The HCAHPS survey is designed to be completed by adult patients who had a recent medical, surgical, or obstetrical visit to an acute care hospital.  There are certain exclusions for patients who should not receive the survey (for example, patients with certain illnesses or diseases or patients who are not discharged alive).

Sample Frame

• All payor types are included in HCAHPS.  Even though this is a CMS initiative, it is not focused on Medicare beneficiaries.
• Patients under age 18 are not included in HCAHPS. 
• Specialty hospitals are not included in HCAHPS.  Only general acute care hospitals are covered by the protocol. 
• Some patient exclusions are required.  These include expired patients, most psychiatric cases, and patients who are discharged to hospice, etc.  Exclusions are based on DRG and admit or discharge status.

Sampling Method

• Sampling will be a simple random sample of all eligible patients (after the exclusions noted above are completed). 
• Sampling will occur at least monthly, but may be more often.  The weekly (or more often) sampling typically conducted by NRC Picker can be accommodated in the HCAHPS protocol.

Sample Size

• A relatively small sample will be drawn each month.  Completed surveys from this sample will accumulate over 12 months for the public report. 
• The number of completed surveys required each year is 300.  Therefore, participating hospitals should have at least 25 completed surveys per month. 
• The number of patients sampled (surveyed) each month will depend on the mode chosen (mail/phone/IVR) and patient characteristics. 
• Small hospitals should conduct a census sample, where all eligible patients receive a survey.  As long as 100 completed surveys are received over the 12 month period, the hospital will be included in the public report.

SURVEY ADMINISTRATION

Mode

The HCAHPS protocol supports several modes of data collection.  There are advantages and disadvantages for each mode.  NRC Picker will help you determine which mode is right for you.  Just ask.

• Mail Only:  An initial survey mailing to the sample followed by a second survey to non-respondents about three weeks later. 
• Mixed Mode:  Using this approach, an initial survey mailing is distributed and then a phone call is made to non-respondents of the mail survey. 
• Phone Only:  This protocol uses exclusively telephone to collect the results.  Five survey attempts must be made to each selected patient. 
• IVR:  IVR stands for Interactive Voice Response.  Using this mode, patients will be contacted by telephone and then transitioned to an electronic survey they can completed with their touch-tone telephone.

Mode Bias Issues

It is widely known in the research community that administering an identical survey to the same types of patients using different modes (mail, phone, IVR) will result in differences in how individuals respond to the survey due to the method of data collection only.  Because of this and with the purposes of hospital comparisons, the results must be adjusted for mode differences prior to public reporting.  CMS/AHRQ has completed a mode bias study. You can read a summary of that study here.

Response Rates

Response rates will vary depending on which mode is used and patient characteristics of the hospital.  During the mode study information was collected about actual response rates by mode. NRC Picker can help you determine the appropriate mode and estimated response rate.  Just ask.