Measures
Our main instrument was a semi-structured interview, which included items on demographics, use of general and mental health services, COVID-19 experience, worsening of symptoms, emergence of new symptoms, and severity of illness (patient’s global impression-PGI and clinician’s global impression-CGI) (Guy, 1976). Clinical data from the patient records (primary and secondary ICD-10 diagnoses, clinic visit data, illness duration) were also extracted. At the time of data collection, there were no established measures in Turkish to assess COVID-19 related anxiety; therefore, we developed an 8-item scale with 4-choice Likert response set to assess that. A general question “How much anxiety/distress did the following cause during the pandemic?” was followed by eight specific questions, such as: “acquiring the virus”, “passing it on to my close ones”, etc. coded as “none, a little, moderately, very much”. The internal consistency (alpha) of the 8 items was .83.