**1. Introduction**

Dietary nutrients play an important role in maintaining the balance of blood glucose as a necessary substance to regulate the normal physiological function of the body, being roughly divided into macronutrients [1–3], dietary fiber (as an independent factor that distinguishes carbohydrates, included in this study) [4–6], minerals [7–9], and vitamins [10–13]. Many studies have shown that there was a positive correlation between recent glycemic control status (RGCS) and serum chromium, zinc, and magnesium levels [7,14–20]. Glycosylated hemoglobin (HbA1c), an irreversible product of blood glucose and hemoglobin, could provide information for long-term glycemic control [21]. Moreover, after the relationship between RGCS and HbA1c concentration was widely confirmed, the serum index was applied to diabetes diagnosis and glycemic monitoring practice [21,22]. Therefore, it was appropriate to use HbA1c as a predictor for RGCS among American adults (specifically referred to non-pregnant participants over 20 years old).

**Citation:** Bai, Y.; Zhang, H.; Yang, J.; Peng, L. The Association between Daily Total Dietary Nutrient Intake and Recent Glycemic Control States of Non-Pregnant Adults 20+ Years Old from NHANES 1999–2018 (Except for 2003–2004). *Nutrients* **2021**, *13*, 4168. https://doi.org/10.3390/ nu13114168

Academic Editor: Lourdes Varela

Received: 18 September 2021 Accepted: 19 November 2021 Published: 21 November 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

At present, most of the research conclusions on the association between daily total nutrient intake and RGCS have been one-sided. They did not analyze the overall effect of various nutrients on RGCS but analyzed minerals, vitamins, and macronutrients separately, which was not complete and systematic, and might have even led to obtaining inconsistent conclusions [23,24]. Although findings on the association between daily total dietary nutrient intake and RGCS were inconsistent and not enough to prove the relationship, these results, to a certain extent, could supply research hypotheses for future large-scale prospective or multi-center verification. Therefore, if we further explored the association between the adults' RGCS and daily total dietary nutrient intakes, it was necessary to construct a holistic and optimal model to combine the macronutrients, minerals, and vitamins of daily total nutrient intake, as well as demographic characteristics, in order to draw a reliable conclusion.

In addition, most studies on dietary factors affecting glycemic control have been conducted on diabetes patients [5,11,25–27], and therefore these conclusions could not be suitable for American adults to control blood glucose. Moreover, it is worth noting that insufficient sample size might also lead to biased conclusions, for instance, in the study of Intra et al. [26], the sample size of cases group was very small (cases group, *n* = 84; controls group, *n* = 2745), and therefore the results of this study might have a larger systematic bias. Therefore, we conducted the follow-up sampling survey study to estimate the association between daily total nutrient intake and RGCS among non-pregnant adults 20+ years old using a large-scale database from National Health and Nutrition Examination Survey (NHANES 1999–2018, except for 2003–2004).

#### **2. Methods**

#### *2.1. Database and Study Population*

We used the NHANES database, a nationally representative survey database collected biennially by the National Center for Health Statistics (NCHS), and employed a complex, multistage, probabilistic sampling design [28]. The database was publicly available on the Internet and can be downloaded by researchers around the world. All details about the database could be efficiently acquired at http://www.cdc.gov/nchs/nhanes/ (accessed on 27 May 2021), including relevant information such as strict quality control measures for the questionnaire data undertaken by NHANES. The 24 h dietary recall data from non-pregnant adults 20+ years of age participating in NHANES 1999–2002 and 2005–2018 surveys were followed biennially for all analyses. The database for analysis consisted of five parts: demographics data, dietary data, examination data, laboratory data, and questionnaire data.

During the 1999–2018 NHANES survey cohorts, 101,316 preliminary participants were included in the study. Individuals without physical examination data (*n* = 2096), under 20 years of age (*n* = 47,208), pregnant (*n* = 2527), without an unusual diet compared food consumed yesterday and without reliable data (*n* = 6431), and 2003–2004 survey cycle data with the missing outcome variable (*n* = 1752) were excluded. Those with complete or reliable 24 h recall data (only day 1 data used) as judged by the United States Department of Agriculture's (USDA) Food Surveys Research Group staff were included in the analyses. In addition, of the 41302 participants, 58 of the dietary survey data contained some missing indices (such as the intake of dietary fiber and folic acid), and we used the median to fill them. Finally, 41,302 subjects (20,458 males and 20,844 females, 50.0 ± 17.9 years for males and 50.2 ± 17.8 years for females) were certainly included in this study (Figure S1). All serum samples were collected under fasting condition. HCHS obtained the written informed consent from all participants and the ethical review committee approved all NHANES protocols.
