AB0803 Accuracy of 'Hospital Militar Central' administrative data base for Psoriatic Arthritis diagnosis in patients at a high complexity center

Autor: A. Beltrán, S. Puentes, S. M. Moreno Lopez, C. A. Camargo Rodríguez, M. Laguna, J. Amador, M. Botero Varon, L. A. Castro
Rok vydání: 2022
Předmět:
Zdroj: Annals of the Rheumatic Diseases. 81:1528.1-1528
ISSN: 1468-2060
0003-4967
DOI: 10.1136/annrheumdis-2022-eular.2102
Popis: BackgroundPsoriatic Arthritis (PsA) is a chronic inflammatory musculoeskeletal disease highly associated with Psoriasis (Pso), with a substantial morbidity and disability. (1,2) Initially considered as a type of Rheumatoid arthritis but afterwards was defined as a distinct entity (3). PsA can cause damage to joints which lead patients to great impairment.(4)The estimated prevalence is variable according to the geographical zone, but it has been established in general population between 0,2-1%, meanwhile in Pso patients could be as frequent as 6-42% (5,6). Patients with PsA present pain and stiffness in the affected joints, they even can have axial skeleton compromise, peripheral arthritis, several nail changes, enthesitis and dactylitis (5,6). A delay in diagnosis as much as 6 months is associated with a much lower treatment response, meanwhile early treatment with anti-inflamatory or immune-modulating drugs improves clinical and radiological outcomes (7). There are several useful tools which allow us to establish easily prevalence, incidence or even chronic associations or acute diseases in public healthcare, these tools are known as administrative databases.(8) In the last few years, there has been developed multiple algorithms for the diagnosis of different pathologies using the information gathered from diagnostic codes, pharmacy reports and surgical procedures in order to identify relevant information for example a disease prevalence of an specific population.(9)ObjectivesDetermine the accuracy of the 10th International Classification of diseases (CIE-10) for Psoriatic Arthritis (PsA) diagnosis in the hospital Militar of Bogotá administrative database and to examine the effects of adding specific pharmacy data to CIE 10 on accuracy of PsA diagnosis.MethodsWe drew a random sample of patients from all ambulatory patients who had at least 1 clinic visit to the hospital between January 1 2015 and February 1 2020 who are 18 years or older. Charts of 972 patients were reviewed. The gold standard for PsA diagnosis was chart documentation of PsA diagnosis by a Rheumatologist. The data definitions of PsA diagnosis included presence of CIE-10 alone or various combinations of CIE-10 an the addition of specific pharmacy data to the algorithm. Accuracy of data definitions of PsA was assessed by calculating sensitivity, specificity, positive and negative predictive values, plausibility reasons and area under the receiver operator characteristics curve (ROC). We used Python and R studio software for the database analysis.ResultsFrom the administrative database we had a prevalence of 5.2% patients with PsA, and was determined the diagnosis algorithm using CIE-10 code L40.5 had a correct classification rate by the algorithm of 97.9%, with 84.3% sensitivity CI 95% (71.4-93), and a higher specificity of 99.3% CI 95% (98.7-99,8). Evaluating positive predictive value of the test was 89.6% CI 95% (77.3 -96.5), while the negative predictive value was 99.1% CI 95% (98.3-99.6). In these results the number of false negatives are higher than the number of false positives, when compared with the gold standard (physician diagnosis in clinical chart). Through a statistical analysis we calculate the ROC curve and the area under the curve which was 0.92 CI 95% (0.868-0.969) when used the CIE-10 code L40.5 for PsA diagnosis. Indicating that the test, taking into account the CIE-10 code L40.5, has a very good performance classifying correctly the PsA patients.ConclusionCIE-10 code L40.5 in the administrative database is a very sensitive and specific screening tool for identifying patients with PsA in the general clinic population. This can be used to evaluate large population samples and classified them for public health research using this kind of databases.Disclosure of InterestsNone declared
Databáze: OpenAIRE
Externí odkaz: