Yes.
Bradley Shapiro, an MIT economics Ph.D. and an assistant professor of marketing at the University of Chicago’s Booth School of Business, has completed a study of direct to consumer advertising (DTCA) of antidepressants. He’s also written a summary of his research for the Cato Institute.
Here’s the issue. I’m sure you’ve noticed how much drug companies advertise their prescription drugs directly to consumers. You know those ads. They’re the ones that tell you what ailment the drug is meant to address and then show you happy people enjoying life while telling you that the side effects may include suicidal impulses, acne, bleeding, etc.
My casual impression is that DTCA has increased in the last decade. But my first big surprise is that actually the trend has gone the other way. Shapiro writes:
Total DTCA of prescription drugs, while significant, has decreased from about $3 billion in 2004 to a little more than $2 billion in 2012. Meanwhile, antidepressant DTCA makes up an important fraction of total DTCA and has increased from about $200 million in 2004 to a peak of about $400 million in 2011, declining to about $300 million in 2012.
Of course, it’s possible that it has increased a lot since 2012. I don’t know.
How do we evaluate such spending? Shapiro nicely details the tradeoffs. On the one hand, people who can benefit from an antidepressant may be motivated to contact the doctor and get him/her to prescribe. On the other hand, most people who do so won’t face the full cost of the drug but, instead, will typically pay a small co-pay (a fixed amount) or co-insurance (a percentage of the price.) So people who value it just more than their share of the price will get it and insurers (which means the rest of us in our premiums) will pay the bulk of the cost. Shapiro actually has a more complete list of costs and benefits. I’ll leave you to read them in his short summary.
Shapiro notes that the subjective benefits are virtually impossible to estimate. But what he can do is measure the effect on labor supply. Here’s his key paragraph:
I find that DTCA induces more patients to be prescribed antidepressants with an elasticity of about 0.031, leading to a direct cost of DTCA to consumers and insurers. Second, I find that conditional on being prescribed in the previous month, current advertising reduces refill prescriptions by a small amount. Next, I find evidence against DTCA having an economically meaningful effect on either the price or the copay of the drug, conditional on prescription. I also find evidence against an economically meaningful effect of advertising on the generic penetration rate. Finally, I find that DTCA causes benefits in the form of increased labor supply. The benefits of increased labor supply outweigh the total cost of additional prescriptions by more than an order of magnitude. The preferred estimates imply that a 10 percent increase in DTCA brings $769.5 million in wage benefits while generating $32.4 million in prescription costs. This finding implies that on average, DTCA is generating prescriptions that are worth more to patients than they cost. In other words, the average DTCA marginal prescription is not a “mistake.”
How does he measure the effect on labor supply? By looking at the number of days of work missed. Go to his longer paper, pp. 14-15, if you want to see more.
I would note two cautions. In one way he overstates costs and in the other way, he overstates benefits.
On the cost side, the amount spent by insurance companies and paid by the rest of us is not pure cost. A large part of it is a gain to the drug company. And even if that doesn’t make you click your heels with delight, that gain to the drug company will, in the longer run, give them an incentive to research and develop new drugs.
On the benefit side, the $796.5 million is not pure benefit. It would be so only if the value of leisure were zero. On the other hand, if you’re missing work because you’re depressed, the value of leisure may well be close to zero.
Bottom line: if Shapiro did his econometrics well–and I bet he did–direct to consumer advertising of antidepressants produces a net gain for the economy.
HT2 Jeffrey Miron of Cato.
READER COMMENTS
The Original CC
Jun 29 2018 at 7:15am
Should that read “that gain to the drug company”?
David R Henderson
Jun 29 2018 at 11:59am
Thanks. Good catch. Will change when I get to my computer.
Alan Goldhammer
Jun 29 2018 at 7:56am
I think the conclusions of the study are all too facile. The modern class of antidepressants are extremely useful but any study that looks at prescriptions without also including therapy as a co-variant is flawed IMO. Depression also runs on a spectrum from mild to severe and there are likely many people taking meds who don’t really need them.
I looked at the SSRN manuscript and did not see any reference about the decrease in DTCA spending. It would be better not to focus on $$ spent but rather on total air time for commercials or print advertising pages which would be far better surrogates (I don’t know how easy it would be to gather such data). It may be that ad costs have come down and quantity has gone up. It would also be nice if there were a linkage to on-patent vs off-patent SSRI antidepressants and this data is easily obtainable. High use of off-patent drugs will be economically beneficial in terms of cost control.
Looking at economic outcomes in the absence of examining the underlying clinical condition is just an example of correlation not being causation.
Justin
Jun 30 2018 at 12:24pm
This completely ignores the empirical strategy of the paper. He didn’t just find a correlation between advertising and the economic outcomes. He found variation in advertising exposure which was as good as random, which created a treatment and a control group. When you have that kind of experiment, you don’t need to include other relevant covariates for the statistics to work out correctly – doing so would increase the precision of the estimates, but it is not strictly necessary.
If you want to make these sorts of criticisms convincingly, I think you need to explain why being just slightly on one side of the television market border is correlated with the other factors which you believe drive the results.
David Henderson
Jun 30 2018 at 8:38pm
Nicely said, Justin. Thanks.
Brad Shapiro
Jul 4 2018 at 1:44pm
Thank you for that, Justin! Well said.
I’d also note that therapy is one possible channel through which these ads could cause increases in labor supply (as a result, you wouldn’t want to “control” for it). The entire effect need not be coming from the drug itself. I have an extended discussion in the manuscript of the potential mechanisms. I simply do not have enough statistical power to pin down the exact share of each mechanism cleanly.
For on vs. off-patent, it is addressed in the paper. I find a precise null effect of advertising on the percentage of prescriptions which are generic. That is, I find evidence that advertising isn’t driving patients disproportionately to expensive brands.
The choice of whether or not to get therapy is a very interesting one. In the data, relatively few people get therapy at all, and that is not unique to this data. Cronin et al. (2017) have a nice paper trying to explain why that’s the case, given that therapy has very good clinical evidence of effectiveness: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2896946
robc
Jun 29 2018 at 9:19am
David,
I think you violated Betteridge’s Law with the first word of the article.
David R Henderson
Jun 29 2018 at 11:57am
Yes I did. On purpose.
Ed Hanson
Jul 2 2018 at 10:55am
I have no complaints about economists justifying their existence with sophisticated econometrics. Someday it may have some value.
But here is another way coming to the same qualitative answer. Just realize that advertising people are some of the smartest and manipulative alive. And those who hire them are very much bottom line oriented. If an advertising campaign has lasted these many years, than those who know the bottom line are certain of the value. A quicker way to YES.
Ed
Alexander Hamilton
Jul 2 2018 at 2:29pm
Yet again the anti-economics post is the worst reply. The purpose of the study was not to determine whether DTCA was worthwhile for the drug company but to look for evidence that DTCA was a net benefit to the economy.
Comments are closed.