Validation in an economic evaluation
Source:    Publish Time: 2012-11-04 06:59   2900 Views   Size:  16px  14px  12px
To ensure the validity of findings, the health economic model should be validated by a couple of approaches. Recently, I received a email regarding my publication, CUA of Intensive Blood Glucose Control with Metformin (Xie & Vondeling, 2008 ). Their questions and my answers are listed below.

Author : Xuanqian Xie

To ensure the validity of findings, the health economic model should be validated by a couple of approaches. Recently, I received a email regarding my publication, CUA of Intensive Blood Glucose Control with Metformin (Xie & Vondeling, 2008 ). Their questions and my answers are listed below.


Q1. Was the model face validated[1], but not reported in the article?

A1: Although there was no separate section of validation in our paper, we tried our best to present the reliable estimates of cost-effectiveness in multiplex ways. Corresponding author, Dr. Vondeling, has considerable research experiences in diabetes. Also, we consulted a number of professionals of diabetes clinical/research on this issue. We reached the consensus that basically we can generalize the UKPDS34 to Chinese, in case of the absence of good evidence for Chinese population in this field.

We did not report these in the publication.

Q 2. If it was face validated, did a third party (i.e., someone from an organization outside of the model building process) perform the face validation?

A2: No

 

Q3. If it was face validated but NOT by a third party, what was the reason why a third party was not used (choose one or more):

(a) third party not necessary,

(b) time constraint,

(c) budget constraints,

(d) lack of third party availability/access

(e) other reason.

A3: (c) & (d).

 

Q4. If it was not face validated, what was the reason why it was not face validated:

(a) face validation not necessary,

(b) time constraint,

(c) budget constraints,

(d) other reason.

A4: NA.

 

Q5. Was the model internally validated[2] but not reported in the article?

A5: We compared the simulated risks of composite events (both fatal and non-fatal) with the original RCT (UKPDS 34), and found they were close. Therefore, we judged that our model reflected the results of UKPDS 34.
We did not report these in the publication.


Q6. If it was internally validated, did a third party (i.e., someone from an organization outside of the model building process) perform the internal validation?

A6: No.

 

Q7. If it was internally validated but NOT by a third party, what was the reason why a third party was not used (choose one or more):

(a) third party not necessary,

(b) time constraint,

(c) budget constraints,

(d) lack of third party availability/access

(e) other reason.

A7: (c) & (d).

 

Q8. If it was not internally validated, what was the reason why it was not internally validated:

(a) internal validation not necessary,

(b) time constraint,

(c) budget constraints,

(d) other reason.

A8: NA.

 

 [1] Face validity is the extent to which a model, its assumptions, and its applications correspond to current science and evidence, as judged by people who have expertise in the problem areas to which the model is being applied.

[2] Internal validity addresses whether the parts of a model behave as intended and the model has been implemented (e.g. coded) correctly.

 

Reference:

Xie X, Vondeling H. Cost-utility analysis of intensive blood glucose control with metformin versus usual care in overweight type 2 diabetes mellitus patients in Beijing, P.R. China. Value Health 2008; 11 Suppl 1:S23-S32 Link: http://www.ncbi.nlm.nih.gov/pubmed/18387063 (This paper was based on my Master thesis.)