ORIGINAL ARTICLE Musculoskeletal Cost-effectiveness analysis of different imaging strategies for diagnosis of haemophilic arthropathy U. KHAN,* C. BOGUE,* W. J. UNGAR,  P. HILLIARD, à M. CARCAO,§ R. MOINEDDIN– and A. S. DORIA* *Department of Medical Imaging;  Department of Child Health Evaluative Sciences; àOrthopaedics Program, Rehabilitation Services; §Department of Haematology/Oncology, The Hospital for Sick Children, Toronto, ON, Canada; and –Department of Public Health, Saint Michael’s Hospital, University of Toronto, Toronto, ON, Canada Summary. Physiotherapy and radiography of the joints are standard diagnostic strategies for assessment of haemophilic arthropathy. The use of ultrasonogra- phy as an adjunct tool for early diagnosis of haemo- philic arthropathy may optimize factor replacement therapy. The objective of this study was to compare costs and effectiveness of physiotherapy, radiography and ultrasonography (intervention strategy, IS) with physiotherapy and radiography alone (standard care strategy, SCS) for diagnosing soft tissue and osteocar- tilaginous changes in haemophilic joints. We retro- spectively compared costs and effectiveness of IS vs. SCS in knees, ankles and elbows of 31 children (age range, 4–17 years) with haemophilia A (n = 30) or B (n = 1) (IS, n = 11; SS, N = 20). Direct health care costs were measured from the provincial health care perspective. Effectiveness was measured by false- negative (FN) rates in each study arm by comparing presence or absence of abnormalities of physiotherapy and imaging exams to the reference standard measure (MRI). In scenario 1, all diagnostic tests matched with MRI. In scenario 2, at least one diagnostic test matched with MRI. The IS was more costly [incre- mental cost/100 patients, Canadian (CND) $4987] and more effective (incremental effectiveness, FNs/ 100 patients for scenario 1, –4.09, and for scenario 2, – 41) for both scenarios. The incremental cost-effective- ness ratios for scenario 1 and for scenario 2 were CND$1166 and CDN$116 per FN result averted per 100 patients, respectively. In conclusion, in the short- term, the incorporation of ultrasonography in a test set for diagnosis of haemophilic arthropathy substan- tially improved the diagnostic performance of this test set, however at an increased cost. Keywords: haemophilia, joints, magnetic resonance imaging, physiotherapy, radiography, ultrasonogra- phy Introduction Haemophilia is characterized by recurrent bleeding into joints, muscles, interfascial, subperiostic and intraosseous spaces [1]. Despite recent improvements in haemophilia management, arthropathy remains a major concern for physicians, patients and their families. Early detection of soft tissue changes in haemo- philic joints that are free of osteochondral changes may lead to a reduction in recurrent episodes due to the initiation or intensification of prophylactic treat- ment [2]. Therefore, imaging plays an important role in the management of haemophilia since it enables non-invasive characterization of joint status. Although conventional radiography remains the standard imaging technique for diagnosis and fol- low-up of arthropathic changes in haemophilic joints, only advanced articular changes are ade- quately assessed by radiography [3]. Soft tissue changes are not well delineated and cartilage loss is not visualized directly, but inferred when the joint space is narrowed [4]. Presently, magnetic resonance imaging (MRI) is considered the most accurate diagnostic technique available for assessment and monitoring of joint status in haemophiliacs [5,6]. It has been used as a reference standard for comparison Correspondence: Andrea S. Doria, MD, PhD, MSc, Department of Medical Imaging, The Hospital for Sick Children, 555 University Avenue, Toronto, ON, Canada M5G1X8. Tel.: +1 416 813 6079; fax: +1 416 813 7591; e-mail: andrea.doria@sickkids.ca Accepted after revision 5 September 2009 Haemophilia (2010), 16, 322–332 DOI: 10.1111/j.1365-2516.2009.02125.x 322 Ó 2009 Blackwell Publishing Ltd