The risk of depression and suicidality in patients with psoriasis, psoriatic arthritis or ankylosing spondylitis: a retrospective United Kingdom database analysis

Date of ISAC Approval: 
08/10/2015
Lay Summary: 
In a previous study using CPRD's predecessor the General Practice Research Database (GPRD), associations were found between psoriasis and first-time depression and first-time suicidality (suicide or suicidal thoughts or behaviour). This was based on patients diagnosed with psoriasis between 1987 and 2002. We wish to repeat this study using more recent data and also to consider patients with two related conditions, psoriatic arthritis and ankylosing spondylitis. There is evidence that these conditions may also have an association with depression. To carry out this study we will broadly follow the methods used in the original paper, selecting patients with any of the three conditions and matching each patient with five other patients of the same sex and similar age who do not have these conditions but attend the same practice. We will then follow the records for these patients and compare the number of patients subsequently diagnosed with depression or suicidality.
Technical Summary: 
This study aims to estimate the incidence of depression and suicidality in patients identified with psoriasis, psoriatic arthritis (PsA) or ankylosing spondylitis (AS), compared with the general population, using a retrospective cohort design. Patients with incident diagnoses between 1st January 2000 and 30th June 2015 will be extracted. Controls will be matched by age (+/- two years), gender and primary-care practice at a ratio of 5:1. All non-exposed controls will be required to have had a primary-care contact within +/-60 days of their respective case's index date. Psoriasis patients will be further divided into mild and moderate/severe groups based on prescribed therapy. Outcomes will be defined by Read or ICD-10 code. Crude incidence will be calculated based upon total observed outcomes divided by total follow-up time. Age- and gender-specific rates will also be calculated. Adjusted hazard ratios will be determined using Cox proportional hazard models adjusted for age, gender, previous history of depression or suicidality, Charlson index and previous health contacts. Sensitivity analyses will be conducted: 1) Including patients with a prior history of depression or suicidality 2) Excluding patients seen less than once per year on average to assess observation bias 3) Excluding patients with co-morbidity measured by the Charlson index.
Health Outcomes to be Measured: 
Incident of depression Incident of suicidality
Collaborators: 

Dr Christopher Morgan - Chief Investigator - Human Data Sciences
Haijun Tian - Collaborator - Novartis Pharmaceuticals Corporation
Paola Primatesta - Collaborator - Novartis Pharma AG
Dr Sara Jenkins-Jones - Collaborator - Human Data Sciences

Linkages: 
HES Admitted;ONS