Constructing the cyber-troll: Psychopathy, sadism, and empathy


Trolling is an online antisocial behaviour with negative psychological outcomes.

Current study predicted trolling perpetration from gender and personality.

Trolls more likely to be male with high levels of trait psychopathy and sadism

Trolls have lower affective empathy, and psychopathy moderates cognitive empathy.

Results have implications for establishing education and prevention programs.


Online trolling is of particular concern due to the harmful negative outcomes its victims experience. The current study sought to explore and extend the personality profile of Internet trolls. After gender was controlled for, psychopathy, sadism, and empathy (affective empathy, cognitive empathy, and social skills) were examined for their predictive utility of trolling behaviour. A sample of 415 participants (36% men, 63% women, 1% other) with a mean age of 23.37 years (SD = 7.19) completed an online questionnaire. Results showed that men were more likely than women to engage in trolling, and higher levels of trait psychopathy and sadism predicted trolling behaviour. Lower levels of affective empathy predicted perpetration of trolling, and trait psychopathy moderated the association between cognitive empathy and trolling. Results indicate that when high on trait psychopathy, trolls employ an empathic strategy of predicting and recognising the emotional suffering of their victims, while abstaining from the experience of these negative emotions. Thus, trolls appear to be master manipulators of both cyber-settings and their victims’ emotions.

Study: Metformin Linked to Higher Risk of Alzheimer’s and Parkinson’s

Metformin and Alzheimer's disease

A recent study found that the use of metformin in people with diabetes increased their risk for developing dementia and Parkinson’s Disease.

This may be surprising as not too long ago, we reported on a different study which found the opposite–that using metformin might lower the risk for dementia in older men.

The study from Taiwanese researchers was presented on March 29, 2017 at The 13th International Conference on Alzheimer’s and Parkinson’s Diseases in Vienna Austria by Dr. Yi-Chun Kuan from the Shuang Ho Hospital, Taipei Medical University, New Taipei City, Taiwan.

The researchers found that long-term use of metformin may raise the risk of neurodegenerative disease in those with type 2 diabetes.

How Harmful Might Metformin Be to the Brain?

As reported by Medscape Medical News, Yi-Chun Kuan and team conducted a cohort study to follow a total 9,300 patients with type 2 diabetes in Taiwan for up to 12 years. They checked records for these patients from the National Health research database of Taiwan including 4,651 who had metformin prescriptions and 4651 matched controls who didn’t take any metformin.

Dr. Kuan told Medscape they adjusted for age, sex, and diabetes severity and that despite this, “the cumulative incidences of Parkinson’s and dementia were significantly higher for our metformin cohort” at 12 years.

In fact, the risk for Parkinson’s disease or Alzheimer’s dementia went up over 50 percent during a 12 year period in those who took metformin when compared to those who did not. Researchers also found that “outcome risks increased progressively with higher dosage and longer duration of treatment.”

Dr. Yi-Chun Kuan said, “We’d heard about a possible protective effect from metformin. However, we found the reverse,” and she added that large-scale, prospective studies would need to be done in other countries to get clarification of the results.

Another detail the researchers noted was that outcomes increased the longer a patient was on metformin and the higher the metformin dose they took, “especially with use for more than 300 days and doses greater than 240 g.”

A limitation of the study was that the patients on metformin might also be taking other diabetes drugs like insulin or sulfonylureas and Dr. Kuan said that she and her team would like to follow up on these other possible associations.

There was also no word on what the patient’s HbA1c levels were to help indicate the state of diabetes management nor an explanation on how factors were controlled for, as medical consultant, Dr. Larry Ereshefsky told Medscape.

Could a B12 Deficiency Have Anything to Do With It?

Recently, metformin has been shown to possibly cause B12 deficiency, particularly in those who take it longterm. One of the side effects of a B12 deficiency is neuropathy, or nerve damage.

A serious B12 deficiency has been known to also cause side effects like cognitive difficulties such as memory loss.

While this study still needs a follow-up, if you are concerned about metformin or your B12 levels, talk to your health care provider who can provide testing and if needed, guidance on how to get your B12 levels up.

Vitamin D Is More Effective Than Flu Vaccine, Study Says.

According to Dr Joseph Mercola and others, vitamin D deficiency may actually be behind your winter flu.

As a result, researchers have begun examining if increased vitamin D might make an effective flu vaccine, and, as it turns out, it does!
n fact, the research found several notable returns:
  • People with the lowest vitamin D levels show significantly more colds and/or cases of the flu.
  • For people with the lowest vitamin D levels, taking a vitamin D supplement cut the risk of respiratory infection by 50% – though researchers also noted that vitamin D was better absorbed through sunlight.
  • Vitamin D proved slightly more effective than the traditional flu vaccine (1 case prevented per 33 people vs 1 case prevented per 40 people), though this was more pronounced in trial participants with severe vitamin D deficiency.
  • Vitamin D is significantly better absorbed from sun exposure than from supplements.
  • Mounting research suggests vitamin D deficiency may actually be a major cause of influenza. People with the lowest vitamin D levels report having significantly more colds or cases of the flu
  • Scientific review confirms vitamin D optimization boosts immunity and cuts rates of cold and flu. Among people vitamin D blood levels below 10 ng/mL, taking a supplement cut risk of respiratory infection by 50 percent
  • To prevent influenza in one person, 40 people must receive the flu vaccine whereas one case of the flu can be prevented for every 33 people taking vitamin D. If you’re severely vitamin D deficient, vitamin D supplementation is 10 times more effective than the flu vaccine


Maternal Retinoids Increase PDGFRα+ Progenitor Population and Beige Adipogenesis in Progeny by Stimulating Vascular Development.


Maternal vitamin A supplementation increases blood vessel density and expands adipose progenitor population in progeny.

Maternal vitamin A supplementation enhances brown-like phenotype in adipose tissues.

Maternal vitamin A supplementation protects offspring from diet induced obesity.

Vitamin A and its metabolite, retinoic acid, play key roles in adipogenesis and energy expenditure of adipose tissues. In mice and humans, vitamin A intake is inversely correlated with adiposity. This study has uncovered a role for maternal retinoids in fetal adipose development. Maternal vitamin A supplementation or RA administration increases adipose progenitor population and promotes beige adipogenesis, which protects offspring from diet induced obesity in later life.


Maternal vitamin A intake varies but its impact on offspring metabolic health is unknown. Here we found that maternal vitamin A or retinoic acid (RA) administration expanded PDGFRα+ adipose progenitor population in progeny, accompanied by increased blood vessel density and enhanced brown-like (beige) phenotype in adipose tissue, protecting offspring from obesity. Blockage of retinoic acid signaling by either BMS493 or negative RA receptor (RARαDN) over-expression abolished the increase in blood vessel density, adipose progenitor population, and beige adipogenesis stimulated by RA. Furthermore, RA-induced beige adipogenesis was blocked following vascular endothelial growth factor receptor (VEGFR) 2 knock out in PDGFRα+ cells, suggesting its mediatory role. Our data reveal an intrinsic link between maternal retinoid level and offspring health viapromoting beige adipogenesis. Thus, enhancing maternal retinoids is an amiable therapeutic strategy to prevent obesity in offspring, especially for those born to obese mothers which account for one third of all pregnancies.


In this study, we found that maternal retinoid supplementation profoundly enhances brown/beige adipogenesis during fetal development, which has long-term effect on BAT and beige phenotype in offspring, and protects offspring from diet-induced obesity. This is an exciting discovery considering the easiness of vitamin A supplementation and the wide existence of vitamin A deficiency worldwide (WHO, 2009), particularly in low income countries. Low income correlates with obesity and metabolic diseases (Pan et al., 2013).

We identified that maternal retinoid status affects fetal and offspring brown/beige adipogenesis through promoting angiogenesis. Maternal vitamin A or RA supplementation enhances angiogenesis through upregulating Vegfa and Vegfr2expression, which consequently increased the population of PDGFRα+ adipose progenitor cells in adipose tissue. Our data are consistent the enhanced angiogenesis in both white and brown adipose tissues of mice exposed to cold stimulus, where Vegfr2blockage abolishes the cold-induced angiogenesis and impairs nonshivering thermogenesis capacity ( Xue et al., 2009). Although it remains controversial whether beige adipocytes are generated by de novo adipogenesis or conversion of existing adipocytes ( Rosenwald et al., 2013 ;  Wang et al., 2013), progenitor cells on endothelial vessels are capable to differentiate into beige adipocytes ( Tran et al., 2012; Min et al., 2016 ;  Vishvanath et al., 2016). Obesity and diabetes impair the angiogenic potential of adipose tissue stem cells ( Rennert et al., 2014 ;  Togliatto et al., 2016), and stimulation of angiogenesis via Vegfa over-expression promotes adipose tissue thermogenesis and protects against diet-induced obesity ( Wu et al., 2011 ;  Sun et al., 2014a). Thus, enhancing angiogenesis is an effective strategy to promote beige adipogenesis in adipose tissue impaired due to obesity and diabetes.

Besides stimulating angiogenesis, our data also show that maternal vitamin A or RA supplementation up-regulates brown/beige adipogenesis of progenitor cells, which is associated with enhanced expression of Prdm16 and other brown adipocyte genes. Using mice with conditional Vegfr2 knockout specifically in PDGFRα+ progenitor cells, the promotion effect of RA on brown adipogenesis was severely reduced, demonstrating an angiogenesis independent effect of RA on beige adipogenesis. These observations are consistent with previous studies showing that the RA increases oxidation and energy consumption of white adipose tissue in mature animals ( Alvarez et al., 1995; Puigserver et al., 1996; Bonet et al., 2003 ;  Mercader et al., 2006). Thus, maternal RA stimulates both angiogenesis and beige adipogenesis during early development, which are mediated by RAR because its functional knockout blocks brown/beige adipogenesis.

Besides beige adipogenesis, increasing the progenitor pool in adipose tissues through angiogenesis has another advantage. It is known that PDGFRα+ progenitor cells are the source of both beige and white adipocytes (Lee et al., 2012; Lee et al., 2013 ;  Lee et al., 2015). Thus, enhancing PDGFRα+ progenitor cell pool will not only increase beige but also white adipogenesis, as shown by the increased expression of white preadipose genes in MVA WAT of this study. Adipose tissue is the organ to store fat, and an insufficient number of adipocytes leads to adipocyte hypertrophy, hypoxia and inflammation, a key cause of metabolic dysfunction (Rosen and Spiegelman, 2014). Thus, adipocyte hyperplasia has protective effects on metabolic dysfunction induced by excessive energy intake. Consistently, there is one subgroup of people who are metabolically health despite being obese, while others exhibit severe metabolic syndromes (Denis and Obin, 2013). People who are called “metabolically healthy obese” (MHO) tend to have smaller adipocytes (Kloting et al., 2010) and higher mitochondrial transcription (Naukkarinen et al., 2014). These individuals have reduced visceral adiposity, reduced inflammation, improved glucose and lipid homeostasis when compared to other equally obese unhealthy subjects (Denis and Obin, 2013). Based on our discovery, maternal vitamin A or RA supplementation increases the progenitor pool in offspring, which reduces average adipocyte sizes and increases adipocyte hyperplasia, improving overall metabolic health of offspring.

In conclusion, offspring adipose tissue health is substantially improved due to maternal vitamin A or RA supplementation. Maternal vitamin A promotes vascular system development, which consequently increases the population of PDGFRα+ adipose progenitor cells. In addition, maternal vitamin A supplementation strongly upregulates beige adipogenesis of PDGFRα+ progenitor cells. In combination, maternal vitamin A treated offspring have increased beige adipogenesis and smaller adipocyte sizes, which protect offspring against diet-induced obesity and metabolic dysfunction.

Detection of Atherosclerotic Inflammation by 68Ga-DOTATATE PET Compared to [18F]FDG PET imaging.


Background Inflammation drives atherosclerotic plaque rupture. Although inflammation can be measured using fluorine-18-labeled fluorodeoxyglucose positron emission tomography ([18F]FDG PET), [18F]FDG lacks cell specificity, and coronary imaging is unreliable because of myocardial spillover.

Objectives This study tested the efficacy of gallium-68-labeled DOTATATE (68Ga-DOTATATE), a somatostatin receptor subtype-2 (SST2)-binding PET tracer, for imaging atherosclerotic inflammation.

Methods We confirmed 68Ga-DOTATATE binding in macrophages and excised carotid plaques. 68Ga-DOTATATE PET imaging was compared to [18F]FDG PET imaging in 42 patients with atherosclerosis.


Results Target SSTR2 gene expression occurred exclusively in “proinflammatory” M1 macrophages, specific 68Ga-DOTATATE ligand binding to SST2 receptors occurred in CD68-positive macrophage-rich carotid plaque regions, and carotid SSTR2 mRNA was highly correlated with in vivo 68Ga-DOTATATE PET signals (r = 0.89; 95% confidence interval [CI]: 0.28 to 0.99; p = 0.02). 68Ga-DOTATATE mean of maximum tissue-to-blood ratios (mTBRmax) correctly identified culprit versus nonculprit arteries in patients with acute coronary syndrome (median difference: 0.69; interquartile range [IQR]: 0.22 to 1.15; p = 0.008) and transient ischemic attack/stroke (median difference: 0.13; IQR: 0.07 to 0.32; p = 0.003). 68Ga-DOTATATE mTBRmax predicted high-risk coronary computed tomography features (receiver operating characteristics area under the curve [ROC AUC]: 0.86; 95% CI: 0.80 to 0.92; p < 0.0001), and correlated with Framingham risk score (r = 0.53; 95% CI: 0.32 to 0.69; p <0.0001) and [18F]FDG uptake (r = 0.73; 95% CI: 0.64 to 0.81; p < 0.0001). [18F]FDG mTBRmax differentiated culprit from nonculprit carotid lesions (median difference: 0.12; IQR: 0.0 to 0.23; p = 0.008) and high-risk from lower-risk coronary arteries (ROC AUC: 0.76; 95% CI: 0.62 to 0.91; p = 0.002); however, myocardial [18F]FDG spillover rendered coronary [18F]FDG scans uninterpretable in 27 patients (64%). Coronary 68Ga-DOTATATE PET scans were readable in all patients.

Conclusions We validated 68Ga-DOTATATE PET as a novel marker of atherosclerotic inflammation and confirmed that 68Ga-DOTATATE offers superior coronary imaging, excellent macrophage specificity, and better power to discriminate high-risk versus low-risk coronary lesions than [18F]FDG. (Vascular Inflammation Imaging Using Somatostatin Receptor Positron Emission Tomography)


We provide gene-, cell-, plaque-, and patient-level data demonstrating that SST2 PET imaging using 68Ga-DOTATATE provides a quantifiable, cell-specific marker of atherosclerotic inflammation that outperforms [18F]FDG in the coronary arteries. Further work is needed to confirm these findings in a larger patient population and to compare imaging with clinical outcomes. 68Ga-DOTATATE PET offers measurement of both generalized atherosclerotic disease activity and detailed information about local plaque functional phenotype to complement multimodal assessments of anatomic, morphologic, and hemodynamic disease severity. This approach, in selected patient populations, has the potential to improve CVD risk prediction, allowing personalized tailoring of therapies aimed to improve clinical outcomes.

The Big Fat Lie You’ve Been Told About What’s Hurting Your Heart

Despite multiple studies showing that carbohydrates hurt your heart, and not saturated fats, misguided advisories and Big Pharma profiteering both persist.

There is no need to stay away from meat, butter, cheese and eggs to keep your heart healthy. Credit: RitaE/pixabay

There is no need to stay away from meat, butter, cheese and eggs to keep your heart healthy.

I’ve been taught since my undergraduate days in medical college that eating saturated fats was to ask for trouble. Meat (red or white), cheese, butter and egg yolk were prohibited. Repeated guidelines from the American Heart Association (AHA), the American College of Cardiology and even the World Health Organisation were clear that fats in general, and saturated fats in particular, were to be strictly avoided to prevent a heart attack. The message was to reduce fats to less than 30% of the total calories consumed in a day, and with saturated fats to be kept well below 10%. Why, most people on the planet followed these dietary commandments from the two most powerful and respected cardiology associations.

The AHA declared in 1961 that saturated fats were bad because they increased blood cholesterol, which blocked coronary arteries and caused heart attacks. Surprisingly, the AHA was driven this conclusion by the hypothesis of one physiologist who didn’t bother to submit a shred of evidence. Ancel Benjamin Keys, a physiologist with a PhD from Cambridge University, stamped his ‘diet heart’ hypothesis into the consciousness of Paul Dudley White, a founder-member of the AHA. White was attending to Dwight Eisenhower, then the US president, who suffered his first heart attack in September 1955. Many middle aged Americans were succumbing to heart attacks in the 1950s and the situation demanded convincing answers from the health community. Eisenhower had helmed NATO and, before that, had been the supreme commander of the Allied forces that wrenched Europe back Europe from the Nazis.

Eisenhower managed the brilliant generals George Patton and Bernard Montgomery, and famously warned the American public in his farewell address of the “military-industrial complex”. But as president, he had no clue of the new and rapidly developing “health-pharmaceutical-industrial complex.”

Keys was able to launch his ‘diet heart’ hypothesis because there was little science available in the 1950s that could explain the near-epidemic heart attack among middle-aged Americans. He presented his “seven countries study” that displayed a clear association between eating greater amounts of saturated fats and deaths due to heart disease. The seven countries were the US, Japan, Yugoslavia, Netherlands, Italy, Greece and Finland. The method behind the study was seriously flawed, however.

The biggest was that Keys had cherry-picked these countries because they supported his hypothesis. He left out 15 countries that did not reveal any association between saturated-fat consumption and heart mortality. He conveniently ignored Denmark, Sweden and Norway, each of which had relatively few deaths from heart attacks in spite of sporting diets with lots of saturated fats. And Chile, on the other hand, had a high cardiac mortality despite eating little saturated fats. An unbiased investigator would have realised these problems in Keys’ hypothesis – as they do now – but didn’t: they hadn’t been presented with the complete data.

Keys also checked food samples for fats in less than 4% of the 12,000 participants he studied, and when the food was studied it was checked for a single day among the American and for less than a week among the European participants. Keys had been impressed by the large number of long-lived people on the Greek island of Crete, but had tested them when they’d been fasting for more than a month during a religious festival. In this period, more than 60% of the population abstained from meat, butter and cheese. This led Keys to the wrong conclusion that a low-fat diet was the key to longevity.

The AHA was so impressed by the ‘diet heart’ hypothesis that it made an official policy of it, and voila! By 1977, more than 220 million Americans were being urged by the US government to adhere to a low-fat diet. The British, true to form, officially imposed the same diet guidelines by 1984 on their subjects.

Remarkably, the AHA ignored no fewer than six randomised studies – including almost 2,500 heart patients – that showed no difference in mortality between the intervention group (low saturated-fat diet) and the control group (which continued with its regular eating habits). Both the intervention and control cohorts had 370 deaths each. Moreover, no women were being studied, and in the absence of a single primary prevention trial, the AHA and the US government had formulated their advisories.

The food industry also got in on the action. Vegetable oils started being manufactured in the millions of tons. Leading them all was Procter and Gamble, which began to aggressively market cottonseed oil – as well as make a sizeable donation to the AHA, an amount worth $20 million today. The corresponding “diet-food-health-industrial complex” has not looked back in the 60 years since.

The largest randomised trial assessing the effects of a low-fat diet on heart and cardiovascular diseases was the Women’s Health Initiative. It followed up 49,000 postmenopausal women who had been on a low-fat diet (alongside an increased intake of fruits, vegetables and grains) for eight years but had failed to lower their risks of death, heart attack, stroke or diabetes.

Two large reviews and meta-analyses (this and this) involving more than 600,000 participants have also failed to show any reduction in cardiovascular events, or death, by replacing saturated fats with vegetable oils. There was an increase in cardiovascular events due to trans-fats.

The Minnesota, DIRECT, Framingham and PURE studies

In 1967-1973, doctors intervened in the diets of a group of people randomly picked from a cohort of 9,000 for the famous Minnesota Coronary Experiment. The intervened group had saturated fats replaced by a polyunsaturated vegetable oil. The control group continued with their regular American diet. These people were from enrolled from mental institutions and from homes for the elderly. More than 2,500 participants continued on their respective diets for at least a year, and autopsy reports were available for about 140 deaths. This trial’s results were never published until a group of investigators got its hands on all the raw data.

They were dumbstruck to learn that the autopsies revealed 42% of the people in the intervention group had suffered a heart attack against only 22% in the control group. Both groups had similar amounts of atherosclerosis in their coronary arteries.

The other major finding was that, in spite of a 13% reduction in blood cholesterol with a vegetable-oil diet, there was a paradoxical 30% higher mortality in people older than 65 years. To explain this, the investigators hypothesised that the lowered cholesterol had the denser LDL particles that are oxidised more easily and so invade the coronary faster. As it happened, the principal investigator of the Minnesota Coronary Experiment was none other than Ancel Keys.

The other distinct possibility (to explain the mortality paradox) is that polyunsaturated vegetable oils produce hundreds of oxidised molecules that are toxic to the human body. For example, the aldehydes are carcinogenic apart from being able to compromise cognition. Another randomised trial assessing the replacement of saturated fats by corn oil also showed an increased mortality against the control group.

More recently, the DIRECT trial finished up in Israel in 2008. It divided participants into three groups. The first was kept on a low-fat diet; the second, a Mediterranean diet; and the third, a low-carbohydrate high-fat diet. At the end of follow-up period, the low carbohydrate high fat group was found to have lost the most weight, have the highest levels of HDL (a.k.a., ‘good cholesterol’) and have triglyceride levels lower than the high-fat group. In fact, the low-carbohydrate high-fat group also had better metabolic markers across the board.

The Framingham study, which began in 1948 and still continues, has been following the consumption of dietary fats and the development of heart disease among its 5000+ inhabitants, chosen from Framingham, Massachusetts. At the end of the first follow-up, the investigators were unable to find any correlation between fat-intake, cholesterol and heart disease.

But like with the Minnesota Coronary Experiment, the data was never deliberately published. In William Kannel, who served as the study’s the chief investigator in 1969-1979, at one point even stated: “That blood cholesterol is somehow intimately related to coronary atherosclerosis is no longer subject to reasonable doubt.” After a 30-year follow-up, the study reported that 1 mg% per year reduction in cholesterol was associated with 14% increased cardiovascular mortality and 11% total mortality.

Finally: the Prospective Urban and Rural Epidemiological (PURE) survey examined cardiovascular risk factors around the world in 2003-2009, with more than 150,000 participants. Though the results are yet to be published, a recently leaked (and now unavailable) video stated that there seemed to be no correlation between saturated fats (red meat, white meat, dairy products) and heart disease but a positive correlation between carbohydrates and heart disease. Moreover, a very sensitive cardiac-risk-factor marker was found to have increased with carbohydrates and reduced by saturated fats. Vegetables and fruits had no effect on the marker.

Though the PURE trial was very large, it was an observational that, strictly speaking, can’t explain causality.

So, based on the evidence obtained from well-conducted clinical trials, Keys’s ‘diet heart’ hypothesis is wrong. However, it remains to be seen when the big cardiac bureaucracies will begin to edit their guidelines. The ‘big cholesterol is bad’ maxim remains firmly in place because its persistence allows drugmakers to persist with large profit margins on drugs that may not even be necessary. Precisely this was confirmed by the FOURIER trial presented in the American College of Cardiology Meeting held in March 2017.

FOURIER was a ‘mega-trial’ that randomised 28,000 cardiac patients to a statin-plus-evelocumab versus a only-statins for two years. The annual cost of an evelocumab regime is $14,000 (Rs 9 lakh). In the end, LDL cholesterol levels had plunged to about 30 mg% in the evelocumab group versus about 90 mg in the only-statins group. There was also a 1.5% absolute reduction in stroke and myocardial infarction risks but – get this – no reduction in mortality. Implication: 75 patients will need to be treated for two years to prevent a single heart attack or stroke, at a total cost of Rs 13.5 crore. You’re likely to get a better deal without spending a penny by following the Copenhagen study: 10 minutes of slow-jogging per day reduced mortality by 70% compared to being sedentary the whole day.

It’s difficult to not feel dizzy when confronted by organisations like the AHA and the WHO, which have converted hypotheses into dogma etched on stone without any evidence in the past. But what then would be good and sane dietary advice to a layperson? There has to be an application of common-sense, a request to continue to eat fruits, vegetables, whole grains and nuts. At the same time, there is no need to stay away from meat, butter, cheese and eggs. There is no evidence that eating saturated fats along with reasonable amounts of proteins, with about 45% of calories as carbohydrates will, trigger a heart attack. Au contraire: evidence has emerged that increasing carbohydrates to 55% or more can actually be harmful to the heart. Even the current obesity epidemic and type-2 diabetes are most likely the handiwork of an increased carbohydrate intake that has replaced fats in people’s diets.

Milk intake and risk of mortality and fractures in women and men: cohort studies

A diet rich in milk products is promoted to reduce the likelihood of osteoporotic fractures. Milk contains 18 of 22 essential nutrients, including calcium, phosphorus, and vitamin D of especial importance for the skeleton. Intestinal uptake of these nutrients is enhanced by the enzymatic capacity to digest lactose into D-glucose and D-galactose by mutation in the lactase gene, a variant common in those with northern European ancestry.1 2 An intake of dairy foods corresponding to three or four glasses of milk a day has been suggested to save at least 20% of healthcare costs related to osteoporosis.3

A high intake of milk might, however, have undesirable effects, because milk is the main dietary source of D-galactose. Experimental evidence in several animal species indicates that chronic exposure to D-galactose is deleterious to health and the addition of D-galactose by injections or in the diet is an established animal model of aging.4 5 6 7 Even a low dose of D-galactose induces changes that resemble natural aging in animals, including shortened life span caused by oxidative stress damage, chronic inflammation, neurodegeneration, decreased immune response, and gene transcriptional changes.5 7 A subcutaneous dose of 100 mg/kg D-galactose accelerates senescence in mice.5 This is equivalent to 6-10 g in humans, corresponding to 1-2 glasses of milk. Based on a concentration of lactose in cow’s milk of approximately 5%, one glass of milk comprises about 5 g of D-galactose. The increase of oxidative stress with aging and chronic low grade inflammation is not only a pathogenetic mechanism of cardiovascular disease and cancer in humans8 9 but also a mechanism of age related bone loss and sarcopenia.9 10 The high amount of lactose and therefore D-galactose in milk with theoretical influences on processes such as oxidative stress and inflammation makes the recommendations to increase milk intake for prevention of fractures a conceivable contradiction.

Because of the high content of lactose in milk, we hypothesised that high consumption of milk may increase oxidative stress, which in turn affects the risk of mortality and fracture. Meta-analyses of cohort studies for the association between dairy and milk intake in relation to mortality11 and fractures12 13 have displayed no clear pattern of risk, and evidence from randomised trials are lacking. Separating milk intake from the consumption of other dairy products may be of importance since a less pronounced induction of oxidative stress and inflammation in humans is expected with cheese and fermented dairy products (for example, soured milk and yogurt) because of their lower or non-existent lactose and galactose content,14 15 possible probiotic antioxidant and anti-inflammatory effects,16 17 18 and effects on gut microbiota.19 20 21 Indeed, a high intake of fermented milk products has been associated with a decreased risk of cardiovascular diseases,18 22 23 24 whereas a high milk intake is related to a tendency of an unfavourable risk profile for the development of diabetes and cardiovascular disease.18 23 24 We therefore assessed the relation between high milk intake with risk of death and fractures in women and men. We also studied biological markers of oxidative stress and inflammation in relation to milk intake in humans.



Objective To examine whether high milk consumption is associated with mortality and fractures in women and men.

Design Cohort studies.

Setting Three counties in central Sweden.

Participants Two large Swedish cohorts, one with 61 433 women (39-74 years at baseline 1987-90) and one with 45 339 men (45-79 years at baseline 1997), were administered food frequency questionnaires. The women responded to a second food frequency questionnaire in 1997.

Main outcome measure Multivariable survival models were applied to determine the association between milk consumption and time to mortality or fracture.

Results During a mean follow-up of 20.1 years, 15 541 women died and 17 252 had a fracture, of whom 4259 had a hip fracture. In the male cohort with a mean follow-up of 11.2 years, 10 112 men died and 5066 had a fracture, with 1166 hip fracture cases. In women the adjusted mortality hazard ratio for three or more glasses of milk a day compared with less than one glass a day was 1.93 (95% confidence interval 1.80 to 2.06). For every glass of milk, the adjusted hazard ratio of all cause mortality was 1.15 (1.13 to 1.17) in women and 1.03 (1.01 to 1.04) in men. For every glass of milk in women no reduction was observed in fracture risk with higher milk consumption for any fracture (1.02, 1.00 to 1.04) or for hip fracture (1.09, 1.05 to 1.13). The corresponding adjusted hazard ratios in men were 1.01 (0.99 to 1.03) and 1.03 (0.99 to 1.07). In subsamples of two additional cohorts, one in males and one in females, a positive association was seen between milk intake and both urine 8-iso-PGF2α (a biomarker of oxidative stress) and serum interleukin 6 (a main inflammatory biomarker).

Conclusions High milk intake was associated with higher mortality in one cohort of women and in another cohort of men, and with higher fracture incidence in women. Given the observational study designs with the inherent possibility of residual confounding and reverse causation phenomena, a cautious interpretation of the results is recommended.


A higher consumption of milk in women and men is not accompanied by a lower risk of fracture and instead may be associated with a higher rate of death. Consequently, there may be a link between the lactose and galactose content of milk and risk as suggested in our hypothesis, although causality needs be tested using experimental study designs. Our results may question the validity of recommendations to consume high amounts of milk to prevent fragility fractures.3 71 72 The results should, however, be interpreted cautiously given the observational design of our study. The findings merit independent replication before they can be used for dietary recommendations.

What is already known on this topic

  • A high milk intake is recommended for the prevention of osteoporotic fractures

  • Milk is the major dietary source of galactose intake

  • The addition of galactose by injection or in the diet is an established animal model of aging by induction of oxidative stress and inflammation

  • Results of previous research on the importance of milk intake for the prevention of fractures and the influence on mortality rates are conflicting

What this study adds

  • A high milk intake in both sexes is associated with higher mortality and fracture rates and with higher levels of oxidative stress and inflammatory biomarkers

  • Such a pattern was not observed with high intake of fermented milk products


Sexually Dimorphic Changes of Hypocretin (Orexin) in Depression.


Hypocretin (orexin) changes were studied in human postmortem brain in depression.

A clear sex-related change was found in the hypothalamic hypocretin-1-immunoreactivity in depression.

A rat depression model did not reflect the changes in the hypocretin system in the human brain in depression.

The stress systems of depressed patients are put into a higher gear by genetic and developmental factors. Over-reaction of these systems to stressful environmental situations makes people vulnerable to depression and suicide. This is the first postmortem study on changes in a relatively novel stress system in depression, consisting of the hypothalamic hypocretin neurons and hypocretin receptors in the prefrontal cortex. A clear sex-related change was found in the hypothalamic hypocretin-1-immunoreactivity in depression. Evaluation of the hypocretin system in a frequently used depression animal model, i.e. chronic unpredictable mild stress rats, did not replicate changes found in the hypocretin systems in the human brain in depression.



Neurophysiological and behavioral processes regulated by hypocretin (orexin) are severely affected in depression. However, alterations in hypocretin have so far not been studied in the human brain. We explored the hypocretin system changes in the hypothalamus and cortex in depression from male and female subjects.


We quantified the differences between depression patients and well-matched controls, in terms of hypothalamic hypocretin-1 immunoreactivity (ir) and hypocretin receptors (Hcrtr-receptors)-mRNA in the anterior cingulate cortex (ACC) and dorsolateral prefrontal cortex. In addition, we determined the alterations in the hypocretin system in a frequently used model for depression, the chronic unpredictable mild stress (CUMS) rat.


i) Compared to control subjects, the amount of hypocretin-immunoreactivity (ir) was significantly increased in female but not in male depression patients; ii) hypothalamic hypocretin-ir showed a clear diurnal fluctuation, which was absent in depression; iii) male depressive patients who had committed suicide showed significantly increased ACC Hcrt-receptor-2-mRNA expression compared to male controls; and iv) female but not male CUMS rats showed a highly significant positive correlation between the mRNA levels of corticotropin-releasing hormone and prepro-hypocretin in the hypothalamus, and a significantly increased Hcrt-receptor-1-mRNA expression in the frontal cortex compared to female control rats.


The clear sex-related change found in the hypothalamic hypocretin-1-ir in depression should be taken into account in the development of hypocretin-targeted therapeutic strategies.


Our study shows, for the first time in postmortem human brain, that hypothalamic hypocretin/orexin is increased in female – but not in male – depressive patients. In addition, there was a diurnal fluctuation in hypothalamic hypocretin-1-ir in the control subjects, which was absent in depression. Moreover, we observed that Hcrt-receptor-mRNA expression showed differences in the ACC and DLPFC depending on age. Male depressive patients who had committed suicide had significantly increased ACC Hcrt-receptor-2-mRNA expression. Our data thus indicate sex-, brain area-, age-, and potentially suicide-related changes in the hypocretin/orexin system in depression. Finally, a significant positive correlation between hypothalamic CRH-mRNA and prepro-hypocretin-mRNA and a significant increase in Hcrt-receptor-1-mRNA expression in the frontal cortex in female – but not male – CUMS rats strengthen the presence of sexually dimorphic hypocretin/orexin system changes in mood disorder.

4.1. Hypocretin-ir in the Hypothalamus in Depression

It should be noted that the increased IOD of hypocretin-ir may indicate an increase in either hypocretin-expressing neuron number (related to the area stained) and/or staining intensity (measured as OD). An increase of either of these parameters indicates an increased expression of hypocretin protein levels. The significantly increased hypocretin-ir in female depressive patients indicates that hypocretin may play a key role in the etiology of depression, which is more prevalent in females than in males (Piccinelli and Wilkinson, 2000). As we have indicated in the Introduction section, studies have found that Hcrt-receptor-1 gene, or a linked locus, may modulate the risk for mood disorders (Rainero et al., 2011) and Hcrt-receptor-1 gene knockout mice showed increased anxiety-like behavior and altered depression-like behaviors (Abbas et al., 2015). One may thus speculate that the higher levels of hypocretin-1-ir in female patients may enhance depressive symptoms. It should be noted that since all the female subjects studied in our study were in their postmenopausal stage, one would not expect to see the hot flash-related hypocretin changes reported earlier (Federici et al., 2016).

The sex difference in the alterations of hypocretin, which happens in the framework of the stress-hypothesis, was further supported by our animal study, showing a significant positive correlation between hypothalamic prepro-hypocretin-mRNA and CRH-mRNA only in female CUMS rats. In view of this, it is of interest to note that the dual Hcrt-receptor antagonist almorexant was found to prevent HPA axis dysregulation caused by CUMS and offers evidence for the possibility that pharmacological blockade of the hypocretin system induces a robust antidepressant-like effect as well as the restoration of the stress-related HPA axis defect (Nollet et al., 2012). It should be noted, however, that other animal models of depression showed different results in terms of changes in the hypocretin system, such as the genetically depressed Wistar-Kyoto male rats, which showed a lower number and size of hypocretin-1 neurons than its control Wistar male rats (Allard et al., 2004). In another genetically depressed rat model, i.e. the Flinders Sensitive Line (FSL), the number of hypothalamic hypocretin-positive neurons was higher in female FSL rats than in the female control rats, i.e. Flinders Resistant Line (FRL) although this publication by Mikrouli et al. offers no data on male rats (Mikrouli et al., 2011). It is a frequent phenomenon that animal models tend to mimic “just a few symptoms rather than a complete psychiatric disorder”. This is a reason to validate the data obtained in animal models on human postmortem material.

We found a clear day-night fluctuation in hypothalamic hypocretin-1-ir in the control subjects, with higher levels at night, which is similar to the pattern reported for lumbar puncture CSF hypocretin-1 levels obtained by continuous in vivo sampling (Salomon et al., 2003). These findings are in agreement with the concept that hypocretin neurons may play a key role in sleep-wake regulation (Saper, 2013), and the absence of the day-night hypocretin-1-ir fluctuation in depression may thus relate to the frequently occurring symptoms of sleep disorders in this condition. It should be noted that in our earlier research we demonstrated a clear diurnal rhythm in the biological clock, the suprachiasmatic nucleus (SCN), for its main neuropeptide, vasopressin, both on the protein level (Hofman et al., 1993) and on the mRNA level (Zhou et al., 2001) in postmortem material, when the patients were grouped according to time of death. Interestingly, a direct projection from the SCN onto the hypocretin neurons was observed in the brains of rat and human (Abrahamson et al., 2001), which indicates that the SCN may directly regulate the function of hypocretin-immunoreactive neurons. Our finding of the absence of a diurnal hypocretin rhythm in depression agrees with our earlier observation of a diminished SCN function in depression (Zhou et al., 2001). Our data (Zhou et al., 2001) and those of others (Li et al., 2013) thus also show that postmortem studies can indeed reflect the day-night fluctuations during life.

It is of interest to note that in rats the maximal activity of the hypocretin system takes place in their active period, i.e. at night (Mileykovskiy et al., 2005 ;Yoshida et al., 2001) Surprisingly, we found the highest hypocretin-1-ir levels in the human hypothalamus at night. However, this observation agrees with two human studies that reported the lowest CSF hypocretin levels during the daytime (Salomon et al., 2003 ;Grady et al., 2006). This means that the nocturnal elevation of hypocretin-1-ir in the hypothalamus is not simply due to a lack of transport or secretion. A similar phenomenon was observed for melatonin, which is also involved in sleep-wake control: in rats, melatonin levels increase during the dark period (their active phase) and decrease during the light period (rest phase) (Gutjahr et al., 2004), while in humans its levels increase during the dark period (rest phase) and decrease during the light period (active phase) (Zeitzer et al., 2007). The possibility that the diurnal regulation systems act in a different way in a diurnal and a nocturnal species (such as rodents) warrants further investigation.

A recent study showed an age-related decline in the number of hypothalamic hypocretin neurons in the range from 0 to 60 years of age in control subjects (Hunt et al., 2015). We did not find such a correlation, but it should be noted that, unlike in the study by others, the control subjects we studied did not contain very young ages.

4.2. PFC Hcrt-receptors in Depression and in Relation to Suicide

Earlier, our group found, with NBB cortex samples (depression patients without suicide), that the ACC seems to be more vulnerable than the DLPFC to alterations in depression-related molecules, such as nitric oxide synthase, gamma-aminobutyric acid and glutamate (Gao et al., 2013 ;Zhao et al., 2012). In our study, with NBB cortex samples, again we observed a trend for lower Hcrt-receptor-1-mRNA expression in the ACC in depression but no changes in Hcrt-receptors in the DLPFC. The novel finding with SMRI cortex samples (containing depressive patients who committed suicide or died of causes other than suicide) that there was a significantly increased Hcrt-receptor-2-mRNA expression in the ACC, but not in the DLPFC, in male MDD patients who had committed suicide is in agreement with previous findings that the ACC is more vulnerable to suicide than the DLPFC (Zhao et al., 2012 ;  Drevets et al., 2008) and that there is a sex difference in the prevalence of suicide (Maguen et al., 2015). Our data concerning increased ACC Hcrt-receptor-2-mRNA expression in male suicide patients are thus interesting, although too limited for a final conclusion. They do represent, however, a strong rationale for further studies on this topic. Finally, the decreased Hcrt-receptor-1-mRNA expression with aging we observed in the DLPFC may at least partly explain the findings that in SMRI cortex samples (younger) both Hcrt-receptors were detectable in ACC and DLPFC, while in the NBB series (older) Hcrt-receptor-1-mRNA expression was only detectable in ACC and Hcrt-receptor-2-mRNA expression was only detectable in the DLPFC.

Some concerns of the present postmortem brain material study should be mentioned. We did not find significant differences in the hypothalamic hypocretin-1-ir expression between MDD and BD patients, which is in accordance with our previous findings for CRH, AVP and OXT and for receptors in the hypothalamus (Bao et al., 2005 ;  Wang et al., 2008), although a final conclusion on this phenomenon should be based upon a larger sample size. Secondly, one of the inherent potential confounding factors in a postmortem study is medication use. However, we do not think that our main conclusions are cofounded by antidepressants, since increased hypothalamic hypocretin-1-ir was only observed in female depressive patients and increased expression of Hcrt-receptor-2-mRNA in the ACC was only observed in male depressive suicides, although all the depressive groups had been on antidepressants. In addition, animal studies showed that benzodiazepines (Panhelainen and Korpi, 2012), haloperidol (Dalal et al., 2003) and fluoxetine (Nollet et al., 2011), taken by the depressive patients in the present study, may inhibit hypocretin neurons and/or decrease hypocretin levels. Therefore, had antidepressants interfered with our measurements, this would have led to an underestimation of the increased hypocretin-1 levels observed in female depressive patients. It is noted that Calegare et al. found that sub-chronic treatment of adult malerats with fluoxetine increased the levels of prepro-hypocretin mRNA in the hypothalamus without affecting the hypocretin-1 CSF levels ( Calegare et al., 2016). In our study, there were 2 out of 10 male depression patients who had taken fluoxetine, while their hypothalamic hypocretin-1-ir levels (IOD: 0.128 and 0.134) were fully within the range of the other male depression patients (IOD range from 0.103 to 0.248, median value 0.166). Finally, it should also be noted that the Hcrt-receptor data are based upon mRNA measurements and have yet to be confirmed on the protein level.

5. Conclusions

A clear sex-related change was found in the hypothalamic hypocretin-1-ir in depression. The CUMS rat depression model did not replicate changes found in the hypocretin systems in the human brain in depression. Since sex-related changes in hypothalamic hypocretin-1-ir expression were observed in depression, this factor should be taken into account in the development of hypocretin-targeted therapeutic strategies.

Breastfeeding, Cognitive and Noncognitive Development in Early Childhood: A Population Study


BACKGROUND AND OBJECTIVES: There is mixed evidence from correlational studies that breastfeeding impacts children’s development. Propensity score matching with large samples can be an effective tool to remove potential bias from observed confounders in correlational studies. The aim of this study was to investigate the impact of breastfeeding on children’s cognitive and noncognitive development at 3 and 5 years of age.

METHODS: Participants included ∼8000 families from the Growing Up in Ireland longitudinal infant cohort, who were identified from the Child Benefit Register and randomly selected to participate. Parent and teacher reports and standardized assessments were used to collect information on children’s problem behaviors, expressive vocabulary, and cognitive abilities at age 3 and 5 years. Breastfeeding information was collected via maternal report. Propensity score matching was used to compare the average treatment effects on those who were breastfed.

RESULTS: Before matching, breastfeeding was associated with better development on almost every outcome. After matching and adjustment for multiple testing, only 1 of the 13 outcomes remained statistically significant: children’s hyperactivity (difference score, –0.84; 95% confidence interval, –1.33 to –0.35) at age 3 years for children who were breastfed for at least 6 months. No statistically significant differences were observed postmatching on any outcome at age 5 years.

CONCLUSIONS: Although 1 positive benefit of breastfeeding was found by using propensity score matching, the effect size was modest in practical terms. No support was found for statistically significant gains at age 5 years, suggesting that the earlier observed benefit from breastfeeding may not be maintained once children enter school.


  • Abbreviations:
    propensity score matching
    strengths and difficulties questionnaire
    structural equation modeling


What’s Known on This Subject:

The medical benefits of breastfeeding for mother and child are considered numerous, yet the effect of breastfeeding on cognitive abilities remains largely debated given selection into breastfeeding. The effect on behavior is even less well understood.

What This Study Adds:

In applying quasi-experimental techniques which mimic random assignment, this study supports limited positive impacts of breastfeeding for children’s cognitive and noncognitive development. Although significant, the effect of breastfeeding on noncognitive development is small in practical terms.

The medical benefits of breastfeeding for both mother and child are considered numerous and well documented.15 Yet the effect of breastfeeding on general cognitive abilities has been a topic of debate for nearly a century.6 The mechanism argued to be responsible for these effects is the nutrients found in breast milk.7,8 Two specific types of long-chain polyunsaturated fatty acids, namely docosahexaenoic (DHA) and arachidonic acid, have been implicated in both visual and neural development and functioning through neural maturation, which is important for cognitive abilities, such as problem solving.911

The link with nutrients may also impact specific cognitive abilities like language development. For example, language abilities, such as vocabulary, are highly dependent on working and long-term memory given the consolidation and retrieval processes needed during acquisition.12,13 In rats, deficiency of fatty acids, such as DHA, during lactation resulted in poor memory retention during learning tasks, whereas supplementation of DHA had reversal effects.14 If the hypothesized “causal” mechanism of superior nutrition in breast milk is true, coupled with the specific impact of DHA on memory, breastfeeding should also impact language abilities. To date, ∼20 studies have investigated this association and all but 115 examined a combined measure of language (receptive and expressive) or receptive language only. There remains debate as to whether expressive and receptive language in early childhood form distinct modalities of language,16,17 raising the question of whether breastfeeding would be equally beneficial to each modality in the case of a 2-factor language model.

Less studied is the impact of breastfeeding on behavior. Breastfeeding may lead to reduced behavioral problems as a result of early skin-to-skin contact, which helps form a secure mother-infant bond.18 Any effects of breastfeeding on cognitive and language development could also prevent the development of behavior problems. The absence of early behavior problems has social, economic, and medical value to society through reduced prevalence of delinquency, incarceration rates, and substance abuse,1921 making this an important area of research. With few exceptions, there remains a dearth of high-quality studies examining behavior,2225 and among them, consensus is not evident.

Without randomization of mothers to breastfeeding and formula conditions, it is challenging to confirm the causal impact of these hypotheses. One study randomized the provision of a breastfeeding intervention, modeled on the Baby-Friendly Hospital Initiative, and found that the children of mothers in the intervention group had higher intelligence scores compared with controls at age 6 years.26 The strongest effects were for verbal intelligence. This study offers the best support to date for a causal link between breastfeeding and cognitive development. However, it is the only cluster randomized trial on human lactation.

The majority of studies in this field are observational, thus the causal implications of breastfeeding are questionable given the inherent difficulty in controlling for selection into breastfeeding. For example, initial associations with cognitive development are often reduced after adjustments for confounders, such as parental education/IQ (ie, from an average 5-point to 3-point difference27), and, in some cases, the associations are no longer statistically significant.28 A variety of observational studies now apply quasi-experimental methods to better address the issue of selection bias, making inroads toward a better understanding of potential causal paths. The techniques used include propensity score matching (PSM), instrument variables, and sibling pair models. This study uses PSM because the sibling pair model limits the available pool of participants and instrument variables are extremely sensitive to the validity of the chosen instrumentation, which should be associated with the exposure but not with the outcome except for via the exposure.

Using a large longitudinal population sample, we applied PSM, which mimics random assignment, in an effort to investigate the potential impacts of breastfeeding on children’s cognitive ability, expressive vocabulary, and behavior problems. Both breastfeeding duration and intensity were examined. Significant advantages for children who were breastfed, after matching, were expected for all outcomes. Grounded in the recommendations of the World Health Organization,29 it was expected that larger effect sizes would be observed for children who were fully breastfed and for longer durations.



Participants included families enrolled in the Growing Up in Ireland infant cohort. Families with infants born between December 2007 and May 2008 were identified from the Child Benefit Register and randomly selected to participate. The overall recruitment response rate was 65% (N = 11 134). A detailed description of the study design can be found elsewhere.30 We used data collected at 9 months and 3 and 5 years of age. Only families with complete data for all confounders when children were 9 months and children who were born full term were included (N = 9854; 88.5% of the initial sample). Boys represented 50.6% (N = 4991) of the sample. Attrition across waves reduced the sample size to 8715 children at 3 years and 8032 at 5 years. Some children had missing data on the cognitive and vocabulary scales, resulting in 8535 and 8241 children respectively at age 3 and 7972 and 7942 children respectively at age 5. Additionally, missing teacher reports for behavior at age 5 years resulted in 7478 children being included in these analyses. Demographic characteristics of the families and rates of breastfeeding engagement can be found in Table 1 and Fig 1. Ethics approval was obtained from the Research Ethics Committee, Department of Children and Youth Affairs Ireland, and written consent was collected from parents/guardians before data collection.


Family, Maternal, Infant, and Medical Characteristics: Infant Cohort at 9 Months


The category “1” on the x-axis represents breastfeeding up to 31 days; “2” represents between 32 and 180 days; and “3” represents ≥181 days.


Children’s cognitive abilities and expressive vocabulary were measured by using 2 scales from the British Abilities Scale31. The pictures similarities scale assessed problem-solving skills and the naming vocabulary scale assessed expressive vocabulary. The construct validity of each scale was derived by using the Wechsler Preschool and Primary Scale of Intelligence-Revised (r = 0.74 and 0.83, respectively).31 Standardized scores that adjusted for performance as compared with other children of the same age, with a mean of 50 and a SD of 10, were used. Age was adjusted in 3-month age bands.

The Strengths and Difficulties Questionnaire (SDQ32) was used to assess children’s problem behaviors. The parent version was used at age 3 years and both the parent and teacher versions were used at age 5 years. The SDQ is comprised of 5 scales (emotional symptoms, conduct problems, hyperactivity/inattention, peer relationship problems, and prosocial behavior) with ratings of applicability of behaviors on a 3-point scale. A total difficulties scale is included, combining the 4 problem scales, to yield an overall difficulties score. We used the conduct problems, hyperactivity/inattention, and difficulties scales given our focus on externalizing problems. Validation of the SDQ has been extensively documented.33Table 2 reports the correlations between parent and teacher SDQ reports and the means and SDs for all child outcomes.

Breastfeeding information was collected retrospectively when infants were 9 months old via maternal report. Support for the reliability of recall in previous breastfeeding studies has been established.34 However, given the lower reliability regarding the timing of the introduction of additional fluids/solids, Labbok and Krasovec’s definition of full (ie, exclusive or almost exclusive) and partial breastfeeding are used.35 Two breastfeeding variables were created to assess whether the infant was fully or partially breastfed and the duration of each. Mothers were asked 4 questions: “Was <baby> ever breastfed,” “How old was <baby> when he/she completely stopped being breastfed,” “Was <baby> ever exclusively breastfed,” and “How old was <baby> when he/she completely stopped being exclusively breastfed?” First, infants were grouped by breastfeeding status, both full and partial (5940) and never breastfed (3914). Of those who had ever been breastfed, 4795 had full breastfeeding at some point. Next, breastfeeding duration was grouped into 3 intervals; breastfed up to 31 days, 32 to 180 days, and ≥181 days. Each category of duration was treated as mutually exclusive, dummy coded, and compared against infants who had never been breastfed for the purpose of matching.

Confounders have been suggested in part to account for the associations found between breastfeeding and child outcomes. We matched groups (breastfed, never breastfed) on 14 of the most pertinent factors. At the child level, factors included sex (boy/girl), birth weight (≥2500 g), and having neonatal intensive care (yes/no). At the maternal level, factors included age (≤24 years, 25–29 years, 30–34 years, or ≥35 years), highest level of education (primary level/no education, second level, or third level), working status before pregnancy (yes/no), ethnicity (Irish, any other white background, African or any other black background, Asian background, or other, including mixed background), depression (a score of ≥11 on the Center for Epidemiologic Studies Depression Scale), and type of delivery (vaginal or caesarean). Family-level factors included having a partner in the residence (yes/no), social class (professional/managerial, other nonmanual/skilled manual, or semiskilled/unskilled), medical card status (free medical care, free general practitioner care, or no free medical care), total number of household members who smoked during the pregnancy (none, or ≥1), and whether the cohort infant had siblings living in the household.

Statistical Analysis

PSM reduces selection bias by matching children who were breastfed to children who were not, but who had a similar probability of being breastfed based on their measured characteristics. We used PSM logit models with nearest neighbor 1:1 matching techniques. In nearest neighbor matching, the sample is randomly ordered with matching occurring sequentially between the treatment (breastfed) and control (not breastfed) group based on participants’ propensity scores. Typically, the pair is then removed from the list and the next match is created. To ensure optimal matches, we imposed a caliper so that pairs could only be matched if the propensity score was within a tenth of a SD of the other. We also allowed matching with replacement given the low rates of longer durations and full breastfeeding in this cohort. Although matching with replacement has been argued to increase variance in the data, it also arguably reduces bias in the sample by ensuring better quality of matches.36 Balance checks in all models revealed substantial reductions of bias between matched groups on all individual confounders (ie, 0%–13.9% remaining bias in partial breastfeeding models, 0%–18.1% remaining bias in full models; data available on request). The remaining overall mean bias across models ranged from 3.2% to 8.5%. The ≤20% remaining bias has been suggested as the acceptable cutoff after matching.37 Thus, we concluded that the analytic matching technique resulted in good matches between conditions. Matching resulted in all participants falling within the area of common support. The average treatment effect on those who were treated (ie, children who were breastfed) is reported. Adjustments were made for multiple hypothesis testing by using the Holmes-Bonferroni method. All statistical analyses for PSM were conducted by using Stata version 13 software (Stata Corp, College Station, TX).

To note, although PSM is advantageous in mimicking random assignment, a drawback is the challenge in evaluating a linear dose-response association, which has previously been found. Structural equation modeling (SEM) offers an alternative approach to examining this dose-response association. Additionally, SEM uses the full sample and has greater power. Thus, the data were also modeled by using SEM, where confounders were treated as correlated exogenous variables, the duration of breastfeeding was treated as a continuous mediating variable, and child outcomes were treated as correlated, which could be influenced by both breastfeeding and confounders. These results can be found in the Supplemental Material.


Postmatching results for children fully breastfed up to 31 days revealed no statistically significant differences between groups on any outcome at age 3 or 5 years (Table 3). Similarly, for children who were fully breastfed between 32 and 180 days, no statistically significant differences were found for any outcomes at either age postmatching (Table 4). Finally, for children who were fully breastfed for ≥6, statistically significant differences were found postmatching for only 2 outcomes, problem solving and hyperactivity at age 3 years. Children who were fully breastfed scored 2.95 (SE = 1.39, P = .048) points higher on the problem-solving scale compared with children who were never breastfed and –0.84 (SE = 0.25, P ≤ .001) points lower on the hyperactivity scale. After adjustment for multiple testing, cognition was no longer statistically significant. However, children who were fully breastfed had slightly lower parent-rated hyperactivity compared with controls, and this remained statistically significant after adjustment (Table 5). Of note, results of the partial breastfeeding models were similar to the full models, however, after adjustment for multiple testing, neither cognitive ability nor hyperactivity at age 3 years remained statistically significant. These results can be found in the Supplemental Material.


Without randomized controlled trials, the issue of causality will necessarily remain open, however the present results contribute important insights to the long-standing debate of potential “causal effects” versus artifacts of confounding that are not properly accounted for. This study also provides new perspectives on breastfeeding and children’s externalizing behavior. To the best of our knowledge, this is among the first studies to examine expressive vocabulary as an individual outcome and to consider externalizing behavior. It should be noted that our results apply only to infants born full term.

After adjustment for multiple testing, the initial support found for breastfeeding and better problem solving at age 3 years if the child was breastfed for a minimum of 6 months was no longer statistically significant. In addition, no statistically significant effects were found for cognitive ability at age 5 years. These results are in contrast to some studies that have used PSM techniques to examine the effects of breastfeeding and general cognitive abilities.3840However, differences in both analytical choices of the PSM approach used (eg, replacement, calipers) and differing selection of covariates may help to explain these differences across studies. Nonetheless, our findings were surprising in the context of the nutrients in breast milk being responsible for increased cognitive development. Regarding expressive vocabulary, no statistically significant advantages were observed for children who were breastfed at either age 3 or age 5.

The limited research on breastfeeding and behavior problems is inconsistent, despite the relatively consistent reliance on the SDQ. Of interest, studies that have dichotomized the SDQ scales into abnormal scores (ie, at the 85th or 90th percentile) have not found statistically significant differences,2325 suggesting that breastfeeding is not likely to be a contributor to behavioral problems at clinical levels. When the SDQ scales are treated as continuous, small effects under certain conditions have been found.22 In this study, we treated all 3 scales as continuous and found that children who were fully breastfed for ≥6 months had lower parent-rated scores on the hyperactivity scale at age 3 years only. This result remained statistically significant after adjustment for multiple testing. Our results suggest that longer durations of breastfeeding might help to reduce hyperactive behaviors for children who display mild to moderate levels in the short term, but that these benefits are not maintained even in the medium term. This result would seemingly support the recommendation of the World Health Organization, suggesting that breastfeeding for at least 6 months is necessary for early gains to be observed.

The inherent strengths of this study include the use of a particularly large longitudinal developmental dataset, the use of a quasi-experimental statistical approach, the use of a repeated measures design, the use of multiple informants and simultaneous standardized assessments thereby limiting potential shared method variance, the comparatively large number of confounders controlled (ie, 14) in contrast to previous studies (ie, an average of 7.7 ± 3.4 in higher-quality studies28;), and assessments in both cognitive and noncognitive domains of child development. Despite these strengths, some limitations must be noted. First, information on breastfeeding was collected retrospectively. Although the reliability of recall has been established,34 it must be acknowledged that recall bias may nevertheless be present, particularly regarding the duration of full breastfeeding. Second, only parent-reported SDQs were collected when children were 3 years of age. Studies have found that parents typically rate their children as having higher levels of problem behaviors as compared with teacher reports, with weak associations between these 2 types of informants,24 as was found in the current study for behavior ratings at age 5 years between parents and teachers. Having access to child care staff reports at age 3 years would have increased the reliability of the maternal-rated hyperactivity finding. Third, no information pertaining to direct breastfeeding versus expressed breast milk feeding was collected. Thus, it is not possible to investigate whether the association with reduced hyperactivity at age 3 years was the result of skin-to-skin contact or due to the nutrients in breast milk. This is an important direction for future studies examining behavioral outcomes. Fourth, although maternal education was included as a confounder, maternal IQ was not collected in this cohort. In the few studies that controlled for maternal IQ, the findings suggested that it accounted for a large part of the association between breastfeeding and cognitive outcomes.39,41 Thus, the inclusion of maternal IQ in future studies that employ PSM is warranted. Finally, PSM does not address selection on unobservables. Causal estimates may only be estimated by using PSM if selection is on observable characteristics or, in cases where unobservable factors influence selection into breastfeeding, the balancing on observables also balances on these unobservables. Despite these limitations, the results of this study add to the growing literature by showing that some statistically significant positive noncognitive benefits may result from longer durations of breastfeeding. Yet, beyond the statistical implications, the practical implications appear minimal and short lived. It is important to note, however, that these findings do not contradict the many medical benefits afforded to both mother and child as a result of breastfeeding.


  1. Source:The American Academy of Pediatrics