Management researchers have observed that health systems have often been encouraged to adopt a product dominant logic for quality process improvement (QPI) initiatives. When health systems view QPI only through a product dominant lens the focus on processes, actions, and outputs risks neglecting key elements within healthcare. Specifically, relationships, outcomes, and individual patient preferences. Reimagining healthcare as a coproduced or a service dominant approach will aid professionals to adopt patient-centered QPI initiatives that reflect the values of their patients and that contribute to the health of individuals and populations (Batalden, 2018).
Justifiable healthcare is an approach that helps health systems align process improvement with technological advancements and patient-centered practices. The model strives for evidence based, efficient, just, and sustainable healthcare. Justifiable healthcare differs from best-practice care due to the requirement of the underlying reasoning is transparent to all stakeholders and decisions are based on a systematic and holistic approach. This differs from authoritarian approaches where best practice is mostly based on biological and technical aspects of care. In traditional models' decisions are often made without regard of societal effects, psycho-social issues, or patient values (van Biesen et al., 2021).
Shared decision-making (SDM) is considered the preferred framework for decisions at the individual patient level. The three pillars in the SDM process are the evidence base, clinical expertise of the healthcare worker, and values of the patient (van Biesen et al., 2021). In recent years artificial intelligence (AI) is starting to play a significant role in the SDM process. AI often supports clinical decision support systems (CDSS), helping clinicians in the diagnosis of disease and treatment decisions (Amann et al., 2020).
To fully exploit opportunities for AI to increase patient outcomes by improving detection, prevention, and treatment of diseases health systems must address issues of data security, patient consent, and autonomy. From a legal point-of-view three core fields must be prioritized for the wide scale adoption of AI for SDM. Specifically, informed consent, certification/approval of AI as medical devices (i.e., FDA and Medical Device Regulation) and liability issues (Amann et al., 2020).
Studies show health systems often focus on three specific practices to accelerate QPI innovation. These include spread, sustainability and scale-up of healthcare innovations. Organizational QPI reforms usually fail to overcome the change-resistant nature of healthcare systems. Levers needed to accelerate innovations include commitment of front-line managers and providers, an emphasis on population needs, supportive policies or incentives, investment in organizational capacity, patient participation, and evidence-based decision-making (Côté-Boileau et al., 2019).
Health systems frequently use routinely collected health data (RCD) to inform QPI initiatives. RCD is commonly obtained from electronic health records, insurance claims, wearables, or apps. Utilization of RCD is driven by the belief that valuable information within the data will improve medical decision-making, assist regulatory approval, and reduce costs. However, this innately hinges on data quality, which is often compromised by missing, incorrect entries, misguided incentives, or server shutdowns. Establishing shared data networks pose added challenges to interoperability and data coordination. Data coming from different sources should be standardized and aligned to common data model before use for analysis (van Biesen et al., 2021) of QPI initiatives.
It is important for health systems to understand how clinical validation is measured. Providers must also take steps to avoid AI bias which leads to systematic errors and deviation from the predictive functions of AI software. Consequently, explain-ability plays an influential role in the clinical setting. Explain-ability facilitates the resolution of disagreement between AI systems and clinicians. If used properly, explainable AI decision support systems can contribute to patients feeling more knowledgeable and it could also promote more accurate risk assessments. This may enhance patient motivation to engage in SDM and act upon risk-relevant information (Amann et al., 2020).
Current SDM and QPI approaches assume a level of eHealth literacy of patients as a pre-requisite participate in care delivery methods. The design of eHealth software and devices for engaging end-users in chronic disease self-management often inadvertently re-enforces inequalities experienced by vulnerable populations. There are large numbers of patients who, because of a variety of situations are not able to engage with these processes or when they do, actual benefits are unidentifiable due to their circumstances. Therefore, a diversity sensitive approach to eHealth should be implemented to include vulnerable and underserved patient populations in designing solutions (Botin et al., 2020).
Healthcare leaders face several QPI challenges including the pace of change, supervisory structure, complexity of data infrastructure, over-reliance on technology-based communication, and lack of leadership training. The organization of team members into silo-ed supervisory structures or service lines makes it difficult to develop shared priorities and impedes communication. While pace of change frequently outpaces the development of application guidance for new clinical policies, processes, or practices. Lack of data infrastructure prevents effective leadership due to fragmentation and volume of data that is available to leaders to make informed decisions. The over-reliance on technology in place of face-to-face communication increases the likelihood of miscommunication. Finally gaps in training available to healthcare leaders impairs the development of effective QPI teams (Abraham et al., 2021).
Instead of propagating a standard vision of process improvement, leaders should focus on ensuring that all stakeholders involved in/impacted by the QPI initiative can answer why they are committed to the innovation. Lags in momentum and interruptions should be expected, but it is vital that stakeholders believe that the innovation adds value to their work and the quality of care they provide to patients. Focusing on the why of a QPI innovation requires sharing data on the advantage of the innovation, highlighting promising results from other systems, and monitoring/measuring performance to see improvement (Côté-Boileau et al., 2019).
Any significant QPI innovation is a source of destabilization and requires commitment from leadership to develop policies to align the innovation with existing protocols to lessen the adverse effects of change. Innovation work is supported by leaders and policies that promote alignment between the qualities of the innovation and system operations and regulations (Côté-Boileau et al., 2019).
Forcing QPI innovations within a short-term agenda hinders sustainability. The focus should be on what people do, instead of what they should be doing. Focusing on what people do, rather than on what they should do, identifies the sources of value and achievability in innovation work. The aim of leadership is for stakeholders to find a way to move a QPI innovation forward that takes their values into account. Strategies to integrate the values stakeholders include forums/meetings that foster dialogue and problem solving, as well as openings for communication between participants from all levels (Côté-Boileau et al., 2019).
Studies have shown that soft skills are especially important for healthcare leaders involved in QPI innovations. Primary among soft skills is the ability to communicate, which help leaders coordinate clinic staff and create a common culture or mission among team members. Soft skills related to mediation and conflict management are also critical for leaders guiding QPI initiatives. Responding positively or proactively to adversity and stress, often referred to as adaptability, resilience, or flexibility, has also proven to be an important soft skill that helps healthcare leaders to overcome QPI challenges (Abraham et al., 2021).
Traditional healthcare analytic approaches often do not support innovation pertinent to clinical and non-clinical practitioners. Design science research (DSR) support health information systems by designing current solutions for effective design methodology. DSR methods can assist health systems through conveying problem definition, suggesting solution methods, validating resolution activities, guiding relevant appraisal, and dissemination of practical information (Miah et al., 2019). DSR can also help providers proficiently utilize AI-based EHR data mining which provides a valuable method to supplement/replace current healthcare data collection methods. Studies show that data obtained from EHRs by AI showed an impressive accuracy of 87.1%. EHR data mining would allow health systems to effectively reallocate resources and reduce performance costs (van Dijk et al., 2021).
The advancement of biotechnologies requires health systems to process a massive amount of data. AI-based analytical tools are essential to sift through all this information to detect patterns and give meaning to the outputs of many procedures currently used in biomedical health care. AI based search engines can retrieve and visualize all the available data on certain molecules, therefore increasing provider understanding of the molecular basis of disease processes. This can help identify patients in whom certain therapeutic interventions will benefit (van Biesen et al., 2021). Such exorbitant amounts of data require the adoption of technologies such as a centralized research data warehouses to store all data necessary to obtain a comprehensive picture of the health of populations before analysis of actionable insights can occur (Cecchetti et al., 2020).
Several steps are necessary for health systems to move toward implementing data warehouse systems. First the integration of data from different types of medical settings such as hospitals, clinics, and specialty centers. Second the linkage of financial data with clinical data which is crucial to high-quality care and positive economic outcomes. Finally, the integration of other factors of health such as environmental, social, and spiritual factors to create longitudinal health data across the care continuum (Cecchetti et al., 2020).
Ultimately, health belongs to the individual. It is their responsibility and difficult to outsource even to a professional (Batalden, 2018). With that in mind it is crucial for providers to develop easily understandable presentation standards that integrate of all available evidence of specific conditions, helping patients to make decisions that are as close as possible to their values. To truly improve outcomes healthcare workers must first elicit the values and life goals of patients before considering treatments. No interventions should be administered to attain outcomes that have no value to patients (van Biesen et al., 2021).
Active patient involvement in SDM illuminates the resources and social support that patients contribute to the process and their health (Batalden, 2018). Digital technologies gives patients access to continuous and in-depth reporting of their symptoms and experiences in a more feasible, sustainable and cost-effective way. Digital symptom reporting and AI-based data collection can have a positive impact on the quality of healthcare with reduced symptom distress through better self-management, improved health-related quality of life, and higher quality of interaction with healthcare professionals (van Biesen et al., 2021).
QPI innovations should be exemplified by quality, safety, and good benefit for money spent to provide value. Health systems can sustain QPI initiatives by integrating system performance with learning into a system that reflects active learning and never-ending process improvement. Successful health systems focus on the development of standardized responses to common needs, tailored responses to specific needs, and adaptable responses to emergent needs (Batalden, 2018).
The regular collection of outcome data enables organizations to be adaptable to the various needs of the communities they serve. It also presents opportunities to evaluate and clarify the performance of healthcare providers at level of the individual provider and the meso level of the organization. The further implementation of Big Data/AI approaches would allow for collection of the data necessary to produce evaluations from different sources and turning them to a meaningful framework (van Biesen et al., 2021) that strengthens partnership between health provider and patient which ultimately leads to better individual and population health.
Abraham, T. H., Stewart, G. L., & Solimeo, S. L. (2021). The importance of soft skills development in a hard data world:Learning from interviews with healthcare leaders. BMC Medical Education, 21(1). Link
Amann, J., Blasimme, A., Vayena, E., Frey, D., & Madai, V. I. (2020). Explainability for artificial intelligence in healthcare: A multidisciplinary perspective. BMC Medical Informatics and Decision Making, 20(1). Link
Batalden, P. (2018). Getting more health from healthcare: Quality improvement must acknowledge patient coproduction—an essay by paul batalden. BMJ, k3617. Link
Botin, L., Bertelsen, P. S., Kayser, L., Turner, P., Villumsen, S., & Nøhr, C. (2020). People centeredness, chronic conditions and diversity sensitive ehealth: Exploring emancipation of the 'health care system' and the 'patient' in health informatics. Life, 10(12), 329. Link
Cecchetti, A. A., Bhardwaj, N., Murughiyan, U., Kothakapu, G., & Sundaram, U. (2020). Fueling clinical and translational research in appalachia: Informatics platform approach. JMIR Medical Informatics, 8(10), e17962. Link
Côté-Boileau, É., Denis, J.-L., Callery, B., & Sabean, M. (2019). The unpredictable journeys of spreading, sustaining and scaling healthcare innovations: A scoping review. Health Research Policy and Systems, 17(1). Link
Miah, S. J., Gammack, J., & Hasan, N. (2019). Methodologies for designing healthcare analytics solutions: A literature analysis. Health Informatics Journal, 26(4), 2300-2314. Link
Moreno-Calderón, A., Tong, T. S., & Thokala, P. (2019). Multi-criteria decision analysis software in healthcare priority setting: A systematic review. PharmacoEconomics, 38(3), 269-283. Link
van Biesen, W., Van Der Straeten, C., Sterckx, S., Steen, J., Diependaele, L., & Decruyenaere, J.(2021). The concept of justifiable healthcare and how big data can help us to achieve it. BMC Medical Informatics and Decision Making, 21(1). Link
van Dijk, W. B., Fiolet, A. T., Schuit, E., Sammani, A., Groenhof, T. J., van der Graaf, R., de Vries, M. C., Alings, M., Schaap, J., Asselbergs, F. W., Grobbee, D. E., Groenwold, R. H., & Mosterd, A. (2021). Text-mining in electronic healthcare records can be used as efficient tool for screening and data collection in cardiovascular trials: A multicenter validation study. Journal of Clinical Epidemiology, 132, 97-105. Link