Review

Received: 29 July 2011 Revised: 12 October 2011 Accepted article published: 28 November 2011 Published online in Wiley Online Library: 19 March 2012

(wileyonlinelibrary.com) DOI 10.1002/jsfa.5639

Comparison of nutritional quality between

conventional and organic dairy products:
a meta-analysis
Eny Palupi,a Anuraga Jayanegara,b† Angelika Ploegera and Johannes Kahla∗

Abstract
As a contribution to the debate on the comparison of nutritional quality between conventional versus organic products, the
present study would like to provide new results on this issue specifically on dairy products by integrating the last 3 years’
studies using a meta-analysis approach with Hedges’ d effect size method. The current meta-analysis shows that organic dairy
products contain significantly higher protein, ALA, total omega-3 fatty acid, cis-9,trans-11 conjugated linoleic acid, trans-11
vaccenic acid, eicosapentanoic acid, and docosapentanoic acid than those of conventional types, with cumulative effect size
(±95% confidence interval) of 0.56 ± 0.24, 1.74 ± 0.16, 0.84 ± 0.14, 0.68 ± 0.13, 0.51 ± 0.16, 0.42 ± 0.23, and 0.71 ± 0.3,
respectively. It is also observed that organic dairy products have significantly (P < 0.001) higher omega-3 to -6 ratio (0.42 vs.
0.23) and
9-desaturase index (0.28 vs. 0.27) than the conventional types. The current regulation on organic farming indeed
drives organic farms to production of organic dairy products with different nutritional qualities from conventional ones. The
differences in feeding regime between conventional and organic dairy production is suspected as the reason behind this
evidence. Further identical meta-analysis may be best applicable for summarizing a comparison between conventional and
organic foodstuffs for other aspects and food categories.

c 2012 Society of Chemical Industry

Keywords: nutrition; organic; comparison; dairy products; meta-analysis

INTRODUCTION argued against that.2,17 – 19 Although systematic reviews have been

The organic food market is growing and expanding all over the conducted on this issue,17,18 most of them did not apply meta-
world,1 and it has been the fastest-growing market in the food analysis.19 Therefore, a valid conclusion from the various studies
sector over the last decade.2,3 This market was estimated to reach has not been possible so far.20
US $55 billion in 2009.1 As the third most important commodity Therefore, based on the description above, the main objective
after fruits and vegetables within the global organic food market, of this study was to compare the nutritional quality between
the organic dairy market is also increasing.4 This suggests that the conventional and organic dairy products. This was conducted by
market players including consumers are becoming increasingly summarizing the related studies for a period from March 2008
interested in this sector. to April 2011, which is not covered by the previous reviews17
The lack of consumer confidence in conventional products due (the period from January 1958 to February 2008 was reviewed by
to issues such as mad cow disease, the dioxin scandal and effects of Dangour et al.17 ), using a meta-analytical technique. The outcome
pesticide and antibiotic use is suspected as one of the driving forces is expected to provide new, additional scientific references in the
of this trend.5 Organic Monitor6 reported that the sales growth organic food quality area.
of dairy products rose over 30% in some countries in Europe
due to the BSE crisis. Apart from that, consumers are becoming
more concerned about the health aspects.7,8 It was shown that
health is the second strongest consideration (after hedonism) for
∗
consumers in choosing organic dairy products, followed by animal Correspondence to: Johannes Kahl, Department of Organic Food Quality
and Food Culture, University of Kassel, Nordbahnhofstrasse 1a, 37213
welfare and environmental issues.2,7,9 Consumers expect better Witzenhausen, Germany. E-mail: kahl@uni-kassel.de
quality in terms of health and nutritional properties from organic
labelled products.10,11
On the other hand, consumers lack information beyond the † Present address: ETH Zurich, Institute of Agricultural Sciences – Animal
Nutrition, 8092 Zurich, Switzerland.
food label. Scientific evidence is highly demanded in order to
promote consumer confidence in organic foods.5,10 However, the
available evidence is not yet able to fill this lack.12 There have been a Department of Organic Food Quality and Food Culture, University of Kassel,
contrasting results among individual experiments. Furthermore, 37213 Witzenhausen, Germany
the final conclusions of some reviews are contradictory. Some
of them inferred that organic products are healthier and have
2774

b Department of Animal Nutrition and Feed Technology, Bogor Agricultural

higher nutritional value than conventional ones,13 – 16 but others University, 16680 Bogor, Indonesia

J Sci Food Agric 2012; 92: 2774–2781 www.soci.org

c 2012 Society of Chemical Industry
Comparison of conventional and organic dairy products www.soci.org

EXPERIMENTAL nutritional quality insight, i.e. ratio of omega-3 to omega-6 fatty

Literature search and selection method acid (n-3/n-6) and -9 desaturase index, which is derived from
Literature search was conducted through a literature database, i.e. [CLA9]/([CLA9] + [VA]).22 The common units that were used for
ISI Web of Knowledge (http://apps.isiknowledge.com) and Science uniformity were: g g−1 dry matter for fresh forage in diet; kg milk
Direct (http://www.sciencedirect.com/) for a period from March per head per day for yield; g kg−1 milk for fat and protein content;
2008 to April 2011. The following keywords were used in the search: mg kg−1 milk fat for α-tocopherol and β-carotene; and g kg−1
‘comparison’, ‘conventional’, ‘organic’, ‘milk’, ‘dairy’, ‘compare’, fatty acid for the fat components.
‘label’, ‘claims’, ‘farm∗ management’, ‘production∗ management’,
‘nutrition’, and ‘quality’. An additional search was conducted Statistical analysis
using the names of the major authors in the field. These Effect size as ‘Hedges’ d’ was applied to quantify parameter
initial searches resulted in 994 potential references. Inaccessible distance between conventional and organic products. This method
full articles were obtained by direct correspondence to the was selected because of its ability to calculate the effect size
authors. Further, the following criteria were used for literature regardless the heterogeneity of sample size, measurement unit,
selection: (1) published in English as full-text articles; (2) peer- and statistical test result, and also suitability for estimating the
reviewed published journals; (3) direct comparison between effect of paired treatments.20,23 The conventional group was
organic and conventional; (4) declared ‘conventional’ (reported pooled into a control group (C) and the organic group was pooled
as conventional, high-input, intensive, non-organic) and ‘organic’ into an experimental group (E). The effect size (d) was calculated as
product (reported as organic label, organic claim, organically
certified, organically regulated, organic farm, organic supplier); (X̄ E − X̄ C
(5) dairy products, including raw milk and milk-based product as d= J
S
well as milk from animals other than bovine; and (6) comparison
of the nutritional quality aspect, including macro- and micro-
where X̄ E is the mean value from the experimental group and X̄ C
nutrients.
is the mean value of the control group. Therefore the positive
According to the initial title screening, as many as 945 references
effect size indicates that the parameter observed is greater in the
were eliminated because the topic was irrelevant to the research.
organic group, and vice versa. J is the correction factor for small
After reviewing the abstracts, 20 out of the 49 articles were
sample size, i.e.
selected. Then, seven articles were eliminated due to the irrelevant
nutritional parameters used (four articles), author not replying to 3
correspondence (two articles), and insufficient data for statistical J =1−
(4(NC + NE − 2) − 1)
meta-analysis (one article). In the end, the screening yielded 13
articles for use in subsequent data coding and statistical data
and S is the pooled standard deviation, defined as
analysis.


(NE − 1)(sE )2 + (NC − 1)(sC )2 )
Studies coding S=
(NE + NC − 2)
Some moderator variables were used for the studies coding and
initial information. Those variables were place of research, type of
where NE is the sample size of the experimental group, NC is the
survey, type of animal, type of product, processing unit, season,
sample size of the control group, sE is the standard deviation of
sample size of the conventional and organic groups, organic
the experimental group, and sC is the standard deviation of the
regulation used, and references. A total of 29 studies were derived
control group. The variance of Hedges’ d (vd ) is described as
from a comprehensive review of the 13 selected articles. The
mean value and standard deviation from several parameters
(NC + NE ) d2
were recorded and tabulated, and the units of measurements vd = +
were homogenized for further data analysis. Those parameters (NC NE ) (2(NC + NE ))
were: fresh forage in diet, yield, fat content, protein content,
and some vitamins (α-tocopherol and β-carotene). Minerals were Cumulative effect size (d++ ) is formulated as
excluded from the parameters because there was a lack of available  
information. Some fat components were recorded in more detail 
n
wi di
owing to the essential parameters for defining the nutritional
quality of the dairy product and the data availability from the d++ = i=1n 

selected articles. Those fat components were: total saturated wi
fatty acid (SFA), total monounsaturated fatty acid (MUFA), total i=1
polyunsaturated fatty acid (PUFA), stearic acid (C18 : 0), oleic acid
(C18 : 1 n-9), linoleic acid (C18 : 2 n-6), α-linolenic acid (ALA, where wi is the inverse of the sampling variance: wi = 1/vd .
C18 : 3 n-3), omega-3 fatty acid (n-3), omega-6 fatty acid (n-6), The precision of the effect size is described by using 95%
conjugated linoleic acid 9 (CLA9, cis-9,trans-11 C18 : 2), vaccenic confidence Interval (CI), i.e. d± (1.96 × sd ). All the above equations
acid (VA, trans-11 C18 : 1), eicosapentanoic acid C20 : 5 n-3 (EPA), are derived from Rosenberg et al.24 and Sanchez-Meca and Marin-
and docosapentanoic acid C22 : 5 n-3 (DPA). Docosahexaenoic Martinez.20 The calculated effect size is statistically significant if
acid was excluded from the list because the number of studies the CI does not reach the null effect size.20
(two) was too few for quantification. A minimum three studies A fail-safe number (Nft ) is calculated to recognize the publication
was needed to quantify the cumulative effect size.21 Additional bias caused by the insignificant studies which are not included on
2775

parameters were calculated from the data available for better the analysis. Nft > 5N + 10 is considered to provide evidence of

J Sci Food Agric 2012; 92: 2774–2781

c 2012 Society of Chemical Industry wileyonlinelibrary.com/jsfa
 www.soci.org E Palupi et al.

Table 1. List of comparison studies used in meta-analysis

Study Type of
code Place survey Animal Product Processing Season Nc No Organic regulation References

1 UK Farm Cow Whole milk Raw W&S 24 34 Organic certified 22

2 UK Farm Cow Whole milk Raw W 21 10
3 Switzerland Farm Cow Whole milk Pasteurized W&S 36 36 Organic farming 42
Switzerland
4 Netherlands Farm Cow Whole milk Raw W 5 5 Organic certified 11
5 UK Farm Cow Whole milk Raw W&S 16 20 Organic certified 43
6 UK Farm Cow Whole milk Raw S 4 5
7 UK Farm Cow Whole milk Raw W 4 5
8 UK Farm Cow Whole milk Raw W 4 5
9 UK Farm Cow Whole milk Raw S 4 5
10 Italy Farm Cow Whole milk Raw W&S 10 10 Organic milk supplier 44
11 Italy Farm Cow Curd Processed W&S 10 10
12 Italy Farm Cow Curd Cooked W&S 10 10
13 Italy Farm Cow Grana Padano Processed W&S 10 10
Cheese
14 Denmark Farm Cow Whole milk Raw W&S 75 50 Danish Ministry of Food, 45
Agriculture and
Fisheries Council
Regulation No.
2092/91
15 Poland Farm Cow Whole milk Raw W&S 21 21 EU Verordnung 1999 46
16 Poland Farm Cow Whole milk Raw W 21 21
17 Poland Farm Cow Whole milk Raw S 21 21
18 US Basket Cow Whole milk Pasteurized W 111 99 USDA standards 47, 10
19 US Basket Cow Whole milk Pasteurized W 82a 99 for organic production
20 Sweden Farm Cow Whole milk Raw W 10 10 Organic certified 48
21 Sweden Farm Cow Whole milk Raw S 15 15
22 Greece Farm Sheep Whole milk Raw W 41 41 Greek Ministry of 49
23 Greece Farm Goat Whole milk Raw W 40 40 Rural Development and
Food
24 UK Basket Cow Whole milk Pasteurized W&S 48 40 Certified organic 50
25 UK Basket Cow Whole milk Pasteurized S 12 10
26 UK Basket Cow Whole milk Pasteurized W 12 10
27 UK Basket Cow Whole milk Pasteurized S 12 10
28 UK Basket Cow Whole milk Pasteurized W 12 10
29 Germany Farm Cow Whole milk Raw W 30 5 Certified by Demeter 51

Nc , sample size from conventional group; No , sample size from organic group; W, winter/indoor feeding period; S, summer/outdoor feeding period;
W&S, whole season.
a rbST-free versus organic milk with the same organic sample as study code 18.

a robust meta-analysis model. Nft was calculated using Rosenthal RESULTS

et al.’s method.24 The least sample size from individual studies Profile of the selected studies
was applied as N. Cohen’s benchmarks are used as the standard The selected studies originated from different countries, i.e.
judgment borders to indicate how large the effect size is. Those UK, Italy, Poland, USA, Greece, Sweden, Denmark, Germany,
benchmarks are 0.2 for small, 0.5 for medium, and 0.8 for large Netherlands and Switzerland (Table 1). All selected studies
effect size.24 Additional cumulative effect size calculation was mentioned that, for the organic group, the samples were
applied on the different clusters from some variables such as organically certified. The regulation or organic certification body
season, processing unit, and type of survey. This was applied used varied, i.e. organic farming Switzerland, Danish Ministry of
to see the influence of the moderator variables on the final Food, Agriculture and Fisheries Council Regulation no. 2092/91,
effect size. All of the above effect size-related calculations were EU Verordnung 1999, USDA standard for organic products, Greek
calculated using MetaWin 2.0.24 Weighted paired-samples t-test Ministry of Rural Development and Food, and Demeter. The season
was applied for the parameters. Least sample size from individual and the period of research were encoded separately. Three season
studies was applied as the weighting factor. The mixed-effect types were used here, i.e. whole season (W&S: 34.5%) if the whole
model of analysis of variance (ANOVA) with Tukey’s test was season from at least one year’s accumulated result was reported,
also applied to see the influence of the moderator variables to summer (S: 20.7%) and winter (W: 44.8%). Some authors clearly
the parameters. These tests were conducted using SPSS software mentioned whether they conducted the evaluation in summer,
2776

version 13.25 winter or whole year, but others did not. Unclear cases were

wileyonlinelibrary.com/jsfa

c 2012 Society of Chemical Industry J Sci Food Agric 2012; 92: 2774–2781
Comparison of conventional and organic dairy products www.soci.org

Figure 1. Forest plot of cumulative effect size (d++ ) and 95% confidence interval (CI) of some nutritional parameters comparing conventional and organic
dairy products. Bold lines indicate the robust model.

categorized according to the typical feeding management and study size. The fail-safe number (Nfs ) briefly shows which one is
research period presented. An outdoor feeding programme was appropriate to be pooled as the final strong conclusions. This
categorized as summer, whereas an indoor feeding programme number expresses how many sample study sizes should be added
was categorized as winter. The sample size of the selected studies to the studies in order to change the initial effect size into a
ranged widely, from four to 111 samples. negligible effect size. If Nfs > 5N + 10, where N is the study
effect size used for initial effect size calculation, then the result
may be taken as the final robust conclusion.24 According to these
Organic versus conventional dairy products fail-safe number rules, robust parameters are fresh forage in diet,
The forest plot of cumulative effect size and 95% CI of all parameters protein, ALA, omega-3 fatty acid, CLA9, VA, EPA, DPA (higher in
(Fig. 1) illustrates the path of comparison of nutritional quality organic dairy product), oleic acid, linoleic acid, and omega-6 fatty
between conventional and organic dairy products. According to acid (lower in the organic product). These robust conclusions are
the cumulative effect size (d++ , ±95% CI), it is clear that organic clearly indicated by the bold lines in Fig. 1.
dairy product contains significantly higher ALA (1.74 ± 0.16) and Further results on the cumulative effect size of milk yield show
total omega-3 fatty acid (0.84 ± 0.14) with large effect size, higher a negative and significant value, i.e. −0.9R ± 0.26 (d++ ± 95%
protein (0.56 ± 0.24), CLA9 (0.68 ± 0.13), VA (0.51 ± 0.16) and DPA CI). Moreover, according to the weighted t-test (Table 2), it
(0.71 ± 0.3) with medium effect size, and higher fat (0.21 ± 0.18), was observed that organic dairy product had a significantly
SFA (0.31 ± 0.15), PUFA (0.18 ± 0.15) and EPA (0.42 ± 0.23) with (P < 0.001) higher ratio of omega-3 to omega-6 (n-3/n-6) than
small effect size, compared to the conventional product. The that of conventional dairy product, with values of 0.42 and 0.23,
result also indicated that organic dairy farming feeds cattle with respectively. Also it had a significantly (P < 0.001) higher 9-
significantly higher fresh forage than that the conventional one desaturase index (0.28 vs. 0.27).
does, with large effect size, i.e. 0.92 ± 0.41.
Negative effect sizes were found in some parameters, which Seasonal factor
indicated that organic product contains smaller amount of the A forest plot of ALA, n-3, CLA 9 and VA from different seasons (Fig. 2)
observed parameters. Those parameters are MUFA (−0.35 ± 0.15), illustrates that season significantly influenced the cumulative
stearic acid (−0.38 ± 0.23), oleic acid (−1.44 ± 0.32), linoleic effect size of omega-3 fatty acid. According to the ANOVA and
acid (−0.71 ± 0.2) and omega-6 fatty acid (−0.53 ± 0.14). All Tukey’s test, the seasonal factor significantly (P < 0.05) influenced
except oleic acid (large effect size) were categorized as medium the -9 desaturase index and n-3/n-6 ratio. Both for conventional
and small effect size. Insignificant cumulative effect size, which and organic dairy products, the n-3/n-6 ratio was higher in summer
was shown if the ±95% CI was higher than or the same as the than that in winter, with the mean value always significantly
cumulative effect size, was also found. Accordingly, insignificant higher in the organic product (Fig. 3). Similarly, the mean value
results were observed for the content of α-tocopherol (0.44±0.62) of the -9 desaturase index was higher in summer, both for
and β-carotene (0.49 ± 0.61). conventional and organic dairy products. However, no difference
Not all of those results enabled us to make strong conclusions in the respective parameter was found between organic versus
2777

owing to the conflicting results among studies and the small conventional products during the summer season.

J Sci Food Agric 2012; 92: 2774–2781

c 2012 Society of Chemical Industry wileyonlinelibrary.com/jsfa
 www.soci.org E Palupi et al.

Table 2. Cumulative effect size, weighted mean value and weighted P-value from all parameters
Cumulative
effect size Studies
Parameters Unit (d++ ) ± 95%Cl size (N) Nfs xc xo SEM P-value Np

Fresh forage in diet g g−1 dry matter 0.9198∗ 0.4076 4 80.1R 0.33 0.60 0.014 ∗∗∗
279
Yield kg milk cow−1 day−1 −0.9028∗ 0.2636 7 154.9R 18.74 15.51 0.154 ∗∗∗
285
Fat content g kg−1 milk 0.2143∗ 0.1797 8 18NR 35.90 36.90 0.081 ∗∗∗
331
Protein content g kg−1 milk 0.5626∗ 0.2409 5 43.6R 31.70 32.50 0.026 ∗∗∗
271
SFA g kg−1 fatty acid 0.3121∗ 0.1490 10 39NR 667.61 675.58 0.933 ∗∗∗ 483
MUFA g kg−1 fatty acid −0.3481∗ 0.1496 10 53.6NR 270.29 259.06 0.765 ∗∗∗ 483
PUFA g kg−1 fatty acid 0.1794∗ 0.1500 10 46.9NR 43.89 45.39 0.309 ∗∗∗ 483
C18:0 (stearic acid) g kg−1 fatty acid −0.3798∗ 0.2347 5 14.8NR 106.57 101.33 0.370 ∗∗∗ 278
C18:1 n-9 (oleic acid) g kg−1 fatty acid −1.4439∗ 0.3228 4 266.7R 227.81 209.42 0.743 ∗∗∗
273
C18:2 n-6 (linoleic acid) g kg−1 fatty acid −0.7151∗ 0.1998 6 125.4R 27.26 21.60 0.204 ∗∗∗
404
C18:3 n-3 (ALA) g kg−1 fatty acid 1.7444∗ 0.1622 16 2875.8R 4.79 7.61 0.048 ∗∗∗
529
n-3 g kg−1 fatty acid 0.8435∗ 0.1373 18 1786.4R 5.61 9.20 0.050 ∗∗∗
574
n-6 g kg−1 fatty acid −0.5283∗ 0.1418 14 89.3R 27.64 23.42 0.197 ∗∗∗ 539
n-3/n-6 ratio – – – – 0.23 0.42 0.004 ∗∗∗ 539
CLA9 g kg−1 fatty acid 0.6851∗ 0.1346 17 686.6R 6.59 8.38 0.061 ∗∗∗ 585
VA g kg−1 fatty acid 0.5147∗ 0.1562 9 163.8R 18.78 22.82 0.157 ∗∗∗ 478
9-desaturase index g kg−1 fatty acid – – – – 0.27 0.28 0.001 ∗∗∗
478
EPA g kg−1 fatty acid 0.4244∗ 0.2318 6 325.1R 0.37 0.64 0.005 ∗∗∗
283
DPA g kg−1 fatty acid 0.7120∗ 0.3009 4 202.2R 0.66 1.04 0.011 ∗∗∗
257
α-tocopherol mg kg−1 milk fat 0.4452NS 0.6227 3 8.5NR 21.04 23.39 0.334 ∗∗∗
84
β-carotene mg kg−1 milk fat 0.4914NS 0.6148 3 10.9NR 4.38 5.29 0.048 ∗∗∗ 84

CI, confidence interval; N, study size for effect size calculation; Nfs , fail-safe number; x̄c , weighted mean value from conventional group; x̄o , weighted
mean value from organic group; SEM, standard error of the mean; Np , cumulative sample size for t-test; Yield, calculated from cow only; SFA, total
saturated fatty acid; MUFA, total monounsaturated fatty acid; PUFA, total polyunsaturated fatty acid; n-3, total omega-3 fatty acid including long-
and medium-chain fatty acid; n-6, total of omega-6 fatty acid including long- and medium-chain fatty acid; n-3/n-6, ratio of total omega-3 fatty acid
and total omega-6 fatty acid; CLA9, conjugated linoleic acid 9 (C18:2 c9t11); VA, vaccenic acid (C18:1 t11);  9-desaturase index, estimated based on
[CLA9]/([VA] + [CLA9]); R, model is robust (Nfs > 5N + 10); NR, model is not robust (Nfs ≤ 5N + 10); NS, not significant.
∗
P-value < 0.05;
∗∗ P-value < 0.01;
∗∗∗ P-value < 0.001.

Figure 2. Forest plot of cumulative effect size (d++ ) and 95% confidence interval (CI) of ALA, n-3, CLA9 and VA from different seasons. Bold lines indicate
the winter season.

DISCUSSION valuable nutrients. It is true that nutritional claims can be made

Defining the ‘premium’ nutritional quality of dairy product if only a significant amount can be proved12 but, unfortunately,
According to Kahl et al.,12 there are two important steps in in most cases significant amounts of valuable milk nutrients have
constructing evidence in the food quality area to ‘define’ the not yet been recognized. ‘Significant’ amount means the amount
substance that is going to be proved and then to ‘prove’ it which enables significant support of human health.
using appropriate tools. A product is considered to have better Comprehensive reviews from Huth et al.26 and Haug et al.27
2778

nutritional quality than others if it contains higher amounts of some enable some estimations to be given concerning a rule for

wileyonlinelibrary.com/jsfa

c 2012 Society of Chemical Industry J Sci Food Agric 2012; 92: 2774–2781
Comparison of conventional and organic dairy products www.soci.org

It appears that the most plausible reason from this evidence

is the difference in feeding regime between organic and
conventional dairy production systems. Higher fresh forage
intake, as observed in organic dairy production, is associated
with higher intake of PUFA, including ALA.31,32 This is due
to the high amount of unsaturated fatty acids, especially ALA
and other omega-3 fatty acids in fresh forage, and it is more
prominent in the early stage of maturity.27 These omega-3
Figure 3. Mean value (± standard deviation) of -9 desaturase index and fatty acids cannot be synthesized in the mammary glands of
n-3/n-6 for different seasons. Note: 9-desaturase index = [CLA9]/([VA]
+ [CLA9]); n-3/n-6, omega-3/omega-6 ratio; C, conventional group; O,
dairy animals33 since -9 desaturase enzyme is not able to
organic group; dark grey bars, summer; light grey bars, winter; Mean make a double bond at the omega-3 position.27 Therefore the
values with superscript letters (a–c) are significantly different (P-value contents are highly dependent on the intake of the respective
<0.05) according to Tukey’s honestly significant difference test. compounds in the diet. However, PUFA, including omega-3 fatty
acids, may undergo transformation processes in the rumen of
ruminants through the action of microbes. The transformation
premium nutritional quality of dairy products. Although Huth
processes are through lipolysis and biohydrogenation of fatty
et al.26 provides a recommendation specifically for the Norwegian
acids.34,35 Unsaturated fatty acids including ALA are hydrogenated
case, it may be used as an initial reference to define the premium
into various isomers of PUFA and MUFA, particularly trans and
dairy product worldwide, since references concerning this issue
conjugated fatty acids, and end up as stearic acid (C18 : 0),
are limited. The premium nutritional quality of dairy product is
which is the ultimate product of the biohydrogenation process.34
applied if it enables the following list to be fulfilled:
Inhibition of the biohydrogenation process, at least partially,
1. High omega-3 to omega-6 ratio, close to 1:2.27 may occur as indicated by higher CLA9 and VA, and lower
2. High CLA9 content,26,27 with a significant amount of CLA9 of C18 : 0 in the organic dairy product in relation to the higher
about 140 to 420 mg d−1 .26 fresh forage intake. Plant secondary compounds, particularly
3. High VA content.26 However, Haug et al.27 stated that a low the phenolic compounds present in forages, may contribute
proportion of VA is preferable. To cover instances that are to this inhibition. Accordingly, phenolic compounds have been
both high in CLA9 and with quite a low proportion of VA, the reported to reduce PUFA biohydrogenation, accumulate VA and/or
-9 desaturase index is applied. The -9 desaturase index = decrease concentration of C18:0.36 – 38 The effects seem to be
[CLA9]/([CLA9] + [VA]).22 related to the phenolic toxicity to microbial species that are
4. At least secure the level of 14 selected key compounds, i.e. involved in fatty acid biohydrogenation, either in the initial steps
protein, vitamins A, D, E, C, B1 , B2 , B11 and B12 , calcium, zinc, of biohydrogenation by inhibiting Butyrivibrio fibrisolvens and/or
iron, potassium and essential oils.28 in the terminal step by inhibiting Clostridium proteoclasticum.36,39
5. Using energy content as the basis for calculation is Further, higher VA supply to the mammary gland may enhance
preferable.29 the content of CLA9 through the action of -9 desaturase.33
Not all of the above points can be covered by the current meta-
analysis. However, a synthesis review should be able to bridge the Influence of seasonal factor
update concept and the results available from primary studies. The availability and variability of feed vary with the season, which
Therefore, an optimum method would be to adapt the above then indirectly influences the nutrient composition of milk.30,40
definition of nutritional quality of dairy product according to the Normally, the nutritional quality of milk, especially the ratio n-3
results available from the selected studies, with only significant to n-6, is reduced during wintertime due to the shifting of the
and robust results acceptable for the final conclusion. Thus the feeding system from outdoors (normally in summer) to indoors
current study takes particular interest in certain parameters, (normally in winter), which then alters the feed from fresh forage
i.e. protein, ALA, omega-3/omega-6 ratio (n-3/n-6), CLA9, 9- to concentrate and conserved forage.27 An open question may
desaturase index, EPA and DPA. then arise as to whether the organic dairy product is enabled to
secure its ‘premium’ nutritional quality during the whole season.
The evidence and a plausible explanation The result presented shows that the effect sizes of some variables
observed, i.e. ALA, n-3, CLA9 and VA, are always positive and
Significantly higher amounts of protein, ALA, n-3, CLA9, VA, EPA
robust both during summer and winter. The result also shows
and DPA in organic dairy products than in conventional products,
that the ratio of n-3 to n-6 and -9 desaturase index are always
as well as a higher ratio of n-3 to n-6 (approximately twofold)
significantly higher in organic dairy product than in conventional
and 9-desaturase index, indicate that the organic dairy product
product, both in winter and summer periods. This evidence may
may have a premium nutritional quality. The current meta-analysis
suggest that organic dairy product seems able to maintain its
also shows that organic dairy farming feeds cattle with higher
‘premium’ nutritional quality during the whole season.
fresh forage levels than does conventional farming, with a large
effect size. This provides evidence that, although the regulation
and also the implementation of organic dairy farming vary among Suitability of meta-analysis as a tool for comparing organic
countries,30 it seems sufficient to guarantee that dairy farms feed versus conventional foods
the cattle a higher fresh forage level than on conventional farms. Meta-analysis, especially the Hedges’ effect sizes technique, seems
The evidence seems able to answer scepticism among the players to be among the best choices for the current review. This technique
in the organic dairy market that the current regulation on dairy is suitable for comparing two or more groups and thus might be
farming indeed drives organic farming to produce an enhanced used for further investigation in comparing conventional and
2779

nutritional quality of organic dairy product. organic foodstuffs other than dairy products or even for wider

J Sci Food Agric 2012; 92: 2774–2781

c 2012 Society of Chemical Industry wileyonlinelibrary.com/jsfa
 www.soci.org E Palupi et al.

meta-analysis. Moreover, this technique has been shown to enable 11 Bloksma J, Adriaansen-Tennekes R, Huber M, van de Vijver LPL, Baars T
small-size studies to be tackled. Furthermore, this meta-analysis and de Wit J, Comparison of organic and conventional raw milk
may offer more objective and repeatable results. Appropriate quality in the Netherlands. Biol Agric Hortic 26:69–83 (2008).
12 Kahl J, van der Burgt GJ, Kusche D, Bügel S, Busscher N, Hallmann E,
presentation of data and results may give opportunities for the et al, Organic food claims in Europe. Food Technol 64:38–46 (2010).
next analyst to reuse information needed for further investigation. 13 Worthington V, Nutritional quality of organic versus conventional
On the other hand, this technique is relatively time consuming. fruits, vegetables, and grains. J Altern Complem Med 7:161–173
Furthermore, it is limited in that only studies which clearly mention (2001).
14 Soil Association, Organic Farming, Food Quality and Human Health: A
the exact values of sample size, mean and standard deviation are Review of the Evidence. Soil Association, Bristol (2000).
eligible for selection. In the case that standard deviation values 15 Magkos F, Arvaniti F and Zampelas A, Organic food: nutritious food
are not given in the papers, mixed models of ANOVA or paired or food for thought? A review of the evidence. Int J Food Sci Nutr
(weighted) t-test, as also done in the present study, may provide 54:357–371 (2003).
an alternative method for conducting meta-analysis studies.41 16 Lairon D, Nutritional quality and safety of organic food: a review. Agron
Sustain Dev 30:33–41 (2010).
17 Dangour AD, Dodhia SK, Hayter A, Allen E, Lock K and Uauy R,
Nutritional quality of organic foods: a systematic review. Am J Clin
Nutr 90:680–685 (2009).
CONCLUSIONS 18 Dangour AD, Lock K, Hayter A, Aikenhead A, Allen E and Uauy
Current meta-analysis of studies conducted over the last 3 years R, Nutrition-related health effects of organic foods: a systematic
presents arguments that the current regulation on dairy farming review. Am J Clin Nutr 92:203–210 (2010).
indeed enables the driving of organic farming to produce organic 19 Woese K, Lange D, Boess C and Bögl KW, A comparison of organically
and conventionally grown foods: results of a review of the relevant
dairy products with different nutritional quality from conventional literature. J Sci Food Agric 74:281–293 (1999).
products. The difference in feeding regime between conventional 20 Sanchez-Meca J and Marin-Martinez F, Meta analysis, International
and organic dairy production is suspected to be the reason Encyclopedia of Education (3rd edn). Elsevier, Amsterdam,
behind this evidence. Further identical meta-analysis may be pp. 274–282 (2010).
best applicable for evaluating and summarizing the comparison 21 Lipsey MW and Wilson DB, Practical Meta-analysis. Applied Social
Research Methods Series, ed. By Bickman L and Rog DJ. Sage
of conventional and organic foodstuffs for other aspects and for Publications, Thousand Oaks, CA (2001).
food categories other than dairy products. 22 Butler G, Nielsen JH, Slots T, Seal C, Eyre MD, Sanderson R, et al, Fatty
acid and fat-soluble antioxidant concentrations in milk from
high- and low-input conventional and organic systems: seasonal
variation. J Sci Food Agric 88:1431–1441 (2008).
ACKNOWLEDGEMENTS 23 Hedges LV and Olkin I, Statistical Methods for Meta-Analysis. Academic
JK thanks the Bundesministerium für Ernährung, Landwirtschaft Press, London (1985).
24 Rosenberg MS, Adams DC and Gurevitch J, MetaWin: Statistical
und Verbraucherschutz (BMELV) for financial support within Software for Meta-analysis: Version 2.0. Sinauer Associates,
project 2810OE096. Sunderland, MA (2000).
25 SPSS 13, Statistical Package for Social Sciences. IBM, Somers, NY (2007).
26 Huth PJ, DiRienzo DB and Miller GD, Major scientific advances with
dairy foods in nutrition and health. Am Dairy Sci Assoc 86:1207–1221
REFERENCES (2006).
1 Willer H and Kilcher L, (Eds.) The World of Organic 27 Haug A, Hostmark AT and Harstad OM, Bovine milk in human nutrition:
Agriculture–Statistics and Emerging Trends 2010. IFOAM, a review. Lipids Health Dis 6:1–16 (2007).
Bonn (D) and FiBL, Frick (CH) Available: http://www.organic- 28 Steijns JM, Dairy products and health: focus on their constituents or
world.net/yearbook-2010.html?&L=0, (06.03.2012). on the matrix? Int Dairy J 18:425–435 (2008).
2 Bourn D and Prescott J, A comparison of the nutritional value, 29 Drewnowski A, Concept of a nutritious food: toward a nutrient density
sensory qualities, and food safety of organically and conventionally score. Am J Clin Nutr 82:721–732 (2005).
produced foods. Crit Rev Food Sci Nutr 42:1–34 (2002). 30 Blair R, Nutrition and Feeding of Organic Cattle. CABI Publishing,
3 Fossel PV, Organic Farming: Everything You Need to Know. Voyageur Wallinford, UK (2011).
Press, Stillwater, MN, pp. 6–10 (2007). 31 Leiber F, Kreuzer M, Nigg D, Wettstein HR and Scheeder MRL, A study
4 Organic Monitor, Global sales of organic food and drink recovering, on the causes for the elevated n-3 fatty acids in cows’ milk of alpine
in The Global Market for Organic Food and Drink: Business origin. Lipids 40:191–202 (2005).
Opportunities and Future Outlook (3rd edn). [Online]. Available: 32 Dewhurst RJ, Shingfield KJ, Lee MRF and Scollan ND, Increasing the
http://www.organicmonitor.com/700340.htm [17 June 2011]. concentrations of beneficial polyunsaturated fatty acids in milk
5 Siderer Y, Maquet A and Anklam E, Need for research to support produced by dairy cows in high-forage systems. Anim Feed Sci
consumer confidence in the growing organic food market. Trends Technol 131:168–206 (2006).
Food Sci Technol 16:332–343 (2005). 33 Chilliard Y, Glasser F, Ferlay A, Bernard L, Roul J and Doreau M, Diet,
6 Organic Monitor, The European Market for Organic Dairy Products. rumen biohydrogenation and nutritional quality of cow and goat
[Online]. Available: http://www.organicmonitor.com/100143.htm milk fat. Eur J Lipid Sci Technol 109:828–855 (2007).
[27 May 2011]. 34 Jenkins TC, Wallace RJ, Moate PJ and Mosley EE, Recent advances in
7 Grunert KG, Bech-Larsen T and Bredahl L, Three issues in consumer biohydrogenation of unsaturated fatty acids within the rumen
quality perception and acceptance of dairy products. Int Dairy J microbial ecosystem. J Anim Sci 86:397–412 (2008).
10:575–584 (2000). 35 Doreau M, Bauchart D and Chilliard Y, Enhancing fatty acid
8 Napolitano F, Girolami A and Braghieri A, Consumer liking and composition of milk and meat through animal feeding. Anim Prod
willingness to pay for high welfare animal-based products. Trends Sci 51:19–29 (2011).
Food Sci Technol 21:537–543 (2010). 36 Khiaosa-Ard R, Bryner SF, Scheeder MRL, Wettstein HR, Lieber F,
9 Hermansen JE, Organic livestock production systems and appropriate Kreuzer M, et al, Evidence for the inhibition of the terminal step of
development in relation to public expectations. Livest Prod Sci ruminal α-linolenic acid biohydrogenation by condensed tannins.
80:3–15 (2003). J Dairy Sci 92:177–188 (2009).
10 Vicini J, Etherton T, Kris-Etherton P, Ballam J, Denham S, Staub R, 37 Vasta V, Mele M, Serra A, Scerra M, Luciano G, Lanza M, et al, Metabolic
et al, Survey of retail milk composition as affected by label fate of fatty acids involved in ruminal biohydrogenation in sheep
claims regarding farm-management practices. J Am Diet Assoc fed concentrate or herbage with or without tannins. J Anim Sci
2780

108:1198–1203 (2008). 87:2674–2684 (2009).

wileyonlinelibrary.com/jsfa

c 2012 Society of Chemical Industry J Sci Food Agric 2012; 92: 2774–2781
Comparison of conventional and organic dairy products www.soci.org

38 Cabiddu A, Salis L, Tweed JKS, Molle G, Decandia M and Lee MRF, The 45 Slots T, Butler G, Leifert C, Kristensen T, Skibsted LH and Nielsen JH,
influence of plant polyphenols on lipolysis and biohydrogenation Potentials to differentiate milk composition by different feeding
in dried forages at different phenological stages: in vitro study. J Sci strategies. Dairy Sci 92:2057–2066 (2009).
Food Agric 90:829–835 (2010). 46 Bilik K and Lopuszanska-Rusek M, Effect of organic and conventional
39 Vasta V, Yanez-Ruiz DR, Mele M, Serra A, Luciano G, Lanza M, et al, feeding of red- and white-cows on productivity and milk com-
Bacterial and protozoal communities and fatty acid profile in the position. Ann Anim Sci 10:441–458 (2010).
rumen of sheep fed a diet containing added tannins. Appl Environ 47 O’Donnell AM, Spatny KP, Vicini JL and Bauman DE, Survey of the
Microbiol 76:2549–2555 (2010). fatty acid composition of retail milk differing in label claims
40 Weller RF and Cooper A, Seasonal changes in crude protein based on production management practices. Am Dairy Sci Assoc
concentration of mixed swards of white clover/perennial ryegrass 93:1918–1925 (2010).
swards grown without fertiliser N in an organic farming system. 48 Larsen MK, Nielsen JH, Butler G, Leifert C, Slots T, Kristiansen GH, et al,
Grass Forage Sci 56:92–95 (2001). Milk quality as affected by feeding regimens in a country with
41 St-Pierre NR, Integrating quantitative findings from multiple studies climatic variation. J Dairy Sci 93:2863–1873 (2010).
using mixed model methodology. J Dairy Sci 84:741–755 (2001). 49 Tsiplakou E, Kotrotsios V, Hadjigeorgiou I and Zervas G. Differences in
42 Collomb M, Bisig W, Bütikoter U, Sieber R, Bregy M and Etter L, Fatty sheep and goats milk fatty acid profile between conventional and
acid composition of mountain milk from Switzerland: comparison organic farming systems. J Dairy Res 77:343–349 (2010).
of organic and integrated farming systems. Int Dairy J 18:976–982 50 Butler G, Stergiadis S, Seal C, Eyre M and Leifert C, Fat composition of
(2008). organic and conventional retail milk in northeast England. J Dairy
43 Butler G, Collomb M, Rehberger B, Sanderson R, Eyre M and Leifert C, Sci 94:24–36 (2011).
Conjugated linoleic acid isomer concentrations in milk from 51 Schröder M and Yousefi F, Investigating the day-to-day variations
high- and low-input management dairy systems. J Sci Food Agric of potential marker fatty acids for organic milk in milk from
89:697–705 (2009). conventionally and organically raised cows. Eur Food Res Technol
44 Prandini A, Sigolo S and Piva G. Conjugated linoleic acid (CLA) and 232:167–174 (2011).
fatty acid composition of milk, curd and Grana Padano cheese in
conventional and organic farming systems. J Dairy Res 76:278–282
(2009).

2781

J Sci Food Agric 2012; 92: 2774–2781

c 2012 Society of Chemical Industry wileyonlinelibrary.com/jsfa