How to recruit Veterans with PTSD for clinical research

Post-traumatic stress disorder (PTSD) can occur following any traumatic event: Violent personal assaults, natural or human-caused disasters, accidents, combat, and other forms of violence. The strategies in this article will focus on military veterans with PTSD but are also relevant for other demographics.

Wayturn often helps clinical research studies find veterans, or active duty personnel, with PTSD. We have experience recruiting for studies on Metformin, different kinds of non-medication therapy, and over-the-phone studies.

First, we will go through some basic considerations for promotional materials, then audience size estimates, and finally cost when recruiting participants with PTSD for clinical trials.

Promotional Materials

Wayturn recommends creating an excess of promotional material because it’s better to have too many options rather than be limited and have to resubmit extra material for IRB approval later. For digital advertising, variation is important because changing just one image can change the cost per enrolment wildly – even from $250 to $20. It is difficult to know without experience which images will work best. The key is to test.

Each research study is unique and the images below are only a starting point. The eligibility criteria, geographical, demographical and many other factors unique to your study will change what will work best.

Veterans with PTSD are often quite hard to enroll, depending on if they believe your research study will actually help them, and their comrades, or not. Compensation is important and can give an excuse for why they’re participating – it can be hard for them to admit to friends and family, or even to themselves that they might need help. They are quick to take offense, as are their families. It’s important to make them feel respected, understood and avoid language that makes them feel as if they’re being used as ‘guinea pigs’.

With this in mind, we will share the images and adcopy that has worked the best for us so far across all PTSD studies.

Images

Sad images are sometimes good at grabbing the attention of people who have experienced traumatic events, but often they instead get ignored or cause offense. Images of female veterans crying often get comments talking about sexism. When the sad angle works though, the images often get no comments at all, just silent clicks and questionnaire completions.

One important consideration is that most stock images feature models that break military uniform codes – resulting in the veterans being offended, cursing at the ads, and not signing up. See this example of what NOT to use:

DO NOT USE this image. The uniform and the hair are both non-compliant with military dress codes. Past and current military personnel will take offense and cause your cost per enrollment to become very high.

So the challenge becomes getting the attention from veterans – showing that the ad is for them – without uniforms. There are two good solutions. The first is using other symbols that people associate with the military at a glance – images of combat, dog tags, camouflage patterns, etc. The other solution is to mention PTSD or Military very visibly in the foreground of an image or in the title.

Wayturn has found the greatest success in using pensive images without faces that give a feeling of isolation – or the complete opposite with close up faces where grief is apparent. Both strategies get attention from the target audience and prime them for reading the headline and adcopy with increased interest.

Language

Most clinical research enrollment agencies send complete ads (image+headline+adcopy) to the ethics committee or IRB for approval. For best results, it’s smarter to structure your promotional material document to allow for many combinations of images and text options to be tested against each other to find what results in the cheapest cost per enrolment.

Even after hundreds of research studies, we still experiment as part of every recruitment process to find the most convincing combination of approved materials – as judged by cost per enrolment.

For PTSD, we typically keep the advertising copy short and clear to avoid causing a debate or offense in the comment section. A hypothetical question grabs people’s attention and if they’re part of the target audience they are often interested enough to hear the answer that they click and read more about the study.

So the ad could look something like this:

We always test longer versions too but usually, the short and concise language wins. Detailed information about the study is better placed on the landing page. The ad on Facebook, or any other platform, is used to generate attention and visits to the landing page – not explain every detail of the study.

Targeting & Audience Size

Deciding who to show the ad to, and estimating how many people there are in the target audience, is a key step in estimating a cost per enrolment. It also influences how many enrollments are likely to be found in a given geographical location, as well as the speed they can be found at.

We usually start by estimating a total audience size given all eligibility criteria, no matter what digital platform we think we can find them on.

There are 18.2 million veterans in the US according a 2018 census. Out of them, around 15% (depending on which war they served during) are diagnosed with PTSD. That gives us roughly 2.7 million veterans with PTSD in the whole United States.

But to estimate a cost per enrolment, we need to figure out where we can reach these veterans. So a second audience size analysis is performed for each platform. Let’s go through Facebook and Instagram as an example.

Note that an actual advertising campaign for a PTSD research study might involve any number of additional platforms, but Facebook is a good example because most clinical research coordinators have some experience with the platform.

Social Media – Facebook & Instagram

If we target everyone who identifies as a veteran in all of the USA on Facebook and Instagram using demographic targeting, we end up with 760,000 people.

Is this a reasonable number?

No, it’s way too low. All veterans will not be on Facebook or Instagram, but we expect more than 0.8 million out of 18.2 million to be. The reason is this: Demographic targeting on Facebook is always smaller than the actual number of people in that demographic on Facebook. That sounds confusing. But think about it this way – not everyone will list in their Facebook profile that they are a veteran.

A way to reach more veterans is to target them by what they actually talk about and share on the platform – what groups they are part of, what topics they discuss in their posts.

Here are some broad veteran interests:

To solve this, we

However, logically more people than just veterans are talking about the above topics. And including all of them gives us an audience of 30 Million, which is clearly too broad compared to our rough estimate of 18.2 Million veterans without PTSD.

To solve this, we need to show the ad only to people with the above interests AND broad US army interests. Imagine it like a Venn diagram: People who have both A and B. Or people who talk about A, B but not C. The result is a more targeted audience who are very active in these topics – not just someone whose uncle is a veteran.

This gives an estimated 5 million people, which seems more reasonable than 30M or 0.8M. So how do we make these 5M veterans more likely to be diagnosed with PTSD – which roughly 15% of them should be?

One option that usually works really well is to make another intersection by targeting all of the above PLUS an interest in therapy or certain medication-related interests. This brings the audience down to 0.5-2M depending on how targeted we want it to be. I would expect this targeting to do reasonably well, but with testing, you could find something even better for your particular study.

Further geographical targeting, and age, still need to be applied. Some common ineligibility criteria like BMI, smoking and gender can also be accounted for quite easily.

If your research study needs active military duty members, and veterans are ineligible, Facebook is still a good option despite one large drawback. Facebook does not allow excluding people with Veteran interests (because of discrimination laws that probably shouldn’t apply to research, but do anyway). One way around this is to use Lookalike audiences – but that is beyond the scope of this article. Contact us for advice.

All of these targeting options are just the beginning of the recruitment efforts. A period of testing different audiences will let us determine which results in the cheapest, and fastest, enrollments for your study. It’s almost never straightforward, but always very satisfying once a solution is found.

Database email campaigns

Veterans are quite commonly targeted or excluded in all military-related research studies. Especially in America, we have a decently sized database of veterans with and without a PTSD diagnosis.

Database emails are often activated gradually throughout the recruitment period to let potential participants trickle in. We avoid sending emails to the whole database at once because research studies generally have a fixed capacity for phone screening appointments. This is however up for discussion in the setup phase of the recruitment plan if you choose to let Wayturn help you with recruitment.

PTSD – Cost per enrolment

The final cost per enrolment will depend a lot on your particular research study. However, in general, PTSD as a condition is quite hard to enroll.

An estimate, given our experience across all our PTSD studies, puts the investment in the range of $40 – $500 per enrolled participant. For a more detailed cost analysis of your research study, contact us and we will help you get started.