We utilized data from the National Health and Nutrition Examination Survey (NHANES), a collaborative effort spearheaded by the Centers for Disease Control and Prevention and the National Centers for Health Statistics (NCHS) in the US. Ethical approval for the NHANES protocol was duly obtained from the Research Ethics Review Board of the NCHS, and all participants provided written informed consent. Our dataset, which was sourced from the NHANES website (www.cdc.gov/nchs/nhanes/), spans the 1988–1994 (NHANES III) and 2007–2012 cycles due to the availability of FEV0.5 data and the fact that these cycles encompassed a comprehensive array of demographic, examination, and questionnaire data.

This study included all baseline data from the participants who had complete data on FEV0.5. The main inclusion criteria were (1) age≥20 years; (2) non-pregnant; (3) acceptable spirometry; (4) qualifying FEV0.5 data; (5) complete demographic data and information on smoking status; and (6) complete follow-up data for all-cause mortality.

The majority of the participants in the NHANES III and NHANES 2007–2012 completed prebronchodilator pulmonary function measurements, while only a small number completed postbronchodilator pulmonary function measurements. Therefore, this study was based on an analysis of the prebronchodilator spirometry data. All pulmonary function measurements were obtained using the Ohio 822/827 Dry Rolled Volume Sealed Spirometer. In the NHANES study, spirometry was performed in accordance with previously recommended guidelines of the American Thoracic Society (ATS).

Stringent quality control measures were applied, necessitating the exclusion of participants who failed to meet the pulmonary function testing criteria. In the NHANES III (1988–1994), individuals with reproducible FEV1 and FVC measurements with ≥2 acceptable trials were included.7 In the NHANES 2007–2012, individuals with FEV1 and FVC that were considered to be of grade A (exceeding the ATS data collection standards) or grade B (meeting the ATS data collection standards) in terms of quality were included.8–10 Lower limit of normal (LLN) for FEV1 and FVC is calculated using the GLI online calculator with a race-neutral approach.11,12

All participants were systematically stratified into four quartiles based on their prebronchodilator FEV0.5 values, designated as group I (0mL to <1937mL), group II (≥1937mL to <2388mL), group III (≥2388mL to <2884mL), and group IV (≥2884mL).

The primary outcome was the all-cause mortality. The secondary outcomes were the risks of comorbidities and chronic respiratory symptoms. The all-cause mortality data were obtained from the comprehensive death certificate records of the National Death Index (NDI) of the NCHS. Information regarding comorbidities and chronic respiratory symptoms was gathered from the questionnaire data on the same date as the pulmonary function tests. This involved inquiring with the patients as to whether they had ever been informed that they had comorbidities by a medical professional, including a doctor or another healthcare provider. Data on chronic respiratory symptoms were gathered via patient self-reporting, wherein the participants were questioned about whether they had experienced cough, cold, sputum production, runny nose, or any other respiratory illness.13–15

Information on various demographic and health-related factors was gathered on the same date as the pulmonary function tests, including age, sex, race, smoking status, educational level and poverty income ratio from the NHANES household interviews. BMI was calculated as weight in kilograms divided by height in meters squared and grouped into four categories: underweight (<18.5kg/m2), normal (≥18.5kg/m2 to <25kg/m2), overweight (≥25.0kg/m2 to <30.0kg/m2), and obese (≥30.0kg/m2). Body surface area (BSA) was calculated using the following formula: BSA (m2)=(body weight [kg])0.425×(height [cm])0.725×0.007184.16 Race was categorized as Mexican–American, non-Hispanic White, non-Hispanic Black, or other race. Educational level was categorized as Less than 9th grade, 9–12th grade or above 12th grade. Poverty income ratio (PIR) was categorized as low-income (PIR<1.3), middle-income (3.50>PIR≥1.30) or high-income (PIR≥3.50). The criteria for classifying smoking status are as follows: “Have you smoked at least 100 cigarettes in your entire life?”, participants who answered “No” were classified as “never smokers.” Those who answered “Yes” were identified as smokers, and based on their answer to the question, “Do you smoke cigarettes now?”, they were classified as “current smokers” (“Yes”) or “former smokers” (“No”).

Continuous variables were compared between the groups by analysis of variance, while categorical variables were compared using Pearson's Chi-square test. The multivariable logistic regression model was used to estimate the association of FEV0.5 with the presence of comorbidities and the presence of chronic respiratory symptoms. Trends in these associations were calculated using quartiles as quasi-continuous variables in the multivariable logistic regression model.

Kaplan–Meier survival analyses were performed to identify differences in all-cause mortality between the groups. The multivariable Cox regression models were used to estimate the association between FEV0.5 and all-cause mortality risk. The proportional-hazards assumption was checked graphically using the Schoenfeld residual test. To further understand the association between FEV0.5 and all-cause mortality, we performed Cox proportional-hazards regression analyses with restricted cubic spline (RCS) analysis utilizing five knots. The trend in the association between FEV0.5 and all-cause mortality risk was calculated using quartiles as quasi-continuous variables in the multivariable Cox regression model.

Subgroup analyses were also performed to evaluate the impact of FEV0.5 on all-cause mortality across various subgroups stratified by age (20–40, 41–60, and 61–80 years), sex, race (Mexican–American, non-Hispanic White, non-Hispanic Black, other race), BMI (underweight, normal, overweight, obese), and smoking status (never smoker, former smoker, current smoker). To verify whether FEV0.5 adds prognostic value beyond FEV1 and FVC, we repeated the analysis in the subgroups with FEV1≥LLN, FVC≥LLN, and both FEV1 and FVC≥LLN. In cases of missing data, we used a deletion measure and abstained from using data interpolation. P<0.05 was considered statistically significant for all tests. R Studio, version 4.3.3, and SPSS, version 29.0, were used to conduct the statistical analyses.

Of the 50,492 participants from the NHANES III (n=20,050) and NHANES 2007–2012 (n=30,442), 25,357 were included in the present analysis after applying the eligibility criteria. Reasons for exclusion are shown in Fig. S1. The baseline characteristics of the participants are presented in Table 1. The median follow-up time was 308 months. The mean age was 46.1±7.2 years, and the mean BMI was 28.0±6.3kg/m2. Fig. S2 illustrates the FEV0.5 distribution at baseline. The mean FEV0.5 was 2412±699mL, with a median (interquartile range) FEV0.5 of 2388 (1937–2884) mL. There were 6352 participants in group I, 6339 in group II, 6328 in group III, and 6338 in group IV. Compared with individuals in groups II, III, and IV, individuals in group I tended to be older and have higher BMI.

Fig. 1 illustrates the results of all-cause mortality in the different groups. Significant differences in all-cause mortality were observed among the four groups (log-rank P<0.05). During follow-up, group I had 3054 deaths (48.1%), group II had 1474 deaths (23.3%), group III had 1051 deaths (16.6%), and group IV had 716 deaths (11.3%). Table 2 reports the results of the association between FEV0.5 and all-cause mortality risk in the different groups. In the univariable regression analysis, compared with individuals in group IV, individuals in groups I, II, and III had a higher risk of all-cause mortality (HRgroupI 6.70, 95% confidence interval [CI] 6.17–7.27; HRgroupII 2.31, 95% CI 2.11–2.53; HRgroupIII 1.56, 95% CI 1.42–1.72; all P<0.05; Ptrend<0.05). After adjustment for age, sex, smoking, body mass index, body surface area, race, educational level, poverty income ratio, and comorbidities (congestive heart failure, stroke, Asthma, chronic bronchitis, emphysema, cancer, diabetes, and hypertension), the multivariable regression analysis results were consistent. Compared with individuals in group IV, individuals in groups I, II, and III had a higher risk of all-cause mortality (HRgroupI 1.77, 95% CI 1.57–2.00; HRgroupII 1.28, 95% CI 1.15–1.43; HRgroupIII 1.15, 95% CI 1.04–1.28; all P<0.05; Ptrend<0.05). FEV0.5 was numerically higher than FEV3 and FEV3/FVC in its ability (the concordance index of the multivariable Cox regression model) to identify the risk of all-cause mortality, but the difference did not reach statistical significance (Fig. S3).

Fig. 1.

Mortality risk stratified by prebronchodilator forced expiratory volume in 0.5s levels at baseline. Definition of abbreviations: BD=bronchodilator; FEV0.5=forced expiratory volume in 0.5s.

(0.4MB).

Table 3 illustrates the results of the association between FEV0.5 and the presence of comorbidities and chronic respiratory symptoms in each group. In terms of chronic respiratory symptoms, the univariable regression analysis revealed that individuals in groups I, II, and III exhibited an elevated risk of chronic cough, wheezing, and dyspnea than those in group IV. These associations persisted in the multivariable regression analysis, which also identified an increased presence of chronic cough, chronic phlegm, wheezing, and dyspnea in groups I, II, and III after adjustment for age, sex, smoking, body mass index, body surface area, race, educational level, and poverty income ratio.

In terms of comorbidities, the multivariable regression model revealed a heightened risk of congestive heart failure, asthma, chronic bronchitis, emphysema, cancer, and diabetes mellitus among individuals in groups I, II, and III compared with those in group IV. Similarly, for stroke and hypertension, an increased presence was observed exclusively in groups I and II when compared with group IV.

Fig. 2 illustrates the association between FEV0.5 and all-cause mortality risk across the various subgroups stratified by sex, age, BMI, race, and smoking status. The association between FEV0.5 and all-cause mortality risk remained almost consistent even after stratifying the participants into subgroups by sex (male, female), age (20–40, 41–60, 61–80 years), race (Mexican–American, non-Hispanic White, non-Hispanic Black), BMI (normal, overweight, obese), and smoking status (never smoker, former smoker, current smoker). Among the male subgroup, individuals in groups I, II, and III exhibited an increased all-cause mortality risk when compared with individuals in group IV. However, among females, only those in group I demonstrated a significantly elevated risk of all-cause mortality when compared with individuals in group IV. In the subgroup stratified by other race and in the underweight subgroup, there was no statistically significant difference in the risk of all-cause mortality between individuals in groups I, II, and III and individuals in group IV. In subgroup with FEV1>LLN, FVC>LLN, or both FEV1 and FVC>LLN, individuals in groups I, II, and III had a higher risk of all-cause mortality after adjustment compared with individuals in group IV (Table S1).

Fig. 2.

Multivariable associations between prebronchodilator forced expiratory volume in 0.5s and risk of all-cause mortality. Definition of abbreviations: BMI=body mass index; BD=bronchodilator; FEV0.5=forced expiratory volume in 0.5s. Adjust for age, sex, smoking, body mass index, body surface area, race, educational level, poverty income ratio, and comorbidities.

(0.81MB).

Fig. 3 illustrates the non-linear association between FEV0.5 and all-cause mortality risk. The association between FEV0.5 and all-cause mortality risk manifested as an L-shaped curve. With the exception of the other race subgroup and the underweight subgroup, a similar trend was observed in all subgroups stratified by sex, smoking status, age, race, and BMI.

Fig. 3.

Nonlinear associations between prebronchodilator forced expiratory volume in 0.5s and risk of all-cause mortality. Definition of abbreviations: BMI=body mass index; BD=bronchodilator; FEV0.5=forced expiratory volume in 0.5s. Adjust for age, sex, smoking, body mass index, body surface area, race, educational level, poverty income ratio, and comorbidities. Panel A shows the non-linear relationship between FEV0.5 and risk of all-cause mortality in all participants. Panel B shows the non-linear relationship between FEV0.5 and the risk of all-cause mortality in the male and female groups. Panel C shows the non-linear relationship between FEV0.5 and the risk of all-cause mortality in populations with different smoking status. Panel D shows the non-linear relationship between FEV0.5 and risk of all-cause mortality for people of different races. Panel E shows the non-linear relationship between FEV0.5 and risk of all-cause mortality in different age groups. Panel F shows the non-linear relationship between FEV0.5 and risk of all-cause mortality for people with different BMI.

(0.48MB).