SAN DIEGO — Primary care physicians use the nonspecific ICD-9 code 496.xx to indicate chronic obstructive pulmonary disease more frequently than specific ICD-9 codes for the disease, Vijay N. Joish, Ph.D., reported in a poster session at the 100th International Conference of the American Thoracic Society.
Most COPD diagnoses in primary care offices are being made without pulmonary function test results or any recorded history of risk factors, said Dr. Joish of the pharmacotherapy outcomes research center at the University of Utah, Salt Lake City.
The findings, part of what he said is the largest study of its kind, confirm previous studies suggesting that patients with COPD continue to be undiagnosed and undertreated. “There is an underuse of spirometry and lung function tests to diagnose COPD,” he said in an interview. “I'm not sure how physicians would get the most effective treatment if they don't have that kind of information.”
He and his associates used GE Centricity, a national electronic medical records database containing more than 16 million patient charts, to identify adults diagnosed with COPD during or after 1990 in office-based family practice, internal medicine, and general practice settings. They identified 35,752 patients by ICD-9 codes 491.xx (bronchitis type), 492.xx (emphysema type), and 496.xx (not elsewhere specified) but studied only the 14,691 who had at least 6 months of health care resource use prior to their first diagnosis of COPD.
The investigators evaluated pulmonary function tests and diagnoses of specified risk factors in the 14,691 patients before and after the first diagnosis of COPD to determine the progression of the disease. They used the Global Initiative for Chronic Obstructive Lung Disease criteria to measure disease severity and defined risk factors as physician recorded diagnosis of smoking, bronchitis, dyspnea/shortness of breath, and cough or abnormal sputum.
Of the 14,691 patients, 10,494 (71%) had an ICD-9 diagnosis of 496, the nonspecific code to indicate COPD. Only 383 patients (less than 3%) had any pulmonary function test before or on their first day of COPD diagnosis, and 426 (less than 3%) had pulmonary function tests after diagnosis. In addition, 4,612 patients (31%) had a recorded risk factor prior to their COPD diagnosis.
In the poster, Dr. Joish and his associates acknowledged that some of the medical records used in the study might have been incomplete. “For instance,” they wrote, “handheld spirometry tests may not [have been] captured and/or appropriately transcribed electronically into the database.”
Pfizer Inc. funded the study.