TY - JOUR
AU - Navas-Acien,, Ana
AB - Abstract Rice accumulates arsenic, an established lung toxicant. Little is known about the association of rice consumption with arsenic-related health effects, particularly interstitial lung disease. Between 2000 and 2002, 6,814 white, black, Hispanic, and Chinese adults from 6 US cities were enrolled in the Multi-Ethnic Study of Atherosclerosis. We included 2,250 participants who had spirometry data, 2,557 with full-lung computed tomography (CT) scans, and 5,710 with cardiac CT scans. Rice consumption and 310 participants with urinary arsenic were assessed at baseline. Spirometry and full-lung CT-derived measures of total lung capacity and high attenuation area (HAA), and interstitial lung abnormalities were measured at examination 5. Cardiac CT-derived HAA was measured at 1–3 visits. Twelve percent of participants reported eating at least 1 serving of rice daily. Comparing data between that group with those who ate less than 1 serving weekly, the mean difference for forced vital capacity was −102 (95% confidence interval (CI): −198, −7) mL, and for forced expiratory volume in 1 second was −90 (95% CI: −170, −11) mL after adjustment for demographics, anthropometrics, dietary factors, and smoking. The cross-sectional adjusted percent difference for total lung capacity was −1.33% (95% CI: −4.29, 1.72) and for cardiac-based HAA was 3.66% (95% CI: 1.22, 6.15). Sensitivity analyses for urinary arsenic were consistent with rice findings. Daily rice consumption was associated with reduced lung function and greater cardiac-based HAA. arsenic, interstitial lung disease, Oryza, spirometry Rice accumulates arsenic (1). Rice absorbs arsenic more readily than other crops, in part because it is grown in flooded conditions (2). Research on the health effects of arsenic has been focused mostly on arsenic-contaminated groundwater, but in populations where levels of arsenic in water are low, the major source of arsenic exposure is rice (3). Thus, it is important to determine the degree to which rice poses a health risk for populations such as those in the United States, especially in urban areas where the primary source of arsenic exposure likely is the diet (4). Arsenic exposure is associated with chronic lung disease. In a recent meta-analysis, exposure to arsenic in water was consistently associated with lower forced vital capacity (FVC) and forced expiratory volume in 1 second (FEV1) but not with the FEV1-to-FVC ratio at high levels of arsenic exposure (ranging from 11 to 860 μg/L) (5). In a study conducted in West Bengal, researchers found a 10-fold higher rate of computed tomography (CT)-identified bronchiectasis, a relatively rare chronic lung disease, in people with versus those without arsenic-induced skin lesions (6) after controlling for chronic infections like tuberculosis. These findings in West Bengal are consistent with markedly increased rates of death resulting from bronchiectasis after an episode of exposure to extremely high levels of arsenic in water in northern Chile from 1958 to 1970 (7, 8). However, little is known about the relation between arsenic and lung function at lower exposure levels (<10 μg/L) in the United States. In the United States, chronic lower respiratory diseases, including interstitial lung disease, are the fourth leading cause of death (9, 10). Characterized by inflammation and fibrosis of the pulmonary parenchyma (11), interstitial lung disease is associated with reduced lung volumes, including FVC without a reduction in the FEV1-to-FVC ratio. Interstitial lung abnormalities (ILAs) are a well-validated, qualitative CT measure of subclinical interstitial lung disease (11, 12), whereas high attenuation area (HAA) is a quantitative CT measure of subclinical interstitial lung disease and lung fibrosis (13). Interstitial lung disease pathogenesis is characterized, in part, by increased lung inflammation and extracellular matrix production, reflected in higher surfactant protein-A (SP-A) and matrix metalloproteinase-7 (MMP-7) levels (14, 15). Although there are several inhaled exposures, including some metals, that are known to cause interstitial lung disease, relatively little is known about ingested environmental exposures contributing to interstitial lung disease (11). Our objective for this study was to determine the relationship between rice consumption and mechanistically relevant markers of subclinical interstitial lung disease. In secondary analyses, we assessed the association of subclinical interstitial lung disease measures with consistent long-term rice intake and urinary arsenic (uAs) in a subset of participants. METHODS Study sample The Multi-Ethnic Study of Atherosclerosis (MESA) is an ongoing, multicenter, longitudinal cohort study of community-dwelling adults that is designed to investigate subclinical cardiovascular disease (16). Between 2000 and 2002, 6,814 men and women aged 45–84 years were recruited to MESA from 6 communities (Baltimore, Maryland; Chicago, Illinois; Los Angeles, California; St. Paul, Minnesota; New York, New York; Salem, North Carolina). Participants are white, black, Hispanic or Chinese/Chinese-American. Exclusion criteria included clinical cardiovascular disease, weight more than 300 lb (136.1 kg), pregnancy, and any impediment to long-term participation. Protocols and examinations were approved by the institutional review boards of all collaborating institutions and the National Heart Lung and Blood Institute. All participants provided informed consent. The sample sizes available for analyses based on spirometry, full-lung CT scans, and lung-related biomarkers are shown in Figure 1. The final sample size for cardiac CT analyses was 5,710 participants with up to 3 CT scans available. Arsenic was measured in a random-stratified sample of 310 participants at baseline, resulting in 310 participants with detected arsenic levels and with cardiac CT scans, 139 with full-lung CT scans, 131 with spirometry data, and 46 with serum biomarker measures. Figure 1. Open in new tabDownload slide Sample size flow charts for analyses based on rice consumption and urine arsenic levels in the Multi-Ethnic Study of Atherosclerosis (MESA). A) A total of 6,814 participants were recruited to MESA between 2000 and 2002. Rice intake was assessed via food frequency questionnaire (FFQ) at baseline and examination 5. Participants missing rice intake data (n = 658 excluded), with unreliable dietary information (n = 438 excluded), or missing adjustment variable data (n = 8 missing education) at baseline were excluded. Cardiac computed tomography (CT) scans were used to estimate percent high attenuation area (HAA) longitudinally. A full-lung CT scan was conducted at examination 5 among 3,113 participants and used to assess HAA and total lung capacity (TLC) cross-sectionally (excluding 368 participants missing rice intake data, 185 with unreliable dietary information, 2 missing weight data, and 1 missing education data). Interstitial lung abnormalities (ILAs) were also read from full-lung CT scans among 2,420 participants (excluding 226 missing rice intake data, 154 with unreliable dietary information, 15 missing weight data, and 1 missing education data). Spirometry (forced vital capacity in 1 second (FEV1), forced vital capacity (FVC), and FEV1-to-FVC ratio) was performed as part of the MESA Lung Study at examination 5 among 2,741 participants (excluding 322 missing rice intake data, 162 unreliable dietary information, 6 with poor spirometry results, and 1 missing education data). Serum biomarkers, including surfactant protein-A and matrix metalloproteinase-7, were measured in 1,228 participants at baseline (excluding 118 missing rice data, 93 with unreliable dietary information, and 1 missing education data). B) Urine arsenic (As) was measured in a stratified random sample of 310 participants at baseline. Among participants with urinary arsenic, data for those with cardiac CT scans, serum biomarkers, spirometry, and full-lung CT scans are shown in the figure. Figure 1. Open in new tabDownload slide Sample size flow charts for analyses based on rice consumption and urine arsenic levels in the Multi-Ethnic Study of Atherosclerosis (MESA). A) A total of 6,814 participants were recruited to MESA between 2000 and 2002. Rice intake was assessed via food frequency questionnaire (FFQ) at baseline and examination 5. Participants missing rice intake data (n = 658 excluded), with unreliable dietary information (n = 438 excluded), or missing adjustment variable data (n = 8 missing education) at baseline were excluded. Cardiac computed tomography (CT) scans were used to estimate percent high attenuation area (HAA) longitudinally. A full-lung CT scan was conducted at examination 5 among 3,113 participants and used to assess HAA and total lung capacity (TLC) cross-sectionally (excluding 368 participants missing rice intake data, 185 with unreliable dietary information, 2 missing weight data, and 1 missing education data). Interstitial lung abnormalities (ILAs) were also read from full-lung CT scans among 2,420 participants (excluding 226 missing rice intake data, 154 with unreliable dietary information, 15 missing weight data, and 1 missing education data). Spirometry (forced vital capacity in 1 second (FEV1), forced vital capacity (FVC), and FEV1-to-FVC ratio) was performed as part of the MESA Lung Study at examination 5 among 2,741 participants (excluding 322 missing rice intake data, 162 unreliable dietary information, 6 with poor spirometry results, and 1 missing education data). Serum biomarkers, including surfactant protein-A and matrix metalloproteinase-7, were measured in 1,228 participants at baseline (excluding 118 missing rice data, 93 with unreliable dietary information, and 1 missing education data). B) Urine arsenic (As) was measured in a stratified random sample of 310 participants at baseline. Among participants with urinary arsenic, data for those with cardiac CT scans, serum biomarkers, spirometry, and full-lung CT scans are shown in the figure. Measurements Rice consumption Usual dietary intake over the past year was assessed at baseline (2000–2002) and at examination 5 (2010–2011) using a modified Block-style, 120-item food frequency questionnaire (17). We abstracted all variables from the food frequency questionnaire related to rice intake (Web Table 1, available at https://academic.oup.com/aje). There were 4 rice-related items, 1) white/Mexican/sticky rice, 2) brown/wild rice, 3) fried rice, and 4) arroz con pollo (“rice with chicken,” a mix of rice, chicken, and vegetables). Brown rice and mixed rice dishes (i.e., fried rice or arroz con pollo) were combined with white rice into 1 variable of overall rice intake and categorized into 3 groups on the basis of how often participants reported eating rice: less than 1 serving per week, 1–6 servings per week, and at least 1 serving per day. Total caloric intake was estimated from the food frequency questionnaire. Participants who provided unreliable dietary information (men with intake of <800 or >4,000 kcal/day or women with <500 or >3,500 kcal/day) were excluded (18). To account for healthy dietary patterns at baseline, we used the Alternate Healthy Eating Index total score (19), using scoring methods adapted to combine the food frequency questionnaire data available in MESA, including questions on intake of vegetable, fruits, nuts and soy protein, red meat, cereal fiber, fats, multivitamin use, and alcohol consumption (20). Long-term rice intake was defined among participants whose food frequency questionnaires at examination 1 and 5 indicated consistent rice intake patterns (n = 2,557) (Web Tables 2–4). Urinary arsenic Baseline urine samples were stored at below −70°C until analyses. Speciated arsenic concentrations (i.e., inorganic arsenic, methylarsenate, dimethylarsenate, and arsenobetaine) were measured using anion-exchange high performance liquid chromatography (Agilent 1100; Agilent Technologies, Waldbronn, Germany) coupled with inductively coupled plasma mass spectrometry (Agilent 7700x ICPMS; Agilent Technologies) (21). The limit of detection was 0.1 μg/L for each of inorganic arsenic, methylarsenate, dimethylarsenate, and arsenobetaine. The percentages of participants with uAs concentrations below the limit of detection were 45.8%, 14.2%, 0%, and 3.9% for inorganic arsenic, methylarsenate, dimethylarsenate, and arsenobetaine, respectively. The interassay coefficients of variation for inorganic arsenic, methylarsenate, dimethylarsenate, and arsenobetaine were 6.0%, 6.5%, 5.9%, and 6.5%, respectively. For participants with uAs species below the limits of detection, we assigned a level of the limit of detection divided by square root of 2. To account for the influence nontoxic seafood arsenicals have on toxic arsenicals (i.e., inorganic arsenic, methylarsenate, dimethylarsenate) at low levels of exposure, urinary concentrations of inorganic arsenic, methylarsenate, and dimethylarsenate were regressed on arsenobetaine (a nontoxic species used as a marker of overall exposure to seafood arsenicals). The residuals of this model reflect inorganic and methylated arsenic species that are not related to recent seafood intake (22). To have levels of exposure that are easily communicable, we added the mean concentration of the corresponding arsenic species (inorganic arsenic, methylarsenate, dimethylarsenate) among participants with low arsenobetaine (<1 μg/L) to the residuals (22). uAs is presented as the sum of arsenobetaine-corrected inorganic arsenic, methylarsenate, and dimethylarsenate. Lung function As part of the MESA Lung Study, 2,741 MESA participants underwent spirometry testing in 2010–2011. The participants were randomly sampled among those who consented to genetic analyses, underwent baseline measures of endothelial function, and attended a MESA follow-up examination during MESA Lung Study recruitment (23). Pre-bronchodilator spirometry was conducted in accordance with American Thoracic Society and European Respiratory Society guidelines and as previously reported (24, 25). Briefly, all participants performed at least 3 acceptable maneuvers on a dry rolling-seal spirometer (Occupational Marketing, Inc., Houston, Texas). Each test was graded for quality on a 5-letter scale (A–D, F) (25). All spirometry examinations were reviewed with automated quality checks and manual clinician review. Low-quality spirometry was defined as a quality score of C (at least 2 acceptable curves with both FVC and FEV1 values repeatable within 250 mL). Lung structure on CT scans CT measures were obtained from a full-lung CT scan in 2010–2011 and from cardiac CT scans at baseline and follow-up examinations (in 2000–2002, 2002–2004, 2004–2005, and 2005–2007). Protocols have been described (26, 27). Briefly, HAA was measured on noncontrast cardiac CT scans on multidetector CT scanners or electron beam tomography scanners. Each participant underwent 2 sequential scans on separate breath holds in succession at full inspiration. The scan with higher air volume was used for analyses, except when scan quality was discordant, in which case the higher-quality scan was used. Quantitative image attenuation, used to calculate HAA and TLC, was measured by trained readers using a modified version of the Pulmonary Analysis Software at the University of Iowa Advanced Pulmonary Physiomic Imaging Laboratory (27). All participants had at least 1 cardiac CT scan from baseline and may have had up to 2 additional cardiac CT scans from follow-up examinations. Cardiac CT scans image approximately 65% of total lung volume, excluding most of the upper lobes but capturing most of the lower lobes. A subset of participants had a full-lung CT scan (n = 2,557). HAA was defined as the percentage of the total imaged lung volume having CT attenuation between −600 and −250 Hounsfield units (28) from cardiac CT and from full-lung CT scans. ILAs were determined from full-lung CT scans by 1 of 5 board-certified radiologists, as previously described (29, 30). Biomarkers MMP-7 and SP-A levels were measured from a subset of participants’ baseline serum samples (n = 1,016) and methods have been described previously (29). Other variables Age, sex, race/ethnicity, study site or city, educational attainment, total caloric intake, and dietary variables used to create the Alternate Healthy Eating Index composite score were assessed at baseline. Height, weight, smoking status, pack-years of smoking, and CT scanner manufacturer were assessed at baseline and at follow-up examinations. Statistical analysis For spirometry (FEV1, FVC, and FEV1-to-FVC ratio) and full-lung CT endpoints (TLC, HAA, ILA), we used linear and/or logistic regression modeling baseline rice consumption as the main exposure variable; with those eating less than 1 serving of rice weekly as the reference group. In model 1, we adjusted for physical factors that influence lung function, including sex, age, height, weight, race/ethnicity, study site, education, total caloric intake, and the Alternate Healthy Eating Index score. In model 2, we additionally adjusted for smoking status and pack-years of smoking. For cardiac-based HAA, which had repeated measures, we used linear mixed-effect models (Web Appendix 1) with similar adjustment to that in model 2 and also adjusted for CT-related factors, including total imaged lung volume, percent emphysema, and weight greater than 220 lb (99.8 kg; for radiation dose level) and scanner manufacturer. For the lung biomarkers, SP-A and MMP-7, which were measured at baseline only, we used linear regression with similar model adjustments. Spirometry endpoints were modeled in their original scales (nontransformed) and the model coefficients reflect mean differences. TLC, HAA, SP-A, and MMP-7 were right-skewed and log-transformed before analysis and results are presented as percent differences using the formula [(eβ)−1] × 100, where β is the regression coefficient for the endpoint per rice category comparison. In exploratory analyses, we used uAs instead of rice. uAs was modeled as continuous (log-transformed) and was standardized by urinary creatinine level. Results are presented per interquartile range increase in uAs. In sensitivity analyses, we restricted analyses to participants who had consistent rice intake as determined at examinations 1 and 5. We assessed effect modification between rice and outcomes by sex, age, race/ethnicity, and smoking status using cross-products of these variables with rice. Statistical analyses were performed using R, version 1.0.136 (R Foundation for Statistical Computing, Vienna, Austria). RESULTS Participant characteristics Twelve percent of participants reported eating at least 1 serving of rice daily (Table 1). Among participants with uAs measurements, rice consumption was positively correlated with uAs (Spearman ρ, 0.37, 95% confidence interval (CI): 0.27, 0.46). Daily rice eaters had lower body mass index and caloric intake, and were more likely to be never smokers. Participants who reported eating at least 1 serving of rice daily were predominantly Chinese (72%) or Hispanic (21%); 5% were black and 2% were white. Participants who reported eating at least 1 serving of rice daily were predominantly located in Los Angeles, California (59%), Chicago, Illinois (23%), and New York, New York (12%); fewer were from St. Paul, Minnesota (4%), Baltimore, Maryland (2%), and Salem, North Carolina (<1%). Table 1. Participant Characteristics at Baseline Stratified by Rice Consumption Category in the Multi-Ethnic Study of Atherosclerosis, 2000–2002 Characteristic Rice Consumption at Baseline Overall (n = 5,710) <1 serving/week (n = 2,312) 1–6 servings/week (n = 2,704) ≥1 servings/day (n = 694) No. % Median (IQR) No. % Median (IQR) No. % Median (IQR) No. % Median (IQR) Female sex 2,618 45.8 1,015 43.9 1,287 47.6 316 45.5 Age, years 62 (53–70) 64 (55–71) 61 (52–69) 63 (54–71) Height, cm 165 (159–173) 167 (160–175) 166 (159–173) 160 (155–168) Weight, lba 169 (145–196) 176 (151–201) 170 (148–196) 140 (125–162) Body mass indexb 28 (24–31) 28 (25–32) 28 (25–31) 25 (22–27) Energy intake, kcal 1,390 (1,028–1,875) 1,304 (980–1,747) 1,507 (1,110–2,015) 1,217 (901–1,702) Alternative Healthy Eating Index 42 (34–51) 40 (32–49) 43 (35–52) 44 (38–51) Study site  Los Angeles, California 1,146 20.1 183 7.9 552 20.4 411 59.2  Chicago, Illinois 1,044 18.3 394 17 489 18.1 161 23.2  New York, New York 815 14.3 221 9.6 514 19 80 11.5  St. Paul, Minnesota 950 16.6 511 22.1 340 12.6 12 1.7  Baltimore, Maryland 863 15.1 498 21.5 425 15.7 27 3.9  Salem, North Carolina 892 15.6 505 21.8 384 14.2 3 0.4 Race/ethnicity  White 2,327 40.8 1,367 59.1 949 35.1 11 1.6  Black 1,432 25.1 680 29.4 716 26.5 36 5.2  Hispanic 1,248 21.9 248 10.7 851 31.5 149 21.5  Chinese 703 12.3 17 0.7 188 7 498 71.8 Education  <High school 2,617 45.8 1,214 52.5 1,150 42.5 253 36.5  High school 1,016 17.8 230 9.9 530 19.6 256 36.9  >High school 2,077 36.4 868 37.5 1,024 37.9 185 26.7 Smoking status  Never 2,648 46.4 899 38.9 1,296 47.9 453 65.3  Former 2,305 40.4 1,075 46.5 1,053 38.9 177 25.5  Current 757 13.3 338 14.6 355 13.1 64 9.2 Pack-years of smokingc 18 (7–36) 21 (8–38) 16 (6–33) 18 (6–37) uAs-to-creatinine ratio, μg/gd 3 (2–5) 2 (1–3) 3 (2–5) 4 (3–6) uAs, μg/Ld 3 (2–5) 2 (1–3) 3 (2–5) 5 (4–7) Urine creatinine, mg/dLd 114 (60–164) 102 (55–168) 117 (54–164) 118 (78–155) Characteristic Rice Consumption at Baseline Overall (n = 5,710) <1 serving/week (n = 2,312) 1–6 servings/week (n = 2,704) ≥1 servings/day (n = 694) No. % Median (IQR) No. % Median (IQR) No. % Median (IQR) No. % Median (IQR) Female sex 2,618 45.8 1,015 43.9 1,287 47.6 316 45.5 Age, years 62 (53–70) 64 (55–71) 61 (52–69) 63 (54–71) Height, cm 165 (159–173) 167 (160–175) 166 (159–173) 160 (155–168) Weight, lba 169 (145–196) 176 (151–201) 170 (148–196) 140 (125–162) Body mass indexb 28 (24–31) 28 (25–32) 28 (25–31) 25 (22–27) Energy intake, kcal 1,390 (1,028–1,875) 1,304 (980–1,747) 1,507 (1,110–2,015) 1,217 (901–1,702) Alternative Healthy Eating Index 42 (34–51) 40 (32–49) 43 (35–52) 44 (38–51) Study site  Los Angeles, California 1,146 20.1 183 7.9 552 20.4 411 59.2  Chicago, Illinois 1,044 18.3 394 17 489 18.1 161 23.2  New York, New York 815 14.3 221 9.6 514 19 80 11.5  St. Paul, Minnesota 950 16.6 511 22.1 340 12.6 12 1.7  Baltimore, Maryland 863 15.1 498 21.5 425 15.7 27 3.9  Salem, North Carolina 892 15.6 505 21.8 384 14.2 3 0.4 Race/ethnicity  White 2,327 40.8 1,367 59.1 949 35.1 11 1.6  Black 1,432 25.1 680 29.4 716 26.5 36 5.2  Hispanic 1,248 21.9 248 10.7 851 31.5 149 21.5  Chinese 703 12.3 17 0.7 188 7 498 71.8 Education  <High school 2,617 45.8 1,214 52.5 1,150 42.5 253 36.5  High school 1,016 17.8 230 9.9 530 19.6 256 36.9  >High school 2,077 36.4 868 37.5 1,024 37.9 185 26.7 Smoking status  Never 2,648 46.4 899 38.9 1,296 47.9 453 65.3  Former 2,305 40.4 1,075 46.5 1,053 38.9 177 25.5  Current 757 13.3 338 14.6 355 13.1 64 9.2 Pack-years of smokingc 18 (7–36) 21 (8–38) 16 (6–33) 18 (6–37) uAs-to-creatinine ratio, μg/gd 3 (2–5) 2 (1–3) 3 (2–5) 4 (3–6) uAs, μg/Ld 3 (2–5) 2 (1–3) 3 (2–5) 5 (4–7) Urine creatinine, mg/dLd 114 (60–164) 102 (55–168) 117 (54–164) 118 (78–155) Abbreviations: IQR, interquartile range; uAs, urinary arsenic. a Metric conversions to kg across row: 76.7 (65.8–88.9); 78.3 (68.5–91.2); 77.1 (67.1–88.9); 63.5 (56.7–73.5). b Weight (kg)/height (m)2. c Among former or current smokers only. d uAs reflects sum of inorganic arsenic, methylarsenate, and dimethylarsenate after correcting for seafood arsenicals among subset of 310 participants. Table 1. Participant Characteristics at Baseline Stratified by Rice Consumption Category in the Multi-Ethnic Study of Atherosclerosis, 2000–2002 Characteristic Rice Consumption at Baseline Overall (n = 5,710) <1 serving/week (n = 2,312) 1–6 servings/week (n = 2,704) ≥1 servings/day (n = 694) No. % Median (IQR) No. % Median (IQR) No. % Median (IQR) No. % Median (IQR) Female sex 2,618 45.8 1,015 43.9 1,287 47.6 316 45.5 Age, years 62 (53–70) 64 (55–71) 61 (52–69) 63 (54–71) Height, cm 165 (159–173) 167 (160–175) 166 (159–173) 160 (155–168) Weight, lba 169 (145–196) 176 (151–201) 170 (148–196) 140 (125–162) Body mass indexb 28 (24–31) 28 (25–32) 28 (25–31) 25 (22–27) Energy intake, kcal 1,390 (1,028–1,875) 1,304 (980–1,747) 1,507 (1,110–2,015) 1,217 (901–1,702) Alternative Healthy Eating Index 42 (34–51) 40 (32–49) 43 (35–52) 44 (38–51) Study site  Los Angeles, California 1,146 20.1 183 7.9 552 20.4 411 59.2  Chicago, Illinois 1,044 18.3 394 17 489 18.1 161 23.2  New York, New York 815 14.3 221 9.6 514 19 80 11.5  St. Paul, Minnesota 950 16.6 511 22.1 340 12.6 12 1.7  Baltimore, Maryland 863 15.1 498 21.5 425 15.7 27 3.9  Salem, North Carolina 892 15.6 505 21.8 384 14.2 3 0.4 Race/ethnicity  White 2,327 40.8 1,367 59.1 949 35.1 11 1.6  Black 1,432 25.1 680 29.4 716 26.5 36 5.2  Hispanic 1,248 21.9 248 10.7 851 31.5 149 21.5  Chinese 703 12.3 17 0.7 188 7 498 71.8 Education  <High school 2,617 45.8 1,214 52.5 1,150 42.5 253 36.5  High school 1,016 17.8 230 9.9 530 19.6 256 36.9  >High school 2,077 36.4 868 37.5 1,024 37.9 185 26.7 Smoking status  Never 2,648 46.4 899 38.9 1,296 47.9 453 65.3  Former 2,305 40.4 1,075 46.5 1,053 38.9 177 25.5  Current 757 13.3 338 14.6 355 13.1 64 9.2 Pack-years of smokingc 18 (7–36) 21 (8–38) 16 (6–33) 18 (6–37) uAs-to-creatinine ratio, μg/gd 3 (2–5) 2 (1–3) 3 (2–5) 4 (3–6) uAs, μg/Ld 3 (2–5) 2 (1–3) 3 (2–5) 5 (4–7) Urine creatinine, mg/dLd 114 (60–164) 102 (55–168) 117 (54–164) 118 (78–155) Characteristic Rice Consumption at Baseline Overall (n = 5,710) <1 serving/week (n = 2,312) 1–6 servings/week (n = 2,704) ≥1 servings/day (n = 694) No. % Median (IQR) No. % Median (IQR) No. % Median (IQR) No. % Median (IQR) Female sex 2,618 45.8 1,015 43.9 1,287 47.6 316 45.5 Age, years 62 (53–70) 64 (55–71) 61 (52–69) 63 (54–71) Height, cm 165 (159–173) 167 (160–175) 166 (159–173) 160 (155–168) Weight, lba 169 (145–196) 176 (151–201) 170 (148–196) 140 (125–162) Body mass indexb 28 (24–31) 28 (25–32) 28 (25–31) 25 (22–27) Energy intake, kcal 1,390 (1,028–1,875) 1,304 (980–1,747) 1,507 (1,110–2,015) 1,217 (901–1,702) Alternative Healthy Eating Index 42 (34–51) 40 (32–49) 43 (35–52) 44 (38–51) Study site  Los Angeles, California 1,146 20.1 183 7.9 552 20.4 411 59.2  Chicago, Illinois 1,044 18.3 394 17 489 18.1 161 23.2  New York, New York 815 14.3 221 9.6 514 19 80 11.5  St. Paul, Minnesota 950 16.6 511 22.1 340 12.6 12 1.7  Baltimore, Maryland 863 15.1 498 21.5 425 15.7 27 3.9  Salem, North Carolina 892 15.6 505 21.8 384 14.2 3 0.4 Race/ethnicity  White 2,327 40.8 1,367 59.1 949 35.1 11 1.6  Black 1,432 25.1 680 29.4 716 26.5 36 5.2  Hispanic 1,248 21.9 248 10.7 851 31.5 149 21.5  Chinese 703 12.3 17 0.7 188 7 498 71.8 Education  <High school 2,617 45.8 1,214 52.5 1,150 42.5 253 36.5  High school 1,016 17.8 230 9.9 530 19.6 256 36.9  >High school 2,077 36.4 868 37.5 1,024 37.9 185 26.7 Smoking status  Never 2,648 46.4 899 38.9 1,296 47.9 453 65.3  Former 2,305 40.4 1,075 46.5 1,053 38.9 177 25.5  Current 757 13.3 338 14.6 355 13.1 64 9.2 Pack-years of smokingc 18 (7–36) 21 (8–38) 16 (6–33) 18 (6–37) uAs-to-creatinine ratio, μg/gd 3 (2–5) 2 (1–3) 3 (2–5) 4 (3–6) uAs, μg/Ld 3 (2–5) 2 (1–3) 3 (2–5) 5 (4–7) Urine creatinine, mg/dLd 114 (60–164) 102 (55–168) 117 (54–164) 118 (78–155) Abbreviations: IQR, interquartile range; uAs, urinary arsenic. a Metric conversions to kg across row: 76.7 (65.8–88.9); 78.3 (68.5–91.2); 77.1 (67.1–88.9); 63.5 (56.7–73.5). b Weight (kg)/height (m)2. c Among former or current smokers only. d uAs reflects sum of inorganic arsenic, methylarsenate, and dimethylarsenate after correcting for seafood arsenicals among subset of 310 participants. Median (interquartile range) lung function test values were as follows: FVC, 3,034 (2,425–3,763) mL; FEV1, 2,242 (1,808–2,766) mL; and FEV1-to-FVC ratio, 0.75% (0.69%–0.80%). Among those with full-lung CT scans, the median (interquartile range) values were as follows: TLC, 4,700 (3,866–5,638) mL; HAA, 4.4% (3.8%–5.3%). A total of 262 participants (13%) had ILA. The median cardiac-based HAA at baseline was 4.2% (interquartile range, 3.6%–5.3%). Among those with serum biomarkers, the median (interquartile range) values were 45.9 (34.0–70.2) ng/L and 3.4 (2.7–4.6) ng/L for SP-A and MMP-7, respectively. Cross-sectional lung function at examination 5 Rice consumption was negatively associated with FVC and FEV1 (Table 2). The adjusted mean difference for FVC comparing eating at least 1 serving of rice daily with eating less than 1 serving weekly was −102 (95% CI: −198, −7) mL; for FEV1 it was −90 (95% CI: −170, −11) mL. The adjusted mean difference for FVC comparing those with consistent long-term rice intake of at least 1 serving of rice daily with those with consistent long-term rice intake of less than 1 serving weekly was −166 (95% CI: −305, −26) mL for FVC and −161 (95% CI: −278, −43) mL for FEV1. The adjusted mean difference for a 2.5-μg/L increase in uAs for FVC was −46 (95% CI: −171, 79) mL; for FEV1 it was −15 (95% CI: −30, 99) mL. There was no evidence of an association between rice consumption or uAs with the ratio of FEV1 to FVC. Excluding participants with airflow obstruction (for whom the FEV1-to-FVC ratio was < 0.7) strengthened the associations between rice intake and uAs with FVC and FEV1 (Table 2). Results were similar when using percent predicted FEV1 and FVC (Web Table 5). There was no evidence of effect modification of rice intake and FVC and FEV1 by race/ethnicity or smoking status (Web Table 6 and Web Figure 1). There was evidence of effect modification of rice intake and FVC by sex, with stronger associations among men. Because of higher water arsenic exposure in Los Angeles, we conducted a stratified analysis in Los Angeles versus the other cities, with consistent findings (Web Table 7). Adjusting the analysis for body mass index, physical activity (metabolic equivalent of task minutes per week), air pollution (particulate matter of diameter <2.5 μm), or income did not change associations (data not shown). Table 2. Mean Difference Between Rice Intake at Examination 1, Consistent Rice Intake at Examinations 1 and 5, and Urinary Arsenic With Spirometry at Examination 5 in the Multi-Ethnic Study of Atherosclerosis, 2000–2002 and 2010–2011 Outcome and Model Rice Intake at Examination 1a Consistent Long-Term Rice Intake at Examinations 1 and 5a Urinary Arsenic 1–6 servings/weekb ≥1 servings/dayb 1–6 servings/weekb ≥1servings /dayb Per 2.5 μg/g Creatinine No. MD 95% CI No. MD 95% CI No. MD 95% CI No. MD 95% CI No. MD 95% CI FVC, mL  Model 1c 1,109 −44 −96, 8 312 −105 −200, −9 651 −40 −107, 26 173 −172 −311, −32 131 −39 −163, 84  Model 2d 1,109 −42 −94, 9 312 −102 −198, −7 651 −33 −99, 34 173 −166 −305, −26 131 −46 −171, 79 FEV1, mL  Model 1c 1,109 −37 −81, 7 312 −93 −174, −13 651 −56 −113, 1 173 −167 −286, −48 131 −10 −124, 105  Model 2d 1,109 −37 −80, 6 312 −90 −170, −11 651 −58 −114, −2 173 −161 −278, −43 131 −15 −30, 99 FEV1-to-FVC ratio, %  Model 1c 1,109 −0.00 −0.01, 0.00 312 −0.01 −0.02, 0.01 651 −0.01 −0.02, 0.00 173 −0.01 −0.03, 0.01 131 0.01 −0.02, 0.03  Model 2d 1,109 −0.00 −0.01, 0.00 312 −0.01 −0.02, 0.01 651 −0.01 −0.02, 0.00 173 −0.01 −0.03, 0.01 131 0.01 −0.02, 0.03 FVC, mL (FEV1-to-FVC ratio > 0.7)e  Model 1c 818 −46 −103, 12 231 −121 −226, −17 485 −59 −133, 15 129 −216 −365, −68 96 −83 −227, 61  Model 2d 818 −45 −104, 11 231 −118 −222, −14 485 −73 −148, 2 129 −221 −370, −73 96 −95 −246, 56 FEV1, mL (FEV1-to-FVC ratio > 0.7)e  Model 1c 818 −31 −76, 14 231 −225 −374, −76 485 −50 −108, 8 129 −168 −284, −51 96 −61 −168, 46  Model 2d 818 −32 −77, 13 231 −81 −162, −0 485 −61 −119, −2 129 −171 −287, −55 96 −74 −187, 38 Outcome and Model Rice Intake at Examination 1a Consistent Long-Term Rice Intake at Examinations 1 and 5a Urinary Arsenic 1–6 servings/weekb ≥1 servings/dayb 1–6 servings/weekb ≥1servings /dayb Per 2.5 μg/g Creatinine No. MD 95% CI No. MD 95% CI No. MD 95% CI No. MD 95% CI No. MD 95% CI FVC, mL  Model 1c 1,109 −44 −96, 8 312 −105 −200, −9 651 −40 −107, 26 173 −172 −311, −32 131 −39 −163, 84  Model 2d 1,109 −42 −94, 9 312 −102 −198, −7 651 −33 −99, 34 173 −166 −305, −26 131 −46 −171, 79 FEV1, mL  Model 1c 1,109 −37 −81, 7 312 −93 −174, −13 651 −56 −113, 1 173 −167 −286, −48 131 −10 −124, 105  Model 2d 1,109 −37 −80, 6 312 −90 −170, −11 651 −58 −114, −2 173 −161 −278, −43 131 −15 −30, 99 FEV1-to-FVC ratio, %  Model 1c 1,109 −0.00 −0.01, 0.00 312 −0.01 −0.02, 0.01 651 −0.01 −0.02, 0.00 173 −0.01 −0.03, 0.01 131 0.01 −0.02, 0.03  Model 2d 1,109 −0.00 −0.01, 0.00 312 −0.01 −0.02, 0.01 651 −0.01 −0.02, 0.00 173 −0.01 −0.03, 0.01 131 0.01 −0.02, 0.03 FVC, mL (FEV1-to-FVC ratio > 0.7)e  Model 1c 818 −46 −103, 12 231 −121 −226, −17 485 −59 −133, 15 129 −216 −365, −68 96 −83 −227, 61  Model 2d 818 −45 −104, 11 231 −118 −222, −14 485 −73 −148, 2 129 −221 −370, −73 96 −95 −246, 56 FEV1, mL (FEV1-to-FVC ratio > 0.7)e  Model 1c 818 −31 −76, 14 231 −225 −374, −76 485 −50 −108, 8 129 −168 −284, −51 96 −61 −168, 46  Model 2d 818 −32 −77, 13 231 −81 −162, −0 485 −61 −119, −2 129 −171 −287, −55 96 −74 −187, 38 Abbreviations: CI, confidence interval; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; MD, mean difference. a Rice models further adjusted for total caloric intake (kcal) and Alternative Healthy Eating Index composite score. b Reference category is <1 serving/week (examination 1, n = 829; participants attending examinations 1 and 5, n = 584). c Adjusted for sex, age (years), race/ethnicity, study site, educational level (<high school, high school, >high school), height (in) and weight (lb). d Further adjusted for smoking status (current, former, never) and pack-years of smoking. e FEV1-to-FVC ratio > 0.7, excluding participants with obstructive-pattern lung disease (<1 serving/week at examination 1, n = 597; participants for whom intake was consistently <1 serving/week at examinations 1 and 5, n = 430). Table 2. Mean Difference Between Rice Intake at Examination 1, Consistent Rice Intake at Examinations 1 and 5, and Urinary Arsenic With Spirometry at Examination 5 in the Multi-Ethnic Study of Atherosclerosis, 2000–2002 and 2010–2011 Outcome and Model Rice Intake at Examination 1a Consistent Long-Term Rice Intake at Examinations 1 and 5a Urinary Arsenic 1–6 servings/weekb ≥1 servings/dayb 1–6 servings/weekb ≥1servings /dayb Per 2.5 μg/g Creatinine No. MD 95% CI No. MD 95% CI No. MD 95% CI No. MD 95% CI No. MD 95% CI FVC, mL  Model 1c 1,109 −44 −96, 8 312 −105 −200, −9 651 −40 −107, 26 173 −172 −311, −32 131 −39 −163, 84  Model 2d 1,109 −42 −94, 9 312 −102 −198, −7 651 −33 −99, 34 173 −166 −305, −26 131 −46 −171, 79 FEV1, mL  Model 1c 1,109 −37 −81, 7 312 −93 −174, −13 651 −56 −113, 1 173 −167 −286, −48 131 −10 −124, 105  Model 2d 1,109 −37 −80, 6 312 −90 −170, −11 651 −58 −114, −2 173 −161 −278, −43 131 −15 −30, 99 FEV1-to-FVC ratio, %  Model 1c 1,109 −0.00 −0.01, 0.00 312 −0.01 −0.02, 0.01 651 −0.01 −0.02, 0.00 173 −0.01 −0.03, 0.01 131 0.01 −0.02, 0.03  Model 2d 1,109 −0.00 −0.01, 0.00 312 −0.01 −0.02, 0.01 651 −0.01 −0.02, 0.00 173 −0.01 −0.03, 0.01 131 0.01 −0.02, 0.03 FVC, mL (FEV1-to-FVC ratio > 0.7)e  Model 1c 818 −46 −103, 12 231 −121 −226, −17 485 −59 −133, 15 129 −216 −365, −68 96 −83 −227, 61  Model 2d 818 −45 −104, 11 231 −118 −222, −14 485 −73 −148, 2 129 −221 −370, −73 96 −95 −246, 56 FEV1, mL (FEV1-to-FVC ratio > 0.7)e  Model 1c 818 −31 −76, 14 231 −225 −374, −76 485 −50 −108, 8 129 −168 −284, −51 96 −61 −168, 46  Model 2d 818 −32 −77, 13 231 −81 −162, −0 485 −61 −119, −2 129 −171 −287, −55 96 −74 −187, 38 Outcome and Model Rice Intake at Examination 1a Consistent Long-Term Rice Intake at Examinations 1 and 5a Urinary Arsenic 1–6 servings/weekb ≥1 servings/dayb 1–6 servings/weekb ≥1servings /dayb Per 2.5 μg/g Creatinine No. MD 95% CI No. MD 95% CI No. MD 95% CI No. MD 95% CI No. MD 95% CI FVC, mL  Model 1c 1,109 −44 −96, 8 312 −105 −200, −9 651 −40 −107, 26 173 −172 −311, −32 131 −39 −163, 84  Model 2d 1,109 −42 −94, 9 312 −102 −198, −7 651 −33 −99, 34 173 −166 −305, −26 131 −46 −171, 79 FEV1, mL  Model 1c 1,109 −37 −81, 7 312 −93 −174, −13 651 −56 −113, 1 173 −167 −286, −48 131 −10 −124, 105  Model 2d 1,109 −37 −80, 6 312 −90 −170, −11 651 −58 −114, −2 173 −161 −278, −43 131 −15 −30, 99 FEV1-to-FVC ratio, %  Model 1c 1,109 −0.00 −0.01, 0.00 312 −0.01 −0.02, 0.01 651 −0.01 −0.02, 0.00 173 −0.01 −0.03, 0.01 131 0.01 −0.02, 0.03  Model 2d 1,109 −0.00 −0.01, 0.00 312 −0.01 −0.02, 0.01 651 −0.01 −0.02, 0.00 173 −0.01 −0.03, 0.01 131 0.01 −0.02, 0.03 FVC, mL (FEV1-to-FVC ratio > 0.7)e  Model 1c 818 −46 −103, 12 231 −121 −226, −17 485 −59 −133, 15 129 −216 −365, −68 96 −83 −227, 61  Model 2d 818 −45 −104, 11 231 −118 −222, −14 485 −73 −148, 2 129 −221 −370, −73 96 −95 −246, 56 FEV1, mL (FEV1-to-FVC ratio > 0.7)e  Model 1c 818 −31 −76, 14 231 −225 −374, −76 485 −50 −108, 8 129 −168 −284, −51 96 −61 −168, 46  Model 2d 818 −32 −77, 13 231 −81 −162, −0 485 −61 −119, −2 129 −171 −287, −55 96 −74 −187, 38 Abbreviations: CI, confidence interval; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; MD, mean difference. a Rice models further adjusted for total caloric intake (kcal) and Alternative Healthy Eating Index composite score. b Reference category is <1 serving/week (examination 1, n = 829; participants attending examinations 1 and 5, n = 584). c Adjusted for sex, age (years), race/ethnicity, study site, educational level (<high school, high school, >high school), height (in) and weight (lb). d Further adjusted for smoking status (current, former, never) and pack-years of smoking. e FEV1-to-FVC ratio > 0.7, excluding participants with obstructive-pattern lung disease (<1 serving/week at examination 1, n = 597; participants for whom intake was consistently <1 serving/week at examinations 1 and 5, n = 430). Cross-sectional lung imaging from full-lung CT scans at examination 5 Baseline rice consumption was not associated with TLC, full-lung HAA, or ILA at examination 5 (Table 3). The adjusted percent difference for TLC comparing those who reported eating at least 1 serving daily with those eating less than 1 serving weekly was −1.33% (95% CI: −4.29, 1.72). The adjusted percent difference for full-lung HAA comparing those who reported eating at least 1 serving daily with those eating less than 1 serving weekly was −0.26% (95% CI: −2.82, 2.36). The odds ratio for ILA comparing those who reported eating at least 1 serving daily with those who never or rarely ate rice was 1.03 (95% CI: 0.54, 1.95). Table 3. Associations Between Rice Intake at Examination 1, Consistent Rice Intake at Examinations 1 and 5, and Urinary Arsenic With Markers From Full-Lung Computed Tomography Scans at Examination 5 in the Multi-Ethnic Study of Atherosclerosis, 2000–2002 and 2010–2011 Outcome Rice Intake at Examination 1a Consistent Long-Term Rice Intake at Examinations 1 and 5a Urinary Arsenic 1–6 servings/weekb ≥1 servings/dayb 1–6 servings/weekc ≥1 servings/dayc Per 2.5 μg/g Creatinine No. % Difference 95% CI No. % Difference 95% CI No. % Difference 95% CI No. % Difference 95% CI No. % Difference 95% CI TLC  Model 1d 1,230 −0.09 −1.67, 1.52 323 −1.38 −4.35, 1.67 710 −0.76 −2.82, 1.34 172 −5.90 −10.15, −1.45 139 0.09 −4.72, 5.15  Model 2e 1,230 0.01 −1.57, 1.61 323 −1.33 −4.29, 1.72 710 −0.71 −2.77, 1.40 172 −5.86 −10.10, −1.41 139 0.23 −4.94, 5.24 HAA  Model 1d 1,230 −0.68 −2.04, 0.69 323 −0.19 −2.78, 2.47 710 −0.31 −2.04, 1.45 172 2.88 −1.02, 6.93 139 −2.00 −5.47, 1.59  Model 2e 1,230 −0.62 −1.95, 0.74 323 −0.26 −2.82, 2.36 710 −0.11 −1.82, 1.64 172 2.98 –0.87, 6.98 139 –2.10 –5.53, 1.45 No. OR 95% CI No. OR 95% CI No. OR 95% CI No. OR 95% CI No. OR 95% CI ILA  Model 1d 127f 1.11 0.81, 1.53 28f 0.99 0.52, 1.87 70g 1.11 0.73, 1.70 13g 1.38 0.48, 3.98 10h 1.64 0.42, 6.43  Model 2e 127f 1.14 0.83, 1.57 28f 1.03 0.54, 1.95 70g 1.15 0.75, 1.77 13g 1.44 0.50, 4.18 10h 1.95 0.44, 8.62 Outcome Rice Intake at Examination 1a Consistent Long-Term Rice Intake at Examinations 1 and 5a Urinary Arsenic 1–6 servings/weekb ≥1 servings/dayb 1–6 servings/weekc ≥1 servings/dayc Per 2.5 μg/g Creatinine No. % Difference 95% CI No. % Difference 95% CI No. % Difference 95% CI No. % Difference 95% CI No. % Difference 95% CI TLC  Model 1d 1,230 −0.09 −1.67, 1.52 323 −1.38 −4.35, 1.67 710 −0.76 −2.82, 1.34 172 −5.90 −10.15, −1.45 139 0.09 −4.72, 5.15  Model 2e 1,230 0.01 −1.57, 1.61 323 −1.33 −4.29, 1.72 710 −0.71 −2.77, 1.40 172 −5.86 −10.10, −1.41 139 0.23 −4.94, 5.24 HAA  Model 1d 1,230 −0.68 −2.04, 0.69 323 −0.19 −2.78, 2.47 710 −0.31 −2.04, 1.45 172 2.88 −1.02, 6.93 139 −2.00 −5.47, 1.59  Model 2e 1,230 −0.62 −1.95, 0.74 323 −0.26 −2.82, 2.36 710 −0.11 −1.82, 1.64 172 2.98 –0.87, 6.98 139 –2.10 –5.53, 1.45 No. OR 95% CI No. OR 95% CI No. OR 95% CI No. OR 95% CI No. OR 95% CI ILA  Model 1d 127f 1.11 0.81, 1.53 28f 0.99 0.52, 1.87 70g 1.11 0.73, 1.70 13g 1.38 0.48, 3.98 10h 1.64 0.42, 6.43  Model 2e 127f 1.14 0.83, 1.57 28f 1.03 0.54, 1.95 70g 1.15 0.75, 1.77 13g 1.44 0.50, 4.18 10h 1.95 0.44, 8.62 Abbreviations: CI, confidence interval; HAA, high attenuation area; ILA, interstitial lung abnormality; OR, odds ratio; TLC, total lung capacity. a Rice models were further adjusted for total caloric intake (kcal) and Alternative Healthy Eating Index composite score. b Reference group is <1 serving/week serving of rice; n = 1,004 for <1 serving/week for TLC and HAA models; n = 107 for <1 serving/week with ILA. c Reference group is <1 serving/week serving of rice; n = 718 for <1 serving/week for TLC and HAA models; n = 79 for <1 serving/week with ILA. d Adjusted for sex, age (years), race/ethnicity, study site, educational level (<high school, high school, >high school), height (in), weight (lb), body mass index (calculated as weight (kg)/height (m)2) > 30 or < 20 (used for radiation dosing level), and scanner manufacturer. HAA models were further adjusted for total imaged lung volume and percent emphysema. e Further adjusted for smoking status (current, former, never) and pack-years of smoking. f Number of participants with ILA is shown (numbers without ILA: 683, 849, and 230 for rice intake at examination 1 of <1 serving/week, 1–6 servings/week, and ≥1 servings/day, respectively). g Number of participants with ILA present shown (number without ILA: 472, 296, 120 for consistent rice intake <1 serving/week, 1–6 servings/week, and ≥1 servings/day, respectively). h Number of participants ILA is shown; n = 101 without ILA. Table 3. Associations Between Rice Intake at Examination 1, Consistent Rice Intake at Examinations 1 and 5, and Urinary Arsenic With Markers From Full-Lung Computed Tomography Scans at Examination 5 in the Multi-Ethnic Study of Atherosclerosis, 2000–2002 and 2010–2011 Outcome Rice Intake at Examination 1a Consistent Long-Term Rice Intake at Examinations 1 and 5a Urinary Arsenic 1–6 servings/weekb ≥1 servings/dayb 1–6 servings/weekc ≥1 servings/dayc Per 2.5 μg/g Creatinine No. % Difference 95% CI No. % Difference 95% CI No. % Difference 95% CI No. % Difference 95% CI No. % Difference 95% CI TLC  Model 1d 1,230 −0.09 −1.67, 1.52 323 −1.38 −4.35, 1.67 710 −0.76 −2.82, 1.34 172 −5.90 −10.15, −1.45 139 0.09 −4.72, 5.15  Model 2e 1,230 0.01 −1.57, 1.61 323 −1.33 −4.29, 1.72 710 −0.71 −2.77, 1.40 172 −5.86 −10.10, −1.41 139 0.23 −4.94, 5.24 HAA  Model 1d 1,230 −0.68 −2.04, 0.69 323 −0.19 −2.78, 2.47 710 −0.31 −2.04, 1.45 172 2.88 −1.02, 6.93 139 −2.00 −5.47, 1.59  Model 2e 1,230 −0.62 −1.95, 0.74 323 −0.26 −2.82, 2.36 710 −0.11 −1.82, 1.64 172 2.98 –0.87, 6.98 139 –2.10 –5.53, 1.45 No. OR 95% CI No. OR 95% CI No. OR 95% CI No. OR 95% CI No. OR 95% CI ILA  Model 1d 127f 1.11 0.81, 1.53 28f 0.99 0.52, 1.87 70g 1.11 0.73, 1.70 13g 1.38 0.48, 3.98 10h 1.64 0.42, 6.43  Model 2e 127f 1.14 0.83, 1.57 28f 1.03 0.54, 1.95 70g 1.15 0.75, 1.77 13g 1.44 0.50, 4.18 10h 1.95 0.44, 8.62 Outcome Rice Intake at Examination 1a Consistent Long-Term Rice Intake at Examinations 1 and 5a Urinary Arsenic 1–6 servings/weekb ≥1 servings/dayb 1–6 servings/weekc ≥1 servings/dayc Per 2.5 μg/g Creatinine No. % Difference 95% CI No. % Difference 95% CI No. % Difference 95% CI No. % Difference 95% CI No. % Difference 95% CI TLC  Model 1d 1,230 −0.09 −1.67, 1.52 323 −1.38 −4.35, 1.67 710 −0.76 −2.82, 1.34 172 −5.90 −10.15, −1.45 139 0.09 −4.72, 5.15  Model 2e 1,230 0.01 −1.57, 1.61 323 −1.33 −4.29, 1.72 710 −0.71 −2.77, 1.40 172 −5.86 −10.10, −1.41 139 0.23 −4.94, 5.24 HAA  Model 1d 1,230 −0.68 −2.04, 0.69 323 −0.19 −2.78, 2.47 710 −0.31 −2.04, 1.45 172 2.88 −1.02, 6.93 139 −2.00 −5.47, 1.59  Model 2e 1,230 −0.62 −1.95, 0.74 323 −0.26 −2.82, 2.36 710 −0.11 −1.82, 1.64 172 2.98 –0.87, 6.98 139 –2.10 –5.53, 1.45 No. OR 95% CI No. OR 95% CI No. OR 95% CI No. OR 95% CI No. OR 95% CI ILA  Model 1d 127f 1.11 0.81, 1.53 28f 0.99 0.52, 1.87 70g 1.11 0.73, 1.70 13g 1.38 0.48, 3.98 10h 1.64 0.42, 6.43  Model 2e 127f 1.14 0.83, 1.57 28f 1.03 0.54, 1.95 70g 1.15 0.75, 1.77 13g 1.44 0.50, 4.18 10h 1.95 0.44, 8.62 Abbreviations: CI, confidence interval; HAA, high attenuation area; ILA, interstitial lung abnormality; OR, odds ratio; TLC, total lung capacity. a Rice models were further adjusted for total caloric intake (kcal) and Alternative Healthy Eating Index composite score. b Reference group is <1 serving/week serving of rice; n = 1,004 for <1 serving/week for TLC and HAA models; n = 107 for <1 serving/week with ILA. c Reference group is <1 serving/week serving of rice; n = 718 for <1 serving/week for TLC and HAA models; n = 79 for <1 serving/week with ILA. d Adjusted for sex, age (years), race/ethnicity, study site, educational level (<high school, high school, >high school), height (in), weight (lb), body mass index (calculated as weight (kg)/height (m)2) > 30 or < 20 (used for radiation dosing level), and scanner manufacturer. HAA models were further adjusted for total imaged lung volume and percent emphysema. e Further adjusted for smoking status (current, former, never) and pack-years of smoking. f Number of participants with ILA is shown (numbers without ILA: 683, 849, and 230 for rice intake at examination 1 of <1 serving/week, 1–6 servings/week, and ≥1 servings/day, respectively). g Number of participants with ILA present shown (number without ILA: 472, 296, 120 for consistent rice intake <1 serving/week, 1–6 servings/week, and ≥1 servings/day, respectively). h Number of participants ILA is shown; n = 101 without ILA. In the subset of participants with consistent self-reported rice intake, the adjusted percent difference for TLC comparing at least 1 serving of rice daily with those eating less than 1 serving weekly was −5.86% (95% CI: −10.10, −1.41). The adjusted percent difference for full-lung HAA for the same comparison was 2.98% (95% CI: −0.87, 6.98). The adjusted odds ratio for ILA comparing those eating at least 1 serving daily with those eating less than 1 serving weekly was 1.44 (0.50, 4.18). Among participants with available uAs measures, the corresponding associations for TLC, full-lung–based HAA, and ILA for an interquartile range increase in uAs (2.5 μg/g creatinine) were not statistically significant (Table 3). Prospective lung imaging from cardiac CT scans at examinations 1–4 Baseline rice consumption was positively associated with baseline differences in cardiac-based HAA but negatively associated with annual change in cardiac-based HAA (Table 4). The fully adjusted percent difference for cardiac-based HAA comparing eating at least 1 serving of rice daily with eating less than 1 serving of rice weekly at baseline was 3.66% (95% CI: 1.22, 6.15), and the annual change in the adjusted percent difference for cardiac-based HAA was −1.80% (95% CI: −2.34, −1.27). The baseline adjusted percent difference for cardiac-based HAA for a 2.5 μg/g creatinine increase in uAs was 3.29% (95% CI: 0.29, 6.28), and the corresponding annual change was −0.80% (95% CI: −1.61, 0.02). There was no effect modification for rice consumption and cardiac-based HAA by age, race/ethnicity, or smoking status (Web Table 8 and Web Figure 2). There was evidence of effect modification by sex, with a stronger association among female patients (Web Table 9). Table 4. Percent Difference of High Attenuation Area (Assessed From Cardiac Computed Tomography Scans at Examinations 1–4) by Self-Reported Frequency of Rice Intake and Per an Interquartile Range Increase (2.5 μg/g Creatinine) in Urinary Arsenic at Examination 1 in the Multi-Ethnic Study of Atherosclerosis, 2000–2002 High Attenuation Area Rice Intake at Examination 1ab uAsb per 2.5 μg/g creatinine 1–6 servings/week vs. <1 serving/week ≥1 servings/day vs. <1 serving/week % Difference 95% CI % Difference 95% CI % Difference 95% CI Baseline difference  Model 1c 1.27 0.01, 2.53 3.55 1.10, 6.06 3.17 0.12, 6.30  Model 2d 1.57 0.33, 2.84 3.66 1.22, 6.15 3.29 0.29, 6.28 Annual change  Model 1c –0.78 –1.13, –0.43 –1.78 –2.31, –1.24 –0.80 –1.62, 0.03  Model 2d –0. 79 –1.14, –0.44 –1.80 –2.34, –1.27 –0.80 –1.61, 0.02 High Attenuation Area Rice Intake at Examination 1ab uAsb per 2.5 μg/g creatinine 1–6 servings/week vs. <1 serving/week ≥1 servings/day vs. <1 serving/week % Difference 95% CI % Difference 95% CI % Difference 95% CI Baseline difference  Model 1c 1.27 0.01, 2.53 3.55 1.10, 6.06 3.17 0.12, 6.30  Model 2d 1.57 0.33, 2.84 3.66 1.22, 6.15 3.29 0.29, 6.28 Annual change  Model 1c –0.78 –1.13, –0.43 –1.78 –2.31, –1.24 –0.80 –1.62, 0.03  Model 2d –0. 79 –1.14, –0.44 –1.80 –2.34, –1.27 –0.80 –1.61, 0.02 Abbreviations: CI, confidence interval; uAS, urinary arsenic. a Rice models were further adjusted for total caloric intake (kcal) and Alternative Healthy Eating Index composite score. b Sample sizes: <1 serving/week, n = 2,312; 1–6 servings/week, n = 694; ≥1 servings/day, n = 694; uAs, n = 310. c Adjusted for sex, age (years), race/ethnicity, study site, education (<high school, high school, >high school), height (cm), weight (lb), total imaged lung volume, percent emphysema, weight >220 lb (99.8 kg; for radiation dosing level) and scanner manufacturer. d Further adjusted for smoking status (current, former, never) and pack-years of smoking. Table 4. Percent Difference of High Attenuation Area (Assessed From Cardiac Computed Tomography Scans at Examinations 1–4) by Self-Reported Frequency of Rice Intake and Per an Interquartile Range Increase (2.5 μg/g Creatinine) in Urinary Arsenic at Examination 1 in the Multi-Ethnic Study of Atherosclerosis, 2000–2002 High Attenuation Area Rice Intake at Examination 1ab uAsb per 2.5 μg/g creatinine 1–6 servings/week vs. <1 serving/week ≥1 servings/day vs. <1 serving/week % Difference 95% CI % Difference 95% CI % Difference 95% CI Baseline difference  Model 1c 1.27 0.01, 2.53 3.55 1.10, 6.06 3.17 0.12, 6.30  Model 2d 1.57 0.33, 2.84 3.66 1.22, 6.15 3.29 0.29, 6.28 Annual change  Model 1c –0.78 –1.13, –0.43 –1.78 –2.31, –1.24 –0.80 –1.62, 0.03  Model 2d –0. 79 –1.14, –0.44 –1.80 –2.34, –1.27 –0.80 –1.61, 0.02 High Attenuation Area Rice Intake at Examination 1ab uAsb per 2.5 μg/g creatinine 1–6 servings/week vs. <1 serving/week ≥1 servings/day vs. <1 serving/week % Difference 95% CI % Difference 95% CI % Difference 95% CI Baseline difference  Model 1c 1.27 0.01, 2.53 3.55 1.10, 6.06 3.17 0.12, 6.30  Model 2d 1.57 0.33, 2.84 3.66 1.22, 6.15 3.29 0.29, 6.28 Annual change  Model 1c –0.78 –1.13, –0.43 –1.78 –2.31, –1.24 –0.80 –1.62, 0.03  Model 2d –0. 79 –1.14, –0.44 –1.80 –2.34, –1.27 –0.80 –1.61, 0.02 Abbreviations: CI, confidence interval; uAS, urinary arsenic. a Rice models were further adjusted for total caloric intake (kcal) and Alternative Healthy Eating Index composite score. b Sample sizes: <1 serving/week, n = 2,312; 1–6 servings/week, n = 694; ≥1 servings/day, n = 694; uAs, n = 310. c Adjusted for sex, age (years), race/ethnicity, study site, education (<high school, high school, >high school), height (cm), weight (lb), total imaged lung volume, percent emphysema, weight >220 lb (99.8 kg; for radiation dosing level) and scanner manufacturer. d Further adjusted for smoking status (current, former, never) and pack-years of smoking. Cross-sectional association with serum biomarkers at examination 1 Rice consumption was positively associated with SP-A level (Table 5). The adjusted percent differences for SP-A comparing persons who reported eating 1–6 servings of rice weekly and those who ate at least 1 serving daily compared with those who ate less than 1 serving weekly were 18.0% (95% CI: 5.4, 32.1) and 12.4% (95% CI: −12.8, 44.9), respectively. The adjusted percent differences for MMP-7 level comparing those who reported eating 1–6 servings of rice weekly and those who ate at least 1 serving daily with those who ate less than 1 serving weekly were −7.3% (95% CI: −12.3, −1.9) and −4.5% (95% CI: −15.7, 8.3), respectively. Table 5. Percent Difference of Serum Biomarkers (Assessed at Examination 1) by Self-Reported Frequency of Rice Intake in the Multi-Ethnic Study of Atherosclerosis, 2000–2002 Outcome Rice Intake at Examination 1 1–6/week vs. <1/weeka ≥1/day vs. <1/weeka % Difference 95% CI % Difference 95% CI MMP-7  Model 1b –7.4 –12.5, –2.1 –4.1 –15.5, 8.8  Model 2c –7.3 –12.3, –1.9 –4.5 –15.7, 8.3 SP-A  Model 1b 18.0 5.4, 32.1 13.8 –11.8, 46.6  Model 2c 18.0 5.4, 32.1 12.4 –12.8, 44.9 Outcome Rice Intake at Examination 1 1–6/week vs. <1/weeka ≥1/day vs. <1/weeka % Difference 95% CI % Difference 95% CI MMP-7  Model 1b –7.4 –12.5, –2.1 –4.1 –15.5, 8.8  Model 2c –7.3 –12.3, –1.9 –4.5 –15.7, 8.3 SP-A  Model 1b 18.0 5.4, 32.1 13.8 –11.8, 46.6  Model 2c 18.0 5.4, 32.1 12.4 –12.8, 44.9 Abbreviations: CI, confidence interval; MMP-7, matrix metalloproteinase-7; SP-A, surfactant protein-A. a Sample sizes: <1 serving/week, n = 2,312; 1–6 servings/week, n = 2,704; ≥1 servings/day, n = 694. b Adjusted for sex, age (years), race/ethnicity, study site, educational level (<high school, high school, >high school), height (cm), weight (lb), total caloric intake (kcal) and Alternative Healthy Eating Index composite score. c Further adjusted for smoking status (current, former, never) and pack-years of smoking. Table 5. Percent Difference of Serum Biomarkers (Assessed at Examination 1) by Self-Reported Frequency of Rice Intake in the Multi-Ethnic Study of Atherosclerosis, 2000–2002 Outcome Rice Intake at Examination 1 1–6/week vs. <1/weeka ≥1/day vs. <1/weeka % Difference 95% CI % Difference 95% CI MMP-7  Model 1b –7.4 –12.5, –2.1 –4.1 –15.5, 8.8  Model 2c –7.3 –12.3, –1.9 –4.5 –15.7, 8.3 SP-A  Model 1b 18.0 5.4, 32.1 13.8 –11.8, 46.6  Model 2c 18.0 5.4, 32.1 12.4 –12.8, 44.9 Outcome Rice Intake at Examination 1 1–6/week vs. <1/weeka ≥1/day vs. <1/weeka % Difference 95% CI % Difference 95% CI MMP-7  Model 1b –7.4 –12.5, –2.1 –4.1 –15.5, 8.8  Model 2c –7.3 –12.3, –1.9 –4.5 –15.7, 8.3 SP-A  Model 1b 18.0 5.4, 32.1 13.8 –11.8, 46.6  Model 2c 18.0 5.4, 32.1 12.4 –12.8, 44.9 Abbreviations: CI, confidence interval; MMP-7, matrix metalloproteinase-7; SP-A, surfactant protein-A. a Sample sizes: <1 serving/week, n = 2,312; 1–6 servings/week, n = 2,704; ≥1 servings/day, n = 694. b Adjusted for sex, age (years), race/ethnicity, study site, educational level (<high school, high school, >high school), height (cm), weight (lb), total caloric intake (kcal) and Alternative Healthy Eating Index composite score. c Further adjusted for smoking status (current, former, never) and pack-years of smoking. DISCUSSION In this multiethnic population of US adults, daily rice consumption was related to lower FVC and FEV1 and higher SP-A levels in cross-sectional analyses. Daily rice consumption was also associated with higher initial differences in HAA, but daily rice consumption was negatively associated with annual changes in HAA in prospective analysis. Rice consumption was not associated with the FEV1-to-FVC ratio or presence of ILA in cross-sectional analyses. When restricting the sample to participants who consistently reported rice intake at examinations 1 and 5, daily rice consumption was related to lower TLC, FVC, and FEV1 and marginally related to higher full-lung HAA and presence of ILA. The subsample with uAs data confirmed that rice intake is associated with higher arsenic exposure; however, although uAs did not reach statistical significance, uAs had a similar direction of association with TLC, ILA, and FVC. Spirometry findings were consistent by sex, race/ethnicity, and smoking status. Although arsenic-contaminated rice poses a potential health risk among people whose diet, due to cultural preferences or dietary restrictions, frequently contains rice (31), little research has been done examining associations between rice consumption and health outcomes. Our findings indicate that, possibly due to arsenic exposure, daily rice consumption may have an adverse relationship with lung function and structure. Similar to meta-analysis results for water arsenic exposure and spirometry, which found an association between water arsenic and lower FVC and FEV1 but not with FEV1-to-FVC ratio (5), we also observed that frequent rice intake was consistently associated with the same restrictive-like spirometry pattern. Although spirometry, TLC, and the cross-sectional cardiac-CT HAA findings largely support the association of arsenic exposure with subclinical interstitial lung disease, findings for annual change in HAA are inconsistent. It is possible that another factor is confounding the association between rice and HAA. Findings regarding race/ethnicity are similar to those for to rice in that Asians have higher cardiac-CT HAA levels at baseline but a negative annual change, as compared to whites (Web Table 9). In our study, although rice consumption patterns differed between racial/ethnic groups, there was no statistical evidence of effect modification in the association between rice intake and the lung outcomes measured. Our interaction tests, however, may be underpowered. Because rice intake and race/ethnicity are closely related and because most HAA studies have been cross-sectional, research is needed to understand the longitudinal changes in HAA, particularly by race/ethnicity (32, 33). Arsenic exposure may affect the respiratory system in part by inducing a low-level inflammatory response and oxidative stress, thus creating a chronic wound-healing environment in the lung (34). In toxicological models, arsenic has also been shown to impair proper wound-healing mechanisms through increased MMP-9 expression and activity (35). Arsenic may also act by reducing the clearance of respiratory pathogens via inducing ubiquitination of cystic fibrosis transmembrane conductance regulator (CFTR), which plays a critical role in mucociliary clearance (36, 37). It is unknown if rice could affect the lungs for reasons beyond arsenic. Potentially, rice consumption could lead to elevated cadmium exposure, a known occupational lung carcinogen (38), but little is known about dietary cadmium and lung diseases (39). Study strengths include our multiethnic study population that includes several racial/ethnic groups with different patterns of rice consumption, a large sample size with a longitudinal population-based assessment, and multiple objective markers of subclinical interstitial lung disease. We also had information on rice consumption 10 years apart, with 63% of participants having consistent rice intake patterns. There are also several limitations. It is unclear why daily rice consumption was associated with higher HAA cross-sectionally but associated with lower HAA longitudinally. This inconsistency may be due to unmeasured confounding or could suggest a potential spurious association between rice consumption and subclinical interstitial lung disease. Information on the sources of drinking water for participants or the arsenic levels in participants’ drinking water is unavailable in MESA (40). We could not confirm that water was not a significant source of exposure; however, because the MESA cohort is drawn from urban-dwelling adults, participants are likely drinking from a community water system for which arsenic levels meet Environmental Protection Agency regulations. The arsenic content of rice also depends on the type of rice and where it is grown. Basmati rice from India, Pakistan, and California, and sushi rice generally have the lowest levels of inorganic arsenic compared with other types of rice, whereas rice from Arkansas, Louisiana, or Texas generally has higher levels (41). Although we had some information on the types of rice people were consuming (white vs. brown), we lacked more detailed information about the rice cultivars (i.e., basmati vs. jasmine) or the country where rice was produced. We also lacked additional information on other rice-based foods (i.e., rice crackers, rice cereal, rice-based sweeteners). It is possible these associations are confounded by other foods commonly eaten with rice, because we were unable to verify arsenic levels in rice; however, we found a positive association between rice consumption and uAs, noted also in other studies (3, 40), indicating that rice consumption contributes to arsenic exposure. In this multiethnic study among US adults, we found some, but not fully consistent, evidence that rice consumption may be related to subclinical interstitial lung disease. Because rice contains arsenic, our results indicate a need to confirm whether associations between rice and health endpoints are related to chronic consumption of arsenic-contaminated rice. Future studies should strive to measure arsenic levels in biological specimens and rice and account for other potential sources of arsenic exposure, including water, to further the scientific understanding of the health effects of dietary arsenic exposure and protect populations with rice-heavy diets. ACKNOWLEDGMENTS Author affiliations: Department of Environmental Health Sciences, Mailman School of Public Health, Columbia University, New York, New York (Tiffany R. Sanchez, Maria Grau-Perez, Matthew S. Perzanowski, Ana Navas-Acien); Department of Medicine, Columbia University, New York, New York (Elizabeth C. Oelsner, David J. Lederer, Christian M. Lo Cascio, R. Graham Barr); Department of Epidemiology, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland (Miranda R. Jones); and Institute of Chemistry, University of Graz, Graz, Austria (Kevin A. Francesconi, Walter Goessler). This research was supported by grants R01HL103676 and K24HL131937 and by contracts HHSN268201500003I, N01-HC-95159, N01-HC-95160, N01-HC-95161, N01-HC-95162, N01-HC-95163, N01-HC-95164, N01-HC-95165, N01-HC-95166, N01-HC-95167, N01-HC-95168, and N01-HC-95169 from the National Heart, Lung, and Blood Institute; by grants UL1-TR-000040, UL1-TR-001079, and UL1-TR-001420 from the National Center for Advancing Translational Sciences; by grants R01ES025216, R01ES021367, P42ES10349, and P30ES009089 from the National Institute of Environmental Health Sciences; and a grant from the Pulmonary Fibrosis Foundation. The authors thank the other investigators, the staff, and the participants of the MESA study for their valuable contributions. A full list of participating MESA investigators and institutions can be found at http://www.mesa-nhlbi.org. Conflict of interest: none declared. Abbreviations CI confidence interval CT computed tomography FEV1 forced expiratory volume in 1 second FVC forced vital capacity HAA high attenuation area ILA interstitial lung abnormality MESA Multi-Ethnic Study of Atherosclerosis MMP-7 matrix metalloproteinase-7 SP-A surfactant protein-A TLC total lung capacity uAs urinary arsenic. REFERENCES 1 Chen Y , Han YH , Cao Y , et al. Arsenic transport in rice and biological solutions to reduce arsenic risk from rice . Front Plant Sci . 2017 ; 8 : 268 . Google Scholar PubMed WorldCat 2 Williams PN , Villada A , Deacon C , et al. Greatly enhanced arsenic shoot assimilation in rice leads to elevated grain levels compared to wheat and barley . Environ Sci Technol . 2007 ; 41 ( 19 ): 6854 – 6859 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Davis MA , Signes-Pastor AJ , Argos M , et al. Assessment of human dietary exposure to arsenic through rice . Sci Total Environ . 2017 ; 586 : 1237 – 1244 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Chung JY , Yu SD , Hong YS . Environmental source of arsenic exposure . J Prev Med Public Health . 2014 ; 47 ( 5 ): 253 – 257 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Sanchez TR , Powers M , Perzanowski M , et al. A meta-analysis of arsenic exposure and lung function: is there evidence of restrictive or obstructive lung disease? Curr Environ Health Rep . 2018 ; 5 ( 2 ): 244 – 254 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Mazumder DN , Steinmaus C , Bhattacharya P , et al. Bronchiectasis in persons with skin lesions resulting from arsenic in drinking water . Epidemiology . 2005 ; 16 ( 6 ): 760 – 765 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Smith AH , Marshall G , Yuan Y , et al. Increased mortality from lung cancer and bronchiectasis in young adults after exposure to arsenic in utero and in early childhood . Environ Health Perspect . 2006 ; 114 ( 8 ): 1293 – 1296 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Zaldívar R . A morbid condition involving cardio-vascular, broncho-pulmonary, digestive and neural lesions in children and young adults after dietary arsenic exposure . Zentralbl Bakteriol B . 1980 ; 170 ( 1–2 ): 44 – 56 . Google Scholar PubMed WorldCat 9 Dwyer-Lindgren L , Bertozzi-Villa A , Stubbs RW , et al. Trends and patterns of differences in chronic respiratory disease mortality among US counties, 1980–2014 . JAMA . 2017 ; 318 ( 12 ): 1136 – 1149 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Centers for Disease Control and Prevention . Health, United States, 2016: With Chartbook on Long-Term Trends in Health . Hyattsville, MD : National Center for Health Statistics ; 2017 . Google Preview WorldCat COPAC 11 Rosas IO , Dellaripa PF , Lederer DJ , et al. Interstitial lung disease: NHLBI Workshop on the Primary Prevention of Chronic Lung Diseases . Ann Am Thorac Soc . 2014 ; 11 ( suppl 3 ): S169 – S177 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Doyle TJ , Hunninghake GM , Rosas IO . Subclinical interstitial lung disease: why you should care . Am J Respir Crit Care Med . 2012 ; 185 ( 11 ): 1147 – 1153 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Podolanczuk AJ , Oelsner EC , Barr RG , et al. High-attenuation areas on chest computed tomography and clinical respiratory outcomes in community-dwelling adults . Am J Respir Crit Care Med . 2017 ; 196 ( 11 ): 1434 – 1442 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Greene KE , King TE Jr , Kuroki Y , et al. Serum surfactant proteins-A and -D as biomarkers in idiopathic pulmonary fibrosis . Eur Respir J . 2002 ; 19 ( 3 ): 439 – 446 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Rosas IO , Richards TJ , Konishi K , et al. MMP1 and MMP7 as potential peripheral blood biomarkers in idiopathic pulmonary fibrosis . PLoS Med . 2008 ; 5 ( 4 ): e93 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Bild DE , Bluemke DA , Burke GL , et al. Multi-Ethnic Study of Atherosclerosis: objectives and design . Am J Epidemiol . 2002 ; 156 ( 9 ): 871 – 881 . Google Scholar Crossref Search ADS PubMed WorldCat 17 Nettleton JA , Steffen LM , Mayer-Davis EJ , et al. Dietary patterns are associated with biochemical markers of inflammation and endothelial activation in the Multi-Ethnic Study of Atherosclerosis (MESA) . Am J Clin Nutr . 2006 ; 83 ( 6 ): 1369 – 1379 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Willett WC . Issues in analysis and presentation of dietary data. In: Nutritional Epidemiology . New York, NY : Oxford University Press ; 2013 : 1 – 53 . Google Preview WorldCat COPAC 19 McCullough ML , Feskanich D , Stampfer MJ , et al. Diet quality and major chronic disease risk in men and women: moving toward improved dietary guidance . Am J Clin Nutr . 2002 ; 76 ( 6 ): 1261 – 1271 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Moore LV , Diez Roux AV , Nettleton JA , et al. Associations of the local food environment with diet quality—a comparison of assessments based on surveys and geographic information systems: the Multi-Ethnic Study of Atherosclerosis . Am J Epidemiol . 2008 ; 167 ( 8 ): 917 – 924 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Scheer J , Findenig S , Goessler W , et al. Arsenic species and selected metals in human urine: validation of HPLC/ICPMS and ICPMS procedures for a long-term population-based epidemiological study . Anal Methods . 2012 ; 4 ( 2 ): 406 – 413 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Jones MR , Tellez-Plaza M , Vaidya D , et al. Estimation of inorganic arsenic exposure in populations with frequent seafood intake: evidence from MESA and NHANES . Am J Epidemiol . 2016 ; 184 ( 8 ): 590 – 602 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Barr RG , Ahmed FS , Carr JJ , et al. Subclinical atherosclerosis, airflow obstruction and emphysema: the MESA Lung Study . Eur Respir J . 2012 ; 39 ( 4 ): 846 – 854 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Miller MR , Hankinson J , Brusasco V , et al. Standardisation of spirometry . Eur Respir J . 2005 ; 26 ( 2 ): 319 – 338 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Hankinson JL , Kawut SM , Shahar E , et al. Performance of American Thoracic Society-recommended spirometry reference values in a multiethnic sample of adults: the Multi-Ethnic Study of Atherosclerosis (MESA) Lung Study . Chest . 2010 ; 137 ( 1 ): 138 – 145 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Hoffman EA , Jiang R , Baumhauer H , et al. Reproducibility and validity of lung density measures from cardiac CT Scans—The Multi-Ethnic Study of Atherosclerosis (MESA) Lung Study . Acad Radiol . 2009 ; 16 ( 6 ): 689 – 699 . Google Scholar Crossref Search ADS PubMed WorldCat 27 Hoffman EA , Ahmed FS , Baumhauer H , et al. Variation in the percent of emphysema-like lung in a healthy, nonsmoking multiethnic sample. The MESA Lung Study . Ann Am Thorac Soc . 2014 ; 11 ( 6 ): 898 – 907 . Google Scholar Crossref Search ADS PubMed WorldCat 28 Lederer DJ , Enright PL , Kawut SM , et al. Cigarette smoking is associated with subclinical parenchymal lung disease: the Multi-Ethnic Study of Atherosclerosis (MESA)-Lung Study . Am J Respir Crit Care Med . 2009 ; 180 ( 5 ): 407 – 414 . Google Scholar Crossref Search ADS PubMed WorldCat 29 Podolanczuk AJ , Oelsner EC , Barr RG , et al. High attenuation areas on chest computed tomography in community-dwelling adults: the MESA study . Eur Respir J . 2016 ; 48 ( 5 ): 1442 – 1452 . Google Scholar Crossref Search ADS PubMed WorldCat 30 Sieren JP , Newell JD , Barr RG , et al. SPIROMICS protocol for multicenter quantitative computed tomography to phenotype the lungs . Am J Respir Crit Care Med . 2016 ; 194 ( 7 ): 794 – 806 . Google Scholar Crossref Search ADS PubMed WorldCat 31 Bulka CM , Davis MA , Karagas MR , et al. The unintended consequences of a gluten-free diet . Epidemiology . 2017 ; 28 ( 3 ): e24 – e25 . Google Scholar Crossref Search ADS PubMed WorldCat 32 Kim JS , Podolanczuk AJ , Borker P , et al. Obstructive sleep apnea and subclinical interstitial lung disease in the Multi-Ethnic Study of Atherosclerosis (MESA) . Ann Am Thorac Soc . 2017 ; 14 ( 12 ): 1786 – 1795 . Google Scholar Crossref Search ADS PubMed WorldCat 33 Podolanczuk AJ , Raghu G , Tsai MY , et al. Cholesterol, lipoproteins and subclinical interstitial lung disease: the MESA study . Ann Am Thorac Soc . 2017 ; 72 ( 5 ): 472 – 474 . WorldCat 34 Lantz RC , Hays AM . Role of oxidative stress in arsenic-induced toxicity . Drug Metab Rev . 2006 ; 38 ( 4 ): 791 – 804 . Google Scholar Crossref Search ADS PubMed WorldCat 35 Olsen CE , Liguori AE , Zong Y , et al. Arsenic upregulates MMP-9 and inhibits wound repair in human airway epithelial cells . Am J Physiol Lung Cell Mol Physiol . 2008 ; 295 ( 2 ): L293 – L302 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Bomberger JM , Coutermarsh BA , Barnaby RL , et al. Arsenic promotes ubiquitinylation and lysosomal degradation of cystic fibrosis transmembrane conductance regulator (CFTR) chloride channels in human airway epithelial cells . J Biol Chem . 2012 ; 287 ( 21 ): 17130 – 17139 . Google Scholar Crossref Search ADS PubMed WorldCat 37 Mazumdar M , Christiani DC , Biswas SK , et al. Elevated sweat chloride levels due to arsenic toxicity . N Engl J Med . 2015 ; 372 ( 6 ): 582 – 584 . Google Scholar Crossref Search ADS PubMed WorldCat 38 IARC Working Group on the Evaluation of Carcinogenic Risk to Humans . Arsenic, metals, fibres and dusts . IARC Monogr Eval Carcinog Risks Hum . 2012 ; 100 ( part C ): 11 – 465 . PubMed WorldCat 39 Hu X , Fernandes J , Jones DP , et al. Cadmium stimulates myofibroblast differentiation and mouse lung fibrosis . Toxicology . 2017 ; 383 : 50 – 56 . Google Scholar Crossref Search ADS PubMed WorldCat 40 Jones MR , Tellez-Plaza M , Vaidya D , et al. Ethnic, geographic and dietary differences in arsenic exposure in the Multi-Ethnic Study of Atherosclerosis (MESA) . J Expo Sci Environ Epidemiol . 2019 ; 29 ( 3 ): 310 – 322 . Google Scholar Crossref Search ADS PubMed WorldCat 41 How much arsenic is in your rice? Consumer Reports Magazine. 2014 . https://www.consumerreports.org/cro/magazine/2015/01/how-much-arsenic-is-in-your-rice/index.htm. Accessed June 28, 2018. © The Author(s) 2019. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Rice Consumption and Subclinical Lung Disease in US Adults: Observational Evidence From the Multi-Ethnic Study of Atherosclerosis
JF - American Journal of Epidemiology
DO - 10.1093/aje/kwz137
DA - 2019-09-01
UR - https://www.deepdyve.com/lp/oxford-university-press/rice-consumption-and-subclinical-lung-disease-in-us-adults-eXqkdO0is9
SP - 1655
VL - 188
IS - 9
DP - DeepDyve
ER -