Cutting Cost by Cutting Unnecessary Testing: Zika in Puerto Rico
While there is an urgent need to address the unsustainable cost of healthcare in the United States and Puerto Rico, rather than relying solely on cuts to services and payments, Dr. Berwick proposes another way in his article "Eliminating Waste in US Health Care."1 Specifically, he recommends identifying and eliminating any procedures, tests, medications, or services that don’t benefit the patient. From an evidence-based medicine standpoint, this sounds like a no-brainer. If the tenants of EBM are providing the highest quality of care based on the best available evidence that it’s effective (and safe), cutting extra interventions that don’t contribute to this should not only save money but lead to improved outcomes. So let's look at a real-world example of Zika virus in Puerto Rico.
An Epidemiology Commentary
Cutting Cost by Cutting Unnecessary Testing: Zika in Puerto Rico
One way of exploring and illustrating this concept is by considering a principle of epidemiology: the positive predictive value of a test. Take a test that claims to have a very high sensitivity (true positive) and also high specificity (true negative). It seems logical to conclude it is an excellent test overall and a clinician should use it without hesitation. The trick is to consider how prevalent the disease is in the population. If a disease is rare, even tests with high sensitivity and specificity aren’t helpful because of the low positive predictive value– you’ll end up with a lot of false positives and it won’t benefit your decision-making or the patient.
This is an especially tricky calculation with Zika because of the trade-off between an infection that is likely mild or asymptomatic and of low prevalence, but with the terribly high consequences of microcephaly or spontaneous abortion. In 2016 the first Zika-related death in the US was actually in Puerto
Rico (a 70M)2. According to the Pan American Health Organization (PAHO), unlike other arboviruses such as dengue or chikungunya, Zika exhibits significantly lower incidence with only 27,000 cases during the same [2023] period3. If we estimate a prevalence of 0.1% (very low) with the test having a positive predictive value of 10%, should an individual test positive there is a 90% chance it’s false. What is the cost of a false positive? Anxiety? Additional testing? Interventions for this person who does not have the disease? Puerto Rico has recorded Zika outbreaks but they’ve been sporadic and even now the prevalence remains low. I don’t presume to know the “right” answer to this, but is continuing to test an inefficient allocation of resources? Perhaps being judicious about who is tested (targeting pregnant women who have the highest burden of consequences) would use both evidence-based approaches plus waste prevention.
Take Away
Does this contribute to the “administrative complexity” Dr. Berwick mentioned? Probably. But isn’t it worth it to take the time to educate providers as well as patients as this also increases health literacy? In short, the example of test usage of Zika in Puerto Rico with a low prevalence illustrates how even a highly sensitive and specific test may not be the perfect answer and can be an opportunity for both waste prevention and patient education.
References
Berwick DM, Hackbarth AD. Eliminating waste in US health care. JAMA. 2012;307(14):1513-1516. doi 10.1001/jama.2012.362
Branswell, Helen. First US death linked to Zika reported from Puerto Rico. STAT Health. April 29, 2016. Accessed October 2023. www.statnews.com/2016/04/29/zika.
Zika: a silent virus requiring enhanced surveillance and control. Pan American Health Organization. September 1, 2023. Accessed October 2023. www.paho.org/en/news/1-9-2023-zika-silent-virus.