A digital health intervention that allowed patients to self-order colorectal cancer (CRC) screening tests doubled the proportion of patients who underwent screening, a randomized trial shows.
“Two components of [Mobile Patient Technology for Health–CRC (mPATH-CRC)] directly encourage screening orders: the decision aid, which increases patients’ intention to receive screening, and the ability to self-order tests, which decreases barriers to order entry,” write David P. Miller Jr, MD, from Wake Forest School of Medicine, Winston-Salem, North Carolina, and colleagues. They report their findings in an article published online today in the Annals of Internal Medicine.
In an accompanying editorial, however, experts applaud the increase in screening rates, but caution that it is hard to know what aspect of the trial drove the large increase.
The researchers note that there is broad support for CRC screening, but that more than one third of Americans remain unscreened. To increase uptake, Miller and colleagues enrolled 450 patients who were due for routine CRC screening and who were scheduled for a visit with their primary care provider in six community-based primary care practices.
They randomly assigned patients to arrive early to their primary care visit and receive either the iPad-based mPATH-CRC (n = 223) intervention or a control (n = 227) intervention.
The mPATH-CRC intervention comprised an 8.6-minute decision aid with information about fecal testing (guaiac-based fecal occult blood testing and fecal immunochemical testing) and colonoscopy, which the patient could self-order. The program then requested a cell phone number or email address so automated follow-up electronic messages tailored to the chosen testing could be sent to the patient to assist them with completing the screening procedure.
The control intervention comprised a 4.3-minute video on diet and exercise, with no opportunity to self-order testing or receive electronic messages.
All participants completed a telephone survey 24 weeks after receiving the intervention.
Those who received the mPATH-CRC intervention were twice as likely as those in the control group to complete a chart-verified screening test within 24 weeks of enrollment (30% vs 15%; difference, 15 percentage points; 95% confidence interval [CI], 7 – 23 percentage points), which was the primary outcome for the trial.
After adjusting for clinic, the odds ratio for completing screening was 2.5 (95% CI, 1.6 – 4.0) in favor of the digital health intervention, with high consistency across clinics (P for interaction = .82).
Colorectal neoplasia detection rates were 2.6% in the control group (six adenomas) and 7.2% in the intervention group (15 adenomas and 1 carcinoma). The difference (4.5 percentage points; 95% CI, 0.6 – 8.5 percentage points) between groups resulted from the increased screening among those in the mPATH-CRC group.
The researchers assessed secondary outcomes at the 24-week postprogram survey. Almost all (97%) of those in the mPATH-CRC group could identify a screening preference compared with 71% of those in the control group (difference, 26 percentage points; 95% CI, 19 – 32 percentage points). Participants in the intervention group were more likely than participants in the control group to prefer fecal testing (53% vs 28%; difference, 26 percentage points; 95% CI, 17 – 34 percentage points).
The proportion of participants who preferred colonoscopies was similar among the groups (43% in the mPATH-CRC group vs 42% in the control group).
More participants in the mPATH-CRC group anticipated being screened within the next 6 months (62% vs 49%; difference, 13 percentage points; 95% CI, 3 – 22 percentage points), and more of those in the mPATH-CRC group reported discussing screening with their primary care provider immediately after receiving the intervention (76% vs 48%; difference, 28 percentage points; 95% CI, 19 – 37 percentage points).
More participants in the intervention group ordered screening tests (69% vs 32%; difference, 37 percentage points; 95% CI, 28 – 45 percentage points). Fecal blood tests and colonoscopies were both ordered more often by those who received the mPATH intervention.
Completion of an ordered colonoscopy was higher than completion of an ordered fecal test among all participants (mPATH-CRC, 61% vs 26%; control, 53% vs 32%).
“This difference could indicate an increased commitment to screening among patients choosing colonoscopy, or it could reflect the routine telephone reminders that patients receive before colonoscopy appointments,” the researchers write.
The most common reasons for not completing a fecal test were losing or not having the test kit (22%), not wanting to complete the kit or thinking it was embarrassing or distasteful (13%), forgetting to complete the kit (11%), and trouble following the required dietary or medication restrictions (11%).
Among those in the intervention group, 118 of 223 participants used the program to order screening tests; of those, 81% signed up to receive follow-up email or text messages to assist them with completing screening.
Primary test providers changed self-orders from fecal tests to colonoscopies for 11% of participants who ordered fecal tests and changed self-orders from colonoscopies to fecal tests for 9% of participants who ordered colonoscopies.
Missed Opportunities?
“Despite its success, this trial missed at least 2 opportunities that might have made it even more transformative,” Shivan J. Mehta, MD, from the Center for Health Care Innovation at the University of Pennsylvania, and David A. Asch, MD, from the Center for Health Care Innovation at the University of Pennsylvania and Corporal Michael J. Crescenz VA Medical Center, all in Philadelphia, write in the editorial.
They say it was not necessary to use a clinic visit to introduce the intervention. “In the end, the goal is not to get more primary care patients in the clinic screened but to get more people screened — a goal that invites the use of smartphones, social networks, and other elements of daily life. Broader conceptions of population health and new financing models help unshackle us from visits and clinics as the denominators of our approaches,” they explain.
Also, Mehta and Asch say, only 70% of the 640 eligible patients agreed to participate in the trial. “The trial meant to answer, ‘Is the intervention better than usual care?’ But the consent requirements meant that the trial answered, ‘Is the intervention better than usual care in a population that has already agreed to participate in a research program designed to answer that question?'” they write.
“This difference rarely affects drug trials but is a major risk in trials of behavioral interventions where patients who consent do not represent those who decline,” Mehta and Asch continue. “Why couldn’t eligible patients simply have been randomly assigned to usual care or this new protocol to see what happened without requiring anyone to opt in (or not opt out) in the first place?”
Meanwhile, the researchers point to the need to build on the current success. “Although screening increased substantially in this study that included many persons with low income and limited health literacy, approximately half of patients did not complete their ordered tests. Incorporating more strategies to help patients complete their tests could further increase the effectiveness of mPATH-CRC. Future research should identify methods for implementing digital health interventions like mPATH-CRC into clinical care.”
The authors and Mehta have disclosed no relevant financial relationships. Asch reports receiving personal fees and other from VAL Health outside the submitted work.
Ann Intern Med. Published online March 12, 2018.
For more news, join us on Facebook and Twitter
Tidak ada komentar:
Posting Komentar