Swedish Dementia Registry and Swedish Total Population Register

SveDem is a Swedish quality-of-care registry for dementia and has been thoroughly described previously [19]. SveDem was established in 2007 with the purpose to register all dementia patients in Sweden at the time of diagnosis and standardize their care. Patient variables include clinical characteristics (e.g., Mini-Mental State Examination [MMSE]), sociodemography (e.g., living arrangements), community support (e.g., daycare), and chronic pharmacological treatment [19]. Dementia disorders recorded in SveDem include Alzheimer’s disease, mixed-pathology dementia, vascular dementia, dementia with Lewy bodies, frontotemporal dementia, Parkinson’s disease dementia, unspecified dementia and other dementia types. In 2018, more than 82,000 dementia patients have been registered, which makes SveDem the largest dementia registry in the world [20].

Population Register is available in computerized form since 1968 and is the basis for official population statistics for large parts of Statistics Sweden’s operations [21]. This register provided dementia-free controls for the patients with dementia.

To improve precision of dementia cases, only diagnoses of dementia from SveDem were included, while patients with dementia diagnoses included in other registers (see exclusion criteria for controls below) and not recorded by SveDem were excluded. Specifically, only patients diagnosed with any dementia and registered to SveDem between May 1, 2007 until October 16, 2018 were included (80,004 patients [18.8% out of all dementia patients in the original data - 424,624 patients]).

In the original data extraction, a pool of dementia-free subjects was extracted from the Population Register. The exclusion criteria for controls were as follows: a) having dementia diagnosis recorded in SveDem; b) or ICD-10 codes F00-F03, G30, G31 (see Supplementary List 1) recorded by the Swedish National Patient Register (Patient Register) or Swedish Cause of Death Register (Death Register); c) or having ATC code N06D (anti-dementia drugs) recorded by the Swedish Prescribed Drug Register (Drug Register); To improve the pool of eligible controls, subjects who had ICD-10 codes F05-F09, G32 (Supplementary List 1) were not considered as controls (nor as dementia cases). Consequently, up to four dementia-free controls per one dementia case were matched with dementia cases on birth year (±3 years), sex, and the county of residence and assigned an index date matching with the dementia diagnosis date (1,328,035 controls matched with 424,624 cases in total).

Afterwards, only subjects diagnosed with diabetes with and without diagnosis of dementia, where dementia diagnosis originated only from SveDem were selected (184,560 subjects in total). After excluding subjects with incorrect or missing data and patients with type 1 diabetes, the cohort consisted of 133,318 subjects with diabetes, while 12,284 (9.2%) had diagnosis of dementia and 121,034 (90.8%) were dementia-free (Supplementary Figure 1). This dementia –dementia-free population with diabetes was the basis for the whole-cohort analysis. In addition, 1:1 propensity score (PS) matching was used to create a pool of comparable dementia –dementia-free pairs with diabetes. As a convention, the term “index date” will refer to both date of dementia diagnosis in the dementia cohort and assigned index date in the dementia-free cohort. Detailed description of the dementia cohort is summarized in Supplementary Table 1.