HA535 Unit 5 Patient Experience Benchmarking Example

Sixty-Three Percent at the Flagship, Seventy-Seven at the Rural Hospital, and How Much Either Number Can Carry: A Benchmarking Analysis of Patient Experience Scores Across a Composite System's Three Hospitals Against a Sixteen-Hospital Peer Group

[Student Name]

Purdue University Global

HA535: Data Analytics for Health Care Managers

Unit 5 Benchmarking Analysis

[Instructor Name]

[Date]

The system, the peer group and every score are composite and bracketed; survey details are simplified for a course analysis.

The Comparison Table

Four top-box patient experience measures are set against the quartiles of a sixteen-hospital peer group of similar size and region:

1. Nurse communication: flagship [74], near the [28th] percentile; suburban [81], above the median; rural [86], highest.

2. Doctor communication: flagship [77]; suburban [82]; rural [85].

3. Overall rating of 9 or 10: flagship [66], tied with the lowest peer, near the [9th] percentile; suburban [74]; rural [80].

4. Definitely recommend: flagship [63], [5.5] points under the peer median of [68.5]; suburban [71]; rural [77].

How Precise Is Each Score?

The flagship returned about [1,140] completed surveys in the reporting year, so its recommend score has a 95 percent margin of about [2.8] points. The rural hospital returned [87], giving a margin near [8.8] points. A dot plot with these intervals shows the flagship's shortfall is clear, while the rural hospital's lead over the suburban hospital is within its margin. The rural hospital's best-in-system score is real enough to celebrate but too uncertain to rank, while the flagship's shortfall is small in points and solid in evidence.

Patient Mix and Survey Mode

Patient experience scores are affected by who answers and how they are surveyed. Research on the hospital patient experience survey found that survey mode, patient characteristics such as age and health, and nonresponse all influence scores, which is why publicly reported scores are adjusted for mode and patient mix (Elliott et al., 2009). The comparison here uses adjusted scores; raw scores would exaggerate the rural hospital's advantage, since its patients are older and report better health on average.

A Funnel Plot

Plotting each peer hospital's recommend score against its number of surveys, with control limits that widen for smaller samples, shows which hospitals truly differ from the group average. Funnel plots are designed for this purpose because they account for the larger chance variation in small institutions (Spiegelhalter, 2005). The flagship falls below the lower limit; the rural hospital falls inside the funnel despite its high score.

Choosing the Peer Group

The sixteen peers were chosen by bed size and region from publicly reported data. A different group would give different percentiles: against all hospitals nationally, the flagship would rank higher, because large urban hospitals tend to score lower. Stating how peers were chosen lets readers judge whether the comparison is fair.

The Cost of Ranking Wrongly

If leaders reward the rural hospital and criticize the suburban one on the basis of a difference within the margin, they may copy the wrong practices and discourage a hospital that is performing well. If they treat the flagship's gap as noise, they miss a real problem affecting more than a thousand respondents a year.

Turning the Gap Into a Target

Closing half of the flagship's gap on recommend, about [three] points, within a year is a realistic target. With more than a thousand surveys a year, a change of that size would be detectable on a quarterly run chart. The target should be tied to the nurse communication work the flagship chooses.

What the Analysis Supports

1. The flagship has a real gap on overall rating and recommend, worth action.

2. The suburban hospital performs above peers and could share its practices.

3. The rural hospital's high scores cannot be ranked against peers with confidence; pooling two or three years would narrow the margin.

What the Flagship Should Examine

Nurse communication, near the [28th] percentile, is the likeliest driver of the overall rating. Unit-level results, where available, can show whether the gap is general or concentrated.

Reporting Recommendations

Show every score with its margin, use funnel plots for peer comparison and avoid ranking hospitals whose intervals overlap. Comparisons over time should be shown on a run or control chart rather than as two numbers a year apart (Provost & Murray, 2011). Pool small hospitals' data over longer periods.

What the Rural Hospital Can Do

Pooling three years of surveys would give about [260] responses and a margin near [five] points, enough to compare with peers more fairly. In the meantime, the hospital can track its own scores over time and collect comments, which are informative even when numbers are few.

Limits

The peer group is chosen by size and region, not by service mix. Top-box scores lose information compared with full distributions, and a year of surveys may reflect a single unusual season.

Conclusion

Benchmarking is only as strong as the precision behind each number. The flagship's gap is real; the rural hospital's lead is promising but uncertain. Showing the margins changes what leaders should do with each comparison and with each hospital.

References

Elliott, M. N., Zaslavsky, A. M., Goldstein, E., Lehrman, W., Hambarsoomians, K., Beckett, M. K., & Giordano, L. (2009). Effects of survey mode, patient mix, and nonresponse on CAHPS hospital survey scores. Health Services Research, 44(2p1), 501-518. https://doi.org/10.1111/j.1475-6773.2008.00914.x

Provost, L. P., & Murray, S. K. (2011). The health care data guide: Learning from data for improvement. Jossey-Bass.

Spiegelhalter, D. J. (2005). Funnel plots for comparing institutional performance. Statistics in Medicine, 24(8), 1185-1202. https://doi.org/10.1002/sim.1970