Identifying optic neuritis and transverse myelitis using administrative data.

OBJECTIVE

We aimed to validate administrative case definitions to identify individuals with optic neuritis (ON) or transverse myelitis (TM), and to distinguish which of these individuals had a monophasic presentation versus multiple sclerosis (MS).

METHODS

Using population-based administrative (health claims) data from Manitoba, Canada, we developed case definitions for ON and TM, and distinguished individuals who had monophasic presentations (ON-nonMS, TM-nonMS) versus those later diagnosed with MS (ON-MS, TM-MS). We compared performance of these case definitions to diagnoses based on medical records review in a reference cohort (n = 1251) using sensitivity, specificity, positive predictive value and negative predictive value. We estimated the annual incidence of these conditions for a three-year period (2011-2013).

RESULTS

When compared to medical records, using ≥1 physician visit, the case definition for ON had good sensitivity (88.5%), and specificity (82.7%) whereas the case definition for TM had low sensitivity (25.9%) and higher specificity (89.0%). Findings for the other case definitions tested were: ON-MS (sensitivity: 84.1%, specificity: 83.9%), ON-nonMS (sensitivity: 66.7%, specificity 98.5%), TM-MS (sensitivity: 22.2%, specificity: 90.4%), and TM-nonMS (sensitivity: 3.7%, specificity: 99.7%). After applying the ON and TM case definitions to administrative data, the average annual incidence of ON over the period 2011-2013 was 75.9 per 100,000 person-years (95%CI: 72.8, 79.1) and of TM was 18.3 per 100,000 person-years (95%CI: 16.8, 19.8).

CONCLUSION

Administrative data can be used to identify individuals with incident ON and TM, and to distinguish those with monophasic syndromes from those with an incident presentation of MS.