Validating a dipstick method for detecting recent smoking.

This report evaluates the validity of a new method for verifying self-reported smoking status in patients presenting for pulmonary medicine treatment. A prospective comparison was made between self-reports of smoking status and a new semiquantitative, enzyme-linked, immunosorbent assay-based method testing for the presence of a prime nicotine metabolite, cotinine. Results were validated by gas chromatography/mass spectrometry. Data were collected in an urban, academic, tertiary health care setting. The study included 76 consecutive new patients presenting to participating clinical practices at the Pulmonology or Thoracic Surgery Services. Before taking a smoking history, patients were informed that their urine would be tested onsite for the presence of nicotine using a new method, the NicoMeter, for determining tobacco product exposure, followed by more standard laboratory testing. The level of agreement between the biochemical measurement types was excellent, kappa = 0.777. The new biochemical measurement type used was easy to use. Self-reported smoking status corresponded closely to biochemical testing. However, there was a 5.3-9.5% misclassification of smoking status among the group studied, depending upon the measurement type used. Among 32 lung cancer patients, 15.6%, most likely misrepresented their current smoking status. The NicoMeter appears to be a valid and useful method for confirming self-reported smoking status. Lung cancer patients had a higher rate of inaccurate nonsmoking compared with patients with nonmalignant pulmonary disease. The findings have implications for investigators who accept self-reported smoking status without biochemical verification.