Digital health technologies, including artificial intelligence, analytics, dashboards, web portals, mobile applications, virtual care, and wearables, hold transformative potential for diagnosis, treatment, and care management. However, their successful implementation and uptake in real-world settings remain challenging, often referred to as the “last mile” problem. This issue arises when digital technologies, despite being designed and developed, fail to be successfully implemented, sustained, and scaled over time.
The Importance of a User-Centered Approach
Understanding the Last Mile Problem
The last mile problem in digital health mainly arises from a lack of a user-centered approach by technology designers. Technologies failing to meet end users’ expectations lead to poor user experience, low uptake, and, ultimately, wasted resources. These issues have been known for a while; however, healthcare stakeholders often find it tough to learn from previous digital health implementation mishaps. Lack of user-centered design causes a disconnect between technology solutions and their practical application in clinical settings.
Clinicians frequently experience frustration with digital tools designed without their input. When technologies don’t align with their workflow or provide meaningful support for their tasks, the result is minimal usage and dissatisfaction. The gap between what designers imagine and what is needed in actual healthcare environments often leads to these tools being underutilized or even abandoned. Therefore, understanding and addressing the last mile problem requires a shift toward user-centered design processes that engage clinicians from the start.
Clinicians as Key Stakeholders
To effectively tackle the last mile challenge, clinicians must be positioned as leading players in the design and implementation of healthcare technologies. Their involvement ensures that developed systems are relevant, usable, and have better long-term success. Clinicians’ unique insights into everyday clinical problems and operational processes are invaluable for creating technologies that genuinely enhance patient care and streamline workflows. By collaborating with technologists, clinicians can help create solutions that minimize disruption, foster trust, and improve care delivery.
Such collaborative efforts not only improve the functionality of health technologies but also ensure these tools align with clinicians’ needs. Including clinicians early in the design phase enables tailoring solutions to address specific clinical challenges and integrate seamlessly into existing workflows. This clinician-led approach encourages adoption, as end users are more willing to engage with tools they helped create and perceive as beneficial to their practice, ultimately leading to better success in digital health initiatives.
Evidence of Success
A systematic review by experts revealed clinician involvement in designing Clinical Decision Support (CDS) systems led to significantly better integration with existing workflows, higher adoption rates, and notable reductions in medication errors. This evidence underscores the crucial role clinician engagement plays in the successful implementation of digital health solutions. When clinicians are included in the design process, the resulting tools not only meet practical needs but also enhance patient outcomes by providing accurate and timely support.
The success stories showcase the tangible benefits of a clinician-centered approach. By embedding tools into the natural workflows of healthcare professionals, health institutions can reduce the cognitive load on clinicians and ensure these technologies are used consistently. Furthermore, this involvement helps identify potential pitfalls and usability issues early, allowing for adjustments before large-scale deployment. This iterative design process, driven by clinician feedback, leads to robust and effective solutions that truly support clinical needs.
Consequences of Excluding Clinicians
Case Study: EPAS Hospital Software
The rampant issue of technology underperforming due to the exclusion of clinicians is exemplified by the implementation of the Enterprise Patient Administration System (EPAS) hospital software in Australia. Staff heavily criticized EPAS for being inefficient and adding to their workloads, primarily due to the absence of meaningful clinician engagement during its design and implementation stages. The EPAS debacle highlights the significant drawbacks of neglecting clinician input, from poor usability to increased operational burdens that ultimately hinder healthcare delivery.
The EPAS case serves as a cautionary tale, demonstrating how excluding key stakeholders like clinicians can derail digital health initiatives. Without their input, the developed systems may become complex, non-intuitive, and more of a hindrance than a help. The inability to streamline or integrate with existing workflows further exacerbates these issues, leaving staff feeling frustrated and overburdened. As a result, such initiatives fail to deliver their intended benefits and fail to gain traction among those who are supposed to use them daily.
Lessons from the COVID-19 Pandemic
The deployment of machine learning tools during the COVID-19 pandemic further illustrates the pitfalls of technology-centric approaches without clinician input. Many algorithms that showed promise in retrospective studies failed to deliver anticipated outcomes in real-world clinical settings. This shortfall was significantly attributed to the lack of early clinician involvement. As a consequence, these tools struggled with poor integration into existing workflows and did not meet the practical needs of frontline healthcare providers.
This disconnect during the pandemic emphasized the necessity for early and ongoing clinician engagement in technology development. The absence of such involvement often results in tools that fail to address the real-world challenges administrators and clinicians face. By integrating clinicians from the beginning, technology developers can ensure the solutions are both practical and effective, ultimately improving patient care and operational efficiency. Building this collaborative foundation is critical to adapting technological advances to the complex dynamics of actual healthcare settings.
Building Trust Through Involvement
Early and sustained involvement of clinicians in the technology development process fosters trust in the tools they help create. Recognizing the practicality, effectiveness, and alignment of these solutions with patient care goals is essential for their successful implementation and uptake. When clinicians see the tangible impact of their contributions, they are more likely to advocate for the adoption of these tools within their institutions, leading to broader acceptance and long-term utilization.
Building this trust is pivotal for the effective deployment of digital health technologies. When clinicians feel a sense of ownership and validation through their involvement, they become champions of these innovations. This advocacy, in turn, supports smoother transitions during rollouts and encourages wider adoption across different levels of healthcare settings. Thus, the engagement of clinicians not only enhances the usability and effectiveness of digital health tools but also ensures they are embraced and deeply integrated into everyday clinical practice.
Supporting Clinicians as Co-Designers
Role of Healthcare Leaders and Policymakers
Healthcare leaders and policymakers have a critical role in enabling clinicians to adopt a co-designer role in the development of healthcare technologies. It is essential for healthcare leaders to provide clinicians with protected time to collaborate with vendors and ensure that technologies being designed are both meaningful and fit for purpose. The findings from the COVID-19 pandemic demonstrated that clinicians often lacked sufficient time for participating in the co-design of Health Information Technologies (HITs), primarily because maintaining patient care was a priority.
Healthcare organizations must reassess their priorities to balance patient care demands with the need for clinician participation in technology design. By allocating protected time and creating structured opportunities for co-design, leaders can empower clinicians to contribute their expertise effectively. This collaboration promises more responsive and adaptable technologies, ultimately benefiting patient care. Policymakers should support these initiatives through guidelines and incentives that encourage meaningful clinician involvement in the design phase.
Addressing the Middleman Problem
Health service organizations need to improve access for vendors to end users directly, without creating a middleman problem. Often, information technology staff intended to bridge the gap between clinicians and vendors inadvertently become gatekeepers, leading to critical user requirements getting lost in translation. Clear outlines of time commitments and expectations around user engagement are crucial for successful co-design and end-user involvement. Health service organizations must ensure transparency and direct communication channels between vendors and end users.
Facilitating direct engagement helps maintain the integrity of user feedback and ensures that the clinicians’ needs are accurately conveyed and addressed. Health organizations should develop frameworks that allow for seamless collaboration between vendors and clinicians, avoiding unnecessary intermediaries. Clear and realistic expectations surrounding time commitments are vital to this process, ensuring that clinicians can provide valuable input without compromising their clinical responsibilities. This balanced approach fosters a more collaborative and effective design process.
Government and Policy Support
Governments and policymakers need to offer timely policies supporting the ongoing implementation of Health Information Technologies (HITs). Existing guidelines, such as the Safety Assurance Factors for EHR Resilience (SAFER) guides and the EHR developer code of conduct, advocate for healthy relationships between vendors and end users, primarily targeting electronic health records (EHRs). However, these guidelines often fall short in mandating user participation in design, allowing technologies to be built based on vendors’ assumptions rather than clinicians’ real-world experiences.
To truly support effective HIT implementations, policymakers must mandate clinician involvement in the design process. By enforcing inclusive guidelines, such as those provided by the Australian Digital Health Agency, and requiring vendors to demonstrate efforts to understand end users’ workflows, governments can ensure that technology solutions are better aligned with clinical practices. Implementing these policies can pave the way for more effective digital health tools that enhance patient care and operational efficiency, bridging the gap between design and practical application.
Learning from Past Failures
Publication Bias Against Failures
Learning from digital health implementation failures is vital for advancing the field, yet publication bias against these failures significantly restricts the ability to share important lessons openly. Many large-scale HIT implementations around the world do not meet their budgets, scope, timelines, or expected benefit targets, but these failures are seldom published in academic journals. This reluctance to publish failures stems from a culture that stigmatizes and rejects unsuccessful projects, leading to a lack of knowledge sharing.
Addressing this issue requires a cultural shift within the academic and healthcare communities. Researchers and practitioners should be encouraged to document and share their experiences with failed implementations to provide valuable insights for future endeavors. By promoting a more open and transparent environment, the healthcare industry can learn from past mistakes, refining strategies, and improving the outcomes of digital health initiatives. This shift toward transparency is crucial for the continuous improvement and advancement of healthcare technologies.
Promoting Open Knowledge Sharing
Applied Clinical Informatics (ACI) stands out as an exception, actively promoting the publication of case studies on IT failures. However, this practice is not widespread in other medical informatics journals, which often deter researchers from documenting failed projects. Stigmatizing and rejecting publications about HIT failures discourage researchers from sharing their experiences, thereby hindering knowledge sharing that could benefit future efforts. Encouraging the publication of such studies can foster a more comprehensive understanding of the challenges faced in digital health implementations.
Creating platforms and encouraging journals to accept and publish detailed accounts of failures can drive open knowledge sharing. This openness allows for the accumulation of practical insights and lessons learned, which can be applied to improve future projects. By destigmatizing failure and promoting candid discussions about what went wrong, the healthcare community can develop more resilient and effective digital health technologies. This inclusive approach ensures continuous learning and progress, ultimately benefiting patient care and healthcare system efficiency.
Conclusion
Digital health technologies, such as artificial intelligence, analytics, dashboards, web portals, mobile apps, virtual care, and wearables, have the power to revolutionize diagnosis, treatment, and care management. These innovations promise significant improvements in health outcomes and efficiency. Nevertheless, their successful implementation and adoption in practical, real-world settings face significant hurdles, often referred to as the “last mile” problem. This challenge arises when digital technologies, despite being expertly designed and developed, struggle to be effectively implemented, maintained, and scaled over time. These difficulties can stem from a variety of factors, including resistance to change among healthcare providers and patients, interoperability issues with existing systems, and a lack of necessary infrastructure or training for optimal use. Furthermore, there can be concerns about data privacy and security, which may hinder widespread acceptance. Overcoming the “last mile” problem is crucial for these technologies to achieve their full potential in transforming healthcare.
