What GlaxoSmithKline Learned From Their Digital Trial of Rheumatoid Arthritis
By Allison Proffitt
March 8, 2018 The Clinical Trials Transformation Initiative (CTTI) conducted a survey of potential research participants as part of its Mobile Clinical Trials program. Hassan Kadhim, a business consultant for clinical operations at Boehringer Ingelheim and a member of CTTI, presented the findings of the Stakeholders survey at the Summit for Clinical Ops Executives or SCOPE* last month.
CTTI surveyed potential research participants to understand patients’ interest in, preferences for, and concerns with using mobile technology in clinical research. CTTI recruited participants with ResearchMatch and a cohort of 193 respondents answered a 92-question survey.
The study presented potential trial participants with two options: a traditional trial with 13 site visits in a year and a battery of tests conducted at the sites, or a “mobile trial” with three site visits in a year, and most data collected via a wrist-worn health monitor and smart phone.
Half the group heard about the traditional trial first, the other half heard about the mobile trial first. After each trial was explained, participants were asked if they’d like to participate. 80% of participants said they would take part in the mobile trial as outlined; only 51% agreed that they would take part in the traditional trial. When given the option to choose the trial of their choice, 76% said they would prefer the mobile trial.
Rubber, Meet Road
With more than three quarters of potential trial participants preferring mobile trials, it would seem, then, that mobile trials are the way of the future. But in practice, engagement is more complicated.
To great fanfare, GlaxoSmithKline and its partners launched the Patient Rheumatoid Arthritis Data from the Real World (PARADE) Study using a customized Apple ResearchKit App in July 2016. The goal was to investigate the feasibility of using a mobile app to recruit and enroll patients into a study and to gain insights about Rheumatoid Arthritis (RA) in a real-world setting.
PARADE was a site-less study, explained Michelle Crouthamel, with the Clinical Innovation & Digital Platforms Unit at GlaxoSmithKline. GSK recruited participants using social media and found an outpouring of patient support almost immediately: more than 400 patients consented to participate in the study in 30 days, she said. We built in a “pretty tedious” medical history questionnaire to help tease out non-RA patients from observers, Crouthamel added. 399 participants completed all demographic questions.
To benchmark the PARADE study, GlaxoSmithKline compared it to Corrona, an RA registry that collects data from both physicians and patients at the time of a clinical encounter. PARADE participants were a bit younger than CORRONA, and the participant population was slightly more diverse. The medications and symptoms reported by the PARADE participants were consistent with the medical literature, Crouthamel said.
She was confident that PARADE had recruited the right patient population. And the cost was unbeatable: the per patient cost was under $100; the entire study cost less than $500,000, Crouthamel said.
Participants were asked to provide demographics, comorbidities, medications, and RA symptoms including pain, fatigue, mood and morning stiffness via the app. Additionally, over the course of 12 weeks, participants were asked to complete validated questionnaires routinely used in RA clinical trials. All data were self-reported and collected via their iPhone.
But after 12 weeks, only 50 of the PARADE study participants were using the app.
Finding What Doesn’t Work
Attrition is not a problem unique to PARADE, Crouthamel explained. Retention in health studies conducted using smartphones is down everywhere, she said, citing work by Dorsey et al. in Academic Medicine (DOI: 10.1097/ACM.0000000000001205). And although the retention rates were disappointing, the PARADE study has been a wealth of information about what does and does not impact participant engagement.
GSK divided participants into two cohorts to study how data return affects engagement. Cohort A received their data every day; Cohort B received no data. Cohort A did retain participants slightly better, Crouthamel said, but neither group did well. Instead, she says, more app features are associated with increased engagement. For an unrelated Parkinson’s Disease trial, a webinar helped boost engagement.
Some participants are simply more inclined to be engaged, Crouthamel said. Women and sicker patients are more altruistic and tend to stay engaged. But for all participants, “convenience is insufficient” to keep patients in trials, she concluded, even if more than 70% of participants say they prefer a digital trial.
We need to align our “why” with theirs, Crouthamel said. Why do people want to participate? How can we give them what they need to stay engaged? Compensation and gamification both help, she said. And for mobile or digital trials, over-recruiting will be necessary to balance attrition.
Armed with new understanding and context, GlaxoSmithKline, through the ViiV Healthcare entity GSK started with Pfizer, has launched The Real World Insights from People Living with HIV Shared Through Electronic Devices (RISE) study.
RISE is to be a 3,000-person trial looking at the real-world needs of people living with chronic HIV, beyond clinical measures like viral load. In less than 20 days, RISE has already enrolled more than 500 participants, Crouthamel said.
RISE will address some of the weaknesses of the PARADE study while maintaining a site-less design, Crouthamel explained. mProve, a US-based mobile technology solutions provider for the clinical research industry, is conducting the trial. Study participants will complete a series of 7 validated surveys within the app either in a single sitting or over the course of 7 days. No study visits or treatments will be given as part of the study.
GSK considered adding a telemedicine component, Crouthamel said, but “we are wondering if we can make this truly digital” while still keeping participants and gathering valuable data.
As a first step, the RISE study is offering participants compensation. Study surveys will take approximately 1 hour to complete and participants will receive up to $35 in Amazon gift credit as compensation for their time after completing the surveys.
It’s an important step forward, Crouthamel believes. “For patients’ benefit, this new approach really needs nurturing and encouragement if we want to ultimately change our healthcare to orbit around patients above all others.”
* The Summit for Clinical Ops Executives, SCOPE. February 12-15, 2018; Orlando, Fla. SCOPE is produced by Cambridge Healthtech Institute, the parent company of Clinical Informatics News.