April 4th, 2018

What the ISCHEMIA Controversy Tells Us About Patient Recruitment in Clinical Trials


Patient recruitment difficulties in a key NIH study demonstrate widespread need for improved enrollment pipelines.

Beginning in July of 2012, the NIH-funded ISCHEMIA trial aimed to test approaches to treating stable ischemic heart disease (SIHD). By the end of the recruitment period, however, it was clear that the trial would not reach its stated goal of recruiting 8,000 patients – and that’s when things got complicated.

Worried that subpar enrollment would nullify the study results, investigators made changes to the protocol to include more patients in the trial – at the very last minute. Even as they’ve faced increasing scrutiny, the ISCHEMIA trial’s investigators have defended their decision as a necessary adaptation to patient recruitment pitfalls beyond their control.

Although the controversy surrounding the ISCHEMIA trial will likely continue unabated for some time, it does offer valuable lessons for sponsors, CROs, and investigators hoping to design effective clinical trials – and to enroll a critical number of patients in the process.

The ISCHEMIA Trial’s Fatal Flaw: Patient Recruitment

The ISCHEMIA trial was designed to answer one of the most important questions in medicine: how best to treat stable ischemic heart disease (SIHD). Specifically, ISCHEMIA was designed to determine whether a conservative course of optimal medical therapy (OMT) – pharmacological intervention and lifestyle changes – could provide better outcomes than the aggressive approach: stenting combined with OMT.

As the January 31, 2018 enrollment deadline came closer, however, investigators encountered a recruitment problem. They had hoped to recruit 8,000 patients, but could only find around 5,000 patients who met the “hard” endpoint criteria of cardiovascular death and heart attacks. This hard endpoint was a major selling point of the original study design, but the lack of patients threatened to make the trial’s results insignificant.

As a result, ISCHEMIA was forced to – as critics put it – “move the goalposts” of the trial by including the soft endpoints of hospitalization for unstable angina or heart failure. For investigators, changing the criteria – even at the last minute – would still allow for actionable results.

But once the decision became public, dissenting cardiologists explained that the study’s conclusions would be significantly and negatively affected. Where “hard” endpoints such as cardiovascular death and heart attacks prevent bias – in that they are irreversible events – “soft” endpoints such as hospitalization for unstable angina or heart failure require diagnoses from physicians, opening the door to interpretation and bias.

Accordingly, critics have argued that the study will be irrevocably skewed, not to mention disappointingly limited given its initial promise and corresponding funding.

Strategizing for Effective Patient Recruitment

Because ISCHEMIA didn’t recruit enough patients from the outset, the trial was forced to make a decision that compromised the significance of the research. This isn’t for lack of trying; the study’s investigators used social media, speaking engagements, and video content to urge the cardiology community to refer their patients to the trial. But it’s clear that ISCHEMIA’s efforts failed to overcome systemic obstacles to greater recruitment.

Studies show that doctors and nurses feel comfortable discussing clinical trials with their patients, but only refer 0.2% of them to relevant studies. In addition to utilizing traditional recruitment methods, trials must invest in digital marketing channels that enable direct contact with potential patients. Social media advertising and paid search allow trials to precisely target patients or caregivers who are interested in or searching for clinical research.

Digital ads aren’t a replacement for conversations with practitioners, but they can help create a new recruitment pipeline that empowers patients to seek out clinical trials on their own.