Discussion

Secondary stroke prevention is an important part of care post-stroke and involves a multimodal approach to risk factor management through lifestyle adjustment, monitoring of comorbidities and use of medications. Our study found that while some secondary stroke care indicators in Ontario appear to be sufficient, such as prescriptions for antihypertensives, others remain suboptimal, such as LDL and HbA1C testing and achievement of targets for these tests.

The reasons behind low rates in some areas are unknown. Although family physicians provide the majority of long-term secondary stroke prevention care,^{Reference Kernan, Viera and Billinger10} they are facing record numbers of burnout.^{Reference Hiefner, Constable, Ross, Sepdham and Ventimiglia23} Guidelines are also becoming more complex and exist for almost every chronic disease, making it difficult to recall targets and interventions for each patient and each condition. A previous qualitative study on the barriers to the use of chronic kidney disease guidelines suggested several reasons for nonadherence from family physicians including cognitive overload, differing priorities and lack of awareness of the guidelines altogether.^{Reference Nash, Garg, Brimble and Markle-Reid24}

Our study found that many individuals do not have their LDL (16%) or HbA1C (20%) checked, even within three years following an ischemic stroke. For many individuals, even when they are checked, the values do not fall within target recommendations. This is despite high levels of lipid-lowering and diabetic medication prescription and despite many individuals seeing their family physician, on average, once every two months in that first year after stroke. Our findings suggest better glycemic compared to lipid control among those tested with over 80% within the target of ≤ 7% HbA1C versus 54.5% for LDL ≤ 1.8 mmol/L. This is in contrast to a previous study from Ontario comparing rates in urban versus rural settings of only around 54% for HbA1C.^{Reference Kapral, Austin and Jeyakumar25} Additionally, compliance to important secondary prevention medication over the year follow-up drops, especially for medications like statins and anticoagulants, which each show a reduction in individuals continuing with medications of approximately 50%. As better adherence to guidelines has been associated with a reduced risk of stroke(7), future studies should address barriers to medication use and laboratory monitoring in the post-stroke setting to improve secondary stroke risk.

Another important factor that has been recommended for secondary prevention is the receipt of an annual influenza vaccine, which appears unique to the Canadian Stroke Guidelines.^{Reference Gladstone, Lindsay and Douketis5} Our results were found to be suboptimal and significantly lower than the target of 80% of high-risk individuals set out by the Public Health Agency of Canada.^{Reference Roumeliotis, Houle, Johal, Roy and Boivin26} These rates also appear lower than previous population averages in the general population, which is typically around 70% of those over the age of 65,^{Reference Sulis, Basta and Wolfson27} and lower than in previous studies of individuals with cardiovascular disease.^{Reference Cho, Houle and Alsabbagh28} Interestingly, despite recommendations in the Canadian guidelines for secondary prevention, most previous research on influenza vaccination is in primary stroke prevention.^{Reference Holodinsky, Zerna, Malo, Svenson and Hill29–Reference Lee, Bae, Hwang, Kim, Suh and Ko31} Future studies could assess the effect of influenza vaccines on secondary prevention specifically,

Most individuals in our study were prescribed antihypertensives and lipid-lowering medications. It is not prudent for all individuals to be on these therapies, as some will have adequate blood pressure or lipid levels without the use of medications, and others still may have allergies or intolerances. Previous studies have shown that the use of statins after stroke, even in those with LDL levels within the target range, is associated with reduced mortality and vascular outcomes.^{Reference Kim, Jeong and Chang32,Reference Kim, Lee and Kim33} Overall, our findings were similar to previous studies. Dalli et al. found that 75% of individuals were prescribed antihypertensives and 84% statins,^{Reference Dalli, Kim and Cadilhac6} while Kapral et al. found over 80% of individuals with a previous stroke were prescribed antihypertensives.^{Reference Kapral, Austin and Jeyakumar25}

This study has several strengths. We obtained a very large sample size of over 50,000 individuals in the largest province in Canada, to capture secondary stroke prevention trends at a population level, resulting in one of the largest studies assessing secondary stroke prevention care indicators. We also were able to incorporate laboratory data to assess rates of meeting recommended secondary stroke prevention targets, which is not available in all administrative database studies. This study also has several limitations. First, administrative data has limits, including capturing that a medication is dispensed, without knowing about compliance, or potential appropriate reasons for nonadherence to guidelines. Additionally, some medications such as low-dose aspirin are available over the counter and would not be captured using the ODB. Further, influenza vaccinations may be given by other prescribers under the guidance of a physician or pharmacist, such as a nurse practitioner, which may not be captured in our dataset. Other important factors for secondary stroke prevention including blood pressure readings, diet and smoking status are also not available through our administrative datasets. Control of these factors is also crucial for secondary stroke prevention. Next, we stopped accrual in 2019 to minimize the effect of the COVID-19 pandemic on secondary stroke prevention care. Because of this, results may not be generalizable to a post-COVID era, and future research may be warranted to assess if the COVID-19 pandemic affected secondary stroke care. Lastly, we only recruited patients in Ontario and those not previously living in an LTC facility, which may limit the generalizability to other jurisdictions and populations.

In summary, secondary stroke prevention care in Ontario remains suboptimal in many areas. Future work should explore barriers to better care.