Battle of the States Over Healthcare

Questions presented in the brief:

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

Analysis of questions:

1. Standardization for improvement in scores while creating incentive is a good place to apply the percent rate change over each year (slope of a line by definition). The problem here is there are 3 groups of answers: Top, Middle, and Bottom. All three groups are dependent upon each other. All three groups together sum to 100%. How do you combine these 3 lines to get one slope for comparison? Can you use just one line to compare all the States?

To solve the comparison problem with slopes, the possibilities of where the percentages were moving was calculated. It was surprising that the States only fell into 4 out of the 12 possible scenarios. Of course Bottom and Middle answers feeding their percentages into the Top is ideal and rated as #1. Worse case scenario would be for both the Top and Middle percentages to go into the Bottom. Now, the hospitals in each State can be ranked and incentive created to move into the #1 group. Arrows were provided to help the users see how their State's ratings are shifting.

2 and 3 are linked together. The words, "hospitals have made more progress than others" gives the indication that the user would like to compare hospitals to each other. The word "progress" suggests the top 2 areas of improvement for each hospital. The words "areas of opportunity" suggests the bottom 2 areas. Obviously, the questions with a higher percentage of answers in the Top-Box are doing better than those with less percentage. This is why percent averages over the Top-Box of all facilities in a State was used in the table for the top 2 and bottom 2 slots. The comparison lines were added to help guide the user in deciding which areas need the most work and give guidance on the direction of the recent actions taken. The obvious shift during Bidenomics/Covid era is an example of recent events.

The dataset listed hospital Id numbers but there was no way to combine that information with the state result table. Therefore, the analysis for the category of questions had to remain on the State level.

4. There were over 3,000 hospitals involved with this survey. It is unreasonable to think all these hospitals will fall under the same recommendations. A table was developed so each State could see their own results and compare them to other States. The State colors, democrat, republican, or swing State, was added out of curiosity. This provides incentive and feedback when political opponents start arguing whose methods of governing is better. Even though the hospitals are run privately, they do receive money from the government and are governed by laws.

Here is a link to the working model: My Heath Care Report