In "Epidemiology of bicycle injuries and risk factors for serious injury" by Frederick P Rivara, Diane C Thompson, and Robert S Thompson, the authors gave a questionnaire to 3,390 bicycle riders who had attended a hospital emergency department in the Seattle area.

They found that cyclists involved in a crash at a speed greater than 15 miles per hour were 1.4 times as likely to have a "severe" injury (defined as an injury severity score greater than 8) as cyclists involved in a crash at a speed of 15 miles per hour or less. The 95% confidence interval was 1.0 to 1.9.

(Caveat: It seems strange to me that the authors only reported odds ratios, when they apparently had the data to compute effect sizes too.)

In "Epidemiology of bicycle injuries and risk factors for serious injury" by Frederick P Rivara, Diane C Thompson, and Robert S Thompson, the authors gave a questionnaire to 3,390 bicycle riders who had attended a hospital emergency department in the Seattle area.

They found that cyclists involved in a crash at a speed greater than 15 miles per hour were 1.4 times as likely to have a "severe" injury (defined as an injury severity score greater than 8) as cyclists involved in a crash at a speed of 15 miles per hour or less. The 95% confidence interval was 1.0 to 1.9.

(Caveats: 1. It seems strange to me that the authors only reported odds ratios, when they apparently had the data to compute effect sizes too. 2. The fast cyclists differ systematically from the slower cyclists: the odds ratio fell to 1.2 when other factors in the study were controlled for. 3. It's Rivara, Thompson and Thompson.)