By Hugo Francisco de SouzaMay 22 2024Reviewed by Susha Cheriyedath, M.Sc.

In a recent review and meta-analysis published in the journal Current Developments in Nutrition, researchers collate and discuss the results of a host of substitution-model-based epidemiological studies aimed at investigating the long-term dietary associations of different food items (mainly dairy products) and non-communicable diseases (NCDs) including all-cause mortality, type 2 diabetes (T2D), and cardiovascular disease (CVD). Their literature search through three online scientific repositories identified 2,544 publications, 34 of which met inclusion criteria and were considered for the meta-analysis.

Substitution of Dairy Products and Risk of Death and Cardiometabolic Diseases: A Systematic Review and Meta-Analysis of Prospective Studies. Image Credit: DUSAN ZIDAR / Shutterstock

After accounting for intra-study biases and specificity and correcting for demographic and medical history confounds, study findings reveal that different dairy items do not statistically alter NCD risk in included study participants. However, substituting dairy items with red meats and processed meats was associated with an increase in NCD risk, while replacing the items with whole grains and olive oils was found to improve long-term NCD outcomes. This study debunks hypotheses regarding ‘healthier dairy products’ while still elucidating the role of dietary regimes in NCD risk alterations.

The association between NCDs and dietary adherence

Chronic, usually age-associated non-communicable diseases (NCDs) represent the number one human killer in the world today, with records reporting ~73% of all deaths in 2017 belonging to this class. Alarmingly, despite substantial advances in medicine, the prevalence of NCDs continues to rise, with recent research suggesting the role of health behaviors (sleep patterns, physical activity levels, and dietary adherence) in NCD risk and progression.

Diet, in particular, has been pinpointed as an easily modifiable potential health risk, with the Global Burden of Disease study suggesting that a suboptimal diet may be the primary determinant of 22% of all global human deaths. Debates into the perceived benefits of ‘healthy’ low-fat dairy items over their ‘unhealthy’ dairy counterparts borrow from evidence from the NCD-associated risks of low-fat milk consumption, one of the 15 dietary risk factors characterizing suboptimal diets. Unfortunately, the relative NCD contributions of different diary (and food) items have never been formally synthesized, with most prior studies focusing on one of a few item comparisons.

About the study

In the present review and meta-analysis, researchers collate and synthesize evidence for more than 30 publications from three online scientific repositories to elucidate NCD risk alterations and general health outcomes accompanying substituting dairy items with 1. Other dairy products, 2. Plant-based foods, or 3. Other animal-derived food products. The review methodology ascribed to the guidelines listed in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist, refers to the PRISMA Statement for Reporting Literature Searches in Systematic Reviews (PRISMA-S) for its validation and has been registered with the International prospective register of systematic reviews (PROSPERO).

The publication dataset was obtained via sequential repository search, title and abstract screening, and full-text screening from three scientific publication databases, MEDLINE, Embase, and Web of Science, from database inspection until 28 June 2023. Prospective publications were included in the review and meta-analysis if they met the exhaustive list of inclusion criteria, such as population type (adult), exposure and comparator (studies specifying substitution food items and models), outcomes (all-cause mortality, CVD, T2D, or their comorbidities), and study design (prospective observational studies). Two reviewers extracted Data from included studies manually and independently and then compared notes to solve any data inclusion conflicts.

Extracted data included study-specific (publication title,  author, publication date, and methodologies employed), demographic, and medical. The Risk Of Bias In Non-randomized Studies - of Exposure (ROBINS-E) tool was used to assess and account for differential risk between included studies and weight statistical models accordingly. The statistical evaluation summarized the relative risks (SRRs) for each substitution interaction across studies.

Study findings

Of the 2,544 publications identified from the database keyword search, 34 publications (representing 15 unique participant cohorts) were observed to meet review inclusion criteria and were included in the review. Of these, 25 were included in the meta-analysis following bias risk assessments.

Of the included studies, 18 only compared the substitution of a single diary item, while the remaining 16 compared two, or rarely, more food items. Alarmingly, only one study was found to be at low risk of bias, with 31 identified as being ‘moderate-risk’ and three as ‘high risk,’ thereby suggesting the need for more prospective studies and clinical trials to use stranded sampling methodologies to improve and facilitate posthoc investigations.

Study findings highlight the surprising lack of change in NCD risk between ‘healthy’ (e.g., low-fat milk) and ‘unhealthy’ (e.g., high total-fat butter), suggesting that, within the sphere of diary product adherence, the quantity of consumption matters more than the specific food item. In contrast, substituting butter with olive oil significantly improved SRR scores, suggesting long-term NCD-aversion benefits when substituting a dairy-based food item with a healthier plant-based one. At the opposite end of the spectrum, substituting dairy items with red meats or processed animal-derived foods substantially increased NCD risk.

Conclusions

In the present study, researchers screen more than 2,500 publications and review more than 30 to elucidate the benefits (or lack thereof) of substituting dairy food items with other dairy products or another plant- or animal-derived food items, particularly in NCD-associated risk metrics such as all-cause mortality, T2D, and CVD. While initially intended to include evaluations of cancer risk, logistical issues prevented the latter. Cancer risk evaluations notwithstanding, the current meta-analysis clarifies that all dairy products carry the same inherent NCD risk, and within-group (dairy) substitutions may not hold long-term benefits for public health.

Substituting milk-based dairy products for plant-based or animal-based alternatives, however, was associated with changes in NCD risk, with healthy plant-based foods (such as olive oil in place of butter) improving risk outcomes, while red and processed meats worsened observations.

Journal reference:

Kiesswetter, E., Neuenschwander, M., Stadelmaier, J., Szczerba, E., Hofacker, L., Sedlmaier, K., Kussmann, M., Roeger, C., Hauner, H., Schlesinger, S., & Schwingshackl, L. (2024). Substitution of Dairy Products and Risk of Death and Cardiometabolic Diseases: A Systematic Review and Meta-Analysis of Prospective Studies. In Current Developments in Nutrition (Vol. 8, Issue 5, p. 102159). Elsevier BV, DOI – 10.1016/j.cdnut.2024.102159,  https://www.sciencedirect.com/science/article/pii/S2475299124000933