Skip to main content

Table 4 EQ-5D responsiveness results

From: A review of the psychometric performance of the EQ-5D in people with urinary incontinence

Author(s), Year

Comparison

Change in clinical measure(s) or other preference based utility

Change in EQ-5D

Agreement with direction?

Agreement with statistical significance?

Ismail et al, 2009 [16]

Change over time

No significant change on any measure (KHQ,1 hr pad test, pad use, leakage episodes)

No significant change

NA

Yes

Rinnie et al, 2008 [22]

Change over time

24 hr pad test significantly improved in both arms

Significant improvement in both arms

Yes

Yes

All condition specific measures (UISS, DIS, VAS, IIQ-7, UDI-6) significantly improved in both treatment groups

   

EQ-VAS significantly improved in both treatment groups

Difference between treatment arms

No significant difference in objective cure, leakage, complication rate, UISS, DIS, VAS, IIQ-7, UDI-6.

No significant difference in EQ-5D

Agreement with some clinical outcomes and not others.

Yes

Haywood et al 2008 [7]

Comparison of means for responders and non-responders

6 week data:

6 week data:

6 week data:

6 week data:

SSI and I-QoL index had difference in expected direction but not statistically significant (at p=0.01). Two of the I-QoL domains had significant difference.

EQ-5D had difference in expected direction but not statistically significant (at p=0.01).

Yes

Not consistent with all

5 mth data:

5 mth data:

5 mth data:

5 mth data:

As for 6 weeks except only one of the I-QoL domains had significant (p<0.01) difference.

EQ-5D had difference in expected direction and statistically significant (p=0.01).

Yes

Not consistent with all.

Mean change scores for patients reporting improvement

6 week data:

6 week data:

6 week data:

6 week data:

Expected direction and significant (at p=0.05) for SSI, I-QoL index, I-QoL domains

Expected direction but p>0.05

Yes

No

5 mth data:

5 mth data:

5 mth data:

5 mth data:

As for 6 weeks but larger changes.

Expected direction and p<0.05.

Yes

Yes

MSRM for patients reporting improvement

6 week data:

6 week data:

6 week data:

6 week data:

SSI, 0.70

0.07

Yes

No

I-QoL index, 1.01

   

I-Qol domains, 0.40 to 0.94

   

5 mth data:

5 mth data:

5 mth data:

5 mth data:

SSI, 0.67

0.26

Yes

Yes

I-QoL index, 1.17

   

I-Qol domains, 0.80 to 1.25

Kobelt et al, 2006 [21]

Median incontinence episodes per day for clinical outcome but change from baseline for EQ-5D

All patients:

All patients:

All patients

All patients

3.0 at baseline, 0.7 at 3mths and 0.9 at 12 mths (p<0.0001 and p<0.001 for differences)

3 mths: 0.048 (p<0.001)6 mths: 0.014 (not significant)

3 mths: Yes

3 mths: Yes

12 mths: “gain remained evident”

12 mths: Yes

12 mths: Yes

Patients with utility<1 at baseline:

Patients with utility <1 at baseline:

Patients with utility <1 at baseline:

3 mths: 0.099 (p<0.01)

6 mths: 0.065 (p<0.001)

  

12 mths: “significant improvements”

As for all patients

As for all patients

Dumville et al, 2006 [17]

Difference between treatment arms:

Objective and subjective cure rates and SF-36 scores showed no significant difference

QALY gain based on EQ-5D utility scores showed no significant difference (CrI crossed zero)

No change in either clinical, generic HRQoL or utility

Yes

Manca et al, 2003 [18]

Differences from baseline to 6mths

Pad weight decreased significantly for both groups.

Utility increased in both arms (significance not reported)

Yes

Not reported

Significant reduction in leakage episodes in both groups (P<0.0001)

Significant reduction in 21/30 symptoms (BFLUTS) in both groups (P<0.0001)

Differences between trial arms:

No significant difference in objective or subjective cure rate between trial arms

QALY difference between arms based on EQ-5D scores non significant at p=0.05

Agreement with clinical outcomes but didn’t detect differences between arms in some SF-36 domains

Yes for clinical outcomes, no for some SF-36 domains

SF-36 scores had significantly smaller improvement/ greater decline lower for colposuspension group vs TVT in four domains at 6 weeks and four domains (three same and one different) at 6 mths.

 

Noble et al, 2002 [11]

Change from baseline:

Improvements in I-PSS, maximum urine flow, and residual volume were significant (p=0.05) for laser and resection but not conservative.

Means increased for laser and resection but not conservative (p values not reported)

Yes

Not reported

Improvements in I-PSS QoL were significant for all three interventions.

Differences between trial arms:

Resection vs conservative and laser vs conservative showed significant difference in all four outcomes.

Gains were greater for resection than laser therapy (p values not reported)

Yes

Not reported

Laser vs resection showed significant difference in only one outcome which was in favour of resection (maximum flow)

Mihaylova et al, 2010 [23]

Comparison between active treatment arms and no treatment:

Number of leaks avoided per week was significantly (p<0.01) better for Duloxetine alone, conservative alone and duloxetine plus conservative (all relative to no treatment).

QALY gains based on EQ-5D utility were significant for Duloxetine alone (p<0.01) and duloxetine plus conservative treatment (p<0.05) but conservative alone was not significant and was negative (all compared to no treatment)

Yes for two of three comparisons against no treatment

Yes for two of three comparisons against no treatment

Comparison between the three active treatment arms:

No significant reduction in number of leaks for 3 comparisons between active treatment arms.

Significant (p<0.05) QALY gains for 2 of 3 comparisons between active treatment arms.

Yes for 2 of 3 comparisons between active treatment arms.

No for 2 of 3 comparisons between active treatment arms.

  1. MSRM=modified standardised response mean.