8,456 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.
Patterns in how people eat and what happens to them. Shows an association, not proof.
1983 / Case-control study
Researchers analyzed food items and food groups in a case-control study including 348 colon cancer cases, 194 rectal cancer cases, 542 matched neighborhood controls, and 535 surgical controls. For colon cancer, saturated fat was the major risk factor, with no protective effect of dietary fiber and only a minor protective effect of cruciferous vegetables in females, to which no individual vegetable made an important contribution. For rectal cancer, a significant effect of saturated fat was observed only in females in the highest consumption category. Rectal cancer risk was significantly increased by eggs, beef, and veal in males, and by pork and beer in females, while dietary fiber and cruciferous vegetables showed no protective effect.
Shows an association, not proof of cause and effect.
Miller AB, Howe GR, Jain M, et al. Int J Cancer, 1983. PMID 6307893
1983 / Matched case-control study
A case-control study in Athens, Greece, evaluated the dietary habits of 100 patients with histologically confirmed colorectal cancer and 100 matched orthopaedic controls using dietary histories of roughly 80 food items. Cases reported significantly less frequent consumption of vegetables (particularly beets, spinach, lettuce, and cabbage) and significantly more frequent consumption of meat (notably lamb and beef). Comparing a high-meat, low-vegetable diet to a high-vegetable, low-meat diet, an estimated risk ratio of about 8 was observed. No significant associations were observed for beer or other alcoholic beverages, and no significant interactions were detected for age, sex, or anatomic site.
Shows an association, not proof of cause and effect.
Manousos O, Day NE, Trichopoulos D, et al. Int J Cancer, 1983. PMID 6862688
1982 / Cross-sectional ecological comparison
Nonstarch polysaccharide (NSP) intake was assessed in 30 men aged 50-59 in two urban and two rural Scandinavian populations with a 3-4 fold difference in large bowel cancer incidence using chemical analysis of 1-day duplicate diets. Intakes were 18.4 +/- 7.8 g/day in rural Finland and 18.0 +/- 6.4 g/day in rural Denmark, compared to 14.5 +/- 5.4 g/day in urban Finland and 13.2 +/- 4.8 g/day in urban Denmark. Calculated intakes from 4-day weighed records using food tables were 2-2.5 g/day higher than direct chemical analysis. Average NSP intake and intake of some component sugars were inversely related to colon cancer incidence in this geographical comparison.
Shows an association, not proof of cause and effect.
Englyst HN, Bingham SA, Wiggins HS, et al. Nutr Cancer, 1982. PMID 6296797
1982 / Comparative observational study
Mean faecal urobilinogen levels and stool pH were compared between subjects from a population group at high risk of developing colon cancer and age-, sex-, and socioeconomic-matched subjects from a low-risk group. Both mean faecal urobilinogen levels and stool pH were higher in the high-risk population group. The authors suggest that an alkaline colon reaction may have a tumorigenic effect whereas an acidic reaction appears protective, with differences dependent on dietary patterns such as roughage, vegetable fibre, and fermented milk products.
Shows an association, not proof of cause and effect.
Malhotra SL J R Soc Med, 1982. PMID 7120255
1981 / Observational study
The relation between frequency of bowel movements and epithelial dysplasia in nipple aspirates of breast fluid was examined in 1481 white women. Women reporting severe constipation (two or fewer bowel movements weekly) had a significant positive association with dysplasia (risk ratio 4.5; 95% confidence interval 1.9-11.9), which was not observed in women with more than one bowel movement daily. Increased risk ratios were also observed in women who had one bowel movement daily or one every other day.
Shows an association, not proof of cause and effect.
Petrakis NL, King EB Lancet, 1981. PMID 6118633
1981 / Comparative observational study
Bowel transit time was examined in vegetarians and non-vegetarians and related to dietary fibre intake and diverticular disease. Vegetarians, who have less diverticular disease than non-vegetarians, had more rapid transit times. Participants with a total dietary fibre intake above 30 g/d all had transit times below 75 h, compared to 38% of those consuming less fibre whose transit times exceeded 75 h (up to 124 h). Individuals with diverticular disease demonstrated faster transit times than those without the disease.
Shows an association, not proof of cause and effect.
Gear JS, Brodribb AJ, Ware A, et al. Br J Nutr, 1981. PMID 6258626
1981 / Observational study
Researchers examined the effects of dietary fiber and personality factors on stool output. Total neutral detergent fiber was predictive of stool weight but not stool frequency. Substantial individual differences in stool output remained after controlling for dietary factors. Personality measures predicted stool weight and frequency independently of diet, accounting for about as much variance in stool output as dietary fiber.
Shows an association, not proof of cause and effect.
Tucker DM, Sandstead HH, Logan GM, et al. Gastroenterology, 1981. PMID 6269944
1980 / Methodological and controlled-diet metabolic study
A method was developed to fractionate human faeces into microflora, dietary-fibre residues, and water-soluble material. In nine healthy subjects consuming a metabolically controlled British-type diet, bacteria accounted for 54.7 +/- 1.7% of total solids, fibre accounted for 16.7 +/- 0.8%, and soluble material accounted for 24.0 +/- 1.3%. The microbial fraction contained 95% of total bacteria, had 6% nitrogen, and accounted for 60% of total faecal nitrogen, indicating bacteria represent a much larger proportion of faecal mass than previously thought.
Shows an association, not proof of cause and effect.
Stephen AM, Cummings JH J Med Microbiol, 1980. PMID 7359576
1979 / Case series
Between 1967 and 1977, fasting therapy was administered to approximately 382 patients with psychosomatic diseases in a Japanese clinic. The therapy achieved a reported efficacy rate of 87%. Suitable indications included irritable colon, eating behavioral disturbances, functional gastrointestinal disorders, neurocirculatory asthenia, labile hypertension, puberty-related psychosomatic symptoms, conversion hysteria, and neuroses with somatic complaints.
Shows an association, not proof of cause and effect.
Suzuki J, Yamauchi Y, Yamamoto H, et al. Psychother Psychosom, 1979. PMID 482552
1979 / Cross-sectional comparative study
A study evaluated the relationship between symptomless diverticular disease of the colon and dietary fibre consumption in vegetarians and non-vegetarians. Vegetarians had a significantly higher mean fibre intake (41.5 g/day) compared to non-vegetarians (21.4 g/day). Diverticular disease was found in 33% of non-vegetarians compared with 12% of vegetarians. Comparisons between participants with and without the condition provided further evidence that a low intake of cereal fibre is associated with the presence of diverticular disease.
Shows an association, not proof of cause and effect.
Gear JS, Ware A, Fursdon P, et al. Lancet, 1979. PMID 85104
1979 / Ecological study
Using food disappearance data from 1954-65 and mortality data from 1967-73 across 20 industrialised countries, researchers examined independent relationships between diet variables and colon cancer. Simple correlation analysis showed total fat, saturated fat, monounsaturated fat, cholesterol, and fibre intakes were each highly correlated with colon-cancer mortality. The partial correlation of dietary cholesterol with colon cancer remained highly significant when controlling for fat or fibre, whereas correlations for fat or fibre were no longer significant when cholesterol was controlled. Cross-classification showed a highly significant main effect for cholesterol, but not for fat or fibre.
Shows an association, not proof of cause and effect.
Liu K, Stamler J, Moss D, et al. Lancet, 1979. PMID 90870
1977 / Cross-sectional comparative study
1259 Tanzanians in a provincial town were examined for varicose veins. Among 1000 people aged 18 years and over, 5.5% (6.1% of men and 5.0% of women) had varicose veins, which is intermediate between rates reported for minimally developed areas (0.1%-0.4%) and Michigan (12.3%). The diet included some refined carbohydrate, and the mean daily stool weight was 228 g, falling between traditional African communities (490 g) and Western countries (115 g). These findings are consistent with the hypothesis relating varicosities to a fiber-depleted diet, while life-long use of the squatting position for defecation did not protect against varicose veins.
Shows an association, not proof of cause and effect.
Richardson JB, Dixon M Lancet, 1977. PMID 66583
1973 / Correlation study
The study examined the correlation between cancers of the breast, colon, and stomach, dietary factors, and various indicators of standard of living. Cancer of the breast and colon was found to be highly correlated with fat and animal protein.
Shows an association, not proof of cause and effect.
Drasar BS, Irving D Br J Cancer, 1973. PMID 4694388
1969 / Ecological correlation study
This study evaluated the hypothesis of a relationship between gastrointestinal cancer mortality and living standards using correlation analysis. An inverse relationship was observed between gastric and intestinal cancer mortality rates (r = -0.85). Higher food intake, measured as animal protein intake, correlated with lower gastric cancer mortality but higher intestinal cancer mortality. The authors noted that these correlation findings do not demonstrate causality or a direct effect on gastrointestinal cancer, and that no time lag was proven between food intake and intestinal cancer mortality.
Shows an association, not proof of cause and effect.
Gregor O, Toman R, Prusová F Gut, 1969. PMID 5366239