In 2018, the global consumer spending on organic foods was estimated at around €80 billion ($95 billion USD). While consumer demand may exist for myriad reasons, within the wider discourse about nutrition and health it is common to hear claims about the superiority of organic food over non-organic food, including (but not limited to):
- Better nutritional profile (higher levels of nutrients and/or greater bioavailability vs. non-organic foods);
- Absence of synthetic herbicides and pesticides (tied to assertions that such compounds are harmful to humans and promote disease, e.g., cancer);
- Better for the environment;
- Greater benefit to human health outcomes.
Consumer behaviour research has identified a number of factors correlating with an overall organic food purchase preferences, in particular health consciousness, environmental considerations, and an assumption of enhanced well-being. It may be said that organic food purchase is a value system and status symbol. However, the price differential between organic and non-organic food can be substantial, and in populations where only small proportions of the total populace meet recommendations of vegetable and fruit intake, there are considerations for food purchase beyond the production method. In this respect, an evidence-based evaluation of the common claims made in relation to organic foods is necessary for an informed choice, which must also factor in pragmatic considerations regarding social, economic, and environmental determinants of choice.
In this Sigma Statement, we will discuss the definitions of organic food, and review the evidence in relation to organic vs. non-organic food under three main areas:
- Organic Food and Nutritional Quality
- Organic Food and Environmental Pollutants
- Organic Food and Health Outcomes
Defining 'Organic' Food vs. 'Conventional' Food
In the European Union, organic food is regulated under Council Regulations No.834/2007 and No.899/2008 on organic production and labelling of organic food products. In the US, organic food is regulated by the Department of Agriculture. We do not propose to review and contrast the respective merits of the different regulatory systems, however, where relevant specific mention will be made to important concepts like the 'maximum residue levels' (MRL) for pesticides in food products (discussed further under the section on Organic Food and Environmental Pollutants, below).
While people think of 'organic food', the term 'organic' in fact refers to the method of production of that food. A food may be labelled organic under EU law and by the US Department of Agriculture (USDA) if 95% of its agricultural ingredients are organic. To be certified as organic under both regulatory systems, minimum time periods are set for which soil must be free of substances prohibited under organic production (e.g., certain synthetic pesticides). Organic regulations in both the EU and US have strict requirements with regard to livestock and animal produce, both in relation to living conditions (i.e., space, air, light, etc.) and in relation to feed (which must itself be organic) and veterinary treatments.
It is important to distinguish the term 'organic' from the term 'natural', which has no legally binding definition but does have certain regulations in place in relation to additives, like flavourings (i.e., 'natural flavourings', which is defined by Council Regulation No.1334/2008). However, for the consumer it is important to note that for the vast majority of food products, 'natural' does not mean anything with regard to method of production. 'Conventional' foods are foods produced through means of production that are outside the definition of organic, but that still meet standard legal requirements for food safety.
Thus, the regulatory systems have strict definitions for what food produce can be classified as 'organic' vs. 'conventional', and these distinctions refer specifically to regulations on methods of production. However, the system is not watertight: in the famous 'Gatto con gli stivali' fraud case, between 2007-2011 food products produced in Italy and Romania were apparently certified as organic. However, the certificates had been falsified, as had production documents to control bodies, resulting in the sale of ~703,000 tons of falsely-labelled conventional products sold as organic, corresponding to estimated financial turnover of around €200-million. Regulations have been significantly updated in the interim, however, the case serves as an example of the growing consumer market for organic-labelled foods.
Organic Food and Nutritional Quality
For the purposes of this section, under the concept of 'quality' we will focus primarily on nutritional content, i.e., actual levels of dietary micronutrients and bioactive food components, i.e., polyphenols, between organic and conventional foods. Aspects relating to food safety, i.e., pesticides, heavy metals, etc., will be addressed separately in a subsequent section.
The assumption of superior nutritional quality is one of the most common claims in relation to organic vs. conventional food. However, is this actually supported by the totality of evidence? Let’s look at both plant and animal produce.
Nutrient Quality - Plant Produce
Dangour et al. conducted a systematic review of studies which compared, through chemical analysis, the nutrient content of foods produced under organic or conventional production methods. From a total of 55 included studies, there was no difference in the nutritional content of crop-produce between organic and conventional methods for 10 of 13 categories analysed. The three for which there were differences - nitrogen, phosphorus, and titratable acidity - likely reflected the difference in fertiliser use and ripeness of the product at time of harvest. For nutrients like vitamin C, magnesium, calcium, potassium, zinc, copper, and phenolic compounds, there were no differences between organic and conventional foods in the included studies.
A further systemic review and meta-analysis in 2012 by Smith-Spangler et al. reported similar findings, with phosphorus and total polyphenol content higher in organic produce, but no difference in other nutrients between organic and conventional production. In particular, there were no significant differences in vitamin C, B-carotene, a-tocopherol, calcium, magnesium, or iron. Conversely, Brandt et al. conducted a systematic review of papers reporting on the vitamin and phenolic compounds in fruits and vegetables, comparing organic to conventional production. They found that the content of vitamin C, and 'secondary plant metabolites' (a definition that encompasses polyphenols, total phenolics, and other non-nutritive bioactive food components), were significantly greater in organic foods, while there was no difference in carotenes, tocopherols, or anthocyanins (a subclass of flavonoids). Specifically, antioxidant 'secondary plant metabolites' were found to be 12% higher in organically grown fruit and vegetables.
The most recent comprehensive systematic review and meta-analysis by Barański et al. included 343 total studies, of which 156 were included in a weighted meta-analysis (i.e., had data available on standard deviations and errors to allow the researchers to provide a greater 'weight' [a percent score out of 100%] to more accurate studies). Antioxidant activity was an average of 17% higher in organic crops. Further, polyphenolic compounds - including total flavonoids and other phenolic compounds - were significantly higher, with a range of 18-69% higher concentrations, depending on the specific compound. Smaller differences were noted with certain vitamins, in particular vitamin C and total carotenoids were higher in organically grown crops. Conversely, levels of proteins and amino acids, fibre, and vitamin E were all lower in organic crops compared to conventional crops.
Taken as a whole, the evidence from the above studies (which collectively synthesised a voluminous body of data) suggest that, compared to conventionally grown crops, organic crops may contain:
- Higher levels of antioxidant compounds (consistently shown)
- Higher levels of vitamin C (inconsistently shown)
- Potentially higher levels of carotenoids (inconsistently shown)
Interestingly, an analysis of broccoli samples obtained from commercial supermarkets demonstrated differences in the nutrient content of vitamin C related to season of harvest, but no difference between organic and conventional production sources. Consequently, much of the focus of the discussion within the literature is on the potential biological relevance of greater intakes of antioxidant polyphenolic compounds for human health outcomes. This will be assessed under the section Organic Food and Health Outcomes, below.
Nutrient Quality - Animal Produce
The previous analyses were primarily focused on the differences in crop produce, i.e., foods of plant origin. However, a number of studies have separately investigated the difference in animal produce between conventional and organic farming methods.
Średnicka-Tober et al. conducted a systematic review and meta-analysis of cross-European milk survey studies, comparing organic to conventional milk and dairy products. They found that organic milk contained higher concentrations of omega-3 polyunsaturated fats, including alpha-linolenic acid (ALA), the long-chain fatty acids eicosapentanoic acid (EPA) and docosahexanoic acid (DHA), and conjugated linoleic acid (CLA). However, iodine and selenium content were significantly higher in conventional milk.
Overall, the compositional differences in omega-3 fatty acids and CLA were modest, and unlikely to be nutritionally relevant. For iodine, although the conventional milk contained significantly greater concentrations, both organic and conventional sources would still contribute relevant levels of intake for this nutrient. The analysis indicated that the primary reason for differences in milk between organic and conventional milk were higher grazing intakes in organic cows. Thus, the nutritional composition may be explained by the level of fresh grass in the diet of a cow. This has been shown before, with the fatty acid composition of milk increasing linearly in relation to the level of fresh grass in the diet of the cow. This is not necessarily an exclusively organic vs. conventional distinction, given that the organic regulations for grazing are for 'when conditions allow', and therefore conserved forage may be used depending on season and/or climate conditions. Ultimately, it does appear that grazing, compared to grass silage feed or no grazing, results in higher amounts of unsaturated fat, protein, and the carotenoid, lutein.
A further systematic review and meta-analysis from the same research group investigated compositional differences in nutrient content in relation to animal meat, specifically analysing differences in fatty acid composition. The overall analysis detected higher concentrations of both omega-3 and omega-6 fatty acids in organic meat, higher overall polyunsaturated fat content and lower monounsaturated fat content, and similar saturated fat content. In analysis by meat type, higher polyunsaturated fat concentrations were detected specifically in pork and chicken, but not beef, lamb, or goat. Lower monounsaturated fat content was also only observed in pork and chicken. Saturated fat was lower in chicken, but there is no difference in other meat types. The magnitude of difference was small, and the body of evidence included in the analysis was weak and inconsistent.
The overall conclusion for animal produce generally, including milk, meat, and eggs, appears to indicate higher levels of polyunsaturated fat, in particular omega-3 content, comparing organic to conventional sources. However, the total body of evidence for meats is inconsistent and, for most nutrients, does not show nutritionally meaningful differences.
Limitations of Current Research Assessing Nutritional Composition
There are a number of limitations to the body of evidence in relation to nutritional composition that warrant consideration. One is that there remains insufficient data to accurately compare individual crops (as opposed to nutrients in an overall class of crop, i.e., 'fruit'), or to compare meat products from different livestock species. As a result, it is not possible to estimate differences in actual dietary intake of both potential beneficial or undesirable compounds from organic and conventional foods. Another is that many nutritionally relevant compounds lack sufficient data to be able to study in a meta-analysis and attempt to derive meaningful conclusions. Each systematic review and meta-analysis above used different methodological approaches, and this has generated debate within the field about the appropriate methodological framework to analyse food composition data, which have yet to be resolved.
Organic Food and Environmental Pollutants
Beyond considerations of nutritional composition, the potential for differences in levels of pesticides, heavy metals, and other synthetic (or non-synthetic but potentially undesirable) compounds, is also relevant to the discussion.
'Pesticides' are used in agriculture to defend crops by warding off infestation, supporting growth, and preventing disease. It is important to note that pesticides may be synthetic or non-synthetic, i.e., may be derived from natural sources, and it is permissible to use organic pesticides. Pesticide levels are regulated through the 'maximum residue level' (MRL), which is the upper level of pesticide residues allowed in or on a food, or in animal feed, and the lowest exposure necessary to protect consumers. In the EU, these are only approved after a risk assessment by the European Food Standards Agency (EFSA), while in the US a similar risk assessment process is conducted by the Environmental Protection Agency (EPA). A 'pesticide residue' is the measurable amount of an active substance, and the related metabolites or degradation byproducts, which may be found on harvested crops or in animal foods.
In the systematic review by Smith-Spangler et al., pesticide residues were found to be 5 times higher in conventional crops compared to organic. The data in the Barański et al. systematic review compared the frequency of occurrence of pesticide residues between organic and conventional crops (as a percentage of samples included). In meta-analysis of 66 data points, they found that residues were detected in 10.5% of organic crops compared to 46.3% of conventional crops, i.e., pesticide residues were 4 times higher in conventional crops. This was a similar pool of studies used in the analysis by Smith-Spangler et al., thus providing consistency in the observed difference. Barański et al. also found that nitrogen and cadmium, a heavy metal, were higher in cereal crops, but no significant levels were detected for vegetables or fruits.
However, these findings require some context. It is important to note that the MRL is a conservative estimate of the potential risk of an exposure in humans, and is set 100 times below the 'No Observable Effect Level' identified in risk assessment studies. While residue levels may be identified on conventional crops, it is a somewhat default finding that they would be higher, given that these same compounds are not used in organic crop production. The most recent EU report on pesticide residues in foods indicated that 95.5% (of 91,015 samples) of foods analysed were below the MRL level, while 2.7% exceeded this amount. In the US, only 0.59% of foods tested were found to exceed the MRL.
A final point in relation to the comparisons between organic and conventional crops is that these comparisons examine regulated compounds. Within organic farming practice, however, is an extensive use (either fully permitted, or with restrictions prior to a justification for use) of different botanical chemicals for which we currently lack any robust evidence for effects on human health. Rather, their use is often coupled with untested assumptions based on their 'natural' status, and the related assumption that 'natural' inherently means safe for humans. Thus, the distinction between organic vs. conventional is not a distinction between 'pesticide' vs. 'no pesticide', because organic farming uses pesticides and other biological control agents. The question is the related potential health effects, and environmental considerations, for which there is more evidence in relation to conventional synthetic pesticides.
Smith-Spangler et al. also found that bacteria resistant to 3 or more antibiotics were up to 3 times higher in conventional pork and chicken compared to organic sources. Many of the included studies were published around the time of the introduction of the EU ban on certain antibiotic uses in animal feed and monitoring of Salmonella and Campylobacter levels - two contaminants examined in the Smith-Spangler review - thus whether altered practices and regulations may have an effect on antibiotic resistant bacteria remains to be further elucidated. However, Smith-Spangler et al. found no difference in the risk for bacterial contamination with pathogenic bacteria, e.g., E.coli and Salmonella, comparing organic to conventional animal products. In sensitivity analysis (removing studies in the meta-analysis to see what effect they have on the overall result), removal of 1 study resulted in a statistically significantly higher risk for E.coli contamination from organic produce.
However, other research shows little difference in risk for bacterial contamination between organic and conventional produce. It is acknowledged that the issue of antibiotic resistance is of global importance for human health, but it is important to state that to date, the primary cause for antibiotic resistance in humans is antibiotic use by humans, and the local animal population is unlikely to be the source of microbial resistance in humans.
Organic Food and Health Outcomes
While the question of whether there are nutritional composition differences between organic and conventional foods is one consideration, a more pertinent question is the effects (if any) on human health outcomes. There is relatively little data, prospectively observational or from interventions, on health outcomes between organic and conventional food consumption.
A number of cross-sectional studies have compared incidence of allergies and atopic sensitisation in children consuming all organic food diets as part of an anthroposophic lifestyle, finding lower prevalence of atopy/allergy in children following such a lifestyle. However, the unique lifestyle of this population subgroup encompasses a number of factors which may lead to lower allergen sensitisation, for example consumption of fermented foods, and lower prevalence of allergy is observed in this population independent of diet.
However, the Dutch KOALA cohort study investigated the relationship between organic consumption vs. conventional food consumption during pregnancy and over the first two years of life on incidence of eczema. The cohort included mothers following a 'conventional lifestyle' (not be confused with conventional food in the context of food production) and mothers following 'alternative lifestyles', i.e., anthroposophic or otherwise. While there was no significant association between overall organic food consumption and eczema risk, there was a significantly lower odds (OR 0.67, 95 % CI 0.46-0.98) for eczema in infants strictly fed organic dairy. There was no significant association for other individual foods for eczema, and no significant associations noted for risk of atopic sensitisation at 2yrs. The lower odds of eczema was attributed to the higher levels of certain dairy fatty acids in breast milk (which was sampled as part of the study), in particular conjugated linoleic acid (CLA) and trans-vaccenic acid (TVA). This correlated (along with omega-3 fatty acids) with lower odds of eczema, atopic dermatitis, and sensitisation assessed by IgE antibodies, in the infants in the cohort.
The Norwegian Mother and Child study found a lower odds (OR=0.75, 95% CI 0.60-0.95) of pre-eclampsia from dietary analysis mid-pregnancy, comparing low organic vegetable to high organic vegetable intake. However, there was no significant association for total organic food consumption, or intakes of fruit. However, the dietary assessment method had not been validated for organic foods, and there was a small number of participants in the exposure group analysed. The frequency categories 'mostly' and 'often' were combined together and compared against the combination of categories 'sometimes' and 'never'. Actual intakes of organic foods, i.e., in grams per day, were not quantified. High healthy eating index scores were also associated with lower odds (OR 0.73, 95% CI 0.64-0.84) for pre-eclampsia. Given that frequent organic food consumption and high healthy eating index scores correlated, both of these were mutually adjusted for, resulting in similar results. Thus, both frequent organic food and a high healthy eating score were independently associated with lower odds for pre-eclampsia. This reflects a challenge in the literature relating organic food to health outcomes, which is that foods associated with reduced risk of disease - vegetables, fruits, wholegrains, less red meat - are higher in consumers who purchase organic foods.
As stated above, most of the human studies are cross-sectional studies in mother-child cohorts. However, there are a number of prospective adult cohorts which have examined chronic disease outcomes. The UK Million Women Study, a large cohort of 623,080 women, examined the relationship between organic vs. conventional food intake and cancer incidence - in particular breast cancer, soft tissue sarcoma, non-Hodgkins lymphoma - over an average of 9.3yrs of follow-up. Organic food intake was defined as 'never', 'sometimes', or 'usually/always'. The only statistically significant findings in the study were in relation to breast cancer and non-Hodgkins lymphoma: 'usually/always' consuming organic food was associated with a 9% (RR 1.09, 95% CI 1.01 - 1.17) increase in risk for breast cancer, while this consumer category was also associated with a 21% (RR 0.79, 95% CI 0.64 - 0.99) decrease in risk for non-Hodgkins lymphoma. Ultimately, of the 14 specific cancer sites analysed and in relation to total cancers incidence, there was no evidence of a reduced risk from organic food consumption, with the possible exception of non-Hodgkins lymphoma. In relation to the finding of increased risk in relation to breast cancer, it is important to note that the study did not adjust for oestrogen-receptor type, menopausal stage, or genetic risk, and as the authors highlighted it could be that women consuming more organic food are also more likely to attend breast cancer screening, i.e., more likely to be diagnosed from screening.
The French Nutri-Net Santé Cohort also investigated cancer incidence in 68,946 adult male and female participants relative to organic food consumption. Similar to the UK Million Women Study, there was a reduction in risk for non-Hodgkins lymphoma and per 5-point increase in organic food score (based on fruits, vegetables, soy-based products, grains and legumes, bread and cereals, and flour), there was a 25% (HR 0.75, 95% CI 0.60 - 0.93) lower risk for non-Hodgkins lymphoma over 4.9yrs. There was also a reduction in risk (HR 0.66, 95% CI 0.45 - 0.96) observed for postmenopausal breast cancer observed with the total organic food score, but this was not significant in the analysis confined to plant-derived organic food score suggesting that other organic foods may have been associated with the outcome in the main analysis. In this cohort, there was no significant association with any other cancer site investigated.
The suggested explanation for these findings of reduced cancer risk related to organic food consumption is lower exposure to pesticides. However, this question currently appears to be contested. In relation to carcinogenic potential of synthetic pesticides, these compounds have been considered to have low possible carcinogenic hazard in humans. However, these assessments often relate to one specific compound, and may not take into account the effect of multiple pesticide residues, in the context of a total diet.
Other concerns raised in the assessment of risk from pesticides include a lack of full consideration of independent science (i.e., non-industry funded), and that gaps in the evidence are too easily accepted in the safety and toxicology assessments. It is commonly cited that occupational exposure to pesticides, e.g. spraying, is associated with increased risk of cardiometabolic disease, Parkinson's disease, and certain cancers. However, this is not the same as exposure to pesticide residues that are consumed through diet.
Recent evidence also challenges the prior evidence regarding carcinogenic potential of pesticides in humans. While it is not incorrect to say that any health risks posed by pesticides would be largely eliminated by consuming no conventional produce, this is not likely to be a reality for a majority of the population. To paraphrase the review by Mie et al., the potential negative effect of pesticide residues consumed through diet should be an argument against fruit and vegetable consumption, yet neither should nutrient content of fruit and vegetables be used to justify exposures with potentially harmful health effects. Confining the assessment to diet alone, however, the current prospective human evidence does not suggest any strong links between organic food consumption and lower risk of cancers.
The extent to which the differences in certain concentrations of compounds, in particular polyphenols, may result in meaningful differences in health outcomes is confined to estimates and speculation. In their systematic review, Brandt et al. posited that if a person substituted all conventional fruit and vegetables in their diet for organic versions, this could increase their intake of 'secondary plant metabolites' like polyphenols by 12%, and that this could be expected to correspond to an increase in life expectancy of 17 days in women and 25 days in men.
However, there are a number of issues with this assumption. It is based on modelling which analysed the predicted effects of increasing fruit and vegetable intake per se, while the assumption by Brandt et al. is based on the idea that a given increase in the levels of polyphenol intake corresponds to an increase in total fruit and vegetable intake. This is a large assumption, given the wide differences between studies in important factors like testing method and seasonal differences.
Second, the model corresponded to a hypothetical equal-weight substitution, such that any health effect would be attributed to the difference in nutrient content rather than a change in gram per day intake. To put this in perspective, the current average fruit and vegetable intake in grams per day in men and women is 260g and 251g, respectively, based on the most recent National Diet and Nutrition Survey Data.
Even if we were to take the estimations of Brandt et al. as valid, and switching these levels to all organic would result in a health benefit corresponding to a 12% increase in bioactive compounds, the health benefit of actually meeting the 400g per day recommendation for fruit and vegetable intake would still be expected to yield a greater magnitude of benefit. For example, in the study which Brandt et al. based their estimates on, increasing fruit and vegetable intake by 100g was associated with a:
- 22% reduction in stomach cancer risk [RR 0.78, 95% CI 0.72–0.84]
- 7% reduction in colorectal cancer risk [RR 0.93, 95% CI 0.88–0.98]
Therefore, the ultimate conclusion, based on these assumptions, is that a benefit would be still derived from increasing total fruit and vegetable intake generally. In other terms, the impact of switching from conventional foods to organic foods could be smaller for diets already high in polyphenol-rich foods. Thus, it is critical that any differences in nutrient content of organic vs. conventional foods are examined in the context of the total diet. Percent change differences in contents for a single or a handful of nutrients may over-inflate apparent differences over what might be nutritionally relevant.
What no review in this area has appeared to address is the financial cost implications, which are important in the consideration of health outcomes. Analysis of the economic costs of complying with the UK Eatwell Guide indicated that households in the bottom 10% of income would have to spend ~74% of disposable income to comply with the dietary recommendations, compared to 6% of income for households in the top decile. These are important considerations for contextualising the pros and cons of organic vs. conventional food.
In populations falling substantially below targets for fruit and vegetables intake, and other foods associated with positive health effects (e.g. whole grains), current data would suggest that the health gains across the population which could be achieved by meeting those fruit/veg targets would significantly exceed the health gains from attempting to change current intake to an all-organic produce diet. In fact, the increase in cost may act as a further barrier to consuming more fruit, vegetables, etc. for many people.
Summary of Key Points
- The term 'organic' denotes a method of production, and is regulated by health regulatory bodies in the EU and US.
- Organically produced crops may yield higher levels of polyphenols. But evidence is more inconsistent regarding specific nutrients of interest, e.g., vitamin C and carotenoids.
- Organically produced pork may have lower levels of antibiotic-resistant bacteria.
- Organically produced animal-source foods generally (including milk, meat, and eggs), appear to have higher levels of polyunsaturated fat, in particular omega-3 content, compared to conventional sources. However, these levels are unlikely to be nutritionally relevant in a total diet.
- Based on recent estimates, between 0.5-2.7% of commercially available conventional foods may have pesticide residue levels above the maximum residue levels (the upper level of pesticide residues allowed in or on a food).
- While there remains concern about the potential carcinogenic effect of chronic exposure to pesticide residues, currently the evidence comparing organic to conventional foods does not suggest any substantially lower risk of overall cancers.
- Many other cohort studies are cross-sectional studies on mother and child. There are suggestions of reduced risk of allergy and atopic sensitisation. However, these studies are confounded by particularly unique lifestyles in the included participants.
- A continuing challenge in the epidemiology of organic vs. conventional food consumption, is the concurrent high levels of health promoting foods in individuals who regularly consume organic foods. Both of these factors may be independently associated with positive health outcomes.
- The overall weight of data does not emphatically support a unique health benefit of organic produce over and above what could be gained from a substantial increase in overall vegetable and fruit consumption across the population.
- The price differential and cost implications between organic vs. conventional food is a highly relevant factor to consider.
Statement Primary Author: Alan Flanagan
Alan is the Research Communication Officer at Sigma Nutrition. Alan is currently pursuing his PhD in nutrition at the University of Surrey, UK, with a research focus in chrononutrition. Alan previuosly completed a Masters in Nutritional Medicine at the same institution.