Modifiable lifestyle, mental health status and diabetic retinopathy in U.S. adults aged 18–64 years with diabetes: a population-based cross-sectional study from NHANES 1999–2018

National Health and Nutrition Examination Survey (NHANES), an ongoing, biennial cross-sectional survey, utilized a complex multistage probability sampling design to investigate the nutritional and health status of the civilian, non-institutionalized U.S. population [12, 13]. Data collected involved physical examinations, laboratory tests, and questionnaires. Informed written consent was obtained from all participants, and the study received ethical approval from the National Center for Health Statistics. The research adhered to the Tenets of the Declaration of Helsinki.

Our study encompassed 3482 participants aged 18–64 years from NHANES 1999–2018, as illustrated in Fig. 1. All participants were non-pregnant and had received a diagnosis of diabetes from a physician or other health professional at 18 years of age or older (excluding diagnosis made solely during pregnancy)

Diabetes in this study was defined based on self-reported diagnoses confirmed by professional physicians, rather than relying on glycosylated hemoglobin A1c (HbA1c) and self-reported results. This approach was selected due to the significant impact of diabetic duration as a risk factor for DR, with newly-diagnosed diabetes not adequately reflecting its substantial influence. Diabetic retinopathy was identified through self-reported diagnosed retinopathy, as confirmed by a dichotomous item as presented consistent with previous literature [14‐17]. In the Diabetes Interview Questionnaires, participants diagnosed with diabetes were asked “Has a doctor ever told you/SP that diabetes has affected your/his/her eyes or that you/s/he had retinopathy (ret-in-op-ath-ee)?” Affirmative responses were considered indicative of self-reported DR. The duration of diagnosed diabetes was calculated as the age at examination minus the age at diagnosis, then subcategorized into three groups: 0–10 years, 11–19 years and ≥ 20 years.

Lifestyle-related variables in NHANES encompassed smoking status, drinking status, physical activity, healthy eating index (HEI), sleeping duration, and self-reported sleeping disorder. Smoking status (never smoker, former smoker, current smoker) was determined by using Smoking-Cigarette Use questionnaires, with never smoking defined as lifetime consumption of fewer than 100 cigarettes. Drinking status (no drinking, mild to moderate drinking, heavy drinking) was based on self-reported drinking frequency and quantity, following officially provided units (1 drink contains 14 g of ethanol) [18]. Mild to moderate drinking was defined as < 1 drink/day for women and < 2 drinks/day for men, while heavy drinking was ≥ 1 drink/day for women and ≥ 2 drinks/day for men. Physical activity level (< 150, 150–1800, > 1800 MET*min/week) was classified by the total MET-min per week, calculated by multiplying the standard metabolic equivalent (MET) value of each activity by its weekly minutes and then summing across activities. Dietary quality was assessed through 24-hour dietary recalls and HEI scores aligned with the 2015–2020 Dietary Guidelines for Americans were used for the 1999–2014 survey cycles. HEI scores were categorized as < 25%, 25-75%, > 75%, with higher HEI indicating healthier eating habits. Sleeping status included sleeping time (< 5, 5–9, > 9 h) and sleeping disorders (with/without sleeping disorder), obtained from the Sleep-Disorders questionnaire since NHANES 2005–2006. Additionally, depression status (with/without depression) was estimated as an indicator of the mental condition, defined using the nine-item depression screening instrument called the Patient Health Questionnaire (PHQ-9) [19], with a score ≥ 10 considered indicative of depression.

Demographic variables included age, gender (male/female), race/ethnicity (Mexican American, Other Hispanic, non-Hispanic White, non-Hispanic Black, other), family economic status (poverty-to-income ratio, ≤ 130%, 130–350%, > 350%), educational level (less than high school, high school or equivalent, college and more), health insurance status (uninsured, any health insurance), and marital status (married, unmarried, other status).

Body mass index (BMI) was calculated as the measured weight in kilograms divided by height in meters squared and categorized as normal (18.5–24.9), overweight (25.0–29.9), obese class I (30.0–34.9), obese class II (35.0–39.9), or obese class III (≥ 40.0). HbA1c level below 7% were considered well-controlled. Uncontrolled hypertension was defined as elevated blood pressure (≥ 140/90 mmHg) measured by a board-eligible physician after the participant had been in a seated position for 5 min of quiet rest. Dyslipidemia was determined by non-high-density lipoprotein (non-HDL) cholesterol, calculated as total measured cholesterol minus HDL cholesterol, with a non-HDL level < 130 mg/dL considered indicative of well-controlled plasma lipids [20].

Diabetic nephropathy (DN) was defined by either a urine albumin to creatinine ratio of 30 mg/g or higher, or estimated glomerular filtration rate less than 60 mL/min/1.73 m². Cardiovascular diseases (CVD) were defined by self-reported any of the following diseases: self-reported congestive heart failure, coronary heart disease, heart attack, or stroke [21].

Data from the NHANES 1999–2000 to NHANES 2017–2018 cycles were analyzed. The distribution of demographic, clinical, and lifestyle characteristics among participants was examined. Lifestyle-related and mental factors were compared by using the student’s t test in a weighted linear regression model (for continuous variables) or the weighted Rao-Scott χ² test (for categorical variables) between the two groups (DR and no DR). Given the well-established association between the duration of diabetes and the risk of DR, subgroup analyses were conducted, stratified by diabetic duration to assess differences and variations in lifestyle. A weighted univariable and multivariable logistic regression model was employed to explore the impact of lifestyle on DR, adjusting for potential confounders including gender, age, race/ethnicity, diabetic duration, DN, CVD, HbA1c, blood pressure, lipids control. Appropriate sample weights for the interview sample, examination sample and fasting subsample were utilized. If the level of missing data for primary analyses was less than 10%, a complete case analysis was applied. All statistical analyses were performed using Statistical Analysis Software procedures (SAS version 9.4). A two-sided p-value < 0.05 was considered statistically significant.