Enhancing healthcare for patients with multiple chronic conditions using machine learning and medical specialist data: a scoping review

Research on Multiple Chronic Conditions

Nearly 16% of pregnant individuals have multiple chronic conditions (MCC), and the risk of severe maternal morbidity and mortality increases in a dose-response manner with the number of preexisting conditions. However, little is known about newborn outcomes in this population. To examine the association of the number of preexisting maternal chronic conditions, as well as the presence of MCC complexity, cardiometabolic MCC, and MCC severity, with the risk of severe neonatal morbidity or mortality (SNM-M). This population-based cohort study was conducted in Ontario, Canada, among all live births from 2012 to 2021. Data were analyzed from September 2024 to November 2025. Maternal MCC measured in the 2 years before conception. Modified Poisson regression was performed to generate adjusted relative risks (aRRs) for SNM-M by the number of chronic conditions, MCC complexity (≥3 chronic conditions in ≥3 body systems), co-occurring cardiometabolic conditions, and MCC severity marked by a prenatal hospitalization for a chronic illness. Multivariable models were adjusted for age, parity, immigration status, income quintile, and rurality. The cohort comprised 1 018 968 newborns, including 20 934 to mothers with 3 or more chronic conditions (mean [SD] maternal age, 30.0 [6.3] years), 73 768 to mothers with 2 chronic conditions (mean [SD] maternal age, 30.3 [5.8] years), 276 765 to mothers with 1 chronic condition (mean [SD] maternal age, 30.7 [5.4] years), and 647 501 to mothers with 0 chronic conditions (mean [SD] maternal age, 31.0 [5.1] years). Compared with newborns of mothers with 0 chronic conditions, the aRR for SNM-M increased in a dose-response fashion in newborns of mothers with 1 (1.26; 95% CI, 1.24-1.28), 2 (1.58; 95% CI, 1.54-1.62), and 3 or more (2.01; 95% CI, 1.94-2.09) chronic conditions. The aRRs were also increased with complex MCC (1.97; 95% CI, 1.88-2.06), cardiometabolic MCC (2.67; 95% CI, 2.24-3.19), and severe MCC (3.11; 95% CI, 2.55-3.79). In this study, risks of SNM-M increased with an increasing number of preexisting maternal chronic conditions. These findings suggest that women and adolescents with MCC may benefit from preconception counseling to optimize chronic disease management, monitoring in pregnancy for earlier identification of complications, and enhanced newborn supports.

This issue of the Annals, as well as the previous one, confronts the enormous public health challenges of multimorbidity. More than 1 in 4 Americans has multiple (2 or more) chronic conditions, including physical and behavioral health problems, accounting for an estimated two-thirds of total US health care spending.1 An individual’s risks for a variety of adverse health outcomes (eg, poor functional status, unnecessary hospitalizations, and adverse drug events) rise as the number of multiple chronic conditions increases.2 The Centers for Medicare and Medicaid Services (CMS) has just released even more detailed information with respect to its Medicare fee-for-service populations,3 exposing the exceptional complexity and sheer burden that multiple chronic conditions pose for patients, health facilities, payers, and clinicians. In its recently released chart book Chronic Conditions Among Medicare Beneficiaries,3 CMS describes detailed demographics and prevalence measures of multiple medical conditions in this population and the dramatic impact on service utilization and spending. Examples of key findings are that two-thirds (20.7 million beneficiaries) had at least 2 or more chronic conditions; about 50% of beneficiaries with stroke or heart failure had 5 or more additional chronic conditions; beneficiaries with 6 or more chronic conditions accounted for about one-half of Medicare spending on hospitalizations; more than one-quarter of beneficiaries with 6 or more chronic conditions had a hospital readmission within 30 days; and the 12% of beneficiaries with 6 or more chronic conditions accounted for 43% of Medicare spending. For health systems that have traditionally focused on research and treatment of single conditions, these tremendous challenges have forced many to escalate efforts to identify and implement solutions. How do we as a society bring a greater sense of order to this vexing challenge? As part of the response, the US Department of Health and Human Services (HHS), in conjunction with partner organizations and other stakeholders, used a deliberative process to develop Multiple Chronic Conditions—A Strategic Framework: Optimum Health and Quality of Life for Individuals with Multiple Chronic Conditions.2 The framework serves as a national-level road map to assist HHS programs and public and private stakeholders in ensuring a more coordinated and comprehensive approach to improving the health status of individuals with multiple chronic conditions.2,4 Released to stakeholders and the public in late 2010, the framework is organized into 4 major goal areas with subsets of objectives and action strategies for use by clinical practitioners, policy makers, researchers, and others. The framework’s goals encompass the interdependent domains of (1) strengthening the health care and public health systems; (2) empowering the individual to use self-care management; (3) equipping health care clinicians with tools, information, and other interventions; and (4) supporting targeted research about individuals with multiple chronic conditions and effective interventions. The 3 articles on multimorbidity included in this issue of the Annals,5–7 which focus on measurement, represent progress in better understanding the epidemiology of multiple chronic conditions, a key aspect of the HHS strategic framework’s fourth goal. The articles further illustrate that there are multiple operational definitions of multimorbidity, each with their respective strengths and weaknesses. Though each may be appropriate according to the outcome of interest, there also may be utility for some degree of standardization in characterizing this heterogeneous population. The definition for multiple chronic conditions used in the HHS strategic framework utilized an approach of simple counts of conditions (ie, 2 or more). The article by Huntley et al, consistent with this approach, finds that simple counts of diseases perform almost as well as complex measures in predicting most outcomes.5 Ultimately, however, from a policy perspective, what generally holds true irrespective of the selected measure is that as the magnitude of multimorbidity increases, patient outcomes decline and costs rise. Thus, targeting care management efforts on multimorbid populations should, in principle, accelerate the country’s progress toward the goals of delivery system reform. We invite readers to further their familiarity with the complex issues posed by multiple chronic conditions by reviewing the HHS strategic framework and related activities (http://www.hhs.gov/ash/initiatives/mcc).8 A current HHS collaborative initiative—led by CMS, the Centers for Disease Control and Prevention, and the Agency for Healthcare Research & Quality—is not only examining options for a conceptual framework to improve definition and measurement of priority chronic conditions, but also is conducting coordinated analyses of multiple data sets to improve descriptive epidemiological characterization of multiple chronic conditions. Increasing life expectancy and the aging of the population will only intensify the challenge for the future. The articles in this issue begin to address an enormous health system challenge that demands our urgent attention.

BackgroundMultimorbidity, defined as the coexistence of multiple chronic conditions, poses significant challenges to health care systems on a global scale. It is associated with increased mortality, reduced quality of life, and increased health care costs. The burden of multimorbidity is expected to worsen if no effective intervention is taken. Machine learning has the potential to assist in addressing these challenges since it offers advanced analysis and decision-making capabilities, such as disease prediction, treatment development, and clinical strategies.ObjectiveThis paper represents the protocol of a scoping review that aims to identify and explore the current literature concerning the use of machine learning for patients with multimorbidity. More precisely, the objective is to recognize various machine learning models, the patient groups involved, features considered, types of input data, the maturity of the machine learning algorithms, and the outcomes from these machine learning models.MethodsThe scoping review will be based on the guidelines of the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews). Five databases (PubMed, Embase, IEEE, Web of Science, and Scopus) are chosen to conduct a literature search. Two reviewers will independently screen the titles, abstracts, and full texts of identified studies based on predefined eligibility criteria. Covidence (Veritas Health Innovation Ltd) will be used as a tool for managing and screening papers. Only studies that examine more than 1 chronic disease or individuals with a single chronic condition at risk of developing another will be included in the scoping review. Data from the included studies will be collected using Microsoft Excel (Microsoft Corp). The focus of the data extraction will be on bibliographical information, objectives, study populations, types of input data, types of algorithm, performance, maturity of the algorithms, and outcome.ResultsThe screening process will be presented in a PRISMA-ScR flow diagram. The findings of the scoping review will be conveyed through a narrative synthesis. Additionally, data extracted from the studies will be presented in more comprehensive formats, such as charts or tables. The results will be presented in a forthcoming scoping review, which will be published in a peer-reviewed journal.ConclusionsTo our knowledge, this may be the first scoping review to investigate the use of machine learning in multimorbidity research. The goal of the scoping review is to summarize the field of literature on machine learning in patients with multiple chronic conditions, highlight different approaches, and potentially discover research gaps. The results will offer insights for future research within this field, contributing to developments that can enhance patient outcomes.International Registered Report Identifier (IRRID)PRR1-10.2196/53761

Cross-sectional study. The purposes of this study were to assess (i) prevalence of self-reported multiple chronic conditions (MCC) in a population-based cohort of persons with traumatic spinal cord injury (TSCI) and (ii) the association between health-related behaviors and MCC. Population-based TSCI cohort. Participants included 716 adults with TSCI of at least 1-year duration who were identified through a population-based TSCI surveillance system. Standard questions from the Behavioral Risk Factor Surveillance System measured cigarette smoking, binge drinking, planned exercises, and 10 chronic health conditions (CHC), including diabetes, heart attack, angina (or coronary artery disease), stroke, cancer, asthma, kidney disease, arthritis, depressive disorder, chronic obstructive pulmonary disease. MCC was defined as having two or more CHCs in this study. Multivariate logistic regression models were used to assess the association between health-related behaviors and MCC. Almost half (45%) of the study sample had MCC. After controlling for demographic and injury characteristics, participants with smoking history of at least 100 cigarettes were 59% more likely to develop MCC, and those who had planned exercises at least three times a week were 36% less likely to have MCC. We found MCC prevalence was high among people with TSCI, and MCC was associated with cigarette smoking and planned exercise.

PSS15 Prevalence and Predictors of Multiple Comorbid Chronic Conditions Among a Nationally-Representative Sample of United States Older Adults with Self-Reported PAIN

The association between an individuals' demographic and health characteristics and the presence of multiple chronic conditions is not well known among older United States (US) adults. This study aimed to identify the prevalence and associations of having multiple chronic conditions among older US adults with self-reported pain. This retrospective, cross-sectional study used data from the 2017 Medical Expenditure Panel Survey. Study subjects were aged≥50 years and had self-reported pain in the past four weeks. The outcome variable was multiple (≥5) chronic conditions (vs. <5 chronic conditions). Hierarchical logistic regression models were used to identify significant associations between demographic and health characteristics and multiple chronic conditions with significance indicated at an a priori alpha level of 0.05. The complex survey design was accounted for when obtaining nationally-representative estimates. The weighted population was 57,074,842 US older adults with pain, of which, 66.1% had≥5 chronic conditions. In fully-adjusted analyses, significant associations of≥5 comorbid chronic conditions included: age 50-64 vs.≥65 years (adjusted odds ratio [AOR]=0.478, 95% confidence interval [CI]=0.391, 0.584); male vs. female gender (AOR=1.271, 95% CI=1.063, 1.519); white vs. other race (AOR=1.220, 95% CI=1.016, 1.465); Hispanic vs. non-Hispanic ethnicity (AOR=0.614, 95% CI=0.475, 0.793); employed vs. unemployed (AOR=0.591, 95% CI=0.476, 0.733); functional limitations vs. no functional limitations (AOR=1.862, 95% CI=1.510, 2.298); work limitations vs. no work limitations (AOR=1.588, 95% CI=1.275, 1.976); little/moderate vs. quite a bit/extreme pain (AOR=0.732, 95% CI=0.599, 0.893); and excellent/very good (AOR=0.375, 95% CI=0.294, 0.480) or good (AOR=0.661, 95% CI=0.540, 0.810) vs. fair/poor physical health. Approximately 38 million of the 57 million US older adults with pain in this study had≥5 chronic conditions in 2017. Several characteristics were associated with multiple chronic conditions, which may be important for health care professionals to consider when working with patients to manage their pain. This study was approved by The University of Arizona Institutional Review Board (2006721124, June 12, 2020).

Evolution of multiple chronic conditions (MCC) follows a complex stochastic process, influenced by several factors including the inter-relationship of existing conditions, and patient-level risk factors. Nearly 20% of citizens aged 18 years and older are burdened with two or more (multiple) chronic conditions (MCC). Treatment for people living with MCC currently accounts for an estimated 66% of the Nation's healthcare costs. However, it is still not known precisely how MCC emerge and accumulate among individuals or in the general population. This study investigates major patterns of MCC transitions in a diverse population of patients and identifies the risk factors affecting the transition process. A Latent regression Markov clustering (LRMCL) algorithm is proposed to identify major transitions of four MCC that include hypertension (HTN), depression, Post- Traumatic Stress Disorder (PTSD), and back pain. A cohort of 601,805 individuals randomly selected from the population of Iraq and Afghanistan war Veterans (IAVs) who received VA care during three or more years between 2002-2015, is used for training the proposed LRMCL algorithm. Two major clusters of MCC transition patterns with 78% and 22% probability of membership respectively were identified. The primary cluster demonstrated the possibility of improvement when the number of MCC is small and an increase in probability of MCC accumulation as the number of co- morbidities increased. The second cluster showed stability (no change) of MCC overtime as the major pattern. Age was the most significant risk factor associated with the most probable cluster for each IAV. These findings suggest that our proposed LRMCL algorithm can be used to describe and understand MCC transitions, which may ultimately allow healthcare systems to support optimal clinical decision- making. This method will be used to describe a broader range of MCC transitions in this and non-VA populations, and will add treatment information to see if models including treatments and MCC emergence can be used to support clinical decision-making in patient care.

The most common chronic condition experienced by adults is multimorbidity, the coexistence of multiple chronic diseases or conditions. In patients with coronary disease, for example, it is the sole condition in only 17% of cases.1 Almost 3 in 4 individuals aged 65 years and older have multiple chronic conditions, as do 1 in 4 adults younger than 65 years who receive health care.2 Adults with multiple chronic conditions are the major users of health care services at all adult ages, and account for more than two-thirds of health care spending.2 Despite the predominance of multiple chronic conditions, however, reimbursement remains linked to discrete International Classification of Diseases diagnostic codes, none of which are for multimorbidity or multiple chronic conditions. Specialists are responsible for a single disease among the patient’s many. Quality measurement largely ignores the unintended consequences of applying the multiple interventions necessary to adhere to every applicable measure. Uncertain benefit and potential harm of numerous simultaneous treatments, worsening of a single disease by treatment of a coexisting one, and treatment burden arising from following several disease guidelines are the well-documented challenges of clinical decision making for patients with multiple chronic conditions.3,4 To ensure safe and effective care for adults with multiple chronic conditions, particularly the millions of baby boomers entering their years of declining health and increasing health service use, health care must shift its current focus on managing innumerable individual diseases. To align with the clinical reality of multimorbidity, care should evolve from a disease orientation to a patient goal orientation, focused on maximizing the health goals of individual patients with unique sets of risks, conditions, and priorities. Patient goal–oriented health care involves ascertaining a patient’s health outcome priorities and goals, identifying the diseases and other modifiable factors impeding these goals, calculating and communicating the likely effect of alternative treatments on these goals, and guiding shared decision making informed by this information.4

Multiple chronic conditions are imposing an increasing health and economic burden on the Chinese health system. While exposure to residential greenness has been shown to provide various health benefits, its relationship with multiple chronic conditions remains largely unexplored. This study aims to investigate this relationship using high-resolution satellite imagery and data from the 6th Health Services Survey (HSS) cohort in Shandong province. We linked health data from the HSS with 12-month average Normalised Difference Vegetation Index (NDVI) measurements based on reported residential geocodes. Multiple chronic condition status was defined as having two or more chronic conditions concurrently, according to the HSS's predefined disease classification. Generalised mixed regression models were utilised to assess both the likelihood and count of multiple chronic conditions in relation to greenspace exposure. Additionally, using the pre-defined disease classes, we also explored how greenspace influences multiple chronic conditions across various physiological systems and disease categories. A total of 28 489 individuals were included in this cross-sectional analysis. After adjusting for potential confounding factors, we found that exposure to greenspace was significantly associated with a reduced prevalence and count of chronic conditions. Adjusted odds ratios (aORs) and 95% confidence intervals (CIs) for were: Q2 (aOR = 0.74; 95% CI = 0.62, 0.88), Q3 (aOR = 0.69; 95% CI = 0.55, 0.86), and Q4 (aOR = 0.70; 95% CI = 0.56, 0.88), respectively, compared against the baseline Q1 quartile. Subgroup analyses revealed that higher residential greenspace exposure reduced risks of blood, endocrine, nutritional and metabolic chronic diseases. No clear associations were found for other chronic disease classes. Additionally, consistent results were observed across spatial and temporal sensitivity analyses. Our findings underscore the potential beneficial effects of residential greenness on multiple chronic conditions, with implications for urban planning, environmental policy, and community development.

Although 1-in-5 adults have multiple (≥ 2) chronic conditions, limited attention has been given to the association between multiple chronic conditions and employment. Cross-sectional data (2011 National Health Interview Survey) and multivariate regression analyses were used to examine the association among multiple chronic conditions, employment, and labor force outcomes for U.S. adults aged 18-64 years, controlling for covariates. Among U.S. adults aged 18-64 years (unweighted, n = 25,458), having multiple chronic conditions reduced employment probability by 11-29%. Some individual chronic conditions decreased employment probability. Among employed adults (unweighted, n = 16,096), having multiple chronic conditions increased the average number of work days missed due to injury/illness in the past year by 3-9 days. Multiple chronic conditions are a barrier to employment and increase the number of work days missed, placing affected individuals at a financial disadvantage. Researchers interested in examining consequences of multiple chronic conditions should give consideration to labor force outcomes.

The use of motivational interviewing is relatively new in multiple chronic conditions (MCCs). A scoping review was conducted according to JBI methodology to identify, map and synthesize existing evidence on the use of motivational interviewing to support self-care behavior changes in older patients with MCCs and to support their informal caregivers in promoting patient self-care changes. Seven databases were searched, from database inception to July 2022, for studies that used motivational interviewing in interventions for older patients with MCCs and their informal caregivers. We identified 12 studies, reported in 15 articles, using qualitative, quantitative, or mixed-method designs, conducted between 2012 and 2022, describing the use of motivational interviewing for patients with MCCs. We could not locate any study regarding its application for informal caregivers. The scoping review showed that the use of motivational interviewing is still limited in MCCs. It was used mainly to improve patient medication adherence. The studies provided scant information about how the method was applied. Future studies should provide more information about the application of motivational interviewing and should address self-care behavior changes relevant to patients and healthcare providers. Informal caregivers should also be targeted in motivational interviewing interventions, as they are essential for the care of older patients with MCCs.

IntroductionUnderstanding longitudinal relationships among multiple chronic conditions, limitations in activities of daily living, and health-related quality of life is important for identifying potential opportunities for health promotion and disease prevention among older adults.MethodsThis study assessed longitudinal associations between multiple chronic conditions and limitations in activities of daily living on health-related quality of life among older adults (≥65 years) from 2004 through 2006, using data from the Medicare Health Outcomes Survey (N = 27,334).ResultsUsing a longitudinal path model, we found the numbers of chronic conditions at baseline and 2-year follow-up were independently associated with more limitations in activities of daily living at 2-year follow-up. In addition, more limitations in activities of daily living at 2-year follow-up were associated with worse health-related quality of life during the follow-up time period. The association between multiple chronic conditions and indices of health-related quality of life was mediated by changes in limitations in activities of daily living.ConclusionBoth baseline and new multiple chronic conditions led to worse health in terms of activities of daily living and health-related quality of life and should be considered important outcomes to intervene on for improved long-term health. In addition, public health practitioners should consider addressing classes of multiple chronic conditions by using interventions designed to reduce the emergence of multiple chronic conditions, such as physical activity, reductions in smoking rates, and improved and coordinated access to health care services.

IntroductionPeople are living longer; however, they are not necessarily experiencing good health and well-being as they age. Many older adults live with multiple chronic conditions (MCC), and complex health issues,...

About half of U.S. adults have at least one chronic health condition, and the prevalence of multiple (two or more) chronic conditions increased from 21.8% in 2001 to 25.5% in 2012. Chronic conditions profoundly affect quality of life, are leading causes of death and disability, and account for 86% of total health care spending. Arthritis is a common cause of disability, one of the most common chronic conditions, and is included in prevalent combinations of multiple chronic conditions. To determine the impact of having arthritis alone or as one of multiple chronic conditions on selected important life domains, CDC analyzed data from the 2013 National Health Interview Survey (NHIS). Having one or more chronic conditions was associated with significant and progressively higher prevalences of social participation restriction, serious psychological distress, and work limitations. Adults with arthritis as one of their multiple chronic conditions had higher prevalences of adverse outcomes on all three life domains compared with those with multiple chronic conditions but without arthritis. The high prevalence of arthritis, its common co-occurrence with other chronic conditions, and its significant adverse effect on life domains suggest the importance of considering arthritis in discussions addressing the effect of multiple chronic conditions and interventions needed to reduce that impact among researchers, health care providers, and policy makers.