Third party insurers generally decide which drugs they will reimburse based upon an assessment of value for money. Many drugs are fully funded, others are reimbursed only for patients who fulfill eligibility criteria, and others are not funded at all.[1, 2] Patients often pay out-of-pocket for drugs not covered by a third party.

Supporters of this approach to drug reimbursement note that all patients have equal access to drugs felt to represent good value for money, and that public resources are not well spent on drugs not deemed cost-effective.[2–5] However, critics suggest that it can take time for public plans to incorporate new evidence about a drug’s benefits in their decision-making. They argue that, in these instances, less affluent patients are less able to access effective new drugs when the cost incurs economic hardship.[6, 7] Data from publicly funded drug programs may provide insight into this issue.

For decades, many patients with atrial fibrillation have been anticoagulated with vitamin K antagonists (VKAs) such as warfarin, to diminish their risk of arterial thromboembolism, particularly stroke. The introduction of the direct oral anticoagulants (DOACs) as alternatives to VKAs has been met with cautious enthusiasm among clinicians [8]. The major advantages of the DOACs include their rapid onset of action, shorter half-lives, lack of requirement for regular laboratory monitoring and the absence of food interactions when compared to VKAs.[9, 10] DOACs are now recommended for the prevention and treatment of thromboembolism. Currently available DOACs include dabigatran which directly inhibits the final effector of coagulation, thrombin (factor IIa), while rivaroxaban and apixaban directly inhibit the rate-limiting enzyme of coagulation, factor Xa. The RE-LY multicenter, non-inferiority trial, which compared the use of dabigatran with warfarin for nonvalvular atrial fibrillation (NVAF), found essentially similar rates of stroke or systemic embolism but lower rates of life-threatening and major bleeding with dabigatran [9, 10].

In Canada, dabigatran etexilate was approved by Health Canada on October 26^th, 2010, for thromboprophylaxis in patients with NVAF. It was added to the provincial formulary almost 18 months later, on April 24^th, 2012. We sought to determine whether older Ontarians who switched from warfarin to dabigatran during this period were more likely to live in wealthier neighborhoods, as compared with those who remained on warfarin. A secondary objective tested whether any identified socioeconomic gradient persisted once dabigatran became available through the public drug program.

We conducted a population-based cohort study of Ontarians aged 66 and older with NVAF who were treated with warfarin between October 28, 2008, and October 26, 2010. Patient information was anonymized and de-identified prior to analysis. Written informed consent was not given by participants for their records to be used in this study. This study was approved by the Research Ethics Board of Sunnybrook Health Sciences Centre, Toronto, Ontario.

We identified 34,797 older Ontarians with NVAF treated with warfarin, including 31,614 “non-switchers” and 3,183 “switchers”. The characteristics of patients are described in Table 3.

In the primary analysis, we found a significant association between increasing SES and switching to dabigatran in the interval between its market approval and inclusion on the provincial formulary (p<0.0001; Table 4). In the highest income quintile, 11.4% of patients switched to dabigatran during that time, compared to only 7.3% in the lowest income quintile (adjusted odds ratio 1.50; 95% confidence interval 1.33 to 1.69; Table 1). Other predictors of early switching to dabigatran included younger age, absence of renal disease or diabetes, lower Charlson comorbidity index, fewer concomitant medications, clopidogrel use, provision of care by a cardiologist or neurologist, and urban place of residence (Table 1; S4 Appendix). In the model, our conclusions did not change after application of the Bonferroni correction.

In the secondary analysis, the 31,614 “non-switchers” were categorized into 28,745 patients who remained on warfarin after dabigatran was listed by the ODBP and 2,869 patients who transitioned to dabigatran in the subsequent 6 months. We found no statistically significant association between income quintile and switching to dabigatran after the drug’s addition to the provincial formulary (8.9% switched in the highest vs. 9.7% in the lowest income quintile, p = 0.085 and adjusted OR 0.93 {95% CI 0.82 to 1.05) among individuals who had remained on warfarin prior to dabigatran’s addition to the provincial formulary (Table 2; S3 and S5 Appendix).

The majority of “switchers” (N = 1942, 61.0%) and “late switchers” (N = 2056, 71.7%) received the lower 110 mg dose of dabigatran rather than the 150 mg dose.

In this population-based study spanning four years, we found that individuals of higher SES receiving warfarin for NVAF were more likely to switch to dabigatran in the period between its market approval and inclusion on the provincial formulary. Importantly, this gradient was eliminated after the drug became available through public reimbursement. These findings accord with a recent retrospective claims analysis that found that those with an enriched health plan were more likely to use dabigatran, and with other studies that suggest SES can influence medication use.[27–32]

While the finding that SES might influence access to uninsured medications is not surprising, our study characterizes the magnitude of this effect in a publicly funded system. Patients living in the highest income quintile neighborhoods were 50 percent more likely to use dabigatran than those living in the lowest income quintile neighborhoods.

Access to specialist care is affected by patient sociodemographic and economic status.[33, 34] In our study, the observation that patients under the care of cardiologists or neurologists were more likely to receive dabigatran is not surprising because of the marketing of direct oral anticoagulants (DOACs) by the pharmaceutical industry to specialist physicians, as well as the likelihood that specialists were more familiar with early data supporting the use of dabigatran in NVAF, compared with general practitioners.[27] Previous studies have found that specialists are more likely than general practitioners to prescribe new drugs. [35, 36]

Although debate exists with regard to its place relative to other DOACs, dabigatran has been shown to be a cost-effective thromboprophylactic strategy relative to warfarin.[37–42] The low number of “switchers” to dabigatran, whether early or late, was not unexpected. We believe that this is reflective of uncertainty regarding the use of dabigatran for NVAF and is in keeping with the cautious early uptake of dabigatran identified in the American Outcomes Registry for Better Informed Treatment of Atrial Fibrillation (ORBIT-AF) study [21]. The higher proportion of individuals in both groups who received lower dose dabigatran likely indicates physician vigilance with dabigatran use in our older study patient population[21].Nonetheless, our findings support the need to ensure up-to-date cost-effectiveness reviews of new drugs, and a timely mechanism to incorporate the results of those reviews into formulary decision making, to prevent disadvantaged access among less affluent patients.

Strengths of our study include its large population-based sample, and the high accuracy of administrative databases for determining drug use and the diagnosis of atrial fibrillation coupled with concomitant anticoagulant use.We obtained concordant results from two different analyses, the Cochran-Armitage trend test and logistic regression, enhancing the robustness of our findings.

This study has several limitations. First, we lacked access to claims data for dabigatran prior to its inclusion on the provincial formulary. We address this limitation by using a strict definition of continuous warfarin use and documentation of dabigatran use in “switchers” after its public formulary listing. Second, neighborhood income was used as a surrogate for patient SES because the administrative databases do not have measures of income at the level of the individual patient. However, this approach has previously been validated as a measure of household income and has been used in many other population-based studies.[12, 13, 43] Third, we could not exclude some patients who were excluded from the RE-LY study due to the constraints of the provincial databases. Therefore, selection bias could have influenced the association found between anticoagulant selection and the absence of renal disease since those who may have had an absolute contraindication to dabigatran (i.e. creatinine clearance < 30 ml/min) were included in the study. However, this does not affect the main analysis assessing the association between SES and prescription of dabigatran. Fourth, we restricted this study to dabigatran because the other DOACs were either not yet approved for use in Canada or not listed on the Ontario formulary at the time of analysis. Finally, we did not have prescription data on younger Ontarians; however, NVAF is principally a disorder of older patients.

Our findings likely apply to many other medications that represent good value for money, so delays in third-party medication coverage collectively have much larger impacts than we have demonstrated for dabigatran alone.[44] While we believe a thorough assessment of a drug’s cost-effectiveness is critical to the sustainability of a healthcare system, we also recognize a need to make this process as efficient as possible to avoid unnecessary delays in patient access to truly effective and cost-effective therapies. Our findings should be a stimulus for Canadian publicly funded drug plans to review their processes to ensure high levels of efficiency, especially removing any duplication in reviews of cost-effectiveness conducted by the national and provincial review panels, expediting price negotiation and ensuring that decisions are made as soon as all the necessary information is available.

In conclusion, we documented socioeconomic inequality in access to dabigatran among NVAF patients receiving warfarin. This finding supports the need to update public formularies in a timely manner based upon high quality evidence about cost-effectiveness. We recognize that a drug can be cost-effective while at the same time associated with considerable costs to payers, either because the drug itself is expensive or because of its use by many patients. However, equity is an important principle in health care and should influence reimbursement policy so that the poor are not required to pay more for prescription drugs as a proportion of their income compared to the rich.[1, 44–47]