__STYLES__

Modernizing the HCAHPS Survey

Tools used in this project
Modernizing the HCAHPS Survey

About this project

Introduction

The Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) is a patient experience survey developed from the partnering of the Centers for Medicare and Medicaid Services (CMS), the Agency for Healthcare Research and Quality (AHRQ), and the CAHPS Consortium. It is managed by CMS with three goals:

• To provide patients with useful information to make better informed decisions on hospital choice;

• To give hospitals incentives to improve the quality of their care through public reporting; and

• To enhance public accountability by increasing transparency of the quality of hospital care.

The HCAHPS survey contains 32 questions: 25 questions focus on patient experience with the care received during their hospitalization, and seven questions in the “About You” section, which elicit patient-level information for use in patient-mix adjustment of HCAHPS survey scores.

The aim of this project is to evaluate whether the HCAHPS survey has been successful in accomplishing this goal.

Analysis Questions

  • Have hospitals' HCAHPS scores improved over the past 9 years?
  • Are there any specific areas where hospitals have made more progress than others?
  • Are there any major areas of opportunity remaining?
  • What recommendations can you make to hospitals to help them further improve the patient experience?

Data Collection

The dataset was downloaded from the Maven Analytics data playground, containing seven tables with primary keys(PK) and foreign keys (FK) - (questions, measures, responses, reports, states, national_results, state_results) and a data dictionary. To be clear, Primary key is a column in a table that uniquely identifies the rows in that table, while foreign keys are columns in a table that refers to the primary keys in another table, therefore linking two or more tables in a relational database.

Data Cleaning and Transformation

The datasets were loaded to power query in Power BI for cleaning. Data cleaning is a crucial step in the data analysis process because it guarantees the validity and accuracy of the conclusions drawn from the data. I considered the following throughout this step: data types, duplicates, null values, incorrect spellings. Also went further to do some data modelling (created relationships between tables using the PK and FK, merged new queries, created measures(Average response rate, average positive, negative and neutral ratings, Net promoter Score (NPS), Rank NPS using DAX. undefined

Exploratory Data Analysis and Visualization

In this phase, I provided answers to the queries that had been posed during the initial collection of data requirements

undefinedInsights

On the dashboard, the insights and suggestions are easily accessible. The following provides an overview of the conclusions:

Response Rates Are Falling: There has been a discernible fall in the response rate since 2021, pointing to changes in patients' attitudes in recent years. Line chart shows response rates between 2014 (31%) and 2022 (23%), with a percentage change of -19% overall and an average 0.1 percentage point drop per year. The NPS score has dropped by 3 points within the last 9 years. Clearly, the NPS score has been impacted by the reduction in response rate. This indicates that a fall in response rates may potentially undermine the reliability of survey results.

Care transitions and communication about medicine measures are not achieving their intended purpose:It is evident from this that care transition, communication about medicines and quietness of hospital environment have all declined. The difficulty in sustaining improvements in the first two aforementioned categories may be attributed to the perceived breadth of interpretation in these questions and patient recall bias.

Research Is Needed on Additional Factors That Influence Patient Experience: More research needs to be done to identify social determinants of health that are outside of the hospital’s influence that impact the HCAHPS survey scores to ensure a level playing field when comparing hospitals.

Recommendations

Communication: The most significant aspects affecting HCAHPS scores are communication and direct patient engagement because five of the reported metrics have a direct relationship with communication. In addition to making sure that patients receive enough medication information, thorough home recovery instructions, and confirmation that they understand their care before being discharged, hospitals should make sure that nurses and doctors communicate with patients regularly and effectively.

Factors beyond hospitals control: Certain factors outside of the hospitals control should be considered in order to ensure a level playing fields when comparing hospitals. These factors include age and construction of hospital facilities, serving communities with higher concentrations of high-need patients, lower access to care, or higher concentrations of racial/ethnic minorities.

Disparities in the capture of patient experience: Address any new gaps that are critical to assessing patient experience, reassess patient's shifting attitudes over evolving domains that matter, and look into patient-perceived ambiguous or perplexing questions.

Discussion and feedback(0 comments)
2000 characters remaining