Introduction

Dairy products have constituted a significant component of the human diet for ~8000 years and are characterized as a distinct food group in many food-based dietary guidelines worldwide [1, 2]. These guidelines acknowledge non-fortified dairy foods as a source of high-quality protein, minerals including calcium, phosphorus, magnesium, selenium, zinc, iodine, and vitamins A, B1, B2, and B12 [2, 3]. Most dietary guidelines recommend a daily intake of 2–3 dairy servings, including low-fat milk, yogurt, and cheese, to fulfill nutrient requirements and enhance dietary quality [4, 5]. Despite these recommendations, research consistently indicates that most individuals fall short of the suggested daily intake [6, 7].

In recent years, there has been a growing interest in exploring the associations between dairy consumption and health outcomes. Systematic reviews indicate that dairy is associated with a reduced risk of various noncommunicable diseases [8]. Despite evidence that dairy intake offers several health benefits, some dairy products are high in saturated fatty acids which have been linked to adverse health effects and are therefore limited in many dietary guidelines [9].

Although less frequently highlighted in food-based dietary guideline messaging, it has been suggested that fermented dairy products such as yogurt and kefir provide health benefits that extend beyond their nutritive components. For instance, these products contain probiotics which may positively influence oral health, gut health, and overall immune function [10,11,12,13]. Considering the cultural significance of dairy in our diet and the increasing prevalence of noncommunicable diseases, it is essential to continuously evaluate the entirety of data regarding the impact of dairy consumption on various health-related outcomes.

Scoping reviews of systematic reviews map the existing evidence and identify research gaps on a broad topic. While they appear similar, umbrella reviews and scoping reviews of systematic reviews are different in scope and quality appraisal. Scoping reviews are broader and more exploratory than umbrella reviews and are intended to provide a descriptive overview rather than a synthesis of results. The objective of this scoping review was to summarize the research landscape on associations between dairy consumption and noncommunicable disease outcomes in adults. Specifically, we investigated cardiovascular health outcomes, various cancer types, body composition, mortality, and other health outcomes including type 2 diabetes mellitus (T2DM), bone and joint health, and cognitive outcomes. Although there are numerous systematic reviews (SR) on the topic, we are unaware of any scoping review that investigates all types of dairy and its effects on a range of health outcomes.

Methods

Protocol and registration

The protocol for this scoping review was established before the beginning of the study and was registered on February 19, 2024, on Open Science Framework (Available at: https://doi.org/10.17605/OSF.IO/YQWVK) [14]. We have followed the PRISMA extension for scoping reviews in composing this review [15].

Search strategy

A systematic literature search was performed on February 19, 2024, in three databases: Ovid Medline, Ovid Embase, and Web of Science. The search strategy was developed for Medline by an experienced information specialist and adapted for Embase and Web of Science (SA). It comprised two main themes: dairy intake in the adult population and health outcomes, and was restricted to systematic reviews in English, German, French, and Spanish. The full details of the search strategy are available in the supplemental file.

In addition to the articles identified through the database searches, all available reports from the World Cancer Research Fund (WCRF) [16] that addressed dairy consumption were included in the scoping review.

Selection of studies

Inclusion and exclusion criteria were created following the Population, Concept, Context framework proposed by the Joanna Briggs Institute (JBI) [17].

Inclusion criteria

SR or meta-analyses (MA) fulfilling the following criteria were included in the scoping review: i) publication date between January 1st, 2014 and February 19, 2024, ii) studies based on randomized controlled trials (RCTs), prospective cohort studies, case-control studies, and cross-sectional studies iii) studies involving primarily adults as study population, iv) studies focusing on exposure to cow’s milk and/or other dairy product consumption in any amount, v) publications reporting any of the following outcomes were included: cardiovascular outcomes (cardiovascular disease (CVD), coronary heart disease (CHD), stroke, and hypertension), cancer types (bladder, breast, colorectal, corpus uteri, esophagus, leukemia, liver, lung, non-Hodgkin lymphoma, oral, ovarian, pancreatic, prostate, stomach/gastric, “combined cancer” as defined by individual reports), body composition (weight gain, overweight, and obesity), all-cause mortality and cardiovascular mortality, and other outcomes (T2DM, bone and joint health, cognitive outcomes). Outcomes were chosen on the basis of three factors: I) establishment as a diet-influenced non-communicable disease or mortality; II) for cancer outcomes, selection of the 20 cancer types with the highest global incidence based on the WCRF [16]; and III) existence of at least one systematic review and/or meta-analysis on the outcome that also meets all other inclusion criteria. All papers—systematic reviews, meta-analyses, and WCRF reviews will henceforth be referred to as “reports”.

Exclusion criteria

Reports were excluded if they met one or more of the following criteria: I) umbrella review, scoping review, and/or systematic reviews or meta-analyses based on in-vitro/animal experiments, II) reports involving infants, children, and adolescents as participants (exception: exposure started in childhood but outcome was assessed in adulthood), III) reports including non-bovine dairy products (sheep, goats, buffaloes, camels, humans, plant-based, formula products), nutritional supplements and dairy products that are explicitly fortified with calcium and/or other nutrients. IV) reports that looked exclusively at risk factors for the disease outcomes of interest.

Selection process of sources of evidence

All identified reports were uploaded to Zotero 6.0.36 [18] in separate libraries and then imported into the basic version of Rayyan systematic review software (https://www.rayyan.ai/) for the deletion of duplicates, title, and abstract screening [19]. Rayyan’s duplicate tool identified suspected duplicates which were then manually resolved. Title and abstract screening were independently conducted by two reviewers (SA, SW) to exclude references that did not meet the inclusion criteria. Full-text publications of all potentially relevant records were obtained and uploaded to Zotero. At the full text level, reasons for exclusion were recorded. Two reviewers independently evaluated half of the full-text articles. A minimum of 10% of the articles were cross-checked. Discrepancies between reviewers were addressed through discussion; a third reviewer (DM) was consulted to resolve remaining questions.

Data extraction

An extraction template was pretested on a subset of five reports. Elicit software was utilized to systematically extract relevant data points from the study texts [20]. Extracted data was subsequently examined in detail by two reviewers, who enhanced and corrected the information. The following data were extracted: author, publication year, design and number of included studies, participant counts, study regions, intervention/exposure, comparison parameter, included health outcomes, direction of effect or association, databases searched, search date range, inclusion of sex stratification. The full details of the extraction tables are available in Supplemental Table A. WCRF Continuous Update Project reports present findings in a standardized way. Their criteria for grading evidence for cancer prevention include convincing (strong evidence), probable (strong evidence), limited—suggestive, limited—no conclusion, and substantial effect on risk unlikely (strong evidence). Further information about WCRF grading evidence criteria can be found in the appendices of WCRF reports [21]. For this study, we charted WCRF reports as follows: convincing and strong evidence were charted as stated (decreases risk/increases risk), and limited evidence was charted as neutral. The WCRF extraction table can be found in Supplemental Table B.

Data categorization and presentation

Table 1 was constructed to present associations between dairy products and all included health outcomes. The numbers in each cell represent the count of SR and/or MA that reported either evidence of a decreased risk, no association, increased risk, or inconclusive evidence (different types of analyses within the report yielded different findings) of the respective dairy product on each health outcome. Several included reports examined multiple health outcomes and/or exposures, resulting in some reports being referenced in multiple cells. Therefore, the total number of “reduced risk,” “no association,” “increased risk,” and “inconclusive” associations may be higher than the number of included reports. Table 2 was created to present subgroup results on the associations of full-fat versus reduced-fat dairy with health outcomes included in some reports.

Results

The literature search yielded 1630 results: 379 from MEDLINE, 716 from EMBASE, and 535 from Web of Science. In addition, ten WCRF evidence synthesis reports, which can be categorized as systematic reviews, were included [22]. Two hundred and fifteen records were assessed as full texts, and a total of 95 reports were included in this scoping review, of which 92 contained meta-analyses. The title, abstract, and full-text phases, along with reasons for exclusion, are presented in a PRISMA flowchart (Fig. 1) [23] A list of excluded reports with reasons for exclusion can be found in Supplemental Table C.

Fig. 1

Flow diagram of the study selection.

Characteristics of included reports

The largest portion of included reports incorporated more than one study design (n = 43), followed by reports that included solely prospective cohort studies (n = 35). Seven reports examined follow-up data from RCTs, two were based exclusively on RCTs and four only included case–control studies (Supplementary Table A). Ten SRs were from the WCRF’s Continuous Update Project and include cohort studies, case–control studies and RCTs. The number of studies analyzed in the included reports ranged from five [24] to 152 [25], whereas participant numbers varied from 1427 [26] to 7,606,009 [27]. WCRF reports did not consistently provide the number of participants per included study in the report body, particularly for individual exposures/interventions such as dairy.

Health outcomes

Twenty-nine diet-related health outcomes were identified in the included reports. They fell into five groups:

1. I.

cardiovascular outcomes including composite cardiovascular disease [27,28,29,30,31,32,33,34], coronary heart disease [9, 29,30,31,32,33, 35,36,37,38,39], all types of stroke [9, 29,30,31, 33,34,35,36,37, 39, 40], ischemic stroke [38], and hypertension [9, 17, 41,42,43],

1. II.

cancer types including bladder [28, 44,45,46,47,48], breast [49,50,51,52,53,54,55,56], colorectal [28, 52, 57,58,59,60,61,62,63,64], corpus uteri [65, 66], esophagus [28, 67, 68], kidney [69], leukemia [70], liver [71,72,73], lung [74, 75], non-Hodgkin lymphoma [76, 77], oral [78,79,80], ovarian [81,82,83], pancreatic [24], prostate [52, 84, 85], stomach/gastric [86,87,88,89,90], and “all types” [28],

1. III.

body composition including overweight and obesity [41, 91], and weight gain [25, 91,92,93],

1. IV.

mortality including all-cause mortality [32, 35, 94,95,96,97,98,99,100] and cardiovascular mortality [94,95,96, 100, 101],

1. V.

other outcomes including T2DM [34, 41, 102,103,104,105,106,[107](https://www.nature.com/articles/s41430-025-01639-5#ref-CR107