Advancing patient-centered hospital care through latent service quality modeling: a longitudinal analysis of HCAHPS data

Monitoring hospital patient experience requires interpretable summaries of ordered ratings. We analyzed ten CMS HCAHPS domain star ratings for 384 hospitals across 2017–2019 and 2021–2022 (1,920 observations), combining a graded response model (GRM) with calendar-time growth modeling and 1,000 hospital bootstrap refits of both stages. Mean fitted latent scores for the sample declined from 2019 to 2021 ( \(-0.106\) ; 95% interval [ \(-0.159\) , \(-0.057\) ]), rebounded from 2021 to 2022 (0.195 [0.152, 0.238]), and exceeded the 2019 mean (0.089 [0.038, 0.138]). Alternative longitudinal models and observed-score analyses showed the same pattern. Overall rating, care transition, and nurse communication had the highest discrimination estimates; their point-estimate ordering is directional. Rank agreement with aggregate scores was high but depended on the comparator. Teaching- and ownership-related contrasts suggested differences in change, but these were inconclusive after adjustment for multiple tests in the primary model. An exploratory model assuming shared earlier trends showed a more negative 2022 deviation from projected trends for for-profit than government hospitals. Overall rating and recommendation remained associated beyond the common score, showing that it does not fully capture relationships among domains. The approach supports patient-experience monitoring by making measurement assumptions and uncertainty explicit; the temporal findings describe associations across study years.

Authors

Institutions

Publication Details

Journal
Scientific Reports
Published
2026-11-14
DOI
https://doi.org/10.1038/s41598-026-75488-y
Primary Topic
Patient Satisfaction in Healthcare
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Advancing patient-centered hospital care through latent service quality modeling: a longitudinal analysis of HCAHPS data

Daniel Zantedeschi, Sajeev Varki, Ling Hu
Scientific Reports
Patient Satisfaction in Healthcare
article

Advancing patient-centered hospital care through latent service quality modeling: a longitudinal analysis of HCAHPS data

Daniel Zantedeschi, Sajeev Varki, Ling Hu
article en

Abstract

Monitoring hospital patient experience requires interpretable summaries of ordered ratings. We analyzed ten CMS HCAHPS domain star ratings for 384 hospitals across 2017–2019 and 2021–2022 (1,920 observations), combining a graded response model (GRM) with calendar-time growth modeling and 1,000 hospital bootstrap refits of both stages. Mean fitted latent scores for the sample declined from 2019 to 2021 ( \(-0.106\) ; 95% interval [ \(-0.159\) , \(-0.057\) ]), rebounded from 2021 to 2022 (0.195 [0.152, 0.238]), and exceeded the 2019 mean (0.089 [0.038, 0.138]). Alternative longitudinal models and observed-score analyses showed the same pattern. Overall rating, care transition, and nurse communication had the highest discrimination estimates; their point-estimate ordering is directional. Rank agreement with aggregate scores was high but depended on the comparator. Teaching- and ownership-related contrasts suggested differences in change, but these were inconclusive after adjustment for multiple tests in the primary model. An exploratory model assuming shared earlier trends showed a more negative 2022 deviation from projected trends for for-profit than government hospitals. Overall rating and recommendation remained associated beyond the common score, showing that it does not fully capture relationships among domains. The approach supports patient-experience monitoring by making measurement assumptions and uncertainty explicit; the temporal findings describe associations across study years.

Scientific Reports
University of South Florida (US), Central Washington University (US)
Patient Satisfaction in Healthcare
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon