Falls occur frequently in patients with a stroke and have serious consequences: discharge delays and hip fractures can result. In order to evaluate the impact of nursing workload on stroke-patient falls, we assessed the patients per nurse ratio in a detailed case-control study carried out in nine Dutch hospitals. There were 49 first falls by 349 stroke patients. Between the cases and controls there was no difference in the means of these ratios for any type of nurse and any type of patient on the day and evening shift. We found that the greatest number of falls (n = 26) occurred on the day shift when most nursing staff were present. On the night shift, we found a significant difference between the means of the patient per nurse ratios (95% confidence interval: 0.28-2.20), but the number of cases was very small. Of the falls, 35% occurred within the first week of admission. We found 19 cases in a subgroup of 138 patients that could be checked with the hospital incidents report committees. This review yielded three additional cases, but nine cases were not reported to these committees. We conclude that simply increasing number of nurses is not likely to diminish the number of falls.