Initial development of the Hemophilia Joint Health Score (HJHS) began in 2003. At the invitation of the International Prophylaxis Study Group, an international group of experienced physiotherapists and physicians convened in Montreal, Canada. They began discussion about developing a new assessment tool. Three scoring systems served as the basis for this evaluation tool. The earliest, known as the World Federation of Hemophilia Orthopaedic Joint Score (WFH), dated to the early 1980s and was developed by a group of orthopedists from Germany, Israel, Venezuela, England and the USA (Gilbert, Seminars in Hematology, 1993). It comprised a physical examination and a pain scale, and was designed to evaluate and score severe joint arthropathy (as was prevalent at that time). It proved insensitive to the earliest signs of joint disease in young children, and psychometric properties of the scale were never established. In 1994, in an effort to address identification of joint disease in young children, the hemophilia center at the University of Colorado, modified the WFH scale to try to make it more sensitive to identifying early joint disease. Earlier modifications of the WFH scale included the Colorado PE-1 and PE - 0.5 scales which added categories of Strength, Gait and Pain to the existing WFH categories. In addition these scales incorporated a category to measure the necessity of the use of splinting or orthotics. The evaluation using either of these two scales was the same but weighting of the scoring in the 1.0 scale was equal, while scoring in the 0.5 scale was unequal. Further modifications of these adult instruments were made to make a tool feasible to use on young children. These modifications included using functional test of strength and the Wong-Baker Faces Pain Rating Scale. Instability was removed from the scale as this was not a problem in young children. (Manco-Johnson, Haemophilia, 2000)This modified scale, was known as the Colorado Young Child scale. It was used in the Joint Outcome Study, a multi-center, prospective, randomized clinical trial which compared joint outcome in children on prophylactic factor infusion with those who received aggressive episodic therapy. (Manco-Johnson, NEJM, 2007. This tool was also used in the Canadian Hemophilia Escalating-Dose Prophylaxis Study (CHPS). (Feldman, JThrombosis &Hemostasis, 2006) In the late 1990s, further modifications of this scale were made by the hemophilia center at Karolinska University Hospital in Stockholm. (Hill, Haemophilia, 2003) This modification was used by the European Paediatric Network for Haemophilia Management in a study which assessed joint outcome in 116 children, comparing the scores to the WFH scale. Issues were identified with both the Colorado scale and the Stockholm scales. These included using a pain measure which was subjective, rather than objective, and difficulty measuring gait on each individual joint, as a problem in any one of the lower extremity joints could impact the gait pattern.
Representatives from Denver and Stockholm met with physiotherapists from Montreal and Toronto in Canada, and Utrecht in The Netherlands to begin discussion on a collaborative, unified scale, in an effort to standardize the methods and tools used for joint evaluation. Strengths and weaknesses of both the Denver and Stockholm scales were identified and a tentative merged system was developed. This scale became known as the Hemophilia Joint Health Score. Studies were initiated to establish reliability (2003) and validity (2006-2007). Reliability of the HJHS was excellent with an interrater coefficient of 0.83 and test-retest reliability of 0.89. (Hilliard, Haemophilia 2006) The HJHS was shown to be 97% more efficient than the WFH scale at differentiating severe from mild and moderate hemophilia, and 74% more efficient at differentiating subjects treated with prophylaxis from those treated on demand. It correlated moderately with a physician global joint score (rs = 0.42, P < 0.0001) and with overall arthropathy impact (rs = 0.42, P < 0.0001). The HJHS and WFH scales both show evidence of strong construct validity, based on their moderate correlation with the physician arthropathy impact score, the physician’s individual joint score and the physician total score. However, the HJHS has been shown to be more sensitive for evaluating mild arthropathy. (Feldman, Arthritis Care & Research, 2011).
Psychometric testing of version 1.0 of the HJHS identified items on the scale that were redundant or rarely endorsed. In version 2.0, the categories of Instability and Gait (per joint) were removed and Axial Alignment was evaluated, but not scored. Additional clarification was made on Crepitus and Muscle Atrophy. Normative tables of ROM were updated for Flexion and Extension Loss. Scoring was revised for the category of Joint Pain and in the Global Gait category, skipping was changed to hopping on one leg. Version 2.1 further clarifies guidelines for scoring Flexion and Extension Loss and Joint Pain. Ankle plantarflexion heel raises have been reduced per reference to the original Daniels & Worthingham 4th edition. The HJHS has been used in studies in Lithuania, Brazil and China, in addition to the countries involved in the testing. A reliability study on use of the HJHS in China was presented as an abstract at the World Federation of Hemophilia Congress in Buenos Aires, Argentina in 2010. (Sun, Haemophilia, 2010). Currently the HJHS is being translated into European Spanish, French and German, in addition to the translations already established in the studies mentioned. Further testing of the 2.1 version is planned in the future to determine the range of its psychometric properties.
The HJHS has been tested in its integral, complete version. The developers strongly recommend that the tool be used in its entirety as outlined in the current instruction manual. If the HJHS is altered in any way we cannot ensure that the reliability and validity of the modified score’s results are equally reliable or valid as the original. Therefore, any modified version of the tool should not be referenced as the HJHS in any publications or presentations.
The Physical Therapy Expert Working Group of the IPSG does not currently have an automated electronic version of the tool available.
Inclusion of the HJHS summary score sheet in an electronic database is acceptable. However, we do not recommend the use of an automated scoring system due to the complexity of specific HJHS items such as Range of motion. Inaccurate item scores can alter both individual joint and overall scores.
Any queries regarding the use of electronic scoring systems should be directed to the PT Expert Working Group through the IPSG website contact page
- Gilbert MS. Prophylaxis: musculoskeletal evaluation. Semin Hematol. 1993 Jul;30(3 Suppl 2):3-6.
- Manco-Johnson MJ, Nuss R, Funk S, Murphy J. Joint evaluation instruments for children and adults with haemophilia. Haemophilia. 2000 Nov;6(6):649-57.
- Hill F, Ljung R. Meeting Report - third & fourth workshops of the European Paediatric Network for Haemophilia Management. Haemophilia. 2003; 9:223-8.
- Feldman BM, Pai M, Rivard GE, Israels S, Poon MC, Demers C, et al. Tailored prophylaxis in severe hemophilia A: interim results from the first 5 years of the Canadian Hemophilia Primary Prophylaxis Study. Journal of Thrombosis & Haemostasis. 2006 Jun; 4(6):1228-36.
- Hilliard P, Funk S, Zourikian N, Bergstrom BM, Bradley CS, McLimont M, et al. Hemophilia joint health score reliability study. Haemophilia. 2006 Sep; 12(5):518-25.