Hostname: page-component-cd9895bd7-dzt6s Total loading time: 0 Render date: 2024-12-13T06:10:08.919Z Has data issue: false hasContentIssue false

Coronavirus disease 2019 is associated with long-term depressive symptoms in Spanish older adults with overweight/obesity and metabolic syndrome

Published online by Cambridge University Press:  05 September 2023

Sangeetha Shyam*
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Departament de Bioquímica i Biotecnologia, Grup Alimentació, Nutrició, Desenvolupament i Salut Mental, Unitat de Nutrició Humana, Universitat Rovira i Virgili, Reus, Spain Institut d'Investigació Sanitària Pere Virgili (IISPV), Reus, Spain
Carlos Gómez-Martínez
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Departament de Bioquímica i Biotecnologia, Grup Alimentació, Nutrició, Desenvolupament i Salut Mental, Unitat de Nutrició Humana, Universitat Rovira i Virgili, Reus, Spain Institut d'Investigació Sanitària Pere Virgili (IISPV), Reus, Spain
Indira Paz-Graniel
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Departament de Bioquímica i Biotecnologia, Grup Alimentació, Nutrició, Desenvolupament i Salut Mental, Unitat de Nutrició Humana, Universitat Rovira i Virgili, Reus, Spain Institut d'Investigació Sanitària Pere Virgili (IISPV), Reus, Spain
José J. Gaforio
Affiliation:
CIBER de Epidemiología y Salud Pública (CIBERESP), Instituto de Salud Carlos III (ISCIII), Madrid, Spain Departamento de Ciencias de la Salud, Instituto Universitario de Investigación en Olivar y Aceites de Oliva, Universidad de Jaén, Jaén, Spain
Miguel Ángel Martínez-González
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Preventive Medicine and Public Health, Instituto de Investigación Sanitaria de Navarra (IdiSNA), University of Navarra, Pamplona, Spain Department of Nutrition, Harvard T.H. Chan School of Public Health, Boston, MA, USA
Dolores Corella
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Preventive Medicine, University of Valencia, Valencia, Spain
Montserrat Fitó
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Unit of Cardiovascular Risk and Nutrition, Institut Hospital del Mar de Investigaciones Médicas Municipal d'Investigació Médica (IMIM), Barcelona, Spain
J. Alfredo Martínez
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Nutrition, Food Sciences, and Physiology, Center for Nutrition Research, University of Navarra, Pamplona, Spain Precision Nutrition and Cardiometabolic Health Program, IEA Food, CEI UAM + CSIC, Madrid, Spain
Ángel M. Alonso-Gómez
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Bioaraba Health Research Institute, Cardiovascular, Respiratory and Metabolic Area; Osakidetza Basque Health Service, Araba University Hospital; University of the Basque Country UPV/EHU, Vitoria-Gasteiz, Spain
Julia Wärnberg
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain EpiPHAAN Research Group, School of Health Sciences, University of Málaga – Instituto de Investigación Biomédica en Málaga (IBIMA), Málaga, Spain
Jesús Vioque
Affiliation:
CIBER de Epidemiología y Salud Pública (CIBERESP), Instituto de Salud Carlos III (ISCIII), Madrid, Spain Instituto de Investigación Sanitaria y Biomédica de Alicante, Universidad Miguel Hernández (ISABIAL-UMH), Alicante, Spain
Dora Romaguera
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Health Research Institute of the Balearic Islands (IdISBa), Palma de Mallorca, Spain
José López-Miranda
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Internal Medicine, Maimonides Biomedical Research Institute of Cordoba (IMIBIC), Reina Sofia University Hospital, University of Cordoba, Cordoba, Spain
Ramon Estruch
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Internal Medicine, Institut d'Investigacions Biomèdiques August Pi Sunyer (IDIBAPS), Hospital Clinic, University of Barcelona, Barcelona, Spain; Institut de Recerca en Nutrició i Seguretat Alimentaria (INSA-UB), University of Barcelona, Barcelona, Spain
Francisco J. Tinahones
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Endocrinology, Virgen de la Victoria Hospital, Instituto de Investigación Biomédica de Málaga (IBIMA), University of Málaga, Málaga, Spain
José Manuel Santos-Lozano
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Family Medicine, Research Unit, Distrito Sanitario Atención Primaria Sevilla, Sevilla, Spain
J. Luís Serra-Majem
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Research Institute of Biomedical and Health Sciences (IUIBS), University of Las Palmas de Gran Canaria & Centro Hospitalario Universitario Insular Materno Infantil (CHUIMI), Canarian Health Service, Las Palmas de Gran Canaria, Spain
Aurora Bueno-Cavanillas
Affiliation:
CIBER de Epidemiología y Salud Pública (CIBERESP), Instituto de Salud Carlos III (ISCIII), Madrid, Spain Department of Preventive Medicine and Public Health, University of Granada, Granada, Spain
Josep A. Tur
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Research Group on Community Nutrition & Oxidative Stress, University of Balearic Islands, Palma de Mallorca, Spain
Vicente Martín Sánchez
Affiliation:
CIBER de Epidemiología y Salud Pública (CIBERESP), Instituto de Salud Carlos III (ISCIII), Madrid, Spain Precision Nutrition and Cardiometabolic Health Program, IEA Food, CEI UAM + CSIC, Madrid, Spain Institute of Biomedicine (IBIOMED), University of León, León, Spain
Xavier Pintó
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Lipids and Vascular Risk Unit, Internal Medicine, Hospital Universitario de Bellvitge-IDIBELL, Hospitalet de Llobregat – Barcelona, Barcelona, Spain
María Ortiz Ramos
Affiliation:
Department of Endocrinology and Nutrition, Instituto de Investigación Sanitaria Hospital Clínico San Carlos (IdISSC), Madrid, Spain
Josep Vidal
Affiliation:
CIBER Diabetes y Enfermedades Metabólicas (CIBERDEM), Instituto de Salud Carlos III (ISCIII), Madrid, Spain Department of Endocrinology, Institut d`Investigacions Biomédiques August Pi Sunyer (IDIBAPS), Hospital Clinic, University of Barcelona, Barcelona, Spain
Maria Mar Alcarria
Affiliation:
Department of Endocrinology and Nutrition, Hospital Fundación Jimenez Díaz, Instituto de Investigaciones Biomédicas IISFJD, University Autonoma, Madrid, Spain
Lidia Daimiel
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Nutritional Control of the Epigenome Group, Precision Nutrition and Obesity Program, IMDEA Food, CEI UAM + CSIC, Madrid, Spain Departamento de Ciencias Farmacéuticas y de la Salud, Faculty de Farmacia, Universidad San Pablo-CEU, CEU Universities, Boadilla del Monte, Spain
Emilio Ros
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Lipid Clinic, Department of Endocrinology and Nutrition, Institut d'Investigacions Biomèdiques August Pi Sunyer (IDIBAPS), Hospital Clínic, Barcelona, Spain
Fernando Fernandez-Aranda
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Psychoneurobiology of Eating and Addictive Behaviors Group, Institut d'Investigació Biomèdica de Bellvitge (IDIBELL), Barcelona, Spain Department of Psychiatry, University Hospital of Bellvitge and University of Barcelona, Barcelona, Spain
Stephanie K. Nishi
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Departament de Bioquímica i Biotecnologia, Grup Alimentació, Nutrició, Desenvolupament i Salut Mental, Unitat de Nutrició Humana, Universitat Rovira i Virgili, Reus, Spain Institut d'Investigació Sanitària Pere Virgili (IISPV), Reus, Spain Toronto 3D (Diet, Digestive Tract and Disease) Knowledge Synthesis and Clinical Trials Unit, Toronto, ON, Canada
Oscar García Regata
Affiliation:
Department of Internal Medicine, OSI ARABA, University Hospital Araba, Vitoria-Gasteiz, Spain
Estefania Toledo
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Preventive Medicine and Public Health, Instituto de Investigación Sanitaria de Navarra (IdiSNA), University of Navarra, Pamplona, Spain
Jose V. Sorli
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Preventive Medicine, University of Valencia, Valencia, Spain
Olga Castañer
Affiliation:
Unit of Cardiovascular Risk and Nutrition, Institut Hospital del Mar de Investigaciones Médicas Municipal d'Investigació Médica (IMIM), Barcelona, Spain
Antonio Garcia-Rios
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Internal Medicine, Maimonides Biomedical Research Institute of Cordoba (IMIBIC), Reina Sofia University Hospital, University of Cordoba, Cordoba, Spain
Rafael Valls-Enguix
Affiliation:
Health Care Centre Raval-Elche, Elche, Spain
Napoleon Perez-Farinos
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain EpiPHAAN Research Group, School of Health Sciences, University of Málaga – Instituto de Investigación Biomédica en Málaga (IBIMA), Málaga, Spain
M. Angeles Zulet
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Nutrition, Food Sciences, and Physiology, Center for Nutrition Research, University of Navarra, Pamplona, Spain
Elena Rayó-Gago
Affiliation:
Health Research Institute of the Balearic Islands (IdISBa), Palma de Mallorca, Spain
Rosa Casas
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Internal Medicine, Institut d'Investigacions Biomèdiques August Pi Sunyer (IDIBAPS), Hospital Clinic, University of Barcelona, Barcelona, Spain; Institut de Recerca en Nutrició i Seguretat Alimentaria (INSA-UB), University of Barcelona, Barcelona, Spain
Mario Rivera-Izquierdo
Affiliation:
Department of Preventive Medicine and Public Health, University of Granada, Granada, Spain
Lucas Tojal-Sierra
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Bioaraba Health Research Institute, Cardiovascular, Respiratory and Metabolic Area; Osakidetza Basque Health Service, Araba University Hospital; University of the Basque Country UPV/EHU, Vitoria-Gasteiz, Spain
Miguel Damas-Fuentes
Affiliation:
Department of Endocrinology, Virgen de la Victoria Hospital, Instituto de Investigación Biomédica de Málaga (IBIMA), University of Málaga, Málaga, Spain
Pilar Buil-Cosiales
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Preventive Medicine and Public Health, Instituto de Investigación Sanitaria de Navarra (IdiSNA), University of Navarra, Pamplona, Spain Atención Primaria, Servicio Navarro de Salud, Pamplona, Spain
Rebeca Fernández-Carrion
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Preventive Medicine, University of Valencia, Valencia, Spain
Albert Goday
Affiliation:
Unit of Cardiovascular Risk and Nutrition, Institut Hospital del Mar de Investigaciones Médicas Municipal d'Investigació Médica (IMIM), Barcelona, Spain
Patricia J. Peña-Orihuela
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Internal Medicine, Maimonides Biomedical Research Institute of Cordoba (IMIBIC), Reina Sofia University Hospital, University of Cordoba, Cordoba, Spain
Laura Compañ-Gabucio
Affiliation:
CIBER de Epidemiología y Salud Pública (CIBERESP), Instituto de Salud Carlos III (ISCIII), Madrid, Spain Instituto de Investigación Sanitaria y Biomédica de Alicante, Universidad Miguel Hernández (ISABIAL-UMH), Alicante, Spain
Javier Diez-Espino
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Preventive Medicine and Public Health, Instituto de Investigación Sanitaria de Navarra (IdiSNA), University of Navarra, Pamplona, Spain Atención Primaria, Servicio Navarro de Salud, Pamplona, Spain
Susanna Tello
Affiliation:
Unit of Cardiovascular Risk and Nutrition, Institut Hospital del Mar de Investigaciones Médicas Municipal d'Investigació Médica (IMIM), Barcelona, Spain
Ana González-Pinto
Affiliation:
Department of Psychiatry, Bioaraba Health Research Institute, Osakidetza Basque Health Service, Araba University Hospital; University of the Basque Country UPV/EHU, CIBERSAM, Vitoria-Gasteiz, Spain
Víctor de la O
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Department of Preventive Medicine and Public Health, Instituto de Investigación Sanitaria de Navarra (IdiSNA), University of Navarra, Pamplona, Spain
Miguel Delgado-Rodríguez
Affiliation:
Precision Nutrition and Cardiometabolic Health Program, IEA Food, CEI UAM + CSIC, Madrid, Spain Division of Preventive Medicine, Faculty of Medicine, University of Jaén, Jaén, Spain
Nancy Babio*
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Departament de Bioquímica i Biotecnologia, Grup Alimentació, Nutrició, Desenvolupament i Salut Mental, Unitat de Nutrició Humana, Universitat Rovira i Virgili, Reus, Spain Institut d'Investigació Sanitària Pere Virgili (IISPV), Reus, Spain
Jordi Salas-Salvadó
Affiliation:
Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health Carlos III, Madrid, Spain Departament de Bioquímica i Biotecnologia, Grup Alimentació, Nutrició, Desenvolupament i Salut Mental, Unitat de Nutrició Humana, Universitat Rovira i Virgili, Reus, Spain Institut d'Investigació Sanitària Pere Virgili (IISPV), Reus, Spain
*
Corresponding author: Sangeetha Shyam; Email: Sangeetha.shyam@urv.cat; Nancy Babio; Email: nancy.babio@urv.cat
Corresponding author: Sangeetha Shyam; Email: Sangeetha.shyam@urv.cat; Nancy Babio; Email: nancy.babio@urv.cat
Rights & Permissions [Opens in a new window]

Abstract

Background

The coronavirus disease 2019 (COVID-19) has serious physiological and psychological consequences. The long-term (>12 weeks post-infection) impact of COVID-19 on mental health, specifically in older adults, is unclear. We longitudinally assessed the association of COVID-19 with depression symptomatology in community-dwelling older adults with metabolic syndrome within the framework of the PREDIMED-Plus cohort.

Methods

Participants (n = 5486) aged 55–75 years were included in this longitudinal cohort. COVID-19 status (positive/negative) determined by tests (e.g. polymerase chain reaction severe acute respiratory syndrome coronavirus 2, IgG) was confirmed via event adjudication (410 cases). Pre- and post-COVID-19 depressive symptomatology was ascertained from annual assessments conducted using a validated 21-item Spanish Beck Depression Inventory-II (BDI-II). Multivariable linear and logistic regression models assessed the association between COVID-19 and depression symptomatology.

Results

COVID-19 in older adults was associated with higher post-COVID-19 BDI-II scores measured at a median (interquartile range) of 29 (15–40) weeks post-infection [fully adjusted β = 0.65 points, 95% confidence interval (CI) 0.15–1.15; p = 0.011]. This association was particularly prominent in women (β = 1.38 points, 95% CI 0.44–2.33, p = 0.004). COVID-19 was associated with 62% increased odds of elevated depression risk (BDI-II ≥ 14) post-COVID-19 when adjusted for confounders (odds ratio; 95% CI 1.13–2.30, p = 0.008).

Conclusions

COVID-19 was associated with long-term depression risk in older adults with overweight/obesity and metabolic syndrome, particularly in women. Thus, long-term evaluations of the impact of COVID-19 on mental health and preventive public health initiatives are warranted in older adults.

Type
Original Article
Copyright
Copyright © The Author(s), 2023. Published by Cambridge University Press

Background

Coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) evolved into a global pandemic since its emergence in 2019 (Guo et al., Reference Guo, Benito Ballesteros, Yeung, Liu, Saha, Curtis and Cheke2022). Despite largely affecting the respiratory system, COVID-19's impact on the cardiovascular, gastrointestinal, and neurologic systems has been recognized (Sparks et al., Reference Sparks, South, Badley, Baker-Smith, Batlle, Bozkurt and Garovic2020). Long-term physical and mental health consequences of COVID-19 are increasingly understood as more data become available (Lopez-Leon et al., Reference Lopez-Leon, Wegman-Ostrosky, Perelman, Sepulveda, Rebolledo, Cuapio and Villapol2021).

Depression is a potentially serious mental health consequence of COVID-19, with its prevalence post-COVID-19 depending on the individual's age and the timing of depression assessment in relation to the infection (Pano et al., Reference Pano, Martínez-Lapiscina, Sayón-Orea, Martinez-Gonzalez, Martinez and Sanchez-Villegas2021). While depression is common in the acute 4-weeks post-infection, there are scarce data on its long-term prevalence (>12 weeks after the diagnosis of COVID-19) (Mazza et al., Reference Mazza, De Lorenzo, Conte, Poletti, Vai, Bollettini and Benedetti2020; Renaud-Charest et al., Reference Renaud-Charest, Lui, Eskander, Ceban, Ho, Di Vincenzo and McIntyre2021). Recently, a meta-analysis determined the frequency of depressive symptoms ≥ 12 weeks post-infection in COVID-19-positive men and women aged over 19 years, with varying degrees of COVID-19 severity, including both hospitalized and non-hospitalized populations and those with and without comorbidities. In this heterogeneous group, the frequency of depressive symptoms ranged between 11% and 28%, while clinically significant depression and/or severe depressive symptoms affected 3–12% of the participants, at >12 weeks after COVID-19 (Renaud-Charest et al., Reference Renaud-Charest, Lui, Eskander, Ceban, Ho, Di Vincenzo and McIntyre2021). While the long-term impact of COVID-19 on depression in the general population may be small, the effects appear to be severe in older age groups (Klaser et al., Reference Klaser, Thompson, Nguyen, Sudre, Antonelli, Murray and Steves2021), specifically in older women and those with prior depressive tendencies (Mazza et al., Reference Mazza, De Lorenzo, Conte, Poletti, Vai, Bollettini and Benedetti2020; Meng et al., Reference Meng, Xu, Dai, Zhang, Liu and Yang2020; Renaud-Charest et al., Reference Renaud-Charest, Lui, Eskander, Ceban, Ho, Di Vincenzo and McIntyre2021).

Older age and cardiometabolic risks increase susceptibility to severe COVID-19 (Channappanavar & Perlman, Reference Channappanavar and Perlman2020; Mueller, McNamara, & Sinclair, Reference Mueller, McNamara and Sinclair2020; Wee, Reference Wee2021), and depression is more prevalent in those with metabolic syndrome and diabetes (Dunbar et al., Reference Dunbar, Reddy, Davis-Lameloise, Philpot, Laatikainen, Kilkkinen and Janus2008; Khaledi, Haghighatdoost, Feizi, & Aminorroaya, Reference Khaledi, Haghighatdoost, Feizi and Aminorroaya2019). The mental health consequence of COVID-19 in older adults with metabolic disturbances is of specific concern because depression is associated with lower adherence to treatment (Castro et al., Reference Castro, Roca, Ricci-Cabello, García-Toro, Riera-Serra, Coronado-Simsic and Gili2021), poorer prognosis (Dunbar et al., Reference Dunbar, Reddy, Davis-Lameloise, Philpot, Laatikainen, Kilkkinen and Janus2008; Khaledi et al., Reference Khaledi, Haghighatdoost, Feizi and Aminorroaya2019), and higher risk of mortality in community-dwelling older adults (Wei et al., Reference Wei, Hou, Zhang, Xu, Xie, Chandrasekar and Goodman2019). While affecting mortality and quality of life in older adults with high cardiometabolic risk, the long-term impact of COVID-19 on depression symptoms will also represent a burden for healthcare systems. Therefore, understanding the impact of COVID-19 on older adults with comorbidities will facilitate early identification and management of the mental health sequelae of COVID-19 in this population. Existing evidence on the topic is highly heterogeneous in terms of the location, age, and sex of individuals and time of depressive symptoms assessment post-COVID-19. Few studies include an unexposed control group to quantify the impact of COVID-19 on depression (Renaud-Charest et al., Reference Renaud-Charest, Lui, Eskander, Ceban, Ho, Di Vincenzo and McIntyre2021). These limitations make it challenging to inform practice with existing information.

Thus, we examined the association of COVID-19 with depressive symptomatology in older adults with overweight/obesity and metabolic syndrome enrolled in the PREDIMED-Plus trial in Spain. We hypothesized that a positive diagnosis of COVID-19 would be associated with higher post-COVID-19 scores for depressive symptoms in comparison to a COVID-19 negative status. We also investigated if sex, time of post-COVID-19 depression assessment, and pre-existing depressive symptomatology affected this association.

Methods

Study design and participants

This analysis involved older women and men enrolled in the PREDIMED-Plus trial, a multicentre, randomized controlled clinical trial in Spain (Martínez-González et al., Reference Martínez-González, Buil-Cosiales, Corella, Bulló, Fitó, Vioque and Salas-Salvadó2019). PREDIMED-Plus is an existing longitudinal cohort of 6874 community-dwelling older adults with overweight/obesity and metabolic syndrome. At enrolment, participants were free from cardiovascular disease, cancers, major depressive disorder, and other major chronic conditions. The trial aims to assess in this cohort the effectiveness of an energy-reduced Mediterranean diet, physical activity, and behavioural support intervention on the primary prevention of cardiovascular disease in comparison to an ad libitum Mediterranean diet without advice to increase physical activity or reduce energy intake. The ongoing trial began its recruitment in 2013 and is scheduled to be completed in 2024. A detailed study protocol has been published earlier (Martínez-González et al., Reference Martínez-González, Buil-Cosiales, Corella, Bulló, Fitó, Vioque and Salas-Salvadó2019; Salas-Salvadó et al., Reference Salas-Salvadó, Díaz-López, Ruiz-Canela, Basora, Fitó and Corella2018) (see supplementary methods).

The PREDIMED-Plus protocol has been approved by the institutional review boards of all participating centres in accordance with the Declaration of Helsinki. All enrolled participants provided written informed consent. This study is registered at the International Standard Randomized Controlled Trial (ISRCT; http://www.isrctn.com/ISRCTN89898870).

The cohort has validated assessments of depression that were obtained before the onset of the COVID-19 pandemic as well as ongoing follow-up measurements. The PREDIMED-Plus database also contains data on demography and clinical status that could confound the relationship between COVID-19 and depression. Thus, the PREDIMED-Plus cohort provides a unique opportunity to evaluate the association of COVID-19 with depressive symptomatology in older adults with overweight/obesity and metabolic syndrome.

Ascertainment of variables

Exposure: SARS-CoV-2 infection

For the primary analysis, the main exposure was a confirmed COVID-19 event in a participant (positive/negative) as adjudicated by the Clinical Event Ascertainment Committee of the trial. The clinical event determination was based on the information from participant medical records reviewed annually by the participating physicians (CDC, 2020). Overall, 410 participants in this analysis were COVID-19 positive. Participants who did not have a confirmed/probable COVID-19-positive diagnosis were considered COVID-19 negative (i.e. assumed to have not experienced the infection). The COVID-19 status accordingly established as a dichotomous variable (positive/negative) was used to define the exposure.

A supplementary analysis was performed using a subsample (n = 3982) of the PREDIMED-Plus participants who had undergone serology testing with SARS-CoV-2 IgG ELISA Kits. These tests obtained between 3 March 2020, and 25 December 2021, classified participants as COVID-19 negative (n = 3698)/COVID-19 positive (n = 287). COVID-19 status was accordingly defined as a dichotomous predictor variable in the supplementary analysis (see supplementary methods for details).

Outcome: depression assessment

Depression assessment in the PREDIMED-Plus

As per the PREDIMED-Plus protocol, participants’ complete annual assessments of depressive symptomatology were performed using the validated 21-item Spanish version of the Beck Depression Inventory-II (BDI-II) (Fernández, Valverde, & Perdigón, Reference Fernández, Valverde and Perdigón2003). Each item in the BDI-II has four possible answers with scores ranging from 0 to 3 in accordance with symptom severity. Thus, the sum total of the BDI-II score ranges between 0 and 63 points, with higher scores indicating a higher propensity for depression.

Identifying pre- and post-COVID-19 measurements

From the annual assessments of depressive symptomatology, a pre-COVID-19 and a post-COVID-19 measurements were identified for each participant based on their COVID-19 event status. For COVID-19-positive participants, the last available BDI-II assessment prior to the COVID-19 diagnosis date was ascertained as the pre-COVID-19 measurement. In these participants, the BDI-II score available from the first post-COVID-19 follow-up visit was ascertained as the post-COVID-19 BDI-II score. For COVID-19-negative participants, the date of identification of the first COVID-19 case in Spain (31 January 2020) was used as a hinge to identify BDI-II scores from comparable time points as COVID-19-positive participants. Thus, in COVID-19-negative participants, BDI-II scores from a visit before 31 January 2020 indicated pre-COVID-19 measurement and those from the subsequent visit were used as post-COVID-19 measurement (online Supplementary Fig. S1). The duration between pre- and post-COVID-19 depression measurements and the time elapsed from the COVID-19 event at post-COVID-19 depression measurements were calculated in weeks. Time elapsed at post-COVID-19 depression measurement was further categorized as ≤12 weeks and >12 weeks to stratify the time-dependent effects of COVID-19 on depression (Klaser et al., Reference Klaser, Thompson, Nguyen, Sudre, Antonelli, Murray and Steves2021; Renaud-Charest et al., Reference Renaud-Charest, Lui, Eskander, Ceban, Ho, Di Vincenzo and McIntyre2021).

Categorizing post-COVID-19 depression risk as the outcome variable

For the primary analysis, the post-COVID-19 BDI-II score was used as a continuous outcome. BDI-II scores have also been categorized to identify the risk of depression: scores 0–13 indicate minimal risk, and scores ≥ 14 identify elevated risk (Becker, Steer, & Brown, Reference Becker, Steer and Brown1996). For a secondary analysis, we categorized elevated depression risk accordingly and treated it as a binary outcome. In addition, since a cut-off ≥ 12 had an adequate specificity index and diagnostic concordance and detects major depressive episodes in 93% of Spanish individuals (Sanz Fernández, Reference Sanz Fernández2013), a supplementary analysis using a cut-off ≥ 12 was also conducted.

Assessment of confounder variables

Data on potentially relevant confounders were also obtained from the PREDIMED-Plus database (See supplementary methods). Pre-COVID-19 visit covariates used in the models included age (years), marital status (levels: single/divorced, married or widow/widower), adherence to the Mediterranean diet (er-MEDAS score), alcohol consumption (g/day), total physical activity (METs min/week), and body mass index (BMI; kg/m2). For other confounders including sex (man/woman), education (<high school, high school, and university), intervention group (A/B), recruitment centre size (>400, 300–400, 250–300, and <250), smoking status (never/former smoker/current smoker), the prevalence of type 2 diabetes mellitus (yes/no), hypercholesterolemia (yes/no), hypertension (yes/no), and cognitive performance [Mini-Mental State Examination (MMSE) scores], and study baseline data were used to reduce missing data. Since the time elapsed since COVID-19 can impact depression assessments (Renaud-Charest et al., Reference Renaud-Charest, Lui, Eskander, Ceban, Ho, Di Vincenzo and McIntyre2021), this duration (weeks) was adjusted for confounding in regression models. Since pre- and post-COVID-19 BDI-II scores were highly correlated, pre-COVID-19 BDI-II scores were adjusted as a covariate in all models.

Statistical analyses

The present analysis was conducted as a prospective cohort study using the PREDIMED-Plus database with the COVID-19 event status updated until 31 December 2021. All other data (depression outcomes and confounder data) were sourced from the database that was updated until 4 November 2022. This allowed for sourcing depression assessments before and after COVID-19 and enabled the inclusion of both acute (<12 weeks) and long-term (≥12 weeks) associations of COVID-19 on depressive symptomatology. We included participants who had completed depression questionnaire assessments both before and after the ascertainment of COVID-19 event status.

In a preliminary cross-sectional exploration, we compared the characteristics and the timing of depression assessment of COVID-19-negative and positive participants using the Chi-Square and Mann–Whitney U tests, as appropriate.

The primary analysis evaluated the longitudinal relationship of COVID-19 on post-infection depression symptomatology (BDI-II scores) using linear regression models, considering the COVID-19-negative status as the reference category. In addition to the unadjusted crude model, three other models were tested. Model 1 adjusted for age, sex, education, marital status, intervention group, recruitment centre size, pre-COVID-19 BDI-II scores, and time since COVID-19 for depression assessments as confounders. Model 2 additionally adjusted for the presence of obesity (BMI ≥ 30 kg/m2), type 2 diabetes mellitus, hypertension, hypercholesterolemia, and cognitive performance on recruitment to the trial. Model 3 also adjusted for lifestyle factors including scores of adherence to the Mediterranean diet, total physical activity levels, smoking status, and alcohol consumption. Alcohol consumption was used as a quadratic term in the model to accommodate for a nonlinear relationship with the outcome. All analyses were conducted with robust estimates of the variance to correct for intracluster correlation. This procedure was used to control for the allocation of household members into the same intervention group without randomization.

A secondary logistic regression analysis using the same models developed for the main analysis was performed with elevated depression risk post-COVID-19 (BDI-II cut-off ≥ 14) as a binary outcome.

Furthermore, to negate over-adjustments, a directed acyclic graph (DAG) (Textor, van der Zander, Gilthorpe, Liśkiewicz, & Ellison, Reference Textor, van der Zander, Gilthorpe, Liśkiewicz and Ellison2016) was modelled (online Supplementary Fig. S2) and a minimal adjustment set was identified for both the linear and logistic regression models. This minimal model adjusted only for pre-COVID-19 depression scores. An additional supplementary logistic regression analysis was undertaken using a BDI-II score ≥ 12 as the cut-off for elevated depression risk.

Effect modification of the association by potential confounders [age group (≤70 or >70 years), sex, intervention group, disease conditions, and time elapsed post-COVID-19] was assessed by introducing product terms in the multivariable model. Further, sub-analyses that stratified results by factors that showed significant interaction (sex, presence of pre-COVID-19 high depression risk, and time elapsed post-COVID-19 during depression assessments) were undertaken. Finally, supplementary linear and logistic analyses were conducted in the sub-sample with serology results to ascertain COVID-19 status.

Data were analysed using the Stata 14 software (StataCorp, College Station, TX, USA), and statistical significance was set at a two-tailed p value <0.05 (see supplementary methods for details).

Results

This analysis included a total of 5486 PREDIMED-Plus participants (51.7% men) with a median [interquartile range (IQR)] age of 69.7 (7.4) years (Fig. 1). Table 1 shows their characteristics stratified by COVID-19 event status. At the pre-COVID-19 visit, participants had a median (IQR) BDI-II score of 5 (8), and these scores did not significantly differ by COVID-19 status. Approximately 14% of the participants included in this analysis had elevated depression risk at the pre-COVID-19 visit with no significant difference in the prevalence between COVID-19-positive and COVID-19-negative individuals. A COVID-19-positive status was associated with the male sex. COVID-19-positive participants were also more likely to report having been former smokers. All other factors evaluated were comparable in COVID-19-positive and COVID-19-negative individuals at the pre-COVID-19 visit.

Figure 1. Flow diagram for PREDIMED-Plus participants included in the analysis to evaluate the impact of COVID-19 on depression. BDI-II, Beck Depression Inventory-II; BMI, body mass index; COVID-19, coronavirus disease 2019; MMSE, Mini-Mental State Examination. aAnalysis used COVID-19 event confirmation data from the PREDIMED-Plus database updated until December 2021. bAnalysis used depressive and covariate assessments from the PREDIMED-Plus database updated until November 2022. #Age, sex, education, intervention group, recruitment centre, smoking status, physical activity, adherence to the Mediterranean diet, BMI, prevalence of baseline diabetes, hypertension, and hypercholesterolemia had no missing data for this analysis. Marital status: 12/5486 (0.2%) missing data. Missing data were replaced with the mode of the variable for the cohort. Alcohol consumption: 15/5486 (0.3%) missing data. Missing data were replaced with cohort mean consumption by gender (*men = 17.47; women = 4.59 g/day). MMSE data: 135/5486 missing data (2.4%). No imputation was performed for missing data.

Table 1. Participant characteristics according to COVID-19 status

BDI-II scores, Beck Depression Inventory-II; BMI, body mass index; COVID-19, coronavirus disease 2019; IQR, interquartile range; MedDiet, Mediterranean diet; MMSE, Mini-Mental State Examination.

Data are n (%) or median (IQR) for categorical and quantitative variables, respectively, unless specified.

a p Values for comparisons between groups were tested using the Mann–Whitney test (owing to the skewed nature of the distribution) or χ2, as appropriate.

b Data are presented as median (IQR).

c Data are from study baseline.

d Data from pre-COVID-19 measurement.

e Notes on scales: BDI-II Scores range between 0 and 63. Elevated depression risk is described as BDI-II scores ≥ 14. MMSE scores range between 0 and 30; the higher the scores greater the cognitive performance. Possible MedDiet scores range between 0 and 17. Higher MedDiet scores represent higher adherence to the Mediterranean diet.

f Duration data are presented as median (25th–75th percentile).

Post-COVID-19 depression assessments in COVID-19-positive participants were on average assessed 23 weeks post-infection. The duration between pre- and post-COVID-19 depression assessments was significantly shorter in COVID-19-positive v. COVID-19-negative participants (p < 0.001, Table 1). However, the mean difference in duration between pre- and post-COVID-19 depression assessments between those who had and did not have the infection was <5 days.

Table 2 evaluates the longitudinal association of COVID-19 on BDI-II scores over a median (IQR) duration of 29.4 (24.7) weeks post-COVID-19 in the cohort. In the fully adjusted model, SARS-CoV-2 infection was significantly associated with post-COVID-19 BDI-II scores [β (95% confidence interval (CI)) 0.65 (0.15–1.15), p = 0.011].

Table 2. Longitudinal association of COVID-19 status with post-infection depression assessments (BDI-II scores)a in the PREDIMED-Plus cohort (β [95% CI])

BDI-II, Beck Depression Inventory-II; CI, confidence interval; COVID-19, coronavirus disease 2019; MMSE, Mini-Mental State Examination.

Linear regression model: exposure = COVID-19 status (positive or negative); outcome: post-COVID-19 BDI-II score. Reference category: COVID-19-negative status.

The crude model only uses COVID-19 status (positive or negative) as the predictor variable in the model.

Model 1: Adjusted for age, sex, education, marital status, intervention group, cluster randomization, recruitment centre size, pre-COVID-19 BDI-II scores, and time since infection for post-COVID-19 depression assessments.

Model 2: Model 1 in addition adjusted for the presence of obesity, diabetes mellitus, hypertension, hypercholesterolemia, and baseline cognition (MMSE scores).

Model 3: Model 2 in addition adjusted for adherence to Mediterranean diet scores, smoking status, physical activity, and alcohol consumption.

a Depression assessment from the first scheduled follow-up visit after COVID-19 was used to calculate post-COVID-19 BDI-II scores.

b β (95% CI) was calculated using linear regression models.

c Sex is included as a predictor for the analysis of the total sample in Model 1. A stratified analysis by sex was conducted to determine differences, if any, in the impact of COVID-19 on depression measurements.

Table 3 summarises the positive association between a SARS-CoV-2 infection and elevated depression risk post-COVID-19 in this group of older adults at high cardiometabolic risk. In the final model, COVID-19 was associated with a 62% increase in the odds of observing elevated depression risk post-COVID-19 in the cohort [odds ratio (OR), 95% CI 1.13–2.30, p = 0.008].

Table 3. Longitudinal association between COVID-19 status and post-infection elevated depression risk in the PREDIMED-Plus cohort [OR (95% CI)]

BDI-II scores, Beck Depression Inventory-II; COVID-19, coronavirus disease 2019; CI, confidence interval; MMSE, Mini-Mental State Examination; OR, odds ratio.

Logistical regression model: exposure = COVID-19 status (positive or negative); outcome = elevated depression risk post-COVID-19. Reference category: COVID-19-negative status.

Model 1: Adjusted for age, sex, education, marital status, intervention group, cluster randomization, recruitment centre size, pre-COVID-19 BDI-II scores, duration post-COVID-19 measurements.

Model 2: Model 1 additionally adjusted for the presence of obesity, diabetes mellitus, hypertension, hypercholesterolemia, and baseline cognition (MMSE scores).

Model 3: Model 2 additionally adjusted for adherence to Mediterranean diet scores, smoking status, physical activity, and alcohol consumption.

a OR (95% CI) was calculated using logistic regression models.

b Sex included as a predictor for the analysis of the total sample in Model 1. A stratified analysis by sex was conducted to determine differences, if any, in the impact of COVID-19 on depression measurements.

cElevated depression risk is defined as BDI-II score ≥ 14, absence of elevated depression risk as BDI-II score < 14. Depression assessment from the first scheduled follow-up visit after the COVID-19 infection was used to categorize post-COVID-19 depressive symptomatology.

These results remained unchanged in the supplementary analysis using a minimal adjustment model (online Supplementary Analysis 1: Supplementary Table S1). In addition, the results of logistic regression quantifying the longitudinal association between COVID-19 and elevated depression risk post-COVID-19 remained unchanged when the cut-off for BDI-II to identify the heightened risk was lowered to 12 points (online Supplementary Analysis 2, Supplementary Table S1).

Evaluation of interactions

Significant interactions (p < 0.01) with COVID-19 were observed for sex and the presence of pre-COVID-19 elevated depression risk (online Supplementary Fig. S3). Fully adjusted predicted post-COVID-19 BDI-II scores and probabilities of elevated depression risk in the stratified sub-analysis undertaken for these factors are visualised in online Supplementary Fig. S4.

Sex

At the pre-COVID visit, 68% of the participants who exhibited depressive symptomatology were women (p < 0.001). In women, a positive COVID-19 event was associated with an increase in BDI-II scores measured post-COVID-19 [β (95% CI) 1.38 (0.44–2.33), p = 0.004, Table 2] in the fully adjusted model. Similarly, a positive COVID-19 status in women was associated with an 82% increase in heightened depression risk post-COVID-19, even when controlled for potential confounders including pre-COVID-19 BDI-II scores (OR, 95% CI 1.17–2.86; p = 0.008, Table 3). However, these associations were not significant in men.

Elevated depression risk pre-COVID-19

Elevated depression risk pre-COVID-19 was positively associated with a similar assessment at the post-COVID-19 visit. Approximately 50% (n = 377) of those who recorded BDI-II scores ≥14 (n = 758) and 6% (n = 284) of those who scored <14 at the pre-COVID-19 visit exhibited elevated depression risk at their post-COVID-19 visit. Table 4 stratifies the prospective association between SARS-CoV-2 infection and elevated depression risk post-COVID-19, by pre-COVID-19 depression risk levels. In individuals with BDI-II scores <14 at the prior visit, a positive COVID-19 event was associated with a 72% increase in the risk of elevated depression post-COVID-19, in the fully adjusted model (OR, 95% CI 1.17–2.62; p = 0.008, Table 4).

Table 4. Longitudinal association between COVID-19 status and depressive symptomatology in the PREDIMED-Plus cohort, stratified by depression risk at pre-COVID-19 assessment (ORa or β b coefficients and 95% CI)

BDI-II scores, Beck Depression Inventory-II; COVID-19, coronavirus disease 2019; CI, confidence interval; MMSE, Mini-Mental State Examination; OR, odds ratio.

Elevated depression risk is defined as BDI-II score ≥ 14, minimal depression risk as BDI-II score < 14. Depression assessment from the first scheduled follow-up visit after the COVID-19 infection was used to evaluate post-COVID-19 depressive symptomatology.

Reference category: COVID-19-negative status.

Model 1: Adjusted for age, sex, education, marital status, intervention group, cluster randomization, recruitment centre size, pre-COVID-19 BDI-II scores, and time since infection for post-COVID-19 depression assessments.

Model 2: Model 1 additionally adjusted for the presence of obesity, diabetes mellitus, hypertension, hypercholesterolemia, and baseline cognition (MMSE scores).

Model 3: Model 2 additionally adjusted for adherence to Mediterranean diet scores, smoking status, physical activity, and alcohol consumption.

a β coefficient (95% CI) was calculated using linear regression models. Exposure = COVID-19 status (positive or negative); outcome: post-COVID-19 BDI-II scores.

b OR (95% CI) was calculated using logistic regression models. Exposure = COVID-19 status (positive or negative); outcome = elevated depressive risk post-COVID-19 (yes/no).

A significant interaction was also observed between the timing of depression assessment and COVID-19 status (p < 0.05). Results stratified by timing of post-COVID-19 depression assessment are shown in online Supplementary Table S2. While the directionality of the relationship between a COVID-19-positive status and depression scores remained consistent, these associations were statistically significant only among participants who had their depression assessment conducted after 12 weeks following COVID-19 diagnosis (COVID-19-positive participants) or after 12 weeks following the first confirmed case of COVID-19 in Spain (COVID-19-negative participants).

In the replication analysis in the subsample with serology results to confirm COVID-19 status, the directionality of the results remained unchanged. However, the association was no longer statistically significant (n = 3801, 284 cases of COVID-19) (online Supplementary Table S3).

Discussion

We examined the association of COVID-19 with depressive symptomatology in older adults with overweight/obesity and metabolic syndrome enrolled in the PREDIMED-Plus trial in Spain. Spain with an increasingly ageing population was among the European countries most affected by the pandemic (Pollán et al., Reference Pollán, Pérez-Gómez, Pastor-Barriuso, Oteo, Hernán, Pérez-Olmeda and Villa2020). COVID-19 was associated with a small but significant and persistent increase in post-infection depression scores in this population. These findings add to the existing global evidence on the mental health consequences of COVID-19 (Deng et al., Reference Deng, Zhou, Hou, Silver, Wong, Chang and Zuo2021; Klaser et al., Reference Klaser, Thompson, Nguyen, Sudre, Antonelli, Murray and Steves2021; Meng et al., Reference Meng, Xu, Dai, Zhang, Liu and Yang2020; Renaud-Charest et al., Reference Renaud-Charest, Lui, Eskander, Ceban, Ho, Di Vincenzo and McIntyre2021).

Post-infection increases in depressive symptomatology associated with infections that have a prolonged convalescence have biological and psychological bases (Kim, Yoo, Lee, Lee, & Shin, Reference Kim, Yoo, Lee, Lee and Shin2018). Biologically, the escalation of depressive symptoms after COVID-19 stems from increased inflammation (Lyra e Silva, Barros-Aragão, De Felice, & Ferreira, Reference Lyra e Silva, Barros-Aragão, De Felice and Ferreira2022; Mazza et al., Reference Mazza, De Lorenzo, Conte, Poletti, Vai, Bollettini and Benedetti2020). COVID-19 is a hyperinflammatory disease with systemic and brain inflammation, leading to acute and persistent neurological and psychological disturbances (Lyra e Silva et al., Reference Lyra e Silva, Barros-Aragão, De Felice and Ferreira2022). COVID-19 could also be a stress-inducing traumatic event, and patients who experience traumatic events are known to have higher inflammation markers (Fernández-Sevillano et al., Reference Fernández-Sevillano, González-Ortega, MacDowell, Zorrilla, López, Courtet and González-Pinto2022). Proinflammatory cytokines are associated with the development of depression, irrespective of baseline scores, indicating that inflammation temporally precedes and increases the depression risk (Martínez-Cengotitabengoa et al., Reference Martínez-Cengotitabengoa, Carrascón, O'Brien, Díaz-Gutiérrez, Bermúdez-Ampudia, Sanada and González-Pinto2016). In addition, the increased depression risk in Middle East Respiratory Syndrome (MERS) patients quarantined in the hospital was ascribed to psychological factors including tension, fear, anger, mistrust, uncertainty, and depressed mood due to the infection itself and the subsequent isolation during quarantine (Kim et al., Reference Kim, Yoo, Lee, Lee and Shin2018), socio-economic and family consequences. These mechanisms could collectively explain the association of COVID-19 with increased depressive symptomatology. In addition, the pandemic nature of the COVID-19 outbreak and the widespread adoption of public health measures could have compounded the association of COVID-19 with depressive symptoms. Therefore, it is likely that the magnitude of the impact of COVID-19 on mental health, specifically among the vulnerable including older adults, is more prominent in comparison to common acute illnesses.

While the association between COVID-19 and depression risk was statistically significant, the effect size was small, and hence, its clinical significance is debatable. Nevertheless, the effect of COVID-19 on depressive symptoms is in line with the repeated calls for mental health interventions in older adults, particularly in older women surviving COVID-19 (Mazza et al., Reference Mazza, De Lorenzo, Conte, Poletti, Vai, Bollettini and Benedetti2020; Meng et al., Reference Meng, Xu, Dai, Zhang, Liu and Yang2020; Renaud-Charest et al., Reference Renaud-Charest, Lui, Eskander, Ceban, Ho, Di Vincenzo and McIntyre2021). Moreover, contrary to the existing understanding that prior mental health conditions make individuals particularly vulnerable to the psychological impact of COVID-19 (Mazza et al., Reference Mazza, De Lorenzo, Conte, Poletti, Vai, Bollettini and Benedetti2020; Meng et al., Reference Meng, Xu, Dai, Zhang, Liu and Yang2020; Renaud-Charest et al., Reference Renaud-Charest, Lui, Eskander, Ceban, Ho, Di Vincenzo and McIntyre2021), we found that COVID-19 was significantly associated with elevated depressive risk post-infection in PREDIMED-Plus participants without a similar risk at the pre-COVID-19 visit. These results provide new insights into the need for holistic management of COVID-19 in older adults who were more vulnerable to infection and had poorer survival rates in the initial phases of the pandemic, owing to senescence and comorbidity-related changes in the immune system (Mueller et al., Reference Mueller, McNamara and Sinclair2020). Aging attenuates coping strategies (Meng et al., Reference Meng, Xu, Dai, Zhang, Liu and Yang2020), while self-awareness of the aging-related increased the risk of mortality from the pandemic and poorer coping tendencies could contribute to increased and persistent depressive tendencies in older adults experiencing COVID-19. Furthermore, poorer physical health increases the risk for poorer mental health post-COVID-19 (Robinson, Sutin, Daly, & Jones, Reference Robinson, Sutin, Daly and Jones2022). Hence, among older adults at high cardiometabolic health risk, preventive mental health interventions to manage depressive symptomatology may be required irrespective of pre-COVID-19 mental health status.

Previous reports suggest that the mental health effects of COVID-19 are transitory and attenuate 12 weeks after the infection (Klaser et al., Reference Klaser, Thompson, Nguyen, Sudre, Antonelli, Murray and Steves2021; Renaud-Charest et al., Reference Renaud-Charest, Lui, Eskander, Ceban, Ho, Di Vincenzo and McIntyre2021). However, we found no evidence to support this contention. The observed lack of significance of the results in the group with post-COVID-19 assessments conducted within 12 weeks of the date of infection could be due to insufficient statistical power in this group. Nevertheless, consistent results in the group that had their depression assessments performed 12 weeks or later after SARS-CoV-2 infection confirms that COVID-19 posed an extended mental health risk in this group of older adults with heightened metabolic risks, even in the absence of depression in pre-COVID-visits.

We could attribute this extended mental health consequence of COVID-19 to both the physiological consequences of COVID-19 and the prolonged lockdown instituted as public health measures to stem the spread of the disease. However, we have recently shown, albeit in a sub-sample of this cohort, that the lockdown was not associated with an increase in depressive symptomatology (Paz-Graniel et al., Reference Paz-Graniel, Babio, Nishi, Martínez-González, Corella, Fitó and Salas-Salvadó2023). Thus, it is highly likely that the persistent depressive symptomatology seen in this group is predominantly a consequence of the disease. These findings reemphasize that COVID-19-induced increases in depressive symptoms could be larger and more persistent in comparison to smaller changes observed for anxiety disorder symptoms and overall mental health functioning measures (Robinson et al., Reference Robinson, Sutin, Daly and Jones2022). With the increasing concern over ‘Long-COVID’, it is important to further monitor the long-term psychological impact of COVID-19 in older adults, specifically concerning depressive symptoms, even in the absence of depression in pre-COVID-visits.

Our study has limitations. First, BDI-II scores were self-reported and are not interpreted as a bonafide diagnosis of the presence/absence of depression. Nevertheless, BDI-II has been validated and used widely in Spain with sufficient specificity to identify individuals at the heightened risk for depression (Sanz Fernández, Reference Sanz Fernández2013). Second, while social and economic outcomes of the pandemic contribute to depression post-COVID-19 (Renaud-Charest et al., Reference Renaud-Charest, Lui, Eskander, Ceban, Ho, Di Vincenzo and McIntyre2021), this analysis did not account for regional variation in lockdown severity and its economic/social consequences. We believe that with adjustments for the recruitment centre size and education, we could have partially accounted for these factors. Third, some COVID-19-negative patients may have had asymptomatic infections that went undiagnosed, resulting in misclassification of cases. This is unlikely because we scrutinized all medical records during 2020 and 2021 when public health strategies for COVID-19 testing were stringent as the nation was in the process of maximizing vaccination coverage. We also recognize that protecting the integrity of the main trial precludes obtaining updated data for covariates such as the prevalence of diabetes, hypercholesterolemia, or hypertension for this analysis. However, the minimal adjustment model shows that the association may be independent of these variables. Furthermore, the results from the sub-sample with positive serology go in the same direction as the primary analysis, suggesting minimal effects of misclassification on this analysis. Finally, this analysis uses data from participants in a clinical trial and may not be widely generalizable.

Nevertheless, this analysis adds strong data to the existing evidence on the mental health sequelae of COVID-19 in a vulnerable group of older adults with overweight/obesity and metabolic syndrome. The sufficiently large PREDIMED-Plus cohort with scheduled data assessments from before the onset of the pandemic and after helps establish the impact of COVID-19 on depressive symptomatology in the cohort while adjusting for the time for depression determinations, an important confounder of this relationship (Renaud-Charest et al., Reference Renaud-Charest, Lui, Eskander, Ceban, Ho, Di Vincenzo and McIntyre2021). Furthermore, the similar time frame within which the pre- and post-COVID-19 assessments were obtained in all participants controls for many extraneous factors that could have increased the depression risk, independently of infection status. Moreover, COVID-19 event adjudication was performed by an independent committee removing any potential bias in the ascertainment of cases. Supplementary analyses using a lower cut-off for depression risk and serology results from a sub-sample confirmed the directionality of the results from the main analysis. Finally, we believe that the identification of a minimal adjustment set using a DAG to investigate the relationships involved in this analysis also removes concerns of over-adjustments in the models.

Our analyses do not consider vaccination status and type, the severity of COVID-19 infection, the infection strain or the treatment modality used, or the need for hospitalization among the COVID-19-positive participants. However, current evidence for the impact of these factors on post-COVID-19 depressive symptoms is inconsistent (Chen, Aruldass, & Cardinal, Reference Chen, Aruldass and Cardinal2022; Mazza et al., Reference Mazza, De Lorenzo, Conte, Poletti, Vai, Bollettini and Benedetti2020; Renaud-Charest et al., Reference Renaud-Charest, Lui, Eskander, Ceban, Ho, Di Vincenzo and McIntyre2021). It is possible that the severity of COVID-19 in the early days of the pandemic differed from those that occurred later. We found that while several of the strains reported in 2020 and 2021 caused severe infections, the omicron variant reported after November 2021 produced milder disease. However, only 46 cases in our cohort were diagnosed after November 2021, and we do not possess data on strain causing COVID-19 in our cohort to tease out these effects. Also, vaccination in Spain started on 27 December 2020, and the possibility that it might have influenced depression outcomes is restricted to approximately 4% of our population who had received at least one dose of the vaccine at the time of post-COVID-19 depression measurements. Nevertheless, considering these factors in future analyses will facilitate identifying sub-groups that would specifically benefit from mental health interventions. We also propose that future studies investigate the trajectory of depressive symptoms in COVID-19 patients using repeated measurements post-infection. Such an evaluation will help better understand the time-dependent mental health effects of COVID-19.

Implications for practice

Overall, our findings support a call for mental health interventions to tackle increased depressive tendencies post-COVID-19 infection in older adults, particularly in women. Furthermore, in this Spanish cohort of older adults with overweight/obesity and metabolic syndrome, the association between COVID-19 and depressive symptoms was persistent and observable after 12 weeks post-COVID-19. Importantly, strategies to mitigate depression should be extended to older adults with cardiometabolic health risks, who do not exhibit heightened depressive symptomatology prior to a SARS-CoV-2 infection.

Supplementary material

The supplementary material for this article can be found at https://doi.org/10.1017/S0033291723002313.

Acknowledgements

The authors wish to thank the PREDIMED-Plus participants and staff for their engagement, as well as the primary care centres involved in the study. We also thank the Cerca Programme of the Generalitat de Catalunya, the CIBEROBN, CIBERESP, and CIBERDEM initiatives of Instituto de Salud Carlos III in Spain. We are grateful to Dr Jesús Francisco García Gavilán for his input on the methodology used in this manuscript. INSA-Ma María de Maeztu Unit of Excellence (grant CEX2021-001234-M funded by MICIN/AEI/FEDER, UE).

Author contributions

All authors (1) made substantial contributions to the study concept or the data analysis or interpretation; (2) drafted the manuscript or revised it critically for the important intellectual concept; (3) approved the final version of the manuscript to be published; and (4) agreed to be accountable for all aspects of the work.

Financial support

This work was supported by a project grant from the Fundación Francisco Soria Melguizo. The PREDIMED-Plus trial was supported by the official Spanish Institutions for funding scientific biomedical research, CIBER Fisiopatología de la Obesidad y Nutrición (CIBEROBN) and Instituto de Salud Carlos III (ISCIII), through the Fondo de Investigación para la Salud (FIS), which is co-funded by the European Regional Development Fund (six coordinated FIS projects leaded by J. S.-S. and J. Vi., including the following projects: PI13/00673, PI13/00492, PI13/00272, PI13/01123, PI13/00462, PI13/00233, PI13/02184, PI13/00728, PI13/01090, PI13/01056, PI14/01722, PI14/00636, PI14/00618, PI14/00696, PI14/01206, PI14/01919, PI14/00853, PI14/01374, PI14/00972, PI14/00728, PI14/01471, PI16/00473, PI16/00662, PI16/01873, PI16/01094, PI16/00501, PI16/00533, PI16/00381, PI16/00366, PI16/01522, PI16/01120, PI17/00764, PI17/01183, PI17/00855, PI17/01347, PI17/00525, PI17/01827, PI17/00532, PI17/00215, PI17/01441, PI17/00508, PI17/01732, PI17/00926, PI19/00957, PI19/00386, PI19/00309, PI19/01032, PI19/00576, PI19/00017, PI19/01226, PI19/00781, PI19/01560, PI19/01332, PI20/01802, PI20/00138, PI20/01532, PI20/00456, PI20/00339, PI20/00557, PI20/00886, PI20/01158); the Especial Action Project entitled: Implementación y evaluación de una intervención intensiva sobre la actividad física Cohorte PREDIMED-Plus grant to J. S.-S.; the European Research Council (Advanced Research Grant 2014–2019; agreement #340918) granted to M. Á. M.-G.; the Recercaixa (number 2013ACUP00194) grant to J. S.-S.; grants from the Consejería de Salud de la Junta de Andalucía (PI0458/2013, PS0358/2016, and PI0137/2018); the PROMETEO/2017/017 and PROMETEO 21/2021 grants from the Generalitat Valenciana; the SEMERGEN grant; Juan de la Cierva-Incorporación research grant (IJC2019-042420-I) of the Spanish Ministry of Economy, Industry and Competitiveness and European Social Funds. This research was also partially funded by EU-H2020 Grants (Eat2beNICE/ H2020-SFS-2016-2); and the Horizon 2020 PRIME study (Prevention and Remediation of Insulin Multimorbidity in Europe; grant agreement #847879). S. G. S. was a recipient of the Maria Zambrano Fellowship with funding support from the Ministry of Universities and the Recovery, Transformation and Resilience Plan, Spain. The Fellowship is ‘Funded by the European Union – NextGenerationEU’. S. K. N. was supported by a postdoctoral fellowship from the Canadian Institutes of Health Research (CIHR, MFE-171207). C. G.-M. was supported by a predoctoral grant from the University of Rovira I Virgili (2020PMF-PIPF-37). J. S.-S. was partially supported by ICREA under the ICREA Academia program. We thank CERCA Programme/Generalitat de Catalunya for institutional support. The funders had no role in study design, data collection and analysis, the decision to publish, or the preparation of the manuscript.

Competing interest

The authors have no conflict of interests to declare.

Ethical standards

The authors assert that the studies involving human participants were reviewed and approved by the study and were conducted in compliance with the guidelines of the Declaration of Helsinki. The study was approved by the Institutional Review Boards of all participating centres. The patients/participants provided their written informed consent to participate in this study.

References

Becker, A. B., Steer, R. A., & Brown, G. K. (1996). BDI-II: Beck depression inventory-second edition manual. San Antonio, TX: The Psychological Corporation.Google Scholar
Castro, A., Roca, M., Ricci-Cabello, I., García-Toro, M., Riera-Serra, P., Coronado-Simsic, V., … Gili, M. (2021). Adherence to lifestyle interventions for treatment of adults with depression: A systematic review and meta-analysis. International Journal of Environmental Research and Public Health, 18(24), 13268. doi:10.3390/ijerph182413268Google Scholar
CDC. (2020, February 11). Labs. Retrieved December 12, 2022, from Centers for Disease Control and Prevention website: https://www.cdc.gov/coronavirus/2019-ncov/lab/serology-testing.htmlGoogle Scholar
Channappanavar, R., & Perlman, S. (2020). Age-related susceptibility to coronavirus infections: Role of impaired and dysregulated host immunity. The Journal of Clinical Investigation, 130(12), 62046213. doi:10.1172/JCI144115Google Scholar
Chen, S., Aruldass, A. R., & Cardinal, R. N. (2022). Mental health outcomes after SARS-CoV-2 vaccination in the United States: A national cross-sectional study. Journal of Affective Disorders, 298(Pt A), 396399. doi:10.1016/j.jad.2021.10.134Google Scholar
Deng, J., Zhou, F., Hou, W., Silver, Z., Wong, C. Y., Chang, O., … Zuo, Q. K. (2021). The prevalence of depression, anxiety, and sleep disturbances in COVID-19 patients: A meta-analysis. Annals of the New York Academy of Sciences, 1486(1), 90111. doi:10.1111/nyas.14506Google Scholar
Dunbar, J. A., Reddy, P., Davis-Lameloise, N., Philpot, B., Laatikainen, T., Kilkkinen, A., … Janus, E. D. (2008). Depression: An important comorbidity with metabolic syndrome in a general population. Diabetes Care, 31(12), 23682373. doi:10.2337/dc08-0175Google Scholar
Fernández, J. S., Valverde, C. V., & Perdigón, A. L. (2003). Adaptación española del Inventario para la Depresión de Beck-II (BDI-II): 2. Propiedades psicométricas en población general. Clínica y salud: Investigación Empírica en Psicología, 14(3), 249280.Google Scholar
Fernández-Sevillano, J., González-Ortega, I., MacDowell, K., Zorrilla, I., López, M. P., Courtet, P., … González-Pinto, A. (2022). Inflammation biomarkers in suicide attempts and their relation to abuse, global functioning and cognition. The World Journal of Biological Psychiatry, 23(4), 307317. doi:10.1080/15622975.2021.1988703Google Scholar
Guo, P., Benito Ballesteros, A., Yeung, S. P., Liu, R., Saha, A., Curtis, L., … Cheke, L. G. (2022). COVCOG 1: Factors predicting physical, neurological and cognitive symptoms in long COVID in a community sample. A first publication from the COVID and cognition study. Frontiers in Aging Neuroscience, 14, 804922. doi:10.3389/fnagi.2022.804922Google Scholar
Khaledi, M., Haghighatdoost, F., Feizi, A., & Aminorroaya, A. (2019). The prevalence of comorbid depression in patients with type 2 diabetes: An updated systematic review and meta-analysis on huge number of observational studies. Acta Diabetologica, 56(6), 631650. doi:10.1007/s00592-019-01295-9Google Scholar
Kim, H.-C., Yoo, S.-Y., Lee, B.-H., Lee, S. H., & Shin, H.-S. (2018). Psychiatric findings in suspected and confirmed Middle East Respiratory Syndrome patients quarantined in hospital: A retrospective chart analysis. Psychiatry Investigation, 15(4), 355360. doi:10.30773/pi.2017.10.25.1Google Scholar
Klaser, K., Thompson, E. J., Nguyen, L. H., Sudre, C. H., Antonelli, M., Murray, B., … Steves, C. J. (2021). Anxiety and depression symptoms after COVID-19 infection: Results from the COVID symptom study app. Journal of Neurology, Neurosurgery and Psychiatry, 92(12), 12541258. doi:10.1136/jnnp-2021-327565Google Scholar
Lopez-Leon, S., Wegman-Ostrosky, T., Perelman, C., Sepulveda, R., Rebolledo, P. A., Cuapio, A., & Villapol, S. (2021). More than 50 long-term effects of COVID-19: A systematic review and meta-analysis. Scientific Reports, 11(1), 16144. doi:10.1038/s41598-021-95565-8Google Scholar
Lyra e Silva, N. M., Barros-Aragão, F. G. Q., De Felice, F. G., & Ferreira, S. T. (2022). Inflammation at the crossroads of COVID-19, cognitive deficits and depression. Neuropharmacology, 209, 109023. doi:10.1016/j.neuropharm.2022.109023Google Scholar
Martínez-Cengotitabengoa, M., Carrascón, L., O'Brien, J. T., Díaz-Gutiérrez, M.-J., Bermúdez-Ampudia, C., Sanada, K., … González-Pinto, A. (2016). Peripheral inflammatory parameters in late-life depression: A systematic review. International Journal of Molecular Sciences, 17(12), 2022. doi:10.3390/ijms17122022Google Scholar
Martínez-González, M. A., Buil-Cosiales, P., Corella, D., Bulló, M., Fitó, M., Vioque, J., … Salas-Salvadó, J. (2019). Cohort profile: Design and methods of the PREDIMED-Plus randomized trial. International Journal of Epidemiology, 48(2), 387388o. doi:10.1093/ije/dyy225Google Scholar
Mazza, M. G., De Lorenzo, R., Conte, C., Poletti, S., Vai, B., Bollettini, I., … Benedetti, F. (2020). Anxiety and depression in COVID-19 survivors: Role of inflammatory and clinical predictors. Brain, Behavior, and Immunity, 89, 594600. doi:10.1016/j.bbi.2020.07.037Google Scholar
Meng, H., Xu, Y., Dai, J., Zhang, Y., Liu, B., & Yang, H. (2020). Analyze the psychological impact of COVID-19 among the elderly population in China and make corresponding suggestions. Psychiatry Research, 289, 112983. doi:10.1016/j.psychres.2020.112983Google Scholar
Mueller, A. L., McNamara, M. S., & Sinclair, D. A. (2020). Why does COVID-19 disproportionately affect older people? Aging (Albany NY), 12(10), 99599981. doi:10.18632/aging.103344Google Scholar
Pano, O., Martínez-Lapiscina, E. H., Sayón-Orea, C., Martinez-Gonzalez, M. A., Martinez, J. A., & Sanchez-Villegas, A. (2021). Healthy diet, depression and quality of life: A narrative review of biological mechanisms and primary prevention opportunities. World Journal of Psychiatry, 11(11), 9971016. doi:10.5498/wjp.v11.i11.997Google Scholar
Paz-Graniel, I., Babio, N., Nishi, S. K., Martínez-González, , Corella, D., Fitó, M., … Salas-Salvadó, J. (2023). How did the COVID-19 lockdown pandemic affect the depression symptomatology in Mediterranean older adults with metabolic syndrome? Depression and Anxiety, 2023, e6765950. doi:10.1155/2023/6765950Google Scholar
Pollán, M., Pérez-Gómez, B., Pastor-Barriuso, R., Oteo, J., Hernán, M. A., Pérez-Olmeda, M., … Villa, A. V. de la. (2020). Prevalence of SARS-CoV-2 in Spain (ENE-COVID): A nationwide, population-based seroepidemiological study. The Lancet, 396(10250), 535544. doi:10.1016/S0140-6736(20)31483-5Google Scholar
Renaud-Charest, O., Lui, L. M. W., Eskander, S., Ceban, F., Ho, R., Di Vincenzo, J. D., … McIntyre, R. S. (2021). Onset and frequency of depression in post-COVID-19 syndrome: A systematic review. Journal of Psychiatric Research, 144, 129137. doi:10.1016/j.jpsychires.2021.09.054Google Scholar
Robinson, E., Sutin, A. R., Daly, M., & Jones, A. (2022). A systematic review and meta-analysis of longitudinal cohort studies comparing mental health before versus during the COVID-19 pandemic in 2020. Journal of Affective Disorders, 296, 567576. doi:10.1016/j.jad.2021.09.098CrossRefGoogle ScholarPubMed
Salas-Salvadó, J., Díaz-López, A., Ruiz-Canela, M., Basora, J., Fitó, M., Corella, D., … PREDIMED-Plus investigators. (2018). Effect of a lifestyle intervention program with energy-restricted Mediterranean diet and exercise on weight loss and cardiovascular risk factors: One-year results of the PREDIMED-Plus trial. Diabetes Care, 42(5), 777788. doi:10.2337/dc18-0836Google Scholar
Sanz Fernández, J. (2013). 50 years of the Beck Depression Inventory: Recommendations for using the Spanish adaptation of the BDI-II in clinical practice [Info:eu-repo/semantics/article]. Retrieved November 23, 2022, from https://eprints.ucm.es/id/eprint/36450/Google Scholar
Sparks, M. A., South, A. M., Badley, A. D., Baker-Smith, C. M., Batlle, D., Bozkurt, B., … Garovic, V. D. (2020). Severe acute respiratory syndrome coronavirus 2, COVID-19, and the renin-angiotensin system. Hypertension, 76(5), 13501367. doi:10.1161/HYPERTENSIONAHA.120.15948Google Scholar
Textor, J., van der Zander, B., Gilthorpe, M. S., Liśkiewicz, M., & Ellison, G. T. (2016). Robust causal inference using directed acyclic graphs: The R package ‘dagitty’. International Journal of Epidemiology, 45(6), 18871894. doi:10.1093/ije/dyw341Google Scholar
Wee, A. K. H. (2021). COVID-19's toll on the elderly and those with diabetes mellitus – is vitamin B12 deficiency an accomplice? Medical Hypotheses, 146, 110374. doi:10.1016/j.mehy.2020.110374Google Scholar
Wei, J., Hou, R., Zhang, X., Xu, H., Xie, L., Chandrasekar, E. K., … Goodman, M. (2019). The association of late-life depression with all-cause and cardiovascular mortality among community-dwelling older adults: Systematic review and meta-analysis. The British Journal of Psychiatry, 215(2), 449455. doi:10.1192/bjp.2019.74Google Scholar
Figure 0

Figure 1. Flow diagram for PREDIMED-Plus participants included in the analysis to evaluate the impact of COVID-19 on depression. BDI-II, Beck Depression Inventory-II; BMI, body mass index; COVID-19, coronavirus disease 2019; MMSE, Mini-Mental State Examination. aAnalysis used COVID-19 event confirmation data from the PREDIMED-Plus database updated until December 2021. bAnalysis used depressive and covariate assessments from the PREDIMED-Plus database updated until November 2022. #Age, sex, education, intervention group, recruitment centre, smoking status, physical activity, adherence to the Mediterranean diet, BMI, prevalence of baseline diabetes, hypertension, and hypercholesterolemia had no missing data for this analysis. Marital status: 12/5486 (0.2%) missing data. Missing data were replaced with the mode of the variable for the cohort. Alcohol consumption: 15/5486 (0.3%) missing data. Missing data were replaced with cohort mean consumption by gender (*men = 17.47; women = 4.59 g/day). MMSE data: 135/5486 missing data (2.4%). No imputation was performed for missing data.

Figure 1

Table 1. Participant characteristics according to COVID-19 status

Figure 2

Table 2. Longitudinal association of COVID-19 status with post-infection depression assessments (BDI-II scores)a in the PREDIMED-Plus cohort (β [95% CI])

Figure 3

Table 3. Longitudinal association between COVID-19 status and post-infection elevated depression risk in the PREDIMED-Plus cohort [OR (95% CI)]

Figure 4

Table 4. Longitudinal association between COVID-19 status and depressive symptomatology in the PREDIMED-Plus cohort, stratified by depression risk at pre-COVID-19 assessment (ORa or βb coefficients and 95% CI)

Supplementary material: File

S0033291723002313sup001.docx

Shyam et al. supplementary material

Download S0033291723002313sup001.docx(File)
File 273.9 KB
Supplementary material: File

S0033291723002313sup002.docx

Shyam et al. supplementary material

Download S0033291723002313sup002.docx(File)
File 321.2 KB
Supplementary material: File

S0033291723002313sup003.tif

Shyam et al. supplementary material

Download S0033291723002313sup003.tif(File)
File 3.6 MB
Supplementary material: File

S0033291723002313sup004.docx

Shyam et al. supplementary material

Download S0033291723002313sup004.docx(File)
File 128.2 KB
Supplementary material: File

S0033291723002313sup005.docx

Shyam et al. supplementary material

Download S0033291723002313sup005.docx(File)
File 74.9 KB
Supplementary material: File

S0033291723002313sup006.docx

Shyam et al. supplementary material

Download S0033291723002313sup006.docx(File)
File 66.7 KB
Supplementary material: File

S0033291723002313sup007.docx

Shyam et al. supplementary material

Download S0033291723002313sup007.docx(File)
File 67.6 KB
Supplementary material: File

S0033291723002313sup008.docx

Shyam et al. supplementary material

Download S0033291723002313sup008.docx(File)
File 67 KB