Table 1. Description of denominator populations and condition definitions.

Condition	Denominator population (condition-specific denominator populations)	Condition definition
Diabetes
Diabetic emergencies	All individuals (aged ≥11 years) with prevalent diagnoses of diabetes mellitus at the start of each week of follow-up. Individuals contributed to the study population from the latest of the start of follow-up in the overall population and the date of their first record indicating a diagnosis of diabetes.	Any record of diabetes-related hyperglycaemia, hypoglycaemia, ketoacidosis, or diabetic coma. Multiple records occurring within seven days of each other were considered as representing the same event.
Mental health
Anxiety	All individuals (aged ≥11 years) from the overall study population.	Any record of symptoms or diagnoses of: social phobia, agoraphobia, panic, generalized anxiety disorder, and mixed anxiety and depression. Multiple records occurring within seven days of each other were considered as representing the same event.
Depression	All individuals (aged ≥11 years) from the overall study population.	Any record of major depressive disorder, dysthymia, mixed anxiety and depression, and adjustment disorders with depressed mood. We also included codes for depressive symptoms. Multiple records occurring within seven days of each other were considered as representing the same event.
Self-harm	All individuals (aged ≥11 years) from the overall study population.	Self-harm defined as records that indicated explicit or undertermined intention to self-harm, non-suicidal or suicidal self-harm (including overdoses with drugs commonly implicated in suicide, e.g. paracetamol). Multiple records occurring within seven days of each other were considered as representing the same event.
Serious mental illness	All individuals (aged ≥11 years) from the overall study population.	Severe mental illness included diagnoses of schizophrenia and other psychotic disorders, and bipolar disorders. Multiple records occurring within seven days of each other were considered as representing the same event.
Eating disorders	All individuals (aged ≥11 years) from the overall study population.	Eating disorders included anorexia nervosa, bulimia nervosa, and other specified feeding and eating disorders. Multiple records occurring within seven days of each other were considered as representing the same event.
Obsessive compulsive disorder	All individuals (aged ≥11 years) from the overall study population.	Obsessive compulsive disorder was defined by codes for body dysmorphic disorders, hypochondriasis, hoarding disorder, and body focused repetitive behaviour disorders. Multiple records occurring within seven days of each other were considered as representing the same event.
Respiratory
Asthma exacerbations	All individuals (aged ≥11 years) with a current asthma diagnosis (i.e. asthma code in the last two or three years if aged <18 years or 18+ years, respectively). Individuals joined the study population from the start of follow-up in the overall population if there was a current asthma diagnosis (i.e. within last 2-3 years) at this time or from the date of their first record indicating an asthma diagnosis within overall follow-up. Participants remained in the study until there was no current asthma diagnosis or the end of overall follow-up. They were able to reenter the study if there was a later diagnostic code for asthma before the end of overall follow-up. Following an existing definition, individuals 40 years and over with asthma were considered as likely to have COPD (and therefore not included in the asthma study population [denominator]) if they had a subsequent COPD diagnosis recorded within the two years following the current asthma record.74	Asthma exacerbations were defined as records for morbidity codes for asthma exacerbations and status asthmaticus, and a primary care prescription for an oral corticoseroid.21 Multiple records occurring within 14 days of each other were considered as representing the same event.
COPD exacerbations	Adults (aged ≥41 years) with an established diagnosis of COPD and evidence of a smoking history.75 Individuals joined the study population from the latest of the start of follow-up in the overall population and the date of their first record indicating diagnosis of COPD.	Exacerbations of COPD were defined using morbidity codes in individuals with existing COPD for COPD exacerbations, lower respiratory tract infections, breathlessness or sputum production, and a new prescription for an oral corticosteroid or antibiotic.22 Multiple records occurring within 14 days of each other were considered as representing the same event.
Cardiovascular
Myocardial infarction	All adults (aged ≥31 years)	Any record for myocardial infarction allowing for a 1-year window between successive records. Multiple records occurring within one year of each other were considered as representing the same event.
Unstable angina	All adults (aged ≥31years)	Any record for unstable angina, allowing for a 6-month window between successive records. Multiple records occurring within six months of each other were considered as representing the same event.
Transient ischaemic attacks	All adults (aged ≥31 years)	Any record for transient ischaemic, allowing for a 6-month window between successive records. Multiple records occurring within six months of each other were considered as representing the same events
Cerebrovascular accident	All adults (aged ≥31 years)	Any record for cerebrovascular accidents, allowing for a 1-year window between successive records. Multiple records occurring within one year of each other were considered as representing the same event.
Cardiac failure	All adults (aged ≥31 years)	Given the complexity with capturing acute events for a chronic condition, we only counted an individual’s first ever diagnosis with cardiac failure.
Venous thromboembolism (pulmonary embolism and deep venous thrombosis)	All adults (aged ≥31 years)	Any record for venous thromboembolism, allowing for a 1-year window between successive records. Multiple records occurring within one year of each other were considered as representing the same event.
Alcohol
Acute alcohol-related event	All adults (aged ≥18 years)	Any record for acute physical or psychological alcohol-related event, including acute alcoholic pancreatitis, Multiple records occurring within 14 days of each other were considered as representing the same event.

⁷⁴Bloom CI, Nissen F, Douglas IJ, Smeeth L, Cullinan P, Quint JK. Exacerbation risk and characterisation of the UK’s asthma population from infants to old age. Thorax 2018; 73: 313–20.

²¹Bloom CI, Palmer T, Feary J, Quint JK, Cullinan P. Exacerbation patterns in adults with Asthma in England A population-based study. Am J Respir Crit Care Med 2019; 199: 446–53.

⁷⁵Quint JK, Müllerova H, DiSantostefano RL, et al. Validation of chronic obstructive pulmonary disease recording in the Clinical Practice Research Datalink (CPRD-GOLD). BMJ Open 2014; 4: 1–8.

²²Rothnie KJ, Müllerová H, Hurst JR, et al. Validation of the recording of acute exacerbations of COPD in UK primary care electronic healthcare records. PLoS One 2016; 11: 1–14.