ACR Meeting Abstracts

ACR Meeting Abstracts

Home » Meeting Abstracts » ACR/ARHP Annual Meeting 2017

Abstract Number: 996

Accuracy of Canadian Administrative Health Data in Identifying Patients with Psoriasis and Psoriatic Arthritis Using Primary Care Medical Records As the Reference Standard

Lihi Eder1,2, Jessica Widdifield3, Cheryl F. Rosen4, Dafna D Gladman2, Raed Alhusayen5, Michael Paterson6, Stephanie Cheng6, Shirin Jabbari6, Willemina Campbell7, Sasha Bernatsky8 and Karen Tu6, 1Women's College Research Institute, Women's College Hospital, Toronto, ON, Canada, 2Medicine, University of Toronto, Toronto, ON, Canada, 3University Health Network, Toronto, ON, Canada, 4Dermatology, Toronto Western Hospital, University of Toronto, Toronto, ON, Canada, 5Sunnybrook Research Institute, Sunnybrook Health Sciences Centre, Toronto, ON, Canada, 6Institute for Clinical Evaluative Sciences, Toronto, ON, Canada, 7Rheumatology, Toronto Western Hospital, Toronto, ON, Canada, 8Divisions of Rheumatology and Clinical Epidemiology, Research Institute of the McGill University Health Centre, Montreal, QC, Canada

Meeting: ACR/ARHP Annual Meeting 2017

Date of first publication: September 18, 2017

Keywords: ,

Session Information

Date: Monday, November 6, 2017

Title: Epidemiology and Public Health Poster II: Rheumatic Diseases Other than Rheumatoid Arthritis

Session Type: ACR Poster Session B

Session Time: 9:00AM-11:00AM

Background/Purpose: We assessed the accuracy of algorithms to identify patients with psoriasis and psoriatic arthritis (PsA) in administrative health data in a validation set derived from primary care electronic medical records (EMRs), and contrasted the effect of different algorithms on the population-based prevalence of psoriasis and PsA in Ontario, Canada.

Methods: We developed a validation set using a sample of 2210 adult patients with suspected psoriasis and PsA. This sample was identified through a targeted search for psoriatic disease-related terms in the EMRs of a random sample of 30,424 patients in the primary care Electronic Administrative data Linked Database (EMRALD) in Ontario, Canada. The reference standard for classifying patients with physician-recorded psoriasis or PsA, was confirmed using a retrospective chart abstraction. All patients were then linked to health administrative data to assess the performance of the different algorithms combining physician billing and hospitalization diagnostic codes, medications and procedures for identification of patients with psoriatic disease. The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of each algorithm were computed. Estimated population prevalence of psoriasis and PsA were then calculated for each algorithm.

Results: Based on our reference standard we identified 1028 patients with psoriasis and 90 patients with PsA which resulted in overall psoriasis and PsA prevalence of 3.4% and 0.29%, respectively. The majority of the patients with PsA (67%) had a documented diagnosis by a rheumatologist, while only 29% of the psoriasis patients had a documented diagnosis by a dermatologist. The accuracy of selected psoriasis and PsA case definition algorithms are presented in Table 1. All algorithms had excellent specificity (97-100%). However, the sensitivity and PPV of the algorithms were low to modest, ranging from 28% to 72% for sensitivity and 43% to 72% for PPV. The population prevalence of psoriasis (1.07-2.4%) and PsA (0.12-0.14%) ranged depending on the algorithm used for case definition.

Conclusion: The accuracy of identifying patients with psoriasis and PsA in Ontario health administrative databases varies widely, however when we applied these algorithms to the entire Ontario population, we observed similar patterns and a steady increase in prevalence, irrespective of the algorithm used. We recommend that PsA receives a distinct outpatient diagnostic code to improve case ascertainment.

 

Table 1 – The accuracy of selected psoriasis case definition algorithms

Algorithm

Sensitivity (%)

Specificity (%)

PPV (%)

NPV (%)

Prevalence

(per 100 population)

1 H or 1 P ever

71%

97%

43%

99%

2.40%

1 H or 1 P ever by a specialist

50%

98%

48%

98%

1.69%

1 H or 2 P ever

52%

99%

62%

98%

1.76%

1 H or 2 P ever at least 1 by a specialist

43%

99%

63%

98%

1.46%

1 H or 2 P in 1 years

44%

99%

63%

98%

1.49%

1 H or 2 P in 2 years

47%

99%

63%

98%

1.58%

1 H or 2 P in 3 years

48%

99%

63%

98%

1.61%

1 H or 3 P ever

41%

99%

71%

98%

1.38%

1 H or 3 P in 2 years

28%

100%

72%

98%

1.07%

1 H or 3 P in 3 years

32%

100%

72%

98%

1.14%

1 H or 1 P ever or 1 prescription of anti-psoriatic treatment

72%

97%

42%

99%

2.42%

1 H or 2 P ever or 1 prescription of anti-psoriatic treatment

55%

99%

60%

98%

1.86%

1 H or 2 P ever or 2 prescription of anti-psoriatic treatment

54%

99%

61%

98%

1.82%

H: Hospitalization psoriasis code; P=physician psoriasis diagnostic code; Specialist = dermatologist; anti-psoriatic medications=tar, topical or oral retinoids, topical vitamin D derivate, IL-17 inhibitor, PDE4 inhibitor, IL-12/23 inhibitor

 

Table 2 – The accuracy of selected PsA case definition algorithms

Algorithm

Sensitivity

(%)

Specificity

(%)

PPV (%)

NPV (%)

Prevalence

(per 100 population)

1 H or (1 P(Ps) and 1 P(SpA)) ever

53%

100%

53%

100%

0.13%

1 H or (1 P(Ps) and 2 P(SpA)) ever

51%

100%

64%

100%

0.13%

1 H or (1 P(Ps) and 3 P(SpA)) ever

48%

100%

66%

100%

0.12%

1 H or ((1 P(Ps) ever or 1 prescription of topical anti-psoriatic treatment) and 1 P(SpA) ever)

55%

100%

54%

100%

0.14%

1 H or ((1 P(Ps) ever or 1 prescription of topical anti-psoriatic treatment) and 2 P(SpA) ever)

52%

100%

65%

100%

0.13%

1 H or ((1 P(Ps) ever or 1 prescription of topical anti-psoriatic treatment) and 2 P(SpA) ever at least 1 by a specialist)

52%

100%

66%

100%

0.13%

1 H or ((1 P(Ps) ever or 1 prescription of topical anti-psoriatic treatment) and 3 P(SpA) ever at least 1 by a specialist)

49%

100%

68%

100%

0.12%

H: Hospitalization PsA code; P(Ps)=physician psoriasis diagnostic code; P(SpA)=physician spondyloarthritis diagnostic code; Specialist = rheumatologist; topical anti-psoriatic treatment=tar, retinoids or vitamin D derivate

 

Disclosure: L. Eder, None; J. Widdifield, None; C. F. Rosen, None; D. D. Gladman, None; R. Alhusayen, None; M. Paterson, None; S. Cheng, None; S. Jabbari, None; W. Campbell, None; S. Bernatsky, None; K. Tu, None.

To cite this abstract in AMA style:

Eder L, Widdifield J, Rosen CF, Gladman DD, Alhusayen R, Paterson M, Cheng S, Jabbari S, Campbell W, Bernatsky S, Tu K. Accuracy of Canadian Administrative Health Data in Identifying Patients with Psoriasis and Psoriatic Arthritis Using Primary Care Medical Records As the Reference Standard [abstract]. Arthritis Rheumatol. 2017; 69 (suppl 10). https://acrabstracts.org/abstract/accuracy-of-canadian-administrative-health-data-in-identifying-patients-with-psoriasis-and-psoriatic-arthritis-using-primary-care-medical-records-as-the-reference-standard/. Accessed .

ACR Meeting Abstracts - https://acrabstracts.org/abstract/accuracy-of-canadian-administrative-health-data-in-identifying-patients-with-psoriasis-and-psoriatic-arthritis-using-primary-care-medical-records-as-the-reference-standard/