Multimorbidity patterns with K-means nonhierarchical cluster analysis

An analysis of the vaginal microbiota and cervicovaginal metabolomics in cervical lesions and cervical carcinoma

Background

To explore the role of vaginal microbiota and metabolomics in the progression of cervical dysplasia.

Methods

The patient group consists of female patients with low-grade, high-grade cervical dysplasia, and cervical cancer. Normal cervix samples from health volunteers were used as controls. The metabolic fingerprints of cervicovaginal lavage were analyzed using liquid chromatography-mass spectrometry, while the vaginal microbiota was examined through 16S rRNA sequencing. Bioinformatic analysis was adopted to investigate the interplay between hosts and microbes. The vaginal metabolic and microbiota profiles of 90 female patients with cervical dysplasia and 10 controls were analyzed to discover the biological characteristics underlying the progression of cervical cancer.

Results

We found that Valyl-Glutamate, N, N'-Diacetylbenzidine, and Oxidized glutathione, which were involved in oxidative stress response, were discriminators to distinguish the normal cervix, invasive cervical carcinomas, and CIN3 from others. Cervical carcinoma was characterized by a large variety of vaginal microbes (dominated by non-Lactobacillus communities) compared to the control. These microbes affected amino acid and nucleotide metabolism, producing metabolites with cervical carcinoma and genital inflammation compared to the control group.

Conclusions

This study revealed that cervicovaginal metabolic profiles were determined by cervical cancer, vaginal microbiota, and their interplays. ROS metabolism can be used to discriminate normal cervix, CIN3, and invasive cervical carcinoma.

Phenotypic Clusters and Multimorbidity in Hypermobile Ehlers-Danlos Syndrome

Objective

To perform a retrospective clinical study in order to investigate phenotypic penetrance within a large registry of patients with hypermobile Ehlers-Danlos syndrome (hEDS) to enhance diagnostic and treatment guidelines by understanding associated comorbidities and improving accuracy in diagnosis.

Patients and Methods

From May 1, 2021 to July 31, 2023, 2149 clinically diagnosed patients with hEDS completed a self-reported survey focusing on diagnostic and comorbid conditions prevalence. K-means clustering was applied to analyze survey responses, which were then compared across gender groups to identify variations and gain clinical insights.

Results

Analysis of clinical manifestations in this cross-sectional cohort revealed insights into multimorbidity patterns across organ systems, identifying 3 distinct patient groups. Differences among these phenotypic clusters provided insights into diversity within the population with hEDS and indicated that Beighton scores are unreliable for multimorbidity phenotyping.

Conclusion

Clinical data on the phenotypic presentation and prevalence of comorbidities in patients with hEDS have historically been limited. This study provides comprehensive data sets on phenotypic presentation and comorbidity prevalence in patients with hEDS, highlighting factors often overlooked in diagnosis. The identification of distinct patient groups emphasizes variations in hEDS manifestations beyond current guidelines and emphasizes the necessity of comprehensive multidisciplinary care for those with hEDS.

Multimorbidity patterns and trajectories in young and middle-aged adults: a large-scale population-based cohort study

Introduction

The presence of multiple chronic conditions, also referred to as multimorbidity, is a common finding in adults. Epidemiologic research can help identify groups of individuals with similar clinical profiles who could benefit from similar interventions. Many cross-sectional studies have revealed the existence of different multimorbidity patterns. Most of these studies were focused on the older population. However, multimorbidity patterns begin to form at a young age and can evolve over time following distinct multimorbidity trajectories with different impact on health. In this study, we aimed to identify multimorbidity patterns and trajectories in adults 18-65 years old.

Methods

We conducted a retrospective longitudinal epidemiologic study in the EpiChron Cohort, which includes all inhabitants of Aragón (Spain) registered as users of the Spanish National Health System, linking, at the patient level, information from electronic health records from both primary and specialised care. We included all 293,923 patients 18-65 years old with multimorbidity in 2011. We used cluster analysis at baseline (2011) and in 2015 and 2019 to identify multimorbidity patterns at four and eight years of follow-up, and we then created alluvial plots to visualise multimorbidity trajectories. We performed age- and sex-adjusted logistic regression analysis to study the association of each pattern with four- and eight-year mortality.

Results

We identified three multimorbidity patterns at baseline, named dyslipidaemia & endocrine-metabolic, hypertension & obesity, and unspecific. The hypertension & obesity pattern, found in one out of every four patients was associated with a higher likelihood of four- and eight-year mortality (age- and sex-adjusted odds ratio 1.11 and 1.16, respectively) compared to the unspecific pattern. Baseline patterns evolved into different patterns during the follow-up.

Discussion

Well-known preventable cardiovascular risk factors were key elements in most patterns, highlighting the role of hypertension and obesity as risk factors for higher mortality. Two out of every three patients had a cardiovascular profile with chronic conditions like diabetes and obesity that are linked to low-grade systemic chronic inflammation. More studies are encouraged to better characterise the relatively large portion of the population with an unspecific disease pattern and to help design and implement effective and comprehensive strategies towards healthier ageing.

Clusters from chronic conditions in the Danish adult population

Multimorbidity, the presence of 2 or more chronic conditions in a person at the same time, is an increasing public health concern, which affects individuals through reduced health related quality of life, and society through increased need for healthcare services. Yet the structure of chronic conditions in individuals with multimorbidity, viewed as a population, is largely unmapped. We use algorithmic diagnoses and the K-means algorithm to cluster the entire 2015 Danish multimorbidity population into 5 clusters. The study introduces the concept of rim data as an additional tool for determining the number of clusters. We label the 5 clusters the Allergies, Chronic Heart Conditions, Diabetes, Hypercholesterolemia, and Musculoskeletal and Psychiatric Conditions clusters, and demonstrate that for 99.32% of the population, the cluster allocation can be determined from the diagnoses of 4-5 conditions. Clusters are characterized through most prevalent conditions, absent conditions, over- or under-represented conditions, and co-occurrence of conditions. Clusters are further characterized through socioeconomic variables and healthcare service utilizations. Additionally, geographical variations throughout Denmark are studied at the regional and municipality level. We find that subdivision into municipality levels suggests that the Allergies cluster frequency is positively associated with socioeconomic status, while the subdivision suggests that frequencies for clusters Diabetes and Hypercholesterolemia are negatively correlated with socioeconomic status. We detect no indication of association to socioeconomic status for the Chronic Heart Conditions cluster and the Musculoskeletal and Psychiatric Conditions cluster. Additional spatial variation is revealed, some of which may be related to urban/rural populations. Our work constitutes a step in the process of characterizing multimorbidity populations, leading to increased comprehension of the nature of multimorbidity, and towards potential applications to individual-based care, prevention, the development of clinical guidelines, and population management.

Multimorbidity in middle-aged women and COVID-19: binary data clustering for unsupervised binning of rare multimorbidity features and predictive modeling

BACKGROUND

Multimorbidity is typically associated with deficient health-related quality of life in mid-life, and the likelihood of developing multimorbidity in women is elevated. We address the issue of data sparsity in non-prevalent features by clustering the binary data of various rare medical conditions in a cohort of middle-aged women. This study aims to enhance understanding of how multimorbidity affects COVID-19 severity by clustering rare medical conditions and combining them with prevalent features for predictive modeling. The insights gained can guide the development of targeted interventions and improved management strategies for individuals with multiple health conditions.

METHODS

The study focuses on a cohort of 4477 female patients, (aged 45-60) in Piedmont, Italy, and utilizes their multimorbidity data prior to the COVID-19 pandemic from their medical history from 2015 to 2019. The COVID-19 severity is determined by the hospitalization status of the patients from February to May 2020. Each patient profile in the dataset is depicted as a binary vector, where each feature denotes the presence or absence of a specific multimorbidity condition. By clustering the sparse medical data, newly engineered features are generated as a bin of features, and they are combined with the prevalent features for COVID-19 severity predictive modeling.

RESULTS

From sparse data consisting of 174 input features, we have created a low-dimensional feature matrix of 17 features. Machine Learning algorithms are applied to the reduced sparsity-free data to predict the Covid-19 hospital admission outcome. The performance obtained for the corresponding models are as follows: Logistic Regression (accuracy 0.72, AUC 0.77, F1-score 0.69), Linear Discriminant Analysis (accuracy 0.7, AUC 0.77, F1-score 0.67), and Ada Boost (accuracy 0.7, AUC 0.77, F1-score 0.68).

CONCLUSION

Mapping higher-dimensional data to a low-dimensional space can result in information loss, but reducing sparsity can be beneficial for Machine Learning modeling due to improved predictive ability. In this study, we addressed the issue of data sparsity in electronic health records and created a model that incorporates both prevalent and rare medical conditions, leading to more accurate and effective predictive modeling. The identification of complex associations between multimorbidity and the severity of COVID-19 highlights potential areas of focus for future research, including long COVID and intervention efforts.

Augmenting K-Means Clustering With Qualitative Data to Discover the Engagement Patterns of Older Adults With Multimorbidity When Using Digital Health Technologies: Proof-of-Concept Trial

BACKGROUND

Multiple chronic conditions (multimorbidity) are becoming more prevalent among aging populations. Digital health technologies have the potential to assist in the self-management of multimorbidity, improving the awareness and monitoring of health and well-being, supporting a better understanding of the disease, and encouraging behavior change.

OBJECTIVE

The aim of this study was to analyze how 60 older adults (mean age 74, SD 6.4; range 65-92 years) with multimorbidity engaged with digital symptom and well-being monitoring when using a digital health platform over a period of approximately 12 months.

METHODS

Principal component analysis and clustering analysis were used to group participants based on their levels of engagement, and the data analysis focused on characteristics (eg, age, sex, and chronic health conditions), engagement outcomes, and symptom outcomes of the different clusters that were discovered.

RESULTS

Three clusters were identified: the typical user group, the least engaged user group, and the highly engaged user group. Our findings show that age, sex, and the types of chronic health conditions do not influence engagement. The 3 primary factors influencing engagement were whether the same device was used to submit different health and well-being parameters, the number of manual operations required to take a reading, and the daily routine of the participants. The findings also indicate that higher levels of engagement may improve the participants' outcomes (eg, reduce symptom exacerbation and increase physical activity).

CONCLUSIONS

The findings indicate potential factors that influence older adult engagement with digital health technologies for home-based multimorbidity self-management. The least engaged user groups showed decreased health and well-being outcomes related to multimorbidity self-management. Addressing the factors highlighted in this study in the design and implementation of home-based digital health technologies may improve symptom management and physical activity outcomes for older adults self-managing multimorbidity.

Associations between multimorbidity burden and Alzheimer's pathology in older adults without dementia: the CABLE study

Studies have shown that multimorbidity may be associated with the Alzheimer's disease (AD) stages, but it has not been fully characterized in patients without dementia. A total of 1402 Han Chinese older adults without dementia from Chinese Alzheimer's Biomarker and LifestylE (CABLE) study were included and grouped according to their multimorbidity patterns, defined by the number of chronic disorders and cluster analysis. Multivariable linear regression models were used to detect the associations with AD-related cerebrospinal fluid (CSF) biomarkers. Multimorbidity and severe multimorbidity (≥4 chronic conditions) were significantly associated with CSF amyloid and tau levels (pFDR < 0.05). Metabolic patterns were significantly associated with higher levels of CSF Aβ40 (β = 0.159, pFDR = 0.036) and tau (P-tau: β = 0.132, pFDR = 0.035; T-tau: β = 0.126, pFDR = 0.035). The above associations were only significant in the cognitively normal (CN) group. Multimorbidity was associated with brain AD pathology before any symptomatic evidence of cognitive impairment. Identifying such high-risk groups might allow tailored interventions for AD prevention.

Learning prevalent patterns of co-morbidities in multichronic patients using population-based healthcare data

The prevalence of longstanding chronic diseases has increased worldwide, along with the average age of the population. As a result, an increasing number of people is affected by two or more chronic conditions simultaneously, and healthcare systems are facing the challenge of treating multimorbid patients effectively. Current therapeutic strategies are suited to manage each chronic condition separately, without considering the whole clinical condition of the patient. This approach may lead to suboptimal clinical outcomes and system inefficiencies (e.g. redundant diagnostic tests and inadequate drug prescriptions). We develop a novel methodology based on the joint implementation of data reduction and clustering algorithms to identify patterns of chronic diseases that are likely to co-occur in multichronic patients. We analyse data from a large adult population of multichronic patients living in Tuscany (Italy) in 2019 which was stratified by sex and age classes. Results demonstrate that (i) cardio-metabolic, endocrine, and neuro-degenerative diseases represent a stable pattern of multimorbidity, and (ii) disease prevalence and clustering vary across ages and between women and men. Identifying the most common multichronic profiles can help tailor medical protocols to patients' needs and reduce costs. Furthermore, analysing temporal patterns of disease can refine risk predictions for evolutive chronic conditions.

Patterns of multimorbidity in primary care electronic health records: A systematic review

Background

Multimorbidity, the coexistence of multiple chronic conditions in an individual, is a complex phenomenon that is highly prevalent in primary care settings, particularly in older individuals. This systematic review summarises the current evidence on multimorbidity patterns identified in primary care electronic health record (EHR) data.

Methods

Three databases were searched from inception to April 2022 to identify studies that derived original multimorbidity patterns from primary care EHR data. The quality of the included studies was assessed using a modified version of the Newcastle-Ottawa Quality Assessment Scale.

Results

Sixteen studies were included in this systematic review, none of which was of low quality. Most studies were conducted in Spain, and only one study was conducted outside of Europe. The prevalence of multimorbidity (i.e. two or more conditions) ranged from 14.0% to 93.9%. The most common stratification variable in disease clustering models was sex, followed by age and calendar year. Despite significant heterogeneity in clustering methods and disease classification tools, consistent patterns of multimorbidity emerged. Mental health and cardiovascular patterns were identified in all studies, often in combination with diseases of other organ systems (e.g. neurological, endocrine).

Discussion

These findings emphasise the frequent coexistence of physical and mental health conditions in primary care, and provide useful information for the development of targeted preventive and management strategies. Future research should explore mechanisms underlying multimorbidity patterns, prioritise methodological harmonisation to facilitate the comparability of findings, and promote the use of EHR data globally to enhance our understanding of multimorbidity in more diverse populations.

A scoping review finds a growing trend in studies validating multimorbidity patterns and identifies five broad types of validation methods

OBJECTIVES

Multimorbidity, the presence of two or more long-term conditions, is a growing public health concern. Many studies use analytical methods to discover multimorbidity patterns from data. We aimed to review approaches used in published literature to validate these patterns.

STUDY DESIGN AND SETTING

We systematically searched PubMed and Web of Science for studies published between July 2017 and July 2023 that used analytical methods to discover multimorbidity patterns.

RESULTS

Out of 31,617 studies returned by the searches, 172 were included. Of these, 111 studies (64%) conducted validation, the number of studies with validation increased from 53.13% (17 out of 32 studies) to 71.25% (57 out of 80 studies) in 2017-2019 to 2022-2023, respectively. Five types of validation were identified: assessing the association of multimorbidity patterns with clinical outcomes (n = 79), stability across subsamples (n = 26), clinical plausibility (n = 22), stability across methods (n = 7) and exploring common determinants (n = 2). Some studies used multiple types of validation.

CONCLUSION

The number of studies conducting a validation of multimorbidity patterns is clearly increasing. The most popular validation approach is assessing the association of multimorbidity patterns with clinical outcomes. Methodological guidance on the validation of multimorbidity patterns is needed.

Epidemiology, mortality, and health service use of local-level multimorbidity patterns in South Spain

Multimorbidity -understood as the occurrence of chronic diseases together- represents a major challenge for healthcare systems due to its impact on disability, quality of life, increased use of services and mortality. However, despite the global need to address this health problem, evidence is still needed to advance our understanding of its clinical and social implications. Our study aims to characterise multimorbidity patterns in a dataset of 1,375,068 patients residing in southern Spain. Combining LCA techniques and geographic information, together with service use, mortality, and socioeconomic data, 25 chronicity profiles were identified and subsequently characterised by sex and age. The present study has led us to several findings that take a step forward in this field of knowledge. Specifically, we contribute to the identification of an extensive range of at-risk groups. Moreover, our study reveals that the complexity of multimorbidity patterns escalates at a faster rate and is associated with a poorer prognosis in local areas characterised by lower socioeconomic status. These results emphasize the persistence of social inequalities in multimorbidity, highlighting the need for targeted interventions to mitigate the impact on patients' quality of life, healthcare utilisation, and mortality rates.

Multimorbidity patterns in COVID-19 patients and their relationship with infection severity: MRisk-COVID study

BACKGROUND

Several chronic conditions have been identified as risk factors for severe COVID-19 infection, yet the implications of multimorbidity need to be explored. The objective of this study was to establish multimorbidity clusters from a cohort of COVID-19 patients and assess their relationship with infection severity/mortality.

METHODS

The MRisk-COVID Big Data study included 14 286 COVID-19 patients of the first wave in a Spanish region. The cohort was stratified by age and sex. Multimorbid individuals were subjected to a fuzzy c-means cluster analysis in order to identify multimorbidity clusters within each stratum. Bivariate analyses were performed to assess the relationship between severity/mortality and age, sex, and multimorbidity clusters.

RESULTS

Severe infection was reported in 9.5% (95% CI: 9.0-9.9) of the patients, and death occurred in 3.9% (95% CI: 3.6-4.2). We identified multimorbidity clusters related to severity/mortality in most age groups from 21 to 65 years. In males, the cluster with highest percentage of severity/mortality was Heart-liver-gastrointestinal (81-90 years, 34.1% severity, 29.5% mortality). In females, the clusters with the highest percentage of severity/mortality were Diabetes-cardiovascular (81-95 years, 22.5% severity) and Psychogeriatric (81-95 years, 16.0% mortality).

CONCLUSION

This study characterized several multimorbidity clusters in COVID-19 patients based on sex and age, some of which were found to be associated with higher rates of infection severity/mortality, particularly in younger individuals. Further research is encouraged to ascertain the role of specific multimorbidity patterns on infection prognosis and identify the most vulnerable morbidity profiles in the community.

TRIAL REGISTRATION

NCT04981249. Registered 4 August 2021 (retrospectively registered).

Multimorbidity clusters and associated health care cost among patients attending psychiatric clinics in Odisha, India

Introduction

There is a dearth of data on common multimorbidity clusters and the healthcare costs for individuals with mental health disorders. This study aimed to identify clinically meaningful physical-mental multimorbidity clusters, frequently occurring clusters of conditions, and healthcare utilization patterns and expenditure among patients attending a psychiatric outpatient clinic.

Materials and Methods

Data were collected in the psychiatric outpatient department among patients aged 18 years and above in February-July 2019 (n = 500); follow-up data on non-communicable disease incidence were collected after 18 months. For analysis, morbidity clusters were defined using two approaches: 1) agglomerative hierarchical clustering method to identify clusters of diseases; and 2) non-hierarchical cluster k mean analysis to identify clusters of patients. Self-reported healthcare costs in these clusters were also calculated.

Result

Two disease clusters were identified: using the 1st approach were; 1) hypertension, diabetes, and mood disorder; 2) Neurotic, stress-related, and somatoform disorders, and acid peptic disease. Three clusters of patients identified using the 2nd approach were identified: 1) those with mood disorders and cardiometabolic, musculoskeletal, and thyroid diseases; 2) those with neurotic, substance use, and organic mental disorders, cancer, and epilepsy; and 3) those with Schizophrenia. Patients in Cluster 1 were taking more than six medicines and had more hospital visits. Within 18 months, 41 participants developed either one or two chronic conditions, most commonly diabetes, hypertension, or thyroid disease.

Conclusion

Cardiometabolic diseases are most commonly clustered with mood disorders. There is a need for blood pressure and sugar measurement in psychiatric clinics and mood disorder screening in cardiac, endocrinology, and primary care clinics.

Social determinants of multimorbidity patterns: A systematic review

Social determinants of multimorbidity are poorly understood in clinical practice. This review aims to characterize the different multimorbidity patterns described in the literature while identifying the social and behavioral determinants that may affect their emergence and subsequent evolution. We searched PubMed, Embase, Scopus, Web of Science, Ovid MEDLINE, CINAHL Complete, PsycINFO and Google Scholar. In total, 97 studies were chosen from the 48,044 identified. Cardiometabolic, musculoskeletal, mental, and respiratory patterns were the most prevalent. Cardiometabolic multimorbidity profiles were common among men with low socioeconomic status, while musculoskeletal, mental and complex patterns were found to be more prevalent among women. Alcohol consumption and smoking increased the risk of multimorbidity, especially in men. While the association of multimorbidity with lower socioeconomic status is evident, patterns of mild multimorbidity, mental and respiratory related to middle and high socioeconomic status are also observed. The findings of the present review point to the need for further studies addressing the impact of multimorbidity and its social determinants in population groups where this problem remains invisible (e.g., women, children, adolescents and young adults, ethnic groups, disabled population, older people living alone and/or with few social relations), as well as further work with more heterogeneous samples (i.e., not only focusing on older people) and using more robust methodologies for better classification and subsequent understanding of multimorbidity patterns. Besides, more studies focusing on the social determinants of multimorbidity and its inequalities are urgently needed in low- and middle-income countries, where this problem is currently understudied.

Can we design the next generation of digital health communication programs by leveraging the power of artificial intelligence to segment target audiences, bolster impact and deliver differentiated services? A machine learning analysis of survey data from rural India

OBJECTIVES

Direct to beneficiary (D2B) mobile health communication programmes have been used to provide reproductive, maternal, neonatal and child health information to women and their families in a number of countries globally. Programmes to date have provided the same content, at the same frequency, using the same channel to large beneficiary populations. This manuscript presents a proof of concept approach that uses machine learning to segment populations of women with access to phones and their husbands into distinct clusters to support differential digital programme design and delivery.

SETTING

Data used in this study were drawn from cross-sectional survey conducted in four districts of Madhya Pradesh, India.

PARTICIPANTS

Study participant included pregnant women with access to a phone (n=5095) and their husbands (n=3842) RESULTS: We used an iterative process involving K-Means clustering and Lasso regression to segment couples into three distinct clusters. Cluster 1 (n=1408) tended to be poorer, less educated men and women, with low levels of digital access and skills. Cluster 2 (n=666) had a mid-level of digital access and skills among men but not women. Cluster 3 (n=1410) had high digital access and skill among men and moderate access and skills among women. Exposure to the D2B programme 'Kilkari' showed the greatest difference in Cluster 2, including an 8% difference in use of reversible modern contraceptives, 7% in child immunisation at 10 weeks, 3% in child immunisation at 9 months and 4% in the timeliness of immunisation at 10 weeks and 9 months.

CONCLUSIONS

Findings suggest that segmenting populations into distinct clusters for differentiated programme design and delivery may serve to improve reach and impact.

TRIAL REGISTRATION NUMBER

NCT03576157.

Multi-Morbidity and Risk of Breast Cancer among Women in the UK Biobank Cohort

(Multi-)Morbidity shares common biological mechanisms or risk factors with breast cancer. This study aimed to investigate the association between the number of morbidities and patterns of morbidity and the risk of female breast cancer. Among 239,436 women (40-69 years) enrolled in the UK Biobank cohort who had no cancer history at baseline, we identified 35 self-reported chronic diseases at baseline. We assigned individuals into morbidity patterns using agglomerative hierarchical clustering analysis. We fitted Cox models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for breast cancer risk. In total, 58.4% of women had at least one morbidity, and the prevalence of multi-morbidity was 25.8%. During a median 7-year follow-up, there was no association between breast cancer risk (5326 cases) and either the number of morbidities or the identified clinically relevant morbidity patterns: no-predominant morbidity (reference), psychiatric morbidities (HR = 1.04, 95%CI 0.94-1.16), respiratory/immunological morbidities (HR = 0.98, 95%CI 0.90-1.07), cardiovascular/metabolic morbidities (HR = 0.93, 95%CI 0.81-1.06), and unspecific morbidities (HR = 0.98, 95%CI 0.89-1.07), overall. Among women younger than 50 years of age only, however, there was a significant association with psychiatric morbidity patterns compared to the no-predominant morbidity pattern (HR = 1.25, 95%CI 1.02-1.52). The other associations did not vary when stratifying by age at baseline and adherence to mammography recommendations. In conclusion, multi-morbidity was not a key factor to help identify patients at an increased risk of breast cancer.

Two-year change in latent classes of comorbidity among high-risk Veterans in primary care: a brief report

Background

Segmentation models such as latent class analysis are an increasingly popular approach to inform group-tailored interventions for high-risk complex patients. Multiple studies have identified clinically meaningful high-risk segments, but few have evaluated change in groupings over time.

Objectives

To describe population-level and individual change over time in latent comorbidity groups among Veterans at high-risk of hospitalization in the Veterans Health Administration (VA).

Research design

Using a repeated cross-sectional design, we conducted a latent class analysis of chronic condition diagnoses. We compared latent class composition, patient high-risk status, and patient class assignment in 2018 to 2020.

Subjects

Two cohorts of eligible patients were selected: those active in VA primary care and in the top decile of predicted one-year hospitalization risk in 2018 (n = 951,771) or 2020 (n = 978,771).

Measures

Medical record data were observed from January 2016–December 2020. Latent classes were modeled using indicators for 26 chronic health conditions measured with a 2-year lookback period from study entry.

Results

Five groups were identified in both years, labeled based on high prevalence conditions: Cardiometabolic (23% in 2018), Mental Health (18%), Substance Use Disorders (16%), Low Diagnosis (25%), and High Complexity (10%). The remaining 8% of 2018 patients were not assigned to a group due to low predicted probability. Condition prevalence overall and within groups was stable between years. However, among the 563,725 patients identified as high risk in both years, 40.8% (n = 230,185) had a different group assignment in 2018 versus 2020.

Conclusions

In a repeated latent class analysis of nearly 1 million Veterans at high-risk for hospitalization, population-level groups were stable over two years, but individuals often moved between groups. Interventions tailored to latent groups need to account for change in patient status and group assignment over time.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12913-022-08757-x.

Patterns of patients with polypharmacy in adult population from Korea

Polypharmacy and its rising global prevalence is a growing public health burden. Using a large representative nationwide Korean cohort (N = 761,145), we conducted a retrospective cross-sectional study aiming to identify subpopulations of patients with polypharmacy and characterize their unique patterns through cluster analysis. Patients aged ≥ 30 years who were prescribed at least one medication between 2014 and 2018 were included in our study. Six clusters were identified: cluster 1 mostly included patients who were hospitalized for a long time (4.3 ± 5.3 days); cluster 2 consisted of patients with disabilities (100.0%) and had the highest mean number of prescription drugs (7.7 ± 2.8 medications); cluster 3 was a group of low-income patients (99.9%); cluster 4 was a group of high-income patients (80.2%) who frequently (46.4 ± 25.9 days) visited hospitals/clinics (7.3 ± 2.7 places); cluster 5 was mostly elderly (74.9 ± 9.8 years) females (80.3%); and cluster 6 comprised mostly middle-aged (56.4 ± 1.5 years) males (88.6%) (all P < 0.001). Patients in clusters 1-5 had more prescribed medications and outpatient visit days than those in cluster 6 (all P < 0.001). Given limited health care resources, individuals with any of the identified phenotypes may be preferential candidates for participation in intervention programs for optimal medication use.

Patterns of multimorbidity and some psychiatric disorders: A systematic review of the literature

Objective

The presence of two or more chronic diseases results in worse clinical outcomes than expected by a simple combination of diseases. This synergistic effect is expected to be higher when combined with some conditions, depending on the number and severity of diseases. Multimorbidity is a relatively new term, with the first fundamental definitions appearing in 2015. Studies usually define it as the presence of at least two chronic medical illnesses. However, little is known regarding the relationship between mental disorders and other non-psychiatric chronic diseases. This review aims at investigating the association between some mental disorders and non-psychiatric diseases, and their pattern of association.

Methods

We performed a systematic approach to selecting papers that studied relationships between chronic conditions that included one mental disorder from 2015 to 2021. These were processed using Covidence, including quality assessment.

Results

This resulted in the inclusion of 26 papers in this study. It was found that there are strong associations between depression, psychosis, and multimorbidity, but recent studies that evaluated patterns of association of diseases (usually using clustering methods) had heterogeneous results. Quality assessment of the papers generally revealed low quality among the included studies.

Conclusions

There is evidence of an association between depressive disorders, anxiety disorders, and psychosis with multimorbidity. Studies that tried to examine the patterns of association between diseases did not find stable results.

Systematic review registration

https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42021216101, identifier: CRD42021216101.

Use of Multiple Correspondence Analysis and K-means to Explore Associations Between Risk Factors and Likelihood of Colorectal Cancer: Cross-sectional Study

BACKGROUND

Previous works have shown that risk factors are associated with an increased likelihood of colorectal cancer.

OBJECTIVE

The purpose of this study was to detect these associations in the region of Lleida (Catalonia) by using multiple correspondence analysis (MCA) and k-means.

METHODS

This cross-sectional study was made up of 1083 colorectal cancer episodes between 2012 and 2015, extracted from the population-based cancer registry for the province of Lleida (Spain), the Primary Care Centers database, and the Catalan Health Service Register. The data set included risk factors such as smoking and BMI as well as sociodemographic information and tumor details. The relations between the risk factors and patient characteristics were identified using MCA and k-means.

RESULTS

The combination of these techniques helps to detect clusters of patients with similar risk factors. Risk of death is associated with being elderly and obesity or being overweight. Stage III cancer is associated with people aged ≥65 years and rural/semiurban populations, while younger people were associated with stage 0.

CONCLUSIONS

MCA and k-means were significantly useful for detecting associations between risk factors and patient characteristics. These techniques have proven to be effective tools for analyzing the incidence of some factors in colorectal cancer. The outcomes obtained help corroborate suspected trends and stimulate the use of these techniques for finding the association of risk factors with the incidence of other cancers.

Sex Differences in Comorbidity Combinations in the Swedish Population

High comorbidity rates, especially mental–physical comorbidity, constitute an increasing health care burden, with women and men being differentially affected. To gain an overview of comorbidity rates stratified by sex across a range of different conditions, this study examines comorbidity patterns within and between cardiovascular, pulmonary, skin, endocrine, digestive, urogenital, musculoskeletal, neurological diseases, and psychiatric conditions. Self-report data from the LifeGene cohort of 31,825 participants from the general Swedish population (62.5% female, 18–84 years) were analyzed. Pairwise comorbidity rates of 54 self-reported conditions in women and men and adjusted odds ratios (ORs) for their comparison were calculated. Overall, the rate of pairwise disease combinations with significant comorbidity was higher in women than men (14.36% vs. 9.40%). Among psychiatric conditions, this rate was considerably high, with 41.76% in women and 39.01% in men. The highest percentages of elevated mental–physical comorbidity in women were found for musculoskeletal diseases (21.43%), digestive diseases (20.71%), and skin diseases (13.39%); in men, for musculoskeletal diseases (14.29%), neurological diseases (11.22%), and digestive diseases (10%). Implications include the need for integrating mental and physical health care services and a shift from a disease-centered to an individualized, patient-centered focus in clinical care.

Practical foundations of machine learning for addiction research. Part I. Methods and techniques

Machine learning assembles a broad set of methods and techniques to solve a wide range of problems, such as identifying individuals with substance use disorders (SUD), finding patterns in neuroimages, understanding SUD prognostic factors and their association, or determining addiction genetic underpinnings. However, the addiction research field underuses machine learning. This two-part narrative review focuses on machine learning tools and concepts, providing an introductory insight into their capabilities to facilitate their understanding and acquisition by addiction researchers. This first part presents supervised and unsupervised methods such as linear models, naive Bayes, support vector machines, artificial neural networks, and k-means. We illustrate each technique with examples of its use in current addiction research. We also present some open-source programming tools and methodological good practices that facilitate using these techniques. Throughout this work, we emphasize a continuum between applied statistics and machine learning, we show their commonalities, and provide sources for further reading to deepen the understanding of these methods. This two-part review is a primer for the next generation of addiction researchers incorporating machine learning in their projects. Researchers will find a bridge between applied statistics and machine learning, ways to expand their analytical toolkit, recommendations to incorporate well-established good practices in addiction data analysis (e.g., stating the rationale for using newer analytical tools, calculating sample size, improving reproducibility), and the vocabulary to enhance collaboration between researchers who do not conduct data analyses and those who do.

Clustering of physical health multimorbidity in people with severe mental illness: An accumulated prevalence analysis of United Kingdom primary care data

BACKGROUND

People with severe mental illness (SMI) have higher rates of a range of physical health conditions, yet little is known regarding the clustering of physical health conditions in this population. We aimed to investigate the prevalence and clustering of chronic physical health conditions in people with SMI, compared to people without SMI.

METHODS AND FINDINGS

We performed a cohort-nested accumulated prevalence study, using primary care data from the Clinical Practice Research Datalink (CPRD), which holds details of 39 million patients in the United Kingdom. We identified 68,783 adults with a primary care diagnosis of SMI (schizophrenia, bipolar disorder, or other psychoses) from 2000 to 2018, matched up to 1:4 to 274,684 patients without an SMI diagnosis, on age, sex, primary care practice, and year of registration at the practice. Patients had a median of 28.85 (IQR: 19.10 to 41.37) years of primary care observations. Patients with SMI had higher prevalence of smoking (27.65% versus 46.08%), obesity (24.91% versus 38.09%), alcohol misuse (3.66% versus 13.47%), and drug misuse (2.08% versus 12.84%) than comparators. We defined 24 physical health conditions derived from the Elixhauser and Charlson comorbidity indices and used logistic regression to investigate individual conditions and multimorbidity. We controlled for age, sex, region, and ethnicity and then additionally for health risk factors: smoking status, alcohol misuse, drug misuse, and body mass index (BMI). We defined multimorbidity clusters using multiple correspondence analysis (MCA) and K-means cluster analysis and described them based on the observed/expected ratio. Patients with SMI had higher odds of 19 of 24 conditions and a higher prevalence of multimorbidity (odds ratio (OR): 1.84; 95% confidence interval [CI]: 1.80 to 1.88, p < 0.001) compared to those without SMI, particularly in younger age groups (males aged 30 to 39: OR: 2.49; 95% CI: 2.27 to 2.73; p < 0.001; females aged 18 to 30: OR: 2.69; 95% CI: 2.36 to 3.07; p < 0.001). Adjusting for health risk factors reduced the OR of all conditions. We identified 7 multimorbidity clusters in those with SMI and 7 in those without SMI. A total of 4 clusters were common to those with and without SMI; while 1, heart disease, appeared as one cluster in those with SMI and 3 distinct clusters in comparators; and 2 small clusters were unique to the SMI cohort. Limitations to this study include missing data, which may have led to residual confounding, and an inability to investigate the temporal associations between SMI and physical health conditions.

CONCLUSIONS

In this study, we observed that physical health conditions cluster similarly in people with and without SMI, although patients with SMI had higher burden of multimorbidity, particularly in younger age groups. While interventions aimed at the general population may also be appropriate for those with SMI, there is a need for interventions aimed at better management of younger-age multimorbidity, and preventative measures focusing on diseases of younger age, and reduction of health risk factors.

Multimorbidity Profiles and Infection Severity in COVID-19 Population Using Network Analysis in the Andalusian Health Population Database

Identifying the population at risk of COVID-19 infection severity is a priority for clinicians and health systems. Most studies to date have only focused on the effect of specific disorders on infection severity, without considering that patients usually present multiple chronic diseases and that these conditions tend to group together in the form of multimorbidity patterns. In this large-scale epidemiological study, including primary and hospital care information of 166,242 patients with confirmed COVID-19 infection from the Spanish region of Andalusia, we applied network analysis to identify multimorbidity profiles and analyze their impact on the risk of hospitalization and mortality. Our results showed that multimorbidity was a risk factor for COVID-19 severity and that this risk increased with the morbidity burden. Individuals with advanced cardio-metabolic profiles frequently presented the highest infection severity risk in both sexes. The pattern with the highest severity associated in men was present in almost 28.7% of those aged ≥ 80 years and included associations between cardiovascular, respiratory, and metabolic diseases; age-adjusted odds ratio (OR) 95% confidence interval (1.71 (1.44–2.02)). In women, similar patterns were also associated the most with infection severity, in 7% of 65–79-year-olds (1.44 (1.34–1.54)) and in 29% of ≥80-year-olds (1.35 (1.18–1.53)). Patients with mental health patterns also showed one of the highest risks of COVID-19 severity, especially in women. These findings strongly recommend the implementation of personalized approaches to patients with multimorbidity and SARS-CoV-2 infection, especially in the population with high morbidity burden.

Identifying multimorbidity profiles associated with COVID-19 severity in chronic patients using network analysis in the PRECOVID Study

A major risk factor of COVID-19 severity is the patient's health status at the time of the infection. Numerous studies focused on specific chronic diseases and identified conditions, mainly cardiovascular ones, associated with poor prognosis. However, chronic diseases tend to cluster into patterns, each with its particular repercussions on the clinical outcome of infected patients. Network analysis in our population revealed that not all cardiovascular patterns have the same risk of COVID-19 hospitalization or mortality and that this risk depends on the pattern of multimorbidity, besides age and sex. We evidenced that negative outcomes were strongly related to patterns in which diabetes and obesity stood out in older women and men, respectively. In younger adults, anxiety was another disease that increased the risk of severity, most notably when combined with menstrual disorders in women or atopic dermatitis in men. These results have relevant implications for organizational, preventive, and clinical actions to help meet the needs of COVID-19 patients.

Empiric segmentation of high-risk patients: a structured literature review

OBJECTIVES

Empiric segmentation is a rapidly growing, learning health system approach that uses large health care system data sets to identify groups of high-risk patients who may benefit from similar interventions. We aimed to review studies that used data-driven approaches to segment high-risk patient populations and describe how their designs and findings can inform health care leaders who are interested in applying similar techniques to their patient populations.

STUDY DESIGN

Structured literature review.

METHODS

We searched for original research articles published since 2000 that identified high-risk adult patient populations and applied data-driven analyses to segment the population. Two reviewers independently extracted study population source and criteria for high-risk designation, segmentation method, data types included, model selection criteria, and model results from the identified studies.

RESULTS

Our search identified 224 articles, 12 of which met criteria for full review. Of these, 8 segmented high-risk patients and 4 segmented diagnoses without assigning patients to unique groups. Studies segmenting patients more often had clinically interpretable results. Common groups were defined by high prevalence of diabetes, cardiovascular disease, psychiatric conditions including substance use disorders, and neurologic disease (eg, stroke). Few studies incorporated patients' functional or social factors. Resulting patient and diagnosis clusters varied in ways closely linked to the model inputs, patient population inclusion criteria, and health care system context.

CONCLUSIONS

Empiric segmentation can yield clinically relevant groups of patients with complex medical needs. Segmentation results are context dependent, suggesting the need for careful design and interpretation of segmentation models to ensure that results can inform clinical care and program design in the target setting.

Multimorbidity patterns of chronic conditions and geriatric syndromes in older patients from the MoPIM multicentre cohort study

OBJECTIVES

To estimate the frequency of chronic conditions and geriatric syndromes in older patients admitted to hospital because of an exacerbation of their chronic conditions, and to identify multimorbidity clusters in these patients.

DESIGN

Multicentre, prospective cohort study.

SETTING

Internal medicine or geriatric services of five general teaching hospitals in Spain.

PARTICIPANTS

740 patients aged 65 and older, hospitalised because of an exacerbation of their chronic conditions between September 2016 and December 2018.

PRIMARY AND SECONDARY OUTCOME MEASURES

Active chronic conditions and geriatric syndromes (including risk factors) of the patient, a score about clinical management of chronic conditions during admission, and destination at discharge were collected, among other variables. Multimorbidity patterns were identified using fuzzy c-means cluster analysis, taking into account the clinical management score. Prevalence, observed/expected ratio and exclusivity of each chronic condition and geriatric syndrome were calculated for each cluster, and the final solution was approved after clinical revision and discussion among the research team.

RESULTS

740 patients were included (mean age 84.12 years, SD 7.01; 53.24% female). Almost all patients had two or more chronic conditions (98.65%; 95% CI 98.23% to 99.07%), the most frequent were hypertension (81.49%, 95% CI 78.53% to 84.12%) and heart failure (59.86%, 95% CI 56.29% to 63.34%). The most prevalent geriatric syndrome was polypharmacy (79.86%, 95% CI 76.82% to 82.60%). Four statistically and clinically significant multimorbidity clusters were identified: osteoarticular, psychogeriatric, cardiorespiratory and minor chronic disease. Patient-level variables such as sex, Barthel Index, number of chronic conditions or geriatric syndromes, chronic disease exacerbation 3 months prior to admission or destination at discharge differed between clusters.

CONCLUSIONS

In older patients admitted to hospital because of the exacerbation of chronic health problems, it is possible to define multimorbidity clusters using soft clustering techniques. These clusters are clinically relevant and could be the basis to reorganise healthcare circuits or processes to tackle the increasing number of older, multimorbid patients.

TRIAL REGISTRATION NUMBER

NCT02830425.

Analysis of multimorbidity networks associated with different factors in Northeast China: a cross-sectional analysis

OBJECTIVES

This study aimed to identify and study the associations and co-occurrence of multimorbidity, and assessed the associations of diseases with sex, age and hospitalisation duration.

DESIGN

Cross-sectional.

SETTING

15 general hospitals in Jilin Province, China.

PARTICIPANTS

A total of 431 295 inpatients were enrolled through a cross-sectional study in Jilin Province, China.

PRIMARY OUTCOME MEASURES

The complex relationships of multimorbidity were presented as weighted networks.

RESULTS

The distributions of the numbers of diseases differed significantly by sex, age and hospitalisation duration (p<0.001). Cerebrovascular diseases (CD), hypertensive diseases (HyD), ischaemic heart diseases (IHD) and other forms of heart disease (OFHD) showed the highest weights in the multimorbidity networks. The connections between different sexes or hospitalisation duration and diseases were similar, while those between different age groups and diseases were different.

CONCLUSIONS

CD, HyD, IHD and OFHD were the central points of disease clusters and directly or indirectly related to other diseases or factors. Thus, effective interventions for these diseases should be adopted. Furthermore, different intervention strategies should be developed according to multimorbidity patterns in different age groups.

Exploring patient multimorbidity and complexity using health insurance claims data: a cluster analysis approach (Preprint)

Background

Although the trend of progressing morbidity is widely recognized, there are numerous challenges when studying multimorbidity and patient complexity. For multimorbid or complex patients, prone to fragmented care and high health care use, novel estimation approaches need to be developed.

Objective

This study aims to investigate the patient multimorbidity and complexity of Swiss residents aged ≥50 years using clustering methodology in claims data.

Methods

We adopted a clustering methodology based on random forests and used 34 pharmacy-based cost groups as the only input feature for the procedure. To detect clusters, we applied hierarchical density-based spatial clustering of applications with noise. The reasonable hyperparameters were chosen based on various metrics embedded in the algorithms (out-of-bag misclassification error, normalized stress, and cluster persistence) and the clinical relevance of the obtained clusters.

Results

Based on cluster analysis output for 18,732 individuals, we identified an outlier group and 7 clusters: individuals without diseases, patients with only hypertension-related diseases, patients with only mental diseases, complex high-cost high-need patients, slightly complex patients with inexpensive low-severity pharmacy-based cost groups, patients with 1 costly disease, and older high-risk patients.

Conclusions

Our study demonstrated that cluster analysis based on pharmacy-based cost group information from claims-based data is feasible and highlights clinically relevant clusters. Such an approach allows expanding the understanding of multimorbidity beyond simple disease counts and can identify the population profiles with increased health care use and costs. This study may foster the development of integrated and coordinated care, which is high on the agenda in policy making, care planning, and delivery.

Combined Multimorbidity and Polypharmacy Patterns in the Elderly: A Cross-Sectional Study in Primary Health Care

(1) Background: The acquisition of multiple chronic diseases, known as multimorbidity, is common in the elderly population, and it is often treated with the simultaneous consumption of several prescription drugs, known as polypharmacy. These two concepts are inherently related and cause an undue burden on the individual. The aim of this study was to identify combined multimorbidity and polypharmacy patterns for the elderly population in Catalonia. (2) Methods: A cross-sectional study using electronic health records from 2012 was conducted. A mapping process was performed linking chronic disease categories to the drug categories indicated for their treatment. A soft clustering technique was then carried out on the final mapped categories. (3) Results: 916,619 individuals were included, with 93.1% meeting the authors’ criteria for multimorbidity and 49.9% for polypharmacy. A seven-cluster solution was identified: one non-specific (Cluster 1) and six specific, corresponding to diabetes (Cluster 2), neurological and musculoskeletal, female dominant (Clusters 3 and 4) and cardiovascular, cerebrovascular and renal diseases (Clusters 5 and 6), and multi-system diseases (Cluster 7). (4) Conclusions: This study utilized a mapping process combined with a soft clustering technique to determine combined patterns of multimorbidity and polypharmacy in the elderly population, identifying overrepresentation in six of the seven clusters with chronic disease and chronic disease-drug categories. These results could be applied to clinical practice guidelines in order to better attend to patient needs. This study can serve as the foundation for future longitudinal regarding relationships between multimorbidity and polypharmacy.

Multimorbidity clusters among long-term breast cancer survivors in Spain: Results of the SURBCAN study

The disease management of long-term breast cancer survivors (BCS) is hampered by the scarce knowledge of multimorbidity patterns. The aim of our study was to identify multimorbidity clusters among long-term BCS and assess their impact on mortality and health services use. We conducted a retrospective study using electronic health records of 6512 BCS from Spain surviving at least 5 years. Hierarchical cluster analysis was used to identify groups of similar patients based on their chronic diagnoses, which were assessed using the Clinical Classifications Software. As a result, multimorbidity clusters were obtained, clinically defined and named according to the comorbidities with higher observed/expected prevalence ratios. Multivariable Cox and negative binomial regression models were fitted to estimate overall mortality risk and probability of contacting health services according to the clusters identified. 83.7% of BCS presented multimorbidity, essential hypertension (34.5%) and obesity and other metabolic disorders (27.4%) being the most prevalent chronic diseases at the beginning of follow-up. Five multimorbidity clusters were identified: C1-unspecific (29.9%), C2-metabolic and neurodegenerative (28.3%), C3-anxiety and fractures (9.7%), C4-musculoskeletal and cardiovascular (9.6%) and C5-thyroid disorders (5.3%). All clusters except C5-thyroid disorders were associated with higher mortality compared to BCS without comorbidities. The risk of mortality in C4 was increased by 64% (adjusted hazard ratio 1.64, 95% confidence interval 1.52-2.07). Stratified analysis showed an increased risk of death among BCS with 5 to 10 years of survival in all clusters. These results help to identify subgroups of long-term BCS with specific needs and mortality risks and to guide BCS clinical practice regarding multimorbidity.

Heterogeneity in COVID-19 Patients at Multiple Levels of Granularity: From Biclusters to Clinical Interventions

Several studies have shown that COVID-19 patients with prior comorbidities have a higher risk for adverse outcomes, resulting in a disproportionate impact on older adults and minorities that fit that profile. However, although there is considerable heterogeneity in the comorbidity profiles of these populations, not much is known about how prior comorbidities co-occur to form COVID-19 patient subgroups, and their implications for targeted care. Here we used bipartite networks to quantitatively and visually analyze heterogeneity in the comorbidity profiles of COVID-19 inpatients, based on electronic health records from 12 hospitals and 60 clinics in the greater Minneapolis region. This approach enabled the analysis and interpretation of heterogeneity at three levels of granularity (cohort, subgroup, and patient), each of which enabled clinicians to rapidly translate the results into the design of clinical interventions. We discuss future extensions of the multigranular heterogeneity framework, and conclude by exploring how the framework could be used to analyze other biomedical phenomena including symptom clusters and molecular phenotypes, with the goal of accelerating translation to targeted clinical care.

Profiles of Frailty among Older People Users of a Home-Based Primary Care Service in an Urban Area of Barcelona (Spain): An Observational Study and Cluster Analysis

Background: The multidimensional assessment of frailty allows stratifying it into degrees; however, there is still heterogeneity in the characteristics of people in each stratum. The aim of this study was to identify frailty profiles of older people users of a home-based primary care service. Methods: We carried out an observational study from January 2018 to January 2021. Participants were all people cared for a home-based primary care service. We performed a cluster analysis by applying a k-means clustering technique. Cluster labeling was determined with the 22 variables of the Frail-VIG index, age, and sex. We computed multiple indexes to assess the optimal number of clusters, and this was selected based on a clinical assessment of the best options. Results: Four hundred and twelve participants were clustered into six profiles. Three of these profiles corresponded to a moderate frailty degree, two to a severe frailty degree and one to a mild frailty degree. In addition, almost 75% of the participants were clustered into three profiles which corresponded to mild and moderate degree of frailty. Conclusions: Different profiles were found within the same degree of frailty. Knowledge of these profiles can be useful in developing strategies tailored to these differentiated care needs.

Patterns of Multimorbidity in a Population-Based Cohort of Older People: Sociodemographic, Lifestyle, Clinical, and Functional Differences

BACKGROUND

The aim of this study is to identify clusters of older persons based on their multimorbidity patterns and to analyze differences among clusters according to sociodemographic, lifestyle, clinical, and functional characteristics.

METHODS

We analyzed data from the Swedish National Study on Aging and Care in Kungsholmen on 2,931 participants aged 60 years and older who had at least two chronic diseases. Participants were clustered by the fuzzy c-means cluster algorithm. A disease was considered to be associated with a given cluster when the observed/expected ratio was ≥2 or the exclusivity was ≥25%.

RESULTS

Around half of the participants could be classified into five clinically meaningful clusters: respiratory and musculoskeletal diseases (RESP-MSK) 15.7%, eye diseases and cancer (EYE-CANCER) 10.7%, cognitive and sensory impairment (CNS-IMP) 10.6%, heart diseases (HEART) 9.3%, and psychiatric and respiratory diseases (PSY-RESP) 5.4%. Individuals in the CNS-IMP cluster were the oldest, with the worst function and more likely to live in a nursing home; those in the HEART cluster had the highest number of co-occurring diseases and drugs, and they exhibited the highest mean values of serum creatinine and C-reactive protein. The PSY-RESP cluster was associated with higher levels of alcoholism and neuroticism. The other half of the cohort was grouped in an unspecific cluster, which was characterized by gathering the youngest individuals, with the lowest number of co-occurring diseases, and the best functional and cognitive status.

CONCLUSIONS

The identified multimorbidity patterns provide insight for setting targets for secondary and tertiary preventative interventions and for designing care pathways for multimorbid older people.

Multimorbidity burden and dementia risk in older adults: The role of inflammation and genetics

Introduction

We investigate dementia risk in older adults with different disease patterns and explore the role of inflammation and apolipoprotein E (APOE) genotype.

Methods

A total of 2,478 dementia‐free participants with two or more chronic diseases (ie, multimorbidity) part of the Swedish National study on Aging and Care in Kungsholmen (SNAC‐K) were grouped according to their multimorbidity patterns and followed to detect clinical dementia. The potential modifier effect of C‐reactive protein (CRP) and apolipoprotein E (APOE) genotype was tested through stratified analyses.

Results

People with neuropsychiatric, cardiovascular, and sensory impairment/cancer multimorbidity had increased hazards for dementia compared to the unspecific (Hazard ration (HR) 1.66, 95% confidence interval [CI] 1.13‐2.42; 1.61, 95% CI 1.17‐2.29; 1.32, 95% CI 1.10‐1.71, respectively). Despite the lack of statistically significant interaction, high CRP increased dementia risk within these patterns, and being APOE ε4 carriers heightened dementia risk for neuropsychiatric and cardiovascular multimorbidity.

Discussion

Individuals with neuropsychiatric, cardiovascular, and sensory impairment/cancer patterns are at increased risk for dementia and APOE ε4, and inflammation may further increase the risk. Identifying such high‐risk groups might allow tailored interventions for dementia prevention.

Five-year trajectories of multimorbidity patterns in an elderly Mediterranean population using Hidden Markov Models

This study aimed to analyse the trajectories and mortality of multimorbidity patterns in patients aged 65 to 99 years in Catalonia (Spain). Five year (2012–2016) data of 916,619 participants from a primary care, population-based electronic health record database (Information System for Research in Primary Care, SIDIAP) were included in this retrospective cohort study. Individual longitudinal trajectories were modelled with a Hidden Markov Model across multimorbidity patterns. We computed the mortality hazard using Cox regression models to estimate survival in multimorbidity patterns. Ten multimorbidity patterns were originally identified and two more states (death and drop-outs) were subsequently added. At baseline, the most frequent cluster was the Non-Specific Pattern (42%), and the least frequent the Multisystem Pattern (1.6%). Most participants stayed in the same cluster over the 5 year follow-up period, from 92.1% in the Nervous, Musculoskeletal pattern to 59.2% in the Cardio-Circulatory and Renal pattern. The highest mortality rates were observed for patterns that included cardio-circulatory diseases: Cardio-Circulatory and Renal (37.1%); Nervous, Digestive and Circulatory (31.8%); and Cardio-Circulatory, Mental, Respiratory and Genitourinary (28.8%). This study demonstrates the feasibility of characterizing multimorbidity patterns along time. Multimorbidity trajectories were generally stable, although changes in specific multimorbidity patterns were observed. The Hidden Markov Model is useful for modelling transitions across multimorbidity patterns and mortality risk. Our findings suggest that health interventions targeting specific multimorbidity patterns may reduce mortality in patients with multimorbidity.

Untangling the complexity of multimorbidity with machine learning

The prevalence of multimorbidity has been increasing in recent years, posing a major burden for health care delivery and service. Understanding its determinants and impact is proving to be a challenge yet it offers new opportunities for research to go beyond the study of diseases in isolation. In this paper, we review how the field of machine learning provides many tools for addressing research challenges in multimorbidity. We highlight recent advances in promising methods such as matrix factorisation, deep learning, and topological data analysis and how these can take multimorbidity research beyond cross-sectional, expert-driven or confirmatory approaches to gain a better understanding of evolving patterns of multimorbidity. We discuss the challenges and opportunities of machine learning to identify likely causal links between previously poorly understood disease associations while giving an estimate of the uncertainty on such associations. We finally summarise some of the challenges for wider clinical adoption of machine learning research tools and propose some solutions.

Patterns of Multimorbidity in Adults: An Association Rules Analysis Using the Korea Health Panel

This study aimed to identify the prevalence and patterns of multimorbidity among Korean adults. A descriptive study design was used. Of 11,232 adults aged 18 and older extracted from the 2014 Korean Health Panel Survey, 7118 had one or more chronic conditions. The chronic conditions code uses the Korean Standard Classification of Diseases. Association rule analysis and network analysis were conducted to identify patterns of multimorbidity among 4922 participants with multimorbidity. The prevalence of multimorbidity in the overall population was 34.8%, with a higher prevalence among women (40.8%) than men (28.6%). Hypertension had the highest prevalence in both men and women. In men, diabetes mellitus and hypertension yielded the highest probability of comorbidity (10.04%). In women, polyarthrosis and hypertension yielded the highest probability of comorbidity (12.51%). The results of the network analysis in four groups divided according to gender and age showed different characteristics for each group. Public health practitioners should adopt an integrated approach to manage multimorbidity rather than an individual disease-specific approach, along with different strategies according to age and gender groups.

Cluster analysis of nutritional factors associated with low muscle mass index in middle-aged and older adults

BACKGROUND & AIMS

Sarcopenia, or age-related muscle loss, is an enormous health problem in an aging world because of its many clinical and societal adverse effects. The uncovering of healthy dietary patterns is an important strategy to prevent or delay sarcopenia. We used K-means clustering to identify subgroups of men and women based on nutritional and health-related factors and investigated risk factors for low muscle mass in the subgroups and in the study population as a whole.

METHODS

We analyzed a total 10,863 participants over 40 years of age who participated in the Korea National Health and Nutrition Survey from 2008 to 2011. Dual energy X-ray absorptiometry was used to determine the appendicular lean mass (ALM) of the participants. Participants with low ALM adjusted BMI (ALM/BMI) were then identified using the criteria of the Foundation for the National Institutes of Health sarcopenia project. K-means clustering and multivariate logistic regression were used to analyze associations between nutritional and health-related variables and low ALM/BMI in the population as a whole and in the individual clusters.

RESULTS

A total of 712 (15.8%) men and 869 (13.7%) women had low ALM/BMI. Five clusters were identified in men and women, respectively. Two clusters of men and one cluster of women exhibited an increased risk of low ALM/BMI. Old age, low total energy intake, low levels of physical activity, and a high number of chronic diseases were consistent risk factors for low ALM/BMI in all Korean men and women. Low protein was a common risk factor for low ALM/BMI in men. After dividing all subjects by the K-means clustering algorithm, two risk factors (high fat intake and smoking) and four factors (low intakes of carbohydrate, protein and fat, and high alcohol consumption) were additionally proposed in Korean men and women, respectively.

CONCLUSIONS

Age, low total energy intake, low level of physical activity, and an increased number of chronic diseases were consistent risk factors for low ALM/BMI in men and women. Cluster-specific risk factors were also noted in men and women.

Multimorbidity of chronic non-communicable diseases and its models of care in low- and middle-income countries: a scoping review protocol

INTRODUCTION

Multimorbidity is the coexistence of two or more chronic non-communicable diseases (NCDs) in a given individual. Multimorbidity is increasing in low- and middle-income countries (LMICs) and challenging health systems. Individuals with multimorbidity are facing the risk of premature mortality, lower quality of life and greater use of healthcare services. However, despite the huge challenge multimorbidity brings in LMICs, gaps remain in mapping and synthesising the available knowledge on the issue. The focus of this scoping review will be to synthesise the extent, range and nature of studies on the epidemiology and models of multimorbidity care in LMICs.

METHODS

PubMed (MEDLINE) will be the main database to be searched. For articles that are not indexed in the PubMed, Scopus, PsycINFO and Cochrane databases will be searched. Grey literature databases will also be explored. There will be no restrictions on study setting or year of publication. Articles will be searched using key terms, including comorbidity, co-morbidity, multimorbidity, multiple chronic conditions and model of care. Relevant articles will be screened by two independent reviewers and data will be charted accordingly. The result of this scoping review will be presented using the Preferred Reporting Items for Systematic Review and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) checklist and reporting guideline.

ETHICS AND DISSEMINATION

This scoping review does not require ethical approval. Findings will be published in peer-reviewed journal and presented at scientific conferences.

Soft clustering using real-world data for the identification of multimorbidity patterns in an elderly population: cross-sectional study in a Mediterranean population

OBJECTIVES

The aim of this study was to identify, with soft clustering methods, multimorbidity patterns in the electronic health records of a population ≥65 years, and to analyse such patterns in accordance with the different prevalence cut-off points applied. Fuzzy cluster analysis allows individuals to be linked simultaneously to multiple clusters and is more consistent with clinical experience than other approaches frequently found in the literature.

DESIGN

A cross-sectional study was conducted based on data from electronic health records.

SETTING

284 primary healthcare centres in Catalonia, Spain (2012).

PARTICIPANTS

916 619 eligible individuals were included (women: 57.7%).

PRIMARY AND SECONDARY OUTCOME MEASURES

We extracted data on demographics, International Classification of Diseases version 10 chronic diagnoses, prescribed drugs and socioeconomic status for patients aged ≥65. Following principal component analysis of categorical and continuous variables for dimensionality reduction, machine learning techniques were applied for the identification of disease clusters in a fuzzy c-means analysis. Sensitivity analyses, with different prevalence cut-off points for chronic diseases, were also conducted. Solutions were evaluated from clinical consistency and significance criteria.

RESULTS

Multimorbidity was present in 93.1%. Eight clusters were identified with a varying number of disease values: nervous and digestive; respiratory, circulatory and nervous; circulatory and digestive; mental, nervous and digestive, female dominant; mental, digestive and blood, female oldest-old dominant; nervous, musculoskeletal and circulatory, female dominant; genitourinary, mental and musculoskeletal, male dominant; and non-specified, youngest-old dominant. Nuclear diseases were identified for each cluster independently of the prevalence cut-off point considered.

CONCLUSIONS

Multimorbidity patterns were obtained using fuzzy c-means cluster analysis. They are clinically meaningful clusters which support the development of tailored approaches to multimorbidity management and further research.