https://data.medicare.gov/Hospital-C...ital/77hc-ibv8
enter 100290 in the search box and it will give you the infection rates for 2013 for the Villages Hospital
You have to scroll to the right on the screen to see the data. I do not know how to find the earlier years, and suspect the 2014 data is not yet available. However the newspaper report you mention says that the hospital is pleased that is has lowered hospital acquired infections. When you look at the 2013 data you will see that except for urinary catheter associated infections, even in 2013 TVRH was average to above average in avoiding acquired infections already.
The Healthgrade website in its methodology reports it used data from 2011 to 2013 in aggregate. Thus three years of data is being lumped into one score. If the data from 2011 was bad and the data from 2013 was improved, the cumulative score would still be poor.
Predicted mortality is the number of patients that would be expected to die who present with a particular illness. There is an attempt to adjust for the other factors in your patient mix. For instance if your hospital admits more elderly patients than the average hospital, you would expect a higher death rate. So, in this sepsis category, TVRH had a higher death rate than would be expected based on the predicted death rate for a similar patient mix. But of course the adjustments can be very off depending on how well the coding clerks add diagnostic codes on the billing forms.
Additionally, the methodology of Healthgrades was to take the first diagnostic code used in billing. So this was likely patients who came to the hospital already septic. It is not a representation of patients who developed infections in the hospital nor does it relate to your original concern with hospital acquired infections and the impact of prophylactic antibiotics.