Mary Costello

Mary Costello

Mary Costello has spent over thirty years in the healthcare and clinical research fields, working in commercial leadership roles with large organizations like Covance and Thermo Fisher and more recently supporting the growth and expansion of eClinical Solutions and Elligo Health Research. She has served on several industry advisory groups and is a Founder and Board Member of BioAustin. At Medable she serves as Head of the Site and Investigator Network and is passionate about ensuring the site voice and experience helps shape improvements to the clinical research industry.

Recent posts by Mary Costello

2 min read

Bias and Decentralized Trials: How to Avoid Selection Bias Through Proactive Action

By Mary Costello on Apr 1, 2021 10:10:09 AM

Avoiding bias in clinical research has been an important concern of drug makers worldwide and is part of what guides oversight on every level.  As we move into an era of opportunity to expand clinical trial inclusion via the power of technology, we will also need to ensure that we don’t create the conditions for bias as a result of access to devices or wifi.

What do I mean by that statement?  Successful decentralized trials will rely heavily on a reliable internet signal capable of transmitting data on a regular basis.  Much of the new technology offered to support DCTs comes with the convenience of collecting data through patients’ laptops or phones, which provides an ease of participation that did not exist in traditional RCT models.  However, these interactions with patients presuppose that they will have phones capable of working with apps (iPhones or Androids), or that they own a laptop and that they have a consistent and reliable internet signal.

It is heartening to see a national conversation turn in the direction of diversity and outline guidance and the necessity of expanding clinical research access to a broader community.  But project planning that fails to account for the variety of scenarios that exist today – especially in some of our lower income communities -- will mean that by default, these same communities that are at the forefront of the conversation will continue to get left behind. To prevent this, project planning must proactively introduce ways of provisioning devices and ensuring access to a strong signal as a key part of widening the inclusion opportunities for all patients.

Sites must be at the forefront of this planning. They will need to understand what challenges may exist and be provided with the support and budget to remove any technical barriers that may prevent participation or continuation in a study. While this may seem like a fairly straight-forward task, there is a danger in not considering the breadth of challenges facing some patients. For example, consider the experience of one site where, at the height of COVID, participants in a trial without home access to adequate wifi were asked to travel to their local library, daily, to upload their diary details.  Yes, that was in the midst of a pandemic.  Realizing the difficulty in this, the site asked for hotspot support but the sponsor did not want to allow this additional request.  The result? Many of the participants dropped out, depleting the numbers and introducing systematic bias in the sample. 

I would encourage sites to begin collecting data that will help gauge the technical situation of their patients.  This could be as simple as emails, whether participants are able and willing to communicate via text, what type of phone they have, whether they have wifi at home, and, importantly,  asking them to rank their comfort level with technology.  Having this information at the ready will help with pre planning, device support, and other critical information that will help assess their ability to support the study while creating a budget to do so.  For research  sites with an associated healthcare practice,  EMRs may already have collected some of this information, providing them with a useful first step in planning.  

In summary: If we know that today’s trials have inherent bias, then we must make sure future trials include everyone, from all types of communities and all types of situations.


2 min read

Technology for Clinical Trials: What can be Learned From the COVID-19 Vaccine Roll Out?

By Mary Costello on Mar 23, 2021 2:20:30 PM

By Mary Costello and Sina Mossayeb

Good news here!

The US has already achieved its target of 100 million Americans vaccinated in 100 days.  Cities, towns and states across the country have pulled off an amazing effort supported by huge teams of volunteers and healthcare professionals who have enlisted stadiums,clinics,mobile transport and public buildings to accomplish their goal. Despite a wide range of processes for rolling out the vaccine, there is one thing they have in common —the registration for the vaccine itself has been anything but simple.

Even “tech centers” such as San Francisco, New York City ,Boston and Austin, where user experience is part of day to day consideration, faced frustrating interfaces for vaccine registrations.  Nonetheless, resilience prevailed, with many of the first wave of eligible patients turning to their children, relatives, and neighbors for support. Tech savvy volunteers on NextDoor, for example, posted offers to help guide often despondent patients through that cumbersome process.

So, what can we learn from this effort?  

Obviously, the strong desire to be vaccinated is enough motivation to ensure that the end user soldiers through. 

But what about our industry? Clinical trials can be an even trickier experience, and studies have shown that frustration in the user experience will cause patients to pause treatment, or give up on participation completely.

Good tech is grounded in human centered design, which considers not only the desired outcome, but also the human journey that defines the experience. This design begins with empathy for end users, considering potential internal or external barriers to entry, their hopes and fears. And, in the best case scenario, it involves co-creating the experience alongside those who will be using the product. The practice is often referenced to as user experience design. Tech as a proposed better solution should support and uphold the experience, not become an impediment or aversion to the goal. 

A well-planned trial considers all the risks to successful study execution prior to finalizing a protocol. To get good data, one might consider making the experience human centered so that engagement increases, by way of a deeper understanding of the people in the trial.  Considerations should include patients’ age, living situations, transportation barriers, abilities, mindsets around health and health providers, support systems and more. For example, are the subjects you are trying to include fearful of hospitals and doctors? Beyond demographics, how do financial difficulties become blockers for participating in a trial or even considering one on a day to day basis?

Remember the registration for vaccines that we started with at the outset of this post? What if the font sizes were bigger. What if less information was on each page, and allow for bite size digestion? What if terminology was made simpler? What if there was an audio or video option to walk seniors through the process? What if there was an ability to hit a button to have a volunteer give you a ride (if you didn’t have someone in your life to help you)? These aren’t just features, these are human centered considerations to create a better experience, which in turn drive results.

The silver lining to every major challenge we face is that those same situations also turn into opportunities to incorporate learnings and continue to improve a more effective process.  Just as we will emerge from the pandemic with a heightened awareness of clinical needs and settings around the world, we will also be in a position to improve public health roll outs for the future. Here’s to the next 100 million and to using human centered design as our starting point.

3 min read

Recommendations for Optimizing Televisits

By Mary Costello on Mar 9, 2021 11:09:40 AM

Like many of you, the past months for me have been learning about how to transfer what was my everyday normal life into online mediums.  Some have worked well, while others have fallen short. For instance, I liked virtual workout classes better than I would have expected. However,  a zoom or facetime call with my 86 year old father ends up feeling more sad than satisfying.  However, only this week did I finally experience my first health televisit, with my physician who wanted to discuss a medication change.  

Given that I have spent the last ten months working with sites interested and eager to introduce digital options into their clinical research realm, I felt this was an opportunity for me to observe how the process works in healthcare.

First of all – what worked?  

I was given instructions verbally over the phone by the nurse about three weeks prior on what to do, none of which I wrote down. But, I did remember the basic expectations! “Check that my registration with the on-line portal was up to date, download the app for the televisit and be ready ten to fifteen minutes prior to the appointment.”  

I was prompted the week before to confirm my visit via text. I did receive an email with a link to download the app, and I remembered my apple password without having to retrieve it, and my doctor called me exactly at the time of my appointment.  It was actually great! I know her, seeing her and having her undivided attention for twenty minutes with no waiting ahead of time or hassles of parking etc. went well!  If I had rated my satisfaction with the visit itself, I would have given it a ten, as I did not feel as though the level of care or attention was any way less than an in person visit would have been and the hassle factor was significantly lower.

What didn’t go as well? And, what were my takeaways when incorporating into research?  

I am not sure that the scheduling system / text reminder system was synchronized with the televisit.  I got a prompt on Sunday night reminding me about safe visit procedures on masking and ensuring I was COVID symptom free before coming to the office.  That was the first red flag.  Since I had also gotten the email asking me to download the app, I figured it was just what it turned out to be – two separate systems for communication, operating independently.  

On Monday morning, I saw a missed call from the doctor’s office.  When I returned it, the message said that due to inclement weather, the office was closed that day.  Austin had had its first snowstorm in about ten + years the night before.  There was no way to leave a message and if I chose to call their on call service for a non-emergency, the standard message warned me that I could incur a charge.  At this point, I am wondering whether my physician, since her office was closed, would expect me to reschedule and would take the day off.  

How did it go? 

She called right at the appointed time which had always stayed as a possibility for me.  The other communication ended up being distracting rather than downright annoying as we are still under Stage Five restrictions here. 

However, it did underscore for me that sites need to think about how they communicate and how they create automated reminders and confirmations.  Before implementing tech for a televisit, walk through how all your patient communications will go.  As it was my first televisit, it would have been quite helpful for the office to have a standard. In fact, it could look a bit like this. 

“How to prepare for your first televisit. Here is the link, here are the steps, here is a screenshot, etc.:

In essence, over communicate! 

If you can’t modify your reminder systems, then point out that a patient may receive a reminder making it seem as though they have an in person appointment, and advise them to ignore that message.

As we migrate to the new world order, we must remember that flexibility is here to stay and our goal should be to enhance that experience at every level! 

Topics: telemedicine TeleVisit
2 min read

Decentralized Shouldn't Mean Disorganized

By Mary Costello on Jan 15, 2021 9:30:06 AM

Decentralized clinical studies are referred to by a variety of names - virtual, digital,  - the list goes on. Yet, whatever you may call it matters not; instead it’s the core tenets of this idea that matter most. Here, the inherent patient-centricity of decentralized studies is both the goal and the reward.

For those who may not know, decentralized trials are executed through a combination of TeleHealth and traditoinal in-office settings. Instead of office visits, DCTs often include visits that may be conducted remotely, with patients at home for some or all of the trial. 

However, simply arming potential participants with access to or actual apps and devices is not a recipe for success. A clinical trial is, at minimum, an exercise in detailed planning and should always be subjected to a “what if” modeling to plan and to identify  what can be prevented or mitigated.

Thus, it comes as no surprise to veterans of clinical trial logistics that the unprecedented challenge of vaccinating waves of Americans with an ultra-cold product was going to be difficult. The decision on how to deploy was left up to states, with seemingly little planning for a central repository to pre-register or track patients nationally.

The goal was to have 20 million vaccinated by December 31st, 2020. The end result, according to Reuters? A much lower 2.8 million. 

The reason? According to multiple reports across Reuters, Newsweek, CNBC and others was simple - a failure in planning and communication.

While the Federal government was responsible for, and relatively successful in providing the allocated number of doses to States, progress halted there. A number of key oversights, such as leaving States to plan their own distribution to residents, as well as developing last minute contracts with retail vendors like Walgreens to administer vaccinations have slowed the pace.

As such, Americans have found themselves left in the dark, with little to no information on how to register for the vaccine, where to go, and most importantly, the timeline to vaccination?

 So, what’s the lesson to be learned here? That decentralized does not have to mean disorganized. 

Clinical trial leaders need to employ the same planning principles that have underwritten successful trial conduct for decades. Proper planning, communication, and room for flexibility are key for all trials - decentralized or traditionally managed. Thankfully, there are several examples in the clinical trial world that sponsors and CROs can look to in order to plan a proper decentralized trial.

New tools in the industry afford us new hope and new access for communities that have historically been denied a chance to participate in research. However, the imperative should be in making all aspects of the study as accessible as possible including thinking through how to communicate, register and prepare participants. 

Here in Texas, we are struggling to move into 1B. Houston, one of the largest and hardest hit areas of the US, has been overwhelmed with too few locations offering the vaccine. In fact, many rural Texas counties have not even received a shipment of product. Imagine if we had used the time we had from February until now to create an online registry for people to register, learn about the vaccines, identify accessible locations, and share confidential health information that would help pre-qualify and stratify vaccine recipients. We would be in a much better position today.

Never shortchange up front planning or underestimate the value of communication. It will be the foundation for success.


3 min read

Why I Joined Medable: a New Generation of Leaders

By Mary Costello on Oct 15, 2020 11:54:08 AM

Like many of you, I was saddened to learn of Ruth Bader Ginsberg’s passing a couple weeks ago. She served as an icon and inspiration to many. I think women who benefited from her work and her decisions, the loss felt almost personal. For me, spending most of my career as a single mother working in a very fast paced industry, I often felt as though in order to succeed professionally, I needed to minimize the impact of balancing motherhood with the demands of work.   

Times are changing and traditional roles are evolving. Now many fathers are sharing the responsibilities of children. But when I was coming up through the ranks of the workforce, that evolution had not yet taken place.

In my 30 plus years of working, every single leader of every single company for which I worked were men. I’ve had the good fortune to work with inspirational company heads and to learn from their styles and their leadership and vision.  

Today I find myself at Medable, and it’s the first time I have worked for a female CEO. So, is there a difference?  

My experience has been similar to those of many others. A recent article in Newsweek, focusing on the difference in responses between female led countries versus male led countries amidst COVID noted that "world leaders who are women have been more likely to use language focusing on compassion and social cohesion”. Forbes published an article last year examining the different ways genders are socialized and found that “women excel in leadership positions because they have developed soft skills necessary for effective leadership...traits like empathy, communication and listening". In 2019, HBR  evaluated 360 degree reviews and found that "women in leadership positions" were perceived just as - if not more - competent as their male counterparts". Yet the percentage of female CEOs of public companies still does not reflect the population.  

There is some evidence of progress. A woman just assumed the CEO position for the BNSF - the first female to hold the leadership position for a railroad. GlaxoSmithKline, one of the largest pharmaceutical companies in the world, is led by a woman.  


So what's been the difference for me, a woman in her late 50s, arguably closer to the final years of my career than I am to its beginning? Medable's culture is shaped by our leader, Michelle Longmire, a physician and founder who believes in the power of passion and drive.  But she also believes strongly in the power of people and connection and works to connect with her teams and collaborators in a very personal way.  

When I was contemplating leaving my former position to join Medable, I asked one of my colleagues for her thoughts on the best thing about Medable and her instant reply was "Michelle. She's inspirational". Despite the stereotype, women want to see one another succeed and it is inspiring to see how our CEO has taken an industry problem as a challenge and an opportunity and built a company around solving those problems. Her skills as a listener mean that the team is drawn to her and willing to share ideas; her openness to multiple perspectives feels different than what I've experienced at other organizations.

We mourn the passing of Justice Ginsberg – a petite, soft voiced judge who will be remembered partly for her belief that "you can disagree without being disagreeable."  Justice Ginsberg showed us that leaning into our traditionally female qualities wasn't limiting and may in fact be liberating.  I appreciate all she did for our country and, in particular, the women in this country.  Equally today I appreciate the opportunity to experience a new generation of female leadership.  

Thanks Michelle and the Medable team – it’s been a great journey so far!