Immigration Status Linked to Delayed Flash Glucose Monitoring Access

Key Takeaways

• Matched immigrants had slower uptake of publicly funded flash glucose monitoring than matched long-term residents.
• Recent immigrants had the longest delay and lowest hazard, while long-term immigrants also showed a smaller persistent lag.

Among older Ontario adults with insulin-requiring diabetes who were eligible for public drug coverage, a JAMA Network Open cohort study found slower access to first publicly funded flash glucose monitoring among immigrants than among long-term residents, with median time to initiation of 583 days versus 514 days in the matched analysis.

The cohort included 109,079 Ontario residents aged 66 years or older who were eligible for the Ontario Drug Benefit program, had insulin-requiring diabetes as of September 15, 2019, and were followed through March 31, 2023. Immigrants since 1985 were classified as recent immigrants after 10 years or fewer in Canada or long-term immigrants after more than 10 years, and they were compared with long-term residents.

Within the full cohort, 13,257 immigrants were matched to 13,257 long-term residents on age within 2 years, sex, and propensity score. Rates were 36.7 versus 38.8 initiations per 100 person-years for immigrants and long-term residents, with an adjusted hazard ratio of 0.95 (95% CI, 0.93-0.98; P=.003). In subgroup analyses, median times were 830 versus 662 days for recent immigrants and their matched long-term resident comparators, with an adjusted hazard ratio of 0.87 (95% CI, 0.80-0.94; P<.001). For long-term immigrants, median times were 544 versus 480 days, and the adjusted hazard ratio was 0.94 (95% CI, 0.91-0.98; P<.001). The greatest delay was concentrated among recent immigrants.

Sensitivity analyses separated adjusted findings from unadjusted comparisons. The crude overall comparison between immigrants and long-term residents was HR 0.99 (95% CI, 0.97-1.01; P=.42), while the crude recent-immigrant comparison was HR 0.83 (95% CI, 0.78-0.88; P<.001). When diabetes-related services were removed from the propensity score, the adjusted estimate was HR 0.96 (95% CI, 0.93-0.99; P=.004). When death was treated as a competing risk, the adjusted estimate was also HR 0.96 (95% CI, 0.93-0.99; P=.009).

Several possible explanations for slower publicly funded access were noted, including challenges with health system navigation, patient-clinician communication, and cultural or knowledge barriers. The two available flash glucose monitoring systems were analyzed together, and earlier private insurance or out-of-pocket use could influence measured timing. Residual confounding may remain, including individual income, beliefs, behavior, and lifestyle.