New Algorithm May Find Misdiagnosis in COPD

Teresa Carvalho, MS avatar

by Teresa Carvalho, MS |

Share this article:

Share article via email

A new algorithm-based approach was developed to identify patients at risk of being misdiagnosed with chronic obstructive pulmonary disease (COPD).

Researchers believe this may become an important tool to help healthcare providers in diagnosing several diseases.

“Our new algorithm can find the patients who have such an unusual disease trajectory that they may indeed not suffer from the disease they were diagnosed with. It can hopefully end up being a support tool for physicians,” Isabella Friis Jørgensen, said in a press release. Jørgensen is a postdoctoral researcher at the Novo Nordisk Foundation Center for Protein Research, University of Copenhagen, Denmark.

Details of the new algorithm are presented in the article “Time-ordered comorbidity correlations identify patients at risk of mis- and overdiagnosis,” published in the journal npj Digital Medicine.

In spite of the differences between COPD and other lung diseases, these disorders have similar symptoms and overlapping clinical features. As such, it can be difficult to distinguish them and sometimes patients may be diagnosed incorrectly.

It has been estimated that 5–60% of all COPD patients are misdiagnosed. An incorrect diagnosis may lead patients to undergo a wrong treatment, so these errors have to be lessened as much as possible.

Having this in mind, researchers at the University of Copenhagen developed an algorithm able to find possible mistakes in COPD diagnoses.

The development of the algorithm was based on the digital disease history of 284,154 COPD patients from a Danish national patient registry. The idea was to discover cases in which disease presentation was unusual compared to a typical COPD disease course.

A significant deviation indicated that something may be wrong with the diagnosis of that particular patient.

Nearly 69,000 typical disease courses were identified by the team in the group analyzed. Interestingly, up to 42,459 patients did not present a typical, time-ordered disease course.

“In the National Patient Registry, we have been able to map what you could call typical disease trajectory. And if a patient shows up with a very unusual disease trajectory, then it might be that the patient is simply suffering from a different disease. Our tool can help to detect this,” said Søren Brunak, PhD, professor at the foundation and the study’s senior author.

Among patients who had no typical disease course, 9,597 had a very different disease profile compared with the general COPD population.

From this group of patients, a subgroup with 2,185 patients died significantly earlier than patients following a typical COPD course. This subgroup did not show the typical COPD symptoms.

Based on the results, researchers suggested these patients might be at risk of misdiagnosed COPD and that their death could be associated with a more serious health problem such as lung cancer.

“When we studied the laboratory values from these patients more closely, we saw that they deviated from normal values for COPD patients. Instead, the values resembled something that is seen in lung cancer patients. Only 10 percent of these patients were diagnosed with lung cancer, but we are reasonably convinced that most, if not all of these patients actually had lung cancer,” Brunak said.

Furthermore, only 3.8% of the subgroup of potentially misdiagnosed patients had a lung function test to confirm a COPD diagnosis.

According to the team, “this percentage is extremely low considering that a lung function test is essential for assessing the airway obstruction and diagnosing COPD and it indicates that the majority of the patients might not be evaluated correctly,” they wrote.

In total, 17.2% of all COPD patients that follow a trajectory are offered rehabilitation, whereas only 10.4% of the over-diagnosed patients, and none of the misdiagnosed patients, are offered COPD rehabilitation. The procedure code to follow the status of COPD rehabilitation was introduced after the data used in this study ended.

In contrast, another subgroup with 2,368 patients had a longer survival after a COPD diagnosis (of at least 5.5 years) and did not show the typical complications of COPD. The team believes this group could be at risk of over-diagnosed COPD.

“Frequent testing and early screening programs are some of the drivers of overdiagnosis, thus the higher the number of tests being done, the higher is the risk of finding a disease,” researchers wrote.

While the algorithm was validated with data from COPD patients, it may be used for other diseases. The team hopes to further conduct clinical trials to validate and test the algorithm so it can be implemented in Danish hospitals.

“Naturally, our most important goal is for the patients to get their money’s worth with respect to their health care. And we believe that in the future, this algorithm may end up becoming a support tool for physicians,” Brunak said.

“Once the algorithm has mapped the typical disease trajectories, it only takes 10 seconds to match a single patient against everyone else,” Brunak concluded.