Development and validation of an image-based questionnaire ARTROVIS to identify risk factors for the articular syndrome and related conditions

Abstract

Background: The importance of timely diagnosis of rheumatic diseases is beyond any doubt. However, in the real practice the time from manifestation of the first symptoms to the assessment by a rheumatologist and treatment administration often exceeds the window of opportunity for disease control. At the moment, there has been no Russian-language questionnaire that would allow a primary care physician to suspect a rheumatic disease manifesting with an articular syndrome. Aim: To develop, validate and test a questionnaire to identify risk factors for joint disorders and related conditions. Methods: From November 2022 to April 2024, we performed the study on the image-based questionnaire on identification of the risk factors for joint syndromes and associated conditions (ARTROVIS), which we have developed for early detection of both joint syndromes, as well as comorbid conditions associated with elevated serum uric acid levels (hyperuricemia). The initial test version of the questionnaire was tested in a focus group of respondents (n = 20) who came for regular medical examination. Based on the results of this face validity assessment, we corrected the wording of 10 questionnaire items. At the second study step, we assessed the primary validity and reliability of the questionnaire in a group of 41 rheumatology patients. The third study step involved 828 respondents (408 men and 420 women, median age 19 [18,0; 20,0] years; 21% with joint syndromes), who came for routine medical examination. The reliability of the questionnaire, its face and discriminant validity, as well as sensitivity and specificity (the ROC analysis) were assessed from 827 questionnaires included in the final processing (one was excluded due to questionnaire completion defects). Results: The self-report questionnaire for patients (ARTROVIS) has 24 items aimed at collecting social and demographic data, identifying risk factors for non-communicable diseases, assessing genetic aspects, identifying complaints from various organ systems, providing information on medical treatments taken by the patient (non-steroid anti-inflammatory agents, diuretics, antihypertensives, glucocorticosteroids, chemotherapy and radiation therapy), as well as an additional question for women on current pregnancy pathology. The questionnaire contains visual aids (colored photos) that facilitate the detection of clinical signs of autoimmune (psoriatic arthritis, Reynolds’s syndrome) and inflammatory rheumatic disease (gout), as well as the differential diagnosis of the joint syndrome. The assessment of face validity of the questionnaire showed the total number of missed answers of 3.42%; 62.85% of the respondents gave their full answers to all items. Good discriminant validity was demonstrated. The ROC analysis showed high correlation between the questionnaire score and the presence of a joint syndrome: the area under the curve was 0.884. The Cronbach’s alfa was 0.798, which is sufficient for a medical questionnaire. Conclusion: We have performed the approbation and standardization of the questionnaire to identify risk factors for the joint syndrome and related comorbidities by assessing its validity and reliability. The ARTROVIS as a self-report questionnaire can be used in primary care as a tool enabling general practitioners and internists to decide on further patient routing to corresponding specialists.