https://orcid.org/0000-0002-8912-7520
Figures & data
Figure 1 Diagram of cohort study design, describing the dates of cohort entry, covariate assessment, follow-up, and study exit. aIdentified using an algorithm where records identified within −1 year to +1 month of the index date are regarded as the best, +1 month to +1 year from the index date as second best, the nearest before −1 year from the index date as the third best, and the nearest after +1 year from the index date as the worst. Smoking status or body mass index recorded after severe mental illness diagnosis were not used.
Figure 2 Flowchart illustrating identification of participants in atopic eczema and psoriasis cohorts. *Numbers of people with and without atopic eczema or psoriasis do not sum to the total number of individuals included in each cohort. Individuals with atopic eczema or psoriasis could be included in the matched comparison cohort up until the date of their first atopic eczema or psoriasis diagnosis.
Table 1 Characteristics of Atopic Eczema and Psoriasis Cohorts at Cohort Entry
Table 2 HRs (95% CI) for the Association Between Atopic Eczema or Psoriasis and Severe Mental Illness. Fitted to Adults with Complete Data for All Variables Included in Each Model and from Valid Matched Setsa
Figure 3 HRs (95% CI) for the association between atopic eczema or psoriasis severity and severe mental illness. Models fitted to adults with complete data for all variables included in each model and from valid matched sets. aModel implicitly adjusted for matching variables. bModel further adjusted for potential confounders (socioeconomic deprivation – using Carstairs index – and calendar time. cConfounder adjusted model additionally adjusted for potential mediators (comorbidity burden [using Charlson comorbidity index], smoking status, harmful alcohol use, body mass index, and in atopic eczema analyses only, problems with sleep and high-dose glucocorticoid use).
![Figure 3 HRs (95% CI) for the association between atopic eczema or psoriasis severity and severe mental illness. Models fitted to adults with complete data for all variables included in each model and from valid matched sets. aModel implicitly adjusted for matching variables. bModel further adjusted for potential confounders (socioeconomic deprivation – using Carstairs index – and calendar time. cConfounder adjusted model additionally adjusted for potential mediators (comorbidity burden [using Charlson comorbidity index], smoking status, harmful alcohol use, body mass index, and in atopic eczema analyses only, problems with sleep and high-dose glucocorticoid use).](/cms/asset/040710c2-6001-4534-abeb-88a8f62f9ae8/dcle_a_12156216_f0003_c.jpg)