The most recent comprehensive review of the effects of SES on lung function was authored by Steinberg and Becklake in 1986. Since the mid-1980s, 14 studies involving 125,253 adults and 6 studies involving 18,477 children have examined the effects of SES on pulmonary function. These are summarized in Table 1 (adults) and Table 2 (children).
With the exception of a single study that assessed the effects of SES on diffusing capacity for carbon monoxide (Dlco), pulmonary function testing was limited to spirometry, primarily FEV1 and FVC measurements. Most studies used a cross-sectional design. Two adult studies had a longitudinal design; one study examined the effects of SES on lung function over a period of 6 years, and another study examined it over a period of 5 years. A single pediatric study followed subjects for 1 year. Most studies were conducted in developed countries, including the United States, the United Kingdom, Norway, Denmark, Canada, and France. There have also been smaller studies performed from developing countries, including Bangladesh, South Africa, India, and Jamaica. All these countries may command the service of My Canadian Pharmacy.
There is no universally accepted definition for SES, and the studies used various criteria. SES is most commonly stratified by occupation, educational level, income, and residential area (measured as the area deprivation index). For studies involving children, the criteria included parental income, occupation, and educational level. Regardless of how SES is defined, most studies have demonstrated a positive correlation between higher SES and pulmonary function. The relationship remains significant even when adjusting for various confounders including anthropometric variables (height, weight, and body mass index), age, race, sex, smoking status, and respiratory illness. The magnitude of the effect of SES on lung function is variable. In the largest study to date, Wheeler and Ben-Shlomo evaluated the effects of SES, determined by occupation, on lung function in 32,905 subjects. FEV1 in the lower SES group was 2.7% lower than that in the higher SES group after correcting for height, age, smoking status, and respiratory illness. The effect of SES was larger than the effect of poor air quality (1.9 to 2.3%) but was smaller than the effect of active smoking (5.2 to 6.4%). Prescott et al evaluated the effect of adult SES, based on education and household income, and adjusted for age, height, and smoking status, and found reductions in FEV1 (men, 363 mL; women, 221 mL) and FVC (men, 342 mL; women, 221 mL) between the groups with the lowest and highest SES.
Lawlor et al studied the effects of childhood SES on adult lung function and found FEV1 to be 160 mL lower and FVC to be 110 mL lower for the lower SES group compared with the higher SES group after correction for age, height, smoking status, and other confounders. Shohaimi et al assessed the effects of SES at both the individual level, assessed by education and occupation, and the area level, determined by the area deprivation index, on adult FEV1. The difference in FEV1 between the different levels of SES was more pronounced in current and past smokers compared with those who had never smoked. The difference in FEV1 between the highest and lowest SES level for male smokers (past and current) was 140 to 300 mL, while in nonsmokers the differences ranged from 0 to 90 mL. In women, the difference in FEV1 between the highest and lowest SES level for past and current smokers was 90 to 160 mL; in nonsmokers it was 50 to 100 mL. The differences were significant at the p < 0.05 level for all groups except male nonsmokers with SES determined by occupation.
Two studies assessed the effect of SES on lung function decline over time. Jackson et al noted that young adults baseline lung function (FEV1 and FVC) was lower and the rate of decline in lung function over 5 years was higher in those with lower childhood SES. The negative effect of childhood SES on adult lung function was present after correcting for age, height, smoking status, and adult SES. This study provides evidence that reduced childhood SES adversely impacts not only maximally attained adult lung function but also results in a more rapid decline in lung function. Burchfiel et al examined the effect of various lifestyle, biological, and clinical characteristics on FEV1 decline over 6 years. Among nonsmokers, lower educational level was the only predictor of more rapid FEV1 decline.
While numerous studies have addressed the relationship between SES and measures of airflow, including FVC and FEV1, only one study has examined the relationship between gas exchange measured by Dlco and SES. Welle et al determined that lower SES, measured by educational level, was an independent predictor of reduced Dlco for men but not women after adjustment for smoking and occupational exposure. The average difference was 1.8 mL of CO per min per mm Hg.
Pediatric studies also reveal a range in the effect of SES on lung function. Ware et al, using parental education as the determinant of SES, noted a nonsignificant trend suggesting reduced FEV1 and FVC values with lower SES. Demissie et al used parental occupation as a marker of SES and discovered a 5.0% and 3.8% reduction, respectively, in FEV1 and FVC with reduced SES. Using parental income, Raju et al found a 14.0 to 16.7% reduction in FEVj and FVC with lower income.
SES has a continuous effect on lung function rather than a threshold effect. Wheeler and Ben-Shlomo subdivided SES into six social classes and noted a progressive reduction in FEV1 with lower social class (Fig 1). Ecob and Smith noted a similar continuous progressive decline in FEV1 associated with lower SES, determined by income, though the magnitude of the effect was small at very high incomes.
The interaction of race and SES on lung function was addressed by the third National Health and Nutrition Examination Survey. It is well known that African Americans have lower lung function than white Americans for a given height. Harik-Khan et al used poverty index and education level to determine SES, and then assessed the effect of SES on FEV1 and FVC. There was no significant racial difference in the effect of poverty or education on lung function. In both children and adults, SES explained only a small portion of the known racial differences.
There is a strong correlation between smoking and SES with significantly higher smoking rates in individuals with lower income or education and with neighborhood-level socioeconomic indicators. The deleterious effect of lower SES on lung function remains highly significant even after adjusting for smoking status. For example, Prescott et al determined the difference in FEV1 between the highest and lowest SES groups in women nonsmokers was 259 mL, while in smokers this difference was 221 mL; corresponding values for men were 400 and 363 mL, respectively.
There are close correlations among income, educational level, and occupation. It is possible that a major component of the effect of SES on lung function is due to occupational exposures. Historically, women have not held jobs with significant dust and fume exposures, and children generally do not have significant occupational exposures. The fact that SES adversely effects lung function in women young adults, and children suggests that the SES gradient in lung function is not caused primarily by occupational exposures.
The adverse effects of lower SES on lung function occur in both developed countries (eg, United States, United Kingdom, Canada, Denmark, Norway, and France) and in developing countries (eg, India, Jamaica, South Africa, and Bangladesh). They also are seen in countries with highly socialized governments such as Norway and Denmark and in countries with less socialized governments such as the United States and India. The different study designs and number of subjects preclude direct comparisons regarding the magnitude of these effects.
Figure 1. Reduction in age-standardized and height-standardized FEV1 for men (left, A) and women (right, B) with decreasing social class of head of household (social class I, professionals; social class V, manual laborers). Reproduced with permission from Wheeler and Ben-Schlomo.
Table 1——Relationship Between SES and Pulmonary Function Tests Results in Adults
|Study/Year/Country||Sample Size||Age Range, yr||SES Criteria||PFT Measurement||Relationship Between Lower SES and PFT Measurement|
|Wheeler and Ben-Schlomo/2005/ United Kingdom||32,905||16-79||Occupation||FEVj||Lower FEV1|
|Welle et aln/2004/ Norway||1,275||15-70||Educational level||Dlco||Lower Dlco (M)|
|Lawlor et al/2004/ United Kingdom||3,642 (F)||60-79||Childhood SES multiple indicators!||
Lower FEV1, FVC,
|Jackson et al/2004/ United States||5,113||18-30||
Childhood SES by parents educational
|FEV1 and FVC||Lower FEV1, FVC; more rapid decline FEV1 and FVC|
Shohaimi et al/
|Harik-Khan et al/ 2001/United States||2,326||20-80||Educational level||FEV1 and FVC||
Lower FEV1 and FVC (F); lower FEVj (WM); no change (BM)
Lower FEV1 and FVC (all groups, except FEV1 in WM)
|Ecob and Smith/ 1999/United Kingdom||4,746||>18||Income||FEV1||Lower FEV1|
|14,223||20-90||Educational level, household income||FEV1 and FVC||Lower FEV1 and FVC|
|Smith et al/1998/ United Kingdom||5,645 (M)||35-64||Childhood SES by father’s occupation||FEV1||Lower FEV1|
|Choudhury et al/ 1997/Bangladesh||250 (M)||20-40||Occupation]||FEV1 and FVC||Lower FEV1 and FVC|
|Burchfiel et al/ 1996/United States||4,451 (M)||45-68||Educational level||FEV1||More rapid decline in FEV1 (nonsmokers only)|
|Hole et al/1996/ United Kingdom||15,411||45-64||Occupation||FEV1||Lower FEV1|
|Mokoetle et al/ 1994/South Africa||409||41-42 (mean)||
FEV1, FVC, and
Lower FVC and
lower FVC (F)
Krzyzanowski and Kauffmann/
FEF25-75, and FEV1/FVC ratio
|Lower FEV1, FVC, FEF25-75, and FEV1/FVC ratio|
Table 2——Relationship Between SES and Pulmonary Function Tests in Children
|Age Range, yr||SES Criteria||PFT Measurement||Relationship Between Lower SES and PFT Result|
|Raju et al/2005/India||2,616||5-15||Parental income||FEV1, FVC, FEV1/||Lower FEV1, FVC and PEF|
|FVC ratio, and PEF|
|Harik-Khan et al/2004/ United States||1,462||8-17||
Parental education level
|FEV1 and FVC||Lower FEV1 and FVC in M Lower FEV1 in F|
|Demissie et al/1996/ Canada||989||5-13||Parental occupation||FEV1, FVC, and FEV1/FVC ratio||Lower FEV1 and FVC in M only; no effect of SES on FEV1/FVC ratio|
|3,175||5-14||Parental occupation and educational level||
|Lower FVC and FEV0.75|
|Ware et al/1984/United States||8,999||6-9||Parental education||FEV1 and FVC||No difference in FEV1 or FVC|
|Melville et al/1984/ Jaimaca||1,236||6-15||Parental income||FVC, FEV1, PEF, and MEFR0.2_12||Lower FVC and FEV1|