0
Original Research: LUNG FUNCTION TESTING |

Interpreting Periodic Lung Function Tests in Individuals*: The Relationship Between 1- to 5-Year and Long-term FEV1 Changes FREE TO VIEW

Mei Lin Wang, MD, MPH; Bipin H. Avashia, MD; Edward L. Petsonk, MD, FCCP
Author and Funding Information

*From the Division of Respiratory Disease Studies (Drs. Wang and Petsonk), National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention, Morgantown, WV; and Medical Department (Dr. Avashia), Bayer CropScience, Charleston, WV.

Correspondence to: Edward L. Petsonk, MD, FCCP, National Institute for Occupational Safety and Health, Mail Stop H-G900.2, 1095 Willowdale Rd, Morgantown, WV 26505; e-mail: elp2@cdc.gov



Chest. 2006;130(2):493-499. doi:10.1378/chest.130.2.493
Text Size: A A A
Published online

Study objective: Spirometry is performed to monitor lung health, but variability between tests can hinder recognition of excessive FEV1 declines. We sought to describe the relationship between FEV1 changes over 1 to 5 years and FEV1 declines over longer terms, using 21,821 test results from 1,884 workers who participated in an annual health monitoring program at a chemical plant between 1973 and 2003.

Methods: Test results from workers with five or more valid results over ≥ 10 years were included in our analysis (mean initial worker age, 35 years; range, 18 to 62 years; 91% male; 35% current smokers and 41% nonsmokers). For each worker, long-term FEV1 slopes (milliliters per year) were calculated by simple linear regression using all available results and compared to changes in FEV1 between two tests over 1 to 5 years, expressed in both milliliters and percentage of initial value.

Results: Long-term (mean, 18 years; range, 10 to 30 years) slopes averaged − 29.1 mL/yr (− 27, − 29, and − 37 mL/yr for male never-smokers, former smokers, and current smokers, and − 20, − 26, and − 27 mL/yr for female never-smokers, former smokers, and current smokers, respectively). Excessive short-term and long-term declines were defined by lower fifth percentile values. Individuals with abnormal short-term declines were found to be 3 to 18 times more likely to ultimately show excessive long-term declines; with the strength of the association increasing with the length of the short-term testing interval. Better test operating characteristics resulted if abnormal short-term FEV1 change was based on percentage change (ie, percentage per year) rather than absolute change (ie, milliliters per year).

Conclusions: Our findings provide guidance for interpreting periodic spirometry results from individuals exposed to respiratory hazards.

Figures in this Article

Spirometry is a widely applied and practical method for assessing lung health, and professional standards are available to guide test performance and interpretation.14 Serial monitoring of lung function changes in individuals exposed to respiratory hazards is recommended for the early detection of conditions that result in airflow limitation, when interventions are most likely to be successful.56 However, because fixed airflow limitation develops gradually over a 20- to 30-year period, the FEV1 loss per year resulting from a disease process is typically small and easily obscured by measurement variability. This may hinder the correct identification of individuals experiencing true excessive declines. Prolonged follow-up is recommended for reliable estimation of the rate of longitudinal change in spirometry measurements in individuals, and only relatively large changes over 1- to 2-year monitoring intervals are confidently identified as abnormal.9 The American Thoracic Society (ATS) recommends that a year-to-year change in FEV1 of ≥ 15% be used to identify a clinically important change in an individual,10 although one study11 has indicated that when spirometry is performed according to professional standards, a yearly decline in FEV1 > 8% or 330 mL occurs in < 5% of healthy working males and should not be considered normal. To effectively interpret serial spirometry results for the purpose of identifying individuals likely to have excessive long-term declines in FEV1, health professionals must understand the relationship between FEV1 changes observed during routine spirometry monitoring and subsequent long-term declines. This study investigated this relationship using data from a large occupational spirometry monitoring program spanning 30 years.

The data that formed the basis of the study were abstracted, without personal identifiers, from an ongoing health monitoring program at the medical department of a large multiproduct integrated chemical facility. The Centers for Disease Control and Prevention Institutional Review Board specified that, since an existing de-identified data set was to be used, informed consent was not required for this study.

Participants and Spirometry Testing

Spirometry was offered annually to all workers at the plant, beginning in 1973; a total of 3,724 workers performed at least two tests between 1973 and 2003. To ensure a reliable estimate of longitudinal decline in FEV1, only workers who had at least five valid results over ≥ 10 years were included in this study (n = 1,884). Spirometry was performed by trained nurses at the company medical department according to ATS standards. In the years before ATS standards were published, training in spirometry was supervised by academic researchers who were familiar with spirometry equipment, techniques, and procedures. Only results that were assessed as valid by the company medical director (B.H.A.) were entered into the database.13

Association Between Short-term Changes and Long-term Declines in FEV1

Short-term FEV1 changes refer here to the differences between any two test results over a 1- to 5-year interval, as would be encountered in a typical spirometry monitoring program, and were evaluated in two ways: the absolute change in FEV1 between two tests over 1 to 5 years (ΔFEV1), and the percentage change in FEV1 between two tests over 1 to 5 years (%ΔFEV1). For all possible 1- to 5-year test intervals for each worker, the absolute change between two serial FEV1 values was calculated by subtracting the first FEV1 value from the subsequent test value and expressed in milliliters. The percentage change in FEV1 was calculated as this absolute change divided by the first FEV1 value multiplied by 100. The distributions of ΔFEV1 and %ΔFEV1 values over 1 to 5 years were also computed, and the fifth percentile values were used as the cut point to define abnormal short-term changes.

Each worker’s long-term FEV1 decline (expressed in milliliters per year, and referred to as the FEV1 slope) was calculated by simple linear regression using all the valid measurements available on that individual. The distribution of long-term FEV1 slopes, stratified by gender, initial smoking status, and baseline age, was computed using the SAS univariate procedure (SAS Institute; Cary, NC).

An individual worker was classified as demonstrating an excessive long-term FEV1 decline if the value of his or her long-term FEV1 slope was at or below the gender-specific fifth percentile. Likewise, for each short-term measurement interval, the monitoring result, ΔFEV1 or %ΔFEV1, was considered abnormal if the value was at or below the corresponding fifth percentile. Odds ratios were used as measures of the strength of associations between an abnormal short-term monitoring result and excessive long-term FEV1 decline. The computation of odds ratios and confidence intervals was based on 2 × 2 tables.12

Diagnostic Accuracy for Spirometry Monitoring

By assuming that the excessive long-term decline category represents the reference standard of disease truly “present” and by designating the short-term FEV1 change as the clinical index test, we assessed the diagnostic values for monitoring spirometry by computing the sensitivity, specificity, and the positive likelihood ratio (LR+) using 2 × 2 tables.1314 Sensitivity is the ability of a test to detect true disease. Specificity is the ability of a test to detect the disease-free state. The likelihood ratio is a useful index of test performance that combines information about the sensitivity and specificity of a test, and provides an indication of how much the odds of disease change based on a positive or a negative result. Application of Bayes theorem using the likelihood ratio produces the following summary equation: the posttest odds of a disease is given by the pretest odds multiplied by the likelihood ratio.13 For example, if the pretest odds of a worker having a disease are 1 in 20, and the LR+ is 10, then after a positive test result, the odds of disease are 10 × 0.05, or 1 in 2. An LR+ of 2 to 5 indicates a fair clinical test; 5 to 10 is considered a good test; and > 10 characterizes an excellent test result.15 The appropriateness of the selection of the fifth percentile cut point for classifying a short-term change as abnormal was examined by comparing the sensitivity, specificity, and LR+ of several other cut points.

Population Characteristics

The 1,884 workers included in this study represented a middle-aged working population: 91% male and 92% white, with 35% current smokers, 24% former smokers, and 41% never-smokers at the baseline test, and 18% current smokers at the final test. The mean follow-up interval between each worker’s first and last test was 18 years (range, 10 to 30 years), with an average of 12 valid spirometry results (range, 5 to 23 results) for each worker.

Long-term FEV1 Declines

The distribution of long-term FEV1 slopes among the workers is shown by gender in Figure 1 . The mean slopes were − 30 mL/yr for men and − 23 mL/yr for women; this gender difference was highly significant (p < 0.0001). For the 1,840 workers excluded from the study, the mean slopes were − 21 mL/yr for men and − 7 mL/yr for women. The values at the fifth percentile, used for categorizing excessive long-term decline as present or absent, were − 68 mL/yr and −55 mL/yr for men and women, respectively.

Figure 2 illustrates the mean FEV1 slope by age and smoking status among men. Male smokers lost more FEV1 than nonsmokers and quitters; the smoking effect was statistically significant by two-way analysis of variance. Men whose baseline age was ≥ 35 years lost significantly more FEV1 than other men, whereas the group with baseline age ≤ 25 years lost significantly less. For women, the patterns of long-term FEV1 declines by smoking and age were similar to the relationships observed for men (data not shown).

Short-term FEV1 Changes Over 1 to 5 Years

Table 1 shows the mean and SD of ΔFEV1 for each short-term monitoring interval (1 to 5 years), and the long-term FEV1 slope (average, 18 years), by gender. In both men and women, the means of ΔFEV1 and 18-year slopes were all similar, although they differed significantly by gender. These results underscore the variability in repeated FEV1 measurements, as reflected in the SD, which is high for the 1-year and 2-year monitoring intervals but decreases as the interval between tests increases.

Table 2 shows the fifth percentile values of ΔFEV1 and %ΔFEV1 by gender and the interval between tests. When expressed as percentage change, the values for men and women are similar, although when expressed as the absolute value (milliliters per year), the decline at the fifth percentile is always greater for men than for women. Table 3 shows the values of ΔFEV1 and %ΔFEV1 by smoking status at the final test and the interval between tests for male participants. The fifth percentile value of ΔFEV1 and %ΔFEV1 for smokers are generally below the values for never-smokers (by approximately 1% for %ΔFEV1), with former smokers always in-between. For women, the trends for smoking are similar (data not shown)

Diagnostic Accuracy of Spirometry Monitoring

The association between excessive long-term FEV1 slopes and abnormal short-term monitoring results (defined using the fifth percentile values from Table 2) was assessed by odds ratio estimates from 2 × 2 tables (Table 4 ). The results indicated that workers with abnormal short-term change had approximately 3 to 18 times higher odds of having long-term excessive FEV1 loss. The odds ratio increased with the length of the short-term monitoring interval and, for each interval, was higher when the %ΔFEV1 criterion was used, compared to the ΔFEV1 criterion.

Table 5 demonstrates the diagnostic values of sensitivity, specificity, and LR+ by interval, again using the cut points from Table 2. Use of the percentage change (%ΔFEV1) resulted in better sensitivity and a higher likelihood ratio, with similar specificity to that obtained using absolute change (ΔFEV1). The results indicate that these diagnostic values increased with the length of the monitoring interval and were consistently better when the percentage criterion was used. Although the specificity of the monitoring results was high (96%), the sensitivity was generally low (17 to 41%).

To investigate the influence of the cut point on the performance of the short-term change as an indication of subsequent excessive long-term decline, other criteria were also investigated (Table 6 ). Using the first percentile cut point achieved an excellent LR+, but sensitivity was at best 13%. In contrast, the tenth percentile showed sensitivities up to 51%, but the LR+ was only fair to good.

Periodic monitoring of pulmonary function is often undertaken in cigarette smokers and other individuals exposed to occupational and environmental inhalation hazards. A principal goal of spirometry monitoring is the prompt and accurate identification of individuals in whom a clinically important lung disorder is developing, in order to permit timely health-protective interventions. However, precise estimation of an individual’s rate of change in FEV1 requires relatively prolonged follow-up, and the reliable detection of excessive longitudinal change in spirometry measurements is complicated by the presence of substantial technical and biological variability and the current dearth of knowledge about the diagnostic accuracy of testing.89,16 For monitoring individuals, it has been recommended that spirometric measurements be made using standardized equipment and testing techniques over at least 4 to 6 years.910,17

Based on 30 years of spirometry monitoring data collected by the medical department staff of a large chemical plant, the current study describes the relationship between short-term FEV1 changes over 1 to 5 years and long-term declines, with the goal of assisting health professionals involved in monitoring programs in the identification of individuals with excessive declines. The relationship between individual changes in spirometry, as recorded through typical monitoring programs, and long-term FEV1 declines has not previously been reported. The fifth percentile values of ΔFEV1 and %ΔFEV1(Table 2) represent reference values for the limit of normal decline for two tests over 1- to 5-year intervals. These values should be most useful for interpreting spirometry that has been performed annually to monitor generally healthy working-age adults. For example, an abnormal FEV1 change during monitoring should trigger further medical assessment and evaluation of the affected individual’s environmental conditions, to permit early interventions for that individual and others exposed in the same environment.

The diagnostic values listed in Table 5 are intended to provide information useful to health-care professionals in understanding the performance characteristics of periodic spirometry testing among apparently healthy persons exposed to respiratory hazards. These values reflect the operating characteristics of gender-specific fifth percentile cut points derived from the chemical plant workers in all smoking and job categories who participated in the annual testing. We did not evaluate the diagnostic values using fifth percentile cut points by smoking status because the smoking status of the individual workers often changed over time. However, the small differences in the fifth percentiles between smoking groups (Table 3) suggest such an approach is unlikely to result in important differences.

Defining an abnormal %ΔFEV1 using gender-specific fifth percentile cut points, our results indicate the LR+ ranges from 3.6 to 10.8 and sensitivity ranges from 17 to 41% for two spirometry tests 1 to 5 years apart, respectively (Table 5). The low sensitivity is not surprising considering the recognized variability between two repeated spirometry measurements. Alternative cut points could be chosen and would influence the diagnostic values. For example, cut points at the tenth percentile of %ΔFEV1 could be used to increase the sensitivity of the test (51% sensitive for the 5-year interval), but this would decrease specificity (eg, increasing misclassification due to acute respiratory tract illnesses) and reduce the LR+ to the fair-to-good range (Table 6). Repeating the spirometry test in 6 to 8 weeks for individuals who meet a cut point criterion was not done for this study but could be evaluated as an approach to improve specificity while maintaining sensitivity. Even without confirmatory repeat testing, the fifth percentile cut point for %ΔFEV1 achieves an excellent LR+ and specificity, with 41% sensitivity, at a short-term interval of 5 years; thus, use of this criterion appears generally preferable at this time.

Occupational health professionals who interpret periodic spirometry results are encouraged to apply interpretation criteria that are effective in identifying individuals who are truly experiencing excessive lung function loss. In doing this, they need to keep in mind that overall test performance is heavily dependent on measurement variability.18 The results from the current study underscore the high variability in indexes derived from repeated FEV1 measurements and emphasize the importance of continuing attention to training, equipment, and procedures in order to reduce unnecessary technical sources of variability.

This study has a number of limitations: first of all, we did not have access to symptom data, medical records, or other independent objective tests that could help assess the presence of a clinical disorder (eg, emphysema) among individuals who had excessive long-term lung function loss. Secondly, a number of factors that may affect the recognition of excessive FEV1 decline were not fully addressed in this analysis, such as birth cohort and changes in body weight.19 Such factors can increase the variability in measurements of FEV1. The findings presented may have been affected by our decision to restrict analysis to workers with at least 10 years of follow-up. Data from workers who had left employment prematurely due to respiratory illness may have been more likely to have been excluded from the study, thus potentially reducing the prevalence and severity of long-term declines in the study population. However, we compared the demographic and spirometry indexes between the 1,884 workers included in the study and the 1,840 excluded due to fewer than five tests or < 10 years of follow-up and observed no evidence of such a bias related to FEV1 decline.

Other recognized potential sources of bias and variation seem unlikely to have affected the results.20 Strict statistical criteria were selected a priori to classify test results, eliminating interpretation and verification bias. Survey biases seem unlikely since workers were tested in the plant medical department on a regular basis throughout the year. Variations in spirometry test protocols can influence measurement variability, and thus affect the interpretation of longitudinal change in lung function. To explore this, the current study data were compared to several other spirometry databases assembled for research and medical surveillance purposes. The fifth percentile values for both year-to-year ΔFEV1 and %ΔFEV1 from the current study data in 1,721 male chemical plant workers (− 380 mL/yr and − 10.4%) are similar to the results from 160 male wood product workers (− 370 mL/yr and − 8.6%) and also to the results from 389 male coal miners and nonmining blue collar workers (− 350 mL/yr and − 9.0%).,11 These similarities suggest that spirometry measurement variability in the current study is similar to studies among other working populations in which accepted professional standards were applied.

The expected rate of change in spirometry results for an individual has not been as well studied as cross-sectional predicted values. There are a few publications8,11,21 that propose methods or specific criteria to determine expected limits of normal longitudinal declines among apparently healthy adults in an occupational or community setting. The current investigation should be useful in this regard, but additional studies in various populations are needed to provide firm guidance in the early detection, assessment, and prevention of lung function loss among individuals at risk for the development of chronic airflow limitation. Data for this study came from an annual occupational health-monitoring program; future studies should investigate if less frequent spirometry testing can achieve comparable test performance. Future studies should also consider the utility of short-term (ie, 6 to 8 weeks) follow-up spirometry, especially for individuals with larger but apparently normal losses (eg, below the tenth percentile but above the fifth percentile cut point) and should also evaluate age-related limits for normal declines.

In conclusion, long-term FEV1 slopes vary significantly by age, gender, and smoking status. Changes in FEV1 between two tests over 1 to 5 years are significantly associated with long-term lung function declines. When classifying excessive short-term FEV1 change over two tests, percentage change (ie, %ΔFEV1, expressed in percentage per year) appears to offer better test operating characteristics than absolute change (ie, ΔFEV1, expressed in milliliters per year) for identifying individuals with excessive long-term FEV1 declines. When using the fifth percentile cutoff values to identify abnormal %ΔFEV1 changes, the diagnostic value of the test, as indicated by the LR+, increases with the test interval from fair (interval of 1 year), to good (2 to 4 years), to excellent (5 years). Annual spirometry in workers and others exposed to respiratory hazards can detect relatively large losses in lung function over short follow-up intervals and, when interpreted using the diagnostic values from this study, can also identify individuals who have a substantial risk of excessive long-term functional decline.

Abbreviations: ATS = American Thoracic Society; ΔFEV1 = change in FEV1 between two tests over 1 to 5 years; %ΔFEV1 = percentage change in FEV1 between two tests over 1 to 5 years; LR+ = positive likelihood ratio.

The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the National Institute for Occupational Safety and Health.

Supported by the National Institute for Occupational Safety and Health and Bayer CropScience.

Figure Jump LinkFigure 1. Distribution of long-term FEV1 slopes by gender among 1,884 chemical plant workers who participated in spirometry screening over ≥ 10 years and had at least five valid results. FEV1 slopes were calculated for each individual by simple linear regression using all valid spirometry results for that individual.Grahic Jump Location
Figure Jump LinkFigure 2. Mean FEV1 slopes by age and smoking status at baseline in 1,721 male chemical plant workers who participated in spirometry screening ≥ 10 years and had at least five valid results. FEV1 slopes were calculated for each individual by simple linear regression using all valid spirometry results for that individual. *Smoker vs former and never-smoker, p < 0.0001; age ≤ 25 years vs others, p < 0.0001; age 26 to 34 years vs age ≥ 35 years, p < 0.0001.Grahic Jump Location
Table Graphic Jump Location
Table 1. ΔFEV1 Values Over 1 to 5 Years and Long-term FEV1 Slope by Test Interval and Gender (n = 1,884)
* 

Mean, 18 years; range, 10 to 30 years.

Table Graphic Jump Location
Table 2. The Fifth Percentile Values of ΔFEV1 and %ΔFEV1 by Interval and Gender
Table Graphic Jump Location
Table 3. Fifth Percentile Values in Men of ΔFEV1 and %ΔFEV1 by Interval and Smoking Status at the Final Test
Table Graphic Jump Location
Table 4. Association Between Excessive Long-term Declines and Abnormal Short-term Change Defined by ΔFEV1 and %ΔFEV1 at Each Short-term Test Interval*
* 

Short-term change (ΔFEV1 or %ΔFEV1) classified as abnormal if value at or below the gender-specific fifth percentile of the distribution from Table 2.

Table Graphic Jump Location
Table 5. Diagnostic Values of Abnormal ΔFEV1 and %ΔFEV1 for Excessive Long-term FEV1 Decline by Interval*
* 

Short-term change (ΔFEV1 or %ΔFEV1) classified as abnormal if value at or below the gender-specific fifth percentile of the distribution from Table 2. LR+ of 2 to 5 indicates a fair clinical test result, 5 to 10 indicates good clinical test result, and > 10 indicates an excellent clinical test result.

Table Graphic Jump Location
Table 6. Diagnostic Values of Abnormal %ΔFEV1 for Excessive Long-term FEV1 Decline Using First and Tenth Percentile %ΔFEV1 Values as Cut Points, by Interval
* 

Short-term change classified as abnormal if the %ΔFEV1 value is at or below the first percentile (− 18.3, − 10.2, − 7.8, − 6.1, and − 5.5%/yr for 1- to 5-year intervals, respectively).

 

Short-term change classified as abnormal if the %ΔFEV1 value is at or below the tenth percentile (− 7.8, − 4.7, − 3.6, − 3.0, and − 2.5%/yr for the 1- to 5-year intervals, respectively).

. American Thoracic Society. (1979) Standardization of spirometry.Am Rev Respir Dis119,4-11
 
American Thoracic Society.. Standardization of spirometry: 1987 update; ATS statement.Am Rev Respir Dis1987;136,1285-1298. [CrossRef] [PubMed]
 
American Thoracic Society.. Standardization of spirometry: 1994 update; ATS statement.Am J Respir Crit Care Med1995;52,1107-1136
 
Quanjer, PH, Tammeling, GJ, Cotes, JE, et al Lung volumes and forced ventilatory flows: Report Working Party Standardization of Lung Function Tests, European Community for Steel and Coal. Official Statement of the European Respiratory Society.Eur Respir J1993;6(Suppl 16),5s-40s
 
Hankinson, J, Wagner, GR Medical screening using periodic spirometry for detection of chronic lung disease.Occup Med1993;8,353-361. [PubMed]
 
NHLBI/WHO workshop summary.. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease.Am J Respir Crit Care Med2001;163,1256-1276. [PubMed]
 
Burrows, B, Lebowitz, MD, Camilli, AE, et al Longitudinal changes in forced expiratory volume in one second in adults.Am Rev Respir Dis1986;133,974-980. [PubMed]
 
Clement, J, Van De Woestijne, KP Rapidly decreasing forced expiratory volume in one second or vital capacity and development of chronic airflow obstruction.Am Rev Respir Dis1982;125,553-558. [PubMed]
 
Wang, ML, Gunel, E, Petsonk, EL Design strategies for longitudinal spirometry studies: study duration and measurement frequency.Am J Respir Crit Care Med2000;162,2134-2138. [PubMed]
 
American Thoracic Society.. Lung function testing:selection of references values and interpretative strategies: ATS statement.Am Rev Respir Dis1991;144,1202-1218. [CrossRef] [PubMed]
 
Wang, ML, Petsonk, EL Repeated measures of FEV1over six to twelve months: what change is abnormal?J Occup Environ Med2004;46,591-595. [CrossRef] [PubMed]
 
Cody, RP, Smith, JK Applied statistics and the SAS programming language 4th ed.1997,83-86 Prentice-Hall. Upper Saddle River, NJ:
 
Gallagher, EJ Clinical utility of likelihood ratios.Ann Emerg Med1998;31,391-397. [CrossRef] [PubMed]
 
Feinstein, AR. Principles of medical statistics. 2002; Chapman & Hall/CRC Press. Boca Raton, FL:.
 
Jaeschke, R, Guyatt, G, Sackett, DL Users’ guides to the medical literature:III. How to use an article about a diagnostics test B: what are the results and will they help me in caring for my patients?JAMA1994;271,703-707. [CrossRef] [PubMed]
 
Glindmeyer, HW, Jones, RN, Diem, JE, et al Useful and extraneous variability in longitudinal assessment of lung function.Chest1987;92,877-882. [CrossRef] [PubMed]
 
ACOEM position statement.. Spirometry in the occupational setting. American College of Occupational and Environmental Medicine.J Occup Environ Med2000;42,228-245. [CrossRef] [PubMed]
 
Hnizdo, E, Yu, L, Freyder, L, et al The precision of longitudinal lung function measurements: monitoring and interpretation.Occup Environ Med2005;62,695-701. [CrossRef] [PubMed]
 
Wang, ML, Avashia, B, Petsonk, EL Factors affecting the recognition of excessive FEV1decline: analysis of 30 years of data from an industry-based monitoring program [abstract].Am J Respir Crit Care Med2005;171,A440
 
Whiting, P, Rutjes, A, Reitsma, J, et al Sources of variation and bias in studies of diagnostic accuracy.Ann Intern Med2004;140,189-202. [PubMed]
 
American College of Occupational and Environmental Medicine... Evidence based statements: evaluating pulmonary function change over time in the occupational setting.  Available at: www.acoem.org/guidelines/article.asp?ID=59. Accessed October 25, 2005.
 

Figures

Figure Jump LinkFigure 1. Distribution of long-term FEV1 slopes by gender among 1,884 chemical plant workers who participated in spirometry screening over ≥ 10 years and had at least five valid results. FEV1 slopes were calculated for each individual by simple linear regression using all valid spirometry results for that individual.Grahic Jump Location
Figure Jump LinkFigure 2. Mean FEV1 slopes by age and smoking status at baseline in 1,721 male chemical plant workers who participated in spirometry screening ≥ 10 years and had at least five valid results. FEV1 slopes were calculated for each individual by simple linear regression using all valid spirometry results for that individual. *Smoker vs former and never-smoker, p < 0.0001; age ≤ 25 years vs others, p < 0.0001; age 26 to 34 years vs age ≥ 35 years, p < 0.0001.Grahic Jump Location

Tables

Table Graphic Jump Location
Table 1. ΔFEV1 Values Over 1 to 5 Years and Long-term FEV1 Slope by Test Interval and Gender (n = 1,884)
* 

Mean, 18 years; range, 10 to 30 years.

Table Graphic Jump Location
Table 2. The Fifth Percentile Values of ΔFEV1 and %ΔFEV1 by Interval and Gender
Table Graphic Jump Location
Table 3. Fifth Percentile Values in Men of ΔFEV1 and %ΔFEV1 by Interval and Smoking Status at the Final Test
Table Graphic Jump Location
Table 4. Association Between Excessive Long-term Declines and Abnormal Short-term Change Defined by ΔFEV1 and %ΔFEV1 at Each Short-term Test Interval*
* 

Short-term change (ΔFEV1 or %ΔFEV1) classified as abnormal if value at or below the gender-specific fifth percentile of the distribution from Table 2.

Table Graphic Jump Location
Table 5. Diagnostic Values of Abnormal ΔFEV1 and %ΔFEV1 for Excessive Long-term FEV1 Decline by Interval*
* 

Short-term change (ΔFEV1 or %ΔFEV1) classified as abnormal if value at or below the gender-specific fifth percentile of the distribution from Table 2. LR+ of 2 to 5 indicates a fair clinical test result, 5 to 10 indicates good clinical test result, and > 10 indicates an excellent clinical test result.

Table Graphic Jump Location
Table 6. Diagnostic Values of Abnormal %ΔFEV1 for Excessive Long-term FEV1 Decline Using First and Tenth Percentile %ΔFEV1 Values as Cut Points, by Interval
* 

Short-term change classified as abnormal if the %ΔFEV1 value is at or below the first percentile (− 18.3, − 10.2, − 7.8, − 6.1, and − 5.5%/yr for 1- to 5-year intervals, respectively).

 

Short-term change classified as abnormal if the %ΔFEV1 value is at or below the tenth percentile (− 7.8, − 4.7, − 3.6, − 3.0, and − 2.5%/yr for the 1- to 5-year intervals, respectively).

References

. American Thoracic Society. (1979) Standardization of spirometry.Am Rev Respir Dis119,4-11
 
American Thoracic Society.. Standardization of spirometry: 1987 update; ATS statement.Am Rev Respir Dis1987;136,1285-1298. [CrossRef] [PubMed]
 
American Thoracic Society.. Standardization of spirometry: 1994 update; ATS statement.Am J Respir Crit Care Med1995;52,1107-1136
 
Quanjer, PH, Tammeling, GJ, Cotes, JE, et al Lung volumes and forced ventilatory flows: Report Working Party Standardization of Lung Function Tests, European Community for Steel and Coal. Official Statement of the European Respiratory Society.Eur Respir J1993;6(Suppl 16),5s-40s
 
Hankinson, J, Wagner, GR Medical screening using periodic spirometry for detection of chronic lung disease.Occup Med1993;8,353-361. [PubMed]
 
NHLBI/WHO workshop summary.. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease.Am J Respir Crit Care Med2001;163,1256-1276. [PubMed]
 
Burrows, B, Lebowitz, MD, Camilli, AE, et al Longitudinal changes in forced expiratory volume in one second in adults.Am Rev Respir Dis1986;133,974-980. [PubMed]
 
Clement, J, Van De Woestijne, KP Rapidly decreasing forced expiratory volume in one second or vital capacity and development of chronic airflow obstruction.Am Rev Respir Dis1982;125,553-558. [PubMed]
 
Wang, ML, Gunel, E, Petsonk, EL Design strategies for longitudinal spirometry studies: study duration and measurement frequency.Am J Respir Crit Care Med2000;162,2134-2138. [PubMed]
 
American Thoracic Society.. Lung function testing:selection of references values and interpretative strategies: ATS statement.Am Rev Respir Dis1991;144,1202-1218. [CrossRef] [PubMed]
 
Wang, ML, Petsonk, EL Repeated measures of FEV1over six to twelve months: what change is abnormal?J Occup Environ Med2004;46,591-595. [CrossRef] [PubMed]
 
Cody, RP, Smith, JK Applied statistics and the SAS programming language 4th ed.1997,83-86 Prentice-Hall. Upper Saddle River, NJ:
 
Gallagher, EJ Clinical utility of likelihood ratios.Ann Emerg Med1998;31,391-397. [CrossRef] [PubMed]
 
Feinstein, AR. Principles of medical statistics. 2002; Chapman & Hall/CRC Press. Boca Raton, FL:.
 
Jaeschke, R, Guyatt, G, Sackett, DL Users’ guides to the medical literature:III. How to use an article about a diagnostics test B: what are the results and will they help me in caring for my patients?JAMA1994;271,703-707. [CrossRef] [PubMed]
 
Glindmeyer, HW, Jones, RN, Diem, JE, et al Useful and extraneous variability in longitudinal assessment of lung function.Chest1987;92,877-882. [CrossRef] [PubMed]
 
ACOEM position statement.. Spirometry in the occupational setting. American College of Occupational and Environmental Medicine.J Occup Environ Med2000;42,228-245. [CrossRef] [PubMed]
 
Hnizdo, E, Yu, L, Freyder, L, et al The precision of longitudinal lung function measurements: monitoring and interpretation.Occup Environ Med2005;62,695-701. [CrossRef] [PubMed]
 
Wang, ML, Avashia, B, Petsonk, EL Factors affecting the recognition of excessive FEV1decline: analysis of 30 years of data from an industry-based monitoring program [abstract].Am J Respir Crit Care Med2005;171,A440
 
Whiting, P, Rutjes, A, Reitsma, J, et al Sources of variation and bias in studies of diagnostic accuracy.Ann Intern Med2004;140,189-202. [PubMed]
 
American College of Occupational and Environmental Medicine... Evidence based statements: evaluating pulmonary function change over time in the occupational setting.  Available at: www.acoem.org/guidelines/article.asp?ID=59. Accessed October 25, 2005.
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

CHEST Journal Articles
PubMed Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543