3: Behavioral triggers for e-HL enhancement

Perceived severity and susceptibility

Lifestyle impact postdiagnosis

Postdiagnosis lifestyle disruption as the primary trigger amplifying perceived severity/susceptibility, increasing threat appraisal by and motivating e-health adoption more than clinical factors.

I have practiced morning tai chi for 40 years — now my teammates say, ‘You'll fake a fall if you're hypoglycaemic…’ Losing this dignity is more deadly than high glucose. (A10)

I broke my ribs falling during my third night-time toilet visit. An hour on the floor taught me the terror of disability. I registered for telemedicine at dawn…. (A4)

Anxiety and worries about long-term complications

Complication-specific anxiety is the dominant trigger for e-health adoption, amplifying the perceived severity and susceptibility and increasing sustained tool usage. Anxiety was experienced through visceral somatization, temporal telescoping, and social contagion mechanisms.

Seeing halos around streetlights—frantically searched ‘retinopathy screening apps’! Doctor called me paranoid, but blindness scares me more than death! (A9)

Neighbor Zhang became a skeleton on dialysis—frequent urination makes my legs weak! Now urine analyzer daily—only sleep if data's normal. (A16)

Cues to action

Recommendations and encouragement from doctors or others

Participants adopted e-health following external recommendations, with physician cues triggering faster adoption than social or family cues. The effectiveness of the recommendations varied according to the credibility of the source, the delivery mode and the relational context.

Chief doctor printed app QR code on prescription—more binding than imperial decree! Registered on way home—can't betray trust. (A7)

Grandson as ‘family digital nurse'—doctor trains him, he trains me! (A13)

Peer modeling and informal learning

When people with diabetes watch peers successfully use digital health tools (like glucose apps), they are more likely to try it themselves. Learning from fellow patients was more effective than formal training.

Old Zhang shared his glucose records with our chat group. He went from ‘danger red’ to ‘safe green’ in just three months! If a 70-year-old can learn, then so can I! (A14)

I finally understood insulin math while teaching newcomers! Explaining to others locks it in your own brain. (A1)

Health deterioration or new symptoms

Participants adopted digital health tools within 72 h of experiencing new symptoms like numbness or vision changes. These physical warning signs were more effective than doctor's advice at prompting behavior change.

My doctor warned about foot ulcers for years—but when I couldn't feel my toes, I downloaded a foot-scan app THAT NIGHT. (A17)

Dropping my grandson's photo because of blurry vision terrified me—I signed up for online eye screening immediately. (A8)

Demand for user-friendly e-health tool features

Participants only adopted digital health tools when their design addressed issues such as complex menus or error messages. Long-term use is increased by simple, respectful interfaces.

When ‘404 Error’ pops up, I panic. Show me a ‘HELP’ video button right there… (A4)

Fancy features mean nothing if … I'm too scared to press anything. Simple = safe = used. (A3)

Suggestions for improving elderly patients’ e-HL

Peer-to-peer coaching improved e-health skills more quickly than traditional classes, while bite-sized “micro-learning” increased retention. Success came from respecting the life experience of seniors, rather than treating them as tech novices.

Grandson became my ‘phone coach'—later I taught my husband who praised ‘Grandma's so cool! Teaching beats learning. (A10)

Group leader posts daily micro-tasks: ‘Tap glucose log button today'! Small wins build big confidence. (A13)

Discussion

This qualitative study indicates that e-HL among older Chinese adults with diabetes is jointly shaped by three interacting domains-facilitators, barriers, and behavioral triggers-whose subthemes align closely with HBM constructs (perceived benefits and barriers, self-efficacy, perceived severity and susceptibility, and cues to action). The consistency between participants’ narratives and the HBM suggests that interventions which amplify salient benefits, reduce task frictions, and strengthen self-efficacy at moments of heightened perceived risk are most likely to initiate and sustain e-health behaviors. This conclusion accords with decades of HBM evidence showing that targeted messages and supports improve both initiation and maintenance of health behaviors when they directly address benefits, barriers, efficacy, and cues.²⁰

Participants identified “ease and convenience” as the predominant perceived benefit-valued for time savings, reduced physical strain, and simpler processes. This finding is consonant with guidance that diabetes-focused digital health technologies enhance self-management and clinical control when tailored to patients’ needs and circumstances (e.g. wireless glucose monitoring, continuous glucose monitoring, and diabetes apps).³³ Notably, respondents did not equate convenience with “more features” but with “fewer steps and clearer feedback,” a nuance emphasized by the WHO Global Strategy on Digital Health 2020–2025, which prioritizes person-centered, context-adapted design.³⁴ The observed trust premium for government-affiliated or hospital-run platforms—perceived as more authoritative and safer than commercial apps—reflects China's maturing “Internet + Healthcare” governance (2018–present), which legitimized internet hospitals and instituted supervisory rules.³⁵ Within this policy context, institutional affiliation appears to function as a potent credibility cue that lowers perceived risk and facilitates platform adoption.³⁶

Self-efficacy emerged as a critical secondary driver of uptake. Prior successful encounters with digital tools, scaffolded training, and support from family and peers reinforced perceived capability; evidence supports micro-learning and peer support as effective means of improving procedural mastery and sustaining engagement in chronic disease self-management.^37,38 Two context-specific insights extend existing literature. First, a “rural precision-skills effect” was observed: participants from rural areas, despite lower baseline exposure to digital tools, reported larger gains in confidence when training leveraged familiar platforms (e.g. WeChat voice messaging) and dialect-adapted prompts. This suggests that limited yet well-practiced digital skill sets can transfer effectively to health tasks when user interfaces resonate linguistically and culturally, echoing findings on older adults and speech interfaces that emphasize adaptable design (speech, format, language use, and voice-first interfaces).^39,40 Second, self-efficacy was reinforced through micro-successes and error-normalization routines (e.g. intentional practice in recovering from technical malfunctions). We conceptualize this as a Competence-Safety Cycle: repeated, incremental successes coupled with low-risk rehearsal transform initial apprehension into routinized competence, lowering the behavioral initiation threshold for novel digital tasks. This mechanism extends the self-efficacy pathway within the HBM and aligns with implementation science frameworks (e.g. RE-AIM, NPT) that stress operational routines and resilience in the face of failures.^41,42

Barriers clustered around interface complexity, fragmented cross-app workflows, time constraints, dependence on assistance, and low perceived credibility or competence of online information. These patterns map to HBM constructs of perceived barriers and reduced self-efficacy, both strong predictors of avoidance and abandonment of health technologies.²⁰ Among older adults, burdens are magnified by small-screen constraints, deep menu hierarchies, and ambiguous error states. Convergent evidence indicates that lowering cognitive load—via plain-language microcopy, single-screen completion of critical tasks, larger touch targets, consistent iconography, and contextual inline help—improves task completion and reduces attrition.^43–45 Our data therefore favor an integrated, task-oriented flow (testing → logging → trend analysis → consultation → follow-up) rather than cross-application navigation, which introduces cognitive discontinuities and heightens reliance on family intermediaries; similar discontinuities are well documented in the nonadoption and abandonment literature.^46,47

Behavioral triggers were situational and acute: new or worsening symptoms, worry about complications, and authoritative cues (e.g. physician recommendations, QR codes at clinical touch points) accelerated initial uptake, while peer modeling supported adherence—consistent with HBM predictions under conditions of elevated perceived severity/susceptibility and credible cues to action.⁴⁸ Empirical work shows that clinician endorsement increases portal and tool adoption among older adults and that Diabetes Self-Management Education and Support (DSMES) programs improve self-management behaviors and glycemic outcomes.^49,50 Participants also repeatedly requested just-in-time assistance at points of uncertainty (e.g. a brief “Help” video adjacent to an error message), aligning with the paradigm of just-in-time adaptive interventions, which deliver context-aware support at the moment of need and have demonstrated feasibility in older populations.⁵¹

Taken together, we propose a practical, testable model of e-HL for older adults that integrates three components (Supplemental Figure S2): HBM targeting, a senior-centered user experience and system-level integration. The HBM layer specifies modifiable targets, such as perceived susceptibility, perceived severity, perceived benefits, perceived barriers, cues to action, and self-efficacy.²⁰ The user experience layer translates these targets into implementation strategies, such as single-screen critical tasks, high-contrast typography, large touch targets, voice input with dialect sensitivity where feasible and explicit error-recovery pathways that are stress-tested in a “glitch lab.” The system layer embeds these interventions into routine care, including physician-as-digital-prescriber workflows (e.g. QR-coded prescriptions and discharge bundles), linkage with DSMES, scheduled telehealth follow-ups and institutional endorsement of platforms. This triad is directionally consistent with the 2025 ADA Standards of Care in Diabetes, which emphasize patient-centered technology and routine assessment of DSMES needs. It is also consistent with the WHO's Global Strategy on Digital Health 2020–2025, which advocates for equitable, context-adapted digital health.^34,52,53

When evaluating these implications, it is also important to recognize the strengths and limitations of the study. The study's primary strength lies in its qualitative depth, providing actionable insights grounded in users’ lived experiences. These insights are further enriched by the integration with the HBM and the focus on the urban–rural divide. While the HBM effectively captured individual-level perceptions, this study did not quantitatively incorporate broader structural determinants (e.g. economic access, device affordability, internet infrastructure) that may mediate the explored relationships. Future research should adopt multi-level approaches to bridge this gap. The generalizability of our findings is constrained by two main factors. First, participants were recruited from a single tertiary hospital in Shenzhen. Although the study included participants with rural household registrations, its urban location may not represent healthcare environments in rural China or other regions, cautioning extrapolation to these populations. Future multicenter studies across diverse geographical and socioeconomic settings are needed to enhance external validity. Second, the applicability of our qualitative tool is confined by its specific sociocultural and healthcare context. It should thus be regarded as a conceptual template outlining key inquiry domains rather than a universal instrument. Meaningful cross-cultural application requires meticulous adaptation and validation to ensure relevance and accuracy in new settings. Methodologically, the study relied solely on investigator triangulation (multiple coders), omitting other forms such as data triangulation (multiple sources) or methodological triangulation (different methods). This restricted triangulation may increase susceptibility to biases associated with a single analytical approach, potentially limiting the internal validity and depth of the findings. Integrating multiple triangulation methods prospectively, such as combining qualitative and quantitative data, could strengthen future causal inferences. Furthermore, our conceptualization of socioeconomic status (SES) was limited, as key individual-level indicators (e.g. income, occupation) were not measured. Subsequent studies would benefit from employing a more comprehensive set of standardized SES metrics to clarify the independent roles of these interrelated factors.

Conclusion

This study validates the HBM as a robust theoretical framework for determining e-HL in older Chinese adults with diabetes, where the facilitators, barriers and behavioral triggers align with HBM constructs. Our findings propose a tripartite intervention framework that integrates HBM targeting, a senior-centered user experience, and system-level integration to enhance e-health adoption sustainably. This integrated approach addresses the urgent need for context-adapted solutions in ageing societies by bridging digital health divides through scalable pathways.

Supplemental Material

sj-doc-1-dhj-10.1177_20552076261432074 - Supplemental material for Toward enhanced e-health literacy: A qualitative exploration of facilitators, barriers, and behavioral triggers among Chinese older adults with diabetes

Supplemental material, sj-doc-1-dhj-10.1177_20552076261432074 for Toward enhanced e-health literacy: A qualitative exploration of facilitators, barriers, and behavioral triggers among Chinese older adults with diabetes by Shiying Cai, Zikai Feng, Min Chen, Huanjiang Liu, Zhiyuan Tian, Yi Fang, Qiuling Xie, Jiaxin Chen, Ren Jie, Xintong Xie, Bin Lin and Jue Li in DIGITAL HEALTH

Footnotes

Acknowledgments

The authors would like to thank all the participants in this study.

ORCID iDs

Huanjiang Liu

Ren Jie

Jue Li

Ethical considerations

The study was approved by the Medical Ethics Committee of Fuyong People's Hospital, Bao'an District, Shenzhen, informed consent form was distributed to participants. The study strictly adheres to the principles outlined in the Declaration of Helsinki, all respondents voluntarily participated and provided signed consent (KY-2024-110).

Contributorship

Conceptualization: SYC, ZKF and MC; methodology: SYC, ZKF, MC, HJL and ZYT; validation: QLX and YF; formal analysis: SYC, ZKF and MC; investigation: SYC, ZKF and MC; writing-original draft preparation: SYC, ZKF and MC; writing-review and editing: SYC, ZKF, MC and XTX; visualization software: JXC and JR; supervision: BL and JL; and project administration: BL and JL. All authors have read and agreed to the published version of the manuscript.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the "Key Laboratory of Digital Traditional Chinese Medicine Culture Research and Communication Innovation,” Guangdong Federation of Social Sciences (Grant No. Guangdong Xuanfa [2022] No. 10) and 2023 Shantou University Medical College High-level Talents Introduction Research Initiation Project (Grant No. 009-510858069).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Supplemental material

Supplemental material for this article is available online.

References

Cole

Florez

. Genetics of diabetes mellitus and diabetes complications. Nat Rev Nephrol 2020; 16: 377–390.

GBD 2019 Diabetes and Air Pollution Collaborators. Estimates, trends, and drivers of the global burden of type 2 diabetes attributable to PM2·5 air pollution, 1990–2019: an analysis of data from the global burden of disease study 2019. Lancet Planet Health 2022; 6: e586–e600.

IDF Diabetes Atlas. Global diabetes data report 2000–2045, https://diabetesatlas.org/data/.

Jia

Chan

Wong

, et al. Diabetes in China: epidemiology, pathophysiology and multi-omics. Nat Metab 2025; 7: 16–34.

GBD 2021 Diabetes Collaborators. Global, regional, and national burden of diabetes from 1990 to 2021, with projections of prevalence to 2050: a systematic analysis for the global burden of disease study 2021. Lancet 2023; 402: 203–234.

LeRoith

Biessels

Braithwaite

, et al. Treatment of diabetes in older adults: an endocrine society clinical practice guideline. J Clin Endocrinol Metab 2019; 104: 1520–1574.

Shan

Sarkar

Martin

. Digital health technology and mobile devices for the management of diabetes mellitus: state of the art. Diabetologia 2019; 62: 877–887.

State Council of China. Healthy China 2030 planning outline. Beijing: Government of China, 2016.

Norman

Skinner

. eHealth literacy: essential skills for consumer health in a networked world. J Med Internet Res 2006; 8: e9.

10.

Xie

Zhang

Xin

, et al. Electronic health literacy and health-related outcomes among older adults: a systematic review. Prev Med 2022; 157: 106997.

11.

Peimani

Stewart

Ghodssi-Ghassemabadi

, et al. The moderating role of e-health literacy and patient-physician communication in the relationship between online diabetes information-seeking behavior and self-care practices among individuals with type 2 diabetes. BMC Prim Care 2024; 25: 442.

12.

Sadeghi

Borzu

Shirzadi

, et al. Investigating e-health literacy and its relationship with self-care behaviors among Iranian middle-aged type 2 diabetes: a cross-sectional study. BMC Public Health 2025; 25: 1376.

13.

Shiferaw

Tilahun

Endehabtu

, et al. e-Health literacy and associated factors among chronic patients in a low-income country: a cross-sectional survey. BMC Med Inform Decis Mak 2020; 20: 181.

14.

Zhang

Sun

, et al. Factors influencing the e-health literacy in cancer patients: a systematic review. J Cancer Surviv 2023; 17: 425–440.

15.

Cui

Yin

, et al. Health-promoting behaviors mediate the relationship between eHealth literacy and health-related quality of life among Chinese older adults: a cross-sectional study. Qual Life Res 2021; 30: 2235–2243.

16.

Seidu

. Using Anderson's model of health service utilization to assess the use of HIV testing services by sexually active men in Ghana. Front Public Health 2020; 8: 512.

17.

Shi

, et al. Factors influencing digital health literacy among older adults: a scoping review. Front Public Health 2024; 12: 1447747.

18.

Kim

Choi

. Association of electronic health literacy with health-promoting behaviors in patients with type 2 diabetes: a cross-sectional study. Comput Inform Nurs 2018; 36: 438–447.

19.

Kerr

Ahn

Waki

, et al. Digital interventions for self-management of type 2 diabetes mellitus: systematic literature review and meta-analysis. J Med Internet Res 2024; 26: e55757.

20.

Janz

Becker

Janz

, et al. The health belief model: a decade later. Health Educ Q 1984; 11: 1–47.

21.

Liu

, et al. Factors influencing self-care behaviours of patients with type 2 diabetes in China based on the health belief model: a cross-sectional study. BMJ Open 2022; 12: e044369.

22.

Pati

Menon

Rehman

, et al. Developing and assessing the “MultiLife” intervention: a mobile health-based lifestyle toolkit for cardiometabolic multimorbidity in diabetes and hypertension management -a type 1 hybrid effectiveness-implementation trial protocol. BMC Public Health 2025; 25: 3.

23.

Zhang

Yang

, et al. How health seeking behavior develops in patients with type 2 diabetes: a qualitative study based on health belief model in China. Front Public Health 2024; 12: 1414903.

24.

Helbing

Brockmann

Chadefaux

, et al. Saving human lives: what complexity science and information systems can contribute. J Stat Phys 2015; 158: 735–781.

25.

Uygun İlikhan

Özer

Perc

, et al. Complementary use of artificial intelligence in healthcare. Med J West Black Sea 2025; 9: 7–17.

26.

Sandelowski

. Whatever happened to qualitative description? Res Nurs Health 2000; 23: 334–340.

27.

Kallio

Pietilä

Johnson

, et al. Systematic methodological review: developing a framework for a qualitative semi-structured interview guide. J Adv Nurs 2016; 72: 2954–2965.

28.

Tong

Sainsbury

Craig

. Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups. Int J Qual Health Care 2007; 19: 349–357.

29.

Shephard

. The 1975 declaration of Helsinki and consent. Can Med Assoc J 1976; 115: 1191–1192.

30.

Alberti

Zimmet

. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabet Med 1998; 15: 539–553.

31.

Hennink

Kaiser

. Sample sizes for saturation in qualitative research: a systematic review of empirical tests. Soc Sci Med 2022; 292: 114523.

32.

Braun

Clarke

. Using thematic analysis in psychology. Qual Res Psychol 2006; 3: 77–101.

33.

American Diabetes Association Professional Practice Committee. 7. Diabetes technology: standards of care in diabetes-2025. Diabetes Care 2025; 48: S146–S166.

34.

World Health Organization. Global strategy on digital health 2020–2025. Geneva: World Health Organization, 2021.

35.

The State Council General Office. State council issues guideline on Internet plus healthcare, 2018.

36.

Chang

Xin

, et al. Strategies for assessing health information credibility among older social media users in China: a qualitative study. Health Commun 2024; 39: 2767–2778.

37.

De Gagne

Park

Hall

, et al. Microlearning in health professions education: scoping review. JMIR Med Educ 2019; 5: e13997.

38.

Liu

Cai

, et al. Peer support in Shanghai's commitment to diabetes and chronic disease self-management: program development, program expansion, and policy. Transl Behav Med 2020; 10: 13–24.

39.

Taylor

Souza

Little

, et al. Enhancing telehealth competency: development and evaluation of education modules for older adults. OTJR (Thorofare N J) 2023; 43: 478–486.

40.

Chen

Lifset

Han

, et al. Screen or no screen? Lessons learnt from a real-world deployment study of using voice assistants with and without touchscreen for older adults. ASSETS 2023; 2023: 52.

41.

Glasgow

Vogt

Boles

. Evaluating the public health impact of health promotion interventions: the RE-AIM framework. Am J Public Health 1999; 89: 1322–1327.

42.

May

Mair

Finch

, et al. Development of a theory of implementation and integration: normalization process theory. Implement Sci 2009; 4: 29.

43.

Gomez-Hernandez

Ferre

Moral

, et al. Design guidelines of mobile apps for older adults: systematic review and thematic analysis. JMIR Mhealth Uhealth 2023; 11: e43186.

44.

Wang

Liu

Zhou

, et al. Usability evaluation of mHealth apps for elderly individuals: a scoping review. BMC Med Inform Decis Mak 2022; 22: 317.

45.

Wildenbos

Jaspers

MWM

Schijven

, et al. Mobile health for older adult patients: using an aging barriers framework to classify usability problems. Int J Med Inform 2019; 124: 68–77.

46.

Greenhalgh

Wherton

Papoutsi

, et al. Beyond adoption: a new framework for theorizing and evaluating nonadoption, abandonment, and challenges to the scale-up, spread, and sustainability of health and care technologies. J Med Internet Res 2017; 19: e367.

47.

Shin

Hamovitch

Gatov

, et al. The NASSS (non-adoption, abandonment, scale-up, spread and sustainability) framework use over time: a scoping review. PLOS Digit Health 2025; 4: e0000418.

48.

Jones

Jensen

Scherr

, et al. The health belief model as an explanatory framework in communication research: exploring parallel, serial, and moderated mediation. Health Commun 2015; 30: 566–576.

49.

Banguilan

Sonnenberg

Chen

. Physicians’ perspectives on inpatient portals: systematic review. Interact J Med Res 2022; 11: e39542.

50.

Werner

Ufholz

Yamajala

. Recent findings on the effectiveness of peer support for patients with type 2 diabetes. Curr Cardiovasc Risk Rep 2024; 18: 65–79.

51.

Nahum-Shani

Smith

Spring

, et al. Just-in-time adaptive interventions (JITAIs) in mobile health: key components and design principles for ongoing health behavior support. Ann Behav Med 2018; 52: 446–462.

52.

American Diabetes Association Professional Practice Committee. 1. Improving care and promoting health in populations: standards of care in diabetes-2025. Diabetes Care 2025; 48: S14–S26.

53.

Davis

Fischl

Beck

, et al. 2022 national standards for diabetes self-management education and support. Diabetes Care 2022; 45: 484–494.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.76 MB