Researchers found that physicians may have a financial conflict of interest when examining a chest x-ray for black lung disease, suggesting that sweeping changes are needed to avoid biases and improve transparency.
Many of the physicians tasked with determining the presence of black lung disease in coal miners may experience financial conflicts of interest (COI) based on the entities that hired them, according to a recent study published in the Annals of the American Thoracic Society1.
The study is the first study to provide empirical evidence on whether financial conflicts of interests are associated with workers’ compensation systems. Previously, news reports, anecdotal accounts, and case reports were the only evidence to suggest that a relationship exists between financial COI and black lung diagnosis.
The US Department of Labor (USDOL) is in charge of administering the Federal Black Lung Program (FBLP), an administrative system that manages workers compensation claims for black lung disease, also known as pneumoconiosis.
“Today’s system of oversite is ineffective at monitoring and regulating the role of physicians in this important compensation program that affects thousands of former and current miners. However, the FBLP program is just a microcosm of the larger workers’ compensation system, and preliminary evidence indicates that the association noted in our study likely pervades this larger system,” wrote the investigators.
Chest radiographs (CXR) are a key tool used by doctors to determine if a claimant is eligible for benefits. Physicians who are licensed and trained to determine the presence of pneumoconiosis by evaluating CXRs are known as B-readers.
The investigators looked at claims data filed between January 1, 2000 and December 31, 2013 and included information on 63,780 radiograph classifications made by 264 B-readers. The B-readers each conducted an average of 241.6 pneumoconiosis classifications, with 20% (n = 53) reading 87.7% (n = 55,960) of the CXRs in the dataset.
Overall, simple pneumoconiosis was found to be positive for 31.4% of records and 3.6% of records were positive for progressive massive fibrosis (PMF).
B-readers with a history of being hired by an employer read CXRs as negative for pneumoconiosis in 84.8% of records (n = 27,780/32,744) compared to 51.3% (n = 18,941/36,898) for B-readers with a history of being hired by a miner and 63.2% (n = 4125/6532) of B-readers exclusively hired by USDOL.
There were 64 B-readers who classified an absence of pneumoconiosis 95% of the time, accounting for 10,505 classifications, of which 9805 (93.3%) were made by B-readers predominately hired by employers.
Conversely, 23 B-readers diagnosed simple pneumoconiosis 95% of the time, accounting for 2412 classifications, of which 530 (22.0%) were determined by B-readers predominately hired by miners.
The adjusted odds of a physician reading a CXR as negative for pneumoconiosis was 1.46 (95% CI, 1.44-1.47) per 10% increase in the proportion of court records specifying that a physician was hired by the employer.
Per 10% increase in court records indicating a miner-hired physician, the adjusted odds ratio was 1.51 (95% CI, 1.49-1.52) for diagnosing simple pneumoconiosis and 1.28 (95% CI, 1.26-1.30) for diagnosis PMF.
“While bias is present in both sets of physicians, a vast majority of classifications submitted to these courts are paid for by mine operators. The employers will throw a lot of resources at these cases to discourage miners from filing for compensation by making the process protracted and painful,” said Lee Friedman, PhD, study author and an associate professor at the Division of Environmental and Occupational Health Sciences of the School of Public Health at the University of Illinois Chicago, in a statement2.
The investigators suggested several ways to improve transparency, oversite, and objectivity such as requiring parties to submit requests and payments for reclassification by a physician of their choice to USDOL so that they can conduct quality control and assess possible bias in classification.
Additionally, the investigators also suggested that B-readers could be paid a flat fee for classifications to avoid payment disparities.
Some limitations of the study included that some classifications may not have appeared in the database and that misclassification of B-readers may have occurred because financial affiliations based on court record may not take into account other types of financial COIs, such as stock ownership or grants.