Enhancing Analytic Abilities to Identify Complex Patients in 225 Practice Partner Research Network (PPRNet) Practices in 42 states : July 2010-July 2012/ Steven M. Ornstein .

Format:

Datafile

Contributor:

Ornstein, Steven M.

Inter-university Consortium for Political and Social Research.

Series:

ICPSR (Series) ; 34554.

ICPSR ; 34554

Language:

English

Physical Description:

1 online resource.

Edition:

2013-03-11.

Place of Publication:

Ann Arbor, Mich. : Inter-university Consortium for Political and Social Research [distributor], 2013.

System Details:

data file

Summary:

Overview Through electronic data collection and improving the efficiency of existing data processes to allow both more complete and specific identification of chronic illness, the study objectives included: Greatly enhance the scope of existing algorithms to permit comprehensive identification of the 20 chronic conditions key to primary care. Improve the specificity of the existing algorithms to permit more precise automated identification of chronic conditions, limiting the amount of human review required. Revise the algorithms to permit identification of more than one condition in a text string. The investigators developed advanced SAS text string search algorithms and developed a modified parsing table that included inclusion and exclusion patterns and resultant diagnoses. The automation searches through each input text string for the inclusion pattern that is not equivalent to the exclusion pattern and maps the string to the corresponding resultant diagnosis. This technique allows the search functions to be easily modified to include additional search criteria and scaled to encompass additional conditions. Data Dictionary A data dictionary for 24 chronic conditions was developed. The dictionary assigns ICD-9 diagnosis codes to problem list text in electronic health record data. The dictionary contains 78,458 records and exists in two forms, a Microsoft Access database and a SAS 9.2 dataset. The Microsoft Access database contains 24 tables, one for each condition. The SAS 9.2 dataset contains four fields. The 24 chronic conditions for which problem list text data were examined and assigned to ICD-9 codes. Conditions include Alcohol Use Disorder, Asthma and Allergic Rhinitis, Atherosclerosis, Atrial Fibrillation, Cerebrovascular Disease, Chronic Liver Disease, COPD, Chronic Renal Disease, Coronary Disease, Dementia, Depression, Diabetes Mellitus, Epilepsy, GERD, Heart Failure, Hyperlipidemia, Hypertension, Migraine Headache, Obesity, Osteoarthritis, Osteopenia/Osteoporosis, Parkinson's Disease, Peptic Ulcer Disease and Rheumatoid Arthritis. Data Access The data dictionary is not available from ICPSR. For use arrangements, please contact Ruth G. Jenkins, PhD (jenkinsr@musc.edu) or Steven M. Ornstein, MD (ornstesm@musc.edu) at the Practice Partner Research Network (PPRNet), Medical University of South Carolina. Cf.: http://doi.org/10.3886/ICPSR34554.v1

Notes:

Title from ICPSR DDI metadata of 2015-01-05.

Access Restriction:

Restricted for use by site license.