Mapping AcroQoL scores to EQ-5D to obtain utility values for patients with acromegaly

Figures & data

Table 1. Patient characteristics in the selected studies.

	Webb et al.^13 (2006); Study 1	Tiemensma et al.^19 (2011); Study 2	Paisley et al.^20 (2007); Study 3
Number of patients	106	72	80
Gender, male	36%	56%	59%
Age, years, mean (SD)	50 (13)	60 (12)	55 (15)
Transsphenoidal surgery	NA	85%	63%
Radiotherapy	NA	24%	68%
Somatostatin analog therapy	43%	36%	34%
Pegvisomant therapy	0	10%	18%
Duration of follow-up, years	NA	16 (10)	NA
Uncontrolled disease	48%	0%	66%
Hypopituitarism	NA	37%	NA
AcroQoL Global, mean (SD)	61.4 (19.2)	66.1 (15.6)	57.3 (17.2)
Physical Dimension, mean (SD)	57.6 (23.9)	62.7 (21.3)	55.1 (19.9)
Psychological Dimension, mean (SD)	63.5 (19.0)	68.1 (14.8)	58.6 (17.9)
EQ-5D utilities, mean (SD)	0.7 (0.3)	0.7 (0.3)	0.7 (0.2)

NA, not assessed.

Table 2. Score distributions on the AcroQoL and EQ-5D.

Values	Mean	Theoretical range	Standard deviation	Minimum	Median	Maximum	Valid n
AcroQoL Global	62.29	0–100	18.48	11.36	63.64	100.00	182
Physical Dimension	59.44	0–100	22.91	6.25	59.38	100.00	184
Psychological Dimension	63.97	0–100	18.49	14.29	66.07	100.00	182
Appearance	53.32	0–100	22.06	3.57	53.57	100.00	183
Relationship	74.45	0–100	19.15	14.29	78.57	100.00	174
EQ-5D utilities	0.71	–0.594–1	0.28	–0.135	0.72	1	184

Table 3. Regression coefficients for the best four models (Model 11 was the model finally selected).

		Model 1	Model 11	Model 12	Model 13
Intercept		0.1729 (0.0397)	0.30778 (0.0568)	0.98733 (0.0456)	0.93615 (0.0234)
Predicted utility**
Physical score		0.00896 (0.0006)	0.00734 (0.0008)
AcroQoL item 03	Category 1			0.06538 (0.1177)	–0.20466 (0.0588)
	Category 2			–0.22476 (0.0694)	–0.20466 (0.0588)
	Category 3			–0.11011 (0.0446)	–0.13133 (0.0330)
	Category 4			0.01747 (0.0467)
AcroQoL item 09	Category 1		–0.23227 (0.0735)	–0.43201 (0.0946)	–0.39247 (0.0861)
	Category 2			–0.17030 (0.0574)	–0.17294 (0.0497)
	Category 3			–0.15940 (0.0472)	–0.14986 (0.0368)
	Category 4			–0.06776 (0.0458)
AcroQoL item 14	Category 1		–0.07128 (0.0335)	–0.21848 (0.0656)	–0.19533 (0.0488)
	Category 2		–0.07128 (0.0335)	–0.13973 (0.0554)	–0.11843 (0.0358)
	Category 3			–0.03822 (0.0518)
	Category 4			–0.02927 (0.0601)
AcroQoL item 22	Category 1			–0.32155 (0.0993)	–0.19668 (0.0788)
	Category 2			–0.06924 (0.0641)
	Category 3			–0.02607 (0.0512)
	Category 4			–0.00708 (0.0497)
Results	R^2	0.5316	0.5646	0.6020	0.5726
	Adj R^2	0.5290	0.5573	0.5630	0.5526
	RMSE	0.1932	0.1873	0.1875	0.1827

Table 4. Analysis of residuals in the different models tested.

	Model 1	Model 11	Model 12	Model 13
n	184	184	180	180
Observed EQ-5D Score	0.705 (0.282)	0.705 (0.282)	0.704 (0.284)	0.704 (0.284)
Predicted EQ-5D Score	0.705 (0.205)	0.705 (0.211)	0.704 (0.220)	0.704 (0.215)
Observed Range	–0.135–1	–0.135–1	–0.135–1	–0.135–1
Predicted Range	0.229–1.069	0.050–1.042	–0.095–1.005	0.020–0.936
Predicted Range (max utility =1)	0.229–1	0.050–1	–0.095–1	0.020–0.936
Error (Quartiles)	–0.069–0.123	–0.055–0.113	–0.068–0.104	–0.087–0.075
Error (Median)	0.035	0.027	0.032	0.046
Mean Error	0.000 (0.193)	0.000 (0.186)	0.000 (0.179)	–0.000 (0.185)
Mean Absolute Error	0.143 (0.129)	0.133 (0.129)	0.127 (0.126)	0.137 (0.125)
AIC	–63.2	–68.1	–32.2	–53.2

Figure 1. Scatterplot of residuals in the selected model.

Table 5. Confirmatory analysis for the different models tested.

	Model 1	Model 11	Model 12	Model 13
n	61	61	60	60
Predicted Range in Construction Sample	0.229–1.069	0.050–1.042	–0.095–1.005	0.020–0.936
Range predicted in Validation sample	0.341–1	0.374–1	0.052–0.987	0.167–0.936
Mean Error	0.046 (0.174)	0.030 (0.168)	0.013 (0.185)	–0.008 (0.174)
Mean Absolute error	0.137 (0.116)	0.128 (0.112)	0.132 (0.130)	0.124 (0.121)
Normality (KS)	<0.01	<0.01	<0.01	<0.01
Correlation (Pearson)	0.598	0.618	0.585	0.570
Correlation (Spearman)	0.723	0.732	0.713	0.744

Webb SM, Badia X, Surinach NL. Spanish AcroQol study group. Validity and clinical applicability of the acromegaly quality of life questionnaire, AcroQoL: a 6-month prospective study. Eur J Endocrinol 2006;155:269-77

 PubMed Web of Science ®Google Scholar

Tiemensma J, Kaptein AA, Pereira AM, et al. Affected illness perceptions and the association with impaired quality of life in patients with long-term remission of acromegaly. J Clin Endocrinol Metab 2011;96:3550-8

 PubMed Web of Science ®Google Scholar

Paisley AN, Rowles SV, Roberts ME, et al. Treatment of acromegaly improves quality of life, measured by AcroQol. Clin Endocrinol (Oxf) 2007;67:358-62

 PubMed Web of Science ®Google Scholar