-
Notifications
You must be signed in to change notification settings - Fork 11
/
NEWS
65 lines (57 loc) · 2.91 KB
/
NEWS
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
NEWS for the comorbidities.icd10 package
Changes for 0.6.1
-----------------
* Fixed bug caused by wrong number of test codes in preprocessor
Changes for 0.6.1
-----------------
* Fixed preprocessor bug
Changes for 0.6
---------------
* Ruijingfang Jiang noted that the HIV-code differs for Sweden
* Fixed is.ICD when ICD-codes are empty
* Renamed argument for cmrbdt.calc from "include_acute_column"" to a more
logical "incl_acute_codes" argument
Changes for 0.5
---------------
* Added tests
* Bugs in AHRQ-matches fixed and hypertension is now properly split into
uncomplicated and complicated
* Added AHRQ v3.6 and 3.7 version and ability to quickly update to newer updates
Changes for 0.4
---------------
* Externalized hierarchy and points calculator
* Added the main cmrbdt.calc() function
* Added a dot preprocessor
* Expanded the tests
* Improved manual - hidden some of the private functions and added families
* Added deprecated warnings for deyo, ahrq, and elixhauser functions
Changes for 0.3
---------------
* Started implementing regular expression alternatives for elixhauser and charlson scores
* Refactored use in order to optimize the use of vectors with icdCodes for a single individual.
* The original NA handling from the comorbidities package is now changed so that the missing codes are simply skipped - should have significant performance improvements
* Renamed the functions to lower-case in for package consistency
* Externalized numeric and regular expression functions as these may be of interest to users and renamed them to codefinder.*
Changes for 0.2
---------------
* Implemented testthat tests
* Bug found in the weighting of the Charlson's Deyo index; the weights that are set to 6 are wrong, note the code below:
- output.frame[, 16] <- output.frame[, 17] * 6
- output.frame[, 16] <- output.frame[, 17] * 6
- both update the 16 column and not their own column
* Bug found in the Deyo aortic aneurysm group:
- 441, 441.9 does not include all 441 codes that correspond to the code according to CMS.goc ICD-9
* Bug found in the Deyo severe liver group:
- 572.3 and 572.4 cause a floating point mismatch -> Switched everyting to integers for Deyo and Elixhausers
* Bug found in the Deyo renal group: 585 is a group and not just 585.00
* Bug - use of seq(xxx.0 to xxx.9, by=0.01) will not include xxx.91 or above
* Split Elixhauser hypertension group into complicated and uncomplicated
* Switched to standard NROW/NCOL
Changes for 0.1
-------------------
* Refactored the code for the comorbidities package to better suit a package-environment
* Instead of looping through the values in each code-list the %in% is now used
* The output.frame is now based on the actual names of the groups, should limit the risk of misclassifying diseases.
* The hierarchy check is now vectorized and also applies to the deyo calculation for diabetes and malignancies
* Converted the Rd documentation to roxygen format
* Allowed for numeric icd-codes