key: cord-0941999-j006ntze authors: Möller, Stephan P.; Apputhurai, Pragalathan; Tye-Din, Jason A.; Knowles, Simon R. title: Longitudinal assessment of the common sense model before and during the COVID-19 pandemic: A large coeliac disease cohort study date: 2021-12-24 journal: J Psychosom Res DOI: 10.1016/j.jpsychores.2021.110711 sha: 710982dc36d47b9a56444c6ffb2ad0973de498ac doc_id: 941999 cord_uid: j006ntze Objective Psychosocial factors likely play a substantial role in the well-being of those living with coeliac disease, especially during the COVID-19 pandemic, however, little research has examined well-being in this cohort using an integrated socio-cognitive model. This study had two aims: (1) Examine changes in gastrointestinal symptoms, psychosocial factors, and well-being outcomes (i.e., psychological distress, quality of life [QoL]) associated with the pandemic, (2) Examine the interrelationship of these variables across timepoints using the Common Sense Model (CSM). Methods 1697 adults with coeliac disease (Time 1, pre-pandemic; 83.1% female, mean age = 55.8, SD = 15.0 years) and 674 follow-up participants (Time 2, pandemic; 82.8% female, mean age = 57.0, SD = 14.4 years) completed an online questionnaire. Hypotheses were tested using repeated-measures MANOVA and cross-lagged panel model analyses. Results Participants reported improved QoL, and reduced gastrointestinal symptoms, negative illness perceptions and maladaptive coping from pre-pandemic to during the pandemic. There was no significant change in pain catastrophising or psychological distress. Cross-lagged effects showed gastrointestinal symptoms to predict negative illness perceptions, which in turn were predictive of poorer outcomes across all variables except pain catastrophising. Consistent with the CSM, there was a reciprocal relationship between illness perceptions and QoL over time. Maladaptive coping and pain catastrophising demonstrated limited predictive utility. Conclusion The COVID-19 pandemic appears to have had a small beneficial effect across several indices of well-being among adults with coeliac disease. Cross-lagged relationships highlight illness perceptions as a predictor of well-being outcomes and a potential target for psychosocial interventions. The COVID-19 pandemic has led to the deaths of millions, caused substantial socioeconomic difficulties [1] , and has been associated with a significant reduction in psychological well-being [2, 3] and quality of life (QoL) [4] . The pandemic has also presented notable challenges to those living the chronic illness due to both increased vulnerability to adverse COVID-19 outcomes [5, 6] and impeded access to relevant medical support structures [7] . Affecting over 1% of individuals globally [8] , coeliac disease is a chronic immune illness characterised by inflammation in the small bowel and various gastrointestinal and extra-intestinal symptoms [9] . The disease is principally treated with life-long and strict exclusion of dietary gluten (i.e., the gluten-free diet) [10] . Concerns regarding the impact of pandemic-related lockdown measures on access to gluten-free food products have been commonly reported by those living with the illness [11] . Moreover, while current evidence indicates that coeliac individuals are not at a greater risk of COVID-19 infection or serious COVID-19 symptoms [12] [13] [14] [15] , individuals with coeliac disease have commonly reported concerns regarding an increased vulnerability to the virus [11, 16] . Findings regarding pandemic-related changes in quality of life (QoL) in coeliac cohorts have been mixed. Comparisons of pre-and mid-pandemic cross-sectional studies have suggested a potential decrease in QoL associated with the pandemic in Italian coeliac cohorts [11, 17] . In contrast, findings in a Brazilian coeliac cohort have indicated a potential increase in QoL associated with the pandemic [18, 19] . The literature to date provides strong evidence for the role of psychosocial factors in post-diagnosis QoL in coeliac cohorts (For review, see [20] ). Studies examining the comparative effects of clinical and psychosocial factors in relation to QoL in coeliac disease have shown psychosocial factors (e.g., perceived gluten-free diet difficulty, coping) to predict QoL over and above clinical disease measures (e.g., dietary adherence, coeliac symptom frequency/severity, histological findings, serology) [21, 22] . Other psychosocial factors found to be associated with reduced QoL include illness perceptions (e.g., poor illness understanding, greater perceived burden and belief in negative illness consequences) [23, 24] , maladaptive coping [25] and psychological distress symptoms [22, 26] . Despite the strong evidence base for psychosocial predictors of QoL, limited research to date has explored this relationship using an integrated socio-cognitive model as seen in coeliac studies of glutenfree diet adherence (e.g., Theory of Planned Behaviour) [27] [28] [29] . The Common Sense Model (CSM) [30, 31] is a widely used socio-cognitive model that describes the cognitive, affective, and behavioural processes by which individuals respond to illness threats (for review, see [32, 33] ). In brief, illness threats (e.g., symptoms) impact patient outcomes (e.g., QoL, psychological well-being) via cognitive/affective representations of illness (e.g., perceived consequences, controllability, coherence, treatment efficacy, emotional representations) and coping patterns (e.g., adaptive/maladaptive coping strategies). Further, the model postulates that illness perceptions, coping strategies and outcomes are dynamically interrelated, changing over time via a process whereby appraisal of outcomes modulates illness beliefs and/or coping strategies [30] . A CSM-guided study of 1,697 coeliac disease adults conducted approximately five months prior to the COVID-19 pandemic identified several potential psychosocial mediators in the relationship between gastrointestinal symptoms and QoL [34] . Specifically, greater gastrointestinal symptoms were associated with more negative representations of illness (e.g., severe perceived life impact and illness concern, limited personal/treatment control), maladaptive coping (i.e., behavioural disengagement, venting, self-blame), pain catastrophising (i.e., tendency to focus on/magnify pain, feeling helpless toward the experience of pain), and psychological distress symptoms (i.e., depressed mood, anxiety, and Extending on the findings of Möller et al. [34] , the present study had two aims. Firstly, to assess changes in the psychosocial factors identified pre-pandemic in Möller et al. [34] by assessing participants 9 to 12 months later during the COVID-19 pandemic. Secondly, to conduct a CSM-guided investigation of the interrelationship between the same model variables across time. It was hypothesised that: Gastrointestinal symptoms were measured using the 10-item coeliac-specific version of the GSRS (i.e., CeD-GSRS) [38] . This version of the GSRS omits five items regarding upper GI and constipation-related GI symptomatology. Respondents are asked about GI symptoms over the previous seven days using a 7-point response format (1 = "No discomfort at all", 7 = "Very severe discomfort"). A total score, ranging from 10 to 70, is calculated by summing all items. Higher total scores indicate greater coeliac GI symptoms. Internal consistency for the CeD-GSRS was good at both timepoints (T1: α = .85, T2: α = .86). Psychological distress was measured using the DASS-21 [39] . The DASS-21 is a composite measure of psychological distress comprised of items relating to symptoms of depression, anxiety, and stress. Each item (e.g., "I felt that I had nothing to look forward to") is measured on a uniform 4-point scale (0 = "Did not apply to me at all", 3 = "Applied to me very much, or most of the time"), a total score (0 -126), wherein higher scores reflect greater J o u r n a l P r e -p r o o f psychological distress, is calculated by summing all items and multiplying by two. The DASS-21 had excellent internal consistency at both timepoints (T1: α = .94, T2: α = .93). Illness perceptions were measured using the Brief-IPQ [40] . Each item (e.g., "How emotions. These items were summed to obtain an overall score (0 -60), wherein a higher score reflected more negative/maladaptive illness perceptions. The scale had good to adequate internal consistency (T1: α = .81, T2: α = .78). Maladaptive coping was measured using the Brief COPE [42] . Each item (e.g., "I've J o u r n a l P r e -p r o o f Pain catastrophising was measured using the 13-item PCS [43] . The PCS measures pain catastrophising across three dimensions: rumination, magnification, and helplessness. All items are scored on a uniform 5-point scale (0 = "not at all", 4 = "all the time"). A total score (0 -52), wherein higher scores reflect greater pain catastrophising, is obtained by summing all items. The PCS had excellent internal consistency at both timepoints (T1: α = .94, T2: α = .95). Quality of life was measures using the 8-item EUROHIS-QOL [44] . The EUROHIS-QOL is a general QoL measure including questions that span multiple domains (i.e., overall quality of life, general health, energy, daily activities, self-esteem, social relationships, finances, and home life. Each item (e.g., "How satisfied are you with your ability to perform your daily living activities") is answered using a question-specific 5-point scale (e.g., 1 = "Very poor", 5 = "Very good"). A total score (8 -40), wherein higher scores reflect greater QoL, is calculated by summing all items. The EUROHIS-QOL had good internal consistency (T1: α = .88, T2: α = .87) at both timepoints. Data were obtained at two timepoints: pre-COVID-19 pandemic (Time 1: August -October 2019) and pandemic (Time 2: May -July 2020). The first wave of data was obtained via email invitation to an online survey, which was sent out to 4,287 individuals that had previously participated in coeliac research [45] . Data reported nationally [46, 47] . A second wave of infection centred in the state of Victoria emerged during mid-late June, resulting in the reintroduction of restrictions [48] . The first case of COVID-19 in New Zealand was reported on the 28 th of February [49] , with a nationwide lockdown enforced on the 25 th March 2020 [50] until the 27 th of April, after which restrictions were gradually lifted until June the 8 th where only international border restrictions remained [51] . As summarised in Table 2 , most of the sample (72%) were engaging in moderate social isolation (i.e., staying at home and only going out for food and engaging in social distancing). Analyses were conducted using SPSS v28 and AMOS v28. Given that the variables and design of the present study were derived from the cross-sectional model in Möller et al. [34], a one-way between-groups MANOVA was conducted to assess for baseline differences across model variables between participants who participated in the follow-up and those who J o u r n a l P r e -p r o o f did not. Furthermore, as the present cross-lagged panel model utilised measures derived from exploratory factor analysis within the original cross-sectional sample (i.e., illness perceptions, maladaptive coping; see [34] ), measurement invariance was assessed between the original cross-sectional sample (N = 1,697) [34] and the subset of participants that completed followup (N = 674) across all model variables. This involved assessing invariance across a series of increasingly constrained, nested models (see [52] ). In ascending order, these included configural invariance (i.e., the baseline model, indicating that the same latent variables are accounted for by the same indicator variables across groups), metric invariance (i.e., that each indicator variable's loading on the latent variable is similar across groups), and finally scalar invariance (i.e., equality of intercepts across groups). As suggested by Cheung and Rensvold [53] , a decrease in the comparative fit index (CFI) of more than .01 was considered to be the threshold for non-invariance when applying an additional constraint to the model. To account for potential measurement error-related biases in the model estimates [54] , the same process was then applied to assess invariance across both timepoints. The first hypothesis was tested using a one-way repeated measures MANOVA. The remaining hypotheses were tested using observed variables within a two-wave cross-lagged model as recommended by Hagger and Orbell [33] (see Figure 1 ). A between-groups MANOVA between participants that completed the follow-up (n = 674) and those that did not (n = 1,023) revealed no significant differences across all model A one-way within-subjects MANOVA found a statistically significant overall effect between the pre-pandemic and pandemic model variables (Pillai's Trace = .090, F (6,668) = 10.98, p < .001, η 2 = .09). As shown in the univariate tests in Table 3 , contrary to hypothesis 1, participants reported small, statistically significant reductions in GI symptoms, negative illness perceptions, and use of maladaptive coping strategies from pre-pandemic to pandemic. Furthermore, contrary to hypothesis 1, QoL had a small statistically significant improvement and there were no significant differences between timepoints for pain catastrophising and psychological distress. The purpose of the present study was two-fold. Firstly, to examine changes across psychosocial factors outlined in Möller et al. [34] associated with the emergence of the COVID-19 pandemic, and secondly, to conduct a CSM-guided cross-lagged panel model analysis examining interrelationships between these same factors over this period. The first hypothesis was unsupported. Contrary to expectation, participants reported small but statistically significant improvements across GI symptoms, illness perceptions, maladaptive coping and QoL. No significant changes were observed for pain catastrophising or psychological distress scores. Our findings did not support the apparent pandemic-related decrease in QoL suggested by cross-sectional pre-pandemic and pandemic Italian coeliac cohorts [11, 17] but were consistent with the increase in QoL observed in Brazilian coeliac J o u r n a l P r e -p r o o f cohorts [18, 19] . Having used a within-subjects design and a larger sample, the present study provided the most robust test of this effect to date. These findings may be due to pandemic-related lockdown measures that necessarily limit the number of meals consumed out of the home [60], likely facilitating greater personal control over the food preparation process, and limiting instances of coeliac-related social interference/exclusion (e.g., feeling left out of social invitations; see [61, 62] [23, 74] and our findings provide further support for this. Secondly, the CSM literature indicates that illness perceptions may impact well-being outcomes directly (e.g., exacerbating psychological symptoms, poorer subjective assessment of well-being/QoL) and indirectly, via coping and illness management behaviours [32] . Our findings did not support the indirect path, finding maladaptive coping to be predictive of only itself over time, and to have the least stability between timepoints of all variables. Given the strong relationship between maladaptive coping strategies and depressive J o u r n a l P r e -p r o o f Journal Pre-proof symptomatology [75] , it is possible that the effects of such strategies on outcomes were largely accounted for by psychological distress symptoms in the present study. Extending on current theory, the findings from this novel study provide crucial longitudinal evidence for the CSM. Overall, the findings reflected a dynamic interrelationship of model variables that was broadly consistent with the foundational principles of the CSM. Specifically, the time-lagged effects reflected a process model wherein illness perceptions play a potential mediating role in the relationship between GI symptoms and well-being outcomes (i.e., psychological distress, QoL). Moreover, QoL was reciprocally related to illness perceptions across time, consistent with the appraisal and feedback process inherent to the CSM [30] . This study had several strengths, including the large repeated-measures sample, use of a cross-lagged panel model methodology, and assessment of several well-being indices (i.e., GI symptoms, psychological distress, QoL), however, there are several limitations. Unlike other studies which compared pre/post COVID QoL in coeliac groups [11, 17, 18] , the present study used a within-subjects design, allowing for a more robust test of this effect. It is nonetheless difficult to make meaningful comparisons between studies given the substantial variation in COVID-19 case numbers, infection rates, duration and severity of restrictions/lockdowns, and socioeconomic conditions across both time and geographical regions during this period. The study sample was also comprised largely of participants that had previously self-selected to take part in coeliac studies, limiting the generalisability of our findings. Because there were only two timepoints, sequential mediation tests could not be conducted. Furthermore, as noted by Hagger et al. [33] , the cross-lagged approach used in the current study is not sensitive to rapid changes in model variables over time. Accordingly, methodologies such as ecological momentary assessment [76] , which rely on regular real-J o u r n a l P r e -p r o o f time assessment of experiences and behaviours in participants' natural environment, may be more instructive [77] . It is also important to note that this study examined illness perceptions and coping as singular, composite scales derived from factor analysis, rather than their constituent dimensions/coping strategies. While this validated these measures within this sample and is consistent with recommendations by the scale developers [41, 42] and past research [65-67, 71, 78] , future research may benefit from exploring these variable dimensions independently, preferably across three or more timepoints. This type of analysis would allow for tests of sequential mediation effects while identifying specific illness beliefs associated with illness management strategies and well-being outcomes. Finally, while this study examined illness perceptions as predicted by GI symptoms, it should be noted that those living with coeliac disease can be asymptomatic. These individuals are nonetheless vulnerable to potential longterm complications associated with ongoing gluten-exposure (for review, see [79] ) and it may therefore be instructive for future investigations to examine non-symptomatic predictors of threatening illness perceptions, such as concerns regarding hidden or inconspicuous gluten exposures. Whilst the literature has emphasised the value of medical and dietary follow-up of those living with coeliac disease [80], this work highlights a key role for the psychologist in facilitating post-diagnosis well-being. Consistent with past recommendations [20] , psychological assessment and follow-up care may be particularly beneficial in this cohort, especially for those experiencing psychological distress symptoms and/or reduced QoL. Importantly, this study demonstrates that illness perceptions were the best predictor of wellbeing outcomes (i.e., psychological distress and QoL) across time, and that illness perceptions are reciprocally related to QoL. These findings also underline the value of illness perceptions J o u r n a l P r e -p r o o f as a potential modifiable target for intervention. At the clinical level, reflecting past recommendations in CSM-based GI cohorts [64, 65, 68] , improvement in psychological distress and QoL are likely to occur through the active targeting of maladaptive illness beliefs (e.g., limited personal control). Psychological strategies such as Cognitive Behavioural Therapy or Acceptance and Commitment Therapy may be therefore helpful to promote personal control and facilitate non-avoidance-based coping strategies. Extending on the findings of Möller et al. [34] , the present study assessed pandemicrelated changes in several key indices of well-being in a large adult coeliac cohort and is the first to our knowledge to test the CSM using a cross-lagged panel model methodology. The findings suggest that the COVID-19 pandemic resulted in improvement across multiple psychosocial factors and well-being indices, including GI symptoms, negative illness perceptions, maladaptive coping, and QoL. Building on the large body of predominantly cross-sectional CSM studies [32, 33] , this study identified several time-lagged relationships supportive of the CSM as a theoretical foundation for illness adjustment in coeliac disease. Specifically, the time-lagged relationships were consistent with the purported mediating role of illness perceptions in the symptom-to-outcome relationship described in the CSM. J o u r n a l P r e -p r o o f J o u r n a l P r e -p r o o f Table 3 . Univariate comparisons showing changes in model variables from pre-pandemic to pandemic. Pre-pandemic J o u r n a l P r e -p r o o f The extent of covid-19 pandemic socio-economic impact on global poverty. a global integrative multidisciplinary review COVID-19 pandemic and mental health consequences: Systematic review of the current evidence Impact of COVID-19 pandemic on mental health in the general population: A systematic review Reported quality of life in countries with cases of COVID19: A systematic review Prevalence of comorbidities among individuals with COVID-19: A rapid review of current literature Comorbidity and its impact on patients with COVID-19 Challenges of non-COVID-19 patients with chronic illness during the pandemic Global prevalence of celiac disease: Systematic review and metaanalysis The Oslo definitions for coeliac disease and related terms Coeliac disease COVID-19 pandemic perception in adults with celiac disease: an impulse to implement the use of telemedicine COVID-19 in celiac disease: a multicentric retrospective cohort study Prevalence, incidence and clinical features of SARS-CoV-2 infection in adult coeliac patients The risk of contracting COVID-19 is not increased in patients with celiac disease Risk of severe covid-19 in patients with celiac disease: A population-based cohort study Risk perception and J o u r n a l P r e -p r o o f Journal Pre-proof knowledge of COVID-19 in patients with celiac disease Quality-of-Life Evaluation in Coeliac Patients on a Gluten-Free Diet Health-Related Quality of Life and Experiences of Brazilian Celiac Individuals over the Course of the Sars-Cov-2 Pandemic Quality of life of celiac patients in Brazil: Questionnaire translation, cultural adaptation and validation Systematic review: Exploration of the impact of psychosocial factors on quality of life in adults living with coeliac disease Quality of life in coeliac disease is determined by perceived degree of difficulty adhering to a gluten-free diet, not the level of dietary adherence ultimately achieved Psychosocial factors are more important than disease activity in determining gastrointestinal symptoms and health status in adults at a celiac disease referral center Psychosocial aspects of coeliac disease: a crosssectional survey of a UK population Living with coeliac disease: controlled study of the burden of illness The relationship between quality of life and coping strategies of adults with celiac disease adhering to a gluten-free diet Predictors of reduced health-related quality of life in adults with coeliac disease Maintenance of a gluten free diet in coeliac disease: The roles of self-regulation, habit, psychological resources, motivation, support, and goal priority Gluten free diet adherence in coeliac disease. The role of psychological symptoms in bridging the intention-behaviour gap Measuring beliefs about gluten free diet adherence in adult coeliac disease using the theory of planned behaviour The common-sense model of self-regulation (CSM): a dynamic framework for understanding illness self-management Illness cognition: using common sense to understand treatment adherence and affect cognition interactions The common sense model of selfregulation: Meta-analysis and test of a process model The common sense model of illness self-regulation: a conceptual review and proposed extended model Quality of life in coeliac disease: relationship between psychosocial processes and quality of life in a sample of 1,697 adults living with coeliac disease Bias in cross-sectional analyses of longitudinal mediation: Partial and complete mediation under an autoregressive model Bringing the social context into research using the common sense model A simple validated gluten-free diet adherence survey for adults with celiac disease A randomized, double-blind study of larazotide acetate to prevent the activation of celiac disease during gluten challenge Manual for the depression anxiety stress scales, Psychology Foundation of Australia1996 The brief illness perception questionnaire A systematic review and meta-analysis of the Brief Illness Perception Questionnaire You want to measure coping but your protocol's too long: Consider the brief cope The Pain Catastrophizing Scale: Development and Validation The EUROHIS-QOL 8-item index: psychometric results of a cross-cultural field study Australians and New Zealanders with coeliac disease Timeline of COVID-19 in Australia: the first year Department of Health and Human Services, Coronavirus update for Victoria -06 Australian Broadcasting Corporation, Victorian coronavirus restrictions tightening as cases continue to increase Single case of COVID-19 confirmed in New Zealand Covid-19: State of emergency declared in New Zealand Level 1 -Border testing and exemptions Measurement invariance conventions and reporting: The state of the art and future directions for psychological research Evaluating goodness-of-fit indexes for testing measurement invariance Process analysis: Estimating mediation in treatment evaluations Longitudinal structural equation modeling Effects of Web-Based Social Connectedness on Older Adults' Depressive Symptoms: A Two-Wave Cross-Lagged Panel Study Longitudinal associations between subjective and psychological wellbeing in Japan: A four-year cross-lagged panel study Evaluating the fit of structural equation models: Tests of significance and descriptive goodness-of-fit measures Effects of sample size, estimation methods, and model specification on structural equation modeling fit indexes Impact of coeliac disease on dietary habits and quality of life Controlled by food' -Lived experiences of coeliac disease Self-compassion directly and indirectly predicts dietary adherence and quality of life among adults with celiac disease Application of the common sense model in inflammatory bowel disease: A systematic review Relations between symptom severity, illness perceptions, visceral sensitivity, coping strategies and well-being in irritable bowel syndrome guided by the common sense model of illness Exploring symptom severity, illness perceptions, coping styles, and well-being in gastroparesis patients using the common sense model Preliminary examination of the relations between disease activity, illness perceptions, coping strategies, and psychological morbidity in Crohn's disease guided by the common sense model of illness Psychological mediators of psychological distress and quality of life in inflammatory bowel disease Extending the Common Sense Model to Explore the Impact of the Fear of COVID-19 on Quality of Life in an International Inflammatory Bowel Disease Cohort Extending the common sense model to explore the impact of visceral sensitivity on quality of life in inflammatory bowel disease Relationship between health status, illness perceptions, coping strategies and psychological morbidity: a preliminary study with IBD stoma patients Illness perceptions and stress: mediators between disease severity and psychological well-being and quality of life among patients with Crohn's disease Examining the mediational role of psychological flexibility, pain catastrophizing, and visceral sensitivity in the relationship between psychological distress, irritable bowel symptom frequency, and quality of life Wellbeing, illness perception and coping strategies in Italian Celiac patients More adaptive versus less maladaptive coping: What is more predictive of symptom severity? Development of a new scale to investigate coping profiles across different psychopathological syndromes Ecological momentary assessment Ecological momentary assessment in behavioral research: addressing technological and human participant challenges Exploring the relationships between illness perceptions, self-efficacy, coping strategies, psychological distress and quality of life in a cohort of adults with diabetes mellitus Complications of coeliac disease The authors thank Dr David Skvarc for his guidance and feedback on statistical analyses. We also thank Coeliac Australia and Coeliac New Zealand for their help in the recruitment process and their ongoing support of coeliac research. Finally, the authors wish to acknowledge and thank the participants of this study. This work was supported by the Australian Government Research Training Program Scholarship.