Table 1

Data sources and definitions

Data sourceDefinition
COVID-19 casesRoutine testing data provided by NHS Scotland and UK Government Lighthouse labs, uploaded to (Electronic Communication of Surveillance in Scotland) and accessed via the Corporate Data Warehouse.Cases were defined as any positive RT-PCR test falling within date ranges of each wave, deduplicated at the individual level. Information on age and sex were derived from CHI prior to anonymisation. Social deprivation was ascertained from the Scottish Index of Multiple Deprivation (SIMD) quintile, a measure based on postcode area (also derived from CHI).
Intensive care and high dependency care unit (ICU and HDU)Data provided by the Scottish Intensive Care Society Audit Group (SICSAG).Admissions linked to a positive laboratory report of COVID-19 either during the 21 days preceding the date of admission, or during their stay. Information on sex, age, and SIMD are presented at patient level, as a single patient may be admitted multiple times. As SICSAG only collect data on adults, data on ICU and HDU patients under 20 years old was not available.
ComorbiditiesLinkage to Scottish Morbidity Records (SMR01) and the Prescribing Information System (PIS) for the 5 years prior to infection.Comorbidities previously shown to be associated with COVID-1915–17 were included for analysis. SMR are hospitalisation discharge records which use International Statistical Classification of Diseases Tenth Revision (ICD-10) diagnostic codes. Codes for the previous 5 years were extracted, excluding discharges less than 25 days before testing positive for SARS-CoV-2. For PIS, British National Formulary (BNF) codes were used to identify primary care prescriptions for the previous 5 years, limited to 15 days prior to their positive test result. A case was defined as having a comorbidity if they had a linked SMR or PIS record with relevant ICD-10 or BNF codes as defined in online supplemental table A1.
DeathsNational Records of Scotland (NRS).NRS registered death within 28 days of the first RT-PCR positive. Postmortem results were also included, where the first positive specimen date is recorded after the date of death. Deaths were grouped into waves by the date of infection, therefore there was a 28-day follow-up at the end of wave 3 to allow for subsequent deaths to be included. COVID-19-related deaths confirmed via postmortem were categorised into each wave by date of death.
  • CHI, Community Health Index; HDU, high dependency unit; ICU, intensive care unit; NHS, National Health Service.