Impact statement

Precision care encompasses a person-centered approach to precision medicine to inform decision making and care planning by taking multimorbidity, functional status, values, goals, preferences, social and societal context into account. This article examines challenges that need to be addressed to realize the promise of precision medicine in the context of multimorbidity. It also discusses how partnerships across the research enterprise; between researchers, practice, and policymakers; with patients, caregivers and communities; between clinical medicine and public health; and across sectors could advance effective real-world adoption of precision medicine.

Continuum of translational research and multimorbidity

There is a long road from scientific discovery to routine application in clinical practice and societal benefit. The uptake of therapeutic advancements often takes decades. To foster and accelerate their adoption, the full continuum of translational research needs to be addressed: developing new approaches, demonstrating their usefulness, and disseminating and implementing their findings, engaging patients throughout the process (Figure 1). This encompasses basic science, preclinical and clinical research, followed by implementation into practice and ultimately impact on public health. Table 1 illustrates how multimorbidity can be considered in each step of the process from animal models, to phase 1 studies, clinical trials, effectiveness studies in real-world practice, and assessments of population health impact.

Table 1. T0–T4 research in the context of multimorbidity

Note: T0–T4 adapted from Tufts Clinical and Translational Science Institute (https://www.tuftsctsi.org/about-us/what-is-translational-science/)

Figure 1. T0–T4 continuum of translational research to improve individual and population health.

Evidence to inform precision care

When considering evidence to inform precision care, the primary focus is on rigorously controlled randomized trials examining the efficacy of precision therapeutics targeting specific genotypes. The evidence needed to inform precision care is broader and richer than effectiveness studies evaluating new diagnostic and therapeutic modalities (Abou-el-Enein et al., Reference Abou-el-Enein, Angelis, Appelbaum, Andrews, Bates, Bierman, Brenner, Cavazzana, Caligiuri, Clevers, Cooke, Daley, Dzau, Ellis, Fineberg, LSB, Gottschalk, Hamburg, Ingber, Kohn, Krainer, Maus, Marks, Mummery, Pettigrew, Rutter, Teichmann, Terzic, Urnov, Williams, Wolchok, Lawler, Turtle, Bauer and Ioannidis2021). Randomized clinical trials (RCTs) typically assess interventions in highly selected and controlled populations with few comorbidities other than the conditions of interest. RCTs report outcomes from the enrolled populations, and at times from pre-specified subgroups. Increased attention is being directed at the need to consider the heterogeneity of treatment effects and the need to understand factors associated with intervention effectiveness in each individual (Kent et al., Reference Kent, Steyerberg and Van Klaveren2018). Evidence is also required to understand the impact of interventions in real-world settings, where these interventions are administered to those living with multiple chronic conditions. Furthermore, social factors that affect access, decision making, and outcomes need to be considered. Assessing evidence in clinically and socioeconomically diverse populations would help us to better understand the long-term benefits and harms and potential trade-offs of interventions in patient populations commonly encountered in the real world, for example, those with diabetes, heart failure and/or depression or a combination of these conditions. Study designs that can provide the needed evidence include pragmatic trials and registry studies. Adoption of a core set of harmonized measures will allow for pooling of data and metanalysis to accelerate evidence generation (Beckmann and Lew, Reference Beckmann and Lew2016; Leavy et al., Reference Leavy, Schur, Kassamali, Johnson, Sabharwal, Wallace and Gliklich2019).

In addition to clinical evidence of effectiveness, additional evidence is needed to inform decision making about when, for whom, and how to use diagnostic and therapeutic advancements. Evidence examining the varied preferences of patients and clinicians can help inform shared decision making (SDM) to facilitate precision care (Umscheid, Reference Umscheid2009), particularly when these preferences reflect real-world concerns, such as polypharmacy and financial toxicity of medical care (Schnipper et al., Reference Schnipper, Davidson, Wollins, Tyne, Blayney, Blum, Dicker, Ganz, Hoverman, Langdon, Lyman, Meropol, Mulvey, Newcomer, Peppercorn, Polite, Raghavan, Rossi, Saltz, Schrag, Smith, Yu, Hudis and Schilsky2015). Evidence examining not only the “what” but also the “how” is also fundamental to informing precision care. Such evidence addresses those challenges frequently encountered by health systems, clinicians, and patients – not “what” to do, but “how” to provide care that makes it easy for patients to get what they need when they need it every time, across their lifespan. Traditional recommendations from clinical practice guidelines are not yet able to address the precision care needs of diverse patient populations, but by leveraging big data available for each patient on their clinical and social determinants of health, it is conceivable that one day the rich and diverse sources of evidence described above could inform individualized recommendations, that would enable more precise care and SDM (Eddy et al., Reference Eddy, Adler, Patterson, Lucas, Smith and Morris2011).

An evidence base of harmonized outcome measures from representative patient samples, together with accounting for aggregate financial and non-financial costs, will allow governments and payers to calculate value and cost-effectiveness of therapeutic innovations from a societal perspective. These macro-level evaluations can both identify opportunities to adapt precision care implementation strategies and enable comparisons of the social impact of precision care investments to other opportunities.

Comprehensive care planning

With advances in computing, artificial intelligence, and patient-generated and high-throughput data, it is important to ensure that patients’ goals, preferences, and social context are not lost among the increasing volume of information. To ensure that patients remain at the center of care, all members of the care team, including patients and their caregivers, will need direct access to a comprehensive longitudinal shared care plan that documents all health conditions, medications, goals, preferences, and social context. The process of care planning requires collaboration among patients and clinicians to proactively discuss and document information critical to providing precision care, including roles, strategies for supporting and empowering patients, plans for engaging in SDM, and plans to balance evidence-based care with patient preferences and treatment burden (Burt et al., Reference Burt, Rick, Blakeman, Protheroe, Roland and Bower2014). Comprehensive care plans enable health care clinicians to focus on proactive, prevention-oriented care, make clinical decisions in the context of the patient’s specific needs, and facilitate communication and coordination among all members of the care team (Holland and Lee, Reference Holland and Lee2019). It is important that care planning incorporates processes for updating plans as new health care options become available, along with new information available from continuous monitoring and tracking of progress on goals. Effective care plans help tailor care to patient goals and preferences during acute, episodic care, as well as during health maintenance and care transitions, helping both patients and clinicians better manage multimorbidity and prevent additional impairments (Baker et al., Reference Baker, Cronin, Conway, DeSalvo, Rajkumar and Press2016).

The need to incorporate many different types and sources of data requires mechanisms to share care plans with all members of the clinical team (including primary care, specialty care, behavioral health, pharmacy, community health workers, and many other relevant team members within and across organizations), as well as patients and their caregivers, to ensure that the right information gets to the right person at the right time. Because care plans themselves can cause harm if a patient ends up with multiple competing and sometimes conflicting plans, the use of interoperable comprehensive shared electronic (e-) care plans (CSeCP), electronic tools that use health information technology to document and share information, can provide all members of the care team with electronic access to information critical to their role (Consumer Partnership for eHealth (CPeH), 2013). The use of interoperable shared electronic care plans is an active area of development, particularly as standards and clinical terminologies become harmonized. For example, the Fast Healthcare Interoperability Resources (FHIR) specification, a standard for exchanging healthcare information electronically, provides a mechanism to exchange the rapidly growing volume of health data in a lightweight, real time, and secure manner, regardless of where the health data are stored (Office of the National Coordinator for Health Information Technology, 2019). Further development of infrastructure and technologies to support data sharing and visualization, along with changes in clinical practice, medical education, and ensuring access in rural and socioeconomically disadvantaged populations will be required to realize the promise of e-care plans in precision care (Backonja et al., Reference Backonja, Haynes and Kim2018; Norton et al., Reference Norton, Ip, Ruggiano, Abidogun, Camara, Fu, Hose, Miran, Hsiao, Wang and Bierman2022). Availability of comprehensive person-centered data from e-care plans can also serve as a rich source of data to study the effectiveness of precision medicine therapeutics in people with multimorbidity.

Clinical decision support can provide clinical teams and patients with knowledge and person-specific information, intelligently filtered or presented at appropriate times, to inform care planning. These digital tools offer clinicians and patients a means to access, understand, and apply personalized data and scientific evidence. Further development, including new advances in artificial intelligence to integrate medical, genetic, patient-reported, and community data can enhance the functionality and utility of clinical decision support (Romero-Brufau et al., Reference Romero-Brufau, Wyatt, Boyum, Mickelson, Moore and Cognetta-Rieke2020). This will aid clinicians in communicating increasingly complicated personalized risk information while facilitating collaborative discussions on the treatment and management of multiple conditions (Zipkin et al., Reference Zipkin, Umscheid, Keating, Allen, Aung, Beyth, Kaatz, Mann, Sussman, Korenstein, Schardt, Nagi, Sloane and Feldstein2014). CDS tools with integrated AI that successfully address the varying, sometimes competing needs of patients with multimorbidity will need to be co-developed with patients, healthcare professionals, health systems, developers and other system level stakeholders as equal partners (Mistry, Reference Mistry2019; Silcox et al., Reference Silcox, Dentzer and Bates2020).

Shared decision making and precision care

SDM is a collaborative process in which patients, their families, and caregivers work together in partnership with members of their care team to make health care decisions. In SDM, healthcare decisions are informed both by the care team’s medical and scientific knowledge and understanding of available interventions and the patient’s individual goals, preferences, and circumstances (Hargraves et al., Reference Hargraves, Montori, Brito, Kunneman, Shaw, LaVecchia, Wilson, Walker and Thorsteinsdottir2019). SDM is essential to the delivery of quality, patient-centered care for persons with multimorbidity who have many opportunities to make and revisit complex medical decisions and are often best positioned to evaluate the tradeoffs in benefits and burden within the context of their lives (Tinetti et al., Reference Tinetti, McAvay, Fried, Allore, Salmon, Foody, Bianco, Ginter and Fraenkel2008; Muth et al., Reference Muth, van den Akker, Blom, Mallen, Rochon, Schellevis, Becker, Beyer, Gensichen, Kirchner, Perera, Prados-Torres, Scherer, Thiem, van den Bussche and Glasziou2014; Chi et al., Reference Chi, Wolff, Greer and Dy2017).

Rapid advances in genetic and molecular testing and parallel advances in the application of patient and community reported measures stand to both enhance and complicate SDM for persons with multimorbidity (Bradley et al., Reference Bradley, Rumsfeld and Ho2016; Krist et al., Reference Krist, Glasgow, Heurtin-Roberts, Sabo, Roby, Gorin, Balasubramanian, Estabrooks, Ory, Glenn, Phillips, Kessler, Johnson, Rohweder and Fernandez2016). SDM should also play an important role in helping patients navigate the expanding landscape of genetic and molecular testing and make decisions about whether to pursue the testing and screening options available to them. There are several features of genetic and molecular testing that position SDM as both a complicated and essential component of consultative care. Genetic or molecular testing can have wide-ranging implications for the patient and their family. It can, for example, affect family dynamics by indirectly informing other relatives of their risk. Genetic and molecular test results may also be perceived by patients as certain when many of these tests can only infer increased risk. Treatments may also not always be available to address the results of predictive genetic or molecular tests. Thus, when patients undergo genetic testing, they may be entering a complicated process where risk probabilities and uncertainty need to be balanced alongside patient goals and preferences.

Furthermore, there are increasing opportunities to combine data sources on social determinants, patient preferences, and genomic and biological markers so patients, in collaboration with their families and clinicians, can make more informed, individualized decisions about their healthcare. These advances will also pose new challenges to the implementation of SDM. To achieve SDM, clinicians and patients need time, skills in facilitated communication, and easy access to clinically useful knowledge (Fraenkel and McGraw, Reference Fraenkel and McGraw2007; Joosten et al., Reference Joosten, DeFuentes-Merillas, de Weert, Sensky, van der Staak and de Jong2008; O’Connor et al., Reference O’Connor, Bennett, Stacey, Barry, Col, Eden, Holmes-Rovner, Llewellyn-Thomas, Lyddiatt, Thomson and Trevena2009). Members of the care team including primary care clinicians will need ready access to information on the benefits and harms of new diagnostic and therapeutic modalities to guide these discussions. This capacity, which currently does not exist in most practices, will become increasingly important as the volume of available and actionable personal health information grows and the landscape of molecular and genetic testing options for patients with multimorbidity increases in size and complexity. For SDM to become a routine component of practice in the age of precision care, workflows will need to be adjusted for more efficient division of responsibility across care teams and digital health solutions, such as by using evidence-based artificial intelligence and clinical decision support tools to access, intelligently filter, understand, and apply personal data alongside scientific evidence.

Health systems and models of care

Health systems play an essential role in developing and implementing models of person-centered precision care. By building infrastructure and aligning incentives across clinicians and payers, learning health systems can create models of care that provide clinicians with the time and access to clinical knowledge needed for precision care processes including comprehensive care planning, addressing social risks, heath equity, and SDM (Mas et al., Reference Mas, Miralles, Heras, Ulldemolins, Bonet, Prat, Isnard, Pablo, Rodoreda, Verdaguer, Lladó, Moreno-Gabriel, Urrutia, Rocabayera, Moreno-Millan, Modol, Andrés, Estrada and Del Rey2021; Easterling et al., Reference Easterling, Perry, Woodside, Patel and Gesell2022). Clinicians can meet the challenges resulting from quickly evolving clinical evidence and the complexities of caring for those with multimorbidity if information technology that provides access to updated evidence is available and easily usable (Guise et al., Reference Guise, Savitz and Friedman2018), through tools such as clinical pathways, decision aids, and other clinical decision support systems (Borsky et al., Reference Borsky, Flores, Berliner, Chang, Umscheid and Chang2019; Flores et al., Reference Flores, Jue, Giradi, Schoelles, Mull and Umscheid2019; Bartlett et al., Reference Bartlett, Makhni, Ruokis, Selling, Hall, Umscheid and Kao2022; Easterling et al., Reference Easterling, Perry, Woodside, Patel and Gesell2022).

Learning health systems are uniquely suited to integrate the two primary functions of health care – caring and learning (Montori et al., Reference Montori, Hargraves, McNellis, Ganiats, Genevro, Miller, Ricciardi and Bierman2019). They can play a particularly important role as they undertake the cycle of evidence synthesis, implementation, and generation. The iterative processes of designing, implementing, and evaluating clinical interventions and models of care can provide the evidence needed for widespread dissemination and implementation. Learning health systems have the opportunity to use their robust data collection and analysis infrastructures to generate new evidence on the effectiveness of diagnostic and therapeutic advancements in real world settings as well as effective ways to organize and deliver care (Easterling et al., Reference Easterling, Perry, Woodside, Patel and Gesell2022). Continuous evidence generation by using qualitative and quantitative data and analytics will help determine both the effectiveness of clinical interventions in the context of co-existing illness and social factors, and strategies for care delivery and quality improvement.

Given their access to data on clinically and socioeconomically diverse populations and a broad spectrum of affiliated clinical specialties, systems are well-positioned to design and evaluate models of care that integrate evidence related to the treatment of multiple chronic conditions. Many learning health systems cultivate ties to organizations and resources within their broader communities, enabling them to provide care that incorporates precision medicine and addresses the unique constellation of needs that extend beyond clinical management by integrating social and community services and supports. As health systems cultivate and disseminate innovative models of precision care, they are creating the standards and expectations for how patients will shape their care in an era of data-driven medicine. The norms established in these early demonstrations – including costs and measures of success – will have a lasting influence on how precision care is delivered across specialties, settings, and geographies.

Social context and precision care

Equitable, patient-centered care includes the active participation of patients in goal setting, ensuring services necessary to address needs are available and accessible, and supporting the development self-management skills. Efforts to achieve this include the tailoring of care plans to ensure the integration of patient preferences, and enhanced community partnerships to coordinate access to care or the provision of services to address social risks. Precision care lives in this framework: it should always be delivered in a human and social context and shaped by the social and behavioral determinants of health, as well as patient preferences, alongside genomic and other biologic information. Integrating precision medicine into frameworks that address social needs is integral to precision care.

Dynamic contextual factors must be assessed research (Bayliss et al., Reference Bayliss, Bonds, Boyd, Davis, Finke, Fox, Glasgow, Goodman, Heurtin-Roberts, Lachenmayr, Lind, Madigan, Meyers, Mintz, Nilsen, Okun, Ruiz, Salive and Stange2014). Furthermore, clinical research should be conducted with an eye toward implementation, integrated with practice in an iterative loop to ensure real-world relevance while informing existing implementation science models (Chambers et al., Reference Chambers, Feero and Khoury2016). This can be achieved through identifying information and service needs (e.g., community resources to address social needs, leveraging information on payer coverage to avoid financial strain and toxicity); iteratively capturing data, especially patient and provider-reported outcomes, and using those data to inform ongoing clinical and community practice. Furthermore, there is a need for actionable tools and resources to facilitate addressing the social determinants of health in decision making and care planning (Glasgow et al., Reference Glasgow, Kwan and Matlock2018). Without a shared decision-making process that accounts for individuals’ specific socioeconomic context, the added costs of precision medicine technologies may exacerbate existing strain. Information on patients’ available resources and supports should inform access and payment models to foster equitable access.