Harvey 2012.

Methods	Randomised, double‐blind, placebo‐controlled trial.
Participants	1200 pregnant women living in the UK, aged 18 years old and older, with a singleton pregnancy with less than 17 weeks' gestation at first assessment (based on last menstrual period and dating scan), aiming to give birth at local maternity hospital, and with serum 25‐hydroxyvitamin D is 25 to 100 nmol/L at nuchal fold/dating scan (10 to 17 weeks' gestation).
Interventions	Participants were randomly assigned to 1 of 2 groups: group 1 (n = 565): received 1000 IU cholecalciferol orally daily and group 2 (n = 569): received placebo, starting from 14 weeks' gestation until delivery. Health worker cadre: the medication was blister packed in a single box for each woman for the duration of pregnancy. Study medication (active/placebo) was supplied to the local pharmacy pre‐randomised by the manufacturer (1:1, unstratified by centre) and sequentially numbered for storage and dispensing. Code break envelopes were supplied to the lead pharmacist, but were not available to the investigative team. Emergency code break access was available through the local principal investigator and on call pharmacist. A single pack for each participant was issued sequentially (containing all pills for duration of the study). Each pack was individually prescribed for each participant. The trials pharmacist allocated a pack to that prescription, documenting both the pack number and the MAVIDOS participant ID; these were checked again by the research nurse on collection, and documented in the participant’s notes; the medication pack came with a tear‐off adhesive label, which was placed in the participant’s notes as an added safeguard against errors in pack allocation. The research nurse collected the medication pack for all participants attending to the clinic that day and issued to the participants directly.
Outcomes	Infant: whole body bone mineral content of the neonate adjusted for gestational age and age at neonatal DXA scan, whole body bone area, bone mineral density, and size corrected bone mineral density (BMC adjusted for BA, length and weight), body composition adjusted for gestational age and age at DXA scan. Laboratory method used for assessment of vitamin D concentrations: A blood sample was taken and plasma was stored at ‐80°C for measurement of 25(OH)‐vitamin D, vitamin D binding protein (DBP), calcium, bone specific alkaline phosphatase and albumin centrally (MRC Human Nutrition Research, Cambridge, UK) at the end of the study.
Notes	Total dose of supplementary vitamin D during pregnancy: less than 200,000 IU; start of supplementation: less than 20 weeks of pregnancy; pre‐gestational BMI (kg/m2): unknown/mixed; supplementation scheme/regimen: daily; skin pigmentation based on Fitzpatrick skin tone chart (Fitzpatrick 1988): mixed/unknown; latitude: north of Tropic of Cancer; season at the start of pregnancy: all year round. Source of funding: Arthritis Research UK, Medical Research Council, Bupa Foundation, and National Institute for Health Research. Study was funded by a combination of research grants, government programmes, non‐governmental organizations (NGOs). Dates of the study and location: October 2008 to February 2014, Southampton, Sheffield, Oxford, the UK. Declarations of interest among primary researchers (or state where this information is not reported by the trial authors): CC reports personal fees, consultancy, lecture fees, and honoraria from Alliance for Better Bone Health, Amgen, Eli Lilly, GlaxoSmithKline, Medtronic, Merck, Novartis, Pfizer, Roche, Servier, and Takeda, outside the submitted work. NCH reports personal fees, consultancy, lecture fees, and honoraria from Alliance for Better Bone Health, AMGen, MSD, Eli Lilly, Servier, Shire, Consilient Healthcare, and Internis Pharma, outside the submitted work. NJB reports remuneration from Internis Pharmaceuticals, outside the submitted work. ATP reports grants from the Arthritis Research Council, during the conduct of the study. NKA has received honoraria, held advisory board positions (which involved receipt of fees), and received consortium research grants from Merck, grants from Roche, Bioiberica, and Novartis, personal fees from Smith & Nephew, Nicox, Flexion, Bioventus, and Freshfields, outside the submitted work. KMG reports reimbursement for speaking at Nestle Nutrition Institute conferences, and grants from Abbott Nutrition and Nestec, outside the submitted work. KMG also has a patent pending for phenotype prediction, a patent pending for predictive use of CpG methylation, and a patent pending for maternal nutrition composition, not directly related to this work. HMI reports grants from the Medical Research Council (MRC), Arthritis Research UK, and European Union’s Seventh Framework Programme, during the conduct of the study; and while not directly receiving funding from other bodies, members of her team have received funding from the following companies from other work: Danone, Nestec, and Abbott Nutrition. RE reports grants and personal fees from Amgen and Alexion; grants from the Department of Health, AstraZeneca, ARUK/MRC Centre for Excellence in Musculoskeletal Ageing Research, National Institute for Health Research, MRC/AZ Mechanisms of Diseases Call, and the MRC; grants, personal fees, and non‐financial support from Immunodiagnostic Systems; grants and membership of a clinical and scientific committee from the National Osteoporosis Society; grants, personal fees, and advisory board membership from Roche; personal fees from Otsuka, Novartis, Merck, Bayer, Johnson & Johnson, Fonterra Brands, Janssen Research, Ono Pharma, Alere (Unipath), Chronos, Teijin Pharma Limited, D‐STAR, and GSK Nutrition; personal fees and advisory board membership from Eli Lilly, and CL Biosystems; and advisory board membership from the European Calcifi ed Tissue Society, IOF CSA, and the American Society for Bone and Mineral Research, outside the submitted work. MKJ reports personal fees from Stirling Anglia, Consilient Health, and Internis, outside the submitted work. All other authors declare no competing interests.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated sequence in randomly permuted blocks of 10.
Allocation concealment (selection bias)	Low risk	The treatments were blister packed in a single box for each woman for the duration of pregnancy and supplied to the local pharmacy pre‐randomised by the manufacturer (1:1, unstratified by centre) and sequentially numbered for storage and dispensing. The lead pharmacist was the only one with access to the code break envelopes.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double‐blinded, matched pills, only lead pharmacist knew about pills.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	All assessments were double‐blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	94% of sample had vitD assessment at term.
Selective reporting (reporting bias)	Unclear risk	There is insufficient information to permit judgement.
Other bias	High risk	Participants were allowed to continue taking their own multivitamin with 400 IU/d of vitamin D but this was not recorded.