Mental health prior to and during the COVID‐19 pandemic in individuals with bipolar disorder: Insights from prospective longitudinal data

Abstract Objectives Many studies have examined the impact of COVID‐19 on the mental health of the public, but few have focused on individuals with existing severe mental illness with longitudinal data before and during the pandemic. Aims: To investigate the impact of the COVID‐19 pandemic on the mental health of people with bipolar disorder (BD). Methods In an ongoing study of people with BD who used an online mood monitoring tool, True Colours, 356 participants provided weekly data on their mental health. Symptoms of depression, mania, insomnia, and suicidal thoughts were compared in 2019 and 2020. From May 2020, participants also provided weekly data on the effect of the COVID‐19 pandemic on anxiety, coping strategies, access to care, and medications. Results On average, symptoms of depression, mania, insomnia, and suicidal thoughts did not significantly differ in 2020 compared to 2019, but there was evidence of heterogeneity. There were high rates of anxiety about the pandemic and its impact on coping strategies, which increased to over 70% of responders in January 2021. A significant proportion of participants reported difficulty accessing routine care (27%) and medications (21%). Conclusions Although mood symptoms did not significantly increase during the pandemic overall, we observed heterogeneity among our BD sample and other impacted areas. Individuals' unique histories and psychosocial circumstances are key and should be explored in future qualitative studies. The significant impacts of the pandemic may take time to manifest, particularly among those who are socioeconomically disadvantaged, highlighting the need for further long‐term prospective studies.


| INTRODUC TI ON
The impact of the COVID-19 pandemic on mental health has been highlighted as an area of concern, particularly for those with preexisting mental health conditions. 1 Studies in the general population find that symptoms of poor mental health have increased compared to pre-pandemic population norms 2,3 and this is more pronounced in people with pre-existing mental illness. [4][5][6][7] Potential explanations for this include reduced access to mental health services and negative effects of social isolation measures. 8,9 Those with pre-existing mental health conditions are, therefore, a key target for research on the effects of the COVID-19 pandemic on mental health. However, most studies on the effect of the COVID-19 pandemic on mental health treated all participants with mental health conditions as one group.
This ignores the heterogeneity in the aetiology and clinical management of different psychiatric disorders, particularly for people with severe mental illnesses.
There has been particular concern about people with bipolar disorder, as it has been proposed that this group could be particularly affected by the consequences of the COVID-19 pandemic due to reduced access to treatment, the impact of social isolation, and increased sensitivity to factors that affect sleep and circadian rhythms. 10 This is also a concern because this group is at high risk of suicide. 11 Existing studies have often grouped individuals with bipolar disorder with those who have other disorders, such as schizophrenia, to form a 'severe mental illness' group (e.g. 12 ). To our knowledge, only three studies, conducted in the United States (US) or Dutch populations, have examined mood symptoms in individuals with bipolar disorder longitudinally during the pandemic. 12,13 These studies found that mood symptoms remained stable or improved during the pandemic, but used limited, short-term comparative measures of pre-pandemic mental health, or examined changes in mental health at early stages in the pandemic. 12,13 A recent scoping review highlighted the lack of longitudinal studies of individuals with bipolar disorder during the pandemic, and called for prospective studies which compare data from before and during the COVID-19 pandemic. 14 Longitudinal research in non-US populations, with participants followed up frequently and for longer periods of time are also needed. Finally, studies in the general population have indicated increases in suicidal ideation 7 (although findings are mixed, 15 ) and insomnia 16,17 during the pandemic, but few studies have specifically examined these symptoms in people with bipolar disorder.

| Aims
Within an ongoing longitudinal cohort of UK individuals with wellcharacterised bipolar disorder, we aimed to test whether rates of mood symptoms, suicidal thoughts and insomnia were worse in 2020 (during the COVID-19 pandemic) compared to 2019 (before the COVID-19 pandemic). Second, we aimed to examine the wider ongoing impact of the COVID-19 pandemic on people with bipolar disorder, including the impact on coping strategies, anxiety about the pandemic, access to mental health care and access to medication.

| Mental health
From January 2015, all participants enrolled in the BDRN were invited to use an online mood monitoring system called True Colours.
Details of the recruitment and implementation of True Colours in the BDRN cohort are provided elsewhere. 19 Once enrolled in True Colours, participants receive weekly email prompts to complete K E Y W O R D S bipolar disorder, COVID-19, depression, longitudinal, mania two online self-report questionnaires: the Altman Self-Rating Mania Scale (ASRM) 20 and the Quick Inventory of Depressive symptomatology (QIDS). 21 The ASRM measures the presence and severity of DSM-IV hypomanic/manic symptoms over the preceding week, and consists of five items with total scores ranging from 0 to 20. It has shown good psychometric properties for detecting episodes of high mood in individuals with bipolar disorder. 22 The QIDS measures the presence and severity of DSM-IV depressive symptoms over the preceding week and comprises 16 items which are combined to give a total score ranging from 0 to 27. The QIDS has been validated in patients with bipolar disorder and major depressive disorder. 23 We dichotomised the ASRM and QIDS based on the scores identified in prior studies that suggest significant symptomatology, specifically a threshold of 5 for the ASRM 20 and 10 for the QIDS. 21 respectively. Analyses comparing individuals who responded at least once on True Colours during December 2018 who were included in the final analyses to those who were not (as they had stopped using True Colours during the pandemic) indicated there were no significant differences between these individuals on demographics (gender, current age, educational attainment) or lifetime measures of severity of illness course (bipolar disorder type, age of bipolar disorder illness onset, history of psychiatric admission, history of psychotic symptoms and rapid cycling illness course). For Aim 2, we analysed data on COVID-19-related questions from May 2020 to January 2021, within the sample of 356 people who had been selected for Aim 1, resulting in a sample of 138 participants.

| Analysis
All analyses were conducted using R version 3. 2020. In contrast to our hypothesis, we did not find an overall pattern of results suggesting that rates of depression, mania, insomnia or suicidal ideation significantly increased in 2020 compared to 2019.

| ASRM
After correcting for multiple testing, we found a significant difference in the proportion of participants who exceeded the cut-off for the ASRM in only one of the 10 months tested: July 2019 compared to July 2020 ( 2 (1)=9.59, p = 0.002,      We also observed no differences in the rates of insomnia or suicidal thoughts across these periods. There was, however, evidence of heterogeneity between participants, with 39% experiencing an increase in average ASRM scores, and 53% experiencing an increase in average QIDS scores in 2020 compared to 2019.

| Aim 2 -Questions about the COVID-19 Pandemic
These results did not support our original hypothesis that peo-  We did, however, find evidence of the pandemic having a significant impact on some of those with lived experience of bipolar disorder. We found that 39% of participants experienced an increase in average mania score and 53% experienced an increase in average depression score in 2020 compared to 2019. A large proportion of participants also reported anxiety about the pandemic and that their usual coping strategies were impacted. These rates appeared to be rising over the pandemic which could point to a longer-term impact on mental health. In addition, up to 27% of participants reported difficulty accessing routine care from mental health services and up to 21% reported difficulty accessing psychiatric medication over the course of the pandemic. Finally, it is possible that the impact of the pandemic may take some time to manifest and people with bipolar could be at risk of poor long-term outcomes in the years to come.
For this reason, further research examining the long-term impact of the COVID-19 pandemic on people with bipolar disorder will be important to conduct.

| Strengths and limitations
The strengths of our study are that we were able to utilise longitudinal data before and during the COVID-19 pandemic from a large sample of people with bipolar disorder from throughout the United Kingdom. These data were also collected at frequent intervals and we were able to examine data on COVID-19-related issues such as access to medication. However, our results should be interpreted in light of the following limitations. First, symptom monitoring was limited to mania and depression. Anxiety is highly comorbid with bipolar disorder 31 and therefore it is possible that symptoms of anxiety showed a different pattern of results compared to what we have observed here. Of note is that a recent study of mental health in service users in the United Kingdom observed increases in generalised anxiety at various points in the pandemic, whereas depression symptoms remained at the same or lower levels compared to previous years. 29 However, despite a lack of pre-pandemic anxiety data, due to its potential importance we added a measure of anxiety to the COVID-

| Clinical implications
A key implication of our results is that we should not assume that the COVID-19 pandemic has had a negative impact in all people with bipolar disorder in all symptom domains. Prospective weekly monitoring of high and low mood, insomnia, and suicidal ideation did not show an overall deterioration over the pandemic to date. It is likely, however, that there are individual patients (e.g. those at socioeconomic disadvantage) for whom the pandemic has had a major impact on their mental health. Responding to the individual patient with their unique history and psychosocial circumstances is key.
Our results do show, however, that anxiety about the pandemic is high and rising and that many participants report difficulty accessing medication and usual routine care. Mental health and primary care services must consider how best to deliver monitoring and care to patients with bipolar disorder, learning the lessons thus far for further waves or subsequent pandemics. It is also important to recognise that the most significant impacts from the pandemic regarding the mental health and wellbeing of this population may take some time to manifest. Reviewing our ongoing data collection again at later dates will be key. Finally, it is possible that individuals who are less likely to engage with remote symptom monitoring may be at higher risk of experiencing poor mental health during the pandemic.
Concerted efforts to engage with these individuals for the purposes of clinical management will be vital.

| Summary and conclusions
In a large UK study of prospective, frequent, longitudinal data on mental health in people with bipolar disorder, we did not find evidence of significantly increased mood symptoms, insomnia or suicidal thoughts during the COVID-19 pandemic compared to the same time points in the same individuals in the previous year.
However, there was evidence of heterogeneity of responses, including subgroups whose symptoms worsened, and data from May 2020 onwards indicated that the COVID-19 pandemic had impacted other areas of functioning, including levels of anxiety, coping strategies and access to mental health care and treatment. Research Network for kindly giving their time to take part in this research.

CO N FLI C T O F I NTE R E S T
None.