Health effects of drinking 100% juice: an umbrella review of systematic reviews with meta-analyses

INTRODUCTION

Low levels of fruit and vegetable consumption are a major modifiable risk factor for noncommunicable disease,¹ contributing to approximately 16 million disability-adjusted life-years and millions of deaths worldwide annually.² Globally, the mean intake of fruits and vegetables is approximately 290 g per day, significantly less than the 400 g recommended by the World Health Organization (WHO).³ The most recent Australian National Health Survey (2020–2021) revealed that only 6.1% of adults and 8.5% of children consumed the recommended amounts of both fruits and vegetables,⁴ with average daily servings of fruit and vegetables for adults at 1.6 and 2.6, respectively,⁴ compared with the recommended 2 servings of fruit and 5–7.5 servings of vegetables (depending on sex and life stage). As such, diets are potentially lacking in healthful dietary fiber, micronutrients, and bioactive compounds.^5–7 Barriers to fruit and vegetable consumption include food preferences, habits, access, spoilage, lack of preparation skills, convenience, and cost.^8,9 One hundred percent fruit and/or vegetable juices may circumvent these barriers; however, their place in dietary recommendations remains controversial.

While lower in fiber and higher in free sugars than whole fruits and vegetables, 100% juices contain key micronutrients of public health interest, including vitamin C, potassium, and vitamin A,¹⁰ and important bioactive compounds, such as flavonoids, carotenoids, and other compounds with antioxidant properties.^11–13 Nutritional properties vary depending on pulp contents, fruit/vegetable variety, ripeness, and processing.^14–16 Importantly, juicing can contribute to improved access, affordability, and palatability relative to whole fruits and vegetables, with no cooking skills or preparation required, longer shelf-life, and less complex storage and shipping requirements.^12,13 Average juice consumption, including 100% juice, in Australia is estimated at 50 mL per capita per day and 275 to 349 mL per capita per day when only juice consumers are considered.¹⁷ In Australia, 100% fruit juice consumption may increase the proportion of people meeting their daily recommended servings of fruit from 10% to 24%.¹⁷ It has been suggested that actively encouraging daily consumption of 100% juice in public health policies could be a practical and cost-effective way of helping populations achieve the recommended intakes for fruit and vegetables.¹² Data from the United States suggest that consumption of 100% fruit juice is associated with improved rates of nutrient adequacy¹⁸ and diet quality,^19,20 potentially through displacement of other energy-dense, nutrient-poor beverages, such as sugar-sweetened soda.^21,22 Notably, the energy density of 100% juice is similar to low-fat milk (∼1.8 kJ/g),²³ and in economic modeling of the Australian diet, juice had the highest nutrient density-to-cost ratio, indicating that it is a cost-effective source of nutrition.²⁴

Despite these potential benefits, the place of 100% juice in dietary guidelines and models of healthy eating remains contentious, and recommendations vary by jurisdiction. The 2020–2025 Dietary Guidelines for Americans include 100% juice as part of a healthy dietary pattern and list it as a recommended primary beverage, with the recommendation that at least half of fruit intake comes from whole fruit.²⁵ Other guidelines, including Australia,²² Spain,²⁶ the Netherlands,²⁶ and the United Kingdom,²⁷ include fruit juice as a core food but recommend limiting servings (eg, 1 serving of 125–150 mL per day). Others, including those in Canada²⁸ and France,²⁹ have excluded juice from healthy eating guides based on the perceived risks of free sugars, acid, and energy levels for weight and oral health, and unfavorable direct comparisons to whole fruits. The New Zealand Ministry of Health eating and activity guidelines include fruit juices as sugary drinks and recommend limiting intake (eg, once a week),³⁰ while the Brazilian guidelines recommend “unsweetened, minimally processed juices” and suggest limiting consumption of sweetened and ultra-processed juices.³¹ Recommendations, regardless of direction, center on fruit juices, with vegetable juices rarely mentioned.^{22,25,26,28,29,31} The WHO targets aim to reduce intakes of free sugars regardless of source, and therefore 100% juices are included in these limits.³² As such, it remains unclear how 100% fruit and vegetable juices should be treated in dietary recommendations.

The rationale surrounding the inclusion of 100% juices in dietary guidelines and models of healthy eating appears to be based on nutrient and bioactive content, which may reduce the risk of some noncommunicable diseases.^12,22,25,26 The rationale for the exclusion of juices appears to be based on free sugars and energy content increasing the risks of other noncommunicable diseases, as well as the loss of dietary fiber relative to the whole-fruit form.^{12,28,29,31,32} However, both approaches represent a reductionist approach to nutrition, as foods and beverages are complex and cannot be reduced to single features (ie, the nutritional composition of 100% juice is more than just free sugars or low in fiber). Similarly, human health outcomes are varied and complex, and considering relationships with individual diseases in isolation may result in conflicting recommendations, despite common health conditions often occurring comorbidly. It is therefore necessary to consider the balance of evidence regarding various 100% juices to properly inform decision making with regard to policy and public health recommendations.

A previous umbrella review of the relationship between juice and health focused only on fruit juice, without consideration of vegetable juices or restriction to 100% juice.³³ In addition, juice is often studied without the separation of sweetened juice (which contains added sugars) and fruit drinks (which only contain a small percentage of fruit juice) from 100% juice, which is undiluted and without added sugars. Therefore, we conducted an umbrella review of systematic literature reviews (SLRs) with meta-analyses (MAs) assessing the relationship between 100% fruit and/or vegetable juice consumption and human health outcomes to collate the highest-level evidence on the health effects of 100% fruit and vegetable juice.

METHODS

This umbrella review (also known as an overview review, meta-review, or review of reviews) summarized SLRs with MAs to enable quantitative conclusions regarding the strength of evidence and dosage effect to be made, and ensured that only health outcomes with a sufficient body of evidence were considered. Systematic literature reviews with MAs were defined as follows: (1) full-text reports that labeled themselves as an SLR or MA anywhere in the text or (2) full-text reports that met the definition of an SLR with MA where it adhered to a strict scientific design based on explicit, prespecified, and reproducible methods and had a specific statistical strategy for assembling the results of multiple studies into a single quantitative estimate.³⁴

This review was prospectively registered with the International Prospective Register of Systematic Reviews (PROSPERO, https://www.crd.york.ac.uk/prospero/; accessed December 1, 2022; registration no. CRD42022380588) and has been reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 checklist (see Table S1 in the Supporting Information online)³⁵ and the Preferred Reporting Items for Overviews of Reviews (PRIOR) checklist (see Table S2 in the Supporting Information online).³⁶

RESULTS

The systematic search strategy identified a total of 498 records, of which 18 SLRs with MAs were eligible for inclusion (Fig. 1). Records excluded after full-text review are listed with reasons in Table S5 (see the Supporting Information online). Three were excluded due to a high or very high degree of overlap of studies included in MAs (see Table S6 in the Supporting Information online), resulting in 15 SLRs with a total of 144 MAs included in this umbrella review. Of the extracted MAs, 51 were primary analyses and 93 were secondary subanalyses by juice type, dose, or duration of intervention.

DISCUSSION

This umbrella review has synthesized the diverse body of quantitative evidence on the health effects of 100% juice. Results from MAs were available for a broad range of doses (50–1200 mL/day, including several interventions using relatively high doses of >500 mL/day) and durations (hours to multiple years). Results from meta-analyses were available for several health outcomes of key relevance to population health, including mortality, CVD, metabolic health, body composition, cancer, and inflammation.

On balance, when multiple juice types, doses, durations of exposure, and health outcomes were considered, limited potential harms associated with 100% juice consumption were identified, with the majority of the available evidence (including MAs from both intervention and prospective cohort studies) being neutral (neither positive or negative). Improved cardiovascular and inflammatory health measures were seen in MAs of intervention studies and, for cardiovascular outcomes, effect sizes were of clinical relevance, while this is more difficult to assess for inflammatory markers. Unfavorable associations were found between 100% juice consumption and type 2 diabetes risk (RR = 1.07), CVD mortality (HR = 1.20), and prostate cancer risk (RR = 1.03), with the totality of this evidence from MAs of cohort studies. The mismatch between intervention and observational studies for CVD-related findings may reflect differences in time frames or confounding in observational trials. This may mean that the true effect size is larger or smaller, with the confidence level being low due to the small number of cohorts included in the meta-analyses. However, these were notably very small effect sizes, even for nutritional epidemiology where smaller RRs are more common than in other epidemiological fields.⁵⁸

Weak associations from cohort studies may be meaningful if supported by randomized controlled trials or laboratory studies.⁵⁸ Interestingly, higher 100% juice consumption was associated with an increased risk for diagnosis of type 2 diabetes and CVD in prospective cohort studies but was not linked to risk factors for these conditions in the meta-analyses of intervention studies. No associations were found for type 2 diabetes risk factors (eg, blood glucose levels and insulin resistance), with some protective effects found for CVD risk factors (eg, blood pressure levels). The polarity and diversity of findings regarding the relationship between 100% juice and health may be explained by differences in primary study methods, with the majority of favorable findings being reported in intervention studies and unfavorable findings being only from prospective cohort studies. Of the 12 included primary MAs of prospective cohort studies, 3 MAs (25%) showed risk and 1 MA (8%) showed benefit, whereas 8 MAs (66%) showed null effects. The single MA that combined cohort and case-control studies showed no associations. Of the 38 primary MAs that included intervention studies, 9 (23%) showed benefit, none showed risks, and the remainder (77%) showed neutral effects, including studies of relatively high daily intakes of more than 500 mL per day.

While prospective cohort studies are the highest level of primary evidence for disease etiology,⁵⁹ data-collection methods, such as food-frequency questionnaires and food recalls, may not adequately measure intake of 100% juices, as juice is a beverage product with significant diversity, and these methods, not designed to capture intakes of specific foods or beverages, are more likely to be used in cohort studies. It is possible that the “juices” captured by these dietary data-collection methods are not solely 100% juice, including fruit drinks or sweetened juices, despite being reported as 100% juice.^60,61 Intervention studies, while only able to consider relatively short-term health outcomes, have the benefit of 100% juice intake being tightly controlled and thus more reliable. It is also possible that the favorable results seen in intervention trials are more pronounced due to the inclusion of participants with a background of low fruit and/or vegetable consumption. The complex and multifactorial nature of chronic disease risk (eg, genetics and additional lifestyle risk factors) may additionally help explain the differences in findings between complex outcomes (as measured in prospective cohort studies) and simple outcomes (as measured in intervention trials). However, it is important to note that, while intervention studies are able to be randomized, they are restricted to short-term outcomes, and may not adequately capture changes or impacts of modulating factors seen over the longer term. Conversely, prospective cohort studies are able to represent long-term outcomes in large cohorts but do not readily allow for precision analyses regarding individual foods or beverages consumed, subpopulations based on genetics, or other potentially modifying factors.

The positive impact of 100% juice consumption on specific health outcomes relevant to population health may be explained by multiple potential mechanisms, including the vitamin, mineral, and bioactive compound content of 100% juice. Data from epidemiological and animal studies show that high potassium intake can reduce blood pressure, arterial stiffness, and associated vascular and organ damage,^49,62–64 with 100% fruit and/or vegetable juice being an affordable, accessible, and palatable source of potassium.^13,65 Furthermore, there is significant evidence linking color-associated compounds to health benefits, and juices are readily available in a variety of colors with bioactive pigments maintained.⁷ Many bioactive compounds, including polyphenols such as hesperidin and naringenin, are highly bioavailable in some juices and may contribute to beneficial cardiovascular and inflammatory effects via modulation of nitric oxide bioavailability and related pathways,^7,49,66,67 with polyphenol intake linked to improved blood pressure outcomes and higher plasma nitric oxide levels.^65,66 The antioxidant activity of bioactive compounds (such as flavonoids, anthocyanins, and carotenoids) and vitamins (such as vitamins C and A) may also contribute to improved cardiovascular and inflammatory outcomes.^49,67 Juices containing folate may benefit health via reduction in homocysteine levels, high levels of which can damage blood vessels.⁶⁸ However, all of these features can vary by juice type, and differing juices included in studies may impact results.

Oral health concerns are commonly cited as reasons to recommend restricting or excluding juice intake.^{22,26,28,29,31} However, no MAs on the relationship between dental caries and other oral health outcomes were eligible for this review when the 100% juice inclusion criteria were applied. This suggests that previous reviews (eg,²¹) may have included data for sweetened juices or other non-100% juices, making the findings not specific to 100% juices. However, a recent SLR without MA on 100% fruit juice and dental health found that the existing data of 5 independent prospective cohort studies and 9 randomized controlled trials was inconclusive regarding the risks, with the randomized controlled trial data demonstrating harm using intra-oral devices lacking normal plaque or saliva action and conditions not representative of normal juice consumption.⁶⁹ Acidic beverages are often associated with tooth erosion⁷⁰; however, it has been argued that assigning risk to dietary factors in this context is inappropriate due to the complexity of behavioral factors involved in oral hygiene and risk for tooth erosion and caries⁷¹ and the modifying effects of toothpastes.^72–74 As such, SLRs with MA investigating the relationships between oral health outcomes and 100% juice consumption are still needed.

One hundred percent juice is well known to be lower in dietary fiber relative to whole fruit,⁷⁵ which, as an isolated comparison, suggests that 100% juice would adversely impact satiety and therefore energy intake and weight management, and this is regularly cited as a reason to recommend limiting intake.^{22,26,28,29,31} Satiety was not included in the present review, as it is not a health outcome directly; however, included intervention studies for adults on body weight, BMI, and waste circumference (with 10–20 interventions per MA) did not find any relationship between 100% juice consumption and body-composition measures, noting that doses ranged from 120 to 1000 mL/day for between 1 and 16 weeks. It is important to note that these included relatively small numbers of participants with MA pools of 411–975 participants, depending on the metric. Where subanalyses were available by juice type, this led to a separation of duration and dose as a consequence, and even the high doses and long durations (eg, 1000 mL/day over 6 wk or 480 mL/day over 12–16 wk) still had no significant associations. Studies in children were cohort studies based on BMI z score and were considerably larger. Importantly, consideration of fiber content alone on satiety does not consider the context of 100% juice in the daily diet, nor the contribution of other features in 100% juice. For example, an intervention study on the timing of juice consumption found weight gain when juice (20% of daily energy requirement) was consumed between meals, but weight loss when the same amount was consumed with meals.⁷⁵ In a study of grape juice, there was no weight gain seen over 12 weeks in those consuming the juice (compared with a no-beverage control); however, there was weight gain in those consuming an energy matched, polyphenol-free beverage,⁷⁶ suggesting that polyphenols impact satiety or energy metabolism.

While this umbrella review was conducted following best-practice protocols and provides a broad overview of quantitative analysis from existing MAs, several limitations need to be considered when interpreting the findings. The protocol was designed to include only 100% juices. However, during review and screening, it was found that many SLRs that purported to include only 100% juice, in fact included primary studies where this restriction was not applied, leading to the inclusion of fruit drinks and sweetened and/or fortified juices. While all efforts were made to ensure the inclusion of only data from studies of 100% juices, it is possible that differences in definitions, descriptions, and interpretations may impact the inclusion of studies in this synthesis. Studies did not include sufficient descriptions to separate juices with pulp from filtered juices.

The application of GRADE revealed many of the included MAs to be of very low or low-quality level. This suggests a need for studies with high-quality design and reporting to better inform decision making. However, a large body of evidence scoring low quality on GRADE is not unusual in nutrition research as the tool relies heavily on the clinical intervention paradigm.⁷⁷ Findings may be limited by including only SLRs with MA; while this is justified based on MAs providing quantitative synthesis and demonstrating a sufficient body of evidence to allow for MA to be conducted, additional juices (such as vegetable juices) and additional health outcomes may be captured in SLRs without MA. This suggests that additional research, including more primary studies and more SLRs with MA, particularly those with dose–response analyses, are required to broaden the umbrella able to be reviewed in the future.

Importantly, this review sought to investigate both 100% fruit and vegetable juices. However, no MAs on 100% vegetable juice that met the inclusion criteria were available. While fruit juices are more commonly produced and consumed, vegetable juices also have beneficial micronutrients and bioactive compounds, and vegetable consumption is even lower than fruit consumption at a population level. As such, it is important to consider the potential for vegetable juices to be incorporated into models of healthy eating to address this gap. Future studies should investigate this possibility. The types of juices included in primary MAs and subanalyses may also need to be considered. Juices are often used in research studies as a delivery method for fruit nutrients and bioactive compounds to address research questions surrounding the fruit and its properties while extending shelf-life, enhancing palatability, and reducing variability in the controlled administration. However, this leads to 2 potential limitations: (1) an imbalance in the types of 100% fruit juices studied, with the predominant juice types (eg, pomegranate, grape, and cherry juices) not necessarily reflecting the most commonly consumed juices in the broader population, and (2) a potential merging of commercial, processed juices with juices prepared fresh or under small-batch conditions, leading to potential variance in pulp (and therefore dietary fiber) content and other nutrient contents due to heat/pressure treatment and storage differences. Meta-analyses can also combine studies of different varieties (eg, different types of oranges used to make orange juice), meaning results may be variable and not apply to all 100% juice types.

CONCLUSION

In conclusion, the balance of evidence does not justify recommendations that exclude or limit 100% juice intake in models of healthy eating and dietary guidelines; however, caution may be warranted for specific subgroups or individuals, depending on risk factors and circumstances. Results include a broad range of doses and durations that have been studied using randomized controlled trials, including several intervention doses that are relatively high (500–1200 mL/day). As such, dietary guidelines for restriction or avoidance of 100% juice are not supported by the current available evidence and may have unintended consequences on population nutrient and bioactive intakes. However, additional well-designed studies are needed focusing on key health outcomes, commonly consumed 100% juice types, as well as 100% vegetable juices. Furthermore, future intervention studies need to be designed to consider actual dietary intakes, and the displacement effect of added juices and cohort studies need to be better designed to separate 100% juices from other juices and fruit drinks.

Importantly, the multiple potential beneficial effects detected at low to moderate levels (50–240 mL/day) of 100% juice consumption can provide exposure to beneficial nutrients and bioactive compounds, without the risks associated with excess consumption of free sugars and calories.⁴⁹ At these low and moderate intake levels, 100% juice can be incorporated into a healthy, balanced diet. When multiple key health outcomes are considered, the balance of evidence suggests that views of 100% juice products formed only on single-nutrient features may be oversimplified and not evidence based, and the inclusion of 100% juice as a core food may be warranted when the potential benefits of the whole beverage are considered. The evidence presented here reflects potential health benefits for 100% fruit juice, regardless of its properties relative to whole fruits; however, the body of evidence is not yet conclusive.

Supplementary Material

Supporting Information

The following Supporting Information is available through the online version of this article at the publisher’s website.

Table S1  PRISMA checklist

Table S2  PRIOR checklist

Table S3  Search strategy

Table S4  Interrater reliability between reviewers at full-text review

Table S5  Studies excluded after full-text review

Table S6  SLRs excluded due to overlap

Table S7  ROBIS summary

Table S8  GRADE

Table S9  Results from primary MAs of 100% fruit juice interventions for metabolic health

Table S10  Results from primary MAs of 100% fruit juice interventions for body composition

Table S11  Results from primary MAs of 100% fruit juice interventions for liver function