By the Numbers: Superior quality in health care research begins with panel enrollment

Jason Freeman is the CEO/founder of Enos Answers LLC. He can be reached at jason.freeman@enosanswers.com.

For the past 20 years, I have worked in the health care market research industry overseeing the global panel fielding of health care studies. One of the biggest challenges I’ve seen during that time is the need for improved detection and elimination of survey fraudsters attempting to enroll in paid survey panels.

Often, you will see and hear conversations about ways to safeguard the quality of primary research data collection. The proposed solutions include asking your online sampling provider the ESOMAR36 questions, optimizing survey design, adding traps, deploying joker checks, adding red-herring questions and purchasing the latest fraud detection technology. While these solutions are important safeguards in their respective areas, they only address the chronic symptoms of survey fraudsters, not the root cause.

One of the best pieces of advice I received in during my health care research career was, “Trust but verify.” If you are passionate about survey data quality, here are some of my panel research rules that will be thought-provoking topics for your next call with your sampling provider. 

Rule #1: Panel enrollment is by invitation only.

Do you allow wedding crashers?

The idea of attending a wedding uninvited became popular after the movie “Wedding Crashers.” These uninvited guests have no concern or respect. They simply arrive to take advantage of the free food and drinks. The best way to avoid such an alarming ruckus is to ensure only invited and desired guests know the time and location of the event. Similarly, the panel enrollment process should be by invitation only. 

It is alarming how many open-enrollment links with “click here to join/register” are posted on websites, social media and paid advertising venues. These links are a beacon for fraudulent activity and an unending supervision nightmare for those attempting to ensure no gaps, holes or security enrollment breaches occur. It is a critical misstep to publicly broadcast the details of how to register. And don’t engage in the flawed thinking that you have the most rigorous and secure vetting process. Eventually, panel enrollment will be compromised by persistent fraudsters who can deploy their arsenal of false registration attempts/hacks when presented with these open-door opportunities. Panels that break this rule can easily have their enrollment database filled with attacks from unverified registrations. Avoid this ticking time bomb by never allowing open enrollment!

Rule #2: Panel enrollment includes the detection and ongoing monitoring of the geolocation of invited and joined panel members.

Where are you located?

Many popular browsers, antivirus software and web applications offer the built-in benefits of virtual private networks and proxy servers (i.e., anonymous IP traffic). While these technologies protect the user's identity and are great tools for accessing internet content securely, fraudsters also use them to obscure and falsify their identity and location. With a simple click, anyone can pretend to be in any country in the world.

Anonymous IP traffic must be detected within panel registration and survey deployment and it must be part of ongoing effort to detect account breaches and intrusions. By implementing a strict policy, invited panelists who ask to join can be monitored against their known clinical practice area, panelist-provided registration data and survey entry IP location.

Rule #3: Panel enrollment requires the full collection of National Provider Identifier or license and deployment of deduplication software.

Have you been here before?

A few months ago, I read recent survey findings on household Internet-connected devices. I was surprised that, on average, there are 7.8 connected internet devices per household in the U.S. Ubiquitous internet access poses a challenge to panels that solely rely on deduplication software to detect multiple registrations and fraudulent activity. One can easily switch devices (i.e., work and home computers) and bypass any security detection system (IP, cookie and browser fingerprinting systems). Those who bypass such checks can easily register multiple accounts under different information.

Fortunately, the health care panel industry has professional license data that can be collected and verified. Collection of this datapoint type allows a secondary check to confirm registration identity and prevents duplication. Moreover, monitoring license data will alert system administrators to potential fraud activity (i.e., multiple registration requests under the same license). With license data collected, panelists can be checked for disbarment and license deactivation, which indicates death or retirement.

Rule #4: Panel enrollment checks the panelist-provided e-mail for fraud flags. Referrals to join should not be allowed. Follow Rule #1.

I have a coupon for that!

Back in the craze for coupon deals, I recall my spouse frantically calling me at work. There was a great coupon deal but the website limited users to only one coupon per e-mail address. She asked me to sign up to get the deal too. I then proceeded to ask my five coworkers in the room to sign up as well.

Panels are inherently targeted like coupon sites. Whether it’s to fight fraud activity or a zealous valid survey member looking for another deal, additional security measures should be deployed during registration.

According to recent statistics, 40% of fraudsters use a newly created e-mail. Security checks such as tumbling (sequentially named e-mail addresses or multiple variations of the same e-mail) and first activity dates (the date any e-mail becomes active) are great ways to detect malicious enrollees.

In addition, beware of panels using referral systems, like asking registered panelists to refer another health care friend to join. They may be referring the same person to register multiple times without multi-factored measures in place to detect duplication.

Rule #5: The panel payment process ensures all prior rules are followed.

What is the getaway plan?

For most panel fraudsters, the end goal is to make a clean getaway. As with a bank robbery, it would be silly not to add a tracking device, dye pouch or marked bills to the loot bag but that’s effectively what happens when panel payment catalogs include anonymous payment options to @handles, e-cards like Amazon or unverified e-mail payments. Rather than make a payment to a Venmo handle, the payment should be made to a collected phone number associated with the panelist’s Venmo account. The collected phone number can then be validated (i.e., reverse phone ownership lookup).

Panels that focus more attention on their robust payment catalog than on secure payment tracking methods will always miss a prime moment to ensure all prior rules are followed. This is the best place to catch a crafty fraudster. For example, sending a paper check allows you to catch those who enter a payment name different (Rule #1) than the one they used to register. In addition, the address location can be verified again (Rule #2) and the address location can be checked to ensure uniqueness across other panel records (Rule #3).

Rule #6: Follow up with your panel provider!

Trust but verify!

We end where we started. It would be incorrect to avoid asking verifying questions and assume safeguards are foolproof. For me, it has uncovered numerous opportunities to prove what I already knew to be true and uncover new understandings I did not know even existed. 

Good rules

The elimination of survey fraudsters in paid research begins with panels following good enrollment rules. Let’s improve the health care data collection industry by ensuring protections are in place. Together, we can eliminate survey fraudsters.