Discrepancies between blood testing performed by Theranos using finger prick samples and testing performed by Quest Diagnostics or LabCorp using standard vein blood draws could affect diagnoses and treatment decisions, according to a study from researchers at the Icahn School of Medicine at Mount Sinai.
As Mount Sinai's Joel Dudley and his colleagues report in the Journal of Clinical Investigation this week, they examined a cohort of 22 healthy adults who underwent blood testing over the course of five days using each company's service. In that way, the researchers collected 14 samples per subject that each underwent 22 clinical lab measurements for a total of 18,480 measurements. They then compared the lab test results they received from Theranos, Quest, and LabCorp.
"While most of the variability we found was within clinically accepted ranges, there were several cases where inaccurate results would have led to incorrect medical decisions," Dudley says in a statement.
The tests performed by Theranos were 60 percent more likely to report a result outside the normal range, Dudley and his colleagues report. In particular, they note that the total cholesterol results provided by Theranos were an average 9.3 percent lower than those given by Quest and LabCorp. This, the Wall Street Journal adds, could influence physicians' decisions regarding whether or not to prescribe statins.
Paperwork the researchers received indicates that Theranos ran the samples from this study at its embattled California lab.
The US Centers for Medicare & Medicaid Services sent the company a letter earlier this year that said that the agency found a number of "condition-level deficiencies" and one hematology testing issue that it said posed "immediate jeopardy to patient health and safety" during its inspection of the company's California lab.
Theranos has said that it has addressed many of the issues brought up by the CMS inspection and that the report doesn't reflect the "current state of the lab." CMS is currently reviewing Theranos' correction plan.
In addition, the Journal has reported that the company only uses its Edison platform for a few of its tests and that to use its finger prick-sized samples, the company has allegedly diluted samples for use on other, traditional lab platforms. It has also reported allegations of irregularities in proficiency testing at Theranos.
Theranos has said that it doesn't dilute samples and that its lab work is accurate.
In a letter to JCI, Theranos says this study was "flawed and inaccurate," according to the WSJ. The company further points out that two study authors have ties to pharmaceutical companies, which it says creates a conflict of interest. The authors denied any conflict, the WSJ says.
This study, the Journal adds, is the first peer-reviewed assessment of Theranos test results in comparison to results from other labs.