Patients
The NHANES employed a sampling design that involved stratification, multiple stages, and probability clusters to evaluate the health and nutritional status of the non-institutionalized US civilian population in a nationally representative manner. The study protocols of NHANES have been approved by the Institutional Review Board of the National Center for Health Statistics, and written informed consent and/or assent is obtained from participants (9).
We collected data from 2007 to 2012 because lung function data was only available for these years. We excluded adolescents(< 18 years old), cancer patients, as well as individuals with missing basic information data (age, gender, race, marital status, family income, education information), dietary data, lung function data, body mass index (BMI) data, physical activity data, smoking information, and individuals with missing comorbidity data. Finally, 6476 eligible participants were included in the analysis. The study was approved by clinical research ethics board of Wenzhou Medical University (NO. 2016–197).
Lung function parameters assessment
Individuals, ages 6 to 79 years of age, were eligible to participate in the spirometry component of the NHANES study. The following exclusion criteria were applied to the spirometry component: had a recent chest pain, stroke, or heart attack, recent eye, chest, or abdominal surgery, exposure to tuberculosis, had recently coughed up blood, or a physical problem with forceful expiration. Individuals with a previous detached retina or collapsed lung, as well as children experiencing painful ear infections, were not included in the study. Spirometry tests were conducted according to the standardized protocol recommended by the American Thoracic Society (ATS) (10), with participants in a standing or, if necessary, seated position. Participants ages 6–10 years were asked to blow out the air for a minimum of 3 seconds of exhalation, whereas those ages 11–79 years were asked to blow out the air for a minimum of 6 seconds of exhalation. Spirometry testing was conducted on an individual until they achieved a reproducible spirogram, or until a maximum of eight spirometry curves were obtained, or until the participant was unable to continue. The three best spirometry readings were assessed using ATS criteria(10) and recorded for each participant. Only participants with FEV1 and FVC values rated as A (Exceeds ATS data collection standards: 3 acceptable curves present and 2 reproducible curves; 2 observed values within 100 ml) or B (Meets ATS data collection standards: 3 acceptable curves present and 2 reproducible curves; 2 observed values within 150 ml) were included in the analysis. The spirometry dataset consisted of raw values for the highest FVC, FEV1, and mid-exhalation forced expiratory flow rate (FEF 25–75%) that met or exceeded ATS standards.
Intake of nutrient
The data on daily nutrient intake (including calcium, vitamin C, vitamin D, vitamin B6) from 2007 to 2012 was collected through in-person or telephone questionnaires. To estimate dietary calcium intake, the USDA's Food and Nutrient Database for Dietary Studies was utilized. The first dietary recall interview is collected in-person in the Mobile Examination Center (MEC) and the second interview is collected by telephone 3 to 10 days later (11). Data on the use of dietary supplements and non-prescription antacid were collected by using 2-day dietary supplement questionnaire.
Dietary nutrient intake was calculated as the average nutrient intake from the two 24-h recalls. Dietary nutrient supplement was also calculated as the average dietary nutrient supplement from the two 24-h recalls. Total nutrient intake was calculated as the sum of daily nutrient intake from diet and supplements.
Covariates
Demographic variables included age (years), sex (male/ female), race (White/ Black/ other race), marital status (married/ widowed/ divorced or separated/never married), BMI, serum calcium level, education level (less than high school/high school/more than high school), physical activity, and family poverty-income ratio (PIR) (0–1.0, ≥ 1 to 2.0, ≥ 2.0 to 4.0, and ≥ 4.0). BMI was calculated as weight (kg)/ height (m2). Hypertension, diabetes mellitus and cerebrovascular disease (CVD) were defined as self-reported physician diagnoses. Smoking status was defined as smoking at least 100 cigarettes in life. Participants were classified as having hypertension if participants taken any antihypertensive drugs, or had systolic blood pressure ≥ 140 mmHg or diastolic blood pressure ≥ 90 mmHg. Diabetes mellitus (DM) was defined if participants had self-reported diagnosis of diabetes, use of diabetes medication or insulin, glycosylated hemoglobin ≥ 6.5%, fasting glucose (mmol/l) ≥ 7.0, random blood glucose (mmol/l) ≥ 11.1, or two-hour oral glucose tolerance test (OGTT) blood glucose (mmol/l) ≥ 11.1. COPD was diagnosed by: After using bronchodilators, FEV1/FVC < 0.7; Participants had self-reported emphysema; Use COPD drugs and be over 40 years old, have a history of smoking or chronic bronchitis. CVD was ascertained if participants had history of CVD, angina or myocardial infarction. Stroke and cancer were determined if participants reported history of stroke or cancer (malignancy), respectively. Asthma was diagnosed by self-reported asthma, use antiasthmatic drug, or use drug (selective phosphodiesterase-4 inhibitors, mast cell stabilizers, leukotriene modifiers, inhaled corticosteroids), age < 40 and have no smoke, chronic bronchitis and emphysema.
Physical activity was expressed as the metabolic equivalent task (MET) and calculated as follows: physical activity (met·min/week) = recommended MET × exercise time for corresponding activities (min/day) × the number of exercise days per week (day) (12). Smoke was categorized into current (participants who smoked at least 100 cigarettes in life and now smoke), former (participants who smoked at least 100 cigarettes in life and quit smoke now), and never.
Statistical methods
All statistical analyses were conducted using the statistical package nhanesR with a complex weighted sampling design of NHANES. To compare differences between groups, we used weighted chi-square tests and linear regression models to analyze categorical and continuous variables, respectively. Participants were characterized according to quartiles of calcium intake (category 1: < 689.875 mg; category 2: 689.875-990 mg; category 3: 990-1406.125 mg; category 4: 1406.125–5399 mg). Multiple linear regression models were used to assess the relationship between calcium intake and lung function (FEV1, FVC, and FEF). Model 1 was unadjusted. Model 2 included age and gender. Model 3 was then calculated based on variables such as age, gender, race, marital status, hypertension, serum calcium level, smoking status, DM, CVD, physical activity, COPD, asthma, education status, BMI, Vitamin B6, Vitamin C, Vitamin D ,and PIR. Analyses were then stratified by age, sex, race, BMI, smoking, and disease status and tested for interactions. In the analysis, P values less than 0.05 were considered statistically significant. The restricted cubic spline (RCS) analysis based on model 3 was performed to test the non-linear relationship between calcium intake and lung function (FEV1 and FVC).