Inside medicine's gray zone

It sounded like a miraculous technology when it was introduced in 1992. The wives of men with low to nonexistent sperm counts could get pregnant using a technique called intracytoplasmic sperm injection.

Usage exploded. It quickly replaced conventional in vitro fertilization as the technique of choice for thousands of couples, many of whom were approaching the end of their childbearing years. Use of the fertility procedure more than doubled between 1996 and 2012 to 76% of all couples seeking help, including many cases where the men did not even have low sperm counts.

While the approach added thousands of dollars in additional out-of-pocket medical expenses for patients, it didn't improve pregnancy, miscarriage and birth rate outcomes compared to the conventional method, according to a study comparing 15 years of data on the two procedures that was published last week in JAMA.

The significant increase of patients for whom the intracytoplasmic procedure wasn't indicated was a surprise. “It can be an effective treatment when there is male-factor infertility. But the problem relates to the expansion of its use, without evidence, in patients who do not have a diagnosis. This is not recommended,” said lead study author Sheree Boulet of the Centers for Disease Control and Prevention. Researchers in the study also noted: “Pregnancies resulting from the use of ICSI (intracytoplasmic sperm injection) have been associated with 1.5 to 4 times increased incidences of chromosomal abnormalities, imprinting disorders, autism, intellectual disabilities and birth defects compared with pregnancies resulting from conventional IVF.”

Technologies like ICSI exist in what researchers, cost-control advocates and consumer groups are beginning to call medicine's gray zone. Many drugs, devices and procedures may have been proven effective in the clinical trials that their developers submitted to the Food and Drug Administration. But after they are widely deployed by physicians and hospitals outside that carefully controlled trial environment, they don't achieve the same outcomes. Sometimes they perform worse than cheaper alternatives already on the market or do more harm to patients.

For years, complaints about the poor performance of pricey and sometimes dangerous new technologies were limited to consumer advocates. “It's an all too common scenario, and the motivation is ultimately money,” said Dr. Michael Carome, director of the Health Research Group for Public Citizen. “But it is not in the best interest of patients or public health.”

But now, their concerns are being echoed by mainstream healthcare leaders with a newfound concern for eliminating waste in the system—a concern that's being driven by payers and employers demanding greater value for their healthcare dollars.

“There are large gray zones in which an intervention is neither clearly effective nor clearly ineffective—zones where benefits are unknown or uncertain and value may depend on patients' preferences and available alternatives,” Amitabh Chandra of the Harvard Kennedy School, Dr. Dhruv Khullar of Massachusetts General Hospital, Boston, and Dr. Thomas H. Lee of Press Ganey Associates wrote in a perspective published this month in the New England Journal of Medicine. “Much healthcare occurs in these gray zones, which are expanding despite insights gained from comparative effectiveness research.”

They proposed wider adoption of several new payment policies to encourage greater scrutiny before physicians and patients use gray zone technologies. They include bundling provider payments for conditions such as unstable angina, prostate cancer and end-of-life care where the patient's deteriorating condition often involves deciding whether to use a questionable technology or procedures.

While bundling payments would give physicians incentives to stop offering treatments that confer minimal benefit, they wrote, it might also create incentives for physicians to “diagnose with a condition that triggers a bundle and then provide less than the standard of care.”

“The solution isn't necessarily more information. We need payment approaches that are designed to work in the gray zone,” said Lee, a cardiologist and chief medical officer at Press Ganey. He complained that comparative effectiveness studies have become “overwhelming—without resulting in a high-value health system” and they usually arrive well after new therapies have already peaked.

In the meantime, physicians remain stuck in a binary world where something is deemed right or wrong, and they make decisions they may not really support. But when their own money is on the line in a bundled-payment arrangement, “They can get tough-minded pretty quickly and make good decisions,” Lee said.

Increased patient cost-sharing for questionable gray zone technologies also can be used to discourage their use. Similar to the tiered payments used by pharmacy benefit managers to encourage greater use of generic drugs, patients can be charged higher copayments when comparative effectiveness or outcomes research has shown that a particular procedure may not be right for the patient or the best use of resources.

Researchers continue to find evidence that popular procedures aren't as effective in practice as they were in clinical trials. A study that appeared this month in JAMA Neurology found that Medicare beneficiaries who underwent carotid artery stenting between 2005 and 2009 had a 32% mortality rate during an average two-year follow-up. The authors of that report concluded that more evidence is needed to support the use of the procedure outside of randomized clinical trials.

Determining which treatment is best for which patient is rarely a simple black-or-white, right-or-wrong choice. But that's how it is treated in healthcare even though many new medical interventions fall into these gray zones, researchers say. Government, professional societies and other organizations all have joined the recent push for comparative-effectiveness studies to drive more precision and clearer choices.

The biggest issue, said Lynn Etheredge, director of the Rapid Learning Project based in Chevy Chase, Md., is not that gray zones exist, “but that we already have a lot of data that could help resolve the questions.” Useful information is often buried in unpublished clinical trial research or stored away in electronic health records, he said. His work focuses on using big data and EHRs to assess the effectiveness of medical treatments.

Several recent initiatives have aimed to tackle the problem by increasing accessibility to information about real- world outcomes. For example, last week, the American Society of Clinical Oncology announced that a few cancer centers have already signed up to participate in a big-data cancer quality initiative that will unlock previously inaccessible information on cancer care using millions of EHRs.

The Institute of Medicine outlined recommendations this month for major changes to how the results of medical studies are handled in the U.S., including timetables for the sharing of adverse-event summaries and analyzable data sets.

The Patient-Centered Outcomes Research Institute, a not-for-profit organization established in 2010 through the Patient Protection and Affordable Care Act, aims to identify information gaps that make it difficult for clinicians to make fully informed choices about available care options, said Dr. Joe Selby, PCORI's executive director. Since 2012, PCORI has awarded 360 grants totaling about $670 million to fund pilot projects and primary comparative-effectiveness studies.

In their NEJM editorial, Chandra, Khullar and Lee criticized PCORI-funded investigations for assuming that the gist of the problem is an inability to manage new technology rather than economics. Areas of medicine that have seen rapid innovation, such as cardiology, oncology and orthopedics, have also seen cost growth. PCORI should begin to evaluate pros and cons of payment approaches such as utilization review, bundled payments and capitation, they say.

Selby responded that PCORI's main focus is to fund research that can address lingering areas of uncertainty within the evidence. It would be interested in economic studies only if they focus specifically on how the incentives would influence treatment choices and outcomes.

Etheredge agreed: “Unless the science is clear, you can't come out with good payment policy. If you get the effectiveness right, insurers and everyone else can figure out the costs.”

Follow Sabriya Rice on Twitter: @sabriyarice