Objective and methods: To assess the validity of the family history obtained at the bedside of patients with recent subarachnoid haemorrhage by subsequently contacting all first and second degree relatives, with verification from medical record data.
Results: In a prospectively collected series of 163 patients with recent subarachnoid haemorrhage the history or cause of death could be ascertained in 1259 (98%) of the first degree relatives and in 3038 (85%) of the second degree relatives. For first degree relatives only, the sensitivity of the family history at the bedside was 0.75 (95% confidence interval (95% CI) 0.35-0.97) and the positive predictive value was 0.55 (95% CI 0.23-0.83); for first and second degree relatives together the sensitivity was 0.58 (95% CI 0.28-0.85) and the positive predictive value was 0.64 (95% CI 0.31-0.89).
Conclusion: The accuracy of the family history taken at the bedside is modest; a more thorough collection of data is crucial if the decision is taken to screen relatives based on the family history.