Examples in the Literature

CAHPS/HCAHPS

(H)CAHPS stands for "Consumer Assessment of Health Plan Series" and is a set of survey tools developed developed jointly by the Agency for Healthcare Research and Quality (AHRQ) and NCQA to assess patient satisfaction with their health plan. It is was designed for use by the Centers for Medicare & Medicaid Services (CMS) as a part of a public reporting and payment incentive initiative. The intent of the CAHPS initiative is to provide a standardized survey instrument and data collection methodology. The official website for content, announcements, update, reminders is www.hcahpsonline.org.

Participation by hospitals in the USA is linked to annual Payment Update (APU). Public reporting of results started March 2008, at the hospital compare website. This website is intended to "provide objective information to help consumers make informed decisions about health care providers".

Methodology: questionnaires are sent to patients by mail after hospital discharge, using a standard random sampling procedure. Example response rates (for state of Maryland 2005) were in the range of 20~30% for adult patients, and 6.3% for physician surveys.

The results are presented by Summary Rates or the percent of respondents who choose the most positive question responses.

• Attributes are the individual questions within each composite.
• Summary Rate Scores represent the percentage of respondents who select one of the top two positive answer choices. For most questions, the Summary Rate Score is the sum of the percentage of respondents who answered "Excellent" or "Very Good" from a five-point scale ranging from "Excellent" to "Poor".
• Composites are calculated by taking the average of the Summary Rate Scores of the attributes in the specified section.
• Overall satisfaction was also rated on a five-point scale with response options consisting of "Very satisfied", “"omewhat satisfied", "Neither satisfied nor dissatisfied", "Somewhat dissatisfied", and "Very dissatisfied". The Summary Rate Score is the sum of the proportion of respondents who selected "Very satisfied" and "Somewhat satisfied".
• Significance testing determined if differences between percentages was attributable to real variations and not to chance.
• Calculation of "mean score": Responses to survey questions are converted to a series of 100-point maximum scales. "very poor" is awarded zero points; "poor", 25 points; "fair", 50 points; "good", 75 points; and "very good" is awarded 100 points. Next, individual item scores within a survey section are averaged to become scores for each section. Finally, section scores are averaged to become the overall satisfaction score. The average of all respondents' overall satisfaction scores is called the hospital's Overall Mean Score.

Definition of a Completed Survey: a survey is considered "complete" even if a patient did not answer all items when at least 50% of the questions applicable to all patients are answered. (Items governed by skip patterns and the basic demographic data are not included in the calculation of percentage complete).

Total Survey Response Rate: For a given four rolling quarter (12-month) public reporting period (note that "refusals" are not removed from the denominator of the response rate calculation):

Press-Ganey