This study employed a multicenter qualitative descriptive design, which is well-suited for exploring human experiences and gaining insights into subjective perspectives on the safety aspects of mHealth usage [28]. A multicenter approach was chosen to capture diverse perspectives, enhance representation, and improve the transferability of findings [29]. The study presentation was guided by the 21-item Standards for Reporting Qualitative Research checklist [30].

This study was conducted in Hong Kong SAR and Wuhan city in mainland China from May 2022 to August 2024. Both locations are characterized by high internet penetration rates and widespread mHealth usage [31,32]. Eligible participants were health care professionals who used mHealth for health service provision as initiated by their institutions, departments, or health governing authorities. Health care professionals working in nondirect care roles, such as technical support or appointment booking hotlines, were excluded. Since both settings lack formal training for staff in using mHealth and some mHealth follow-ups require 6 to 12 weeks, those with less than 3 months of experience in providing mHealth services were excluded to ensure a thorough understanding of the topic. For sample homogeneity, only health care professionals offering real-world mHealth care services in public health care institutions (eg, hospitals and community health care centers) were included.

Participants were recruited through purposive sampling, leveraging poster advertisements distributed via alumni groups, social media platforms, and word of mouth. This approach allowed the study researchers to intentionally select participants with substantial knowledge and experience in mHealth usage for in-depth interviews [33].

Semistructured individual in-depth interviews were conducted to capture the participants’ perspectives while ensuring relevance to the study objective [34]. An interview guide was developed based on the authors’ experiences, the literature, and a framework (Multimedia Appendix 1). Two female authors (JJS and MHSC), both with extensive experience in mHealth service provision and qualitative interviewing, conducted the interviews. Their familiarity with the topic enabled a nuanced understanding of participants’ experiences, aligning with the notion that interviewer-participant dynamics exist on a continuum rather than an insider-outsider dichotomy [35]. Both interviewers received extensive training in conducting qualitative interviews to ensure neutrality. Additionally, a standard interview guide was used and a pilot interview was conducted, which was followed by a briefing, to ensure data integrity [36]. To maintain confidentiality, no identifying information, such as participants' names or institutional details, was collected [37]. Probing questions, neutral interview locations with privacy (eg, a quiet coffee shop or garden), and a nonjudgmental and respectful attitude were employed to minimize social desirability bias and encourage authentic responses [38].

Interviews lasted for 40 to 60 minutes, were audio-recorded, and were accompanied by field notes to capture nonverbal cues. Depending on participants’ preferences, interviews were conducted either in person or online (via videoconferencing) to accommodate their schedules. The use of 2 interview channels was supported by the literature, as both methods yield similar levels of self-disclosure and equivalent thematic content [39,40]. The sample size was determined based on the principle of data saturation, with 2 additional interviews conducted after no new findings emerged from the 25th interview to confirm saturation [41].

Thematic analysis was used, which can identify patterns within and across data regarding participants’ views, experiences, and practices [42]. The audio recordings were transcribed verbatim alongside field notes. Transcripts were reviewed for accuracy against the original audio recordings by interviewers. Two researchers (JJS and MHSC) independently read and reread the transcripts to immerse themselves into the data and made notes to identify the initial codes [36]. After independently coding 4 interviews, the researchers compared their codes for consistency and resolved discrepancies through discussion, updating the codebook accordingly. Subsequent transcripts were coded independently, with regular adjudication sessions held to refine the codebook. Data saturation was reached at the 16th interview with no themes emerging. However, the researchers completed 27 interviews to ensure a richly textured understanding of the topic [43]. Codes with similar meanings or conceptual relationships were grouped into subthemes and themes using constant comparative analysis. Final themes were determined through extensive team discussions. Dependability was ensured by maintaining an audit trail documenting the rationale for coding decisions and theme development. Member checking was conducted with 7 participants from Hong Kong (n=3) and Wuhan (n=4) to validate the findings.

Ethics approval was obtained from the Institutional Review Boards of Hong Kong Polytechnic University (HSEARS20240131005) and Shanghai University (ECSHU2022046). Participants were informed of their right to withdraw at any time without any consequences. Written informed consent was obtained from participants before interviews commenced.

A total of 27 health care professionals were recruited and interviewed, including 20 professionals (5 physicians and 15 nurses) from hospitals in Wuhan and 7 nurses from community health care settings in Hong Kong SAR. The mean age of the participants was 32.19 (SD 3.74) years, with a mean working experience of 8.04 (SD 4.05) years. Most participants were female (20/27, 74%). Nearly half of the participants had a bachelor’s degree (13/27, 48%), some had a master’s degree (9/27, 33%), and few had a diploma degree (3/27, 11%) or a doctoral degree (2/27, 7%). Participants from Hong Kong SAR shared using various mobile apps published by the Hospital Authority for home-based care, such as Stoma Care, Hip Fracture, and HA Go. On the other hand, participants from Wuhan shared using consultation apps of their hospitals, WeChat mini-programs, and the WeChat Official Accounts Platform (a subapp within the WeChat ecosystem).

Four overarching themes and 4 subthemes emerged, specifically addressing the multifaceted dimensions of safety concerns, including the theoretical, technical, organizational, and patient-centered aspects of mHealth use. All 4 themes were covered by participants from 2 different health care systems, with some differences in subthemes further elaborated. Figure 1 illustrates the key themes and subthemes identified. 

Liability emerged as a central concept for health care professionals when discussing patient safety in the context of mHealth usage. While participants used varied terms, such as “who,” “what if,” and “how,” their insights collectively illuminated the multifaceted risks tied to liability. These risks spanned the design, content, compatibility with existing health care services, promotion, and usage of mHealth technologies. Participants described liability as encompassing daily practice challenges, including patient deterioration due to delayed responses, unclear service scopes, and fragmented health care systems. These liability issues were described as extending beyond individual responsibilities, necessitating systematic or structural identification and resolution. Health care professionals across regions and institutions consistently highlighted liability as a potential barrier to the broader application and acceptance of mHealth if left unaddressed.

Health care professionals emphasized that the safe integration of mHealth into real-world settings requires effective change management at both individual and institutional levels. Four subthemes emerged: appropriate referral, resources and funding, informed mHealth usage, and working process efficiency. While the first 3 were viewed as primarily benefiting patients, working process efficiency was regarded as critical for health care professionals. Comparative analysis revealed a notable absence of concerns regarding resources and funding among participants from Hong Kong SAR. In contrast, these participants notably focused more on informed mHealth usage and efficiency improvements compared to their counterparts in Wuhan.

Participants highlighted that patient safety in mHealth depends not merely on providing information but on ensuring that patients truly understand it. They noted that while mHealth offers convenience and popularity, significant digital disparities, especially among older adults and underserved populations, pose serious risks. Participants stressed that overlooking these disparities could turn mHealth into a hazard by excluding vulnerable groups or providing generalized information that might mislead and harm patients.

Without ensuring understanding, participants expressed that mHealth places professionals in challenging positions, leaving patients to independently interpret and adopt health behaviors, often with insufficient support. To address these challenges, participants advocated for face-to-face guidance before introducing mHealth tools.

Few participants shared that they tend to avoid mentioning mHealth to patients who have difficulty understanding the technology or educational content in order to avoid risks.

The need for proper training was especially pronounced when using wearable devices to remotely track health data. Participants emphasized empowering patients or their caregivers to correctly use these devices, including providing hands-on training and asking for return demonstrations.

Participants consistently highlighted the importance of information security, emphasizing that health information is sensitive in nature and that information leakage could bring both ethical and economical risks for patients. They shared that mHealth usage involves a health communication platform, which includes sociodemographic and clinical backgrounds, and diagnostic and treatment suggestions. Participants identified different scenarios where information leakage might occur, including device loss by professionals, patients inadvertently sharing information while seeking technical support, and oversharing through social media apps. Notably, differences emerged between the perspectives of professionals from Hong Kong SAR and those from Wuhan. Participants from Hong Kong SAR emphasized existing regulatory policies designed to protect patient information. They described strict measures for accessing medical systems and protocols for handling devices containing sensitive data. In contrast, professionals from Wuhan expressed the need for stricter oversight and governance of information security, particularly given the widespread use of social media platforms like WeChat and apps developed by commercial entities. They advocated for regulatory authorities, such as the public security bureau, to monitor and record activities to mitigate risks. Despite the differences, participant narration delineates the need for standard practice guidelines or standards across regions to ensure secure use of mHealth, which is tailored to local contexts and technological infrastructure. Participants from both regions recognized the potential risks of using mHealth platforms that store and communicate sociodemographic, clinical, and treatment-related data. Participants noted the shared responsibility of health care providers, regulatory bodies, and developers in ensuring robust data protection mechanisms.

The integration of mHealth into health care systems has rapidly gained momentum as a global trend, offering promising opportunities to enhance both patient care and operational efficiency. As health care systems increasingly adopt mHealth solutions, it becomes crucial to understand the implications for patient safety in these new care models. This qualitative study is among the first to explore health care professionals’ perspectives on the safety concerns associated with mHealth implementation in real-world settings. The findings emphasize the importance of patient safety in ensuring that mHealth is integrated into a reliable and sustainable care model. Across participants from Hong Kong SAR and Wuhan, common themes emerged regarding safety challenges and the strategies for managing these concerns. However, regional differences also became apparent, highlighting the need for context-specific adaptations to effectively address local health care needs and conditions. These findings not only contribute to the growing body of knowledge on mHealth but also provide valuable insights for policy makers and health care providers as they navigate the complex process of integrating mHealth into diverse health care environments.

Findings from this study highlighted liability considerations pertaining to mHealth across settings, which differ from the liability statements of traditional health care settings that detail professional malpractice and compensation to patients [44]. Health care professionals’ discussions on liability were not intended to identify a responsible individual when risks occur, which could drastically hinder mHealth usage, but rather to acknowledge the inherent limitations and advocate for caution when adopting mHealth. Unlike a previous study on real-world mHealth usage where professionals and managers expressed concerns about the trustworthiness of apps [45], findings from this study highlighted the intrinsic limitations of mHealth in ensuring accurate health information exchange to enable prompt patient evaluation and decision-making. This difference may be attributed to the fact that participants from Hong Kong SAR used mHealth apps developed, operated, managed, and financed by the Hospital Authority, while participants from Wuhan were regarded as high-caliber professionals from top hospitals who translated their expertise in delivering in-person care to mHealth. Moreover, owing to the self-finance attributes of mHealth care provision in Wuhan [46], participants emphasized the importance of explicitly communicating the estimated timeframe and responding in a timely manner to enhance reliability.

The study also revealed the vital role of change management in implementing mHealth in real-world settings. Consistent with existing literature [47,48], participants unanimously recognized mHealth as an inevitable trend due to its potential to enhance patient outcomes and professional efficiency. Growing acceptance and usage of technology among patients further motivated professionals to recommend appropriate mHealth platforms [47]. Likewise, health care professionals should be trained and informed about existing reliable mHealth options and be prepared to educate patients about the potential benefits and risks of using them [49]. Effective change management, as suggested by participants, should account for variations in the readiness of professionals to adopt mHealth, emphasizing the importance of training and education to inform both professionals and patients about the benefits and limitations of mHealth. Even health care professionals who hesitate to use mHealth should be empowered to learn and prepare themselves to partner with colleagues and patients who desire using mHealth for health purposes to ensure care consistency and patient safety.

Findings from this study affirmed the literature’s emphasis that mHealth care provision should ensure the understanding of patients to empower engagement and generate health benefits [50,51]. Unlike interventional studies that consider multiple factors to manage confounding variables, participants in this study consistently suggested mobilizing various resources, such as volunteers for technology coaching, informal caregivers, and in-person teachers for return demonstrations, to improve comprehension and prevent risks in real-world mHealth implementation. Theoretically, mHealth literacy should be evaluated and promoted to ensure equity in mHealth implementation, while in practice, health care professionals’ evaluations of patients’ mHealth literacy could determine whether they use or recommend mHealth. Consistent with the literature, this study mentioned the usage challenges faced by patients, such as small font sizes for older adults, small touch screens to operate, and learning difficulties [14]. The need for professionals to repeatedly teach patients to operate technology and address potential misunderstandings could dampen the confidence of professionals and compromise their willingness to use mHealth for care provision.

Information security, though less prominently discussed, emerged as a high priority to meet legislative requirements, which is consistent with a previous study that revealed the perspectives of stakeholders in real-world mHealth service provision [52]. This aspect was considered particularly important when participants mentioned the tendency of linking health records from health care settings with mHealth platforms, the collection of telemonitoring data, and the exchange of multimedia information between professionals. However, contrary to a previous study where managers considered data security to be outside their direct role [52], participants in this study shared experiences of themselves proactively implementing strategies to prevent information leakage.

The findings from this study could enhance the understanding of the safety aspects associated with real-world mHealth service provision and inform essential change management at the individual, process, and institutional or policy levels. Despite collecting data regionally, the study findings could have international applicability as patient safety is holistic and systematic, highlighting the universality of human perspectives and strategies to enhance safety through the establishment, interactions, experiences, and outcomes of mHealth. Patient safety remains the first and foremost priority in mHealth care provision, necessitating effective communication, care coordination, and multidisciplinary collaboration to ensure care continuity [53]. The results could potentially inform developers and user interaction designers to include more safety-enhancing features for real-world mHealth implementation, especially to make it user-friendly and inclusive for people with pre-existing ailments or older adults. Despite variations in mHealth apps and their integration into health care systems, the professional skills and user views identified in this study could hold significant potential for advancing international health care. This study also has policy implications. Health policy makers in Hong Kong SAR may consider redefining the liability of mHealth and cautioning against the inherent limitations of this model to promote informed usage and appropriate mHealth referral. Health policy makers in mainland China may consider expanding guiding policies to include not only diagnosis and treatment but also patient consultation, self-care empowerment, and related data security measures. The settings may support the training of health care professionals to cultivate smart manpower and streamline care processes to provide safe mHealth services.

While this study offers valuable insights into the patient safety aspects of mHealth in real-world health care settings, it is not without limitations. First, the study’s qualitative design limits the generalizability of the findings, as the data were collected from health care professionals in 2 specific regions, Wuhan and Hong Kong SAR. The findings may not fully reflect the experiences or perspectives of health care providers in other regions with differing health care systems or technological infrastructure. Moreover, excluding health care professionals in nondirect care roles may omit important perspectives. Future studies should consider involving leaders, health informatics personnel, and other related parties to capture broader views. While the multicenter approach enhanced the breadth of perspectives, the findings might not capture the full diversity of health care professional experiences across different health care settings, particularly in smaller or rural environments. While the study focused on health care professionals’ perspectives, it did not include the views of students, support staff, leaders or managers, patients, or caregivers, who play critical roles in mHealth adoption and usage. Future research should include a broader range of stakeholders. Lastly, the samples from the 2 health care systems were not exactly homogeneous, as we recruited both physicians and nurses from Wuhan and only nurses from Hong Kong SAR (despite several rounds of attempts to recruit physicians), which may challenge the stability of the results. This difference arose because mHealth platforms are co-created and co-operated by physicians and nurses from the same departments in Wuhan, whereas the division of care provision in Hong Kong SAR is clear.

The widespread adoption of mHealth technologies is reshaping health care delivery, bringing both significant opportunities and challenges. Health care professionals recognized the importance of patient safety in mHealth usage and described key considerations for ensuring patient safety. They recommended different competencies and shared commendable practices and policies for delivering mHealth care safely. Institutional or system-level support is needed to clarify the scope and liability of mHealth to facilitate informed usage, improve change management, and ensure information security. Furthermore, it is crucial for health care professionals to foster greater technology acceptance, make informed referrals, and strengthen their competencies in ensuring patient comprehension in order to maintain the safety and sustainability of mHealth care.