International Society of Hypertension

New Investigators Network

Our Fellows Work

Back

Hypertension specific patient-reported outcome measure. Part II: validation survey and item selection process

 

Abstract in English:

Hypertension specific patient-reported outcome measure. Part II: validation survey and item selection process. Russian Journal of Cardiology. 2019;(7):40-46

https://doi.org/10.15829/1560-4071-2019-7-40-46

 

English full-text:

https://russjcardiol.elpub.ru/jour/article/view/3682/2720

Collection of selected articles 2019. Russian Journal of Cardiology. 2019;(12s3):46-68. https://doi.org/10.15829/1560-4071-2019-s3


  1. 1.     Summarize your work in one sentence.

 

The present work is the second and the key step in the development of hypertension-specific patient-reported outcome measure (HYpertension imPact quEstionnaire, HYPE), which is the most labor-intensive, but at the same time, however, extremely exciting.

 

  1. 2.     Summarize your findings in one sentence.

 

We carried out a large-scale survey with HYPE test version, distributed to more than 400 respondents (50 healthy volunteers and 360 patients with uncontrolled essential hypertension), whose answers have formed the basis for the interim HYPE version, cut in half (163 to 80 items), after several stages of psychometric, statistical and expert analyses implemented.

 

  1. 3.     Which were the more important methods you used in this work? If it is not a traditional method you can briefly explain the concept of that methodology.

 

 

This «item-selection» step was carried out based on the principles and criteria of the classical test theory (CTT) and the item response theory (IRT). These are two currently popular statistical frameworks for patient-reported outcomes development, test-score equating, and the identification of biased test items. In the case with CTT, six different criteria were applied to each question in the HYPE test version. In addition to purely mathematical ones, the Cronbach’s α coefficient (a measure of reliability) was adopted.

In the IRT analysis, item parameters of discrimination (coefficient a) and difficulty (classically, four degrees) were estimated with maximum likelihood estimation. Moreover, item characteristic curves (ICC) were used to graphically show item’s informativeness.

What is an ICC? In each item, five curves represent five options. Ideally, while the first (corresponds to the possible answer according to Likert scale) curve is steadily decreasing, then the last curve is steadily increasing and the middle three curves follow the normal distribution. The closer the ICC of an item is to the ideal case, the more information it contains.

 

  1. 4.     What did you learn from this paper, what was your take-home message?

 

This step in development of the hypertension-specific questionnaire (HYPE) served as a key one. The results obtained led to its twofold reduction due to the exclusion of inappropriate items and has paved the way for the next step which contained validity, reliability and sensitivity analyses. Furthermore, the primary structure and conceptual framework have been remained mostly unchanged. The take-home message from the article reminds an old saying «He that would eat the fruit must climb the tree».

 

Back

Latest Tweets:

The ISH is positively committed to opposing discrimination against people on the grounds of gender, race, colour, nationality, religion, marital status, sexual orientation, class, age, disability, having dependants, HIV status or perceived lifestyle.