A Concept Analysis of Person-Centered Care and Personalized Medicine/Health Care as a Basis for an Indivualized AI and Big Data Driven Nursing and Health Care

Research Aricle

J Fam Med. 2024; 11(4): 1364.

A Concept Analysis of Person-Centered Care and Personalized Medicine/Health Care as a Basis for an Indivualized AI and Big Data Driven Nursing and Health Care

Hasseler M1,2*

1Faculty of Health Science, Ostfalia University of Applied Sciences, Germany

2Visiting Scholar, School of Nursing, University of Maryland, Baltimore, USA

*Corresponding author: Hasseler M Faculty of Health Sciences, Ostfalia University of Applied Sciences, Germany. Tel +49 5361 8922 23170 Email: [email protected]

Received: March 08, 2024 Accepted: April 29, 2024 Published: May 06, 2024

Abstract

Background: Digitalization and artificial intelligence are playing an increasingly important role in healthcare. But so far, there is a lack of evidence as to whether the promises can be kept and whether patient-centredness and the personalization of healthcare can be implemented. Up to now, discussions have been dominated by personalized medicine and precision medicine. The impression arises that the biopsychosocial approach is not being integrated into the discussion about AI and big data analysis in healthcare.

Methodology: A concept analysis of personalized medicine and person-centered care according to Rogers (2000) is carried out. The basis is a structured literature review.

Results: Personalized medicine initially appears to have a purely biomedical approach, but evidence is emerging in the literature that further data about patients is necessary if tailor-made healthcare is to be enabled. Person-centered care is based on various scientific theoretical approaches, but aims to perceive and care for the patient in a biopsychosocial way.

Conclusion: The concept analysis makes it clear that the concepts of personalized medicine and person-centered care should be merged for AI development and big data analysis so that the relevant characteristics can be incorporated into this development and the data relevant to patients can be collected and evaluated using a bio-psycho-social approach.

Keywords: Personalized medicine; Precision medicine; Person-centered nursing; Person-centered health care; AI; Digitalization

Introduction

Digitalization in the healthcare sector, particularly in the form of artificial intelligence and big data, is an increasingly important topic. Wiljer et al [58] assume that a decisive point has now been reached in healthcare where AI and its algorithms can be used safely, including to improve the quality of healthcare. The extent to which AI is changing prevention, diagnosis, disease care and predictive healthcare is considered significant [58]. According to Couture et al. [12], AI in healthcare has the potential to improve population health due to the large-scale use of AI-based diagnostic tools, use of AI-based health apps, especially if they are developed specifically for defined populations and their health, or if AI is used for automated population health screening. The great promise of AI in healthcare and health services research is that, due to the reliance on big data, it enables an understanding of what works for whom and when [36]. A scoping review by Sharma et al. [46] shows that most of the research on AI in health care covers very different clinical settings and focuses more on healthcare providers and less on patient or people in need of care. Furthermore, they predominantly examine clinical care provided by physicians rather than other healthcare professionals. Similarly, the AI tools examined so far support more human decision-making and and do not inquire about people's autonomy in the healthcare system. Furthermore, the authors point out that most studies examine very defined interventions in health care that have so far had little to do with the clinical reality or the patients' real lives, needs and preferences [46]. The scoping review shows that research on AI in healthcare is predominantly conceptual in nature, often commentaries, opinion pieces and conceptual frameworks that address relevant questions but do not provide any empirical evidence. The evidence of empirical evidence of AI in healthcare is narrow and underdeveloped, limiting the potential for generalization of findings to healthcare practice and further development of methodological approaches. In a slightly similar way, von Gerich et al. [58] summarize their results of their scoping review. Accordingly, most developed and trialed AI technology was only “working as intend or showing potential”. The clinical benefit of the new AI-based tools has not been validated in the studies. In addition, only four studies have examined the relationship between technological functionality and the end-user perspective. Many studies on AI in nursing and healthcare do not specify the target group or the intended benefits [58]. Gama et al. [16] question the qualita of data used for AI and big data since it depends on many factors and local conditions that have not yet been sufficiently researched, such as the availability of technologies, documentation systems, healthcare professionals, patients, family members and so on. However, there are also a number of requirements if AI is to be used in healthcare [57]. Firstly, this includes ensuring that people retain autonomy over all healthcare processes and their data. The right to informed decision-making and the protection of privacy and confidentiality should be preserved. This also includes protecting human well-being, personal safety and the public interest. Other requirements include transparency, comprehensibility and accountability for people in healthcare. Furthermore, inclusion and equity are essential criteria for an individual-centered AI. This indicates that no one should be excluded from the development of AI and big data analysis in healthcare. WHO [57] emphasizes that AI in health care must be independent of age, gender, income and other social determinants [57].

Background

Current literature discusses the potential of artificial intelligence in nursing, e.g. how it will change nursing care [38]. It is assumed that artificial intelligence can, among other things, improve and support the organization of patient processes and treatment plans and/or provide all relevant information that doctors and nurses need to make appropriate decisions and/or assist with repetitive or routine tasks in healthcare or medication management [35,50]. It is hypothesized that AI will take over routine tasks in the healthcare process, such as providing patient information to nurses and doctors, creating care plans, performing medication management, etc. [35]. Other authors assume that the integration of artificial intelligence into nursing care will change the nurse-patient interaction. There is discussion in the international literature that AI-based processes in nursing can support clinical decisions or even generate automatic warning systems and thus also systematically support the nursing workflow and enable personalized patient care [45,8]. These technologies need to be examined in more detail, in particular how they change nursing care processes qualitatively and quantitatively, how they change every day processes and whether and how the outcomes of nursing care are influenced [41]. In their consensus statement, Ronquillo et al. [40] elaborate on the positive potential of AI in nursing care as well as possible negative influences, e.g. that system-related biases can be reflected in the AI algorithms, including social and health inequalities. The question also arises as to how these AI-based systems improve patient safety and outcomes and to what extent they contribute to the prevention of nursing errors [45]. For example, there are opportunities for clinical pathways, disease progression or the prediction of deterioration or risks to be taken over by AI, thus paving the way for nursing action [42]. Other potential uses are also seen in the fact that AI can combine diagnosis with scientific evidence and guidelines and lead to personalized treatment plans [31]. The qualitative survey by Rony et al. [41] shows that the nursing professions surveyed generally consider the implementation of AI to be very positive and supportive of nursing care. In principle, they anticipate that AI in nursing care will, for example, enable tailor-made healthcare, detect any deterioration in health and improve the well-being of patients through this personalized or person-centered AI-based support. In another paper Rony, Parvin & Ferdousi [42] claim, that AI will “revolutionize nursing practice and enhance patient care” since “real-time insights”, more accurate diagnosis and personalized care plans will be possible.

However, some recent publications such as Rony et al. [41], Rony, Parvin & Ferdousi [42] and Ronquillo et al. [40] emphasize that nursing professions in particular, with their more holistic approach in health care have a special role to play in the process of developing and implementing AI in health care. The above outlined debate on the development of AI in healthcare has shown that, on the one hand, AI offers many promises for improving and transforming nursing and healthcare. On the other hand, however, it appears that the perspective of the people, the patients, as well as their needs, wishes and factors that shape their lives and which are relevant in the provision of high-quality and person-centered healthcare, are missing. Shang (2021) states that the topic of AI and nursing is still in its infancy. This can be recognised, among other things, by the still low degree of application of AI in practical nursing care as well as in research publications on AI and nursing, which remains very underrepresented. Buchanan et al [8] question how person-centred nursing care can be ensured with the development of AI and big data analysis. This includes not only ensuring that a nurse-patient interaction and relationship continues to take place and is established, but also that the relevant information of a person-centred AI development is included. There is some discussion in the literature that there is a risk in AI that systemic biases or social inequalities will be perpetuated, thus making AI less accurate [33]. So, after all, the algorithms or AI-based decisions, diagnoses, treatment suggestions and more, require person-cenetred information so that a bio-psycho-social health care can be made in the interests of the individual patient. A purely personalized medical AI will lack relevant information.

The impression is that the development and testing of AI in the health and nursing care of patients is currently directed from a predominantly one-dimensional and rather biomedical and monodisciplinary perspective. As early as 2018, Abdul-Hasn & Kenny [1] formulated that the developments in personalized medicine should go hand in hand with the huge genetic findings and databases with electronic documentation systems that have been determined to date. The two authors view the electronic patient record as a real-time, patient-centered digital documentation system that is managed responsibly by healthcare professionals. They are assuming that this combination could considerably improve the clinical healthcare of patients.

Aims

Although almost all publications on AI in healthcare state the potential for better and more accurate diagnosis, personalized and person-centered as well as tailor-made planning and implementation of healthcare, the literature lacks the conceptual foundations of how the approaches of individualized, person-centered and personalized nursing healthcare should be realized with AI. So far, the published literature has not specified the basis on which the required data should be derived for AI development if all the promises of AI in healthcare and nursing care are to be achieved.

In this light, the aim of this conceptual analysis is to clarify the meaning of the terms personalized medicine and person-centered nursing/healthcare so that they can be used as a basis for AI development and big data analysis in order to fill the gaps mentioned above. The aim is to provide the opportunity to define the relevant aspects of person-centered and personalized health care in such a way that the pertinent characteristics can be integrated into AI development and big data analysis. This concept analysis is intended to provide a basis for the question of how artificial intelligence and big data can be used to achieve person-centered and personalized nursing and health care that meets people's needs and requirements and can support the process of interprofessional and cross-sectoral healthcare.

Methods

Concept analysis is seen as a relevant basis for further research, since a precise description of a concept provides a theoretical basis for further research and practice. Nuopponen (2010:4) defines concept analysis as follows: "Concept analysis could be basically defined as an activity where concepts, their characteristics and relations to other concepts are clarified." A concept analysis analyses and presents the state of knowledge for a concept of interest. Guiding questions include: How is the concept defined, described? Is it used in defined areas? How is it used? What predictions, explanations do the concept enable? In which generalizations or descriptions/descriptions of patterns does the concept appear? If the theory/concept makes a difference in practice, what difference does it make in use?

Concept analysis is therefore useful to approach conceptual and terminological problems and thus understand them as part of "terminology work". Since the concepts of personalized medicine and person-centered care have evolved in understanding in recent years, but are also so fundamental to enabling AI and big data analysis in healthcare that is useful to people in need of healthcare, it makes sense to analyze both concepts with the help of a concept analysis as a basis for further AI development and research,

Concept analysis is therefore an analytical and synthesizing approach to define and describe relevant concepts [39]. There are various methodological approaches to concept analyses. In this thesis, the approach of Rodgers [39] is applied.

Following the methodology proposed by Rodgers (39), the following steps were carried out (Table 1).

• While conducting the core analysis, the following questions were addressed to the included publications [52].

Surrogate terms: Do other words say the same thing as the chosen concept? Do other words have something in common with the concept?

Antecedents: Which events or phenomena have been associated with the concept in the past?

Attributes: What are the concept’s characteristics?

Examples/Use of Concept: Are concrete examples of the concept described in the data material?

Consequences: What happens after or as a result of the concept?

This approach was supplemented by the additional items definition of the concept according to Walker & Avant [56]: What kind of definitions offer the authors on the concept?

The results of this core analysis can be seen in Table 2 &3.

Literature Search

Following the approach of Rodgers [52], the literature search is broad-based. This is also done against the background that the terms "personalized medicine", "personalized nursing" and "person-centered nursing" are very wide-ranging.

The literature search was conducted iteratively. Initially, general literature sources are sought in order to gain an overview of the current state of discussion on this topic. It can be seen that all terms are sometimes very generic in publications, but then again very specifically related to population groups, settings and diseases. Some already refer to digitalization and AI development, especially in relation to personalized medicine. Based on these findings, the decision was made to define the following inclusion and exclusion criteria

Inclusion criteria: Personalized medicine, precision medicine; personalized nursing, person-centered nursing, person-centered medicine in the title. The articles should deal with the concepts in general. The search is initially conducted for the last 5 years. If it becomes apparent that relevant publications date back a further 5 years, the search is carried out a further 5 years into the past. Furthermore, not only studies are included, but also commentaries, opinion papers and grey literature, i.e. also publications that were produced on behalf of e.g. ministries, health insurance companies or similar and were created as pdf or other publication options. Only publications in English are integrated.

Exclusion criteria: All contributions that deal with specific population groups, settings, diseases in the context of the selected concepts. In addition, no articles older than 10 years will be included.

In accordance with the iterative procedure, the PubMed/Medline database is first searched after the first orientation research for each term, the title and abstract of the articles searched are screened to determine whether they are suitable for the concept analysis and transferred to the literature management program CITAVI. At the same time, an evaluation is carried out in accordance with the above-mentioned areas of analysis and entered into the tables. The search terms are then entered into the Livivo database and the articles are screened in the title and abstract to determine whether they fulfil the inclusion criteria and transferred to CITAVI. The same procedure is followed with Psycho-Info Cinahl and Embase. For each search, the included articles are analyzed according to the criteria listed in table 2 & 3. The advantage of this iterative search and analysis process is that any articles that already exist and have been analyzed are immediately recognized and no longer included. Furthermore, it is also possible to compare the analysis results, recognize overlaps or differences or any further developments in order to be able to assess that all relevant findings on both concepts have been included and data saturation can be determined. This is followed by a snowball procedure, i.e. if it becomes apparent within an analyzed article that a source is cited that has not yet appeared in the database search but appears to be relevant, it is manually searched for and researched and assessed as to whether it meets the inclusion criteria. Due to the aim and intention of this publication to perform a concept analysis, it is not necessary to apply a method to assess the quality of the included publications.

In the end, the concept analysis includes the publications that contain descriptions and definitions of the desired concepts, that make statements about how these concepts are used, that describe the elements, attributes and characteristics of the concepts.

The Following PRISMA Flow Charts Show the Iterative Process of the Literature Search

Prisma Flow Charts 1 & 2

Citation:Hasseler M. A Concept Analysis of Person-Centered Care and Personalized Medicine/Health Care as a Basis for an Indivualized AI and Big Data Driven Nursing and Health Care. J Fam Med. 2024; 11(4): 1364.