Abstract
Aims This study had two primary aims: (1) to assess change in alcohol use and hazardous drinking since the start of the COVID–19 pandemic in the UK; (2) to investigate the extent to which poor inhibitory control and stress were associated with a change in alcohol use and hazardous drinking since the pandemic began.
Methods We interrogated cross-sectional data from the first sweep of the COVID-19 survey, comprising four birth cohorts (13, 453 respondents, aged 19-62 years). Alcohol use, stress, and inhibitory control were self–reported. Change in drinking and hazardous alcohol use were regressed on stress and inhibitory control, adjusting for sex, ethnicity and socioeconomic status.
Results Between fourteen to thirty percent of individuals reported an increase in alcohol use since the start of the pandemic, depending on age and sex. Of these, a substantial proportion (approximately four to ten percent) reported drinking hazardously, thus were at high risk of alcohol–related harm. The effects of stress and inhibitory control were age dependent. In thirty–year–olds, those reporting feeling stressed since the start of the pandemic were around four times more likely to engage in hazardous drinking (OR = 3.92, 95% CI = 1.17–13.15). Among nineteen–year–olds, a one unit increase in impatience was associated with a fourteen percent (ORadjusted = 1.14, 95% CI = 1.05–1.23) increase in the likelihood of drinking more during the lockdown, and a twenty–one percent (ORadjusted= 1.21, 95% CI = 1.05–1.41) increase in the likelihood of drinking hazardously. For those aged thirty or fifty, a one-unit increase in risk-taking was associated with an eighteen (ORadjusted = 1.18, 95% CI = 1.05–1.33) or six percent (ORadjusted = 1.06, 95% CI = 1.01, 1.12) increase in the likelihood of drinking hazardously, respectively. The findings related to the interaction between stress and inhibitory control were relatively complex. For example, thirty–year–olds that rated themselves as impatient and reported less stress tended to drink more (ORadjusted = 1.22, 95% CI = 1.00–1.48).
Conclusions Data from four nationally representative birth cohorts suggests that alcohol use during the pandemic increased in up to thirty percent of individuals. Hazardous drinking was observed in up to ten percent of respondents. Stress and inhibitory control were related to this but effects were age dependent. Governments should carefully consider the impact of personality and stress that may affect alcohol consumption in at-risk individuals.
Introduction
Since being first identified in Wuhan, China, in December 2019, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the virus that causes COVID-19, has caused a significant threat to global health (Sohrabi et al., 2020). Governments around the world responded by imposing ‘lockdowns’ (orders to remain at home, and socially isolate) on their populations, and available evidence supports this action as a means of mitigating the rate of spread of the virus (Anderson et al., 2020). However, the indirect impact of lockdown on public health has raised concern, particularly relating to mental health and wellbeing (Pfefferbaum & North, 2020).
Concerns that the lockdowns may increase alcohol misuse have been raised, particularly concerning people at high-risk of developing, or re-establishing, hazardous alcohol use (Clay & Parker, 2020). An example of individuals who are at high risk of alcohol misuse are people that display poor inhibitory control (see Lee et al., 2019 for review). Inhibitory control is generally conceptualised as one of the core executive functions (Diamond, 2013). It is a complex and multifaceted construct made up of several subcomponents: response inhibition (i.e., action inhibition, action cancellation), sensitivity to delay (i.e., delay discounting, patience), sensitivity to risk/reward (risk-taking, sensation seeking), and attention (i.e., capacity to focus and avoid interference) (Strickland and Johnson, 2020). Indeed, overwhelming evidence from pre-clinical translational work (e.g., Kreek et al., 2005), neuroimaging research (e.g., Voon et al., 2020), and heritability studies (e.g., Khemiri et al., 2016), suggests that poor inhibitory control is both a risk factor for the development, and consequence, of substance misuse and addiction.
The association between stress and alcohol use is also well–established (Jose et al., 2000; Ruisoto & Contador, 2019). Similar to inhibitory control, stress plays a critical role in both the onset and maintenance of alcohol misuse and addiction (see Becker, 2017 for review). On the one hand, the acute anxiolytic properties of alcohol motivate some individuals to drink (Kwako and Koob, 2017). On the other, perhaps counterintuitively, alcohol acts as a stressor, whereby acute exposure stimulates the hypothalamic-pituitary-adrenocortical (HPA) axis through direct activation of the paraventricular nucleus (Armario, 2010). Further, exposure to either chronic stress or chronic alcohol misuse both lead to blunted stress responses, including dysregulation of the HPA axis – a known risk factor for hazardous drinking and addiction (see Milivojevic & Sinha, 2018 for review).
We have recently found evidence that suggests a complex interplay between inhibitory control, stress, and alcohol use, where an experimentally-induced acute psychosocial stressor increased craving for alcohol, and subsequent alcohol consumption, in healthy (non-addicted) individuals (Clay et al., 2018; Clay and Parker, 2018). The strength of these stress–induced increases in alcohol craving and consumption were predicated on individual differences in risk-taking personality traits, stress-reactivity, and stress-recovery. Collectively, this suggests these underlying (e.g., poor inhibitory control), and environmental (e.g., stress) factors may combine to make particular individuals more at risk.
Here, we analysed the first sweep of the COVID-19 survey (University of London, Institute of Education, Centre for Longitudinal Studies, 2020) – which was answered by individuals from five nationally representative cohorts who have been providing data since childhood – to investigate: (1) alcohol use during the pandemic in the UK; and (2) the extent to which poor inhibitory control and/or stress were associated with any change in alcohol use or hazardous drinking.
Methods
The Centre for Longitudinal Studies COVID–19 Survey
The survey design, recruitment procedure, and fieldwork processes have been described in detail elsewhere (Brown et al., 2020). Briefly, the CLS COVID-19 survey was administered to five nationally representative cohorts, each from a different generation, all born in the UK, who have been completing surveys about their lives and development since childhood. These included: (1) the Millennium Cohort Study (MCS), who are part of ‘Generation Z’, and were aged 19; (2) Next Steps, who are part of the ‘Millennial’ generation, who were aged 30; (3) the 1970 British Cohort Study (BCS70), who belong to ‘Generation X’ – aged 50; (4) the National Child Development Study (NCDS), who were aged 62 and were born in the latter part of the ‘Baby Boomer’ generation; and (5) the National Study of Health and Development (NSHD), who were born at the beginning of the ‘Baby Boomer’ era, and were aged 74. The survey was issued to 50,479 individuals across the five cohorts (including parents of the children recruited into the MCS) between 2 and 31 May 2020. Overall, 18,042 of those invited responded, achieving a response rate of 35.7%.
Study sample
Due to data availability at the time of analysis, four of the five cohorts included in the COVID-19 survey were analysed. Namely: the MCS cohort members (n = 2,645), Next Steps (n = 1,907), the BCS70 (n = 4,223), and the NCDS (n = 5,178). The study was restricted to UK-based respondents, thus non-UK residents (n = 500) were excluded prior to analysis. This left 13,453 cases for analysis. Selected sample characteristics are shown in Table 1.
Outcome measures
Alcohol use behaviour was measured using five questions taken from the Alcohol Use Disorders Identification Test (AUDIT; Babor et al., 2001, 1992) The questions administered during the survey were: (1) “How often have you had a drink containing alcohol?”; (2) “How many standard alcoholic drinks have you had on a typical day when you were drinking?”; (3) “How often have you found you were not able to stop drinking once you had started?”; (4) How often have you failed to do what was expected of you because of drinking?; and (5) “Has a relative, friend, doctor, or health worker been concerned about your drinking or advised you to cut down?”. Questions one and two were repeated, prefaced by either “in the month before the Coronavirus outbreak”, or “since the start of the Coronavirus outbreak”. This provided an assessment of alcohol use prior to, and during, the pandemic. Questions one to five were posed in the context of the pandemic, thus were worded using the latter phrasing, offering an assessment of alcohol misuse during the outbreak – see Supplementary Materials for further details about outcome measures (e.g., psychometric properties and scoring).
Potential risk factors
Stress was assessed using a single question: “Since the Coronavirus outbreak, please indicate how the following have changed… The amount of stress I’ve been feeling”. The possible responses included “More than before”, “Same – no change”, and “Less than before”. Two measures of inhibitory control were used in this survey: patience and risk-taking. Each was measured using a single ten-point Likert scale item. The risk-taking question said “On a scale from 0 – 10, where 0 is ‘never’ and 10 is ‘always’, how willing to take risks would say you are?”. The patience item was phrased “On a scale from 0 – 10, where 0 is ‘never’ and 10 is ‘always’, how patient would you say you are?” See Supplementary Materials for further details about potential risk factors (e.g., psychometric properties).
Potential confounders
A selection of covariates, guided by the literature on associations with hazardous alcohol use, were included in models, including the respondent’s sex, ethnicity, National Statistics Socio-economic Class (NS-SEC) prior to the outbreak of Coronavirus, and economic activity during the pandemic. See Supplementary Materials for further details about potential confounders.
Statistical analysis
Stata IC (version 16.1) and R (version 3.6.2) were used for all analyses. The ‘patience’ item was reverse scored such that higher scores reflected greater impatience. Change in alcohol use and the hazardous drinking score were calculated using the method described in the Supplementary Material.
Inverse probability weighting was used to account for bias introduced due to missing data, and to ensure the results were as representative as possible. The overall percentage of missing data was 23.43%. The median percentage of missing data by variable was 5.29% (IQR = 8.01%). See Supplementary Table 3 for further details on missing data.
The association between inhibitory control, stress, and alcohol use was investigated using ordinal logistic regression. We first regressed change in alcohol use and alcohol misuse during the pandemic on inhibitory control, stress, and the interaction between inhibitory control and stress. We then adjusted the model estimates by including our list of covariates in our models. We noticed that the standard error among fifty-year-olds that reported being in education during the pandemic was inflated, leading to implausible results. This was due to only two fifty-year-olds females falling into this category. These cases were omitted for all regression-based analyses. Excluding this data did not change model results. For brevity, model estimates for the covariates are not included in text. However, full model output is presented in Supplementary Tables 4 – 11.
Results
Change in alcohol consumption
Across all cohorts, the majority of respondents reported drinking the same amount or less alcohol since the start of the pandemic (Figure 1A, Table 1). Approximately, one third of thirty–year–olds, one quarter of fifty–year–old males, one third of fifty-year-old females, one fifth of sixty–two–year– olds, and one tenth of nineteen–year–olds reported drinking more.
Hazardous drinking during the pandemic
The majority of participants fell into the low risk category, regardless of age or sex (Figure 1B, Table 1). Approximately one third of all participants were in the increasing risk category, except those aged nineteen, and thirty–year–old females. For those aged thirty or older, around one tenth of males and five percent of females were among the high or highest risk groups. Overall, fifty–year–olds were at the greatest risk of alcohol related harm with approximately seven percent of males and three percent of females falling into the highest risk group.
Change in stress
Females were disproportionally affected by stress (Figure 1C), with between forty and fifty percent of all female respondents reporting feeling more stressed since the start of the pandemic. In comparison, between one fifth and one third of males reported feeling more stressed.
Across all age groups, those aged thirty had the largest proportion of individuals that reported feeling more stressed since the start of the pandemic (Table 1). Those aged sixty–two were least affected by pandemic–related changes in stress with over two–thirds of respondents reporting feeling either no change or less stress.
Associations between stress, inhibitory control and drinking
Stress
Among thirty-year-olds, those that reported feeling more stressed since the start of the pandemic were 3.51 (95% CI = 1.07–11.56) times more likely to be at increasing, high, or highest (versus low) risk of alcohol-related harm, compared to those that reported feeling no change in stress. This effect remained after adjusting for potential confounding factors (ORadjusted = 3.92, 95% CI = 1.17–13.15).
Impatience
For nineteen-year-old participants, greater impatience was associated with increased alcohol consumption during the pandemic. For a one unit increase in impatience, the odds of reporting increased alcohol use were 1.09 (95% CI = 1.03–1.14) times greater than reporting no change in, or less, alcohol use. When adjusted for the effects of potential confounding factors, the effect remained (ORadjusted = 1.14, 95% CI = 1.05–1.23). Nineteen-year-olds that described themselves as more impatient were also more likely to fall into higher categories of risk of alcohol-related harm based on their alcohol misuse scores (OR = 1.14, 95% CI = 1.04–1.25). This effect remained after controlling for potential confounders (ORadjusted = 1.21, 95% CI = 1.05–1.41).
Risk–taking
Among those ages thirty and fifty, there was a positive association between increased risk-taking and alcohol misuse. For a one unit increase in risk-taking in those aged thirty, the odds of being categorised as having increasing, high, or highest risk of alcohol-related harm was 1.19 (95% CI = 1.07–1.34) times greater than falling into the low risk category. This effect remained after adjusting for potential confounds (ORadjusted = 1.18, 95% CI = 1.05–1.33). Among fifty-year-olds, the odds of falling into the higher categories of risk of alcohol related harm were 1.09 (95% CI = 1.03–1.16) times greater than falling into the lower categories. This effect remained after controlling for potential confounds (ORadjusted = 1.06, 95% CI = 1.01–1.12).
Stress x personality interactions
Interactions between stress and inhibitory control were age-specific. For nineteen-year-olds that reported feeling less stressed during the pandemic (compared to the same level of stress), a one unit increase in risk-taking was associated with 1.30 (95 CI = 1.02–1.67) times greater odds of also falling among the higher categories of the alcohol misuse scale, indicating a greater risk of alcohol-related harm. However, when this estimate was adjusted by adding potential confounders into the model, the effect was no longer significant (ORadjusted = 1.32, 95% CI = 0.88–1.99). Conversely, a one unit increase in risk-taking among fifty-year-olds who reported feeling more stressed since the outbreak was associated with thirteen percent greater odds (OR = 1.13, 95% CI = 1.02–1.24) of also reporting an increase in alcohol use compared to those that reported no change in stress. However, this effect was not present after controlling for confounders (ORadjusted = 1.06, 95% CI = 0.98–1.15).
There were several interactions between self-reported pandemic-related stress and self-reported impatience, but only in the thirty- and fifty-year-old participants. Among thirty-year-olds that reported feeling less stressed, a one unit increase in impatience was associated with 21% (OR = 1.21, 95% CI = 1.01–1.46) greater odds of also reporting an increase in drinking compared to those that reported no change in stress. This effect remained after controlling for potential confounding variables (ORadjusted = 1.22, 95% CI = 1.00–1.48). For thirty-year-olds that reported feeling more stressed, a one unit increase in impatience was associated with a 12% (OR = 0.88, 95% CI = 0.80–0.98) decrease in the odds of reporting an increase in alcohol use compared to those who reported no change in stress. This effect remained after controlling for potential confounds (ORadjusted = 0.88, 95% CI = 0.80–0.98).
This pattern was similar in terms of alcohol misuse. For thirty-year-olds that reported experiencing less stress, a one unit increase in impatience was associated with a 30% (OR = 1.30, 95% CI = 1.07– 1.59) increase in the odds of falling into a higher risk category of alcohol-related harm. This effect remained after controlling for potential confounds (ORadjusted = 1.31, 95% CI = 1.09–1.57). Finally, among the fifty-year-olds, a one unit increase in impatience alongside reporting less stress was associated with a 17% increase in the odds of placing among a higher risk category of alcohol-related harm, but only after adjusting for the effects of potential confounding factors (ORadjusted = 1.17, 95% CI = 1.04–1.31).
Discussion
This was the first study to explore the extent to which stress and inhibitory control impacted on alcohol use and misuse during the period of social isolation (lockdown) related to the COVID-19 pandemic. Across all age–groups in the UK, from four national birth cohorts. we found that the majority of respondents reported drinking the same amount or less since the start of the pandemic. Nevertheless, approximately one third of thirty–year–olds, one quarter of fifty–year–olds, one fifth of sixty–two–year olds, and one seventh of nineteen–year–olds reported drinking more. These figures corroborate emerging evidence which suggests that between one fifth and one third of individuals in the UK report drinking more during the pandemic (Institute of Alcohol Studies, 2020). In terms of hazardous drinking, the majority fell into the low risk group, regardless of age or sex. However, approximately ten percent of males and five percent of females aged thirty or older were at either high or highest risk of alcohol–related harm. In addition, the relationship between stress, inhibitory control, and pandemic–related alcohol use behaviour was assessed. In the thirty–year–old group, stress was a significant risk factor, where those who reported feeling more stressed since the start of the pandemic were approximately four times more likely to be at risk of alcohol related harm after controlling for covariates. Further, thirty– and fifty–year–olds who rated themselves as ‘risk-takers’ were more likely to engage in hazardous drinking behaviour, and nineteen-year-olds who rated themselves as impatient were at increased odds of drinking more since the pandemic and displaying more hazardous drinking behaviour. Among sixty–two–year–olds, neither stress nor inhibitory control were related to alcohol use behaviour during the pandemic.
Thirty-year-olds had the highest proportion of individuals that reported increased stress since the start of the pandemic. This group also had the highest proportions of individuals that reported increased alcohol use, and were particularly sensitive to the effects of stress during the outbreak on their hazardous drinking. Analogous to this finding, previous research has suggested that the Millennial generation struggle with stress management considerably more than previous generations (Bland et al., 2012). Similarly, a recent data from the UK Household Longitudinal Survey (Etheridge & Spantig, 2020) suggests that young individuals have seen larger declines in well–being than the old during the first lockdown. Surprisingly, despite the well-established link between substance use and stress (Jose et al., 2000; Ruisoto & Contador, 2019), a main effect of stress was not observed in any other group. Instead, self-report personality factors were more important in predicting increased alcohol use and hazardous drinking, and in some cases, a complex interaction between stress and personality.
In thirty– and fifty-year–olds, for example, risk-taking personality was associated with an increased propensity to consume more alcohol and to have higher hazardous drinking scores. This corresponds to a large volume of literature which associates poor inhibitory control with alcohol misuse (see for example Kreek et al., 2005; Khemiri et al., 2016; Lee et al., 2019; Voon et al., 2020). The majority of nineteen–year–olds reported drinking less since the start of the pandemic. This was unsurprising considering the recent evidence of the ‘devaluation of alcohol’ among Generation Z (Kraus et al., 2020). This finding may also have been driven by the closure of on–trade drinking locations since drinking at venues such as pubs and bars is more common among young people (Ally et al., 2016), and reduced exposure to environments associated with alcohol consumption has been associated with a reduction in drinking among young individuals during the pandemic (Winstock et al., 2020). However, critically, nineteen–year–old participants that rated themselves as impatient drank more during the pandemic and showed higher hazardous drinking scores. This group also had the highest levels of impatience across all cohorts. Taken together, these findings raise a concern about the potential for adults who have poor inhibitory control to be at particular risk of an escalation of alcohol misuse following the pandemic situation. Public health officials and health workers should be particularly vigilant over the coming months (and following any subsequent national or local lockdowns) for increased alcohol use – especially among individuals with poor impulse control. It may be prudent to include measures of inhibitory control during routine assessments of alcohol use in in primary care settings, during hospital admissions, or through third sector interventions (e.g., online self-assessment tools; White et al., 2010).
It is clear from previous research that there is an interaction between stress and personality factors that influence drinking behaviour. For example, people who experience acute stress show increases in craving for, and consumption of, alcohol (Clay et al., 2018; Clay & Parker 2018). Here, counter– intuitively, we found that thirty–year–old participants who rated themselves as impatient, and who experienced less stress during the pandemic, drank more alcohol. Similarly, those that rated themselves impatient and experienced high levels of stress also reported lower alcohol consumption during the pandemic. In addition, both thirty– and fifty–year old respondents who reported experiencing less stress during the pandemic, had higher hazardous drinking scores. As ‘drinking to cope’ was a prominent feature related to alcohol use during lockdown in the USA (Rodriguez et al., 2020), it may also be the case here. For instance, individuals with poor inhibitory control tend to use alcohol as a method of dealing with stress (Hamilton et al., 2013; Fede et al., 2020). Therefore, these individuals may have reduced stress levels due to their reported increased alcohol use. Alternatively, as the physiological response to long–term (chronic) and short–term (acute) stress differs (see Stephens & Wand, 2012 for review), for instance the acute stress response is characterised by a significant increase in catecholamines, it may be that the interaction between inhibitory control and chronic stress also differs. Therefore, future research should endeavour to investigate the impact of the interaction between different types of stress and inhibitory control in the context of alcohol use.
Limitations
We acknowledge several limitations in our study that must be considered. First, the survey was designed to capture information across several domains other than those relevant here (see Brown et al., 2020 for details). Therefore, to mitigate known issues related to respondent burden (e.g., satisficing), brevity was prioritised which meant less detail in some of the measures used. For instance, single-item measures were used to assess risk-taking, impatience, and stress which may not capture the full scope of these construct (i.e., these measures may suffer from reduced content validity). This increases the uncertainty surrounds estimates calculated using these measures. Therefore, the use of single–item measures may also inflate standard errors and risk for type II error. Some of this potential error is offset by our large sample size; however, we found some effects that were not statistically significant despite moderate effect sizes (e.g., among thirty-year-olds that reported increased stress, ORadjusted = 2·18, 95% CI = 0·97 to 4·89). Second, there may be individual differences in the way each question was interpreted. For instance, feelings of stress are subjective and vary between–individuals (Sommerfeldt et al., 2019). Therefore, while some may find the pandemic and related period of social isolation as extremely stressful, others will find lockdown less stressful than pre–pandemic life. This may offer another explanation for why some that reported poor inhibitory control and lower levels of stress also reported increased alcohol use. Third, there is no way to independently verify self-report drinking, and it is well-known that people under-estimate the alcohol consumption when asked on questionnaires due to social desirability bias, and often a lack of detailed memory of drinking episodes (e.g., Northcote & Livingston, 2011). It may, therefore, be that our data under-represents the true extent of drinking during the pandemic. Finally, the longitudinal nature of birth cohort data allows for bias introduced by missing data to be minimised using sample weights calculated by the CLS team (Brown et al., 2020). However, there is a possibility that unobserved predictors of missing data may still influence results.
Conclusions
This study demonstrates that in the UK, the first wave of the COVID–19 pandemic was associated with an increase in alcohol use in between fourteen and thirty percent of individuals, depending on age and sex. Of these, a substantial proportion (between approximately four and ten percent) reported drinking hazardously, thus were at high risk of alcohol–related harm. Again, these results varied based on age and sex. Increased stress was a significant risk factor among thirty–year–olds, with those reporting feeling more stress since the start of the pandemic being around four times more likely to engage in hazardous drinking. Among nineteen–year–olds, a one unit increase in impatience was associated with a fourteen percent increase in the likelihood of increasing alcohol use during the lockdown, and a twenty–one percent increase in the likelihood of drinking hazardously after controlling for age, sex, ethnicity and socioeconomic status. Similarly, for those aged thirty or fifty, after controlling for covariates, a one-unit increase in risk-taking was associated with an eighteen or six percent increase in the likelihood of drinking hazardously, respectively. The findings related to the the interaction between stress and inhibitory control were relatively complex. Here, for example, thirty–year–olds that rated themselves as impatient and reported less stress tended to drink more. Overall, these findings compliment early emerging evidence that reports on changes in alcohol use during the pandemic (Institute of Alcohol Studies, 2020), and other studies that demonstrate that patients with existing alcohol use disorder were at risk of relapse during the pandemic (Kim et al., 2020). It is critical that, in future lockdowns, governments carefully consider the impact of change of lifestyle and stress that might impact on alcohol consumption in at-risk individuals. The UK government may, for example, consider age-specific strategies, (Tanner-Smith and Lipsey, 2015) or limiting alcohol sales, the latter of which has been the case in other countries. (Matzopoulos et al., 2020).
Data Availability
All data used in this study are available from the UK Data Service Website (https://ukdataservice.ac.uk/) under the "Safeguarded" data access policy. The primary dataset used is referenced under study number 8658. Ethnicity data associated with the MCS, Next Steps, BCS70, and NCDS cohorts are referenced under study numbers 6073, 5545, 5558, and 5565 respectively. All analysis code is available on the Open Science Framework website (https://osf.io/wf2rj/).
Financial support
JMC is funded by an ESRC Doctoral Training Partnership grant (ES/P000673/1). MOP receives funding from the Foundation for Liver Research. The data collection was carried out by the Centre for Longitudinal Studies at University College London, and was funded by the ESRC. The funders had no role in study design, data analysis, data interpretation, writing of the report, or the decision to submit for publication.
Conflicts of interest
Conflicts of Interest: None
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of1975, as revised in 200.
Availability of data and materials
All data used in this study are available from the UK Data Service Website (https://ukdataservice.ac.uk/) under the “Safeguarded” data access policy. The primary dataset used is referenced under study number 8658. Ethnicity data associated with the MCS, Next Steps, BCS70, and NCDS cohorts are referenced under study numbers 6073, 5545, 5558, and 5565 respectively. All analysis code is available on the Open Science Framework website (https://osf.io/wf2rj/).
Acknowledgements
We would like to thank the Centre for Longitudinal Studies team at University College London for collecting the data and for making the data available through the UK Data Service. We would also like to thank all of the participants that completed the survey for their cooperation.
Footnotes
Updated manuscript and supplementary materials.
References
