Background: There is growing concern about possible cognitive consequences of COVID-19, with reports of 'Long COVID' symptoms persisting into the chronic phase and case studies revealing neurological problems in severely affected patients. However, there is little information regarding the nature and broader prevalence of cognitive problems post-infection or across the full spread of disease severity. Methods: We sought to confirm whether there was an association between cross-sectional cognitive performance data from 81,337 participants who between January and December 2020 undertook a clinically validated web-optimized assessment as part of the Great British Intelligence Test, and questionnaire items capturing self-report of suspected and confirmed COVID-19 infection and respiratory symptoms. Findings: People who had recovered from COVID-19, including those no longer reporting symptoms, exhibited significant cognitive deficits versus controls when controlling for age, gender, education level, income, racial-ethnic group, pre-existing medical disorders, tiredness, depression and anxiety. The deficits were of substantial effect size for people who had been hospitalised (N = 192), but also for non-hospitalised cases who had biological confirmation of COVID-19 infection (N = 326). Analysing markers of premorbid intelligence did not support these differences being present prior to infection. Finer grained analysis of performance across sub-tests supported the hypothesis that COVID-19 has a multi-domain impact on human cognition. Interpretation: Interpretation. These results accord with reports of 'Long Covid' cognitive symptoms that persist into the early-chronic phase. They should act as a clarion call for further research with longitudinal and neuroimaging cohorts to plot recovery trajectories and identify the biological basis of cognitive deficits in SARS-COV-2 survivors.

Case studies have revealed neurological problems in severely affected COVID-19 patients. However, there is little information regarding the nature and broader prevalence of cognitive problems post-infection or across the full spread of severity. We analysed cognitive test data from 84,285 Great British Intelligence Test participants who completed a questionnaire regarding suspected and biologically confirmed COVID-19 infection. People who had recovered, including those no longer reporting symptoms, exhibited significant cognitive deficits when controlling for age, gender, education level, income, racial-ethnic group and pre-existing medical disorders. They were of substantial effect size for people who had been hospitalised, but also for mild but biologically confirmed cases who reported no breathing difficulty. Finer grained analyses of performance support the hypothesis that COVID-19 has a multi-system impact on human cognition.

Ninety-five per cent of mapped CRTs (n = 233) completed the survey. Few CRTs adhered fully to national policy guidelines. CRT implementation and local acute care system contexts varied substantially. Access to CRTs for working-age adults appears to have improved, compared with a similar survey in 2012, despite no evidence of higher staffing levels. Specialist CRTs for children and for older adults with dementia have been implemented in some areas but are uncommon.Clinical implicationsA national mandate and policy guidelines have been insufficient to implement CRTs fully as planned. Programmes to support adherence to the CRT model and CRT service improvement are required. Clearer policy guidance is needed on requirements for crisis care for young people and older adults.Declaration of interestNone.

1The sensitisation model suggests paranoia is explained by over-sensitivity to 2 perceived threat in social environments. However, this has been difficult to test 3 experimentally. We report two pre-registered studies that tested i) the sensitisation 4 model as an explanation of paranoia, and; ii) the role of purported maintaining 5 factors in supporting social sensitisation. In study one, we recruited a large general 6 population sample (N=987) who serially interacted with other participants in multi-7round Dictator games, matched to fair, partially fair, or unfair partners. Participants 8 rated attributions of harmful intent and self-interest after each interaction. In study 9 two (N=1011), a new sample of participants completed the same procedure and 10 additionally completed measures of anxiety, worry and interpersonal sensitivity. As 11 predicted, paranoid ideation predicted higher and faster overall harmful intent 12 attributions, whereas attributions of self-interest were unaffected, supporting the 13 sensitisation model. Contrary to predictions, neither worry nor anxiety predicted 14 harmful intent attributions while interpersonal sensitivity predicted decreased harmful 15 intent attributions. In a third exploratory study we combined data sets to examine the 16 effect of paranoia on trial by trial attributional changes when playing fair and unfair 17dictators. Paranoia predicted a greater reduction in harmful intent attributions when 18 playing a fair but not unfair dictator, suggesting paranoia may also exaggerate the 19 volatility of beliefs about the harmful intent of others. 20 21

The sensitization model suggests that paranoia is explained by over-sensitivity to social threat. However, this has been difficult to test experimentally. We report two preregistered social interaction studies that tested (i) whether paranoia predicted overall attribution and peak attribution of harmful intent and (ii) whether anxiety, interpersonal sensitivity and worry predicted the attribution of harmful intent. In Study 1, we recruited a large general population sample (N = 987) who serially interacted with other participants in multi-round dictator games and matched to fair, partially fair or unfair partners. Participants rated attributions of harmful intent and self-interest after each interaction. In Study 2 (N = 1011), a new sample of participants completed the same procedure and additionally completed measures of anxiety, worry and interpersonal sensitivity. As predicted, prior paranoid ideation was associated with higher and faster overall harmful intent attributions, whereas attributions of self-interest were unaffected, supporting the sensitization model. Contrary to predictions, neither worry, interpersonal sensitivity nor anxiety was associated with harmful intent attributions. In a third exploratory internal meta-analysis, we combined datasets to examine the effect of paranoia on trialby-trial attributional changes when playing fair and unfair dictators. Paranoia was associated with a greater reduction in harmful intent attributions when playing a fair but not unfair