Objective

To illustrate methodological considerations when assessing the relationship between patient care experiences and mortality.

Data source

Medical Expenditure Panel Survey data (2000-2005) linked to National Health Interview Survey and National Death Index mortality data through December 31, 2006.

Study design

We estimated Cox proportional hazards models with mortality as the dependent variable and patient experience measures as independent variables and assessed consistency of experiences over time.

Data extraction methods

We used data from respondents age 18 or older with at least one doctor's office or clinic visit during the year prior to the round 2 interview. We excluded subjects who died in the baseline year.

Principal findings

The association between overall care experiences and mortality was significant for deaths not amenable to medical care and all-cause mortality, but not for amenable deaths. More than half of respondents were in a different care experience quartile over a 1-year period. In the five individual experience questions we analyzed, only time spent with the patient was significantly associated with mortality.

Conclusions

Deaths not amenable to medical care and the time-varying and multifaceted nature of patient care experience are important issues to consider when assessing the relationship between care experience and mortality.

There is widespread interest in assessing care coordination to improve overall care quality. We evaluated a five-item measure of care coordination included in the 2012 Consumer Assessment of Healthcare Providers and Systems (CAHPS) Medicare survey (n = 326,194 respondents, 46% response rate). This measure includes patient reports of whether their personal doctor discusses their medicines, has medical records and other relevant information, and is informed about care from specialists, and whether the patient gets help in managing care and timely follow-up on test results. A one-factor categorical confirmatory factor analytic model indicated that five items constituted a coherent scale. Estimated health-plan-level reliability was 0.70 at about 102 responses per plan. The composite had a strong unique association with the CAHPS global rating of health care, controlling for the CAHPS core composite scores. This measure can be used to evaluate relative plan performance and characteristics associated with better care coordination.

Objective

To evaluate the dimensionality of hospital quality indicators treated as unidimensional in a prior publication.

Data source/study design

Pooled cross-sectional 2010-2011 Hospital Compare data (10/1/10 and 10/1/11 archives) and the 2012 American Hospital Association Annual Survey.

Data extraction

We used 71 indicators of structure, process, and outcomes of hospital care in a principal component analysis of Ridit scores to evaluate the dimensionality of the indicators. We conducted an exploratory factor analysis using only the indicators in the Centers for Medicare & Medicaid Services' Hospital Value-Based Purchasing.

Principal findings

There were four underlying dimensions of hospital quality: patient experience, mortality, and two clinical process dimensions.

Conclusions

Hospital quality should be measured using a variety of indicators reflecting different dimensions of quality. Treating hospital quality as unidimensional leads to erroneous conclusions about the performance of different hospitals.

Measures of patients' care experiences are increasingly used as quality measures in accountability initiatives. As the prominence and financial impact of patient experience measures have increased, so too have concerns about the relevance and fairness of including them as indicators of health care quality. Using evidence from the Consumer Assessment of Healthcare Providers and Systems (CAHPS®) surveys, the most widely used patient experience measures in the United States, we address seven common critiques of patient experience measures: (1) consumers do not have the expertise needed to evaluate care quality; (2) patient "satisfaction" is subjective and thus not valid or actionable; (3) increasing emphasis on improving patient experiences encourages health care providers and plans to fulfill patient desires, leading to care that is inappropriate, ineffective, and/or inefficient; (4) there is a trade-off between providing good patient experiences and providing high-quality clinical care; (5) patient scores cannot be fairly compared across health care providers or plans due to factors beyond providers' control; (6) response rates to patient experience surveys are low, or responses reflect only patients with extreme experiences; and (7) there are faster, cheaper, and more customized ways to survey patients than the standardized approaches mandated by federal accountability initiatives.

Distal limb injuries are common in racing horses and track surface properties have been associated with injury risk. To better understand how track surfaces may contribute to equine limb injury, we developed the first 3D computational model of the equine hoof interacting with a racetrack and simulated interactions with model representations of 1) a dirt surface and 2) an all-weather synthetic track. First, a computational track model using the Smoothed Particle Hydrodynamics (SPH) method with a Drucker-Prager (D-P) elastoplastic material model was developed. It was validated against analytical models and published data and then calibrated using results of a custom track testing device applied to the two racetrack types. Second, a sensitivity analysis was performed to determine which model parameters contribute most significantly to the mechanical response of the track under impact-type loading. Third, the SPH track model was coupled to a biomechanical model of the horse forelimb and applied to hoof-track impact for a horse galloping on each track surface. We found that 1) the SPH track model was well validated and it could be calibrated to accurately represent impact loading of racetrack surfaces at two angles of impact; 2) the amount of harrowing applied to the track had the largest effect on impact loading, followed by elastic modulus and cohesion; 3) the model is able to accurately simulate hoof-ground interaction and enables study of the relationship between track surface parameters and the loading on horses' distal forelimbs.

Background

The Agency for Healthcare Research and Quality held a research meeting on using Consumer Assessment of Healthcare Providers and Systems (CAHPS®) data for quality improvement (QI) and evaluating such efforts.

Topics covered.

Meeting addressed: 1)What has been learned about organizational factors/environment needed to improve patient experience? 2)How have organizations used data to improve patient experience? 3)What can evaluations using CAHPS data teach us about implementing successful programs to improve patient experience?

Key themes

Providers and stakeholders need to be engaged early and often, standardize QI processes, complement CAHPS data with other data, and compile dashboards of CAHPS scores to identify and track improvement. Rigorous study designs are valuable, but much can be learned and accomplished through practical organization-level studies.

Objective

To assess the effect of changing survey questions on plan-level patient experience measures and ratings.

Data source

2015 Medicare Advantage CAHPS Survey respondents.

Study design

Ninety three randomly selected beneficiaries in each of 40 MA plans received a revised (5.0) CAHPS survey; 38 832 beneficiaries received version 4.0. Linear mixed-effect regression predicted CAHPS measures from fixed effects for survey version and beneficiary characteristics and random effects for plan and plan-by-version random slope.

Principal findings

Response rates were 42 percent for both versions. Removal of "try to" from screeners increased the percentage of respondents eligible for follow-up questions. Version 5.0 caused a small increase (1-3 points on a 0-100 scale, P < 0.05) in the mean of three altered measures and a moderate increase (>3 points) in one. There was a small statistically significant increase in two unaltered measures. These changes were uniform across plans, so there would be no expected change compared to results using the legacy survey in the score distributions other than uniform mean shifts, and no expected effect on summary measures.

Conclusions

These analyses illustrate how to assess the impact of seemingly minor survey modifications for other national surveys considering changes and highlight the importance of screeners in instrument design.

Background

Data from surveys of patient care experiences are a cornerstone of public reporting and pay-for-performance initiatives. Recently, increasing concerns have been raised about survey response rates and how to promote equity by ensuring that responses represent the perspectives of all patients.

Objective

Review evidence on survey administration strategies to improve response rates and representativeness of patient surveys.

Research design

Systematic review adhering to the Preferred Reporting Items for Systematic reviews and Meta-Analyses guidelines.

Study selection

Forty peer-reviewed randomized experiments of administration protocols for patient experience surveys.

Results

Mail administration with telephone follow-up provides a median response rate benefit of 13% compared with mail-only or telephone-only. While surveys administered only by web typically result in lower response rates than those administered by mail or telephone (median difference in response rate: -21%, range: -44%, 0%), the limited evidence for a sequential web-mail-telephone mode suggests a potential response rate benefit over sequential mail-telephone (median: 4%, range: 2%, 5%). Telephone-only and sequential mixed modes including telephone may yield better representation across patient subgroups by age, insurance type, and race/ethnicity. Monetary incentives are associated with large increases in response rates (median increase: 12%, range: 7%, 20%).

Conclusions

Sequential mixed-mode administration yields higher patient survey response rates than a single mode. Including telephone in sequential mixed-mode administration improves response among those with historically lower response rates; including web in mixed-mode administration may increase response at lower cost. Other promising strategies to improve response rates include in-person survey administration during hospital discharge, incentives, minimizing survey language complexity, and prenotification before survey administration.

Using data from 335,249 Medicare beneficiaries who responded to the 2007 Medicare Consumer Assessment of Healthcare Providers and Systems (CAHPS) survey, along with data from 22 cognitive interviews, we investigated the reliability and validity of an instrument designed to assess beneficiaries' experiences with their prescription drug plans. Composite measures derived from the instrument had acceptable internal consistency and sufficient plan-level reliability to inform consumer choice, quality improvement, and payor oversight. These measures were positively associated with members' overall rating of the plan and their willingness to recommend the plan. Moreover, each was independently useful in predicting beneficiaries' global ratings of their plan. This instrument can be an important tool for helping beneficiaries to choose a plan that best meets their needs.

Patient care experience surveys evaluate the degree to which care is patient-centered. This article reviews the literature on the association between patient experiences and other measures of health care quality. Research indicates that better patient care experiences are associated with higher levels of adherence to recommended prevention and treatment processes, better clinical outcomes, better patient safety within hospitals, and less health care utilization. Patient experience measures that are collected using psychometrically sound instruments, employing recommended sample sizes and adjustment procedures, and implemented according to standard protocols are intrinsically meaningful and are appropriate complements for clinical process and outcome measures in public reporting and pay-for-performance programs.