Using publicly available hospital data, I developed a map to that ranks US hospitals with respect to how well they follow guidelines, standards of care or practice parameters. This process of care metric is based on medical information from patient records that reflects the rate or percentage across 12 procedures related to surgical care.  These percentages were translated into scores that ranged from 0 (worse) to 100 (best).  Higher scores indicate that the hospital has a higher rate of following best practices in surgical care. Details of how these metrics were calculated appear below the map.

First, for each hospital, I calculated an overall Process of Care Metric by averaging each of the 12 process of care scores (excluding when the data indicated the sample size was too small).  The process of care metric was used because it has good measurement properties (internal consistency was .75) and, thus reflects a good overall measure of process of care. Using this process of care metric, the hospitals were divided into three groups:

• Red  (0 through 95)
• Yellow (greater than 95 through 97.5)
• Green (greater than 97.5)

These three colors were used to color code the hospitals on the maps. Hospitals that did not have process of care scores were coded as blue on the map.

Clicking on one of the buttons will provide detailed information about the process of care metrics for that hospital. NOTE: Hospitals that do not have any process of care scores are coded as blue.

Map of US Hospitals and Process of Care Measures

[iframe_loader src=”https://www.google.com/fusiontables/embedviz?viz=MAP&q=select+col2%3E%3E0+from+1zyiuuIrVlYz1QiKfHEgr51dLaC_6QxYuzl0-Y3A&h=false&lat=38.651198332299536&lng=-92.577263975&z=4&t=1&l=col2%3E%3E0″]

Process of Care

Quality measures are used to gauge how well an entity provides care to its patients. Measures are based on scientific evidence and can reflect guidelines, standards of care, or practice parameters. A quality measure converts medical information from patient records into a rate or percentage that allows facilities to assess their performance.

The following 12 questions were used to map hospitals with respect to their process of care. For each of the 12 items, hospitals received a score that reflected the percent of patients who received this type of care.

1. Surgery Patients given an antibiotic at the right time (within one hour before surgery) to help prevent infection
2. Surgery Patients whose preventive antibiotics were stopped at the right time (within 24 hours after surgery)
3. Surgery Patients who were given the right kind of antibiotic to help prevent infection
4. Surgery Patients who got treatment at right time (within 24 hours before or after surgery) to help prevent blood clot
5. Surgery Patients whose doctors ordered treatments to prevent blood clots after certain types of surgeries
6. Heart Surgery Patients whose blood sugar is kept under good control in the days right after surgery
7. Surgery Patients needing hair removed from the surgical area before surgery who had hair removed using a safer method
8. Surgery Patients whose urinary catheters were removed on the first or second day after surgery
9. Surgery patients who were taking heart drugs called beta blockers before coming to the hospital, who were kept on them
10. Outpatients having surgery who got an antibiotic at the right time – within one hour before surgery (higher numbers are better)
11. Outpatients having surgery who got the right kind of antibiotic (higher numbers are better)
12. Patients having surgery who were actively warmed in the operating room or whose body temperature was near normal

In addition to the percentages for each of the 12 process of care metrics, the data set included information on each metric that described the quality of the sample on which the percentage was based, essentially indicating when data were unreliable (number of cases too small) or were not.

Calculating the Process of Care Metric

This classification metric (e.g., better, same, worse) was used to help color-code the different hospitals for mapping purpose. Better was coded as green. Same was coded as yellow. Worse was coded as red. Hospitals that did not have data for that metric were coded as blue.

The Process of Care metrics were transformed into a 0 to 100 scale by simply multiplying the original value by 1000.  Mortality Rates and Readmission Rates are scores so higher scores mean worse hospital performance.  To rescale the values to a 0 to 100 scale, where higher score s indicate better hospital performance, I simply rescaled each hospital’s mortality/readmission rates (out of 1000) into a new metric (0 to 100) using the following formula (Hospitals that had zero death rate would receive a Survival Rate Metric score of 100. Higher death rates lead to lower Survival Rate scores.).

New metric = Original metric * (1 – (Rate / 1000)) * 100

Survival Rate Metrics were calculated as:

Survival Rate Metric = Mortality Rate * (1 – (Rate / 1000)) * 100

Non-Readmission Metrics were calculated as:

Non-Readmission Metric = Readmission Rate * (1 – (Rate / 1000)) * 100

This transformation does not result in any loss of information. The rescaled values (Survival and Non-Readmission Rates) are correlated perfectly (r = 1.0) with their original counterpart (Mortality and Readmission Rates).

Ranking Hospitals by Process of Care Color Codes

I created one map for the process of care outcomes. First, for each hospital, I calculated an overall Process of Care Metric by averaging each of the 12 process of care scores (excluding when the data indicated the sample size was too small).  The process of care metric was used because it has good measurement properties (internal consistency was .75) and, thus reflects a good overall measure of process of care. Using this process of care metric, the hospitals were divided into three groups:

• Red  (0 through 95)
• Yellow (greater than 95 through 97.5)
• Green (greater than 97.5)

These three colors were used to color code the hospitals on the maps. Hospitals that did not have process of care scores were coded as blue on the map.