https://orcid.org/0000-0002-0015-0725
Figures
Abstract
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.
Citation: Lleal M, Corral-Vazquez C, Baré M, Comet R, Herranz S, Baigorri F, et al. (2023) Multimorbidity patterns in COVID-19 patients and their relationship with infection severity: MRisk-COVID study. PLoS ONE 18(8): e0290969. https://doi.org/10.1371/journal.pone.0290969
Editor: Ikechukwu Innocent Mbachu, Nnamdi Azikiwe University, NIGERIA
Received: June 14, 2022; Accepted: August 19, 2023; Published: August 31, 2023
Copyright: © 2023 Lleal et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Data cannot be shared publicly because of confidentiality. Data used for this analysis are available from the Government of Catalonia Institutional Data Access (PADRIS Program) for researchers who meet the criteria for access to confidential data. Inquiries of access to aggregated data should be addressed to AQuAS director (direccio.aquas@gencat.cat), under the following conditions: http://aquas.gencat.cat/ca/ambits/analitica-dades/padris/.
Funding: MaB received funding from Institut d'Investigació i Innovació Parc Taulí (http://www.tauli.cat/institut/) [Grant number: CIR2020/023] and Red de Investigación en Servicios de Salud en Enfermedades Crónicas (REDISSEC) (http://www.redissec.com/) [Grant number: RD16/0001/0002]. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Abbreviations: AQuAS, Agency for Health Quality and Assessment of Catalonia; CC, Chronic condition; COPD, Chronic obstructive pulmonary disease; CV, Cardiovascular; DM, Diabetes mellitus; HC3, Shared Clinical History of Catalonia; ICD-10-CM, International Classification of Diseases, 10th edition, Clinical Modification; ICD-10-PCS, International Classification of Diseases, 10th edition, Procedure Coding System; MM, Multimorbidity; O/E, Observed/Expected; PADRIS, Public Data Analysis for Health Research and Innovation Program; SUVEC, Catalan Epidemiological Surveillance Emergency Service
Background
The first wave of the COVID-19 pandemic caused high rates of death and critical infection. The first studies quickly identified a differential vulnerability depending on age and sex, so that higher rates of death and severity were detected in older and male individuals [1, 2].
The study of the relationship between isolated chronic diseases and COVID-19 infection has resulted in the identification of diverse morbidities as potential risk factors, such as diabetes mellitus (DM), obesity, hypertension, heart failure, chronic obstructive pulmonary disease (COPD), chronic kidney disease, and cardiovascular (CV) conditions [3–10]. Noticeably, many of these conditions share some pathophysiological pathways, suggesting that their interactions could increase the risk of a poor COVID-19 prognosis, and that their identification would help in the management of the disease. Nevertheless, few research works have examined the interaction between multiple diseases and its impact on COVID-19 outcomes [8, 11–13].
Multimorbidity (MM), which is traditionally defined as the concomitance of two or more chronic conditions (CC) in the same individual, is a highly prevalent issue that increases with age and society ageing [14–17]. Multimorbid patients may present complex profiles associated to pathophysiological interactions that could complicate medical treatment [18, 19]. However, there is no gold standard in the characterization of MM due to the novelty of the research field, and many studies analyse the mere presence or absence of MM, the sole number of CC, pre-established scores such as the Charlson Multimorbidity Index, or the aggregation of CC by factor analysis or hard hierarchical-clustering methods [20–22].
Remarkably, new strategies that more accurately reflect the complex, non-random association of CC in a patient-centred point of view are emerging, such as soft (or fuzzy) clustering analysis [23–25]. This technique has already provided relevant results in previous studies, so that several relationships between MM patterns and clinically relevant outcomes such as disability, adverse drug reactions, hospitalization, or mortality have been identified [26–30].
In order to perform a solid profiling of the MM patterns, detailed medical information is needed. In this sense, the use of Big Data provides a quick and exhaustive covering of large datasets. Indeed, the access to public healthcare Big Data facilitated by several state organisms enabled rapid analyses, yielding relevant results, and leading to a better handling of the pandemic [13, 31–33].
Given the previous statements, the aim of the present study, encompassed in the MRisk-COVID Big Data project, was to provide an age- and sex-centred characterization of the MM clusters of an adult COVID-19 cohort and to assess the relationship between these clusters and the severity and mortality of the infection.
Methods
Study design and cohort
The MRisk-COVID study is an observational Big Data research project based on a cohort of 14286 COVID-19 patients aged >20 years, residing in a healthcare zone of about 400000 inhabitants in the Northeast region of Spain (Vallès Occidental est). This healthcare zone is situated within the autonomous community of Catalonia, specifically within the province of Barcelona, and encompasses a single reference hospital, Parc Taulí Hospital Universitari. Data were provided by the Agency for Health Quality and Assessment of Catalonia (AQuAS) in the framework of the Public Data Analysis for Health Research and Innovation Program (PADRIS), under an extraordinary call that took place in April 2020 in order to provide healthcare Big Data for COVID-19 research studies.
The cohort of this study is composed by all COVID-19 cases registered between 27th February and 15th June of 2020 by the Catalan Epidemiological Surveillance Emergency Service (SUVEC) as either confirmed cases (identified through positive PCR, ELISA or rapid antibody test), or suspected cases (no positive test result, but exhibited symptoms and were classified as potential cases by physicians). The study did not involve sampling, as it encompasses all individuals who met the inclusion criteria. The International Classification of Diseases, 10th edition, Clinical Modification (ICD-10-CM) codes considered as COVID-19 diagnoses are summarized in S1 Table.
Ethics statement
The study was approved by the Ethics Committee (CEIm) of the Parc Taulí University Hospital (reference 2021/5067), which waived the requirement of informed consents due to the epidemiological nature of the study and the use of anonymized data provided by PADRIS. This study was conducted according to the principles expressed in the Declaration of Helsinki.
Big data processing, data linkage and study outcomes
Demographic information and clinical records were compiled, integrated and analysed. Demographic data were obtained from the Shared Clinical History of Catalonia (HC3) and included age and sex. In order to prevent re-identification of the patients, age was provided by PADRIS in quinquennia, and groups with high risk of re-identification were not included (women aged >95 and men aged >90). Clinical records were obtained from any primary healthcare centre and hospital of Catalonia that the individuals attended. The provided records covered from 1923 to 2020 in primary healthcare centres, and from 2016 to 2020 in hospitals. These data encompassed all diagnoses and hospital procedures, including the index hospitalization episode of COVID-19 infection. Information about admission and discharge circumstances was also provided (S1 Fig). Patients’ data were stratified by sex and age, thereby obtaining eight different datasets (groups of 21–45, 46–65, 66–80 and 81–95 years, separated into male and female individuals).
Infection severity was defined as the occurrence of at least one of the following conditions during any of the registered COVID-19 episodes: severe respiratory affection (including insufficiency, failure, or distress); use of respiratory support (including mechanical ventilation or oxygen therapy); septic shock; multiple organ failure (the combination of respiratory failure and any other organ failure); inflammatory response; admission to intensive care unit; and mortality associated with COVID-19 episodes (either registered by epidemiological surveillance, primary care or hospital records). This definition was based on the COVID-19 Treatment Guidelines provided by National Institutes of Health [34], with some modifications under the research group physicians’ criteria. The ICD-10-CM and ICD-10 Procedure Coding System (PCS) codes considered as severity conditions are compiled in S1 Table. Besides, mortality was also assessed independently.
Data compilation, processing and statistical analysis were performed using R [35] and RStudio [36].
Multimorbidity cluster analysis
The ICD-10-CM codes of the complete diagnoses records were categorized into chronic or acute diseases by the Chronic Condition Indicator v.2021.1 [37], and only CC were selected. Then, in order to reduce the number of variables for the cluster analysis, diagnoses were grouped and classified by the Clinical Classification Software Refined v.2021.1 [37]. This software provides a hierarchical classification of the diagnoses into general CC categories (e.g. Asthma, or Chronic obstructive pulmonary disease and bronchiectasis), which, in turn, belong to disease families (e.g. Diseases of the respiratory system). The CC of the family Neoplasms were manually distributed into two general categories: Neoplasia (solid tumour), and Hematologic neoplasia.
The clustering procedure was independently applied to each sex-age group. For each stratum, CC were filtered by prevalence (>2%) and only patients with MM, as defined by the presence of 2 or more CC, were included in the analyses. The clustering methodology was adapted from the one described by Vetrano et al [28]. An initial dimension reduction of the CC datasets was performed by a Multiple Correspondence Analysis, using the elbow criteria in the Scree plot for dimension selection. The resulting data were subjected to soft clustering analysis by the fuzzy c-means algorithm. This algorithm requires two main parameters: the number of desired clusters (C) and the fuzziness (m), which indicates the degree of overlapping membership of the patients to the clusters. This parameter can range from 1 (equivalent to hard, non-overlapping clustering) to infinite. Several values of C (4, 5, 6 and 7) and m (1.1, 1.2, 1.4, 1.5, 2, and 4) were tested. Since fuzzy c-means is a stochastic method, 100 iterations were performed for each combination of C and m in order to obtain mean and reproducible results. The optimal m was estimated by the mean calculation of five indexes: Calinski–Harabasz, Partition coefficient (both optimal at their highest value), Partition entropy, Fukuyama, and Xie-Beni (the three of them optimal at their lowest value).
The final C for each sex and age stratum was determined through consensus among 11 professionals of the multidisciplinary research group. This agreement was achieved following a Delphi-like selection method that consisted on equal voting through two selection rounds. Consensus was defined when one of the options reached a majority of votes (≥6); cases with lack of consensus or similar voting count were decided through a clinical debate session.
Several indicators were calculated for the CC in each cluster in order to characterize the specificity and composition of the MM patterns: a) Prevalence within the cluster (%); b) Observed/Expected (O/E) ratio, estimated by dividing the prevalence in the cluster by the prevalence in the corresponding age-sex group cohort; and c) Exclusivity (%), obtained by dividing the number of patients that presented the CC in the cluster by the total number of patients affected by the CC in the corresponding age-sex group cohort. A generic label was assigned to each cluster in an attempt to summarize the most prevalent and overexpressed CC as well as to facilitate the clinical interpretation of the results. Regarding patients, membership percentages were calculated, indicating the degree of belonging of each patient to the selected clusters.
Statistical analysis
Statistical bivariate differences between severe and non-severe cases, as well as dead and survivor cases, were measured by chi-square tests in the case of categorical/dichotomous variables (age, sex, and MM) and Wilcoxon signed-rank tests in the case of continuous variables (number of CC).
The statistical relationship between mortality / severity and the MM clusters was assessed by weighted chi-square tests, where the weight of the variables was the percentage of membership of the patients to each cluster. Patients with only one or no CC were evaluated as an additional “No multimorbidity” group. These groups were discarded from the analysis of the 66–80 and 81–95 years strata, due to the scarce number of non-multimorbid cases.
Results
Characteristics of the study population, infection severity and mortality
The study cohort comprised a total of 5411 males (37.9%) and 8875 females (62.1%). The age distribution revealed a mode of 41–45 years (n = 1438). Among the total cohort, 3140 patients had a confirmed COVID-19 infection (22.0%) and 11 146 were suspected cases (78.0%). The 78.9% of the cohort presented MM, and the mean number of CC was 6.0 (standard deviation of 5.4) (Table 1).
SD = standard deviation; OR = Odds Ratio; CI = Confidence Interval.
According to the selected codes, 1352 of the patients (9.5%; 95% CI: 9.0–9.9) suffered a severe infection. The most frequent severity criteria were oxygen therapy (n = 891, 6.2%), and respiratory failure (n = 880, 6.2%). Regarding isolated mortality, 555 patients (3.9%; 95% CI: 3.6–4.2) died during a COVID-19 associated episode, with 368 cases (66.3%) occurring at a hospital.
The relationship between severity / mortality and the demographic and medical variables (age, sex, MM, and number of CC), identified by bivariate statistical analyses, is presented in Table 1. Descriptive statistics of severe COVID-19 infection stratified by confirmed or suspected cases are shown in S2 Table.
Chronic conditions of the COVID-19 patients
A total of 215 CC were identified, belonging to 20 different families of diseases. The most represented families were Diseases of the circulatory system (which contained 33 CC), Diseases of the genitourinary system (22 CC), Mental, behavioural and neurodevelopmental disorders (22 CC), and Diseases of the musculoskeletal system and connective tissue (21 CC). The most prevalent CC were Essential hypertension (30.3%), Anxiety and fear-related disorders (26.62%), Obesity (21.8%), and Disorders of lipid metabolism (21.3%).
For each age-sex stratum, different sets of CC with prevalence equal to or greater than 2% were identified and are listed in S3 Table.
Multimorbidity patterns
Fuzzy c-means cluster analysis was performed with a m value of 1.1, and C values were set to 4 clusters (males of 21–45 years; females of 81–95 years), 5 clusters (males of 81–90 years; females of 21–45, 46–65, and 66–80 years), and 6 clusters (males of 46–65 and 66–80 years).
The resulting MM clusters showed distinct CC distribution in each age-sex stratum. Detailed information on the prevalence, exclusivity, and O/E ratio of each CC per cluster is provided in S4 Table. S2 and S3 Figs illustrate the most representative CC per cluster. The distribution of patients among the different clusters, estimated based on their membership percentages, is summarized in Fig 1 and Table 2.
The size of the circles represents the population size and the color indicates the percentage of severe COVID-19 cases (upper figure) and mortality (lower figure). COPD = Chronic Obstructive Pulmonary Disease; CV = Cardiovascular.
COPD = Chronic obstructive pulmonary disease; DM = Diabetes mellitus; OR = Odds ratio; CI = Confidence interval; (*) = not included in the bivariate analysis.
Clusters involving DM or CV conditions / risk factors were found in all age-sex groups. Neurodevelopmental disorders clusters were present in both groups of 21–45 year-olds, and Neuropsychiatric patterns were found in the 46–65 year-olds’ groups. Clusters grouping Psychogeriatric diseases were identified in patients of >65 years, along with the Cardiorespiratory and DM—cardiorespiratory patterns. Some clusters were sex-specific, like Cognitive-motor disorders in males (46–65 and 66–80 years), and osteoinflammatory conditions in females, encompassing the clusters Osteoinflammatory—psychosomatic (46–65 years), Osteoinflammatory—psychosomatic—CV risk factors (66–80 years), and Osteoinflammatory—neoplasia (81–95 years).
Relationship between multimorbidity clusters and severe COVID-19 infection or mortality
The distribution of severe cases and mortality among MM clusters is represented in Fig 1 and detailed in Table 2 along with the P-values and odds ratios.
The weighted bivariate analysis showed that clusters of patients in the age ranges of 21–45 and 46–65 displayed P≤0.001 in relation to severe infection, and clusters of males in the 66–80 cohort presented P = 0.016. Among male patients, the clusters with the highest percentage of severe cases were CV risk factors (21–45 years, 7.5%), Cognitive-motor disorders (46–65 years, 32.9%), Unspecific with COPD predominance (66–80 years, 30.4%), and Heart—liver—gastrointestinal diseases (81–90 years, 34.1%). In female patients, the clusters associated with the highest percentages of severity were DM and CV risk factors (21–45 years, 6.2%), DM—CV diseases (46–65 years, 21.2%), Neoplasia—haematologic (66–80 years, 19.7%), and DM—CV diseases (81–95 years, 22.5%).
Regarding mortality, clusters of females aged 21–65 and males aged 46–65 obtained P<0.001, and males aged 66–80 presented P = 0.02. The clusters with the highest percentage of mortality among males were Substance abuse disorders (21–45 years, 1.7%), DM—liver diseases (46–65 years, 7.2%), Cognitive-motor disorders (66–80 years, 20.5%), and Heart—liver—gastrointestinal diseases (81–90 years, 29.5%). Clusters with the highest percentage of mortality in females were DM and CV risk factors (21–45 years, 1.4%), DM—CV diseases (46–65 years, 9.1%), Cerebrovascular diseases (66–80 years, 12.5%), and Psychogeriatric (81–95 years, 16.0%).
Discussion
The present study assessed the MM patterns associated to the population affected by the first wave of COVID-19 in a Spanish area. The results revealed well-defined MM clusters taking into account age and sex, some of which showed a high severity of COVID-19 infection and mortality, with a stronger relationship in younger age groups.
First of all, defining the study cohort was a vital aspect in order to establish reliable MM profiles. Many studies from the first wave of the pandemic focused only on PCR positive individuals [1, 3, 9, 38]; nevertheless, the critical situation at that time resulted in a small fraction of patients undergoing confirmatory tests in the studied geographic area. The test positivity rate in Catalonia during the selected period peaked at 43.5% and did not drop below the World Health Organization recommendation of 5% until mid-May, thereby showing the shortage of testing resources [39, 40]. This situation particularly affected older individuals, who suffered the highest mortality rates in nursing homes without testing nor hospital admission [41]. Therefore, excluding these individuals from the cohort could introduce biases in the definition of multimorbidity patterns.
Taking the whole situation into account, the inclusion of suspected cases in the study cohort allowed for a more accurate approximation of the actual affected population. Although this criterion may have resulted in some possible false positives, it helped avoid potential biases associated with factors like age, access to medical testing or socioeconomic situation.
Using fuzzy c-means cluster analysis allowed for the simultaneous classification of patients into different clusters with a certain probability. This method is robust for pattern recognition in situations where clusters tend to overlap [42], such as in individuals with a high prevalence of co-existing conditions. It should be taken into account that the number of clusters C depends on the researchers’ subjective election, which could affect the reproducibility of the study. Nevertheless, a Delphi-like method was carried out in order to enhance the robustness to the cluster selection, taking advantage of the multidisciplinarity of the research team, which includes members with technical or clinical expertise [43].
In addition, another key methodological aspect of this work was the stratification of the cohort. Given that the accumulation of CC is more frequent at older ages [17, 44], and that males and females have differential tendencies to suffer certain pathologies [45], it was necessary to consider both age and sex perspectives. In fact, previous studies on COVID-19 data from different countries already revealed that men have a higher risk of admission to intensive treatment units (from 14.4% to 44.9%) [1] or adverse outcome (OR: 2.05, 95% CI: 1.39–3.04) [2], which was supported by our bivariate analysis of demographic data. Indeed, several sex-related differences were observed in the MM clusters: patterns related with psychosomatic conditions or osteoinflammatory diseases were more frequent in females, while cognitive-motor, digestive and respiratory disease patterns were more associated to males, consistent with previous findings [23, 46, 47].
Regarding age-related differences, previous works based on the fuzzy c-means algorithm already identified certain MM clusters in cohorts of older patients (≥60 years) that were similar to some patterns observed in the 66–80 and 81–95 age groups in the present study. For example, Vetrano et al identified clusters of cardiometabolic diseases, respiratory diseases, and cognitive / sensory impairment [28], and Akugizibwe et al found clusters of psychiatric conditions and sensory impairments/cancer [27]. Regarding the younger cohorts, a previous stratified study by Prados-Torres et al found some similar clusters by factor analysis, like psychiatric-substance abuse (males of 15–45 years), or cardio-metabolic (all strata) [48].
On the basis of these clusters, the results of the bivariate analyses suggest that certain MM patterns could be associated to higher COVID-19 infection severity or mortality rates in young-middle ages. The fact that this relationship was not strong in older patients may be explained by the inherent risk of old age itself [49, 50], as indicated by the bivariate analysis of demographic parameters. Additionally, other factors could also explain this mild association in older individuals, such as the simple accumulation of morbidities [17, 51, 52], difficulties in accessing healthcare services during the first wave of the pandemic [53], or the adequacy of therapeutic effort policy in age-related cases [54]. Furthermore, geriatric syndromes [55], which are barely coded in medical records, represent an important factor of vulnerability [25]. For example, frailty and disability are commonly detected in older patients and have been described as potential risk factors of critical COVID-19 infection [56, 57]. Nevertheless, in spite of the lack of strong statistical association, several MM clusters of patients in the 66–80 and 81–95 age ranges showed elevated infection severity and mortality rates compared to other clusters.
Therefore, the high percentages of infection severity / mortality displayed by certain MM clusters suggest that these patients may be more vulnerable to poor prognosis caused by COVID-19 infection. Previous pre-pandemic studies already showed some similar MM patterns associated with other adverse outcomes. For example, clusters of psychiatric disorders and CV diseases have been linked to a higher risk of unplanned hospitalization [27], while cognitive / sensory, complex cardiometabolic, respiratory, and age-associated chronic clusters showed increased risk of mortality [28, 58]. Taking into account other COVID-19 studies focusing on individual diseases, several CC that showed high prevalence and O/E ratio in the vulnerable clusters of our study were already identified as risk factors of adverse outcomes, such as COPD [4, 5, 8], liver disease, substance consumption [7], neoplasia [5, 33], and CV pathologies [31, 33]. Nevertheless, the present results indicate that beyond the occurrence of isolated pathologies, the accumulation of CC may represent an additional dimension to consider when facing patient-centered risk evaluations, as suggested by other authors [12].
In any case, the described results should always be interpreted considering the limitations of the study. Although the inclusion criteria took into account the characteristics of the first wave of the pandemic, there are inherent limitations related to this period (concerning data collection, codification and testing). Additionally, it should be considered that Big Data analyses imply some disadvantages, such as difficulties in ensuring data quality and homogeneity [59]. On the other hand, the study data corresponds to a specific region, which ensures a certain uniformity in the characteristics of the population, as well as their clinical management, protocols and codification, mitigating the potential heterogeneity of Big Data. Lastly, it should be noted that the occurrence of COVID-19 severe infection or mortality, or even their relationship with MM clusters, might be influenced by additional factors that were not included in the analyses, such as healthcare-associated features, socioeconomic status, dependency or frailty.
Therefore, these results provide insights into the relationships between MM clusters and severity or mortality, but do not provide a risk prediction. Our methodological approach allowed to explore if a comprehensive set of conditions forming multimorbidity clusters was related to the risk of COVID-19 infection severity or mortality. This approach may offer a complementary picture to the findings of possible further studies with multivariate analyses.
Conclusions
In summary, the present study provided a complete stratified profiling of MM clusters in COVID-19 patients from the first wave of the pandemic, revealing that males, older individuals and patients with certain MM patterns displayed higher probabilities of suffering severe COVID-19 infection or mortality. These findings may contribute to enhance the current knowledge about MM definition as well as to identify the most vulnerable population. This might especially benefit the preventive care of high-risk patients or those with limited access to medical treatment, and might help optimize healthcare policies and future strategies.
Supporting information
S1 Table. List of codes for COVID-19 registration, primary care mortality and severe infection.
ICD-10-CM = International Classification of Diseases, 10th edition, Clinical Modification. ICD-10-PCS = International Classification of Diseases, 10th edition, Procedure Coding System.
https://doi.org/10.1371/journal.pone.0290969.s001
(PDF)
S2 Table. Number of patients and prevalence of severe COVID-19 infection according to the type of case (confirmed or suspected).
https://doi.org/10.1371/journal.pone.0290969.s002
(PDF)
S3 Table. Chronic conditions with >2% prevalence per age and sex stratum.
https://doi.org/10.1371/journal.pone.0290969.s003
(XLSX)
S4 Table. Prevalence (%), Observed/Expected (O/E) ratio, and exclusivity (%) of the chronic conditions per multimorbidity clusters.
https://doi.org/10.1371/journal.pone.0290969.s004
(XLSX)
S1 Fig. Source and distribution of the analysed data.
The included patients were submitted to an anonymization process. A patient Id code was assigned to each individual, which was maintained through all the different datasets in order to enable data compilation.
https://doi.org/10.1371/journal.pone.0290969.s005
(PDF)
S2 Fig. Representation of the Observed/Expected ratio (O/E) and exclusivity (%) of the chronic conditions composing the multimorbidity clusters of male patients.
O/E >1 and exclusivity > 1/number of clusters are displayed.
https://doi.org/10.1371/journal.pone.0290969.s006
(PDF)
S3 Fig. Representation of the Observed/Expected ratio (O/E) and exclusivity (%) of the chronic conditions composing the multimorbidity clusters of female patients.
O/E >1 and exclusivity > 1/number of clusters are displayed.
https://doi.org/10.1371/journal.pone.0290969.s007
(PDF)
Acknowledgments
We would like to thank people working at PADRIS, from AQuAS, at the Department of Health of Catalonia, and specially Borja Velasco, for coordinating and performing the extraction of the data files that were used in this project. This work was carried out as part of the PhD programme in Biomedical Research Methodology and Public Health at the Autonomous University of Barcelona.
References
- 1. Peckham H, de Gruijter NM, Raine C, Radziszewska A, Ciurtin C, Wedderburn LR, et al. Male sex identified by global COVID-19 meta-analysis as a risk factor for death and ITU admission. Nat Commun 2020;11:6317. pmid:33298944
- 2. Booth A, Reed AB, Ponzo S, Yassaee A, Aral M, Plans D, et al. Population risk factors for severe disease and mortality in COVID-19: A global systematic review and meta-analysis. PLoS One 2021;16:e0247461. pmid:33661992
- 3. Grasselli G, Greco M, Zanella A, Albano G, Antonelli M, Bellani G, et al. Risk Factors Associated with Mortality among Patients with COVID-19 in Intensive Care Units in Lombardy, Italy. JAMA Intern Med 2020;180:1345–55. pmid:32667669
- 4. Atkins JL, Masoli JAH, Delgado J, Pilling LC, Kuo C-L, Kuchel GA, et al. Preexisting Comorbidities Predicting COVID-19 and Mortality in the UK Biobank Community Cohort. J Gerontol A Biol Sci Med Sci 2020;75:2224–30. pmid:32687551
- 5. Guan W, Liang W, Zhao Y, Liang H, Chen Z, Li Y, et al. Comorbidity and its impact on 1590 patients with COVID-19 in China: a nationwide analysis. Eur Respir J 2020;55:2000547. pmid:32217650
- 6. Richardson S, Hirsch JS, Narasimhan M, Crawford JM, McGinn T, Davidson KW, et al. Presenting Characteristics, Comorbidities, and Outcomes Among 5700 Patients Hospitalized With COVID-19 in the New York City Area. JAMA 2020;323:2052–9. pmid:32320003
- 7. Reilev M, Kristensen KB, Pottegård A, Lund LC, Hallas J, Ernst MT, et al. Characteristics and predictors of hospitalization and death in the first 11 122 cases with a positive RT-PCR test for SARS-CoV-2 in Denmark: a nationwide cohort. Int J Epidemiol 2020;49:1468–81.
- 8. Iaccarino G, Grassi G, Borghi C, Ferri C, Salvetti M, Volpe M. Age and Multimorbidity Predict Death Among COVID-19 Patients: Results of the SARS-RAS Study of the Italian Society of Hypertension. Hypertension 2020;76:366–72. pmid:32564693
- 9. Poblador-Plou B, Carmona-Pírez J, Ioakeim-Skoufa I, Poncel-Falcó A, Bliek-Bueno K, Cano-Del Pozo M, et al. Baseline chronic comorbidity and mortality in laboratory-confirmed COVID-19 cases: Results from the PRECOVID study in Spain. Int J Environ Res Public Health 2020;17:5171. pmid:32709002
- 10. Tartof SY, Qian L, Hong V, Wei R, Nadjafi RF, Fischer H, et al. Obesity and Mortality Among Patients Diagnosed With COVID-19: Results From an Integrated Health Care Organization. Ann Intern Med 2020;173:773–81. pmid:32783686
- 11. Tisminetzky M, Delude C, Hebert T, Carr C, Goldberg RJ, Gurwitz JH. Age, Multiple Chronic Conditions, and COVID-19: A Literature Review. J Gerontol A Biol Sci Med Sci 2021:glaa320.
- 12. Maddaloni E, D’Onofrio L, Alessandri F, Mignogna C, Leto G, Pascarella G, et al. Cardiometabolic multimorbidity is associated with a worse Covid-19 prognosis than individual cardiometabolic risk factors: A multicentre retrospective study (CoViDiab II). Cardiovasc Diabetol 2020;19:164. pmid:33004045
- 13. Clark A, Jit M, Warren-Gash C, Guthrie B, Wang HHX, Mercer SW, et al. Global, regional, and national estimates of the population at increased risk of severe COVID-19 due to underlying health conditions in 2020: a modelling study. Lancet Glob Heal 2020;8:e1003–17. pmid:32553130
- 14. Barnett K, Mercer SW, Norbury M, Watt G, Wyke S, Guthrie B. Epidemiology of multimorbidity and implications for health care, research, and medical education: A cross-sectional study. Lancet 2012;380:37–43. pmid:22579043
- 15. Yarnall AJ, Sayer AA, Clegg A, Rockwood K, Parker S, Hindle J V. New horizons in multimorbidity in older adults. Age Ageing 2017;46:882–8. pmid:28985248
- 16.
The Academy of Medical Sciences. Multimorbidity: a priority for global health research. 2018.
- 17. Mounce LTA, Campbell JL, Henley WE, Tejerina Arreal MC, Porter I, Valderas JM. Predicting incident multimorbidity. Ann Fam Med 2018;16:322–9. pmid:29987080
- 18.
National Institute for Health and Care Excellence. Multimorbidity: clinical assessment and management. NICE guideline (NG56). 2016.
- 19. Valderas JM, Starfi B, Sibbald B. Defining Comorbidity: Implications for Understanding Health and Health Services. Ann Fam Med 2009;7:357–63. pmid:19597174
- 20. Makovski TT, Schmitz S, Zeegers MP, Stranges S, van den Akker M. Multimorbidity and quality of life: Systematic literature review and meta-analysis. Ageing Res Rev 2019;53:100903. pmid:31048032
- 21. Chudasama Y V., Khunti KK, Zaccardi F, Rowlands A V., Yates T, Gillies CL, et al. Physical activity, multimorbidity, and life expectancy: A UK Biobank longitudinal study. BMC Med 2019;17:108. pmid:31186007
- 22. Ng SK, Tawiah R, Sawyer M, Scuffham P. Patterns of multimorbid health conditions: A systematic review of analytical methods and comparison analysis. Int J Epidemiol 2018;47:1687–704. pmid:30016472
- 23. Guisado-Clavero M, Roso-Llorach A, López-Jimenez T, Pons-Vigués M, Foguet-Boreu Q, Muñoz MA, et al. Multimorbidity patterns in the elderly: A prospective cohort study with cluster analysis. BMC Geriatr 2018;18:16. pmid:29338690
- 24. Tazzeo C, Rizzuto D, Calderón-Larrañaga A, Roso-Llorach A, Marengoni A, Welmer A-K, et al. Multimorbidity patterns and risk of frailty in older community-dwelling adults: a population-based cohort study. Age Ageing 2021:afab138. pmid:34228784
- 25. Baré M, Herranz S, Roso-Llorach A, Jordana R, Violán C, Lleal M, et al. Multimorbidity patterns of chronic conditions and geriatric syndromes in older patients from the MoPIM multicentre cohort study. BMJ Open 2021;11:49334. pmid:34782339
- 26. Marengoni A, Akugizibwe R, Vetrano DL, Roso-Llorach A, Onder G, Welmer A-K, et al. Patterns of multimorbidity and risk of disability in community-dwelling older persons. Aging Clin Exp Res 2021;33:457–462. pmid:33580869
- 27. Akugizibwe R, Calderón-Larrañaga A, Roso-Llorach A, Onder G, Marengoni A, Zucchelli A, et al. Multimorbidity Patterns and Unplanned Hospitalisation in a Cohort of Older Adults. J Clin Med 2020;9:4001. pmid:33321977
- 28. Vetrano DL, Roso-Llorach A, Fernández S, Guisado-Clavero M, Violán C, Onder G, et al. Twelve-year clinical trajectories of multimorbidity in a population of older adults. Nat Commun 2020;11:3223. pmid:32591506
- 29. Violán C, Foguet-Boreu Q, Fernández-Bertolín S, Guisado-Clavero M, Cabrera-Bean M, Formiga F, et al. Soft clustering using real-world data for the identification of multimorbidity patterns in an elderly population: Cross-sectional study in a Mediterranean population. BMJ Open 2019;9:e029594. pmid:31471439
- 30. Lleal M, Baré M, Ortonobes S, Sevilla-Sánchez D, Jordana R, Herranz S, et al. Comprehensive Multimorbidity Patterns in Older Patients Are Associated with Quality Indicators of Medication-MoPIM Cohort Study. Int J Environ Res Public Health 2022;19. pmid:36497976
- 31. Banerjee A, Pasea L, Harris S, Gonzalez-Izquierdo A, Torralbo A, Shallcross L, et al. Estimating excess 1-year mortality associated with the COVID-19 pandemic according to underlying conditions and age: a population-based cohort study. Lancet 2020;395:1715–25. pmid:32405103
- 32. Jin J, Agarwala N, Kundu P, Harvey B, Zhang Y, Wallace E, et al. Individual and community-level risk for COVID-19 mortality in the United States. Nat Med 2021;27:264–9. pmid:33311702
- 33. Semenzato L, Dray-spira R, Weill A, Zureik M. Chronic diseases, health conditions and risk of COVID-19-related hospitalization and in-hospital mortality during the first wave of the epidemic in France: a cohort study of 66 million people. Lancet Reg Heal—Eur 2021;8:100158.
- 34.
COVID-19 Treatment Guidelines Panel. Coronavirus Disease 2019 (COVID-19) Treatment Guidelines. Natl Institutes Heal 2021.
- 35.
R Core Team. R: A language and environment for statistical computing. R Found Stat Comput Vienna, Austria 2014.
- 36.
RStudio Team. RStudio: Integrated Development for R 2015.
- 37.
Agency for Healthcare Research and Quality. Rockville. MD. Healthcare Cost and Utilization Project (HCUP) 2021. www.hcup-us.ahrq.gov/home.jsp.
- 38. Castro VM, McCoy TH, Perlis RH. Laboratory Findings Associated With Severe Illness and Mortality Among Hospitalized Individuals With Coronavirus Disease 2019 in Eastern Massachusetts. JAMA Netw Open 2020;3:e2023934. pmid:33125498
- 39.
Roser M, Ritchie H, Ortiz-Ospina E, Hasell J. Coronavirus Pandemic (COVID-19) 2020. https://ourworldindata.org/coronavirus/country/spain.
- 40.
IDESCAT (Statistical Institute of Catalonia). Statistical information to track the impact of Covid-19. https://www.idescat.cat/dades/covid19/?lang=en (accessed June 14, 2023).
- 41.
Comas-Herrera A, Zalakaín J, Litwin C, Hsu AT, Lemmon E, Henderson D, et al. Mortality associated with COVID-19 outbreaks in care homes: early international evidence. 2020.
- 42. Bezdek JC, Ehrlich R, Full W. FCM: The fuzzy c-means clustering algorithm. Comput Geosci 1984;10:191–203.
- 43. Humphrey-Murto SM, Wood TJ, Gonsalves C, Mascioli K, Varpio L. The Delphi Method. Acad Med 2020;95:168. pmid:31335812
- 44. Salive ME. Multimorbidity in older adults. Epidemiol Rev 2013;35:75–83. pmid:23372025
- 45. Mauvais-Jarvis F, Merz NB, Barnes PJ, Brinton RD, Carrero J-J, DeMeo DL, et al. Sex and gender: modifiers of health, disease, and medicine. Lancet 2020;396:565–82. pmid:32828189
- 46. Violán C, Roso-Llorach A, Foguet-Boreu Q, Guisado-Clavero M, Pons-Vigués M, Pujol-Ribera E, et al. Multimorbidity patterns with K-means nonhierarchical cluster analysis. BMC Fam Pract 2018;19:108. pmid:29969997
- 47. Foguet-Boreu Q, Violán C, Rodriguez-Blanco T, Roso-Llorach A, Pons-Vigués M, Pujol-Ribera E, et al. Multimorbidity patterns in elderly primary health care patients in a South Mediterranean European region: A cluster analysis. PLoS One 2015;10:e0141155. pmid:26524599
- 48. Prados-Torres A, Poblador-Plou B, Gimeno-Miguel A, Calderón-Larrañaga A, Poncel-Falcó A, Gimeno-Feliú LA, et al. Cohort Profile: The epidemiology of chronic diseases and multimorbidity. The EpiChron cohort study. Int J Epidemiol 2018;47:382e–384e. pmid:29346556
- 49. Girard N, Odille G, Sanchez S, Lelarge S, Mignot A, Putot S, et al. Comment on: “Beyond Chronological Age: Frailty and Multimorbidity Predict In-hospital Mortality in Patients With Coronavirus Disease 2019.” J Gerontol A Biol Sci Med Sci 2021;76:e155–6.
- 50. Ho FK, Petermann-Rocha F, Gray SR, Jani BD, Vittal Katikireddi S, Niedzwiedz CL, et al. Is older age associated with COVID-19 mortality in the absence of other risk factors? General population cohort study of 470,034 participants. PLoS One 2020;15:e0241824. pmid:33152008
- 51. Buja A, Claus M, Perin L, Rivera M, Corti MC, Avossa F, et al. Multimorbidity patterns in high-need, high-cost elderly patients. PLoS One 2018;13:e0208875. pmid:30557384
- 52. Vinjerui KH, Bjerkeset O, Bjorngaard JH, Krokstad S, Douglas KA, Sund ER, et al. Socioeconomic inequalities in the prevalence of complex multimorbidity in a Norwegian population: findings sectional HUNT Study from the cross-sectional HUNT Study. BMJ Open 2020;10:e036851.
- 53. Martins Van Jaarsveld G. The Effects of COVID-19 Among the Elderly Population: A Case for Closing the Digital Divide. Front Psychiatry 2020;11:577427. pmid:33304283
- 54. Borsellino P. Limitation of the therapeutic effort: Ethical and legal justification for withholding and/or withdrawing life sustaining treatments. Multidiscip Respir Med 2015;10:5. pmid:25705381
- 55. Inouye SK, Studenski S, Tinetti M, Kuchel G. Geriatric Syndromes: Clinical, Research and Policy Implications of a Core Geriatric Concept in older adults related to primary and secondary prevention. J Am Geriatr Soc 2007;55:780–91.
- 56. Bellelli G, Rebora P, Valsecchi MG, Bonfanti P, Citerio G, Galimberti S, et al. Frailty index predicts poor outcome in COVID-19 patients. Intensive Care Med 2020;46:1634–6. pmid:32451583
- 57. Hussien H, Nastasa A, Apetrii M, Nistor I, Petrovic M, Covic A. Different aspects of frailty and COVID-19: points to consider in the current pandemic and future ones. BMC Geriatr 2021;21:389. pmid:34176479
- 58. Zheng DD, Loewenstein DA, Christ SL, Feaster DJ, Lam BL, McCollister KE, et al. Multimorbidity patterns and their relationship to mortality in the US older adult population. PLoS One 2021;16:e0245053. pmid:33471812
- 59. Andreu-Perez J, Poon CCY, Merrifield RD, Wong STC, Yang G-Z. Big Data for Health. IEEE J Biomed Heal Informatics 2015;19:1193–208. pmid:26173222