4,771 peer-reviewed studies, each with a short note written from its published abstract: what kind of study it was, who took part, what was measured and what the authors reported. Observational studies show associations, not proof.
2017 / Prospective multicenter cohort study
In a prospective multicenter cohort of 4,470 participants followed for 8 years, researchers evaluated whether baseline dietary fiber intake assessed via FFQ was related to knee pain trajectories. Four pain patterns were identified: no pain (34.5%), mild pain (38.1%), moderate pain (21.2%), and severe pain (6.2%). Highest vs. lowest quartile of total fiber intake was linked to a lower risk of moderate pain (OR 0.76, 95% CI 0.61-0.93) and severe pain (OR 0.56, 95% CI 0.41-0.78), with P <= 0.006 for trend for both. Similar inverse associations were observed for grain fiber intake.
Shows an association, not proof of cause and effect.
Dai Z, Lu N, Niu J, et al. Arthritis Care Res (Hoboken), 2017. PMID 27899003
2017 / Cross-sectional survey study
Using NHANES 2007-2010 data, researchers examined the association between self-reported cannabis use and DXA bone mineral density (BMD) of the proximal femur and lumbar spine among 4,743 participants aged 20-59 years. In the population, 60% reported ever using cannabis (47% former, 5% light, and 7% heavy users). After adjusting for age, sex, BMI, race/ethnicity, and other BMD determinants, no association between cannabis and BMD was observed for any level of use (p ≥ 0.28).
Shows an association, not proof of cause and effect.
Bourne D, Plinke W, Hooker ER, et al. Arch Osteoporos, 2017. PMID 28286929
2017 / Two prospective cohort studies
Researchers analyzed data from the Osteoarthritis Initiative (OAI) and the Framingham Offspring Osteoarthritis Study to assess associations between baseline dietary fiber intake and knee osteoarthritis outcomes over 48 months (OAI) or 9 years (Framingham). In both cohorts, total dietary fiber was inversely associated with incident symptomatic OA (p trend <0.03), with a lower risk in the highest versus lowest quartile (OR 0.70, 95% CI 0.52 to 0.94 in OAI; OR 0.39, 95% CI 0.17 to 0.88 in Framingham). Total and cereal fiber were also significantly inversely associated with knee pain worsening in OAI (p trend <0.02). No apparent association was observed between fiber intake and radiographic OA.
Shows an association, not proof of cause and effect.
Dai Z, Niu J, Zhang Y, et al. Ann Rheum Dis, 2017. PMID 28536116
2017 / Prospective longitudinal study
In a prospective longitudinal study, 29 patients were monitored while on a medically supervised ketogenic diet for an average of 2.1 years (range 0.5-6.5 years). There was a trend toward a reduction in lumbar spine BMD Z score of 0.1562 per year (p=0.071), with 20 patients (68%) exhibiting lower BMD Z scores by the end of treatment. The rate of bone loss was greater in more mobile patients (0.28 SD loss per year, p=0.026), and height-adjusted BMAD Z scores showed a mean reduction of 0.19 SD in 13 evaluated patients. Two patients experienced fractures, and while mean urinary calcium-creatinine ratios were elevated (0.77), only one patient developed renal calculi.
Shows an association, not proof of cause and effect.
Simm PJ, Bicknell-Royle J, Lawrie J, et al. Epilepsy Res, 2017. PMID 28778055
2017 / Prospective cohort study (20-year follow-up)
Using China Health and Nutrition Survey data from 15,572 adults followed from 1989 to 2011, researchers evaluated associations between dietary and nutrient patterns and fracture risk. Compared with the lowest tertile, participants in the highest tertile of cumulative scores for modern dietary and animal-sourced nutrient patterns had higher fracture risks (HR = 1.34; 95% CI: 1.06-1.71 and HR = 1.37; 95% CI: 1.08-1.72, respectively). Overall mean factor scores showed similar or stronger patterns of association, whereas recent and baseline scores showed no association with fracture risk.
Shows an association, not proof of cause and effect.
Melaku YA, Gill TK, Appleton SL, et al. Nutrients, 2017. PMID 29088104
2016 / Cross-sectional study
Researchers examined the cross-sectional association between dietary soy milk intake and radiographic knee joint space narrowing and osteophytes in 5,764 subjects. In fully adjusted models, soy milk intake of ≥once a day (OR 0.49, 95% CI 0.26-0.92, P = 0.026) and <once a day (OR 0.79, 95% CI 0.67-0.94, P = 0.009) was inversely associated with osteophytes compared to never intake (P for trend = 0.001). No significant associations were observed between soy milk intake and joint space narrowing in any model.
Shows an association, not proof of cause and effect.
Li H, Zeng C, Wei J, et al. Rheumatol Int, 2016. PMID 27193467
2014 / Cross-sectional survey
In a cross-sectional study of 1,218 men and 907 women aged ≥60 years, dietary acid load was evaluated for associations with bone mineral density. An inverse association between PRAL from diet and proximal femur bone mineral density was observed in men consuming <800 mg/d of dietary calcium (p = 0.02), but not in men consuming ≥800 mg/d (p = 0.98). In women, PRAL from diet was positively associated with proximal femur BMD (p trend = 0.04). When supplemental calcium was considered, no association was observed between dietary acid load and bone mineral density.
Shows an association, not proof of cause and effect.
Mangano KM, Walsh SJ, Kenny AM, et al. J Bone Miner Res, 2014. PMID 23873776
2014 / Prospective cohort study
The study evaluated the relationship between high-protein foods (legumes, meat, and meat analogues) and hip fracture incidence in 33,208 participants from the Adventist Health Study-2. Compared with consuming legumes less than once weekly, legume intake once daily or more was associated with a 64% lower risk of hip fracture (HR = 0.36, 95% CI 0.21, 0.61). Consuming meat analogues once daily or more was associated with a 49% reduced risk (HR = 0.51, 95% CI 0.27, 0.98), and meat intake of four or more times weekly was associated with a 40% reduced risk (HR = 0.60, 95% CI 0.41, 0.87) compared to intakes of less than once weekly.
Shows an association, not proof of cause and effect.
NANO counterpoint →
Lousuebsakul-Matthews V, Thorpe DL, Knutsen R, et al. Public Health Nutr, 2014. PMID 24103482
2014 / Prospective cohort study
Researchers examined the relationship between teenage milk consumption (ages 13 to 18 years) and later hip fracture risk over 22 years of follow-up in over 96,000 men and postmenopausal women. During follow-up, 1,226 hip fractures occurred in women and 490 in men. In men, each additional glass of milk per day during teenage years was associated with a 9% higher risk of hip fracture (RR = 1.09; 95% CI, 1.01-1.17), which was attenuated after adjusting for attained height (RR = 1.06; 95% CI, 0.98-1.14). Teenage milk consumption was not associated with hip fractures in women (RR = 1.00 per glass per day; 95% CI, 0.95-1.05).
Shows an association, not proof of cause and effect.
Feskanich D, Bischoff-Ferrari HA, Frazier AL, et al. JAMA Pediatr, 2014. PMID 24247817
2014 / Prospective cohort study
The Singapore Chinese Health Study examined the relationship between dietary patterns and hip fracture risk in 63,257 Chinese men and women aged 45-74 years using a validated food-frequency questionnaire. Principal components analysis identified vegetable-fruit-soy (VFS) and meat-dim-sum (MDS) patterns, and diet quality was evaluated using the AHEI-2010. In both genders, participants in the highest versus lowest quintile had lower hip fracture risk for the VFS pattern (HR: 0.66; 95% CI: 0.55, 0.78; P-trend ≤ 0.008) and the AHEI-2010 (HR: 0.68; 95% CI: 0.58, 0.79; P-trend ≤ 0.008). The MDS pattern was not associated with hip fracture risk.
Shows an association, not proof of cause and effect.
Dai Z, Butler LM, van Dam RM, et al. J Nutr, 2014. PMID 24572035
2014 / Prospective cohort study
A prospective study followed 1,662 community-dwelling men aged 70 to 97 years for a mean of 4.3 years to examine the link between serum 25-hydroxyvitamin D (25OHD) and fracture risk. Over follow-up, 123 first-incident fragility fractures occurred, revealing a U-shaped association. In multivariate analysis, fracture risk was highest in the lowest 25OHD quintile (≤36 nmol/L; HR 3.5; 95% CI 1.7-7.0) and in the highest quintile (>72 nmol/L; HR 2.7; 95% CI 1.4-5.4) compared with the 4th quintile (≥60 to ≤72 nmol/L). These relationships were not explained by bone mineral density, fall risk, physical activity, or other examined lifestyle or anthropomorphic factors.
Shows an association, not proof of cause and effect.
Bleicher K, Cumming RG, Naganathan V, et al. J Bone Miner Res, 2014. PMID 24677358
2014 / Prospective cohort study
Researchers analyzed data from 73,572 postmenopausal women in the Nurses' Health Study followed for up to 30 years, during which 1,873 incident hip fractures occurred. In multivariable models, each additional daily serving of total soda was associated with a 14% increased risk of hip fracture (RR: 1.14; 95% CI: 1.06, 1.23). Risk was significantly higher for both regular soda (RR: 1.19; 95% CI: 1.02, 1.38) and diet soda (RR: 1.12; 95% CI: 1.03, 1.21), with no significant differences between colas and noncolas, or caffeinated and decaffeinated sodas. The attributable risk in the cohort for total soda intake was 12.5%.
Shows an association, not proof of cause and effect.
Fung TT, Arasaratnam MH, Grodstein F, et al. Am J Clin Nutr, 2014. PMID 25099544
2014 / Cross-sectional bioarchaeological / isotope and trace element study
Researchers analyzed stable carbon, nitrogen, and sulphur isotopes along with inorganic bone elements (strontium and calcium) in 53 individuals, including 22 gladiators, from Ephesus dating to the 2nd and 3rd century AD. All individuals consumed C3 plants like wheat and barley as staples, with relatively low δ15N values possibly reflecting frequent legume intake. Sr/Ca-ratios of gladiators were significantly higher than those of contemporary Roman inhabitants, which might suggest frequent use of a plant ash beverage.
Shows an association, not proof of cause and effect.
Lösch S, Moghaddam N, Grossschmidt K, et al. PLoS One, 2014. PMID 25333366
2014 / Prospective cohort studies
Two large Swedish cohorts evaluated the association between milk consumption and time to mortality or fracture. Over a mean follow-up of 20.1 years in women and 11.2 years in men, adjusted all-cause mortality hazard ratios per glass of milk were 1.15 (95% CI 1.13 to 1.17) in women and 1.03 (1.01 to 1.04) in men. In women, each glass of milk was associated with hazard ratios of 1.02 (1.00 to 1.04) for any fracture and 1.09 (1.05 to 1.13) for hip fracture, with corresponding hazard ratios of 1.01 (0.99 to 1.03) and 1.03 (0.99 to 1.07) in men. The authors noted that residual confounding and reverse causation are possible given the observational design.
Shows an association, not proof of cause and effect.
Michaëlsson K, Wolk A, Langenskiöld S, et al. BMJ, 2014. PMID 25352269
2013 / Cross-sectional study
This cross-sectional study evaluated bone health and its association with nutritional and biochemical variables in 33 adults with classic galactosemia. Hip BMD was significantly lower in women compared to men (0.799 vs. 0.896 g/cm², p = 0.014), and more women had a BMD-Z < -2.0 (33% vs. 18% at the spine; 27% vs. 6% at the hip). In final regression models, higher gonadotropin levels were linked to lower spinal BMD in women (p = 0.017), and serum calcium was a significant predictor of both hip (p = 0.014) and spine (p = 0.013) BMD in both sexes.
Shows an association, not proof of cause and effect.
Batey LA, Welt CK, Rohr F, et al. Osteoporos Int, 2013. PMID 22525982
2013 / Prospective cohort study
A study evaluated 157 postmenopausal women categorized by low (≤0.76 μM), intermediate (0.76-1.42 μM), or high (≥1.42 μM) urinary phytate (InsP6) concentrations, measuring densitometry over 12 months and 10-year fracture risk via FRAX. Women with low InsP6 had significantly greater lumbar spine bone mass loss than those with high levels (3.08 ± 0.65% vs. 0.43 ± 0.55%), with more women in the low group losing over 2% bone mass (55.6% vs. 20.7%). The 10-year fracture probability was also significantly higher in the low-phytate group than the high-phytate group for hip fracture (0.37 ± 0.06% vs. 0.18 ± 0.04%) and major osteoporotic fracture (2.45 ± 0.24% vs. 1.83 ± 0.11%).
Shows an association, not proof of cause and effect.
López-González AA, Grases F, Monroy N, et al. Eur J Nutr, 2013. PMID 22614760
2013 / Cross-sectional observational study
Researchers investigated the relationship between sodium intake, calcium intake, and blood lead concentration in 1090 pregnant Korean women at mid-pregnancy. Dietary intakes were estimated using a 24-hour recall method including supplements, and blood lead was measured via graphite furnace atomic absorption spectrophotometry. Multiple regression analysis adjusted for covariates showed that maternal sodium intake was positively associated with blood lead concentration only when calcium intake was below the estimated average requirement (P=0.001).
Shows an association, not proof of cause and effect.
Lee YA, Hwang JY, Kim H, et al. Br J Nutr, 2013. PMID 22784671
2013 / Nationwide registry-based cohort study
Using national registries from 1980 through 2006, researchers examined 4.4 million Danish individuals aged 40 and older to assess associations between skin cancer diagnoses and health outcomes. For non-melanoma skin cancer vs without, adjusted odds ratios were 0.96 (95% CI: 0.94-0.98) for myocardial infarction and 1.15 (1.12-1.18) for hip fracture (though risk was reduced below age 90), and the hazard ratio for all-cause mortality was 0.52 (0.52-0.53). For cutaneous malignant melanoma, odds ratios were 0.79 (0.74-0.84) for myocardial infarction and 0.84 (0.76-0.93) for hip fracture, with an all-cause mortality hazard ratio of 0.89 (0.87-0.91), though melanoma was positively associated with death in some individuals. The authors noted causal conclusions cannot be made from these data.
Shows an association, not proof of cause and effect.
Brøndum-Jacobsen P, Nordestgaard BG, Nielsen SF, et al. Int J Epidemiol, 2013. PMID 24038635
2013 / Pharmacokinetic modeling with observational pre- and post-flight data
Researchers modified a physiologically based pharmacokinetic model to assess blood lead levels (PbB) resulting from microgravity-accelerated bone loss and spacecraft drinking water lead concentrations (PbW). The model predicted that for an average American astronaut, Pb release from bone is offset by reduced ingested or inhaled environmental Pb, causing PbB to decrease when PbW is below approximately 9 µg/L. Measured PbB in astronauts before and after 6-month missions on the International Space Station (ISS), where PbW averages 0.6 µg/L compared to 5 µg/L at Johnson Space Center, supported these results. Most astronauts will not be adversely affected, though a small percentage with high bone lead could be at risk depending on individual bone loss rates.
Shows an association, not proof of cause and effect.
Garcia HD, Hays SM, Tsuji JS Aviat Space Environ Med, 2013. PMID 24459792
2011 / Prospective longitudinal cohort study
Researchers evaluated whether plasma 25(OH)D concentrations change during a systemic inflammatory response in 33 patients undergoing primary knee arthroplasty, taking blood samples preoperatively, daily for 5 days postoperatively, and at 3 months. Preoperatively, most patients were 25(OH)D deficient (<50 nmol/L). By day 2 post-surgery, a large rise in C-reactive protein (P < 0.001) was accompanied by an approximate 40% reduction in 25(OH)D (P < 0.001), with levels remaining below baseline at day 5 (P < 0.001). At 3 months, 25(OH)D and calculated free 25(OH)D remained significantly lower by 20% and 30%, respectively (P < 0.01), indicating 25(OH)D may not reliably measure vitamin D status during systemic inflammation.
Shows an association, not proof of cause and effect.
Reid D, Toole BJ, Knox S, et al. Am J Clin Nutr, 2011. PMID 21411617
2010 / Cross-sectional histomorphometric autopsy study
Researchers analyzed 675 iliac crest biopsies from male and female autopsies without signs of secondary bone disease to assess mineralization defects and serum 25(OH)D levels. An osteoid surface per bone surface (OS/BS) over 20% was found in 36.15% of patients, and 25.63% had osteoid volume per bone volume (OV/BV) greater than 2%. Although no minimum 25(OH)D level was inevitably linked to mineralization defects, pathologic osteoid accumulation was not observed in any individual with circulating 25(OH)D above 75 nmol/L (30 ng/mL).
Shows an association, not proof of cause and effect.
Priemel M, von Domarus C, Klatte TO, et al. J Bone Miner Res, 2010. PMID 19594303
2010 / Cross-sectional multicenter study
Data from 3,075 men aged 40-79 years were analyzed to examine the link between musculoskeletal pain and 25-hydroxyvitamin D3 levels categorized as normal (>= 15 ng/ml) or low (< 15 ng/ml). Compared to pain-free participants (18.9% with low levels), low vitamin D was more frequent in those with 'other pain' (23.3%) and chronic widespread pain (CWP, 24.1%, p < 0.05). Age-adjusted models showed having 'other pain' was associated with a 30% increase and CWP with a 50% increase in the odds of low 25-(OH)D, persisting after adjusting for physical activity. Adjusting for BMI, smoking, alcohol use, and depression attenuated the link, though pain remained associated with a 20% increased odds of low vitamin D.
Shows an association, not proof of cause and effect.
McBeth J, Pye SR, O'Neill TW, et al. Ann Rheum Dis, 2010. PMID 20498201
2010 / Qualitative phenomenological study
Researchers used Husserlian phenomenological methodology to explicate the phenomenon of ginger compresses among 10 adults with osteoarthritis for at least a year. Data were collected using daily diaries, drawings, follow-up interviews, and telephone conversations. Seven themes emerged: meditative-like stillness and relaxation of thoughts; constant penetrating warmth throughout the body; positive change in outlook; increased energy and interest in the world; deeply relaxed state progressing to a gradual shift in pain and increased interest in others; increased suppleness within the body; and more comfortable, flexible joint mobility. The authors concluded that controlled research is needed with larger numbers of older people.
Shows an association, not proof of cause and effect.
Therkleson T J Adv Nurs, 2010. PMID 20626491
2010 / Cross-sectional study
Researchers analyzed NHANES 1999-2002 data from 2,671 women aged 20-85 to examine associations between bone turnover markers, micronutrient intake, and blood lead levels (BLLs). In adjusted analyses, postmenopausal women in the highest vs. lowest tertiles of NTx and BAP had 34% (95% CI, 23%-45%) and 30% (95% CI, 17%-43%) higher mean BLLs, respectively. Among premenopausal women, highest tertiles of NTx and BAP were linked to 10% (95% CI, 0.60%-19%) and 14% (95% CI, 6%-22%) higher BLLs. Calcium intake above the DRI was linked to lower mean BLLs in postmenopausal women across tertiles, but most associations were not statistically significant, with similar findings for vitamin D supplements.
Shows an association, not proof of cause and effect.
Jackson LW, Cromer BA, Panneerselvamm A Environ Health Perspect, 2010. PMID 20688594