**1. Introduction**

The world's population is aging at a faster speed than it used to be according to a report from the World Health Organization [1]. It is predicted that the number of older people worldwide would more than double by 2050 [2]. The aging process is accompanied by declines in functional capacity and cognition [3,4]. Disability in older people is associated with low quality of life [5], increased burden on caregivers [6], rise in morbidity and mortality [7], and elevated health care costs [8], which finally put more burdens on the whole society.

Activities of daily living (ADL) is an index to represent the capability of skills essential to manage basic physical needs, comprised of the following areas: grooming/personal hygiene, dressing, toileting/continence, transferring/ambulating, and eating [9]. The prevalence of ADL disability in Chinese individuals aged 60 years and above was 10.03% in 2015 [10]. Actions aimed at preventing ADL dependence may reduce the burden on older people themselves, caregivers, and the health care system [11]. Several factors are reported to be associated with ADL disability, including gender, slow gait speed, weight loss, muscle strength, etc. [5,12]. In addition to that, diet and nutrition are found to have an impact on ADL or functional capacity. Previous prospective studies indicated that higher consumption of dairy products is associated with a lower risk of functional disability [13]. Nutrients, such as vitamin D [14] and dietary fiber [15], are also reported to be related to ADL.

Dietary diversity is an important aspect of our diet. Keeping a diverse diet is one of the key recommendations of Chinese dietary guidelines [16]. Epidemiological studies showed that a higher dietary diversity score (DDS) is inversely associated with some age-related diseases, such as diabetes [17], cognitive decline [18]. Additionally, it has been shown that adherence to a diverse diet reduced the risk of death [19,20]. However, prospective evidence on DDS and ADL disability is limited. In this study, we prospectively analyzed the effect of DDS on the risk of ADL disability in a prospective cohort study.

#### **2. Materials and Methods**

### *2.1. Study Population*

This study was based on the China Health and Nutrition Survey (CHNS). Details about the CHNS has been described elsewhere [21]. In brief, the CHNS is an open cohort study, and participants from twelve geographically diverse areas of China were involved in this study. The CHNS was designed to understand the health and nutrition of the Chinese and how they are affected by social transformation. It was initiated in 1989, and follow-up surveys were conducted in 1991, 1993, 1997, 2000, 2004, 2006, 2009, 2011, and 2015. Data collected in phase 1997 and beyond were used in the present study. The inclusion criteria included answered the ADL survey and aged 60 years and above at the end of the survey (report of ADL disability, loss to follow-up, the phase of 2015, whichever occurred first). The exclusion criteria included absent from dietary survey, report of ADL at baseline survey, report of previously diagnosed cancer at baseline survey, and missing values on covariates. In the end, 5004 participants were included this study (Figure 1). The CHNS was approved by institutional review boards at the University of North Carolina at Chapel Hill and the National Institute of Nutrition and Food Safety, Chinese Center for Disease Control and Prevention. All participants gave written informed consent before they participated in the survey.

#### *2.2. Dietary Survey and Dietary Diversity Score*

Participants' diet intakes in the past 3 days were recorded by dietary recall and household food weight inventory. Details about the dietary survey process have been published [22]. All food items that a participant consumed in a 24 h period were divided into eight food categories (cereals and tubers, vegetables, fruits, meat, aquatic products, eggs, soybeans and nuts, and dairy products). Intake of any food from each of the food categories will add one point to DDS, with a total score of eight. Average daily DDS was calculated for each participant at each phase. Cumulative average DDS across phases before the end of the survey (report of ADL disability, loss to follow-up, the phase of 2015, whichever occurred first) was computed to represent long-term diet exposure. Subsequently, average DDS was grouped into tertiles from low to high (T1: 1.33–3.25; T2: 3.27–4.03; T3: 4.06–8.00). Besides, participants' nutrient intakes were estimated according to the Chinese food composition tables, and cumulative average daily intakes were also computed.

#### *2.3. Ascertain of Disability in Activities of Daily Living*

In phase 1997, 2000, 2004, 2006, and 2015, participants aged 55 years and above were asked whether she or he could finish some self-care tasks (standing up after sitting for a long time, dressing, toileting, bathing, and feeding). For each question, participants were asked "Do you have any difficulty doing this?": no difficulty; having some difficulty, but can still do it; need help to do it; cannot do it at all; and unknown. Participants who chose "need help to do it" or "cannot do it at all" in at least one of the five activities were defined as ADL disability [23].

**Figure 1.** Flow chart of sample selection.
