Patient recruitment automation

Patient recruitment is a major step in a clinical studies and typically tends to be a bottleneck. For drugs that are targeted for large consumption (more than a million prescriptions a year), around 5000 patients are needed to clinically test the drug. FDA requires such high numbers to assess all the data gathered and asertain the clinical utility and safety of the drug. If the sample size of patients is too small then FDA might place restrictions on the drugs’ use in Phase IV (Drug marketing phase).

Finding and enrolling study subjects delays 60% of trials by at least one month and 13% of trials by over six months (Sinackevich, Nick, et al.; “Speeding the Critical Path,” Applied Clinical Trials, January 2004). Slow enrollment is the number one reason for the failure of clinical trials. The process of getting candidates for a clinical trial involves many steps. From putting up ads about the study to getting volunteers, answering their concerns, filtering the right set of volunteers based on the study protocol and keeping them informed and retaining them during the study. Some steps of the process can be automated and some steps need manual intervention. But in many clinical trials being run, the automation piece is completely missing. Every step of the process is manual.

For example, San francisco based HIV organization that runs ads on BART for HIV cure volunteers ( does exactly this ! They ask volunteers to call (415) 554 – 9068 to leave their name by spelling it twice and a number to reach them, and also a good time to call back !!Now this method of getting volunteers has a lot of issues. Someone from the clinical study office has to sift through all the voicemails left and note down all the names and times to call back. Voice mails can get cut off, mistakes can be made when noting down names and numbers etc. and so this list for all practical purpose could be missing important volunteer contact details. The other issue is information dissemination. If someone can answer questions and give precise answers then volunteers might be less averse to joining the clinical study program. Just by looking at an advertisement in the train, not many people are going to opt-in. They need answers and that too immediate ones to keep them informed about the study.

HummingBytes has been working on solving the patient recruitment process and have devised solutions to automate the process. Our products suite for patient recruitment consists of the following :

1. patientAnswers (Outreach and Answers solution)

Our  SMS based customer service product is a perfect and effective way to communicate with study volunteers. It starts off with asking volunteer to TEXT in their questions and concerns to a SMS shortcode. This question is queued to a web-interface where the CRO agents can reply to the questions from the web. This is a real-time conversation and not an email like conversation. As soon as the question comes it pops up on the screen of the CRO agent who can then immediately answer the question. The product has the capability to put multiple agents to answer questions, in case there is a flood of questions from volunteers. Just like a contact center handles incoming voice calls, SMS agents get to answer questions based on pre-built call distribution algorithms.

Once the volunteer is satisfied with their answers, he or she can be opted-in for the study immediately from the web interface. In case the volunteer wants to opt-in at a different time then they can do so using our study opt-in products. 

2. patientOptin  (Study Opt-in solution)

Study opt-in product lets a volunteer to opt-in to a clinical study using Voice, SMS or Email. All channels are opened up simultaneously so that the volunteer chooses the preferred mechanism and opts-in. Now the volunteer is ready to undergo a basic set of questions which is the inclusion / exclusion criteria of the study (the clinical study protocol)

3. patientFilter  (Inclusion / Exclusion criteria filtering solution)

Inclusion / Exclusion criteria can be a complex set of questions that determine if the volunteer is a viable study candidate. Finding a man in their 40s whose has had a heart attack in the past and undergoing no treatment currently can be tedious for a manual process. But an automated process can find the right set of volunteers using this inclusion / exclusion criteria. The automated engine makes calls to all the opted-in volunteers and asks them questions setup for the study. Then based on the responses and the rules of the study protocol volunteers are included or excluded. The system is smart enough to stop asking questions as soon as it finds someone is excluded in a study.

All the rules that govern the flow of the product is set up on the website and it is an easy to use the interface to create and build rules. Rules are built on the responses given by the volunteer.

Here is an example of a rule that is an exclusion criteria. This rule is set up on the second question which asks for the age of the volunteer.

If (Question2 Response < 40) Then end call

Rules can be complex or simple depending on the clinical study protocol. The system makes a call to every opted-in volunteer and poses them the question based on the rules. When someone is deemed unfit for the study the system immediately stops asking any more questions and thanks the volunteer for their time. Once all the volunteers are culled and the good candidate pool is created then based on the study protocol they will have to be reminded about doctor’s appointment or even remotely monitored.

4. patientReminders, patientMonitor, patientDiary   (Reminders, Remote monitoring and eDiary solution)

Volunteers of a study need to be frequently reminded about their medications or appointments. They also will have to be monitored remotely to see how well they are accepting the medication / placebo being given to them. The Reminders and Remote monitoring system is capable of sending scheduled reminders to volunteers via Voice, SMS or Email. It can even accept a response from the volunteer they remotely monitoring them.

Reminders can be sent to Site investigators and even patients. The content of the message might differ based on the purpose of the alert, but the mechanism to alert is still the same and available via the same web-interface.

Each patient can also keep a diary of their daily activities, food consumption, exercise, their mood, sleep habits etc. This electronic patient diary (EPD)  is a record of the patients condition while under a clinical trial. This self-reporting mechanism is a patient reported outcome or (PRO) and can provide a dramatic improvement of the trial process over the conventional paper method of collecing data. Our EPD is a Voice & SMS based self-reporting system which can be accessed at any time from anywhere.

The above automation products from HummingBytes can make the patient recuitment process more efficient and digital friendly (since all information is kept up to date and is collected through automated processes). Visit for more information on our products and offerings.


Tags: , , , , , , , , , , , , ,

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: