Pharmacoeconomic Model of Serotonin Reuptake Inhibitors: CREDIBILITY OF THE MODEL
The Sullivan model appears credible, the participants said, in that the mathematical equations did not seem to have been forced into the service of a desired result. Participants felt that this model seems to have been built on conservative, common-sense assumptions that are easily understood. When the user of a model understands that the model is driven by realistic assumptions, the model begins to make sense conceptually instead of seeming to be just a “numbers game.”
Some participants indicated that they had previously seen similar decision trees that were constructed for cyclo-oxygenase-2 (COX-2) inhibitors, osteoporosis drugs, and anti-depressants. The depression model incorporated rates of partial effectiveness, which led to additional visits for titration of the original agent or the addition of a second agent— followed by reduced rates of compliance owing to poly-pharmacy. Such a model is complicated, but it does take into account real-life issues.
cialis soft tablets
One participant stated that the high rate of side effects leading to a medical intervention in the Sullivan model is surprising, because the only SRI-related side effects that seem to concern physicians are sexual ADRs. Patients complain about sexual dysfunction, but not about gastrointestinal side effects.
A pharmacy consultant questioned the model’s reliance on varying ADR rates. Considerable patient variability is seen among these drugs, that participant said, and it is difficult to quantify the effect of ADRs. For example, diarrhea might develop, but it usually resolves within a week. Although diarrhea is an ADR, the panel wondered whether it is an ADR that leads to further spending by the health plan. In the real world, ADRs are not the only reason people switch or stop using an SRI. In its current form, the model is a “great academic exercise,” a participant said, but it would be surprising to see the model hold up when health plan data are incorporated.
One difficulty in constructing a model on the basis of the incidence of ADRs reported in the prescribing information is that such an approach captures only the ADRs that occurred during the clinical trial phase. Broader ADR rates need to be measured, because once a drug is on the market, different ADR rates are observed. After a drug has been marketed, it also is likely that rare but costly ADRs will emerge, and their cost will eclipse that of the simple ADRs seen during the clinical trial phase of drug development.
buy antibiotics online without prescription
A participant pointed out that instead of relying on ADR data from U.S. prescribing information, such data could be acquired from European countries, where the reporting of ADRs is required by regulatory bodies, in contrast to the voluntary reporting of ADRs in the U.S. The model’s authors considered European ADR data as a possible source of information, but the main problem with this approach is that the populations and treatment patterns can vary greatly from country to country. The regulatory environments, reporting methods, and incentives also can differ greatly, and ultimately, decision-makers in the U.S. rarely seem to want to rely on European economic data to make decisions in their populations. Therefore, U.S. data were used when available.
A participant suggested that although the cost of ADRs might be quantifiable in a large population, it is difficult for a smaller health plan to examine its data and to attribute an adverse event to a drug. Another participant, however, explained that the Sullivan model was based on a study that did just that—it examined the electronic medical records from a Texas health plan to find the association between SRI-related ADRs and physician visits.
“SILOS”
Regardless of how credible and useful a pharmacoeconomic model may seem, the reality is that in many health plans, the pharmaceutical costs still are widely “siloed”; that is, they are considered separately from medical costs. A participant pointed out that if pharmacy managers are under pressure to work within a given pharmacy budget, the potential of a drug to affect costs outside the pharmacy silo will receive only minimal consideration. Thus, the nature of an organization’s business model is a factor in determining whether a pharmaco-economic model such as the Sullivan model would be of interest to a P&T committee.
canadian antibiotics
In a system where claims for outpatient and inpatient visits are carved out separately, this particular model would not work, the participants said. It might have some potential for justifying pharmacy expenses, but contracts that are negotiated in silos do not allow for the overall lowering of expenditures.
CONCLUSION
The Sullivan pharmacoeconomic model of the SRI class was generally well received by managed care officials with experience in the P&T decision-making process. The extent to which the model might be used by a P&T committee depends, first, on whether the committee gives consideration to pharmacoeconomic arguments. If so, the potential utility of the model next depends on whether the model can be customized to reflect the organization’s demographic and utilization profiles, and whether it can be updated to reflect changes in the marketplace.
