Seeking that (hypodermic) needle in a haystack

Editor's note: Chris Lee is president of Chicago research firm MedQuery Inc.

As the world has evolved to communication via the Internet and e-mail, the job of recruiting market research respondents has also evolved. Pre-registered market research panel members for every category conceivable have made recruiting for studies quicker and more predictable. Unfortunately it has its limits, particularly when rare types of respondents are required.

The medical industry is one of the most complex and challenging categories for which to recruit market research respondents. Due to highly specialized medical professionals and oftentimes extremely rare patient conditions, it is not as simple as blasting out e-mails to a panel to fill your quotas.

The panels of recruiting companies are filled with the more common medical professional titles and patient conditions because this is where the money is for them. Less-common specialties and patient conditions often need to be recruited via phone, support groups and other more challenging methods.

Even with a sizable panel available, the screening requirements may be significant enough that you find out after the job starts that you need to spend more money and find another way to recruit.

Even worse, you may think a panel is available and the screener is reasonable, only to find out that the condition is only treated by a small subspecialty of physicians and you now have no chance of reaching the client’s quota.

I can’t tell you how many times a client has started a conversation by saying something such as “We just need doctors who treat this very rare medical condition and, by the way, they need to be published authors, work only in large group practices at a prestigious hospital, treat 1,000 patients per month, have green eyes and be able to juggle with one hand … AND we have never had trouble in the past recruiting them.”

Look out! There go your profits falling off a cliff…

Should you just walk away from a difficult project like this? What if this is a key client you have been trying to work with for a long time? Do you take a chance and just hope for the best?

Is it even possible to feel comfortable bidding on this type of medical project? How do you know when a job is worth it and when to take a pass?

So let’s assume you don’t yet have a good understanding of the disease, the medical professionals involved, the percentage of respondents who will qualify for the project and whether an existing panel can achieve the quotas needed. There is no way you should take on this job – or is there?

Several key methods

Guaranteeing project success in these situations is impossible but there are definitely several key methods to make sure you: understand what you are getting yourself into; set client expectations accordingly and; most importantly, price the project to ensure it is profitable. 

Research. The first and most important task is to learn more about the disease. Two minutes of Googling can uncover a great deal about it. I always start by Googling the name of the disease and incidence per 100,000. For example, “hemophilia incidence per 100,000.” This is a standard method the medical community uses to gauge the disease incidence in the general population.

Any time the condition is less than 100 per 100,000 you are dealing with a rare disease. Anything under 20 per 100,000 is extremely rare. Now, here is the interesting part. I find that this number strongly correlates to the highest number of patient recruits you can achieve for the disease. For example, myocardial infarction (heart attack) has an incidence of 274 per 100,000. Check around with recruiting companies and the range of recruits they can achieve is typically 250-300. On the other hand, hemophilia has an incidence of 20 per 100,000. Guess what: Most of their panels can’t support quotas higher than 20. This gives you a starting point to what is possible.

Understanding physician specialties. Often the specialty of the physician is far less obvious than you think. For example, you have a big asthma study so you think you will just target pulmonologists (lung specialists), right? Yes, but nearly every intensive-care unit physician is also a trained pulmonologist. It doesn’t seem obvious that a doctor working in an ICU is a target for this study but many also have private practices where they practice only pulmonology.

Using the hemophilia example again, hematologists are the specialty that would treat a rare blood disorder like hemophilia but this is a somewhat common example of a rare disease that has special treatment centers designed specifically for this disease. There are only 134 hemophilia centers nationwide and about 277 physicians total who regularly treat hemophilia – roughly 12 percent of all the hematologists in the U.S. So targeting hematologists in general would be an ineffective strategy. In fact, you could easily find yourself way below your target quota. It is always fun to tell a client you won’t get close to meeting their quotas, right?

So how do you avoid these types of mistakes? The American Medical Association maintains a list of all specialties and subspecialties along with counts of how many there are in the U.S. You can easily Google to find a copy of this information. A copy of this list is an invaluable reference tool!

Googling “who treats asthma or hemophilia” or variations of this can also be helpful but typically won’t uncover the fact that critical-care medicine or hemophilia specialists are what you need. I find instead that if you Google “hemophilia treatment center” you will then be able to find some Web sites of the actual centers so that you can review which physicians are managing the disease and how the disease is managed in more detail. By reviewing a few of these centers’ sites, it becomes much less likely that you will overlook the actual specialty or subspecialty of the physician you need. You might even find out how many centers treat the disease nationally.

Calculating screener incidence. Of all the details that can make a good (read: profitable) job go bad, it is incorrectly gauging how many people will qualify for the project. If the client’s statement “We never have had trouble with this in the past” can send a chill down your spine, a client saying “Just estimate 70 percent will qualify and we can see how it goes” can be even more dangerous. I have rarely seen the percent of respondents that qualify to a study to be higher than expected. It is usually lower – sometimes much lower; sometimes non-existent.

The challenge of course is that you often find this issue out after a week of significant recruiting effort and by that time the schedule is running out of time, your client is frustrated and you look like an amateur.

So how do you solve this issue? Unfortunately this one can’t be solved but it can be managed. It is impossible to be an expert on all the nuances of medical professionals and those they treat. I have worked in medical market research for 20 years now and still am learning new facts. For example, off the top of your head, guess the answer to this common screening question: What percent of physicians are hospital-based? Drumroll please: 10 percent.

I am guessing you overestimated, maybe by a lot. How could this be? Well, first, the term “hospital-based” is not very well-defined. Do we mean their office is in the hospital? That they are employed by the hospital? Or, do they just need to spend a lot of time at the hospital? The definition that physicians typically use (the one that counts here) is whether they are employed by the hospital. The definition your client has may be something very different.

The point with this example is that screener questions are sometimes not as simple as they appear. Taking a few extra minutes to think through how the question might be misinterpreted can save you a lot of headaches (and money).

Another screener question that can cause a lot of issues is the one in which the patient has to be taking a certain medication or the physician has to be prescribing a certain volume of a medication. Clients generally provide a good guess based on market share figures to predict this incidence percentage but this has the possibility of being quite a bit different than the actual percentage who screen in to the study. Even worse, the client doesn’t know and just asks you to price assuming “most will qualify.”

The first and most important strategy here is to create some caveats. For example, you can price this one for, say, a 70 percent qualifying incidence but point out contingency plans. If the incidence falls below 50 percent, this means we might have to increase the cost-per-recruit by $25. This won’t help you avoid an unpleasant conversation if this in fact does happen but you have set the client’s expectation (hopefully!) so that this scenario should not be a surprise to them.

The second strategy here is to do your own research. Although it is not easy to find actual figures of how many people use the medication, you can usually find out how many people have the condition in question and you often can determine whether the medication in question is the market leader, a newer drug or an older drug falling out of favor. This information can help you roughly estimate how much volume exists for the drug. Not a perfect solution but it is likely to help get you in the right ballpark for the estimate.

Make a huge difference

All of these suggestions should only take minutes at the point of bidding or launching a new job but the time and cost savings these tips will provide should make a huge difference to you and your client. So the next time a client overestimates how easy a difficult medical recruit will be, hopefully you’ll be armed with the information you need to make your difficult medical recruit easy.