Data Limitations
The extent to which the data on people visiting physicians for arthritis and related conditions capture the full spectrum of people with arthritis in Canada is unknown. The data presented in Chapter 4 cover only the fiscal year 1998/99, and since not all people with arthritis see a physician in the course of a year the data do not account for any potential patients not visiting in that time period.
Table 4A-1 Arthritis and related conditions diagnostic codes |
|||
Disease Category |
Condition |
Diagnostic Categories |
Diagnostic Code(s) |
Arthritis and Related Conditions |
Osteoarthritis |
Osteoarthritis |
715 |
Rheumatoid Arthritis |
Rheumatoid arthritis, Still's disease |
714 |
|
Connective Tissue Disorders |
Disseminated lupus erythematosus, generalized scleroderma; polyarteritis nodosa, temporal arteritis |
710; 446 |
|
Ankylosing Spondylitis |
Ankylosing spondylitis |
720 |
|
Gout |
Gout |
274 |
|
Other Arthritis and Related Conditions |
Traumatic arthritis; pyogenic arthritis; joint derangement, recurrent dislocation, ankylosis; Dupuytren's contracture; arthropathy associated with other disorders classified elsewhere; internal derangement of the knee; other and unspecified disorder of the joint; polymyalgia rheumatica; peripheral enthesopathies and allied syndromes; synovitis, tenosynovitis, bursitis, bunion, ganglion; fibrositis, myositis, muscular rheumatism; other diseases of the musculoskeletal system and connective tissue |
716; 711; 718; 728; 713+#; 717#; 719#; 725#; 726#; 727; 729; 739+ |
|
+ Diagnostic code not used in Saskatchewan # Diagnostic code not used in Ontario |
Diagnostic codes provided in physician claims were not validated. Further, many types of visit, such as visits to discuss negative test results and visits for non-specific conditions, may have been difficult to code by diagnosis. Individual physicians may have used a small subset of codes as a matter of routine or convenience. On the other hand, infrequently used codes, such as for rheumatoid arthritis, may have been more likely to be used appropriately, particularly in a primary care setting where the physician may have had to look up the proper code.
In this chapter, individuals were included in the data and analyses for a particular condition if they made at least one visit to any type of physician for which the diagnostic code corresponded to that condition. As a result, this may have included patients with only tentative diagnoses. When initially investigating a patient's condition, physicians may have entered on the claim form a diagnosis that was later ruled out by test results or further examination.
While the diagnostic codes used by the provinces were all based on the International Classification of Diseases (ICD), each province has modified this classification to some degree. Some used 3-digit diagnostic codes (Saskatchewan, Manitoba, Ontario, Nova Scotia) and others used 4-digit codes (British Columbia, Alberta, Quebec). Some provinces were missing codes, and the conditions associated with each code varied somewhat among provinces. If a code was not available for a particular arthritis condition, it is likely that the physician used another arthritis-related code instead. Missing arthritis codes may have been replaced by more general musculoskeletal diagnostic codes or coded in a less predictable manner. Such coding differences may explain at least some of the provincial variations in rates presented in this chapter. Large provincial differences in coding some conditions, such as fibrositis, prohibited the presentation of data on these conditions. As a result they were grouped as “other arthritis and related conditions” (Table 4A-1). Data were not presented on this grouping because of the heterogeneity of the conditions included.
Physicians in all the participating provinces, except Alberta and Nova Scotia, were allowed to enter only one diagnosis for each visit. While physicians in Alberta and Nova Scotia were able to provide three diagnoses per visit, only the first diagnosis was included in the data to achieve comparability with the other provinces. Using only a single diagnostic code means that if a patient had more than one reason for visiting, some diagnoses were missed. Since arthritis is often seen as a co-morbid condition, a physician may have been less likely to provide an arthritis code than that of another disease.
Provincial health insurance claims typically include only fee-for-service claims, so that physicians and patients enrolled in alternative payment plans are not usually included. However, some of these physicians submit “shadow bills” to the provincial health insurance plan with diagnostic information. If submitted, these claims were included in the data presented for Ontario, Saskatchewan and Nova Scotia. Data missing from alternative payment plans are not likely to have had a major effect on data validity in this chapter, as only a small minority of Canadians are enrolled in such plans. However, omission of those covered by alternative payment plans means that the findings in this report are likely to be underestimates of ambulatory care for arthritis.
An additional limitation to consider is that physician specialty was determined solely by registered specialty in all of the provinces, with the exception of Ontario and Nova Scotia, where billing specialty was also considered. Registered specialties may not have been accurate if physicians did not update the provincial health insurance plan once specialty and subspecialty training, such as internal medicine and rheumatology, was completed. The presented groupings of “all medical specialists” and “all surgical specialists” are therefore more likely than separately grouped “internal medicine” and “rheumatology” or “orthopedic surgery” to be accurate and comparable provincially.
[Previous][Table of Contents][Next]