Disease | Administrative data | GP data |
---|
 | HOSP | DRUG | EXE | PROBLEM |
 | (ICD9CM)+
| (ATC)++
| (ICD9CM) | (ICD9CM) |
Diabetes mellitus | 250* | A10 | 250 | 250* |
Treated diabetes | Â | A10 | Â | 250* AND |
 |  |  |  | A10++++
|
Ischaemic heart disease | 410-*414* | C01DA | 414 | 410-*414* |
Heart failure | 428*, 40201, | - | 428 | 428*, 40201, |
 | 40211, 40291, |  |  | 40211, 40291, |
 | 40401, 40403, |  |  | 40401, 40403, |
 | 40411, 40413, |  |  | 40411, 40413, |
 | 40491, 40493 |  |  | 40491, 40493 |
COPD | 490*-492*, | R0+++
| - | 490*-492*, |
 | 494*, 496* |  |  | 494*, 496* |
- Algorithms for case ascertainment of diabetes, ischaemic heart disease, heart failure, and chronic obstructive pulmonary disease (COPD), respectively from regional administrative databases and from GP databases. Regional administrative databases link Hospital discharge records (HOSP), Drug dispensation records (DRUG), and Disease-specific exemptions (EXE) from 2003 to 2008, and a patient was classified as having the selected disease if at least one of the listed conditions were met, ie condition 1 OR condition 2 OR condition 3. GP databases were queried in the PROBLEM field of the clinical database, where diagnosis are coded.
-
+Either in main or in one of the secondary diagnoses.
-
++At least two dispensations in different dates in a single year.
-
+++A specific algorithm involving number, heterogeneity of ATC codes and time span of dispensations is used, see [21].
-
++++Patients having at least 2 prescriptions in one of the previous 2 years.