The spiritual “Dem Dry Bones” with catchy lines like “Hip bone connected to the back bone” provides an overly simplistic overview of the human anatomy from head to toe. The song’s broad appeal is likely based, in part, on our shared anatomy and associated common experience.
But payers know that when it comes to the safety and efficacy of a drug, device, or test (which, collectively, we will refer to as medical technology), spotting differences in outcomes is based on controlling for differences between treatment and comparison groups. Random assignment of patients between treatment and comparison groups, in theory, controls for these differences. But outcomes are also a function of other differences across nations at the patient, provider, and system levels — differences that are neither neatly controlled for in global clinical trials that pool patients across disparate geographies nor adjusted for when using ex-US clinical trial outcomes to predict US real-world outcomes. The lack of controls and adjustments for national differences is a basis for payers discounting or flat-out ignoring ex-US data.
Patients
Globally, patient populations differ in myriad ways. There are basic demographic differences, such as age or gender, that make some medical technologies more, or less, relevant. Other kinds of differences, like socioeconomic status or health coverage status, can make some technologies more, or less, accessible. Issues that impact global economic development — think hunger and extreme poverty, or infectious diseases like malaria, polio, and HIV — can also impact health outcomes. Some populations or subgroups are at increased risk for certain genetic disorders, such as sickle cell and Tay-Sachs, or are more susceptible to developing certain conditions, like breast cancer. Patient populations vary in terms of comorbidities, use of other treatments, and drug metabolism, and all of these too, in turn, can impact outcomes.
Cultural differences may be associated with lifestyle behaviors, such as diet or tobacco use, that can impact outcomes. Cultural differences may affect care-seeking behavior, resulting in overtreatment, undertreatment, or delayed treatment, all of which can result in poor outcomes. After treatment is initiated, cultural differences can influence adherence to a treatment regimen. Cultural differences may also impact whether, or the extent to which, patients perceive and report outcomes like pain or depression.
Providers
There are international and regional differences in how providers use medical technology — and as Wennberg and others have documented, there are even wide and unexplained variations in resource input, utilization of services, and expenditures among neighboring communities! There are geographic differences in the provider workforce regarding training, size, and distribution. In Bangladesh, for instance, doctors receive only 5 hours of training about diabetes. And now, with the diabetes epidemic rampaging across Asia, there’s a severe shortage of GPs who know what to do about it. There are cultural differences in how much time is spent with patients and how that time is spent. Further, providers around the globe have differing levels of access to medical technologies.
Even when clinical protocols attempt to standardize the way providers use the technology being assessed or manage the specific condition being studied, there may be uncontrolled or less-controlled differences in how other drugs, devices, and tests are used and how other comorbidities are managed.
Systems
Beyond the patients and providers and patient–provider interactions, differences in how care is organized, delivered, recorded, and tracked can make a difference. To what extent do differences in information systems, health records management, or surveillance systems affect outcomes? In the United States, sophisticated care- and disease-management programs may move the needle for certain outcomes, but their effects are apples to the oranges in countries that lack a basic infrastructure for administering or distributing vaccines or that are more focused on infectious disease threats than management of chronic disease.
What on Earth?
Across nations, we see variances in patients and providers; utilization of immunizations, diagnostic exams, surgical procedures, and prescription drugs; and morbidity and mortality rates. So, what doesn’t vary between nations? And to what extent do these variables — measured or not — impact outcomes and our assessments of efficacy and safety?
Here’s the rub: The tighter the clinical trial protocol, including how medical-technology use is standardized and how inclusion and exclusion criteria are employed, the more comparable the outcomes across nations — apparently comparable enough to win FDA approval. Ironically, this greater control moves the results further and further away from the real-world evidence that payers really, really want. Yes, it’s a Catch-22 for manufacturers.
If payers are looking for an easy reason not to prefer a new technology, then ex-US data is an easy objection. On the other hand, payers could choose, conveniently, to embrace ex-US data if it supports their worldview or interests. That said, payers’ scrutiny of ex-US data depends on the context, including perceived need and cost of the technology. Further, manufacturers should remember that payers may question the generalizability of trial results to their populations, even when the trial is based in the United States.
If manufacturers are having difficulties recruiting a sufficient number of patients because the condition is rare or there are too few treatment-naïve patients, then they should look abroad and, of course, proceed in an ethical manner. If, however, manufacturers are conducting international clinical trials simply with an eye toward saving time or money, then they should also think about the potential costs associated with US payers devaluing those clinical trial results.
As per the FDA, the purpose of Phase 3 studies is to demonstrate whether a product offers a treatment benefit to a specific population (emphasis added). But as we’ve discussed, global clinical trials are arguably not limited to a specific population and ex-US trials may not be generalizable — even if all patients have a hip bone and a back bone.
No Comments