Lower 24‐h urinary potassium excretion is associated with higher prevalent depression and anxiety status in general population

Abstract Background Uncertainty remains about the association of potassium (K) intake with depression and anxiety status. We explored their relationship using 24‐h urinary K, reflecting K intake, in general population. Methods We collected 24‐h urine and performed self‐rating depression and anxiety scales (SDS, SAS) cross‐sectionally in adults selected by random sampling in China. SDS and SAS standard score ≥50 defined depression and anxiety status. Participants were divided into three groups (T1, T2, and T3) by 24‐h urinary K tertile. Odds ratios (OR) and 95% confidence intervals were calculated. Sensitivity analysis was performed by excluding anti‐hypertensive agent takers. Results 546 participants comprised current analytical sample. First, T1 and T2 groups showed higher SDS scores (40.0 vs 40.0 vs 36.0, p = .001), prevalence (19.8 vs 15.9 vs 7.1%, p = .002), whereas increased adjusted odds for depression status only in T1 group (OR = 2.71, p = .017), compared with T3 group. Second, T1 and T2 groups showed higher SAS scores (38.0 vs 40 vs 35.0, p < .001) and prevalence (14.8 vs 21.4 vs 8.8%, p = .003), whereas increased adjusted odds for anxiety status only in T2 group (OR = 2.07, p = .042), compared with T3 groups. Third, T1 and T2 groups showed higher prevalence (10.4% vs 11.5% vs 2.7%, p = .004) and adjusted odds (OR = 3.71, p = .013; OR = 3.66, p = .014) for co‐existent anxiety and depression status, compared with T3 group. Most results remained consistent in sensitivity analysis. Conclusions Lower K intake is implicated in presence of anxiety and depression status in general population; this may provide basis for programs to increase K intake and prevent disease.


INTRODUCTION
Mental disorders have become one of the serious public health problems in the world, with depression and anxiety as the two most common mental disorders in the general medical setting (Andrade et al., 2003;Kroenke et al., 2007). Studies show that the lifetime prevalence of depression and anxiety averaged 11.2% and 3.7%, respectively (Kessler et al., 2015;Ruscio et al., 2017). Since the 1990s, depressive disorders, second only to ischemic heart disease, have become the second greatest contributor to global disease burden, quantified as years of life lived in less than ideal health (Vos et al., 2012). In addition, disability-adjusted life-years of depression and anxiety increased by 14. 3% and 12.8% during 2007(GBD 2017Dalys & Hale Collaborators, 2018. Therefore, exploring modifiable risk factors is necessary to prevent depression and anxiety. Although accumulating studies have been conducted in the past few decades, the etiologies for depression and anxiety are not understood clearly. Female gender, low socioeconomic status, less social support, stress, alcohol and drug abuse, genetic and epigenetic factors, dysregulation of gut microbiota, and several disease conditions contribute to increased risk for depression and anxiety (Hosseinzadeh et al., 2016).
In addition, balanced diet or dietary patterns play important roles in human model of thinking and human behavior, as the intake of foods affects human cognition, memory capacity, and emotions (Huang et al., 2019); balanced dietary patterns such as the Mediterranean diet have been uniquely associated with a lower risk of depression or depressive symptoms (Parletta et al., 2019;Rienks et al., 2013).
Besides a balanced diet, isolated nutrients are another element that might also be involved in mental disorders. Emerging evidence suggests that potassium (K) intake might be one of the modifiable risk factors for depression and anxiety, whereas inconclusive. In 1423 Japanese elderly population, K intake is significantly and negatively correlated with depressive symptoms among female participants (Thi Thu Nguyen et al., 2019). Furthermore, K intake is found lower in 59 patients with depression aged ≥18 years in another study (Kaner et al., 2015) and serum K is decreased slightly in 200 preoperative patients with anxiety (McCleane & Watters, 1990). Moreover, consumption of vegetables and fruits, well-known as sources of K and recommended for the prevention of depression (Opie et al., 2017), is negatively correlated with severity of depressive symptoms (Mamplekou et al., 2010) and presence of anxiety (Sadeghi et al., 2021;Saghafian et al., 2018).
Importantly, an intervention study reports that a low-sodium and high-K diet seems to have positive effects on general mood state including assessment of depression and anxiety (Mrug et al., 2019;Torres et al., 2008). Therefore, it is reasonable to speculate that K intake is associated with the development of depression and anxiety and a proper K intake may have positive effects on the prevention/reduction of the disease status, if the association between the two can be established.

Study population
In this cross-sectional study, we obtained study participants aged ≥18 years using multistage proportional random sampling method from Emin county, Xinjiang, China between March and June 2019. Eligible populations were asked to participate in 24-h urine collection and post examination. Details were described in our previous study (Abudoureyimu et al., 2021;Li et al., 2022;Wang et al., 2021). In brief, the county was divided into 20 sites at the first stage. At the second stage, 10 sites were selected. At the third stage, participants were selected from locals, based on inclusion and exclusion criteria as in Figure 1.

Measures
2.2.1 24-h urine sample collection and measurement As described previously (Abudoureyimu et al., 2021;Wang et al., 2021), a complete scheme was applied to the collection and detection process of 24-h urine samples. Urine samples meeting any of the following exclusion criteria were not included in this study: the urine F I G U R E 1 Flowchart for study participants. (a) Laboratory data include serum potassium test value, 24-h urinary potassium test value, and 24-h urinary sodium test value. (b) Questionnaire information included general information, SAS data, SDS data, and PSQI data. volume < 500 ml; the duration of specimen collection < 20 h; reported more than 100 ml of urine lost during collection; the 24-h urinary creatinine (24-h UCr) per kilogram of body weight was not up to standard (Mohammadifard et al., 2019;Nerbass et al., 2014). The detection of electrolytes (i.e., K, sodium, etc.) and creatinine in urine samples were carried out uniformly.

2.2.2
Data and blood sample collection

SDS and SAS scales, depression, and anxiety status
The SDS is a self-report scale, which contains 20 items reflecting subjective feelings of depression, of which 10 are positive and 10 are negative. Each item was rated with respect to how participants felt using a four-point Likert-type scale. Options include: 1 = little or no time, 2 = a small part of the time, 3 = quite a lot of time, 4 = most or all of the time. Forward scoring questions were scored as 1, 2, 3, and 4; reverse scoring questions were scored as 4, 3, 2, and 1. Reverse scoring question number: 2, 5, 6, 11, 12, 14, 16, 17, 18, and 20. The original total score of SDS is between 20 and 80, and the result is usually expressed as the standard total score of SDS, which is obtained by multiplying the original score by 1.25 and taking the integer part (Zung, 1965). The SAS is also a self-report scale with 20 items covering a series of anxiety symptoms (Zung, 1971). In the current study, participants were instructed to choose how often they experienced each symptom over the past week given on a four-point Likert-type scale, ranging from 1 (little or no time) to 4 (most or all of the time). Items include positive and negative experiences. Forward scoring questions were scored as 1, 2, 3, and 4; reverse scoring questions were scored as 4, 3, 2, and 1. Reverse scoring question number: 5, 9, 13, 17, and 19. The standard total score of the SAS ranges from 25 to 100. The higher the standard score, the more serious the symptom.
Depression status was defined as SDS standard score ≥50, anxiety status as SAS standard score ≥50, and coexistence of the two as SDS standard score ≥50 and SAS standard score ≥50 (Zung et al., 1990).

2.2.4
Definitions of other covariates . Poor sleep quality was defined as a PSQI score > 6 ( Antza et al., 2018;Buysse et al., 1989;Zheng et al., 2013). Education attainment status was categorized as middle school and lower and high school and higher. Occupations were divided into two types as mental and manual work. Alcohol intake was defined as drinking at least once a week within 1 month at the survey time point . Cigarette consumption was defined as smoking more than 20 packs of cigarettes and smoking currently ).

Statistical analysis
Participants were divided into three groups by the tertile of 24-h UK as Logistic regression analysis was used to assess the association of the tertiles of 24-h UK (T3 group of 24-h UK as the reference) and the presence of depression status, anxiety status, and co-existence of the two, and results were presented as unadjusted and adjusted odds ratios (ORs) and the 95% confidence intervals (CI).
Independent variables significantly relevant to depression and or anxiety status (p < .1) in univariate Logistic regression analysis were adjusted in multivariate logistic analysis. Tolerance and the variance inflation factor were examined to identify multicollinearity, which could be concerned if the variance inflation factor was > 10 and the tolerance was < 0.10.
Sensitivity analysis was conducted to compare SDS, SAS score, and logistic regression analysis by excluding participants with hypertension under anti-hypertensive treatment.
Results were considered statistically significant if two-tailed p value was less than .05. All analyses were performed with SPSS statistical software, version 20.0 (Chicago, IL, USA).

Participant characteristics at baseline
As in Figure 1,

The SAS score and anxiety status
Participants in T1 (38.0 vs 35.0, p = .019) and T2 groups (40.0 vs 35.0, p < .001) showed significantly higher SDS scores than did those in T3 group, which remained consistent in sensitivity analysis by excluding hypertensives under anti-hypertensive treatment (Table 2).
In logistic regression analysis (

Co-existence of anxiety and depression status
As in Table 3, compared with T3 groups, participants in T1 (10.4% vs 2.7%, p = .023) and T2 (11.5% vs 2.7%, p = .007) groups showed significantly higher presence of co-existent anxiety and depression status in total participants, which is largely consistent in women (T1 vs T2 vs T3: 14.3% vs 18.9 vs 3.8%, p = .003) but not in men.
In logistic regression analysis (Table 4)

DISCUSSION
To our knowledge, current study is the first to explore the relationship between 24-h UK excretion and depression status, anxiety status, and co-existence of the both in relatively large sample community- For depression status, adjusted model was adjusted for age, gender, education attainment status, occupation, body mass index, systolic and diastolic blood pressure, diabetes, dyslipidemia, sleep quality, serum K, and 24-h urinary sodium excretion. For anxiety status, adjusted model was adjusted for gender, education attainment status, occupation, cigarette and alcohol use, systolic and diastolic blood pressure, dyslipidemia, sleep quality, and 24-h urinary sodium excretion. For co-existent depression and anxiety status, adjusted model was adjusted for gender, education attainment status, occupation, systolic and diastolic blood pressure, and sleep quality.
their co-existence, which remained consistent in sensitivity analysis by excluding potential confounding of anti-hypertensive agents for depression status and for co-existence of the both. These results imply that lower K intake, indicated by lower UK excretion, may be involved in the development of depression status and or anxiety status in the general population.
Observations from the current study add evidence on the ongoing uncertainty of K and depression and anxiety status. Current findings are consistent with results of previous reports. For example, some studies based on DASH diet model have shown that low sodium and high potassium intake can significantly reduce depression and anxiety score and improve depression and anxiety status (Mrug et al., 2019;Torres et al., 2008). In addition, the current findings may extend some previous studies from the clinical environment to the general adult population.
For example, consumption of foods high in sodium and low in potassium contributes to the development of depressive and anxiety symptoms in early adolescence (Mrug et al., 2019), and low-sodium and highpotassium diet seems to have an overall positive effect on depressive and anxiety mood state (Torres et al., 2008).
Urinary potassium excretion has been established as a marker of overall diet quality, with higher potassium excretion positively correlated with greater intake of vegetables, fruits, whole grains, fish, and poultry, and negatively correlated with the intake of fast food and red meat (Mente et al., 2009). Therefore, with regard to public health, prevention, control, or improvement of depression, anxiety, and the comorbidity by K supplements could have public health implications since it is feasible to increase K intake by fresh vegetables and fruits, and even salt substitute, which is a practical and cost-effective approach to supplement K and has been shown to slow the incidence of hypertension (Bernabe-Ortiz et al., 2020), another risk factor for depression and anxiety.
In the current study, we included several confounding factors for parameters of interest. For example, we had data on anthropometric (BMI) and socioeconomic indicators (education attainment status and occupation), disease history (diabetes, hypertension), and lifestyle factors (cigarette and alcohol use and sleep quality), which have effects on interest of outcome. We also collected data on anti-hypertensive agent use, most of which exerts effects on K and Na. Furthermore, we performed univariate and multi-variate logistic regression and sensitivity analysis to obtain objective association of 24-h UK and depression and/or anxiety status, although some of the data showed significant differences among groups. For example, BMI increased significantly from T1 to T3 group of UK tertile, whereas its effects on the results were ruled out using regression analysis.
The main strength of this study is the objective measurement of K excretion using 24-h urine samples. Second, the current study was conducted in general population with wide age ranges from both genders. Therefore, results may be generalizable, although limited to a certain county. Third, we adjusted for the sleep quality, an independent risk factor for depression and anxiety (Huang & Zhu, 2020), in logistic regression analysis, and the association still remained significant. Nonetheless, the current study also contains some limitations.
A key limitation is that we failed to obtain causality between K with depression and or anxiety due to the cross-sectional nature of the study. Second, due to seasonal or daily changes in urinary potassium excretion, a single 24-h urine sample may not reflect the usual dietary intake or pattern of participants. Nevertheless, we collected urine samples on weekdays and weekends, and from spring to summer, which may have reduced the relevant variability. In addition, we used SDS and SAS to evaluate the depressive and anxiety mood of the study population and roughly screen out possible depression and anxiety status patients. Nevertheless, SDS and SAS have been widely used including scientific researches with higher diagnostic sensitivity (SDS: 92%; SAS: 89.0%) and specificity (SDS: 77%; SAS: 69.0%) (Dunstan et al., 2017;Gabrys & Peters, 1985). In addition, both questionnaires are confirmed to be reliable and valid in Chinese population (Liu et al., 2018). Moreover, we failed to assess overall dietary intake pattern in this study population, which may provide more information on the association of K intake and depression and or anxiety. However, the main objective of the study was to assess the association of K and depression and or anxiety status, and K in 24-h urine sample may provide more accurate assessment of 24-h K intake, since studies show that 77% of K ingested is excreted through urine (World Health Organization, 2012).

CONCLUSIONS
Lower 24-h UK excretion shows independent association with higher SDS and SAS scores and with higher prevalent depression and or anxiety status in general adults from China, suggesting lower K intake may be involved in depression and or anxiety, and this could lead to optimization programs focused on increasing potassium intake at the population level, which has been shown to be feasible and result in disease prevention.

CONFLICT OF INTEREST
The authors declare no conflict of interest.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.