Study Design and Population
The present study used data from five cycles (2007–2016) of National Health and Nutrition Examination Survey (NHANES) (http://www.cdc.gov/nchs/nhanes.htm). NHANES collects demographic, behavioral, clinical, social, dietary, and laboratory data from non-institutional people in the United States, which is a cross-sectional survey of a representative population of the nation with purpose of making assessment of the health and nutrition condition of adults and minors across the country. The survey is programmed of National Center for Health Statistics (NCHS). Detailed information on participants' demographics, dietary factors, socioeconomic status, and medical status was collected in-home by professional interviewers. Standardized medical physical examinations as well as the biospecimen collection were performed by experienced medical personnel in mobile examination centers (MECs). Patients without laboratory data of urinary PAHs metabolites and without UI questionnaire information were excluded. The study was approved by Ethics Review Board of NCHS, and all participants had been required to sign an informed consent form before the study began.
Polycyclic Aromatic Hydrocarbons
Through the survey, urine samples were collected by professionally trained technicians and then frozen in -20 ° C environment according to Laboratory/Medical Procedures pending further analysis. PAHs in urine samples were tested using capillary gas chromatography combined with high resolution mass spectrometry (GC-HRMS), isotope dilution gas chromatography/tandem mass spectrometry (GC-MS/MS), and isotope dilution online solid phase extraction high performance liquid chromatography/tandem mass spectrometry (online SPE-HPLC-MS/MS). The specific test method varied with the year. (http://www.cdc.gov/nchs/nhanes.htm). Ten hydroxylated urinary PAH, namely, 1-hydroxynapthalene, 2-hydroxynapthalene, 2chydroxyfluorene, 3-hydroxyfluorene, 9-hydroxyfluorene, 1-hydroxypyrene, 1-hydroxyphenanthrene, 2-hydroxyphenanthrene, 3-hydroxyphenanthrene, and 4-hydroxyphenanthrene were observed.
Outcome Information
UI was determined and classified by following questionnaires among all participants over the age of 20. Participants with positive answer to “Have you leaked or lost control of even a small amount of urine with activity like coughing, lifting, or exercise during the past 12 months?” were categorized as SUI. UUI was diagnosed as answering yes to “Have you leaked or lost control of even a small amount of urine with an urge or pressure to urinate and you couldn’t get to the toilet fast enough during the past 12 months?” People with answer YES to both UUI and SUI questions were diagnosed with MUI. The above questions were asked at MECs by trained interviews, using the Computer-Assisted Personal Interview system.
Covariates Measurement
The covariates adjusting association between urinary PAHs and UI include age, race (Mexican Hispanic, other Hispanic, non-Hispanic white, non-Hispanic black, and others), body mass index (BMI), ratio of family income-poverty, marital status (living alone and married or living with partner and married or living with partner), educational level (under high school, high school or general educational development, and above high school), physical activity level (under moderate, moderate, and vigorous), Charlson comorbidity index, smoking status (never, former, and current), and alcohol drinking level (none, moderate, heavy). Smoking level was classified as never smoking (< 100 cigarettes during the life), former smoking (≥ 100 cigarettes during the life, currently not smoking) and current smoking (≥ 100 cigarettes during the life, currently smoking). The alcohol consumption categories were defined as none (0 g per day), moderate (0.1–27.9 g per day for men and 0.1–13.9 g per day for women), and heavy (no less than 28.0 g per day for men and no less than 14.0 g per day for women). All information is available on NHANES website (http://www.cdc.gov/nchs/nhanes.htm).
Statistical Analysis
The characteristics of participants were presented in the form of numbers (percentage) for categorical variables and mean ± standard deviation (SD) for continuous variables. Differences between base line characteristics of people with UI and without UI were calculated using either Kruscal Whallis H test on continuous variables or a weighted chi-square test on categorical variables.
Since the concentrations of PAHs were diluted by urine, we adjusted it and used the creatinine corrected PAHs concentrations instead of the original concentrations for analysis. Because the urine PAHs showed a highly skew distribution, log transformation was performed then. Due to the same reason, PAHs levels were presented as geometric mean (GM), median, upper and lower quartiles. When the urine PAHs concentrations is lower than the lower limit of detection (LLOD), it is replaced by the square root of 2. The continuous variables of urine PAHs concentrations were separated into four ranges by the median, upper and lower quartiles, and were transformed into categorical variables. Multivariable logistic regression models were used and the relationships between urinary PAHs and three UI subtypes were estimated with odds ratios (ORs) and corresponding 95% confidence intervals (95%CI). Three models were established: a non-adjusted model (Model I), a partially adjusted model (Model II) with age, race, and BMI adjusted, and a fully adjusted model (Model III) with age, race, BMI, ratio of family income-poverty, educational level, marital status, physical activity, Chalson comorbidity index, smoking status, and alcohol drinking adjusted.
Missing values were entered by the median of existing continuous variables or by mode of existing categorical variables. All statistical analyses were performed with R (http:// www.R-project.org, The R Foundation) and Empower (www.empowerstats.com). Considering the complex sampling design of NHANES, weighted analyses were conducted. The p-values of multiple testing were all adjusted and statistically significant were defined as two-tailed p < 0.05.