From: Masnick M et al. Improving Understanding of Publicly Reported Healthcare-Associated Infection (HAI) Data.
This is the final version of the survey instrument used in the manuscript cited above.
Patients were randomly assigned to one of two arms: (1) the new presentation method; (2) the existing "detailed" presentation method from CMS Hospital Compare.
The same underlying data is used for both arms; only the presentation method differs. Questions are presented in the same order for both arms.
Each question asked participants to select one possible answer from a list of options. The options were the same for all questions across both arms.
The survey questions for both arms follow below.
Questions are divided into 4 scenarios of 3 questions each:
Questions are in order from 1 to 12 below, so questions in the same scenario are grouped together. In the actual survey administered to patients, they were shuffled into the following order for all respondents:
Question 8 Question 11 Question 2 Question 4 Question 5 Question 10 Question 9 Question 3 Question 12 Question 7 Question 6 Question 1
The new presentation method uses "dot ratings" to describe the performance of a hospital:
|Infection dot rating||Description|
|●○○○||95% upper CI < 1; SIR point estimate < 0.42 (this is 1 SD for CAUTIs in the CMS Hospital Compare data)|
|●●○○○||95% upper CI < 1; SIR point estimate between 0.42 and 1|
|●●●○○||95% upper CI > 1; SIR point estimate < 1|
|●●●●○||95% upper CI > 1; SIR point estimate between 1 and 2.4 (1 SD)|
|●●●●●||95% upper CI > 1; SIR point estimate > 2.4 (1 SD)|
|Confidence dot rating||Description|
|High||High confidence: 1/2 of IQR or smaller; IQR = 6.63 for SIR 95% CI width for CAUTI|
|Moderate||Moderate confidence: Between 1/2 IQR and 1 IQR|
|Low||Low confidence: > 1 IQR|