Our study showed that regular use of aspirin in people with newly diagnosed type 2 diabetes over their lifetime cost $8,801 per QALY gained. Medicare and Medicaid programs cover many drugs and medical technologies with much higher ICERs than our estimated ICERs for aspirin (20
). If we use the conventional $50,000/QALY as the threshold for cost-effectiveness, aspirin use is a very cost-effective intervention for people with newly diagnosed diabetes. Both one-way and probabilistic sensitivity analyses showed that the baseline results were robust. All the ICERs in one-way sensitivity analyses were below $25,000/QALY. There is near certainty that the ICER of aspirin use for primary prevention is <$30,000 per QALY, whether parameters come from the ATT study or the De Berardis et al. meta-analysis.
Our study reported higher ICERs of aspirin use for primary prevention of CVD in people with diabetes than in members of the general population who have a similar risk for CHD or stroke. For example, Pignone et al. (5
) reported that aspirin use was cost saving in men with a 10-year risk of CHD of 25%; our study reported an ICER of $5,752/QALY for men with slightly higher risk of CHD (28%). For women, Pignone et al. (6
) reported an ICER of $2,532/QALY in women with a 10-year stroke risk of 5.5% and predicted an even lower ICER for women with higher risk for stroke. In our study, the ICER was $13,833/QALY for women with an average 10-year risk of 8.7% for stroke.
Some have postulated that aspirin use might be cost saving in people with diabetes (21
). Our results did not support this. There are three possible reasons. First, aspirin's effect on gastrointestinal bleeding increased the total medical costs of the group taking aspirin. Second, the aspirin treatment group lived longer and required additional resources for treatment of diabetes and hypertension. Our study results showed that the treatment cost for glycemic and hypertension control was $1,411 higher in the group taking aspirin than the group not taking aspirin (). Third, aspirin treatment affects diabetes macrovascular complications but not microvascular ones. Those receiving aspirin therapy live longer due to reduced risk of CHD and therefore are more likely to develop diabetic microvascular complications later in life. We found that the cumulative incidence of other diabetes complications (i.e., end-stage renal disease, nephropathy, blindness, and lower-extremity amputations) at year 10 were similar in both the aspirin and nonaspirin groups. However, the lifetime cumulative incidence of these complications was higher in the aspirin group, indicating a total excess risk of 0.87% in men and 0.66% in women. However, our results are consistent with results previously reported by Gaspoz et al. (21
) and Kahn et al. (22
) on the cost-effectiveness of aspirin use in secondary prevention of CVD.
Our study has several limitations. First, the effectiveness of aspirin use for people with diabetes needs to be further demonstrated by large clinical trials. The difference in the effects of aspirin used as primary prevention of CVD in people with and without diabetes is debated (23
). The recently published Prevention of Progression of Arterial Disease and Diabetes Trial in the U.K. and the Japanese Primary Prevention of Atherosclerosis With Aspirin for Diabetes trials (24
) reported that aspirin did not reduce the risk of cardiovascular events. However, questions about aspirin effectiveness remain, and two additional trials are underway to address the issue of aspirin use for CVD prevention in people with diabetes (14
). Our cost-effectiveness analysis of the aspirin therapy is subject to change, pending results from ongoing clinical trials. Second, although our cost-effectiveness model was validated against clinical trials, it, like all models, is based on simplifying assumptions. However, we have tried to make our assumptions transparent and performed sensitivity analyses to show the impact of our assumptions.
Our study shows that regular aspirin use appears to be very cost-effective in people aged ≥40 years with newly diagnosed type 2 diabetes. However, use of aspirin in primary prevention is still controversial. Future clinical trials are needed to better understand if aspirin is efficacious for people with type 2 diabetes. Additional cost-effectiveness analyses, accounting for these studies, might be needed.