Diet Quality as Assessed by the Healthy Eating Index, Alternate Healthy Eating Index, Dietary Approaches to Stop Hypertension Score, and Health Outcomes: A Second Update of a Systematic Review and Meta-Analysis of Cohort Studies

Background Suboptimal diet quality has a large impact on noncommunicable disease burden. Objective This study aimed to update the body of evidence on the associations between diet quality, as assessed by the Healthy Eating Index, Alternate Healthy Eating Index, and the Dietary Approaches to Stop Hypertension score, and health status. Moreover, results of the previous systematic reviews and meta-analyses were extended by evaluating the credibility of the evidence. Methods PubMed, Embase, and Scopus databases were searched to identify eligible studies published between May 15, 2017 and March 14, 2020. Pooled relative risk (RR) with 95% CI for highest vs lowest category of diet quality were estimated using a random-effects model. Heterogeneity was explored using Cochran ’ s Q test and I 2 statistic with 95% CI. Presence of publication bias was detected by using funnel plots and Egger ’ s regression test. The NutriGrade tool was used to assess the credibility of evidence. Results The current update identi ﬁ ed 47 new reports, resulting in a total of 113 reports including data from 3,277,684 participants. Diets of the highest quality, as assessed by the Healthy Eating Index, Alternate Healthy Eating Index, and Dietary Approaches to Stop Hypertension scores, were inversely associated with risk of all-cause mortality (RR 0.80, 95% CI 0.79 to 0.82, I 2 ¼ 68%, n ¼ 23), cardiovascular disease incidence or mortality (RR 0.80, 95% CI 0.78 to 0.82, I 2 ¼ 59%, n ¼ 45), cancer incidence or mortality (RR 0.86, 95% CI 0.84 to 0.89, I 2 ¼ 73%, n ¼ 45), incidence of type 2 diabetes (RR 0.81, 95% CI 0.78 to 0.85, I 2 ¼ 76%, n ¼ 16), and incidence of neurodegenerative diseases (RR 0.82, 95% CI 0.75 to 0.89, I 2 ¼ 71%, n ¼ 12). In cancer survivors, the highest diet quality was linked with lower risk of all-cause (RR 0.83, 95% CI 0.77 to 0.88, I 2 ¼ 45%, n ¼ 12) and cancer mortality (RR 0.82, 95% CI 0.75 to 0.89, I 2 ¼ 44%, n ¼ 12). The credibility of evidence for identi ﬁ ed associations between overall healthy dietary patterns and included health outcomes was moderate. Conclusion This updated systematic review and meta-analysis suggests that high diet quality (assessed by the Healthy Eating Index, Alternate Healthy Eating Index, and Di- etary Approaches to Stop Hypertension) is inversely associated with risk of all-cause mortality, cardiovascular disease incidence or mortality, cancer incidence or mortality, type 2 diabetes, and neurodegenerative disease, as well as all-cause mortality and cancer mortality among cancer survivors. Moderate credibility of evidence for identi ﬁ ed associations complements the recent 2020 Dietary Guidelines Advisory Committee report recommending healthy dietary patterns for disease prevention.

dietary patterns are considered crucial to disease prevention.
In 2015, we published a systematic review and metaanalysis on associations between diet quality, as assessed by the Healthy Eating Index (HEI), Alternate Healthy Eating Index (AHEI), and Dietary Approaches to Stop Hypertension (DASH) score, and risk of all-cause mortality, CVD mortality or events, cancer mortality or incidence, T2D, and neurodegenerative disease. 3 Moreover, in an updated version of this systematic review, the associations of these diet quality indices with allcause mortality and cancer mortality among cancer survivors were also investigated. 4 Pooled data from 68 prospective cohort studies suggested that diets that scored highly on HEI, AHEI, and DASH were inversely associated with risk of allcause mortality, CVD, cancer, T2D, and neurodegenerative disease by 22%, 22%, 16%, 18%, and 15%, respectively. 4 Among cancer survivors, a high-quality diet was inversely associated with risk of all-cause mortality and cancer mortality. 4 Numerous new studies have appeared since this last publication that are eligible for inclusion in this systematic review. In addition, addressing the credibility of the evidence is crucial in health decision-making. Currently, there is heightened awareness of the need for an appropriate evaluation of the certainty of evidence synthesis in nutrition research. 5 For this reason, this update aimed to extend previous analysis by assessing the credibility of evidence using the NutriGrade tool, 6 which was applied previously in a series of systematic reviews on food groups. [7][8][9][10] METHODS A predefined protocol for previous versions of this systematic review and meta-analysis was registered in PROSPERO International Prospective Register of Systematic Reviews (https://www.crd.york.ac.uk/PROSPERO/; ID: CRD42013006561). Changes made to the protocol are annotated in the text. The present update of the review was conducted and reported according to the Meta-Analysis of Observational Studies in Epidemiology guidelines. 11

Data Sources and Searches
An updated literature search for studies published from May 15, 2017 through March 14, 2020 was conducted in electronic databases (PubMed, Embase, and Scopus) with no restrictions on the language of publication. The search strategy for the PubMed database was adopted as follows: ("healthy"[All fields] AND ("eating"[All fields] OR "eating"[MeSH Terms])) AND ("abstracting and indexing as topic"[MeSH Terms] OR ("abstracting"[All fields] AND "indexing"[All fields] AND "topic"[All fields]) OR "abstracting and indexing as topic"[All fields] OR "index"[All fields]) OR ("dash"[All fields] AND ("diet"[All fields] OR "diet"[MeSH Terms]). Reference lists from included articles were checked in order to identify additional eligible studies. One author (J.M.) performed the literature search and any uncertainties were resolved through discussion with another author (L.S.).

Study Selection
Studies were considered as eligible for inclusion if they met the following criteria: conducted on adult populations (aged 18 years or older); evaluated the association of diet quality as assessed by the HEI, AHEI, or DASH score on risk of all-cause mortality, CVD incidence or mortality, cancer incidence or mortality, T2D, and neurodegenerative disease in the general population and on all-cause mortality or cancer mortality among cancer survivors; and had a prospective observational design. Studies performed exclusively in patients with chronic diseases (such as CVD, diabetes, chronic kidney disease, or frailty syndrome) were excluded. In the case of 2 reports based on the same study, the one with the longer follow-up or with a larger number of cases was included. Two authors (J.M. and A.D.) independently performed the title and abstract screening, as well as study selection, and any disagreements were resolved by discussion with the third author (L.S.).

Data Extraction
After selecting the studies, the following information was extracted from each report: name of the first author, year of publication, study location and name, sample size, age at entry to the study, sex, length of follow-up, outcome, number of cases, diet quality index, adjustment factors, risk estimate (multivariable-adjusted odds ratio [OR], risk ratio [RR], or hazard ratio [HR]), and study quality. In contrast to the previous version of this review, if a study provided separate estimates for men and women, they were pooled using the fixed-effects model before inclusion in the analysis (according to the methodological recommendations of the World Cancer Research Fund 12 ). When a study reported multiple risk estimates, the one adjusted for the highest number of confounders was selected. Study data were extracted by one author (J.M.) and verified by another (L.S.).

Study Quality and Credibility of Evidence Assessment
The methodological quality of studies was evaluated using the Newcastle Ottawa Assessment Scale for Cohort Studies, as in the previous versions. 13 The NutriGrade tool was used to quantify the credibility of evidence for the association between diet quality and predefined outcomes. 6 In brief, a summary score was calculated and interpreted as very low (0 to <4 points), low (4 to <6 points), moderate (6 to <8 points), or high (8 to 10 points) credibility of the evidence (details of the NutriGrade tool items are provided in the footnotes for Table 1). Separate judgments on credibility of evidence were made for overall high diet quality (all 3 indices), HEI, AHEI, and DASH score. Table 1. Item-level scoring for NutriGrade tool and credibility of evidence for association between the diet quality (overall high diet quality, HEI, a AHEI, b or DASH c score) and all-cause mortality, CVD d incidence or mortality, cancer incidence or mortality, T2D e , neurodegenerative disease, as well as all-cause mortality and cancer mortality among cancer survivors.

Statistical Analysis
In accordance with previous versions of this review, studies were grouped according to the different clinical outcomes (i.e., all-cause mortality, CVD incidence or mortality, cancer incidence or mortality, T2D, and neurodegenerative disease and all-cause mortality, or cancer mortality among cancer survivors). Extracted risk estimates (OR, RR, and HR) comparing the highest and lowest category of dietary indices were interpreted as RR and were pooled using a random-effects model with the DerSimonian-Laird method. 83 The weighting of each study was assigned by calculating the standard error of the log-transformed RR, interpreted as an estimated variance of the log-transformed RR. 83 Between-study heterogeneity was explored using Cochran's Q test and I 2 statistic. An I 2 statistic >50% was regarded as a substantial amount of heterogeneity. Moreover, 95% CI for I 2 was calculated with the heterogi command in STATA software. 84 Subgroup analyses were performed for dietary indices (HEI, AHEI, and DASH score) and their versions valid for the corresponding time period (HEI, HEI-2005, HEI-2010, HEI-2015, AHEI, AHEI-2010, and DASH-Fung score). 81 The DASH-Fung score is composed of 8 components (fruits, vegetables, low-fat dairy, whole grains, nuts/seeds/legumes, red and processed meat, sugarsweetened beverages, and sodium) scored by quantiles of intake. 81 For CVD and cancer, separate analyses were applied to compare mortality and incidence. Additional subgroup analyses were conducted for distinct CVD outcomes (nonfatal or fatal coronary heart disease, stroke, and heart failure), site-specific cancers, and neurodegenerative disease (cognitive impairment and Parkinson disease). To check the robustness of results, analyses were restricted separately to US studies, long-term follow-up (8 years or more), high-quality (Newcastle Ottawa Assessment Scale score 8 points or higher), and men or women. Moreover, pooled estimates were recalculated using the fixed-effect model.
For comparisons with more than 10 eligible studies available, the publication bias was explored with Egger's regression test and funnel plots. 85 All analyses were conducted in Review Manager (RevMan) 86 and STATA software. 84

Literature Search and Study Characteristics
Detailed steps of the database search and study selection are illustrated in Figure 1. The updated search revealed 50 new eligible reports,  which were assessed for overlapping with results from the 68 reports 14-81 included in 2 previous versions of this systematic review. Five reports 66,70,[134][135][136] were excluded at this step because they overlapped other included studies, resulting in a total of 113 reports included in the current update (47 additional reports not identified previously). [67][68][69][71][72][73][74][75][76][77][78][79][80][81] Characteristics of the 47 studies identified in the current update are presented in Table 2. Including the previous reports, analyses pooled data from 3,277,684 participants. Considering different clinical outcomes, all-cause mortality risk was assessed in 23 reports, CVD incidence or mortality in 45 reports, cancer incidence or mortality in 45 reports, T2D in 16 reports, neurodegenerative diseases in 12 reports, and allcause mortality and cancer mortality among cancer survivors in 12 reports (those included breast, colorectal, ovarian, and overall cancer, as well as multiple myeloma survivors).

Subgroup and Sensitivity Analysis
Stratification of analyses for dietary scores showed that all 3 scores were associated with a lower risk of all-cause mortality, CVD, cancer, and T2D in the general population, and all-cause mortality among cancer survivors. There was a trend towards (P ¼ 0.04) a greater reduction in T2D risk using the DASH score (RR 0.78, 95% CI 0.72 to 0.83, The original HEI, unlike the newer versions (HEI-2005 and 2010), was not found to be related to risk of all-cause mortality, T2D, and cancer. The HEI-2015 was inversely associated with all-cause mortality, CVD, and cancer, but not with T2D. The original version of AHEI was not associated with cancer, whereas its newer adaptation was associated with decreased risk. After limiting the analyses to studies using only the DASH-Fung score, the observed associations remained except for neurodegenerative diseases and all-cause mortality among survivors (Table 4; available at www.jandonline.org).
Estimates for men and women did not differ except for allcause mortality and cancer-specific mortality among cancer survivors, which were found to be nonsignificant in men (RR 0.95, 95% CI 0.86 to 1.04, I 2 ¼ 34% and RR 0.91, 95% CI 0.82 to 1.01, I 2 ¼ 12%, respectively) (Tables 5 and 6; available at www. jandonline.org). Findings from analyses conducted in USbased cohorts remained significant except for neurodegenerative diseases (RR 0.90 95% CI 0.80 to 1.01, I 2 ¼ 47%) ( Table 7; available at www.jandonline.org). Sensitivity analysis conducted by the inclusion of studies with long-term  Table 2. Summary characteristics of 47 prospective studies (identified in the current update of review) evaluating association between diet quality (assessed by HEI, a AHEI, b or DASH c score) and all-cause mortality, CVD d incidence or mortality, cancer incidence or mortality, T2D, e neurodegenerative disease, as well as all-cause mortality and cancer mortality among cancer survivors.                   . Forest plot showing pooled relative risks with 95% CI for association between highest vs lowest diet quality (assessed by HEI, AHEI, and DASH score) and risk of cardiovascular disease incidence or mortality in prospective cohort studies. a HEI ¼ Healthy Eating Index. b AHEI ¼ Alternate Healthy Eating Index. c DASH ¼ Dietary Approaches to Stop Hypertension. Table 3. Main and subgroup analyses for association between diet quality (assessed by HEI a , AHEI b or DASH c score) and all-cause mortality, CVD d incidence or mortality, cancer incidence or mortality, T2D e , neurodegenerative disease, as well as all-cause mortality and cancer mortality among cancer survivors.

Outcome
No. of reports Diet Quality Index/Score Relative risk 95% CI I 2 , % f (95% CI) P for subgroup differences g (continued on next page) Table 3. Main and subgroup analyses for association between diet quality (assessed by HEI a , AHEI b or DASH c score) and all-cause mortality, CVD d incidence or mortality, cancer incidence or mortality, T2D e , neurodegenerative disease, as well as all-cause mortality and cancer mortality among cancer survivors. (continued)

Outcome
No. of reports Diet Quality Index/Score Relative risk 95% CI I 2 , % f (95% CI) P for subgroup differences g g P for differences between HEI, AHEI and DASH score, unless other indicated. h P for differences between cardiovascular disease incidence and mortality. i Coronary heart disease, stroke and heart failure estimates included both fatal and non-fatal cases. j P for differences between coronary heart disease, stroke, and heart failure. k P for differences between cancer incidence and mortality. l Presented are only site-specific cancer with at least two studies per comparison. m P for differences between site-specific cancers. n NA¼not applicable. o P for differences between Parkinson's disease and cognitive impairment.
follow-up (median 8 years) or high-quality studies (Newcastle Ottawa Assessment Scale 8 points) did not change results from the main analyses (Tables 8 and 9; available at www.jandonline.org). Furthermore, the fixed-effect model suggested robustness of findings from primary analysis (Table 10; available at www.jandonline.org).

Publication Bias
Results of Egger's regression test indicated no evidence of publication bias for all-cause mortality (P ¼ 0.16), CVD incidence or mortality (P ¼ 0.72), cancer incidence or mortality (P ¼ 0.97), T2D (P ¼ 0.87), neurodegenerative diseases (P ¼ 0.34), and all-cause mortality among cancer survivors (P ¼ 0.13). However, some indication for publication bias was found for cancer mortality among cancer survivors (P ¼ 0.02).
Visual inspection of funnel plots revealed general symmetry for CVD incidence or mortality ( Figure 10; available at www. jandonline.org), cancer incidence and mortality ( Figure 11; available at www.jandonline.org), T2D ( Figure 12; available at www.jandonline.org), and neurodegenerative diseases ( Figure 13; available at www.jandonline.org). However, small asymmetry could be observed for all-cause mortality ( Figure 14; available at www.jandonline.org), all-cause mortality among cancer survivors ( Figure 15; available at www. jandonline.org), as well as cancer mortality among cancer survivors ( Figure 16; available at www.jandonline.org), indicating that risk of publication bias cannot be excluded.

Credibility of the Evidence
Considering overall high adherence to healthy dietary patterns (assessed by HEI, AHEI, and DASH score), the credibility of evidence for all 7 included outcomes was moderate according to the NutriGrade tool (Table 1). Judgments for HEI ranged from very low (neurodegenerative diseases) and low (all-cause mortality, cancer incidence or mortality, T2D, allcause and cancer mortality among cancer survivors) to moderate (CVD incidence or mortality). Regarding AHEI, the credibility of evidence was moderate, except for overall cancer and cancer survivor outcomes. Similar judgments were made for DASH score, except for cancer (moderate credibility of evidence) and neurodegenerative diseases (low credibility of evidence).

DISCUSSION
The current update of this systematic review and metaanalysis summarized data from 113 reports, including more than 3.2 million participants, on associations between diet quality assessed by HEI, AHEI, and DASH, and multiple chronic diseases outcomes. The highest vs lowest quality of diet described by all 3 dietary scores combined was associated with lower risk of all-cause mortality (20%), CVD incidence or mortality (20%), cancer incidence or mortality (14%), T2D (19%), and neurodegenerative diseases (18%) in the general population, as well as all-cause mortality (17%) and cancer mortality (18%) among cancer survivors.
In general, the results from the present update correspond with the main findings in the first update of this review published in 2018. 4 However, the updated literature search added new substantial evidence to prior analyses by identifying 47 prospective cohort studies that have not yet been considered. For the first time it was possible to observe an inverse association of all-cause mortality among cancer survivors for the AHEI and DASH scores, as well as cancer mortality among cancer survivors for the DASH score. Moreover, it was possible to meta-analyze incidence of cancer subtypes represented only by single studies in the previous review, such as hepatocellular, lung, and prostate cancers. 4 Analysis of dietary patterns is recognized as a more comprehensive approach compared with focusing on single nutrients or food. 137 As distinct dietary patterns show differences in their composition, it is important to clarify that meta-analysis compares similar constructs. The dietary indices included in this systematic review share common characteristics-high intake of food and nutrients considered as beneficial with simultaneous low intake of those detrimental for health. 137 Thus, HEI, AHEI, and DASH scores promote high consumption of fruits, vegetables, whole grains, and healthy fats with simultaneous low intake of solid fats, added sugar, and sodium. With respect to T2D, the trend toward a greater benefit from adhering to DASH score compared with HEI might be attributed to the fact that DASH, unlike HEI, penalizes intake of red and processed meat, which is linked with a higher risk of T2D. 8,138 Both the HEI and AHEI scores are updated regularly, considering the latest available evidence. The benefit of adhering to a diet might depend on chronological development of scores. 4 Thus, the original HEI was not associated with risk of all-cause mortality and cancer, probably due to lack of distinction between refined and unrefined grains. Considering the explicit statement of the 2015-2020 Dietary Guidelines for Americans to cut the intake of saturated fat and added sugar, the developers of the HEI-2015 included them as separate moderation components. 82 Advantages of adhering to high-quality diets can also vary depending on disease etiology. Analyses stratified for specific CVD outcomes suggested an inverse association of high diet quality with coronary heart disease, stroke, and heart failure. These findings are in line with recent meta-analyses, which identified healthy/prudent dietary patterns as beneficial in the prevention of stroke and heart failure. 139,140 Consistent with results from the 2018 update, there was an inverse association between highest diet quality adherence and colorectal cancer risk. Comparable associations for healthy dietary patterns were reported by the recent 2020 scientific report from the 2020 Dietary Guidelines Advisory Committee (DGAC) and a meta-analysis of cohort studies by Feng and colleagues. 141,142 Reduced risks of hepatocellular and prostate cancers in the highest diet quality category were also reported by another systematic review. 143 For neurodegenerative diseases, high diet quality was linked with a lower chance for cognitive decline, but not Parkinson disease. Whereas current findings for cognitive impairment correspond with a previous meta-analysis by Solfrizzi and colleagues, 144 a relationship between the quality of dietary patterns and Parkinson disease has not been established.
Addressing the credibility of the evidence is a crucial aspect of delivering evidence-based nutrition recommendations. Despite the clear advantage in inferring causality, the use of randomized trials in studying dietedisease relationships is limited by a lack of blinding, as well as difficulty ensuring proper adherence in a long-term follow-up. Therefore, prospective cohort studies remain an essential source of evidence in nutritional epidemiology. 5 The NutriGrade tool used in the current update was previously implemented to evaluate the credibility of evidence from dietedisease associations. [7][8][9][10] The judgments of this study suggested moderate credibility of evidence for the association between healthy dietary patterns reflected by the HEI, AHEI, and DASH score and multiple health outcomes (Table 1). Moreover, considering specific indices, moderate credibility of evidence was identified for associations between HEI, AHEI, and DASH for CVD, AHEI and DASH for all-cause mortality and T2D, and DASH for cancer, as well as AHEI for neurodegenerative diseases. The main reasons that credibility could not be judged as high include small effect sizes, low number of studies, indirectness caused by differences in specific scores estimates, and evidence for the presence of publication bias for some outcomes. In contrast to current findings, the DGAC in its 2020 scientific report suggested the presence of strong evidence for a relationship between dietary patterns and all cause-mortality, as well as confirmed findings from its previous 2015 report suggesting the presence of strong evidence for a relationship between dietary patterns and CVD. 141 Difference in judgment can be explained by the application of a different framework to assess the credibility of evidence. In that context, an advantage of the NutriGrade tool is the application of clear and transparent guidance on scoring. Furthermore, the DGAC statement pertaining to CVD involved supportive evidence from randomized controlled trials focusing on surrogate outcomes, such as blood pressure or blood lipids, whereas our judgment focused only on prospective studies with hard clinical end points. Moderate credibility of the evidence for cancer and T2D corresponded with findings from the 2020 DGAC scientific report. 141 In addition, moderate credibility for T2D corresponded with judgments from a recent umbrella review. 138 Moderate to high credibility of evidence is considered plausible to make a strong recommendation on particular health interventions. 145 In particular, consistent moderate credibility of evidence for both specific scores and overall high diet quality in the context of CVD suggests promotion of healthy diet as a part of disease prevention. Therefore, findings from this systematic review complement the recent 2020 DGAC report. 141

Strengths and Limitations
A major limitation of the current findings is the substantial amount of heterogeneity. 4 Potential sources of this heterogeneity include varying characteristics of cohorts as well as heterogeneous dietary scoring systems. All 3 dietary indices/scores were developed on the basis of data from the US population. More than one-third of the reports included in the current update of the systematic review provided data from European and Asian cohort studies. Both food availability and culinary tradition can vary greatly across countries and continents. 146 Thus, the use of predefined dietary patterns in their original form may not be appropriate in other cultural settings. Several studies had to redefine food classification and components of indices due to lack of intake data or food items infrequently consumed in specific communities. For example, one of the Chinese modifications of AHEI did not include intake of whole grains, sugar-sweetened beverages, and trans-fatty acids, 64 and the Swedish adaption of the DASH score did not include sodium restriction. 77 Therefore, participants categorized in 2 different studies as adhering to high-quality diets could potentially differ in the intake of healthful and less healthful foods and nutrients. Likewise, the majority of studies included in the present metaanalysis adopted the DASH score created by Fung,81 which uses scoring based on the quantile distribution of intake. This could lead to differences in cutoff points for components between studies. This approach was previously recognized as a potential disadvantage of other a prioriedefined diet quality indices, such as the Mediterranean diet score. 147 In addition to being an important source of heterogeneity, different scoring did not allow for conducting a doseeresponse meta-analysis, which downgraded the credibility of evidence for identified associations. Another limitation is the fact that the majority of included cohort studies used food frequency questionnaires as a dietary assessment method. The semi-quantitative nature of these data limits the ability to obtain an accurate estimate of intake. 148 Future studies should consider focusing on new assessment methods, such as multiple source method and biomarkers of intake. 149,150 A strength of multiple-source method is the ability to provide an unbiased estimate of individual usual intake by combining data from 24-hour dietary recalls and food frequency questionnaires. However, validated biomarkers of intake could provide an additional objective measure of long-term exposure to a specific food or nutrient.
Only a few studies included in this review collected longitudinal data on dietary intake. Individual food consumption can change substantially during long-term follow-up, leading to misclassification if only baseline data are available. 125 Future updates of this systematic review should address differences in the evidence generated from baseline and repeated measures to assess changes in intake. Finally, for the analyses of cancer survivor data, different cancer subtypes were pooled, which might be a limitation because of different responsiveness of distinct tumors to the diet. However, in a previous meta-analysis, scores from a prioriedetermined diet quality indices showed a consistent inverse association for different tumor subtypes. 151 A particular strength of this analysis is the large number of included studies (113 reports pooling data from 3,277,684 participants). The majority of included studies were highquality prospective cohort studies. The current update was conducted and reported according to a predefined protocol. Furthermore, the credibility of evidence was assessed using the NutriGrade tool, which has already been adopted in previous systematic reviews.

CONCLUSIONS
This updated systematic review and meta-analysis suggests that high diet quality (as assessed by HEI, AHEI, and DASH) is inversely associated with risk of all-cause mortality, CVD incidence or mortality, cancer incidence or mortality, T2D, and neurodegenerative diseases, as well as all-cause mortality and cancer mortality among cancer survivors. Moderate credibility of evidence for identified associations complements the recent 2020 DGAC report recommending healthy dietary patterns for disease prevention. 27