Table 1.
Characteristics of included studies
| Reference, study design, study name (period) | Location, setting, funding | Population, baseline characteristics | Exposure (measurement), collection period | Follow-up period | Risk of bias |
|---|---|---|---|---|---|
| Afzal et al., 2014 [1], prospective cohort study, Copenhagen City Heart Study (1981–1983) | Denmark, general population (population register), public research funding and material sponsorship from Diarosin Liasion | 10,186 individuals without dementia Women % (n): 56.1% (5718) Age in y (median, range): total: n.s.; E1: 57 (47–64), E2: 58 (49–65), E3: 58 (50–65) |
Plasma 25(OH)D measurement: DiaSorin Liaison 25(OH)D total assay (Immunoassay) Sample collection: 1981–1983 stored until 2009 to 2010 E1: no vit D deficiency: ≥50 nmol/L (>20 ng/mL)a E2: vit D deficiency: ≥25 to <50 nmol/L (≥10 to 20 ng/mL) E3: serious vit D deficiency: <25 nmol/L (<10 ng/mL) |
Median 21 y (Range: 0.03–30 y) until diagnosis of AD, vascular dementia, death, emigration or May 2011 |
unclear |
| Annweiler et al., 2011 [2], prospective cohort study, EPIDémiologie de l’OStéoporose (EPIDOS) study Toulouse (1992–1994) | France, general population, public research funding | 40 women without dementia from the EPIDOS Toulouse study Women (n): 100% (40) Age in y (median, 25./75. percentile): 78.4 (76.4/82.0) |
Serum 25(OH)D measurement: Radioimmunoassay Sample collection: 1992–1994 E1: ≥25 nmol/L (≥10 ng/mL) E2: <25 nmol/L (<10 ng/mL) |
7 y | high |
| Graf et al. 2014 [3], prospective cohort study | Switzerland, geriatric hospital, public research funding and material sponsorship from AstraZeneca Switzerland (2004–2005) | 246 patients, of these 200 cognitively normal, 46 with mild cognitive impairment (MCI) Women % (n): 75.6% (147 cognitively normal and 39 MCI) Age in y (mean, SD): total: n.s.; cognitively normal: 84.4 (7.1), MCI: 85.3 (6.6) |
Plasma 25(OH)D measurement: Electrochemiluminescence-Immunoassay Sample collection: 2004–2005 E1: optimal vit D status: ≥75 nmol/L (≥30 ng/mL)a E2: sub-optimal vit D status: 50–75 nmol/L (20–30 ng/mL) E3: vit D insufficiency: 25–49,9 nmol/L (10–19,96 ng/mL) E4: vit D deficiency: <25 nmol/L (<10 ng/mL) Reclassification for Meta-Analysis: e1: no vit D deficiency: ≥50 nmol/L (≥20 ng/mL) e2: vit D deficiency: ≥25 to <50 nmol/L (≥10 to 20 ng/mL) e3: serious vit D deficiency: <25 nmol/L (<10 ng/mL) |
2 y | unclear |
| Knekt et al., 2014 [4], retrospective cohort study, Mini Finland Health Survey (1978–1980) | Finland, general population (population register), public research funding | 5010 subjects without hospitalisation due to dementia Women % (n): 54.7% (2738) Age in y (median): total: n.s.; E1: 54, E2: 55, E3: 56, E4: 59 |
Serum 25(OH)D measurement: Radioimmunoassay (DiaSorin) Sample collection: 1978–1980 stored until 2003 E1: 4. quartile: 54–159 nmol/L (21.6–63.6 ng/mL)a E2: 3. quartile: 40–53 nmol/L (16–21.2 ng/mL) E3: 2. quartile: 29–39 nmol/L (11.6–15.6 ng/mL) E4: 1. quartile: 7–28 nmol/L (2.8–11.2 ng/mL) |
17 y | unclear |
| Littlejohns et al., 2014 [5], prospective cohort study, Cardiovascular Health Study (1992–1993) | USA, 4 communities, ambulatory participants, private and public research funding, | 1658 subjects without dementia, cardiovascular diseases or stroke Women % (n): 69.2% (1148) Age in y (median, SD): 73.6 (4.5) |
Serum 25(OH)D measurement: LC-MS/MS Sample collection: 1992–1993 stored until 2008 E1: no vit D deficiency: ≥50 nmol/L (≥20 ng/mL) E2: vit D deficiency: ≥25 to <50 nmol/L (≥10 to 20 ng/mL) E3: serious vit D deficiency: <25 nmol/L (<10 ng/mL) |
Average 5.6 y (SD 1.6) | unclear |
| Schneider et al. 2014 [6], prospective cohort study, Atherosclerosis Risk in Communities (ARIC) Brain MRI Study (1993–1995) | USA, general population from 2 regions, public research funding | 1652 subjects (white or black ethnic background) without hospitalisation due to dementia, cardiovascular diseases or stroke Women % (n): 60.3% (996) Age in y (mean, SD): total: 62 (n.s.).; E1: Whites 63.1 (4.3); Blacks 62.2 (4.4), E2: Whites 63.3 (4.5); Blacks 61.4 (4.6), E3: Whites 62.9 (4.4); Blacks 61.0 (4.5) |
Plasma 25(OH)D measurement: LC-MS/MS Sample collection: 1993–1995 stored until 2012 E1: highest tertile Whites ≥70.8 nmol/La (≥28.3 ng/mL); Blacks ≥48.3 nmol/L (≥19.3 ng/mL); E2: middle tertile: Whites 54.5 to <70.8 nmol/L (21.8 to <28.3 ng/mL); Blacks 35.0 to <48.3 nmol/L (14.0 to <19.3 ng/mL) E3: lowest tertile: Whites <54.5 nmol/L (<21.8 ng/mL); Blacks <35.0 nmol/L (<14.0 ng/mL) Reclassification for Meta-Analysis: e1: no vit D deficiency: ≥50 nmol/L (≥20 ng/mL) e2: vit D deficiency: ≥25 to <50 nmol/L (≥10 to 20 ng/mL) e3: serious vit D deficiency: <25 nmol/L (<10 ng/mL) |
Median 16.6 y | unclear |
Abbreviations: AD Alzheimer’s disease, nmol/L nanomoles per litre, ng/mL nanograms per millilitre, SD standard deviation, Vit D Vitamin D, n number, n.s. not specified, E or e exposure, y years, LC-MS liquid chromatography-tandem mass spectrometry, MCI mild cognitive impairment
aTo convert 25(OH)D to nanomoles per litre from nanograms per litre, multiply values by 2.5