Office Spirometry for Lung Health Assessment in Adults: A Consensus Statement From the National Lung Health Education Program

Gary T. Ferguson, MD, FCCP; Paul L. Enright, MD; A. Sonia Buist, MD; and Millicent W. Higgins, MD, Honorary FCCP, *From the University of Arizona (Dr. Enright), Tucson, AZ; Botsford Pulmonary Associates (Dr. Ferguson), FramingtonHills, MI; Oregon Health Sciences University (Dr. Buist), Portland,OR; and the University of Michigan (Dr. Higgins), AnnArbor, MI.


CHEST. 2000;117(4) 

In This Article

Appendix 1: Office Spirometer ValidationStudies

The NLHEP recommends the widespread use of spirometry by PCPs for detecting COPD in adult smokers and describes a new type of spirometer for this purpose: the office spirometer. The value of spirometry for aiding the diagnosis of COPD and asthma, when performed by trained technicians using diagnostic spirometers that meet current ATS recommendations, is widely accepted. The accuracy and precision of diagnostic-quality spirometry performed in the hospital PF laboratory, pulmonary research clinic, and occupational clinic settings by technicians who are trained and have considerable experience performing spirometry have been studied by many investigators and found to be acceptable for the purposes of detecting airways obstruction in individuals and for detecting abnormal declines in FEV1 in groups of adults. However, the first prospective study of > 1,000 spirometry tests performed by nurses in the outpatient clinics of 30 randomly selected primary-care physicians in New Zealand found that less than one third of the test sessions included at least two acceptable maneuvers.[118] About one third of the maneuvers had a "slow start" (peak expiratory flow time [PEFT], > 85 ms [a substantially stricter criterion than those in the NLHEP document]). About two thirds of the maneuvers lasted for < 6 s (forced expiratory time [FET], < 6 s), and visual inspection of the volume-time curves suggested that most of these short-duration maneuvers underestimated the FVC (did not achieve an end-of-test plateau). On the positive side, after attending a 2-h spirometry training workshop, nurses were much more likely to obtain acceptable test sessions. These results confirm the necessity for each new office spirometry system to have a "real-world" validation study before it is marketed.

Several factors, other than instrument accuracy, are known to influence the real-world accuracy and repeatability of spirometry tests. These factors include the following: the technician's training, experience, number of tests performed per month, motivation, motivational skills, and patience; the patient's coordination, cooperation, strength, endurance, and motivation; maneuver and test session quality feedback (to the technician and patient); the training materials that accompany the spirometer; the type and frequency of calibration checks and actions taken to remedy equipment and sensor problems; the testing environment (space, lighting, noise, time constraints, and other stressors); and changes in these factors over the time period of measurement (eg, differing technicians, updated software, new flow sensors, etc).

The goal of an office spirometer validation study is to compare the spirometer's screening and tracking performance in adult patients in the PCP setting with that of diagnostic spirometers used by trained technicians. The following study protocol is designed to apply to any model of office spirometer. It is designed to be performed in a reasonable amount of time (6 months) and with reasonable resources (< $50,000 if a price of <$20 per test is negotiated with the PF laboratory). In order to minimize poststudy criticism, the limits of acceptable outcomes have been predetermined. The manufacturer or distributor of all office spirometers that claim to meet NLHEP specifications must conduct this study for that model and include the published results of the study with each office spirometer sold.

A study coordinator with experience in clinical trials, without a conflict of interest (such as one that an employee of the office spirometer manufacturer or distributor would have), shall be selected. Exactly the same instrument, sensors, manuals, calibration tools, accessories, and training materials that are sold (or provided) commercially as the spirometry system shall be used in the validation study. The same amount of in-service training with the same type of personnel shall be used during the study that will be used for actual commercial training (for instance, 45 min with a local distributor). The setting, staff, and patients will be selected to optimize the generalizability of the results to the real world. A single, representative sample of the office spirometer shall first undergo (and pass) independent testing for FEV1 and FEV6 accuracy and reproducibility, which will be performed by a third party using current ATS recommendations 1 and a spirometry waveform generator, including four waveforms generated using BTPS conditions (body temperature humidified air). All disposable flow sensors used for testing shall be saved in a plastic bag, labeled with the date and patient identification number, and sent to the study coordinator at the end of the study.

Thirty PCPs shall be recruited from advertisements offering "a free spirometer and 6 months of spirometry supplies." At least two regions of the United States shall be represented. At least five PCPs (either medical doctors or doctors of osteopathy) shall be selected from each of the following specialties: family practice, general internal medicine, and general surgery. Allergists and pulmonary specialists shall be excluded. Staff who report that they have personally performed > 100 spirometry tests during the previous 5 years shall be excluded. PCPs who have used a spirometer in their office during the previous year shall be excluded. Each PCP must agree to perform spirometry testing for at least 20 adult patients per month (an average of one patient per weekday) for 6 months. The altitude of each office (within 500 feet) shall be recorded.

Inclusion criteria are consecutive outpatients, aged 45 to 85 years, who are current cigarette smokers or who quit smoking during the previous year. Patients with asthma (according to self-report or the medical record) and those previously noted to have a significant response to inhaled bronchodilators (FEV1 increases, > 12% and > 0.2 L) shall be excluded from the study, since their FEV1 values have inherently high short-term variability.

At least one patient per week shall be asked by each PCP to return to their clinic within 1 month for repeat spirometry testing. A contract shall be made with a local hospital-based PF laboratory to perform follow-up diagnostic spirometry (including printed volume-time and flow-volume curves, but without a physician interpretation) on a subset of study patients. All patients with abnormal spirometry test results shall be scheduled to perform diagnostic spirometry testing at a local hospital-based PF laboratory within 2 weeks of the screening spirometry test. The cost of the diagnostic testing, and a $20 reimbursement for each patient, shall be paid for by the study. The PF lab shall send a copy of the results to the study coordinator. The results of the follow-up spirometry tests shall not be sent to the PCP until the end of the study.

The long-term accuracy of a random sample of five of the study spirometers shall be measured by a third party using a waveform generator at the beginning and at the end of the 6-month study. A random sample of five used flow sensors shall be obtained for the long-term accuracy testing at the end of the study.

The demographics of all patients tested shall be determined and stored for analysis. The demographics must include a unique patient identification number, age, gender, height, weight, race, smoking status, asthma status, date, and technician identification number. The following parameters shall be stored digitally for all (or the best three) spirometry maneuvers: FEV1 , FEV6 , back extrapolation volume, PEFT, PEF, FET, sequence number, and the 50-point flow-volume curve (the average flow during each 2% segment of the FEV6 ). This may require modifications to the office spirometers used in the study (when compared to those that will be sold commercially). These modifications should be designed to minimize technician interaction with the recording device. A written log shall be kept by the office staff of any problems with the spirometer, any calibration checks performed, any preventative maintenance, and any repairs.

The study coordinator shall determine the long-term accuracy of the office spirometer instruments by comparing the baseline and 6-month FEV1 and FEV6 measurements from the "gold standard" waveform generator and the records of repairs, up-dates, or replacements. The quality of all spirometry test sessions (screening, follow-up, and diagnostic) shall be graded by the study coordinator using the stored data and the criteria listed in Table 4 of this document. The rates, trends, and correlates of unacceptable-quality test sessions (QC grades, D or F) shall be determined using logistic regression.

The false-positive and false-negative rates for detecting airways obstruction (after allowing for 3% error in the measured FEV1 / FEV6 ratio) shall be determined by comparing the office spirometry results with the valid follow-up diagnostic tests performed in the PF laboratory. Results from the diagnostic-quality spirometry tests that are determined by the study coordinator (using the printed reports from the PF lab) to be valid (QC grades, A or B) are assumed for the purposes of this study to be the "gold standard." Both the false-negative rate and the false-positive rate shall be < 5% for the office spirometry system to be considered acceptable.

The value of office spirometry for "tracking" purposes shall be determined by calculating the short-term coefficient of repeatability of FEV1 for the subset of patients who performed repeat tests. Acceptable repeatability is for ≥ 95% of the patients to have repeat FEV1 values that match within 0.30 L. The predictors of poor repeatability shall be determined by logistic regression.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: