Table 1.
Baseline characteristics of studies included in the meta-analysis
| First author | Publication year | Region | Study design | Study name | Subjects (cases) | Sex | Mean/median age (years) | Follow up period (years) | Diet assessment | Exposure | Quantity | OR/RR (95% CI) | Ascertainment method of hip fracture | Adjustment for confounders |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Feskanich | 2014 | USA | cohort study | The Nurses’ Health Study (NHS); the Health Professionals Follow-up Study (HPFS) | 96,927 (1716) | M/F | F (30–55) M (40–75) | 22 | FFQ | Milk | ≥ 4 vs. 1 glass/day | Males 1.21(0.86–1.64) Females 1.01(0.78–1.31) | Self -reporting | Age, questionnaire cycle, adult milk consumption, calcium supplements, vitamin D supplements, retinol from supplements, total protein, alcohol and caffeine intakes, total energy intake, physical activity, BMI, smoking, use of thiazide diuretics, use of furosemide diuretics and oral steroids (men only], use of hormone replacement therapy, incident diagnoses of osteoporosis and cancer, teenage measures |
| Michaëlsson | 2014 | Sweden | cohort study | The Swedish Mammography Cohort; the Cohort of Swedish Men | 106,772 (5425) | M/F | 56.5 (39–79) | F: 20.1 M:11.2 | FFQ | Milk | ≥ 600 vs. < 200 g/day | Males: 1.01 (0.85–1.20) Females: 1.60(1.39–1.84) |
Registers | Age, BMI, height, total energy intake, total alcohol intake, healthy dietary pattern, calcium and vitamin D supplementation, ever use of cortisone, educational level, living alone, physical activity level estimated as metabolic equivalents, smoking status, and Charlson comorbidity index; for women only, use of estrogen replacement therapy and nulliparity |
| Yogurt | ≥ 400 vs. < 1 g/day | Males: 0.75 (0.63–0.90) Females: 0.70 (0.57–0.86) |
||||||||||||
| Cheese | ≥ 60 vs. < 20 g/day | Males: 0.75 (0.62–0.92) Females: 0.64 (0.55–0.74) |
||||||||||||
| Sahni | 2014 | USA | cohort study | The Framingham Original Cohort | 764 (97) | M/F | 76.9 (68–96) | 11.6 | Validated FFQ | Milk | ≥ 7 vs. ≤ 1 servings/week | 0.58 (0.31–1.06) | Self-reporting confirmed by review of medical records and radiographic and operative reports | Age, sex, weight, height, total energy intake, current cigarette smoking, calcium supplement use, vitamin D supplement use |
| Yogurt | > 0 vs. 0 servings/week | 1.09 (0.65–1.81) | ||||||||||||
| Cheese | > 1 vs. ≤ 1 servings/week | 0.72 (0.48–1.08) | ||||||||||||
| Cream | ≥ 3 vs. < 1 servings/week | 1.04 (0.59–1.86) | ||||||||||||
| Benetou | 2011 | European | cohort study | The Cancer and nutrition (EPIC) study | 29,122 (275) | M/F | 64.3 (60–86) | 8 | Validated FFQ | Total dairy products | Yes vs. no | 1.02 (0.93–1.12) | Both registers and self-reporting | Sex, age, BMI, height, educational level, smoking status, physical activity at leisure, supplement use, history of diabetes at enrolment, total energy intake |
| Feart | 2013 | France | cohort study | The Three-City (3C) study | 1482 (57) | M/F | 75.9 (67.7–94.9) | 8 | FFQ and a 24-h dietary recall | Dairy products | Highest vs. lowest category | 1.05 (0.60–1.85) | Self-reporting | Each individual food group component of the Mediterranean diet score, age, gender, physical activity, total energy intake, educational level, marital status, BMI, osteoporosis treatment, calcium and/or vitamin D treatment |
| Milk | Highest vs. lowest category | 0.86 (0.50–1.49) | ||||||||||||
| Yogurt | Highest vs. lowest category | 0.90 (0.50–1.61) | ||||||||||||
| Cheese | Highest vs. lowest category | 0.78 (0.44–1.39) | ||||||||||||
| Cumming | 1997 | USA | cohort study | Study of Osteoporotic Fractures (SOF Study) | 9704 (306) | F | 71 (65+) | 6.6 | Validated FFQ | Milk | ≥ 3 vs. rarely/never glasses/day | 0.90 (0.50–1.70) | Self-reporting | Age, clinic, weight, history of osteoporosis, history of fractures since age 50, fall in past 12 months, protein intake, caffeine intake, recreational physical activity, take walks for exercise, impaired low frequency contrast sensitivity, estrogen replacement therapy, thiazide use, use of calcium and Vitamin D supplements, use of Turns antacid tablets |
| Fujiwara | 1997 | Japan | cohort study | the Adult Health Study (AHS) | 4573 (55) | M/F | 58.5 | 14 | Standardized questionnaire | Milk | ≥ 5 vs. ≤ 1 times/week | 0.54 (0.25–1.07) | Registers | Age, alcohol, BMI, prevalent vertebral fracture, number of children, age at menarche |
| Meyer | 1997 | Norway | cohort study | the National Health Screening study | 39,787 (213) | M/F | 47.1 (42.9–65.9) | 11.4 | FFQ | Milk | ≥ 5 vs. < 1 glasses/day | Males: 0.46 (0.22–0.98) Females: 0.83 (0.44–1.56) |
Self-reporting confirmed by review of medical records and radiographic and operative reports | Age, body height, BMI, self-reported physical activity at work and during leisure time, diabetes mellitus, disability pension, marital status, smoking |
| Owusu | 1997 | USA | cohort study | The Health Professionals Follow-up Study | 43,063 (56) | M | 54 (40–75) | 8 | Validated FFQ | Milk | 2.5 vs. ≤1 glass/week | 0.97 (0.39–2.42) | Self-reporting | Age, alcohol consumption, smoking, BMI, physical activity, total energy intake |
| Kanis | 2004 | Europe, Australia, and Canada |
Meta-analysis of cohort study | The European Vertebral Osteoporosis Study (EVOS);The Canadian Multicentre Osteoporosis Study (CaMos);The Dubbo Osteoporosis Epidemiology Study (DOES);The Rotterdam Study;The Sheffield Study;The Gothenburg study | 39,563 (413) | M/F | 66.7(58.9–80.0) | 3–8 | NR | Milk | 'Highest vs. lowest category of consumption | Males:0.66 (0.39–1.12) Females: 0.92 (0.69–1.22) |
Both registers and self-reporting | NR |
| Jha | 2010 | India | case-control study | NR | 200 (100) |
M/F | 65.2 | NR | Standardized questionnaire | Milk, | > 1 vs. ≤ 1 glass/day | 0.30 (0.13–0.72) | Self-reporting | NR |
| Cheese | > 1 vs. ≤ 1 servings/week | 0.48 (0.24–0.93) | ||||||||||||
| Yogurt | > 2 vs. ≤ 2 cups/week | 0.77 (0.39–1.51) | ||||||||||||
| Lan | 2010 | Taiwan | case-control study | NR | 725 (228) | M/F | (60+) | NR | Standardized questionnaire | Milk | ≥ 6 vs. none or <1 drink/week | 0.58 (0.37–0.91) | Self-reporting | Socio-demographic, disease history, self-assessed health, anthropometry and health habits, diet habits, injury-related experience, physical functioning, cognitive and other functioning, physical performance, female reproductive history, bone mineral density |
| Jitapunkul | 2001 | Thai Chinese | case-control study | NR | 120 (60) | F | 71.4 | NR | Structured questionnaire | Milk | Yes vs. no | 0.26 (0.09–0.76) | Self-reporting | No regular intake of milk, low number of pregnancies, thin body appearance, low serum calcium |
| Kanis | 1999 | Southern Europe | case-control study | The MEDOS study | 1862 (730) | M | 74 | NR | Standardized questionnaire | Milk | Highest vs. lowest category of consumption | 0.82 (0.61–1.11) | Registers | BMI, recreational physical activity, consumption of tea, alcohol consumption, coffee consumption, smoking, sunlight exposure |
| Cheese | Yes vs. no | 0.75 (0.49–1.14) | ||||||||||||
| Johnell | 1995 | Southern Europe | case-control study | The MEDOS Study | 5618 (2086) |
F | 77.8 | NR | Standardized questionnaire | Milk | Highest vs. lowest category | 0.71 (0.58–0.87) | Registers | Mental score, BMI, menarche, menopause, sunlight exposure, tea consumption |
| Tavani | 1995 | Italy | case–control study | NR | 960 (241) | F | 63.3 (45–74) | NR | Standardized questionnaire | Milk | > 7 vs. < 7 drinks/week | 1.00 (0.60–1.60) | Registers | Age, education, BMI, estrogen replacement therapy |
| Cheese | > 6 vs. < 4 portions/week | 1.00 (0.70–1.50) | ||||||||||||
| Cumming | 1994 | Australia | case-control study | NR | 416 (209) | M/F | (65–100) | NR | Standardized questionnaire | Dairy product | Highest vs. lowest category | 1.70 (0.50–5.40) | Registers | Age, sex, country of birth, mental state score, psychotropic drug use, smoker status, work in the house and garden, weight |
| Nieves | 1992 | USA | case-control study | NR | 329 (161) | F | (50–103) | NR | Validated FFQ | Milk | ≥ 7 vs. no glasses/week | 1.10 (0.63–1.94) | Registers | Hospital site, age and BMI, smoking status, alcohol consumption |
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); F, Female; M, male; FFQ, food frequency questionnaire; NR, not reported