Allan M. Green, MD,PhD,JD
Market behavior is more often a subject of belief than of clear fact. The success of Robert Putnam’s compelling study “Bowling Alone”, which chronicles the evolution of American social behavior over the last 75 years, owes a great debt to market surveys of consumer behavior and preference carried out by advertising agencies on behalf of various clients starting in the 1930’s. That data strongly suggests that shared social activity of all sorts in the United States has markedly declined over the past decades. People, it would seem, prefer to spend time alone with their televisions or PC’s than have dinner parties, play bridge, join fraternal organizations or participate in neighborhood gatherings.
Recent debate about the behavior of consumers of medical care has focused on the cost-sensitivity of health care decision-making. Plans are being floated to provide patients with more decision-making authority, despite their lack of technical understanding, in the belief that this might allow a more efficient weighing of alternatives for many medical issues. Such suggestions are generally based on the presumption that physicians are not sensitive to the costs of their decisions to the health care system.
As Bob Putnam demonstrated, market survey data may provide evidence of the extent of physician sensitivity to pharmaceutical pricing.
The Gerson Lehrman Group performs custom surveys of medical practitioners for drug development clients. Physician attitudes towards specific drug-associated adverse events may be better understood by such surveys. Perceptions of the clinical benefits demonstrated in FDA-reviewed clinical trials can be assessed relative to available therapeutic alternatives. Prescribing patterns can be better understood so that future education and advertising can be directed towards demonstrably more effective use. While drug labeling is closely circumscribed by the Food and Drug Administration, drug advertising—though still reviewed by the FDA-- has somewhat more scope for manufacturer discretion.
Such surveys have demonstrated that there is significant physician sensitivity to drug pricing when there are therapeutic alternatives. For example, a recent Gerson Lehrman study asked 25 cancer physicians about their attitudes towards a new anti-emetic which is intended to reduce the nausea and vomiting associated with certain anti-cancer drug therapies. Like several of the current leading antiemetics, the new agent exerts its effect by blocking serotonin action at the so-called 5HT3 receptors. However, the new drug has been shown to have a much longer serum half life than any of the other competing agents, suggesting that it may be effective after a single dose for delayed, as well as immediate, symptoms. FDA studies demonstrated some advantage over existing agents for chemotherapy that is normally associated with moderate symptoms, but not for chemotherapy that is associated with severe symptoms.
Physician assessment of the clinical data was such that, if the price of the new drug were the same as the current 5HT3 drug of choice, they would use it in about 45% of their patients. However, if the price of the new drug were 75% of their current drug of choice, they would use the agent in 60% of patients; that is, a 25% price advantage would lead to a 33% increase in utilization. On the other hand, if the new drug were priced 10% higher than the current drug of choice, it would be used in 32% of patients; in other words, a 10% price hike would lead to a 13% loss of market share, reflecting a 29% decreased in units sold. A 25% price premium over existing agents, would lead to a 48% loss of unit sales and a 23% utilization rate among eligible patients. Thus, at least for a drug with therapeutic competition, small price increases compared to current medications may lead to disproportionate loss of unit sales.
Interestingly, larger sales premiums do not a linear response in utilization. The survey showed that, if the new drug were 50% higher in price than the current medication, market share would only be cut by 20% to 25%, reflecting about a 45% decrease in unit sales. If the price of the new drug were double the price of current agents, physicians suggested that they still would use the drug in 20% of eligible patients; thus, a 100% price premium only leads to a 56% loss in unit sales. One may speculate that physician demand is relatively inelastic for a group of patients who they perceive would have unequivocal benefit from a newer agent. This result is not inconsistent with the earlier data showing relatively high demand elasticity for the new drug for the bulk of patients in whom added benefits of the new agent were uncertain.
Thus, oncology physicians appeared to act as patient advocates, both economically and clinically: they rejected the higher cost of a new drug in the absence of reasonable expectation of unique patient benefit, but they accepted a significantly higher price for the 20% of their patients whom they perceived would get significant unique benefit from the new agent.
A second physician survey by this group asked 25 academic cardiologists about a new agent for the treatment of chronic angina. The drug is not yet FDA approved, demonstrated only modest clinical improvement in exercise tolerance and caused an abnormal change in ECG response sometimes associated with abnormal heart rhythms and fainting. Given the modest clinical impact expected of this drug, physicans thought they might use such an agent in 2% to 10% of their eligible patients, depending upon price. At $60 per patient per week for the new medicine, it might be used in 10% of eligible patients. If the price were doubled to $120 per week, physicians felt that they would only offer the new agent to about 2% of eligible patients, an 80% loss of unit sales. Once again, physicians showed what economists call highly elastic demand---a strong connection between price and demand—for a drug which might have equivocal benefits for their patient population. The fact that 2% of patients might be considered for the drug, even at the highest price, suggests that physicians will disregard price to provide benefit for their difficult-to-treat patients who might not be controlled on currently available therapy.
One corollary of these findings is that investors in later stage clinical studies need to be cognizant that the robustness of clinical results—not just the achievement of primary endpoints in an FDA-mandated clinical trial—will be a prime determinant of investment value.
Such pricing studies are useful in developing credible market plans for new pharmaceuticals and support long-term financial forecasting. Drug pricing strategy clearly depends upon the clinical benefits demonstrated in clinical studies and the perceived advantages of new agents compared to their therapeutic alternatives.
Incidentally, such physician surveys also can contribute to the debate about whether health insurance, or its lack, affects the quality of care one receives. In the first survey discussed above, it was noted that there is one antiemetic agent on the market with a different mechanism of action from that of the 5 HT3 inhibitors. That agent is rather costly, but it is said to prevent delayed, severe nausea and vomiting. As one physician characterized this agent: “It has a unique role to treat patients who are at risk for delayed nausea and vomiting. It is efficacious for this group only. It is quite expensive.” Another physician noted that: the use of this agent ”definitely has an advantage over conventional 5HT3 therapy alone. However, it is only used in the non-Medicare populations since Medicare doesn’t pay $300 for it.” Think about it.