There are precious few cost differences between the main tests used to diagnose coronary artery disease in chest pain patients, concludes a new study of more than 9,000 patients published online May 24 in the Annals of Internal Medicine. Diagnostic strategies certainly differed between the tests, but costs over the first 90 days and the first three years showed no significant differences, the authors wrote.
The group analyzed costs for treating chest pain from nearly 200 centers across the U.S., including the initial outpatient testing strategy for suspected coronary artery disease using either coronary CT angiography (CCTA) or functional stress testing. Over three years, the choice of diagnostic test had little effect on cost (Ann Intern Med, May 24, 2016).
"Costs varied a little bit depending on the diagnostic process but they were pretty close to a wash," lead author Dr. Daniel Mark from Duke University told AuntMinnie.com. "There was $200 to $300 extra associated with the CTA arm, but really after about 90 days there was no evidence that the choice of which test they used had any influence on the cost."
Best options unclear
Chest pain is a frequent and expensive cause of patients reporting to the emergency room. Recently, the Prospective Multicenter Imaging Study for Evaluation of Chest Pain (PROMISE) found that various diagnostic strategies being used for individuals with chest pain yielded little difference in patient outcomes.
When patients present with chest pain, doctors rely on patient history and noninvasive tests to assess for the possibility of coronary artery disease. However, evidence on the effectiveness of the various testing strategies has mostly been limited to observational studies comparing diagnostic accuracy or the prognostic value of the results. As a result, consensus is lacking with regard to the testing strategy that will produce the best outcomes at the lowest cost, Mark and colleagues wrote.
PROMISE, on whose data the present economic analysis was based, looked at the effect of different diagnostic strategies for coronary artery disease on patient outcomes. The study included 10,003 patients at 193 centers in the U.S. and Canada who were evaluated for coronary artery disease following presentation for chest pain. The patients were randomly assigned to either initial CCTA or functional stress testing, with the choice of functional testing left to the provider.
PROMISE showed that complications occurred in 3.3% of patients from the CCTA group and 3% of those in the functional testing group, a difference that was not statistically significant (p = 0.75). The individual components of the primary end point, combinations of components, and subgroup analyses did not differ significantly between the groups.
After excluding some PROMISE patients from their cost-effectiveness analysis, the Duke researchers were left with 9,649 patients.
Analyzing costs
Economic costs in the study were based on a U.S. perspective, and only fee-for-service patients from the U.S. healthcare system were included. Hospital-based costs were from 2014 data after application of charge-cost correction formulas. A cost-weighted formula was used for outpatient care, as there is no charge-cost conversion formula available for these costs.
The results showed a mean initial testing cost of $174 for an exercise electrocardiogram (ECG), $404 for CCTA, about $500 for pharmacologic and exercise stress echocardiography, $946 for exercise stress nuclear testing, and $1,132 for pharmacologic stress nuclear testing. Mean costs at 90 days were $2,494 for a CCTA strategy versus $2,240 for the functional strategy.
| Costs of CCTA vs. functional stress testing | |||
| Initial mean testing cost | Mean cost at 90 days | Mean cost at 3 years | |
| Exercise ECG | $174 | ||
| Exercise stress echo | $514 | ||
| Pharmacologic stress echo | $501 | ||
| Exercise stress nuclear | $946 | ||
| Pharmacologic stress nuclear | $1,132 | ||
| Total functional group cost | $2,240 | $6,856 | |
| CCTA cost | $404 | $2,494 | $7,213 |
Cost differences over time, all higher for the CCTA strategy, were $254 during the first 90 days, $99 for the first year, $26 for the second year, and $249 for the third year, the authors wrote. Three-year cumulative costs were $7,213 for the CCTA strategy and $6,586 for the functional strategy. Higher costs for CCTA in year 3 were due to outliers not associated with cardiovascular disease, they noted.
"An initial CTA strategy had costs similar to those of a functional stress testing strategy, although the patterns of care differed," Mark and colleagues wrote. "Specifically, the CCTA group had less follow-up noninvasive testing and more invasive catheterization and revascularization. Our data suggest that, after 90 days, little happened to these patients out to three years that was driven specifically by which testing strategy they received."
Reducing referral to cath
Few large studies to date have looked at the long-term costs and resource utilization effects of coronary CTA versus functional testing strategies, they wrote. But, overall, the literature seems to show that CCTA modestly reduces referral to catheterization for patients with normal arteries, but increases the use of invasive catheterization and revascularization, with associated increases in costs.
On the other hand, results of the recent Scottish Computed Tomography of the Heart (SCOT-HEART) trial showed pretty conclusively that a CCTA strategy halved the incidence of myocardial infarction versus a functional strategy. However, these results must be interpreted cautiously, Mark said.
"The SCOT-HEART thing is hard to explain, and I'm not sure why" it showed sharply reduced event rates with a CCTA-based strategy, he said. One factor may be SCOT-HEART's potentially less-stable patient population versus PROMISE. Another may be SCOT-HEART's widespread use of treadmill testing versus nuclear stress echo for the functional arm, a group that represented only about 10% of the functional testing patients in the PROMISE analysis.
The take-home message from this analysis is that individual patient factors should guide the approach for coronary artery disease diagnosis, he said.
"I think that the reason to prefer one test over another would not be economic, so you have to make a choice based on the individual circumstances of the patient," he said.
Before the current study was done it wasn't obvious at all if a particular diagnostic approach had important advantages, and each modality had its advocates. Some felt that CCTA's higher referral rate would harm patients in the long run, while others saw advantages in CCTA's reduction in the negative invasive angiography rate. Now, with this study, there is more certainty, though it isn't the last word, Mark said.
An accompanying editorial by Dr. Joe Xie and Leslee Shaw, PhD, from Emory University said that the relatively low costs of care for both strategies may be a reflection of a relatively low-risk, stable cohort, and that more study is needed to assess the value of no-testing options for those patients who lack compelling reasons for additional tests.
"From a cost perspective, symptom-guided treatment without diagnostic testing may dominate economically because it may eliminate the commonplace finding of 'testing begetting more testing,' " Xie and Shaw wrote. "Treatment strategies for suspected CAD have sought to delay or selectively use coronary angiography leading to revascularization ... and similar strategies may be valuable in the de novo evaluation of patients with chest pain."
