Diet and depression in teens

A recent longitudinal study found that the type of diet in teens is associated with differences in BMI but also with chronic inflammation and the development of depression several years later (see doi.org/10.1016/j.bbi.2018.01.002 ).

Details:
--843 adolescent offspring in the Western Australian Pregnancy Cohort (Raine) Study were followed longitudinally at ages 1,2,3,6,8,10,14, and 17. The current study assessed data in those at ages 14 and 17.
--88% Caucasian, 72% maternal education >12 yrs, physical activity <1 time per month in 10%/1-3x per week in 57%/4+ times per week in 33%, 97% nonsmokers, 87% no alcohol in past 6 months, family income <$40K in 25%/$40-60K in 20%/$60-105K in 33%/>$104K in 20%
--assessed: dietary patterns (from food frequency questionnaire), energy intake, and BMI at age 14; adipokines (leptin, adiponectin) and inflammatory marker hs-CRP (high sensitivity C-reactive protein) at age 17;  and the relationship between these findings to depressive symptoms (Beck Depression Inventory), and Internalizing and Externalizing Behavioral Problems (Child Behavior Check List Youth Self-Report of 118 items for behavior in past 6 months) at age 17. 

Results:
--overall data: mean BMI age 14 was 21.3/age 17 was 22.9, serum markers: leptin 18.2 mg/L, hs-CRP 1.7 mg/L (with hs-CRP 2-3 in 5.5%/>3 in 13.9%); Beck Depression score normal in 91.5% at age 14, 85.0% at age 17, clinically concerning Internalizing Score at age 17 in 16.2%, Externalizing Score in 19.9%
--higher Western diet (high intake of red meats, fast foods, processed/refined foods, confectionary, soft drinks, full-fat dairy) at age 14 was associated with higher energy intake and BMI at age 14. And, this higher BMI at age 14 was associated with (at age 17): higher BMI (p<0.01), leptin (p<0.05), hs-CRP (p<0.05), depressive symptoms (p<0.05) and mental health problems (p<0.05)
--a healthy diet (fruits, veges, legumes, fish, whole-grains) was inversely assoc with BMI and inflammation at age 17 (p<0.05)
--in terms of reverse causality (depression leading to bad diet): depression at age 14 did not predict dietary patterns at age 14. however, depressed affect and externalizing/internalizing behaviors at age 14 was highly predictive of mental health disorders at age 17
--no significant difference in outcomes if stratify by sex or by smoking. there was a notable change in cigarettes and alcohol: at age 14, 3% smoking and 10% had alcohol in prior 6 months; by age 17, the numbers were 15% smoking and 50% consumed alcohol
--there was a small but significant difference in hs-CRP by maternal ethnicity: mean for Caucasian mothers was 1.8 mg/L, but 1.12 for non-Caucasian

Commentary:
--some background: in many Western societies, the lifetime prevalence of mental health disorders is close to 50%, with up to half beginning prior to 14yo and another 25% before age 24.
--in this prospective, longitudinal study, a Western diet was associated with increased risk of high BMI at age 14, which correlates with high BMI, leptin and hs-CRP levels, as well as depressive and mental health disorders at age 17. Several other studies have also found that Western diet is associated with increased BMI and also with increased inflammation; the opposite was true with a healthy diet (as defined above).​ also, there may be a synergy: a Western diet itself can lead to low-grade inflammation as well as obesity; and a Western diet in the setting of obesity may lead to even higher inflammatory responses to a high-fat meal.
--The current study fits in with lots of data over many years linking depression with inflammation. Many studies have shown that inflammation precedes depression in a dose-response manner, though some have found a bidirectional effect. for the opposite age spectrum: in a prospective study in 85-year olds without neuropsych symptoms at baseline, higher baseline CRP levels preceded an accelerated increase in depressive symptoms (p<0.001), despite no relationship with cognitive decline 5 years later. these depression scores reached significance 2 years later and increased further over time (see van den Biggelaar. Exper Geront 2007; 42: 693)
--this study did assess the potential role of reverse causation, where depression leads to a bad diet which then leads to obesity, though would be useful to have longer term data on this (does this all translate into adult obesity???). and more data for the year or two before the 14 yo questionnaires, to confirm the relationship
--there are pretty consistent relationships between both the metabolic syndrome and Mediterranean diet with inflammation. for example one study showed reduced inflammatory markers, including CRP, in those on a Mediterranean diet (see Estruch R. Proc Nutr Soc. 2010; 69 (3):333). And many studies have found that inflammation is frequently part of the metabolic syndrome and may be part of the link with atherosclerosis/diabetes (eg see Eckel RH. Lancet 2010; 375: 181). other studies have found that obesity itself is associated with inflammatory markers (eg, it turns out that 30% of IL-6, an inflammatory marker, is derived from adipose tissue). and there are studies finding that increasing adiposity leads to leptin signaling and the release of hs-CRP from the liver. monounsaturated fats themselves seem to decrease inflammation.
--as a tangentially related issue: there also is some interesting and conflicting studies on high carbohydrate diets and depression. The Women's Health Initiative found an association between higher glycemic index diets and increasing odds of incident depression in postmenopausal women (see Gangwisch J. Am J Clin Nutr 2015; 102: 454). On the other hand, there is some argument that at least some depressed patients may be self-medicating by consuming high carbohydrate diets to treat their depression, probably by increasing brain serotonin levels (sort of similar to giving SSRIs), with purported mechanism that carbs increase insulin, which increases brain tryptophan levels (indirectly, it turns out: insulin lowers the "large neutral amino acids" such as leucine, isoleucine and valine, which compete with tryptophan for passage through the blood-brain barrier, and this tryptophan accelerates the production and release of serotonin), see Wurtman RJ. Obes Res. 1995; 3 (Suppl ): 477S. There are several animal studies, and some human ones which support this (eg smokers quitting smoking, may have attendant dysphoria, and have carbohydrate-craving/weight gain.
--limitations of this study: this was observational study, so cannot predict causality. also, the data are largely self-reported, and there may be inherent inaccuracies in that (eg obese and overweight individuals in other studies have been found to under-report dietary intake). i am not sure about the literature on this, but it might well be that mental health questionnaires have different validities in 14yo and 17yo teens, that perhaps either hormonal changes or changes in brain structure/maturity affect the individuals’ responses to the questionnaires. also the baseline group of  teens in this study was pretty healthy, exercised pretty regularly, came from economically pretty well-off families, had mostly pretty normal BMI, did not consume lots of cigarettes or alcohol, and had pretty low depression scores

so, the finding that inflammation was associated both with obesity and with depression does not necessarily translate into inflammation being the causal link (does obesity itself at age 14 lead to more depressive symptoms later on, independent of inflammation??? what about issues about self-image?? bullying??).  And the role of inflammation in other diseases (diabetes, heart disease) is even more supported than for depression.  but, the big picture conclusion, that diet seems to be related to subsequent depression, does seem to have some validity, and that decreasing depression should probably be added to the litany of good things associated with eating well.
other relevant blogs:
--see http://gmodestmedblogs.blogspot.com/2014/12/mediterranean-diet-and-longevity.html  which looks at Mediterranean diet and longevity, specifically at telomere length, which is also associated with inflammation
--see http://gmodestmedblogs.blogspot.com/2016/08/non-alcoholic-fatty-liver-disease-2.html for some studies suggesting that high fructose diets are associated with hepatic inflammation

to be added to the email list, contact me at: gmodest@uphams.org

Comments

Popular posts from this blog

cystatin c: better predictor of bad outcomes than creatinine

diabetes DPP-4 inhibitors and the risk of heart failure

UPDATE: ASCVD risk factor critique