Follow-up score, change score or percentage change score for determining clinical important outcome following surgery? An observational study from the Norwegian registry for Spine surgery evaluating patient reported outcome measures in lumbar spinal stenosis and lumbar degenerative spondylolisthesis

Table 4 Sensitivity and specificity for estimated cut-off values. Correct classification rate in different PROM baseline groups

	Spinal stenosis				Degenerative spondylolisthesis
	Estimated cut-off	Correct classification	Sensitivity	Specificity	Estimated cut-off	Correct classification	Sensitivity	Specificity
ODI follow-up score	≤24		0.83	0.80	≤24		0.85	0.84
Low baseline		80%				87%
Medium		85%				85%
High baseline		80%				84%
ODI change score	≥13		0.78	0.77	≥13		0.83	0.71
Low baseline		72%				77%
Medium		84%				86%
High baseline		78%				75%
ODI percentage change	≥30		0.87	0.77	≥30		0.89	0.77
Low baseline		83%				88%
Medium		85%				85%
High baseline		80%				82%
EQ-5D follow-up score	≥0.692		0.76	0.81	≥0.692		0.80	0.88
Low baseline		75%				81%
Medium		79%				80%
High baseline		80%				82%
EQ-5D change score	≥0.105		0.77	0.68	≥0.105		0.81	0.63
Low baseline		73%				74%
Medium		75%				80%
High baseline		72%				71%
Leg pain follow-up score	≤3		0.80	0.83	≤3		0.79	0.78
Low baseline		82%				81%
Medium		82%				76%
High baseline		81%				79%
Leg pain change score	≥3		0.78	0.75	≥3		0.78	0.72
Low baseline		69%				70%
Medium		82%				76%
High baseline		78%				80%
Leg pain percentage change	≥40		0.82	0.80	≥40		0.80	0.73
Low baseline		79%				75%
Medium		81%				76%
High baseline		81%				81%
Back pain follow-up score	≤4		0.84	0.79	≤4		0.85	0.79
Low baseline		81%				82%
Medium		83%				80%
High baseline		82%				87%
Back pain change score	≥2		0.82	0.66	≥3		0.78	0.80
Low baseline		72%				67%
Medium		83%				81%
High baseline		71%				83%
Back pain percentage change	≥33%		0.81	0.79	≥33%		0.83	0.78
Low baseline		76%				78%
Medium		83%				80%
High baseline		80%				85%

The sensitivity describes the probability of correctly classifying an individual replying ‘completely recovered’ or ‘much improved’ (GPE) as a ‘success’ when assessed by the estimated cut-offs for the PROMs. The specificity describes the probability for detecting a ‘non-success’ patient (one with a lower response at the GPE scale)
For each estimated cut-off values the percentage of correctly classified patients (correct classification) into ‘success’ and ‘non-success’ according to the anchor are given separately for patients with low (ODI; 0–32, EQ-5D; −0.59-0.1, NRS leg and back pain; 0–5), medium (ODI; 32–46, EQ-5D; 0.1–0.6, NRS leg and back pain; 6–7), and high (ODI; 46 to 100, EQ-5D; 0.6–1.0, NRS leg and back pain; 8–10) baseline scores

ISSN: 1471-2474