Refining smart healthcare care for mental health and substance use disorders: A patient-centred, evidence-based approach

Abstract

In this article, we comment on the article by Zhang et al, which explores the familiarity, awareness, and usage of smart medical care and its correlation with mental health and personality traits. The use of intelligent healthcare technologies in treating mental disorders and substance use disorders shows significant promise, but involves certain challenges, such as limited access, low technological literacy, and privacy concerns. These barriers disproportionately affect deprived populations and individuals with severe mental health conditions. We highlight the positive impact of smart healthcare solutions, such as telemedicine and wearable technologies, on patient engagement, remote monitoring, and treatment adherence. To overcome these challenges, we propose strategies, such as improving user-friendliness, ensuring equitable access to digital interventions, enhancing cybersecurity, and integrating smart healthcare into clinical workflows. Training healthcare providers and developing policies to ensure the ethical use of patient data are essential. When implemented thoughtfully, smart healthcare technologies can revolutionize mental health and substance use disorder treatment, improve patient outcomes, and reduce healthcare inequities.

Key Words: Mental health disorder; Smart healthcare; Patient-centered; Substance use disorder; Digital health

Core Tip: This article explores the integration of smart healthcare technologies, such as wearable devices, telemedicine, and electronic health records, into psychiatric and substance use disorders. It highlights innovative approaches, such as ecological momentary assessments and smartphone-based monitoring that enable personalized and proactive care. We discuss challenges, including data complexity, usability, and ethical concerns, and provide recommendations - such as user-friendly designs with gamification, privacy safeguards, adaptive interfaces for mobility issues, and scalable stepped-care models - to enhance engagement. The need for certification, clinician support tools, and two-way patient communication is also addressed, optimizing smart healthcare for better outcomes in mental health and substance use disorder treatment.

INTRODUCTION

In this paper, we comment on the article by Zhang et al[1]. Mental health disorders and substance use disorders (SUDs) affect over 1 billion individuals worldwide, accounting for approximately 7% of the total disease burden measured in disability-adjusted life years as of 2016[2]. Despite this significant burden, a substantial treatment gap persists, with more than 70% of individuals with mental disorders failing to receive timely and early assistance[3]. This treatment gap is further exacerbated by the socio-economic determinants of mental health, which are often aggravated by health, social, economic, and environmental crises, as highlighted during the coronavirus disease 2019 pandemic. The digitization of healthcare offers a promising avenue to bridge this gap, particularly through the integration of smart healthcare technologies. Smart healthcare involves the application of advanced technologies and data analytics to enhance efficiency, diagnostic accuracy, and patient outcomes. The key innovations include wearable devices for real-time monitoring, telemedicine for remote consultations, and electronic health records for seamless access to patient information. By utilizing smart healthcare solutions, healthcare providers can offer more personalized and proactive care, leading to better overall patient health outcomes.

The application of smart healthcare has transformative potential in the field of mental health and SUDs. Devices, such as implanted biosensors, ingestible medications, neurostimulators, and tracking tools can monitor mood states, stress levels, activity patterns, and cognitive functions. These technologies create opportunities for early detection, continuous monitoring, and individualized intervention. However, their implementation is challenging. Ensuring data privacy and security, managing the complexity of passive and continuous data collection, and addressing the heterogeneity of psychiatric conditions are the critical barriers. Additionally, disparities in access to these technologies, particularly in resource-limited settings, emphasise the need for equitable solutions[4].

Patient acceptance of and willingness to engage with smart healthcare tools are pivotal for their successful integration into psychiatric practice. Addressing concerns, such as usability, trust, and perceived benefit, is essential for fostering engagement with the tools. Similarly, a patient-centred approach that prioritizes individual needs, preferences, and active participation is critical for improving treatment outcomes, enhancing patient satisfaction, and ensuring long-term adherence. Despite its importance, the implementation of patient-centred care in smart healthcare remains inadequately explored. To this end, this article highlights the transformative potential of smart healthcare in mental health and SUDs with a focus on addressing implementation challenges, enhancing patient-centred methodologies, and ensuring equitable access. By addressing these issues, smart healthcare can advance mental healthcare delivery and improve global outcomes.

ROLE IN PSYCHIATRIC COMORBIDITY

Given the chronic nature of mental health conditions, it is critical to meticulously monitor symptoms to enable early intervention, prevent relapse, and reduce hospitalization. Smartphones, owned by 72% of the individuals with severe mental disorders[5], facilitate ecological momentary assessments, enabling real-time data collection in natural settings. The integration of sensing technologies with conventional care can shift mental healthcare from a reactive to a proactive approach. Sensing technologies include movement tracking, location, speech and technology use patterns, and the level of physical activity. In addition, physiological signs, such as facial expression, heart rate variability, eye movement, and electrodermal activity can be monitored. For example, eye movement analysis can be used to differentiate patients with depressive disorders from controls, as well as patients with bipolar disorder (BD) from those with unipolar depression, where an increase in reaction time in prosaccade and antisaccade tasks is observed in both unipolar disorder and BD[6]. A wrist-worn accelerometer has been used in patients with mood disorders to measure physical activity and distinguish activity patterns between adults with BD and those with major depressive disorder[7]. Researchers have also used it to examine the relationships between activity, sleep, energy, and mood in individuals with and without mood disorders[8].

A growing number of computer scientists are analyzing social media data as linguistic and behavioral indicators to create digital phenotypes that may indicate the presence of mental health issues. To forecast the existence of particular mental disorders and symptomatologies, such as despair, suicidality, and anxiety, signals are extracted from the posting and behavioral history of social media platforms, such as Facebook, Reddit, and Twitter. Standard methods to ensure the validity of this research are suggested to obtain meaningful results[9]. Digital monitoring is an electronic mood-monitoring method that aims to improve compliance by sending reminders or improving user-friendliness[10-12]. Patients diagnosed with BD, experience a significant alleviation of depressive and manic symptoms with smartphone-based therapy and monitoring[13]. Numerous randomized controlled trials have investigated the efficacy of mobile applications, such as MONARCA[14] and SIMPLe[15] for self-monitoring of symptoms in individuals diagnosed with BD. A study involving 17 patients with BD utilized the MONARCA application over a three-month period. The results indicated a significant correlation between self-reported mood scores on the application and scores on the Hamilton Depression Rating Scale-17. However, no correlation was observed with the Young Mania Rating Scale[16]. Faurholt-Jepsen et al[10] conducted a systematic review and found that electronic self-monitoring is a reliable technique for assessing mood in individuals with depression but is not as effective for detecting mood in individuals with mania. Additionally, preliminary studies on the use of mobile applications have been conducted with very small sample sizes. For instance, MONARCA[14] was tested on only 78 patients aged 18-60 years, whereas SIMPLe was tested on only 30 participants aged > 18 years[15]. Although the preliminary studies are promising, it remains uncertain whether similar results can be observed in larger populations. Additionally, certain patient groups were excluded from these trials, including individuals outside the specified age ranges as well as those with severe depression or mania, and comorbid mental health conditions, such as schizophrenia and learning disabilities[16]. A systematic review of mobile device applications for BD management assessed various categories, including information (n = 32), screening and assessment (n = 10), symptom monitoring (n = 35), community support (n = 4), and treatment (n = 1). The findings indicated that the content rarely aligned with the practice guidelines (2/13, 15%) or established core psychoeducation principles (4/11, 36%). Furthermore, only 22% of the applications (18/82) addressed privacy and security concerns by offering privacy policies[17].

Nonetheless, studies in this field have several methodological shortcomings, primarily because the research area is new. A recent systematic review assessing the effectiveness of digital tools for passive monitoring of depression revealed that the majority of studies utilized opportunistic designs characterized by small sample sizes and brief follow-up durations, predominantly involving student populations, thereby restricting their generalizability[18]. A significant number of variables are generated in sensing studies, and it is probable that published papers selectively report and emphasize ‘positive’ findings. Furthermore, remote sensing technology, as a relatively novel advancement, imposes the responsibility of reporting on patients, with limited studies addressing the acceptability of study protocols among participating patients[18]. The validity of the devices employed to accurately measure target behavior requires further examination. Some behaviors, such as movement can be reliably inferred from global positioning system sensors or accelerometers. In contrast, more complex behaviors, including sleep and sociability, require multisensory data and greater inferential reasoning. Evidence regarding the efficacy of actigraphy for sleep detection is inconsistent. A review indicated a moderate-to-poor correlation with polysomnography[19], whereas another recent systematic review suggested its potential as a marker for sleep-wake patterns, noting that it tends to overestimate sleep duration and underestimate wake time[20]. Systematic reviews of the use of digital technology in BD are similarly constrained by studies characterized by small sample sizes, unclear generalizability of results, and heterogeneity in measures and outcomes[21]. In a study on self-monitoring practices, 45% individuals with BD (n = 552) had already been monitoring their mood, two-thirds utilized self-tracking data in consultation with healthcare professionals, and 80% perceived technology as having the potential to assist in managing their condition, expressing a specific interest in more automated digital symptom-tracking methods. Individuals with BD exhibit heightened sensitivity to stigma and may decline to use any technology that may label them as ‘different’. The potential of passive sensing has not been actualized in clinical practice; the data collected by sensors have predominantly been confined to research rather than therapeutic applications[22].

Both guided and self-guided mHealth psychotherapy applications have shown efficacy in treating mild-to-moderate depressive and anxiety symptoms, although their effectiveness in treating severe affective disorders remains uncertain[23,24]. Telehealth, including videoconferencing, internet-based programs, and mobile health applications, is feasible and generally accepted by service users and providers. Telemedicine was on the rise in mental healthcare prior to the Corona Virus Infectious Disease-2019 pandemic, but it was still in its infancy[25]. Although concerns about exacerbating symptoms in patients with psychosis are rare, many prefer teleconsultation because of its convenience and physical separation. Telehealth can reduce anxiety in patients with psychosis by providing a sense of security and control[26]. A systematic review evaluating the use of televideos in the management of major depressive disorder found that televideo treatment significantly improved satisfaction and relief from depressive symptoms compared to face-to-face interventions. Despite the initial costs, televideos are more cost-effective because of reduced travel expenses. However, the limited randomized control trials data, which mostly utilize collaborative treatment models, limit their generalizability[27].

High levels of user satisfaction have been observed when interventions aimed at influencing healthy lifestyles are delivered by voice-based conversational agents[28]. Smart healthcare is being increasingly utilized by older adults and in home-based care. Severe dementia can be detected early using non-invasive, easy-to-wear wearable technology and classification algorithms with a basic cognitive function test. Deep learning and machine learning techniques combined with smart medical devices can be effectively used in healthcare to monitor, classify, and diagnose various diseases, including Alzheimer’s disease[29]. The European Psychiatric Association recommends mobile-based interventions to improve the outcomes of patients with schizophrenia, depression, and their quality of life. Web-based psychoeducational interventions are favorable for family members and friends to increase their knowledge about schizophrenia and empower patients to discuss treatments. Online peer support groups are effective for patients and caregivers. However, modifications by mental health professionals are necessary. This association also suggests the need for the development of quality standards, ethical guidelines, and legal frameworks to regulate the provision of e-mental health interventions for individuals with schizophrenia and other psychotic disorders[30].

SUD AND SMART HEALTHCARE

Several mobile-based alcohol use disorder (AUD) treatments exist and are categorized into four types: Monitoring and reminders, intervention, full recovery management, and game-based systems. Although text-based applications are promising for supporting AUD management, their effectiveness remains limited. A systematic review found improved clinical outcomes, medication compliance, and active participation in peer support groups in most researches[31]. These programs focused on improving appointment attendance, motivation, self-efficacy, relapse prevention, and social support. Mobile phone messaging has also been used to deliver interventions, mainly in the domains of motivational messages, assessment, adherence to appointments and medication, affirmative messages for positive behavioral change, risk reduction, self-report of craving, and psychoeducation[32]. These interventions can be unidirectional, bidirectional, or even group-based, operating on principles, such as social learning, self-determination, social support, or the cognitive behavioral therapy model[33-35].

They have been utilized for the management of AUDs, opioid and stimulant use, and polysubstance use in combination with anti-retroviral therapy compliance. Personalized text messages have been shown to have greater effects on clinical outcomes and lower attrition rates. The interventions that provide immediate support have been considered highly acceptable. A systematic review highlighted improved clinical outcomes, including reduced alcohol, methamphetamine, and opioid use. Although most studies involved small, male-dominated (9/11) samples (n = 5-125), they included populations with people of color (n = 8) as well[32]. The barriers include the lack of standardized participant feedback models and usability issues, particularly with early generation phones. Moreover, certain studies have revealed that the benefits of the intervention cease to exist after delivery of those interventions[34].

A Cochrane review indicated a significant impact of digital interventions on tobacco use when compared to inactive controls [Risk Ratio: 1.15, 95% confidence interval (CI): 1.01-1.30, n = 6786]; however, the evidence quality was assessed as low[36]. Similarly, a meta-analysis of individual patient data indicated that internet interventions result in an average decrease in weekly alcohol consumption of 50.2 standard units (95%CI: -75.7 to -24.8) and a greater treatment response rate (odds ratio: 2.20, 95%CI: 1.63-2.95, P < 0.001) relative to various control groups[37]. For cannabis use, digital prevention programmes (g = 0.33, 95%CI: 0.13-0.54) and treatment programmes (g = 0.12, 95%CI: 0.02-0.22, P = 0.02) led to decreased use compared to controls[38]. Similarly, opioid use (g = 0.36, 95%CI: 0.20-0.53) and illicit drug use (g = 0.35, 95%CI: 0.24-0.45) showed significant reductions after digital interventions, although central stimulant drug use showed no significant change (4 studies, n = 481, P = 0.164)[39]. The evidence of newer interventions utilizing videoconferencing, mobile sensors, virtual reality, chatbots, and artificial intelligence remains insufficient. Economic evaluations suggest potential cost-effectiveness; however, further research is required to validate these findings[40].

BARRIERS FOR USE AND RECOMMENDATIONS

Technological interventions face significant barriers, including technical difficulties, device reliability, connectivity, and accuracy concerns. Lack of technological literacy, insufficient user competence, low education levels, and diminished expectations of interventions further reduce adherence and may contribute to these issues. For example, a meta-analysis of randomized controlled trials indicated that only 17% of patients completed all modules of unguided psychotherapy[41], and mental health application retention rates dropped to 3.9% after 15 days[42]. Research dedicated to enhancing smart healthcare and utilizing it to efficiently provide services to marginalized communities is scarce. More high-quality research is required to establish whether digitally delivered interventions can fully bridge the so-called ‘digital divide’, although there is moderate evidence that they can improve mental health outcomes among economically and digitally marginalized youths[43]. Through the co-operation of community organizations and healthcare providers, intelligent healthcare can be customized to address the distinct requirements of marginalized communities. Excessive reliance on smart devices can lead to digital fatigue, anxiety, and reduced interpersonal connections. Deficits in trust in therapists and unrealistic expectations further undermine the effectiveness of interventions. It is also imperative to gather additional definitive evidence regarding the effectiveness of newer technologies, such as voice-based conversational agents, in preventing and managing chronic mental health issues, both in isolation and when compared with conventional medical treatment[44]. It is essential to foster trust through transparent communication and patient-centric design.

RECOMMENDATIONS

Health monitoring systems that rely on sensors should possess usability, ergonomic design, and adherence to industry standards, and should be incorporated into fabrics or the skin of the users, guaranteeing exceptional signal precision and longevity, as well as being comfortable, pliable, and inconspicuous. Adoption of user-friendly, customizable designs that integrate gamification elements and automated reminders, such as levels, reward systems, social characters, and contests[45]. Rigorous privacy requirements and adherence, inclusion of early intervention tactics, and a high degree of customization[46]. Frequent retraining for the users is recommended to improve comfort with technology. Screen gadgets should enable a two-way communication, allowing patients to respond to notifications and request assistance. Alternative devices for people with mobility issues and other comorbidities like tremors[31] can improve utility. Collaboration with community organizations to develop tailored solutions and focus on cost-effective, scalable models, such as a stepped care approach where validated, low-intensity, self-guided digital mental health interventions could be provided to those with mild-to-moderate symptom severity, bridging the accessibility gap[24]. The establishment of a certification process for digital health technologies and digital mental health interventions specifically focused on mental health is recommended[24]. As per European Psychiatric Association recommendations, digital platforms must be designed to facilitate straightforward data entry and retrieval, seamless enrichment with procedural metadata, information from patient-owned digital health technologies and other external medical and non-medical sources, actionable information generation, including automated screening tools for at-risk patients and alerts for clinicians and patients, and encouragement for clinicians to adhere to care guidelines and quality recommendations[24].

ETHICAL CONCERNS

Unlike secure clinical networks, patients often lack the knowledge to follow security guidelines and store information on mobile platforms, which increases the risk of theft or hacking. For instance, the FBI advises connecting medical devices to a separate network from other smart home devices because shared networks create vulnerabilities[47]. Medical devices are considered the weakest link in healthcare cybersecurity, requiring significant investment in IT security, continuous monitoring, timely updates, and effective collaboration between manufacturers and users. Additionally, healthcare organizations should provide ongoing smart healthcare education to all physicians, staff, and linked patients, including those who are busy, uninterested, compromised, or facing budgetary challenges[48].

In addition to cybersecurity, extensive data collection from health applications has raised ethical concerns. Many applications track user movements via global positioning system or cell tower identification, enabling continuous location monitoring. Although healthcare providers are generally trusted in handling sensitive patient information, the involvement of third-party application developers and analysts introduces ambiguities and potential risks. Studies show that patients with severe mental disorders, such as BD frequently express concerns about the misuse of their data. For instance, one-third of patients in a cohort of people with severe mental disorder voiced anxiety about privacy risks[49]. A systematic review of mHealth applications in high-income countries revealed that patients often fear data breaches related to sensitive information including their identity and health conditions[50]. These concerns deter application usage despite its perceived benefits, such as improved health outcomes and enhanced patient-provider relationships. Patients consistently recommend measures to enhance trust and security, such as ensuring strict confidentiality, providing training on data privacy, and using discreet notifications to protect user anonymity. For instance, notifications from mental health applications can inadvertently reveal a user’s condition if viewed by others, causing distress and reducing application adoption. Transparent policies for data handling, secure application design, and user education can address these concerns. Involving patients in the design and development of digital health tools ensures alignment with their needs, while safeguarding their privacy. Establishing trust through these measures is crucial for the broader adoption and effectiveness of smart healthcare technologies[50].

CONCLUSION

Smart healthcare technologies hold immense promise for transforming medical practices by addressing the limitations of traditional methods. Innovations, such as telemedicine and wearable technologies have enabled remote monitoring, provided quicker access to medical advice, and improved the overall efficiency of healthcare systems, ultimately leading to improved patient outcomes. However, for these technologies to reach their full potential, it is crucial to make them user-friendly, particularly for individuals with severe mental health conditions. Involving patients in the development and refinement of these technologies is essential for ensuring their effectiveness and accessibility.

Despite these benefits, challenges persist in terms of affordability, accessibility, and usability, particularly in deprived communities and among patients with comorbidities. Future research should focus on addressing these gaps, investigating the long-term efficacy of smart healthcare interventions, and exploring ways to better integrate them into existing clinical workflows. Furthermore, policy formulation must prioritize equitable access to these technologies, ensuring that they are not only available, but also tailored to meet the diverse needs of different populations. To enhance practical applications, healthcare systems should invest in training healthcare providers, fostering digital literacy among patients, and developing clear guidelines for the ethical use of personal health data. Further studies are required to evaluate the impact of smart healthcare on mental health outcomes, patient engagement, and long-term care continuity. As the healthcare landscape continues to evolve, smart healthcare technologies can play a pivotal role in reducing health disparities, enhancing the quality of mental healthcare, and improving the overall well-being of individuals worldwide.