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

Indications for Office Spirometry

Primary-care providers (PCPs) should perform an office spirometry test for patients ≥ 45 years old who report smoking cigarettes (current smokers and those who quit during the previous year) in order to detect COPD.

Rationale: Several well-recognized criteria have been established for the use of medical tests that have been proposed for the early detection of disease, [30,31,32,33,34] and spirometry for the detection of COPD in adult cigarette smokers fulfills all of these criteria:

1. The disease, if not detected early, would go on to cause substantial morbidity or mortality;
2. Treatment is available that is more effective when used at the early stage before the development of symptoms than when used after the symptoms develop; and
3. A feasible testing and follow-up strategy is available that
   a. minimizes the false-positive and false-negative rates,
   b. is relatively simple and affordable,    c. uses a safe test, and
   d. includes an action plan that minimizes potential adverse effects.

The above criteria are usually applied to screening tests, defined as medical tests done for individuals who have no symptoms or signs that suggest the possibility of disease. Office spirometry is considered to be a part of a clinical evaluation and does not fall under the definition of a screening test when performed for patients with respiratory symptoms who are seen during a clinical encounter (whether or not they have a history of cigarette smoking). Also, if the patient has been diagnosed as having tobacco addiction (a disease with a code in the International Classification of Diseases, ninth revision), office spirometry may be indicated to assess the severity of that disease and is not then considered to be a screening test. Although the NLHEP does not recommend office spirometry for screening unselected populations or for testing patients who have no cardiopulmonary risk factors, the next section of this document provides evidence that office spirometry fulfills all of the criteria listed above when it is used to detect COPD in adult smokers.

COPD is the most important lung disease encountered and the fourth leading cause of death in the United States, and it affects at least 16 million people.[7,35] Of the top causes of mortality in the United States, only the death rate for COPD continues to rise, increasing by 22% in the past decade. The 10-year mortality rate for COPD after diagnosis is > 50%.[36] In addition, the number of patients with COPD has doubled in the last 25 years, with the prevalence of COPD now rising faster in women than in men.[37] Although the frequency of hospitalization for many illnesses is decreasing, the number of hospital discharges for COPD rose in the last decade. COPD causes 50 million days per year of bed disability and 14 million days per year of restricted activity.[38,39] COPD causes about 100,000 deaths per year, 550,000 hospitalizations per year, 16 million office visits per year, and $13 billion per year in medical costs, including home care.[35]

COPD is a slowly progressive, chronic disease characterized by cough, sputum production, dyspnea, airflow limitation, and impaired gas exchange. 40 The early and common symptoms of chronic cough and sputum production usually are ignored by the patient (and often their physicians) as normal or expected for a smoker, and no intervention is deemed necessary. The disease usually is not diagnosed until the patient experiences dyspnea with only mild exertion, which interferes with the patient's quality of life. The diagnosis of COPD is made by clinicians (1) by noting the presence of at least one risk factor in the patient's medical history (usually . 20 pack-years of cigarette smoking), (2) by documenting moderate-to-severe airflow limitation using a diagnostic spirometry test, and (3) by excluding heart failure and asthma as the causes of airflow limitation.[12]

The LHS was a randomized clinical trial that demonstrated that COPD could be detected in its early stages in smokers with few symptoms.[28] Spirometry tests were performed for > 70,000 women and men who were current smokers (without regard to symptoms), 35 to 59 years old, from nine United States communities and Winnipeg, Canada.[41] About 25% of those tested were found to have borderline to moderate airflow obstruction. An additional 5% had severe airflow obstruction ( < 50% of predicted), and they were excluded from the study and referred for treatment. Those taking medications for asthma also were excluded. About 6,000 smokers with borderline to moderate airflow obstruction were recruited and were followed up for 5 years. About half of the participants reported chronic cough (with a wide range of 26 to 81%, depending on gender, age group, and clinic site). Wheezing on most days and nights was reported by about one third of participants; only 2.8% reported a current diagnosis of asthma but were not taking any prescription medications for asthma.[42] Those who continued to smoke were documented to have faster rates of decline in lung function. Importantly, participation in a smoking cessation program significantly decreased the rate of decline in lung function in these individuals relative to those who continued to smoke. Those participants who continued not to smoke (sustained quitters) showed a small improvement in lung function over the first year compared to continuing smokers (mean rise in FEV1 , 57 mL vs mean fall in FEV1 ,38 mL, respectively) and had reduced rates of decline over the remaining 4 years of study (mean rate of decline in FEV1 , 34 vs 63 mL/yr, respectively).[28] Thus, the rate of decline of FEV1 following successful smoking cessation was very similar to that seen in healthy nonsmoking adults (28 to 35 mL/yr).[43,44]

In addition to documenting the benefits of smoking cessation in modifying the natural history of COPD, the LHS documented the ability to successfully intervene with an intense smoking cessation program in relatively asymptomatic smokers.[28] At least 35% of the subjects studied were able to quit smoking for extended periods of time, and 22% of the subjects were able to quit and sustain smoking cessation for 5 years (as compared to 6% in the usual care group). The smoking recidivism rates during the 5 years equaled the repeat quitter rates, such that 35% of the subjects were nonsmokers at any cross-sectional period of time. Of course, smoking cessation rates are likely to be lower in primary-care settings when compared to a clinical trial.[33,34]

Effective smoking cessation methods available to primary-care practitioners have dramatically improved in the last several years. Detailed recommendations are now available that synthesize the expanding smoking cessation knowledge base.[45,46] Awareness of different stages in the process of behavioral change have allowed for more focused efforts on subjects likely to quit smoking.[47,48] In addition, increasing success with repeated attempts at smoking cessation now is recognized. Significant advances in the understanding and treatment of nicotine addiction also have occurred.[49] Nicotine gum and patches 50 are now available over the counter in the United States. Bupropion hydrochloride (Zyban; Glaxo Wellcome; Research Triangle Park, NC), an oral medication that is even more effective than nicotine patches,51,52 now is available by prescription in the United States. Comprehensive and effective community-based smoking cessation programs also are available in most communities in the United States.[53]

Recognizing that individual rates of decline in lung function vary, the LHS clearly documents that spirometry can identify large numbers of adult smokers at risk for COPD, and that smoking cessation programs can impact positively on the progression of COPD in those smokers who successfully quit. The regular use of b-agonists or ipratropium in current or former smokers with airways obstruction, but without asthma, apparently has no effect on COPD progression.[28,54,55] However, there is some recent evidence that high-dose inhaled corticosteroids given to smokers with spirometric evidence of mild-to-moderate airflow limitation reduces morbidity, improves quality of life.[56,57,58]

Previous studies of lung function testing in the general population have had mixed results, with some showing no effect [19] and others suggesting that knowledge of an abnormal lung function test doubled the likelihood of quitting smoking, even when no other interventions were applied.[59,60,61,62] A recent review 63 concluded that spirometry meets all the criteria for a test for the early detection of COPD, except that there is no conclusive evidence that spirometry adds to the efficacy of standard smoking cessation advice, which is based on current clinical practice guidelines.[45] Two randomized clinical trials that address this issue have been performed. The first study of 923 Italian smokers found a 1-year quitting rate of 6.5% in those who received counseling with spirometry, 5.5% in those with counseling alone, and 4.5% in those who received only brief physician advice.[64] These rates did not differ significantly, but only half of the study participants who were asked to visit a laboratory for spirometry testing ever did so, and there was no evidence that the spirometry results even were discussed with those who performed the test; therefore, the study probably had inadequate power to show a difference (a type II error). The second study was population based and identified 2,610 young men who were current smokers, were aged 30 to 45 years, had low FEV1 values, and were from 34 cities in Norway.[65] A random half of the men were mailed a personalized letter from a physician stating that they should quit smoking because they were at increased risk for smokingrelated lung disease because of their low lung function. A 15-page smoking cessation pamphlet that emphasized behavioral modification was included in the letter. The self-reported 12-month sustained smoking cessation rates were 5.6% in the minimalist intervention group vs 3.5% in the control group (who were not informed of their spirometry results). After adjusting for age of smoking onset, cigarettes smoked per day, and history of asbestos exposure, the letter describing the abnormal spirometry results was responsible for a 50% improvement in the smoking cessation rates (p < 0.01). Even a 1 to 2% improvement in smoking cessation rates would result in a very large absolute number of lives saved each year in the United States.[66]

Various studies have determined COPD risk factors. COPD occurs predominantly in current and former cigarette smokers, and there is a dose-response relationship. The risk of COPD is strongly associated with the intensity and duration of smoking. 42,67,68 Other factors that also increase COPD risk, but less commonly or to a lesser degree, include occupational dust exposure,69 environmental tobacco smoke,68 exposure to environmental air pollution,70 a rare genetic deficiency of a 1 -antitrypsin,71 a history of childhood respiratory infections,72 and the presence of airway hyperresponsiveness, as measured by spirometry.[73,74] Even moderate COPD cannot be detected reliably by a medical history or physical examination.[75,76,77]

Abnormal spirometry (ie, limitation of expiratory airflow, airways obstruction, or a low FEV1 /FVC ratio) is a strong predictor for rapid progression of COPD.[28] The degree of airways obstruction correlates closely with pathologic changes in the lungs of smokers and patients with COPD.[78] Spirometry results are also a strong independent predictor of morbidity and mortality due to COPD,79,80 mortality due to cardiovascular disease,81 lung cancer,82,83 as well as all-cause mortality.[84,85]

The accuracy of a test for the early detection of disease is measured in terms of two indexes: sensitivity and specificity.[5] A test with poor sensitivity will miss cases (true-positive results), producing false-negative results, while a test with poor specificity will result in healthy persons being told that they have the disease, producing false-positive results.

An accepted reference standard (a "gold standard") must be available to provide the means for distinguishing between true-positive and false-positive results from the new test. The traditional "gold standard" for the diagnosis of COPD is the pathologic examination of lung tissue,[78] but this confirmation of the disease is inappropriate in routine practice due to the invasive nature of a lung biopsy. The finding of abnormally low lung densities on a high-resolution CT (HRCT) lung scan in adult smokers is very highly correlated with the pathologic grading of emphysema [86] and, therefore, may soon be considered a secondary reference for COPD, but HRCT lung scans are infrequently performed clinically due to their high cost. COPD, as determined by HRCT lung scans, is moderately correlated with lung function testing (FEV1 /FVC ratio and diffusing capacity of the lung for carbon monoxide) in adult smokers,[87] but emphysema (lung tissue destruction accompa by lung hyperinflation) is only one component of COPD and may not be an important predictor of morbidity and mortality, independent of airflow obstruction. The widely accepted definition of COPD progression is an abnormal rate of decline in lung function.[54,80] The normal annual decline in FEV1 in healthy, never-smoking adults who are 35 to 65 years old has been determined by several longitudinal studies to be a mean of 30 mL/yr with an upper limit of the normal range of 50 mL/yr, which may be used to define "rapid fallers."[88]

It is important that a high proportion of those who test positive actually have disease (positive predictive power). This proportion is higher when the prevalence of disease is high. The best estimates of the prevalence of airflow obstruction and COPD in the US population are now available from NHANES III (conducted from 1988 to 1994). In NHANES III, spirometry was measured in a sample of > 16,000 adults who represented the noninstitutionalized population of the United States. About 29% of all the adult participants reported current smoking, and 24% were former smokers. Normal reference values of several spirometry variables were developed from the "healthy" subset of the nonsmoking men and women who were free of respiratory symptoms and diseases. Lower limit of normal (LLN) values, which were specific for age, sex, and height, were set at the fifth percentile of the reference population values.[27] For this report, prevalence rates of low lung function in the US population were estimated by defining low lung function as a FEV1 /FEV6 ratio less than the LLN and an FEV1 value less than the LLN. See Table 1 for the results.

Prevalence rates of low lung function increase with age and are highest in current smokers, intermediate in former smokers, and lowest in never smokers. Rates are similar in men and women. Compared with rates in never smokers, rates are more than five times as high in current smokers at ≥ 45 years old and are more than three times as high in former smokers ≥ 55 years old. Prevalence rates also were compared in men and women who reported any respiratory condition or symptom with those who did not. A report of any of the following placed the individual in the symptomatic group: a doctor's diagnosis of asthma, chronic bronchitis, or emphysema; cough or phlegm on most days for ≥ 3 con months during the year; shortness of breath on mild exertion; or chest wheezing or whistling apart from colds. Rates of low lung function were consistently three or more times higher in symptom men and women than in those who were asymptomatic.

We recommend that all patients ≥ 45 years old who are current smokers, as well as those with respiratory symptoms, perform office spirometry or diagnostic spirometry. Based on the NHANES III study, the numbers of patients eligible for spirometry under these recommendations, and the expected yield of abnormal spirometry tests are given in Table 2. About one quarter of current cigarette smokers with a respiratory symptom, a total of 9 million persons in the United States, can be expected to have low lung function (airway obstruction). Smokers ≥ 45 years old without respiratory symptoms also have a relatively high abnormality rate: about 9% of men and 14% of women. On the other hand, current and former smokers < 45 years old have spirometry abnormality rates that are similar to those of healthy never smokers (about 5%), reducing the value of spirometry testing of young adult smokers. Asymptomatic former smokers ages ≥ 55 years also have a sprirometry abnormality rate of 5%.

Spirometry is a relatively simple, noninvasive test. Office spirometry takes only a few minutes of the patient's and technician's time and includes a few athletic-type breathing maneuvers of 6 s duration. The economic costs of a spirometry test include the cost of the instrument and the cost of personnel time (both training and testing). Diagnostic spirometers currently cost about $2,000, and about $10 of time per test is spent in testing (including training time) and disposable supplies. Office spirometers will cost , $800 and require even less testing time than diagnostic spirometers. Adding a post-bronchodilator spirometry test for asthma adds about 15 min to the test time (but is not needed for COPD evaluations).

Any medical test has both tangible and intangible costs. Adverse effects may occur (1) due to the procedure itself, (2) due to the investigation of abnormal results, or (3) due to the treatment of detected abnormalities or diseases.[33,34] There are no adverse side effects from spirometry testing, other than occasional minor discomfort. However, investigation and confirmation of abnormal spirometry results in some patients will cost both time and money and may result in psychological and social harm in some patients. The cost of diagnostic spirometry to confirm airflow obstruction when performed in a hospital-based pulmonary function (PF) laboratory ranges from $20 to $60. The estimated travel time, waiting time, and testing time spent by the patient ranges from 1 to 3 h. The possible psychological impact of being labeled as "ill" by self and others related to false-positive or even true-positive test results could lead to alterations in lifestyle and work and to seeking medical attention. Another potential adverse effect is the unmeasured risk of reinforcing the smoking habit in some of the four of five adult smokers who are told that they have normal results for spirometry testing. However, the clinician should counteract this possibility by taking the opportunity to tell the patient that normal results for spirometry testing do not mean that the patient's high risk of dying from a heart attack, lung cancer, or other smoking-related diseases is substantially reduced; therefore, smoking cessation remains very important.

Finally, the risk of an adverse effect caused by the intervention for COPD (smoking cessation) is very small. The side effects of over-the-counter nicotine replacement are minimal. Successful smoking cessation leads to a small average increase in body weight,89 but the slight increase in medical risk from minor weight gain is far exceeded by the benefits due to reduced morbidity and mortality and the economic savings in cigarette and cleaning costs.

Even when test quality seems good, diagnostic spirometry is highly recommended to confirm abnor office spirometry findings prior to initiating an expensive workup or an intervention with negative economic consequences (such as a recommendation to change jobs or to prescribe a medication). The key focus of the NLHEP program is preven and early intervention. Validated abnormal test results in a smoker should lead to a more detailed history and examination for pulmonary disease and cardiovascular risk factors (including hypertension, diabetes mellitus, obesity, hypercholesterolemia, etc). Consideration should be given to the presence of pulmonary diseases other than COPD, including asthma, restrictive lung and chest wall diseases, neuromuscular diseases, and cardiac disease. When airway obstruction is identified in a smoker, the primary intervention is smoking cessation. In the event that a patient with airway obstruction contin to smoke cigarettes, a renewed or increased effort to assist with smoking cessation is essential. Future research may determine that other interven such as anti-inflammatory therapy, are effective in selected patients with airway obstruction. Referral to a subspecialist for further diagnostic testing should be considered in some patients, such as those in whom bronchiectasis or other lung diseases are suspected. Pre- and post-bronchodilator diagnostic spirometry is indicated if asthma is suspected.

Primary-care physicians should perform an office spirometry test in patients with respiratory symptoms such as chronic cough, sputum production, wheezing, or dyspnea on exertion in order to detect asthma or COPD.

Rationale: Analyses of data from a population sample of 25- to 75-year-old white men in Tucson, AZ, found that spirometry abnormality rates increased in those who reported respiratory symptoms, after excluding those who reported a physician diagnosis of asthma, chronic bronchitis, or asthma.[90] Abnormal spirometry was defined as an FEV1 below the LLN, using the reference equations from the study by Crapo et al,91 which reported spirometry reference values very similar to the NHANES III values. The comparison subjects, never smokers without respiratory symptoms, had a 3.8% spirometry abnormality rate, while asymptomatic former smokers and current smokers had abnormality rates of 9.2% and 11%, respectively. Former smokers and current smokers with any of three respiratory symptoms (chronic cough and sputum, dyspnea walking on level ground, or attacks of dyspnea with wheezing) had abnormality rates of 25.6% and 14.1%, respectively. These abnormality rates, and those from NHANES III (Tables 1, 2), demonstrate that the presence of respiratory symptoms in a former or current cigarette smoker substantially increases their pretest probability (risk) of having airflow obstruction (low lung function) or COPD.

The National Health Interview Survey (conducted from 1993 to 1995) estimated that 4 million adults (4.5% of those aged 35 to 65 years) have asthma (by self-report) and that 630,000 emergency department visits for asthma occur each year in this age group.[92] A survey of 59 primary-care practices with 14,000 patients in Wisconsin reported an asthma prevalence of 6.2% in adults (≥ 20 years old), half of whom reported adult onset of the disease.[93] An additional 3.3% of the patients without a diagnosis of asthma reported attacks of wheezing with dyspnea during the previous year, suggesting, along with other investigations, that asthma is underdiagnosed in adults.[17] Spirometry is recommended by current clinical guidelines for patients with symptoms that suggest asthma in order to help confirm the diagnosis.[94]

Primary-care physicians may perform an office spirometry test for patients who desire a global health assessment (risk assessment).

Rationale: Lung function testing is now recognized as a measure of global health, predicting all-cause mortality and morbidity in adults.[85,95,96,97] In addition, lung function test results and changes in lung function over time have been shown to identify patients at high risk for lung cancer [82,83] and at increased risk for coronary artery disease,[98] congestive heart failure,[99] stroke and other heart and blood vessel disaeses,[100] and altered mental function in later years of life.[101] Early identification and recognition of increased global health risks also may allow for evaluation and for prevention and early intervention in other risk areas appropriate to each of these nonpulmonary disease categories. Office spirometry also may identify patients with subclinical asthma or restrictive lung processes in both adults and children, leading to the institution of appropriate evaluations and treatments. Although prophylactic interventions such as vaccination are recommended for patients with respiratory illnesses, only a small percentage of them receive influenza and pneumococcal vaccines.[102] In adults, early intervention following early identification of lung function abnormalities can lead to improved smoking cessation, to occupational, avocational, or environmental changes, and to increased awareness and attention to cancer, cardiac, and other nonpulmonary health issues associated with abnormal lung function. Early identification of lung function abnormalities in relatively asymptomatic patients may provide "teachable moments" or specified times for a given patient when there is an increased awareness and response to medical education and intervention. Such moments may lead to an increased responsiveness to smoking cessation and to enhanced opportunities for other preventive therapies or modification of identifiable risk factors.

Assuming that lung function testing of selected individuals is a useful part of health care, it is essential that the test chosen is the best available. First, it must be able to detect mild disease. Although many lung function tests are available, previous studies examining the value of these tests have shown that most of them are unacceptable or ineffective as tools for the early detection of COPD.[19,20] The exceptions are peak expiratory flow (PEF) and spirometry. PEF measurements are recommended for asthma management by current clinical practice guidelines, but spirometry is recommended to help make the diagnosis of asthma.[94] Likewise, we do not recommend the use of PEF to evaluate patients for COPD. The advantages of PEF tests are the following: measurements within a minute (three short blows) using simple, safe, hand-held devices that, typically, cost < $20. On the other hand, the disadvantages of using PEF when compared to spirometry are as follows: PEF is relatively insensitive to obstruction of the small airways (mild or early obstruction); PEF is very dependent on patient effort; PEF has about twice the inter- and intrasubject variability 103 ; and mechanical PEF meters are much less accurate than spirometers.[13]

Tracking of lung function over time has potential advantages over a single test.[104] However, there are no published data demonstrating that when the results of the first spirometry test are normal in a high-risk patient the measurement of annual changes in lung function (tracking) in the primary-care setting is better than simply repeating office spirometry at 3- to 5gyear intervals, which we recommend. In occupational medicine, diagnostic-quality spirometry tests often are performed regularly for the surveillance of employees at high risk.[104,105] Annual tests increase the likelihood of detecting changes in lung function earlier when compared to less frequent testing intervals. Infrequent testing (eg, every 5 years) may delay identification of lung function abnormality, reducing the benefits of identification, prevention, and early intervention in lung disease. However, when testing is performed more frequently, and when a less-than-optimal spirometry quality-assurance program is used, the false-positive rate increases. Office spirometry may be indicated for patients who report workplace exposures to chemicals, dusts, or fumes that are known to cause lung disease; however, a discussion of testing for occupational lung disease is beyond the scope of this document.


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: