NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Balk E, Chung M, Raman G, et al. B Vitamins and Berries and Age-Related Neurodegenerative Disorders. Rockville (MD): Agency for Healthcare Research and Quality (US); 2006 Apr. (Evidence Reports/Technology Assessments, No. 134.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

Cover of B Vitamins and Berries and Age-Related Neurodegenerative Disorders

B Vitamins and Berries and Age-Related Neurodegenerative Disorders.

Show details



This evidence report on B vitamins and berries and age-related neurodegenerative disorders is based on a systematic review of the literature. The Tufts-New England Medical Center Evidence-based Practice Center (Tufts-NEMC EPC) held meetings and teleconferences with a technical expert panel (TEP) to identify specific issues central to this report. The TEP was comprised of technical experts in basic and clinical research in neuroscience, nutrition, B vitamins, and berries. A comprehensive search of the medical literature was conducted to identify studies addressing the key questions. Evidence tables of study characteristics and results were compiled, and the methodological quality and the applicability of studies were appraised. Study results were summarized with both qualitative and quantitative reviews of the evidence, evidence and summary tables

A number of individuals and groups supported the Tufts-NEMC EPC in preparing this report. The TEP served as our science partner. It included technical experts, representatives from the Agency for Healthcare Research and Quality (AHRQ), and both the National Center for Complementary and Alternative Medicine (NCCAM) and the Office of Dietary Supplements (ODS) at the National Institutes of Health (NIH). The TEP worked with the EPC staff to refine key questions, identify important issues, and define parameters for the report. Additional clinical domain expertise was obtained through local experts who joined the EPC. A draft version of this report was critically appraised by a panel of peer reviewers.* Revisions were made based on their comments; although all statements within the report are those of the authors only.

The review process and the report have been structured to account for the separate, but parallel, issues related to the effects of B vitamins and of berries. Processes related to neuroscience and to understanding animal and in vitro studies occurred in conjunction with all team members and relevant TEP members, whereas those related to either B vitamins or berries specifically occurred separately. Because of the small amount of literature related to berries and neurocognitive outcomes, the report encompasses both interventions. The report Introduction, Results, and Discussion chapters are structured such that common issues and topics are discussed first, followed by B vitamins, and then berries.

Key Questions Addressed in This Report

B Vitamins


What is the evidence regarding mechanisms of action of the B vitamins B1, B2, B6, B12, and folate (singly and in combination) for preventing, decreasing the rate of progression of, or reversing the neurological changes associated with age-related neurodegenerative conditions such as Parkinson's or Alzheimer's disease?


What is the evidence that the B vitamins B1, B2, B6, B12, and folate can prevent, decrease the rate of progression of, or reverse the neurological changes associated with age-related neurodegenerative conditions such as Parkinson's or Alzheimer's disease in humans


What adverse events in humans have been reported in the literature for supplementation with the B vitamins B1, B2, B6, B12, and folate?


Do the frequency of adverse events vary with source, dose, or other evaluated factors?



What are the constituents in berries with beneficial nerve- and brain-related health effects (from in vitro, animal, and human studies)?


In what other food sources are these constituents found?


What is the evidence regarding mechanisms of action of berry constituents for preventing, decreasing the rate of progression of, or reversing the neurological changes associated with age-related neurodegenerative conditions, including Parkinson's or Alzheimer's disease?


What is the evidence that the constituents of berries can prevent, decrease the rate of progression of, or reverse the neurological changes associated with age-related neurodegenerative conditions, including Parkinson's or Alzheimer's disease in humans


Is the source, species, dose, composition, characteristics, or processing of berries and berry constituents related to the effect of the intervention?


What adverse events in humans have been reported in the literature for the constituents in berries?


Do the frequency of adverse events vary with source, dose, or other evaluated factors?

Approach To Analyzing the Literature

To guide the assessment and synthesis of the literature, we used an expanded version of the generally-referred-to “PICO” method (Population, Intervention, Comparator, Outcomes) to define the parameters of interest. We used this approach for analysis of both human, animal, and in vitro studies. With input from the TEP, we asked the following questions to establish the literature review criteria:

  • What are the populations of interest?
  • What are the interventions of interest?
  • What are the comparators of interest?
  • What are the (marker/intermediate and clinical) outcomes of interest?
  • What are the health conditions of interest?
  • What are acceptable study designs?

Topic Refinement

In regards to both studies that examine putative mechanisms of action on neurodegenerative disorders and to studies that examine associations and effects in humans on neurodegenerative disorders, there is a very broad range of related topics that have been studied. In an iterative process, the EPC worked with the TEP to focus the questions and the topics on those that are most likely to shed light on mechanisms of action and effects related to Alzheimer's disease (AD), Parkinson's disease (PD) and related neurocognitive disorders. Thus this report does not evaluate all mechanisms of action or all associations related to neurological function. Given the very large number of studies (both human and animal) related to B vitamins, and the small number of studies of berries, these caveats apply primarily to B vitamin topics.

The following topics were chosen, in consultation with the TEP, for evaluation:

B Vitamins: Human Studies

  • Association between B vitamin treatment/intake with diagnosis of AD or PD, cognitive function, or histopathology (primary prevention of disease)
  • Association between B vitamin treatment/intake with severity of AD or PD, cognitive decline, or histopathology. (secondary prevention/treatment)
  • Association of B vitamin levels and AD or PD diagnosis, or histopathology
  • Association of B vitamin levels and AD or PD severity
  • Association of B vitamin levels and cognitive function

B Vitamins: Animal / In Vitro Studies

  • Effect of B vitamin supplementation or deficiency on cognitive function, movement disorders, histopathology, L-dopa and pre-cursor levels, etc., in appropriate models
  • Effect of B vitamins on the expression or function of AD-related genes (presenilin, alpha-2 macroglobulin, amyloid precursor protein, Apo E4)
  • Blood brain barrier function in relation to B vitamins
  • Cerebrovascular endothelial function in relation to B vitamins

Thus, the following potential topics (among others) are not reviewed: B vitamin-dependent enzyme levels or function; markers of inflammation or other potential causes of neurocognitive decline, including homocysteine, except as they relate to the association between B vitamins and neurocognitive status; B vitamin megadose-related toxicity; animal studies using B vitamin antagonists, brain lipid metabolism, animal perinatal and growth-related brain/nerve/cognition development; genes related to B vitamin function or enzymes such as MTHFR; GABA metabolism, or neuron ion channels.

Berries. Given the small size of the relevant literature, all studies evaluating the effect or association of berries or constituents of berries with any neurological or cognitive outcome were included.

Eligibility Criteria

This report encompasses evaluations of both clinical human studies and basic science studies performed in animal and in vitro models. Therefore, specific eligibility criteria were needed for each topic. We first describe the common eligibility criteria for any study included in this report, followed by additional specific criteria for each topic.

Human Studies. The common inclusion criteria for human studies analyzed in this report consist of primary studies; English language publication, human adult subjects; analysis of the predictor or description, including quantification, of the intervention, and analysis of the following categories of outcomes: diagnosis or severity (degree) of AD, PD, other age-related neurocognitive disorder, or cognitive impairment; test of cognitive function. We excluded studies of mental retardation, including Down syndrome, Wernicke's encephalopathy, subacute combined degeneration, vascular dementia, acute encephalopathy, and mixed causes of dementia lacking separate analyses for disease types. Also excluded were studies of peripheral neuropathy and other lower motor neurodegeneration not related to PD. However, studies that compared groups of patients with age-related neurocognitive disorders with groups of patients with other dementias were included. We also excluded case reports and studies of non-applicable populations, such as young patients with diabetes. Abstracts without an associated full report were excluded. Where studies were reported in multiple publications, the more completely reported and/or the report with the longer duration of follow-up were used; although data from multiple publications of the same study may be combined.

Animal / In Vitro Studies. Animal and in vitro studies had to be published in full form, excluding abstracts, in English language journals. We included all animal and in vitro models of diseases of interest and all outcome measurements related to the outcomes and/or associations of interest. We excluded studies that used inappropriate animal or in vitro models, such as immature animals and non-neuronal cells.

B Vitamin Topics

Common criteria. The following B vitamins were investigated:

  • B1 (thiamine)
  • B2 (riboflavin)
  • B6 (pyridoxine and related compounds)
  • B12 (cyanocobalamin)
  • Folate

We included evaluations of the single vitamins and combinations of the B vitamins. We excluded evaluations of “multivitamins” that included vitamins other than B vitamins. Evaluation of B vitamins could be from supplements (given by any route), food sources, or specific tissue concentrations. Evaluated body levels included blood, serum, plasma, cerebrospinal fluid, or tissue sample (including red blood cell) levels of the specific vitamins and commonly measured metabolites (i.e., pyridoxal-5′-phosphate, the active coenzyme form of B6, and thiamine pyrophosphate, the active coenzyme form of B1). We allowed any measurement methodology. We did not include other proxies for B vitamin levels (e.g., thiamine-dependent enzyme activity).

Human intervention studies (trials). We included only prospective trials of clearly defined B vitamin interventions. We allowed randomized controlled trials (RCTs), prospective non-randomized comparative trials, and prospective cohort studies (single arm studies without a control group). We allowed trials of both supplements and food sources. We excluded studies of the effect of B6 intake on Parkinsonian symptoms and L-dopa levels in patients using L-dopa treatment. (This issue is discussed in the adverse events section of the results.)

Human association studies. Among studies that reported associations between B vitamin levels and neurocognitive outcomes, we included only those that included subjects with either AD or PD, or neurocognitive impairment, excluding studies focusing on cognitively normal populations. All studies, regardless of sample size, were included regarding PD or vitamins B1, B2, or B6 levels. For cross-sectional studies of either B12 or folate levels and subjects with either AD or cognitive impairment, we included only studies that evaluated both at least 100 subjects total and 30 subjects with AD or cognitive impairment (not including vascular dementia, mental retardation, etc.). However, we included all longitudinal studies, regardless of sample size.

For studies evaluating B vitamin intake (i.e., by food frequency questionnaires), we included only studies with at least 50 subjects. We chose this arbitrary threshold to as a minimum number of subjects required to ensure adequate power for associations to be investigated in these retrospective studies. Studies of food intake (from food frequency questionnaires) must have had comparison groups of subjects with different levels of neurocognitive function. In addition we excluded cross-sectional intake studies that examined only dietary intake of patients with dementia. These studies evaluated nutritional deficiencies caused by poor diet due to dementia, which was not considered to be of interest.

For both human intervention and association studies, we did not include evaluations of outcomes related to depression, other psychiatric conditions, sleep, appetite, or other somatic conditions. We evaluated only diagnoses or measures of cognitive function or symptoms of PD.

Animal / in vitro studies. We excluded animal or in vitro models specific to Wernicke's encephalopathy; namely models of thiamine deficiency combined with ethanol. Although, if sufficient data regarding thiamine deficiency without ethanol was also included, these studies were reviewed. We also excluded animal and in vitro models that caused or exacerbated B vitamin deficiency with B vitamin antagonists. In addition, “case reports” or “case series” of B vitamin deficiencies in farm animals were excluded.

Berry Topics

Common criteria. After reviewing various definitions of berries and in consultation with the TEP, the following berries were included:

  • Bilberry
  • Black raspberry
  • Blackberry
  • Blueberry
  • Boysenberry
  • Cranberry
  • Currants
  • Gooseberry
  • Lingonberry
  • Marionberry
  • Raspberry
  • Strawberry

We recognized that these common terms for berries do not always match one-for-one with specific species. We allowed all fruits that are commonly designated among these berries. We included studies that used whole berries or specific constituents of berries. We did not include studies that evaluated constituents found in berries that were not derived from berries (e.g., purified quercetin).

Human studies. We included any study that examined the effect of or association between berries and any neurocognitive outcome in any population.

Animal / in vitro studies. We exclude studies using amphetamine- or lithium chloride-induced conditioned taste avoidance (CTA) as rats' learning or behavioral outcome. The CTA paradigm measures the avoidance by rats of a sucrose solution that has been paired with a high dose of a drug, such as amphetamine. The LiCl is used as a control. “Learned safety” theory is the mechanism of CTA results;31 it is not related to age-related cognitive or behavioral function.

Other Topics

Constituents in berries. Regarding the berry Key Question 1 on the constituents in berries related to neurological effects, we evaluated introduction and discussion sections from articles reviewed for other berry topics and also searched for both systematic and general reviews of the topic.

Adverse events. For both B vitamins and berries we included any adverse event data from otherwise evaluated human studies. We also reviewed other human studies that did not meet criteria for inclusion for other topics. In addition, we searched for both systematic and general reviews regarding adverse events in humans. We included all systematic reviews. General reviews were included on an ad hoc basis, depending on generalizability and adequacy of source material. We excluded adverse events related to pregnancy, children, contraception, cancer, and specific drug interactions (methotrexate, colon cancer chemotherapy, etc.). For berries, we also excluded allergies and issues related to food contaminants.

Literature Search Strategy

We conducted a comprehensive literature search to address the key questions.* Final literature searches for English language publications on B vitamins were conducted in MEDLINE® and the Commonwealth Agricultural Bureau (CAB) Abstracts™ on February 2, 2005 and for berries, in the same databases, on March 3, 2005. Search terms included subject headings and textwords with filters to limit the publications to English language. Subject headings and text words were selected so that the same set could be applied to both databases. The searches included both human, animal, and in vitro studies. Among the articles in MEDLINE®, specific article types were excluded, such as editorials, letters, and case reports, and other types that would not meet eligibility criteria.

Both the B vitamin and berry searches used a common neurocognitive model that included the following terms: nervous system diseases, cognitive disorders, neurodegeneration, dementia, Alzheimer, Parkinson, Lewy body, neuron/nerve cells, brain, and related terms.

The B vitamin search included both common and chemical names for all the B vitamins of interest. The berry search included both common and botanical names for all the berries of interest and the term “fruit,” excluding “fruit fly.” In addition, we included a list of 33 chemical terms for known berry constituents.

Additional studies were sought by contacting members of the TEP, and from reference lists of selected included articles and review articles and meta-analyses. Although the large majority of evidence regarding berries was from a single group of investigators, the decision was made with the TEP to maintain the restriction of eligible literature to published, peer-reviewed articles.

Study Selection and Data Extraction

All citations identified through the literature search were screened according to the inclusion criteria. A low threshold for acceptance was used at this stage to maximize the retrieval of potentially useful studies. Retrieved articles were evaluated against the complete inclusion criteria.

A single reviewer extracted each eligible study.* Data extraction problems were addressed during weekly meetings. Occasional sections were re-extracted to ensure that uniform definitions were applied across extracted studies. Problems and corrections were noted through spot checks of extracted data and during the creation of summary and evidence tables. A second reviewer independently verified the data in the summary tables using the original article.

Data Extraction

The same data extraction forms were used for both the B vitamin and berry articles.

Human Studies. Two data extraction forms were created for human studies; one for interventions, and one for associations. These forms were designed in the format of an evidence table to allow simple conversion to these tables. In both forms, items extracted included: factors related to study characteristics (study design, duration, country, setting, funding source), population (age, sex, race), eligibility criteria, definitions of neurocognitive disorders, study sample (number enrolled, number analyzed, reasons for dropout), descriptions of interventions or predictors and of outcomes, limitations, comments, and an assessment of both study quality and applicability (see below).

Intervention forms also captured results data related to baseline, follow-up, change, and net change in outcomes, along with standard deviation or standard error and statistical significance. Association forms captured results data related to mean outcome values of different groups, correlation values (r, odds ratio, relative risk, hazard ratio, etc.), and statistical significance of either differences or associations.

Animal and In Vitro Studies. Animal and in vitro studies are usually designed to examine the proposed mechanisms or pathways for the observed effects of a substance on defined diseases or conditions in humans. These studies are generally not meant to provide precise estimates of effects, but instead to test alternative hypotheses. Therefore, the process in reviewing animal and in vitro studies is different than reviewing human clinical or epidemiological studies. In contrast to traditional systematic reviews of human studies where large heterogeneity across studies related to different models and outcomes being examined can be problematic, in basic science studies heterogeneity (such as different models) across studies is essential to test and eliminate alternative hypotheses (such as different outcomes), so long as the central hypothesis (e.g., the physiological application) is related.

Thus, the goal of data extraction for these articles was not to extract the exact quantitative findings of each study. Instead, we extracted the following information to capture the concepts of importance. Namely,

  • What is the central hypothesis or stated purpose of the study?
  • What is the authors' assessment of the gap between what is known and unknown?
  • What is the working model used in the study?
  • What is the study design (including characteristics, intervention, comparator, and outcomes, sample size, duration)?
  • What are the measurements or outcomes?
  • What are the results and authors' conclusions?
  • What is the quality (including limitations) of the study?

Grading of the Evidence

Studies accepted in evidence reports have been designed, conducted, analyzed, and reported with varying degrees of methodological rigor and completeness. Deficiencies in any of these components can lead to biased reporting and interpretation of the results. While it is desirable to grade individual studies to highlight the degree of potential bias, the grading of study quality is a challenging process. Most factors commonly used in quality assessment of RCTs do not demonstrate a consistent relationship to estimates of treatment effects.32 Thus, there is still no uniform approach to grade studies. For human studies of both B vitamins and berries, our EPC has adopted the following approach, as used in previous evidence reports.

Methodological Quality Grade (Human Studies)

We used a 3-category grading system (A, B, C) to denote the methodological quality of each study. This grading system has been used in most of the previous evidence reports from the Tufts-NEMC EPC as well as in evidence-based clinical practice guidelines.33 This system defines a generic grading system that is applicable to varying study designs including RCTs, non-randomized comparative trials, cohort, and case-control studies:


Category A studies have the least bias and results are considered valid. A study that adheres mostly to the commonly held concepts of high quality including the following: clear description of the population, setting, interventions and comparison groups; sufficient power (arbitrarily defined as minimum sample size of 10 subjects); clear description of the content of the intervention or predictor used; appropriate comparator; appropriate measurement of outcomes; appropriate statistical and analytic methods and reporting; no reporting errors; less than 20% dropout; clear reporting of dropouts; and no obvious bias. Intervention trials must be double-blinded RCTs. Correlation analyses must use prospectively gathered data and must perform appropriate adjustment for potential confounders.


Category B studies are susceptible to some bias, but not sufficient to invalidate the results. They do not meet all the criteria in category A because they have some deficiencies, but none likely to cause major bias. The study may be missing information, making it difficult to assess limitations and potential problems.


Category C studies have significant bias that may invalidate the results. These studies have serious errors in design, analysis or reporting, have large amounts of missing information, or discrepancies in reporting. Specific criteria included large (>20%) or unequal dropout rate, large discrepancy in baseline and final numbers of subjects, non-randomized or single-cohort intervention studies, dissimilar baseline values among cohorts, unclear duration or numbers of subjects, missing baseline data, or irreconcilable apparent differences between data in figures, tables, and text.

In addition, cross-sectional association studies (between vitamin B level and either diagnosis or cognitive test score) that did not adjust for any potential confounders (i.e., performed only univariate analyses without relevant sub-analyses).

Methodological quality scoring was performed near the end of the review when we had the most experience and knowledge about the included studies. Each included study was graded by at least 2 people (with the exception of studies with major deficiencies, such as a non-comparative study design). When there were disagreements, 1 or 2 additional reviewers graded the studies and consensus was reached. Approximately half the studies had quality scoring by 3 or more reviewers.

Methodological Quality Grade (Animal / In Vitro Studies)

Although we used the same 3-category grading system (A, B, C) to denote the methodological quality of each study, the criteria used to assess the methodological quality of animal or in vitro studies are different from those used for human studies. Compared to human clinical trials, randomization of treatments and blinded analysis may be essential, but is often not applicable to animal or in vitro experiments. Therefore, we did not incorporate these factors into our quality grading system. This system defines a generic grading system that is applicable to both animal and in vitro studies:


Category A studies have the least bias and results are considered valid. A study should report comprehensive background information on animals or cell lines used. For animals, the information should include the animal source, strain, sex, age, body weight, housing condition (diet, light/dark cycle, number of animals per cage), and experimental environment (ambient temperature, time of day, and season). For cell lines, the information should include the origin, growth media, and experimental environment. The number of animals in the experiments and animals excluded from a study, and the reasons for their exclusion, must be reported. Controls should be contemporary and preferably be approximately equal in group size to the intervention groups. Treatments (e.g., the compositions of experimental and control diets) and outcome measures should be clearly defined and reported. Experimental models should be independent of each other. All experiments should have at least one repetition.


Category B studies are susceptible to some bias, but not sufficient to invalidate the results. They do not meet all the criteria in category A because they have some deficiencies, but none likely to cause major bias. The study may be missing information, making it difficult to assess limitations and potential problems.


Category C studies have significant bias that may invalidate the results. These studies have serious errors in design, analysis or reporting, have large amounts of missing information, discrepancies in reporting or irreconcilable apparent differences between data in figures, tables, and text.

Applicability Grade

Only human studies were assessed for applicability. For animal and in vitro studies, no assessment was made as to the applicability of the experimental model.

Applicability addresses the relevance of a given study to a population of interest. Every study applies certain eligibility criteria when selecting study subjects. Most of these criteria are explicitly stated (e.g., disease status, age, sex). Some may be implicit or due to unintentional biases, such as those related to study country, location (e.g., community vs. specialty clinic), or factors resulting in study withdrawals. The question of whether a study is applicable to a population of interest (such as Americans) is distinct from the question of the study's methodological quality. For example, due to differences in the background diets, an excellent study of Japanese men may be very applicable to people in Japan, but less applicable to Japanese American men, and even less applicable to African American men. The applicability of a study is thus dictated by the questions and populations that are of interest to those analyzing the studies.

In this report, the focus is on individuals at increased risk for, or diagnosed with, age-related neurocognitive disorders; in particular AD or PD. Even though a study may focus on a specific target population, limited study size, eligibility criteria, and the patient recruitment process may result in a narrow population sample that is of limited applicability, even to the target population. To address this issue, we categorized studies within a target population into 1 of 3 levels of applicability that are defined as follows:

  • Image fig134fu3.jpg Sample is representative of the target population. It should be sufficiently large to cover both sexes, an appropriate age range, and other important features of the target population (e.g., general health status). At least 30 subjects analyzed.
  • Image fig134fu2.jpg Sample is representative of a relevant sub-group of the target population, but not the entire population. Limitations include such factors as narrow age range, single ethnicity, setting that applies to only a portion of the population (e.g., nursing home). At least 10 subjects analyzed.
  • Image fig134fu1.jpg Sample is representative of a narrow subgroup of subjects only, and is of limited applicability to other subgroups. For example, a study of the oldest-old men or a study of a population on a highly controlled diet.

Reporting Results

Outcomes Reported

For both human intervention studies and animal / in vitro studies we evaluated all outcomes relevant to neurocognitive function that were reported in studies. However, for human association studies regarding cognitive function, in consultation with the TEP, we focused the detailed evaluation to a limited number of outcomes. A large number of tests of cognitive function have been used by different study groups. Few of these have been validated in any systematic way. Interpretation of tests used by single groups or that have not been validated can be problematic. Thus we evaluated in detail the following tests of cognitive function:

  • Mini-mental status examination (MMSE) and modifications
  • Alzheimer's Dementia Assessment Scale (ADAS)
  • Mattis' Dementia Rating Scale (DRS)
  • Wechsler Adult Intelligence Scale (WAIS)

Other cognitive tests are summarized qualitatively only. All relevant outcomes in studies of patients with PD are reported in detail, as are all associations between B vitamin levels and diagnoses of cognitive disorders.

Metrics Included

Human Intervention Trials. For controlled intervention trials the summary tables describe 3 sets of data: the mean baseline levels in both intervention and control arms, within-cohort changes (e.g., InterventionFinal - InterventionInitial), the net change of the outcome, and the reported P values of both the within-cohort change and the net change. The net change of the outcome is the difference between the change in the intervention arm and the change in the control arm:

Net change = (InterventionFinal - InterventionInitial) - (ControlFinal - ControlInitial). For non-controlled interventions, we report the within-cohort changes and P values. For both types of studies we did not calculate any P values, but, when necessary, used provided information on the 95% confidence interval or standard error (SE) of the net difference to determine whether it was less than 0.05. We included any reported P value less than 0.10; those above 0.10 and those reported as “non-significant” were described as “NS” (non-significant) in the tables.

Human Association Studies. For studies reporting mean B vitamin levels, mean cognitive function scores, or prevalence of disease in different groups of patients, these values are included in summary tables along with reported P values of differences among the groups. For studies that reported further analyses, such as odds ratio or relative risk, or correlation (r) between, for example, B vitamin level and cognitive test score, these values are reported, along with their statistical significance. When available, both unadjusted and adjusted values are included.

Animal / In Vitro Studies. Numerical results are not reported for the basic science studies. We aimed to capture the direction and the statistical significance of all outcomes. For each analysis we report a symbol for the effect and the statistical significance (when reported). We used the following symbols:

+ Normal B vitamin animals/tissue performed better than B vitamin-deficient animals/tissue, or

Berry-fed animals/tissue performed better than non-berry-fed animals/tissue

0 No difference in performance

- Normal B vitamin animals/tissue performed worse than B vitamin-deficient animals/tissue, or

Berry-fed animals/tissue performed worse than non-berry-fed animals/tissue

The assessment of whether animals or tissue receiving the intervention performed better than controls was made based on a combination of the reported results, the statistical significance, and the conclusions of the authors.

Units. For measures of B vitamin levels, the original units reported in the study were included in the evidence tables. However, all such measurements were converted to standard units (e.g., mg/dL) in the summary tables to facilitate comparisons.

Evidence and Summary Tables

The evidence table offers a detailed description of the studies that addressed each of the key questions. The evidence table is available via the internet.* The tables provides all the information that was extracted from each study (as described above, under Data extraction). Each study appears once regardless of how many interventions or outcomes were reported. The evidence tables of human studies are ordered alphabetically by the first author, then by publication date. The evidence tables of animal and in vitro studies are categorized by topic and ordered chronologically, so as to capture the sequence of the research.

Summary tables are included in each Results section. They succinctly report summary measures of the main outcomes evaluated. They include information regarding study duration (as applicable), study size, intervention and control, outcomes, results, methodological quality, and study applicability. They are designed to facilitate comparisons and synthesis across studies. Studies reporting multiple predictors (e.g., B vitamins) may appear several times in summary tables. Blank cells indicate that the relevant data were not reported in the articles.

Studies that did not report detailed reports of interest to this report are included in the summary tables. The qualitative results - whether a significant or non-significant association - are included either as a paraphrase of or direct quote from the authors.

Within summary tables of human studies, studies were ordered first by outcome test (for cognitive tests: MMSE, ADAS, DRS, WAIS, then others), then from highest quality to lowest, then from highest applicability to lowest, then from largest to smallest number of subjects. Summary tables of animal and in vitro studies are ordered chronologically.

Adverse Events Reporting

We used the term adverse event as defined by the World Health Organization (WHO) International Conference on Harmonization. An adverse event is “any untoward medical occurrence in a patient or clinical investigation subject administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment. An adverse event can therefore be any unfavorable and unintended sign (including an abnormal laboratory finding, for example), symptom, or disease temporally associated with the use of a medicinal product, whether or not considered related to the medicinal product.” An adverse drug reaction is any “noxious and unintended response to a medicinal product related to any dose...” (

We reviewed all accepted and rejected human studies of either B vitamins, berries, or berry constituents being used as an intervention for data on adverse events and drug interactions. These reports included randomized trials, cohorts, case-control studies, and individual case reports and series. We excluded allergies (except for anaphylaxis) and occupational exposures.

Since adverse event reporting was very limited among the reviewed studies, we also performed searches for both systematic reviews and review articles regarding adverse events due to either B vitamins or berries. We also performed a search of articles on berries that have been tagged by MEDLINE® or CAB Abstracts™ as addressing adverse events.



Appendixes cited in this report are provided electronically at http://www​​/pub/evidence/pdf/berry/berry​.pdf.


  • PubReader
  • Print View
  • Cite this Page

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...