Abstract
BACKGROUND/OBJECTIVES
The association between coffee and tea consumption and the risk of hyperuricemia and gout remains controversial. This meta-analysis of observational studies evaluated this association.
MATERIALS/METHODS
We conducted a literature review in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. We searched the PubMed, Embase, and the Cochrane Library databases through December 2024 using related keywords. Pooled odds ratios (ORs) with 95% confidence intervals (CIs) were calculated using a random-effects model.
RESULTS
We included 13 observational studies, 10 cross-sectional studies, and 3 cohort studies involving 27,740 hyperuricemia and gout cases among 936,827 participants. While coffee consumption was significantly associated with a decreased risk of hyperuricemia and gout (OR, 0.73; 95% CI, 0.63–0.85; I2 = 77.5%), tea consumption was not (OR, 1.02; 95% CI, 0.84–1.24; I2 = 84.2%). In the subgroup analyses, coffee consumption was significantly associated with decreased risk of hyperuricemia and gout in cohort (OR, 0.51; 95% CI, 0.40–0.66; I2 = 26.9%; n = 3) and cross-sectional (OR, 0.87; 95% CI, 0.79–0.96; I2 = 49.1%; n = 6) studies. Tea consumption was significantly associated with an increased risk of hyperuricemia and gout in both male (OR, 1.18; 95% CI, 1.00–1.41; I2 = 56.5%; n = 5) and female (OR, 1.19; 95% CI, 1.03–1.39; I2 = 0.0%; n = 5).
CONCLUSION
The results of this meta-analysis suggest that dietary coffee, but not tea, consumption reduces the risk of hyperuricemia and gout. However, further prospective cohort studies are needed to confirm the confounding factors of the association between coffee and tea consumption and hyperuricemia and gout.
Keywords: Hyperuricemia, gout, coffee, tea, meta-analysis
INTRODUCTION
Hyperuricemia is characterized by abnormally elevated serum uric acid levels and is a primary risk factor for gout [1]. Recent studies have indicated a continuous increase in the prevalence of hyperuricemia and gout, which are closely associated with advancing age, obesity, metabolic syndrome, high-purine diets, and lifestyle factors [2]. Hyperuricemia is also strongly correlated with metabolic disorders, such as insulin resistance, obesity, and renal dysfunction [3], as well as various chronic conditions, including hypertension and cardiovascular diseases [4]. Gout predominantly manifests as severe pain and inflammation in the joints, particularly in the toes, knees, and ankles, with recurrent acute attacks that can progress to chronic arthritis [5]. Therefore, preventive measures and management strategies for hyperuricemia and gout are imperative.
Coffee and tea are consumed globally. While several cross-sectional and cohort studies have suggested that their consumption may be associated with reduced risks of gout and hyperuricemia [6,7], no consistent conclusions have been established [8,9]. For instance, Guo et al. [7] reported an association between coffee and tea consumption and reduced gout risk, whereas Bae et al. [8] observed no association between coffee and tea consumption and hyperuricemia risk. Although previous meta-analyses [10,11,12] have examined the association between coffee and tea consumption and the risk of hyperuricemia or gout, they were limited by small numbers of included studies and inconsistent findings. For instance, their analysis of one cohort study and 4 cross-sectional studies, Zhang et al. [12] reported no significant association between tea consumption and hyperuricemia (odds ratio [OR], 0.84; 95% confidence interval [CI], 0.65–1.09). Moreover, Zhang et al. [10] also observed that coffee consumption was not associated with hyperuricemia but significantly reduced the risk of gout (relative risk [RR], 0.43; 95% CI, 0.31–0.59). Similarly, Li et al. [11] demonstrated that coffee consumption was not significantly associated with hyperuricemia or gout (OR, 0.72; 95% CI, 0.48–1.08). These findings indicate the need for further meta-analyses, including recent studies and well-designed prospective cohorts, to clarify these associations.
Therefore, this study aimed to present an updated systematic review and meta-analysis of the association of tea and coffee consumption with hyperuricemia and gout by incorporating recent observational research evidence.
MATERIALS AND METHODS
Search strategy
We conducted this systematic review and meta-analysis of hyperuricemia and gout according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines [13]. We systematically searched the Medline, Embase, and Cochrane Library databases for eligible studies published until December 2024. We selected medical subject headings (MeSH) terms for Medline and keywords related to coffee and tea consumption and hyperuricemia and gout risks. The complete search strategy applied in PubMed was as follows: (“Gout”[MeSH Terms] OR “Hyperuricemia”[MeSH Terms] OR (“Gout”[Text Word] OR “gouts”[Text Word] OR “Hyperuricemia”[Text Word] OR “hyperuricemias”[Text Word] OR “uric acid”[Text Word] OR “urate”[Text Word] OR “hyperuricaemia”[Text Word])) AND (“Coffee”[MeSH Terms] OR “Tea”[MeSH Terms] OR (“Coffee”[Text Word] OR “caffeine”[Text Word] OR “cichorium intybus”[Text Word] OR “coffea”[Text Word] OR “caffe”[Text Word] OR “green tea”[Text Word] OR “green teas”[Text Word] OR “tea green”[Text Word] OR “teas green”[Text Word] OR “black tea”[Text Word] OR “black teas”[Text Word] OR “tea black”[Text Word] OR “teas black”[Text Word])).
Eligibility criteria and study selection
Two authors independently evaluated all studies retrieved from the databases and bibliographies. Disagreements between the reviewers during the selection process were resolved by discussion. The study inclusion criteria were: 1) observational studies (cross-sectional, case-control, cohort); 2) coffee or tea consumption as the exposure of interest; 3) information on hyperuricemia and gout at the corresponding OR and 95% CI; 4) studies conducted in the general population; and 5) published in the English language. The exclusion criteria were 1) serum uric acid level as the outcome of interest; 2) data from the same study; 3) clinical trials, reviews, case reports, and animal studies; and 4) studies not published in English. If multiple articles were published from the same study, the most comprehensive study was included.
Data extraction
Observational studies that met all the above criteria were included in the data extraction. Two independent authors extracted the following data from the studies: first author, publication year, country of the included participants, study name, number of participants, age range, type of outcome, dietary assessment, beverage consumption, definitions of coffee and tea consumption, categories of exposure (highest versus lowest category), OR with 95% CI, and adjusted variables.
Main and subgroup analyses
In the primary analysis, we investigated the association between coffee and tea consumption and the risk of hyperuricemia and gout using adjusted ORs with 95% CIs. The groups with the highest coffee and tea consumption levels were compared with those with the lowest levels in each study. We also performed subgroup analyses according to study design, sex, age, type of outcome, dietary patterns, geographical region of participants, and methodological quality.
Quality assessment based on the risk of bias
We applied the Newcastle-Ottawa Scale (NOS) [14] for cross-sectional and cohort studies to evaluate the risk of bias. Each included study was assessed in terms of 3 subscales: selection (up to 3 points), comparability (up to 2 points), and exposure (up to 3 points). The NOS scores ranged from 0 to 9, with higher scores reflecting the lowest risk of bias. Studies with scores of 6 or less and 7 or higher were classified as low and high quality, respectively. Studies with scores of 7 or higher were classified as “high quality,” studied with score of 4 to 6 were classified as “moderate quality,” and studied with score of 3 and less were classified as “low quality.”
Statistical analysis
We used the adjusted ORs and 95% CIs reported in individual studies, comparing the highest with the lowest (reference) exposure, to calculate the pooled OR with 95% CI. The pooled effect was calculated using the inverse variance weighted mean of the logarithm of the adjusted OR with 95% CI. We evaluated heterogeneity in the results across studies using Higgins I2, which measures the percentage of total variation across studies [15]. I2 ranged between 0% (no observed heterogeneity) and 100% (maximal heterogeneity), and I2 value > 50% was considered evidence of substantial heterogeneity. We used a random-effects model meta-analysis based on the DerSimonian and Laird method in all analyses, as individual studies have high variability in design and were conducted in different populations [16]. We also examined the publication bias in the studies included in the final analysis using Begg’s funnel plot and Egger’s test. If publication bias existed, the Begg’s funnel plot was asymmetrical, or the P-value was less than 0.05, as determined by Egger’s test. All statistical analyses were performed using Stata SE version 16.1 software package (StataCorp, College Station, TX, USA).
RESULTS
Selection of relevant studies
A total of 1,668 studies were retrieved from the preliminary database search (Fig. 1). A total of 502 studies were excluded because of duplication, and 1,080 were excluded because they did not meet the selection criteria during title and abstract screening. The remaining 86 studies were assessed through a full-text review, among which 73 studies were subsequently excluded for the reasons shown in Fig. 1. The remaining 13 studies [7,8,17,18,19,20,21,22,23,24,25,26,27] were included in the final meta-analysis.
Fig. 1. PRISMA 2020 flow chart of the article selection.
PRISMA, Preferred Reporting items for Systemic Review and Meta-Analysis.
Characteristics of the included studies
Table 1 shows the characteristics of the studies included in the meta-analysis. Thirty observational studies involved 27,740 patients with hyperuricemia and gout among 936,827 participants. The studies included 10 cross-sectional [8,19,20,21,22,23,24,25,26,27] and 3 cohort studies [7,17,18]. 7 studies were conducted with participants aged ≥ 40 yrs [7,8,17,19,20,22,25], and 4 studies were conducted with participants aged ≥ 18 yrs [21,23,24,26]. Regarding exposure, 9 studies assessed coffee consumption [7,8,17,18,19,20,22,25] and 10 studies assessed tea consumption [7,8,17,18,20,21,23,25,26,27]. Regarding outcome types, 9 studies investigated hyperuricemia [8,19,20,21,23,24,25,26,27], and 4 studies investigated gout [7,17,18,22]. Eight studies were conducted in Asia [8,19,20,21,23,25,26,27], 3 in America [17,18,24], and 2 in Europe [7,22].
Table 1. General characteristics of the studies included in the final analysis.
| Studies | Country (study name) | Participants (gout and hyperuricemia/no gout and hyperuricemia) | Age (yrs) | Type of outcome | Dietary assessment | Consumption items of beverage | Definition of coffee and tea consumption | Categories of expose (highest vs. lowest category) | OR (95% CI) | Adjusted variables | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Cross-sectional study | |||||||||||
| Pham et al. (2010) [19] | Japan | 11,662 (1,020/10,642) | 49–76 | Hyperuricemia | FFQ. Self-administered | Coffee | Coffee | ≥ 4 cups/day vs. 0 cups/day | 0.70 (0.53–0.93) | Age, BMI, smoking, alcohol use, work-related physical activity and leisure-time physical activity, hypertension, diabetes, eGFR, and seafood intake | |
| Teng et al. (2013) [20] | China (The Singapore Chinese Health Study) | 483 (171/312) | 45–74 | Hyperuricemia | Validated FFQ. Interviewer administered | Coffee, green tea. black tea | Coffee | Daily drinkers vs. non-drinkers | 0.93 (0.49–1.76) | Cholesterol, creatinine, HbA1c, triglycerides, age, gender, BMI, education, cigarette smoking status, physical activity status, hypertension at baseline, daily products, red meat, fish, alcohol, and other beverages | |
| Green tea | Daily drinkers vs. non-drinkers | 0.82 (0.38–1.75) | |||||||||
| Bae et al. (2015) [8] | Korea (The Korean Multi-Rural Communities Cohort Study in Rural Communities) | 9,400 (1,021/8,379) | ≥ 40 | Hyperuricemia | Validated FFQ. Interviewer administered | Coffee, tea, caffeine | Coffee | ≥ 8.2 g/day vs. < 0.1 g/day | 0.92 (0.47–1.83) | Age, education, marital status, cigarette smoking, alcohol drinking, regular exercise, BMI, triglyceride, fasting serum glucose, hypertension medication, glomerular filtration rate, total energy, vitamin C, meat intake, seafood intake, dairy food intake, soft drink intake, added sugar in coffee, and added cream in coffee | |
| Tea | ≥ 51.5 mL/day vs. < 0.1 mL/day | 1.21 (0.98–1.50) | |||||||||
| Li et al. (2015) [21] | China | 1,372 (195/1,177) | 18–61 | Hyperuricemia | Semi-quantitative FFQ. Interviewer administered | Tea | Tea | ≥ 7 times/week vs. < 1 time/week | 0.56 (0.33–0.93) | Age, smoking and drinking status | |
| Hutton et al. (2018) [22] | Europe (The UK Biobank Resource) | 130,966 (2,135/128,831) | 40–69 | Gout | One question. Self-administered | Coffee | Coffee | per cup/day vs. < 1 cup/day | 0.85 (0.82–0.87) | Age, sex, BMI, hypertension, kidney disease, meat intake, fish intake, cheese intake, tea intake, fruit intake, vegetable intake, bread intake, and cereal intake | |
| Zeng et al. (2020) [23] | China | 918 (130/788) | 18–60 | Hyperuricemia | FFQ. Interviewer administered | Tea | Tea | Low vs. high | 0.51 (0.36–0.74) | None | |
| Agoons et al. (2021) [24] | USA (The National Health and Nutrition Examination Survey) | 2,139 (227/1,912) | ≥ 18 | Hyperuricemia | FFQ. Interviewer administered | Coffee | Coffee | ≥ 4 cups/day vs. none | 0.32 (0.09–1.14) | Age, alcohol consumption, smoking, diabetes, hypertension, eGFR, and BMI | |
| Lee et al. (2021) [25] | Korea (The Korean Genome and Epidemiology Study) | 167,752 (11,750/156,002) | ≥ 40 | Hyperuricemia | Validated FFQ. Interviewer administered | Coffee, green tea | Coffee | ≥ 3 times (d) vs. no drink | 0.95 (0.87–1.04) | Age, sex, BMI, income, past medical histories, smoking, alcohol consumption, nutritional intake, frequency of coffee, green tea, and soft drinks | |
| Green tea | ≥ 3 times (d) vs. no drink | 0.99 (0.91–1.08) | |||||||||
| Li et al. (2022) [26] | China | 7,644 (3,149/4,495) | ≥ 18 | Hyperuricemia | Questionnaire. Self-administered | Tea | Tea | ≥ 3 times (w) vs. no drink | 1.35 (1.11–1.64) | Age, sex. BMI, SCr, CHOL, TG, LDL-C, HDL-C, FBG, hypertension, education level, smoking status, alcohol drinking status, salty diet, midnight snack, and overeating corned or smoked food | |
| Ding et al. (2023) [27] | China (The China Multi-Ethnic Cohort Study) | 22,449 (3,236/19,213) | 30–79 | Hyperuricemia | One question. Interviewer administered | Tea | Tea | No harmful alcohol use but tea consumption vs. No harmful alcohol use and no tea consumption | 1.34 (1.10–1.63) | Age, sex, socioeconomic status, family history of diabetes, smoking, drinking beverage, physical activity, insufficient of vegetable and fruit intake, spicy food, and hemp diet | |
| Cohort study | |||||||||||
| Choi et al. (2007) [17] | USA (The Health Professional Follow-up Study) | 45,869 (757/45,112) | 40–75 | Gout | Validated FFQ. Self-administered | Coffee, Decaffeinated coffee, tea, caffeine | Coffee | ≥ 6 cups/day vs. 0 cup/day | 0.41 (0.19–0.88) | Age, total energy intake, BMI, diuretic use, history of hypertension, history of renal failure, alcohol, total meat, seafood, purine rich vegetables, dairy foods, total vitamin C, and the beverages presented in this table | |
| Tea | ≥ 4 cups/day vs. 0 cup/day | 0.82 (0.38–1.75) | |||||||||
| Choi et al. (2010) [18] | USA (The Nurses’ Health Study | 89,433 (896/88,537) | 30–55 | Gout | Validated FFQ. Self-administered | Coffee, Decaffeinated coffee, tea, caffeine | Coffee | ≥ 4 cups/day vs. 0 cup/day | 0.43 (0.30–0.61) | Age, total energy intake, BMI, menopause, use of hormonal replacement, diuretic use, history of hypertension, alcohol, sugar-sweetened soft drinks, total meals, seafood, chocolate, dairy foods, total vitamin C, and beverages presented in the table | |
| Tea | 4 cups/day vs. 0 cup/day | 1.55 (0.98–2.47) | |||||||||
| Guo et al. (2023) [7] | UK (The UK Biobank) | 447,658 (3,053/444,605) | 40–69 | Gout | Questionnaire. Self-administered | Coffee, tea | Coffee | > 6 cups/day vs. none | 0.60 (0.47–0.77) | Age, sex, income, TDI, ethnicity, BMI, qualification, smoking status, alcohol status, total physical activity level, sedentary time, duration of sleep, water intake, intake of processed meat, fresh fruits and vegetables, fish, prevalent hypertension, prevalent diabetes, HDL, LDL, serum urate level at baseline, diuretic use, use of antihypertensive drugs, use of antihyperlipidemic drugs, use of antidiabetic drugs, and tea consumption in coffee analysis | |
| Tea | > 6 cups/day vs. none | 0.77 (0.66–0.91) | |||||||||
OR, odds ratio; CI, confidence interval; FFQ, food frequency questionnaire; BMI, body mass index; eGFR, estimated glomerular filtration rate; HbAlc, glycosylated hemoglobin; SCr, serum creatinine; CHOL, cholesterol; TG, triglyceride; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; FBG, fasting blood glucose; TDI, Townsend deprivation index; HDL, high-density lipoprotein; LDL, low-density lipoprotein.
Methodological quality
Table 2 presents the NOS scores for each criterion in the included studies. The scores of all the studies were > 5. The mean scores were 6.4 and 7.6 for cross-sectional and cohort studies. respectively. We defined high quality studies as those with scores above 7 scores, and moderate quality studies as 4 to 6 scores for each study design. Based on this definition, 7 studies were categorized as high quality (4 studies in cross-sectional studies and 3 studies in cohort studies) and 6 as moderate quality (6 studies in cross-sectional studies).
Table 2. Quality assessment of the included studies based on the NOS.
| Studies | Selection | Comparability | Exposure | Total | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5A | 5B | 6 | 7 | 8 | |||
| Cross-sectional study1) (n = 10) | |||||||||||
| Pham et al. (2010) [19] | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 8 | |
| Teng et al. (2013) [20] | 1 | 1 | 1 | 0 | 1 | 1 | 0 | 1 | 0 | 6 | |
| Bae et al. (2015) [8] | 1 | 1 | 1 | 0 | 1 | 1 | 0 | 1 | 0 | 6 | |
| Li et al. (2015) [21] | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 8 | |
| Hutton et al. (2018) [22] | 0 | 1 | 1 | 0 | 1 | 1 | 0 | 1 | 0 | 5 | |
| Zeng et al. (2020) [23] | 0 | 1 | 1 | 1 | 0 | 0 | 0 | 1 | 1 | 5 | |
| Agoons et al. (2021) [24] | 1 | 1 | 1 | 0 | 1 | 1 | 0 | 1 | 0 | 6 | |
| Lee et al. (2021) [25] | 1 | 1 | 1 | 0 | 1 | 1 | 0 | 1 | 1 | 7 | |
| Li et al. (2022) [26] | 1 | 1 | 1 | 0 | 1 | 1 | 0 | 1 | 1 | 7 | |
| Ding et al. (2023) [27] | 1 | 1 | 1 | 0 | 1 | 1 | 0 | 1 | 0 | 6 | |
| Cohort study2) (n = 3) | |||||||||||
| Choi et al. (2007) [17] | 1 | 1 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 8 | |
| Choi et al. (2010) [18] | 1 | 1 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 8 | |
| Guo et al. (2023) [7] | 1 | 1 | 0 | 1 | 1 | 1 | 1 | 1 | 0 | 7 | |
Cross-sectional studies are presented as below: 1, Adequate definition of cases; 2, Cases are consecutive or obviously representative; 3, Selection of controls; 4, Definition of controls; 5A, Comparability of cases and controls on the basis of the design or analysis adjusted for age; 5B, Comparability of cases and controls on the basis of the design or analysis adjusted for additional factor; 6, Ascertainment of expose; 7, Same method of ascertainment for participants; 8, Non-response rate.
Cohort studies are presented as below: 1, Representativeness of the expose cohort; 2, Selection of the non-exposed cohort; 3, Ascertainment of expose; 4, Demonstration that outcome was not present at start of study; 5A, Comparability of cases and controls on the basis of the design or analysis adjusted for age; 5B, Comparability of cases and controls on the basis of the design or analysis adjusted for additional factor; 6, Ascertainment of outcome; 7, Follow-up long enough for outcome to occur; 8, Adequacy of follow-up of cohort.
NOS, Newcastle-Ottawa Scale.
1)NOS for cross-sectional studies (yes = 1, no = 0).
2)NOS for cohort studies (yes = 1, no = 0).
Meta-analysis results
Fig. 2 shows the association between coffee consumption (highest versus lowest) and the risk of hyperuricemia and gout in a random-effects meta-analysis of 9 observational studies. Coffee consumption was significantly associated with a decreased risk of hyperuricemia and gout (OR, 0.73; 95% CI, 0.63–0.85; I2 = 77.5%). Fig. 3 shows the association between tea consumption (highest versus lowest) and the risk of hyperuricemia and gout in 10 studies. Tea consumption was not associated with the risk of hyperuricemia and gout (OR, 1.02; 95% CI, 0.84–1.24; I2 = 84.2%). Publication bias assessed using funnel plots and Egger’s test revealed an asymmetric Begg’s funnel plot (Fig. 4). Therefore, we observed no publication bias in 9 studies on coffee consumption (Egger’s test, P for bias = 0.12) and 10 studies on tea consumption (Egger’s test, P for bias = 0.12).
Fig. 2. Forest plot of meta-analysis of studies on coffee consumption with the risk of hyperuricenemia and gout.
OR, odds ratio; CI, confidence interval.
1)Random-effects model.
Fig. 3. Forest plot of meta-analysis of studies on tea consumption with the risk of hyperuricemia and gout.
OR, odds ratio; CI, confidence interval.
1)Random-effects model.
Fig. 4. Funnel plots for identifying publication bias in the meta-analysis of coffee and tea consumption studies.
OR, odd ratio; SE, standard error.
Fig. 5 shows the association between coffee and tea consumption and the risk of gout. Coffee consumption was significantly associated with a decreased risk of gout (OR, 0.59; 95% CI, 0.40–0.85; I2 = 87.9%), but not in tea consumption (OR, 0.99; 95% CI, 0.60–1.61; I2 = 74.5%). Fig. 6 shows the association between coffee and tea consumption and the risk of hyperuricemia. Coffee (OR, 0.84; 95% CI, 0.68–1.04; I2 = 41.1%) and tea consumption (OR, 1.04; 95% CI, 0.83–1.30; I2 = 85.2%) were not associated with the risk of hyperuricemia.
Fig. 5. Forest plot of meta-analysis of studies on coffee and tea consumption with the risk of gout.
OR, odds ratio; CI, confidence interval.
1)Random-effects model.
Fig. 6. Forest plot of meta-analysis of studies on coffee and tea consumption with the risk of hyperuricemia.
OR, odds ratio; CI, confidence interval.
1)Random-effects model.
Subgroup meta-analyses
We performed subgroup meta-analyses according to sex, age, outcome type, dietary pattern, geographical region of the included participants, and methodological quality. Table 3 presents the results of the subgroup analyses of coffee consumption. In the subgroup analyses by study design, coffee consumption was significantly associated with a decreased risk of hyperuricemia and gout in cohort (OR, 0.51; 95% CI, 0.40–0.66; I2 = 26.9%; n = 3) and cross-sectional (OR, 0.87; 95% CI, 0.79–0.96; I2 = 49.1%; n = 6) studies. In the subgroup analyses by sex, coffee consumption was not associated with the risks of hyperuricemia and gout in male (OR, 0.79; 95% CI, 0.56–1.13; I2 = 72.4%; n = 4) or female (OR, 0.85; 95% CI, 0.42–1.71; I2 = 90.3%; n = 3). In subgroup analyses of dietary patterns, coffee (OR, 0.73; 95% CI, 0.63–0.85; I2 = 77.5%; n = 9) and decaffeinated coffee (OR, 0.76; 95% CI, 0.63–0.92; I2 = 0.0%; n = 2) consumption significantly reduced the risk of hyperuricemia and gout.
Table 3. Subgroup meta-analyses of studies on coffee consumption and the risk of hyperuricemia and gout.
| Factors | Number of studies | Summary OR (95% CI) | Heterogeneity, I2 (%) | |
|---|---|---|---|---|
| Study design | ||||
| Cohort study | 3 | 0.51 (0.40–0.66) | 26.9 | |
| Cross-sectional study | 6 | 0.87 (0.79–0.96) | 49.1 | |
| Sex | ||||
| Male | 4 | 0.79 (0.56–1.13) | 72.4 | |
| Female | 3 | 0.85 (0.42–1.71) | 90.3 | |
| Age | ||||
| 18 yrs and older | 1 | 0.32 (0.09–1.14) | 0.0 | |
| 40 yrs and older | 7 | 0.80 (0.71–0.91) | 68.4 | |
| Type of outcome | ||||
| Gout | 4 | 0.59 (0.40–0.85) | 87.9 | |
| Hyperuricemia | 5 | 0.84 (0.68–1.04) | 41.1 | |
| Dietary patterns | ||||
| Caffeine | 3 | 0.77 (0.50–1.18) | 83.1 | |
| Coffee | 9 | 0.73 (0.63–0.85) | 77.5 | |
| Decaffeinated coffee | 2 | 0.76 (0.63–0.92) | 0.0 | |
| Region | ||||
| America | 3 | 0.42 (0.31–0.57) | 0.0 | |
| Asia | 4 | 0.88 (0.74–1.04) | 27.1 | |
| Europe | 2 | 0.73 (0.52–1.02) | 86.7 | |
| Methodological quality | ||||
| High quality | 5 | 0.62 (0.45–0.88) | 87.9 | |
| Moderate quality | 4 | 0.85 (0.83–0.88) | 0.0 | |
OR, odds ratio; CI, confidence interval.
Table 4 presents the results of the subgroup analysis of tea consumption. In the subgroup analyses by study design, tea consumption was not associated with risk of hyperuricemia and gout in cohort (OR, 0.99; 95% CI, 0.60–1.61; I2 = 74.5%; n = 3) or cross-sectional (OR, 1.04; 95% CI, 0.83–1.30; I2 = 85.2%; n = 7) studies. In the subgroup analyses by sex, tea consumption was significantly associated with increased risks of hyperuricemia and gout in male (OR, 1.18; 95% CI, 1.00–1.41; I2 = 56.5%; n = 5) and female (OR, 1.19; 95% CI, 1.03–1.39; I2 = 0.0%; n = 5). In subgroup analyses of dietary patterns, green tea (OR, 1.33; 95% CI, 0.64–2.74; I2 = 76.3%; n = 2) and tea (OR, 0.97; 95% CI, 0.74–1.26; I2 = 86.6%; n = 8) consumption were not associated with risk of hyperuricemia and gout.
Table 4. Subgroup meta-analyses of studies on tea consumption and the risk of hyperuricemia and gout.
| Factors | Number of studies | Summary OR (95% CI) | Heterogeneity, I2 (%) | |
|---|---|---|---|---|
| Study design | ||||
| Cohort study | 3 | 0.99 (0.60–1.61) | 74.5 | |
| Cross-sectional study | 7 | 1.04 (0.83–1.30) | 85.2 | |
| Sex | ||||
| Male | 5 | 1.18 (1.00–1.41) | 56.5 | |
| Female | 5 | 1.19 (1.03–1.39) | 0.0 | |
| Age | ||||
| 18 yrs and older | 3 | 0.74 (0.36–1.54) | 92.7 | |
| 40 yrs and older | 5 | 1.01 (0.82–1.25) | 76.5 | |
| Type of outcome | ||||
| Gout | 3 | 0.99 (0.60–1.61) | 74.5 | |
| Hyperuricemia | 7 | 1.04 (0.83–1.30) | 85.2 | |
| Dietary patterns | ||||
| Black tea | 1 | 0.56 (0.25–1.27) | 0.0 | |
| Green tea | 2 | 1.33 (0.64–2.74) | 76.3 | |
| Tea | 8 | 0.97 (0.74–1.26) | 86.6 | |
| Region | ||||
| America | 2 | 1.22 (0.66–2.23) | 48.8 | |
| Asia | 7 | 1.04 (0.83–1.30) | 85.2 | |
| Europe | 1 | 0.77 (0.66–0.91) | 0.0 | |
| Methodological quality | ||||
| High quality | 6 | 0.98 (0.78–1.23) | 81.8 | |
| Moderate quality | 4 | 1.10 (0.71–1.69) | 87.9 | |
OR, odds ratio; CI, confidence interval.
DISCUSSION
This systematic review and meta-analysis included 13 observational studies involving 936,827 participants. The results of our quantitative synthesis showed that although tea consumption was not associated with an overall risk of hyperuricemia and gout, coffee consumption was associated with a reduced overall risk of these conditions.
The results of subgroup analyses according to study design, age, disease, type of tea consumed, and methodological quality showed no significant associations between tea consumption and the risk of hyperuricemia and gout. However, analyses stratified by sex revealed that tea consumption was associated with increased risks of hyperuricemia and gout in both males and females. Studies conducted in Europe demonstrated that tea consumption reduces the overall risk of hyperuricemia and gout, whereas studies from the United States and Asia observed no such association. We observed consistent results across study designs and methodological qualities regarding the association between coffee consumption and the risk of hyperuricemia and gout; decaffeinated coffee consumption reduced the overall risk of hyperuricemia and gout, whereas caffeine consumption did not. Studies conducted in the United States demonstrated that coffee consumption reduced the overall risk of hyperuricemia and gout; however, studies from Asia and Europe reported no association. Analysis by disease classification showed that coffee consumption reduced the risk of gout but showed no association with hyperuricemia.
Consistent with previous meta-analyses [10,11,12], we also found that coffee consumption reduced the risk of gout and showed no association with hyperuricemia risk [10,11]. Additionally, our results regarding the absence of an association between tea consumption and the risk of hyperuricemia and gout aligned with previous findings [12]. However, contrary to the findings of the meta-analysis by Li et al. [11] of no association between coffee consumption and the overall risk of hyperuricemia and gout, we observed a correlation between coffee consumption and reduced overall risk of hyperuricemia and gout.
Our meta-analysis differs from previous meta-analyses in several important aspects. First, previous meta-analyses [10,11,12] included limited numbers of studies, particularly 2 prospective cohort studies reporting on gout risk. In contrast, we included 13 studies, with 9 related to coffee and 10 related to tea, including 3 prospective cohort studies, thus incorporating more large-scale prospective cohort studies than previous research and reducing the potential for recall bias.
Although the mechanisms underlying the association between coffee consumption and the risk of hyperuricemia and gout have not yet been clearly elucidated, several potential explanations have been proposed. Coffee polyphenols (e.g., chlorogenic and caffeic acids) and roasting-derived lactones inhibit xanthine oxidase, thereby reducing urate production [23,28], while diterpenes such as cafestol and kahweol exert anti-inflammatory effects by suppressing COX-2/nuclear factor-κB signaling [29]. In animal models, aqueous extracts of Coffea arabica alleviated pain behaviors and reduced pro-inflammatory cytokine levels (interleukin [IL]-1β, IL-6, tumor necrosis factor-α), with decaffeinated extracts showing comparable efficacy [30]. In a randomized crossover trial, decaffeinated coffee lowered serum urate, whereas caffeinated coffee transiently increased urate and xanthine oxidase activity, indicating divergent effects of caffeine and non-caffeine compounds [31]. More recently, metabolomics research demonstrated that metabolites involved in the “caffeine metabolism” pathway were significantly enriched in individuals with hyperuricemia, suggesting a potential contribution of caffeine-related pathways [32]. Collectively, these findings suggest that coffee may contribute to a reduced risk of hyperuricemia and gout through mechanisms involving both inhibition of urate production and modulation of inflammatory pathways.
Our results showed that decaffeinated coffee consumption was associated with a reduced overall risk of hyperuricemia and gout, whereas caffeine consumption showed no association. In contrast, the UK Biobank cohort [7] reported a nonlinear protective association at moderate intake (≤ 3 cups/day) for both caffeinated and decaffeinated coffee, suggesting potential differences by population characteristics or study design. Mendelian randomization analyses in Japanese cohorts [33] further suggested that the protective effect of coffee may not be fully explained by serum urate levels, implying urate-independent pathways such as anti-inflammatory mechanisms.
In our study, the consumption of black tea and green tea was not associated with the overall risk of hyperuricemia and gout, in contrast to the report by Zhang et al. [12] that green tea consumption is associated with increased serum uric acid levels. Tea extracts may increase serum uric acid levels in healthy individuals but decrease levels in patients with hyperuricemia [34]. This phenomenon is interpreted as a dual effect, as polyphenols in tea can inhibit uric acid production and promote excretion [35,36,37], whereas uric acid itself can act as an antioxidant that prevents oxidation [6,17,18]. Additionally, green tea components are complex, and despite the effects of polyphenols such as epigallocatechin gallate, other elements may increase serum uric acid levels [34]. Chen et al. [38] suggested that catechin and caffeine in tea may increase uric acid levels. Therefore, better-designed studies that distinguish between tea varieties are needed.
In sex-specific subgroup analyses, tea consumption was significantly associated with an increased overall risk of hyperuricemia and gout in both males and females, contrary to the findings in the meta-analysis by Zhang et al. [12] that reported no association between tea consumption and gout risk regardless of sex. Long-term prospective studies designed to differentiate between sexes are needed to interpret these divergent results.
The results of the regional analysis in this study showed that tea consumption in Europe and coffee consumption in the United States were associated with a reduced overall risk of hyperuricemia and gout. These results may have been significantly influenced by regional differences in dietary habits, coffee and tea culture [23], as well as genetic factors [22]. Hutton et al. [22] reported that coffee consumption can modulate uric acid excretion according to glucokinase regulatory protein (GCKR) and ATP binding cassette subfamily G member 2 (ABCG2) variants, suggesting that interactions with genetic factors should also be considered. Additionally, as prospective cohort studies have only been conducted in Europe and the United States, large-scale prospective studies in Asia are needed to draw reliable conclusions.
This meta-analysis has several strengths. First, compared with previous studies, the addition of large-scale prospective cohort studies reduced the potential for recall bias. Second, most of the included studies were analyzed based on multivariate-adjusted results, thus minimizing confounding variables and leading to more reasonable conclusions by utilizing large sample sizes. Third, this is the first meta-analysis to comprehensively and simultaneously compare the associations between the effects of tea and coffee and the risk of hyperuricemia and gout. Finally, we observed no publication bias.
However, this meta-analysis also has certain limitations. First, owing to limitations in the literature, few suitable studies qualified for inclusion in this meta-analysis. Second, the adjusted confounding variables in the included studies were inconsistent, and the high heterogeneity among the studies may have distorted the meta-analysis results. Third, coffee and tea intake was assessed using self-reported frequency, and considerable variation in exposure definitions (e.g., intake units and cut-off points) prevented dose–response meta-analysis. These factors limit the precision and clinical applicability of our findings. Future studies using standardized intake metrics and harmonized exposure categories are warranted to clarify dose–response relationships.
In conclusion, this study presents evidence from an updated meta-analysis incorporating recent research and prospective study data compared with previous meta-analyses. The results demonstrate the association of coffee consumption with a reduced overall risk of hyperuricemia and gout, with a particular association observed for a decreased risk of gout. However, we observed no significant association between coffee consumption and hyperuricemia, or between tea consumption and the risk of hyperuricemia and gout. Nevertheless, owing to the methodological limitations of the present investigation, caution is warranted when drawing definitive conclusions. Better-designed large-scale prospective studies and randomized controlled clinical trials are needed to elucidate the associations between coffee and tea consumption and the risks of hyperuricemia and gout.
Footnotes
Funding: This work was supported by the Shinhan University Research Fund, 2024.
Institutional Review Board Statement: Not applicable.
Conflict of Interest: The authors declare no potential conflicts of interests.
- Conceptualization: Hong SH, Kim JM.
- Data curation: Hong SH, Kim JM.
- Supervision: Hong SH, Kim JM.
- Visualization: Hong SH, Kim JM.
- Writing - original draft: Hong SH, Kim JM.
- Writing - review & editing: Hong SH, Kim JM.
References
- 1.Ni Q, Lu X, Chen C, Du H, Zhang R. Risk factors for the development of hyperuricemia: a STROBE-compliant cross-sectional and longitudinal study. Medicine (Baltimore) 2019;98:e17597. doi: 10.1097/MD.0000000000017597. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Dehlin M, Jacobsson L, Roddy E. Global epidemiology of gout: prevalence, incidence, treatment patterns and risk factors. Nat Rev Rheumatol. 2020;16:380–390. doi: 10.1038/s41584-020-0441-1. [DOI] [PubMed] [Google Scholar]
- 3.Sluijs I, Beulens JW, van der A DL, Spijkerman AM, Schulze MB, van der Schouw YT. Plasma uric acid is associated with increased risk of type 2 diabetes independent of diet and metabolic risk factors. J Nutr. 2013;143:80–85. doi: 10.3945/jn.112.167221. [DOI] [PubMed] [Google Scholar]
- 4.Feig DI, Madero M, Jalal DI, Sanchez-Lozada LG, Johnson RJ. Uric acid and the origins of hypertension. J Pediatr. 2013;162:896–902. doi: 10.1016/j.jpeds.2012.12.078. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hainer BL, Matheson E, Wilkes RT. Diagnosis, treatment, and prevention of gout. Am Fam Physician. 2014;90:831–836. [PubMed] [Google Scholar]
- 6.Choi HK, Curhan G. Coffee, tea, and caffeine consumption and serum uric acid level: the third national health and nutrition examination survey. Arthritis Rheum. 2007;57:816–821. doi: 10.1002/art.22762. [DOI] [PubMed] [Google Scholar]
- 7.Guo H, Wang S, Peng H, Wang M, Li L, Huang J, Wu T. Dose-response relationships of tea and coffee consumption with gout: a prospective cohort study in the UK Biobank. Rheumatology (Oxford) 2023;62:3043–3050. doi: 10.1093/rheumatology/kead019. [DOI] [PubMed] [Google Scholar]
- 8.Bae J, Park PS, Chun BY, Choi BY, Kim MK, Shin MH, Lee YH, Shin DH, Kim SK. The effect of coffee, tea, and caffeine consumption on serum uric acid and the risk of hyperuricemia in Korean Multi-Rural Communities Cohort. Rheumatol Int. 2015;35:327–336. doi: 10.1007/s00296-014-3061-8. [DOI] [PubMed] [Google Scholar]
- 9.Teng GG, Pan A, Yuan JM, Koh WP. Food sources of protein and risk of incident gout in the Singapore Chinese Health Study. Arthritis Rheumatol. 2015;67:1933–1942. doi: 10.1002/art.39115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Zhang Y, Yang T, Zeng C, Wei J, Li H, Xiong YL, Yang Y, Ding X, Lei G. Is coffee consumption associated with a lower risk of hyperuricaemia or gout? A systematic review and meta-analysis. BMJ Open. 2016;6:e009809. doi: 10.1136/bmjopen-2015-009809. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Li R, Yu K, Li C. Dietary factors and risk of gout and hyperuricemia: a meta-analysis and systematic review. Asia Pac J Clin Nutr. 2018;27:1344–1356. doi: 10.6133/apjcn.201811_27(6).0022. [DOI] [PubMed] [Google Scholar]
- 12.Zhang Y, Cui Y, Li XA, Li LJ, Xie X, Huang YZ, Deng YH, Zeng C, Lei GH. Is tea consumption associated with the serum uric acid level, hyperuricemia or the risk of gout? A systematic review and meta-analysis. BMC Musculoskelet Disord. 2017;18:95. doi: 10.1186/s12891-017-1456-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, Shamseer L, Tetzlaff JM, Akl EA, Brennan SE, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. doi: 10.1136/bmj.n71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Wells GA, Shea B, O’Connell D, Peterson J, Welch V, Losos M, Tugwell P. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses [Internet] Ottawa: Ottawa Hospital Research Institute; 2024. [cited 2024 December 31]. Available from: http://www.ohri.ca/programs/clinical_epidemiology/oxford.htm. [Google Scholar]
- 15.Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med. 2002;21:1539–1558. doi: 10.1002/sim.1186. [DOI] [PubMed] [Google Scholar]
- 16.DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7:177–188. doi: 10.1016/0197-2456(86)90046-2. [DOI] [PubMed] [Google Scholar]
- 17.Choi HK, Willett W, Curhan G. Coffee consumption and risk of incident gout in men: a prospective study. Arthritis Rheum. 2007;56:2049–2055. doi: 10.1002/art.22712. [DOI] [PubMed] [Google Scholar]
- 18.Choi HK, Curhan G. Coffee consumption and risk of incident gout in women: the Nurses’ Health Study. Am J Clin Nutr. 2010;92:922–927. doi: 10.3945/ajcn.2010.29565. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Pham NM, Yoshida D, Morita M, Yin G, Toyomura K, Ohnaka K, Takayanagi R, Kono S. The relation of coffee consumption to serum uric Acid in Japanese men and women aged 49–76 years. J Nutr Metab. 2010;2010:930757. doi: 10.1155/2010/930757. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Teng GG, Tan CS, Santosa A, Saag KG, Yuan JM, Koh WP. Serum urate levels and consumption of common beverages and alcohol among Chinese in Singapore. Arthritis Care Res (Hoboken) 2013;65:1432–1440. doi: 10.1002/acr.21999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Li X, Song P, Li J, Wang P, Li G. Relationship between hyperuricemia and dietary risk factors in Chinese adults: a cross-sectional study. Rheumatol Int. 2015;35:2079–2089. doi: 10.1007/s00296-015-3315-0. [DOI] [PubMed] [Google Scholar]
- 22.Hutton J, Fatima T, Major TJ, Topless R, Stamp LK, Merriman TR, Dalbeth N. Mediation analysis to understand genetic relationships between habitual coffee intake and gout. Arthritis Res Ther. 2018;20:135. doi: 10.1186/s13075-018-1629-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Zeng J, Zhang J, Li Z, Li T, Li G. Prediction model of artificial neural network for the risk of hyperuricemia incorporating dietary risk factors in a Chinese adult study. Food Nutr Res. 2020;64:3712. doi: 10.29219/fnr.v64.3712. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Agoons DD, Agoons BB, Kaze AD, Komanduri S. Effect of serum ferritin on the association between coffee intake and hyperuricemia among American women: the national health and nutrition examination survey. Cureus. 2021;13:e13855. doi: 10.7759/cureus.13855. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Lee JS, Kim TJ, Hong SK, Min C, Yoo DM, Wee JH, Choi HG. Impact of coffee/green tea/soft drink consumption on the risk of hyperuricemia: a cross-sectional study. Int J Environ Res Public Health. 2021;18:7299. doi: 10.3390/ijerph18147299. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Li R, Zeng L, Wu C, Ma P, Cui H, Chen L, Li Q, Hong C, Liu L, Xiao L, et al. Tea consumption is associated with an increased risk of hyperuricemia in an occupational population in Guangdong, China. Int J Gen Med. 2022;15:2747–2757. doi: 10.2147/IJGM.S355253. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Ding X, Chen L, Tang W, Chen T, Xu J, Yang X, Ding R, Tang X. Interaction of harmful alcohol use and tea consumption on hyperuricemia among Han residents aged 30–79 in Chongqing, China. Int J Gen Med. 2023;16:973–981. doi: 10.2147/IJGM.S401889. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Honda S, Miura Y, Masuda A, Masuda T. Identification of crypto- and neochlorogenic lactones as potent xanthine oxidase inhibitors in roasted coffee beans. Biosci Biotechnol Biochem. 2014;78:2110–2116. doi: 10.1080/09168451.2014.946397. [DOI] [PubMed] [Google Scholar]
- 29.Ren Y, Wang C, Xu J, Wang S. Cafestol and kahweol: a review on their bioactivities and pharmacological properties. Int J Mol Sci. 2019;20:4238–4252. doi: 10.3390/ijms20174238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Matosinhos RC, Bezerra JP, Barros CH, Fernandes Pereira Ferreira Bernardes AC, Coelho GB, Carolina de Paula Michel Araújo M, Dian de Oliveira Aguiar Soares R, Sachs D, Saúde-Guimarães DA. Coffea arabica extracts and their chemical constituents in a murine model of gouty arthritis: how they modulate pain and inflammation. J Ethnopharmacol. 2022;284:114778–114788. doi: 10.1016/j.jep.2021.114778. [DOI] [PubMed] [Google Scholar]
- 31.Towiwat P, Tangsumranjit A, Ingkaninan K, Jampachaisri K, Chaichamnong N, Buttham B, Louthrenoo W. Effect of caffeinated and decaffeinated coffee on serum uric acid and uric acid clearance, a randomised within-subject experimental study. Clin Exp Rheumatol. 2021;39:1003–1010. doi: 10.55563/clinexprheumatol/gd9nia. [DOI] [PubMed] [Google Scholar]
- 32.Liu H, Xie R, Dai Q, Fang J, Xu Y, Li B. Exploring the mechanism underlying hyperuricemia using comprehensive research on multi-omics. Sci Rep. 2023;13:7161–7172. doi: 10.1038/s41598-023-34426-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Shirai Y, Nakayama A, Kawamura Y, Toyoda Y, Nakatochi M, Shimizu S, Shinomiya N, Okada Y, Matsuo H Japan Gout Genomics Consortium (Japan Gout) Coffee consumption reduces gout risk independently of serum uric acid levels: mendelian randomization analyses across ancestry populations. ACR Open Rheumatol. 2022;4:534–539. doi: 10.1002/acr2.11425. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Peluso I, Teichner A, Manafikhi H, Palmery M. Camellia sinensis in asymptomatic hyperuricemia: a meta-analysis of tea or tea extract effects on uric acid levels. Crit Rev Food Sci Nutr. 2017;57:391–398. doi: 10.1080/10408398.2014.889653. [DOI] [PubMed] [Google Scholar]
- 35.Roth M, Timmermann BN, Hagenbuch B. Interactions of green tea catechins with organic anion-transporting polypeptides. Drug Metab Dispos. 2011;39:920–926. doi: 10.1124/dmd.110.036640. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Fuchikami H, Satoh H, Tsujimoto M, Ohdo S, Ohtani H, Sawada Y. Effects of herbal extracts on the function of human organic anion-transporting polypeptide OATP-B. Drug Metab Dispos. 2006;34:577–582. doi: 10.1124/dmd.105.007872. [DOI] [PubMed] [Google Scholar]
- 37.Vaidyanathan JB, Walle T. Transport and metabolism of the tea flavonoid (-)-epicatechin by the human intestinal cell line Caco-2. Pharm Res. 2001;18:1420–1425. doi: 10.1023/a:1012200805593. [DOI] [PubMed] [Google Scholar]
- 38.Chen DD, Dong XX, Yang XJ, Sun HP, Liang G, Chen X, Pan CW. Tea consumption and serum uric acid levels among older adults in three large-scale population-based studies in China. BMC Geriatr. 2021;21:267. doi: 10.1186/s12877-021-02216-8. [DOI] [PMC free article] [PubMed] [Google Scholar]