This article was originally published magazines you know.
Death from colorectal cancer can be prevented with regular screening. Controlling high blood pressure could extend the lives of the approximately half a million Americans who die from the disease each year. Vaccination helps prevent tetanus, which can be fatal.
It’s clear that preventive medicine can make a big difference to your health.
Yet most people don’t understand it preventive medicine It may save their lives. In fact, in 2015, only 8% of U.S. adults over the age of 35 received all recommended immunizations, cancer screenings, and other high-priority services.
Researchers looking to change this situation are renting pages from Facebook, Google and other tech companies. By quickly comparing small differences in how they communicate with patients (a process known as A/B testing), healthcare professionals can quickly learn what works and what doesn’t. . This approach has already yielded some practical improvements, but not everyone is convinced of its value.
Tech-oriented companies use A/B testing to determine marketing slogans, web page colors, and many other options. Randomization is key. This means that people are randomly assigned to view different versions to test. Does a larger “subscribe” button on your website get more clicks than a smaller one, and does his one headline in an article attract more readers than the other?
Leora Horwitz, a physician and health services researcher at New York University Langone Health, and her colleagues are using this technique, which they call rapid randomized controlled trials, to learn how to improve the delivery of health services. ) was adopted. Randomized controlled trials (RCTs) are widely used in medicine, usually to test new drugs or treatments for other diseases. For example, patients are randomly assigned to receive either a new drug or the current standard of care and then followed for months or years to assess whether the new drug works better. But these trials are time-consuming and expensive, partly because researchers have to recruit people willing to participate in medical experiments.
In contrast, rapid RCTs are not used to study new treatments and therefore do not need to recruit participants. Rather, Horwitz’s goal is to improve healthcare delivery through rapid trials that allow researchers to iterate and fine-tune healthcare delivery changes based on what they learn from each test.
“we Randomize what we do so that it can be evaluated quickly and accurately Whether what we’re doing is working,” said Horwitz, who wrote about the 2023 approach. Public Health Annual Review.
For example, Horwitz and colleagues wanted to find a way to get patients to book appointments to address gaps in care, or overdue preventive services. Due to the sheer number of patients, the clinic is unable to reach everyone by phone or through his online portal, which NYU Langone uses to communicate with patients. So the healthcare system needed to understand what kinds of reminders worked best.
In an A/B test, we divided patients with care gaps into two groups. Patients who have signed up for an online portal account and those who have not. Patients in each set were then classified into different groups based on their health care history. Based on past behavior, patients who were less likely to initiate an appointment on their own were classified as a high-risk group. Those who eventually booked themselves in the past were assigned to the lower-risk group.
In part of the test, thousands of patients without portal accounts were randomly assigned, and some patients were sent phone reminders while others did not. Patients who received a call made an appointment to address 6.2% of care gaps, compared to only 0.5% who did not.
In another part of the test, some patients with portal accounts received reminder messages through that channel, while others did not. Of those who received the message, she booked the service she needed in 13 percent, compared to 1.1 percent who did not hear back.
Importantly, the study found that telephone reminders were the most effective way to contact subgroups of patients who were at high risk and least likely to receive preventive services without nudging. is.Shortly after the test results became available, New York University Langone Prioritized all patients at highest risk You can now receive phone reminders and have greatly expanded your ability to send messages through the patient portal.
“When we learn something, we immediately apply it to all our messages,” Horwitz says. This allows what they learn to quickly spread to tens of thousands of people. “That makes me happy.”
Many of the female patients at the medical center are now being tested, thanks to A/B testing at New York University Langone. short A message prompting you to book a mammogram appointment. Researchers used a rapid RCT to test reminder text sent through an online portal. Do short messages yield better results?In fact, patients who received a 78-word reminder booked almost twice as many mammograms as those who received her previous 155-word message.
In another study, to find ways to increase vaccination rates among young children, Horwitz and her team compared one and two text reminders to parents with no text reminders at all. , turned to rapid randomized testing. Only two text reminders (one sent at 6pm and hers another two days after hers at noon) made a difference. Triple the number of scheduled bookings. Most of the appointments were made after the second text message, suggesting that this boosting reminder prompted parents to take action.
Although still new in medicine, the idea of rapid RCTs is gaining popularity. A research team (an economist, a doctor, and a public policy expert, none of whom belonged to Horwitz’s group) used this technique to learn how to increase their income. Use of preventive care services by black menwith the lowest life expectancy of any US demographic group.
They recruited more than 1,300 black men from barbershops and flea markets in the Oakland, California area, asked them to complete health questionnaires, and gave them coupons for free health checks. A pop-up clinic staffed by 14 black and non-black male physicians was set up to offer testing, and participating men were randomly assigned to black or non-black physicians. The results: Black men assigned to black doctors were more likely to get infected. diabetes testinfluenza vaccination, and other preventive services other than those assigned to non-Black physicians.
Some experts doubt that rapid A/B testing will become commonplace in healthcare. Dr. Darren DeWalt, director of the Institute for Quality Improvement in Medicine at the University of North Carolina, likes the concept, but perhaps because people tend not to support randomization, even if it means improving quality of care. We believe most medical institutions will avoid this concept for ethical reasons. Context for innocuous things like appointment reminders. “People in this country don’t like the idea of being randomly assigned to something, even something as simple as that,” says DeWalt. “There are many doubts about researchers in the medical field.”
Some people criticize A/B testing for playing around with its limits. Pierre Barker, chief scientific officer at Boston’s nonprofit Institute for Better Health Care, says that significant improvements in health care delivery require a thorough analysis of the problem to be solved, which requires systems to: We believe that many changes may be required. In contrast, rapid randomized controlled trials are not broad efforts to understand why patients are not receiving preventive services and what can be done to change them, but rather single, discrete changes, such as Focuses on words used in phone scripts.
“It’s more about how fast you can move than how big the impact is,” he says. From rapid randomized controlled trials, “I’m not yet convinced that we can get much more than small incremental changes.”
While it’s true that the new reminders didn’t close most of the care gaps at New York University of Langone, Horwitz said the test provided information that led to the implementation of hundreds of potential life-saving services. . This is what persuades her that the medical industry should adopt rapid randomized trials.
“If you work for a web company or an airline or any other industry, you would naturally randomize, and that’s standard practice,” she says. “But it is still foreign in medicine, and it shouldn’t be.”
