Table 2.

Estimated sample sizes per group, based on different outcome measures. For function and symptoms, standard deviations are from the Maine Lumbar Spine study, with 3-month follow-up (roughly equal proportions treated surgically and non-surgically). Return to work proportion is based on a study by the Washington State Department of Labor and Industries.

Category	Measure	Hypothesized difference between groups	N/group, 2-sided alpha = 0.05, power = 0.80
Functional Status	Difference in change in Roland score (0–23 scale, SD = 7.37)	2.5 points	136
		3.0 points	95
Symptoms	Improvement in pain (1 = gone, 5 = same, 7 = much worse) (SD = 1.841)	0.75 point	94
		1.0 point	53
Return to Work	Proportion currently employed, if lower proportion is 0.40	0.15	173
		0.20	97