BACKGROUND AND OBJECTIVES: Although nutrition is important to bone health, the impact of different dietary patterns on bone density and fracture is unclear. The aim of this study was to synthesise conflicting evidence from observational studies to determine associations of empirically derived dietary patterns with bone density and fracture in healthy adults. METHOD: A systematic review (PROSPERO CRD42017071676) with meta-analysis where possible (for hip fracture) and otherwise with best-evidence synthesis. RESULTS: Twenty-one studies were included in the best-evidence synthesis and four in the meta-analysis. Meta-analysis demonstrated a protective association between 'healthy' pattern score and hip fracture (risk ratio 0.73; 95% confidence interval: 0.56, 0.96; I2 = 95%) for highest compared to lowest 'healthy' pattern score category. In best-evidence synthesis, there was conflicting evidence for associations of both pattern scores with bone density at all sites and total fractures and for 'Western' score and hip fracture. No study reported detrimental effects of 'healthy' patterns, or beneficial effects of 'Western' patterns. DISCUSSION: The results suggest that general practitioners promoting a 'healthy' dietary pattern is, at worst, unlikely to be detrimental for bone health and, at best, may reduce hip fracture.