|
|
|
Detailed instructions:
General issues:
Data Tools do not work at all.
The Data Tools are designed to work in Microsoft Internet Explorer and Mozilla FireFox. If you are using another browser, it may not be compatible with the Data Tools. If you are using one of these browsers and are still unable to use the tools, please contact us by email with a description of the tool you trying to use, the problem you are having, and your browser and version number.
Desired event variable or area/hospital is not listed.
There are several reasons that an option might not be listed:
- The variable is not available for the chosen geography level. For example, surgical and medical discharge rates are not available for hospitals. Please check back often, as we add data frequently.
- The variable is not available for the chosen hospital. Provider-specific physician measures, which use the 20% Part B Medicare claims file, are available only for hospitals with at least 400 deaths. Certain hospitals had incomplete ICU data during the study period, so measures related to ICU utilization are not available for those hospitals.
- Rates are heavily suppressed for the chosen geography level. If you choose hospital service areas (HSAs), some events will not appear among your choices because very few HSAs have sufficient sample size to report rates for that event.
- Rates are heavily suppressed for the chosen area. Some of the 3,436 HSAs are not included in the Data Tools because rates are suppressed for most events.
Requested data are suppressed.
For provider-specific data, inpatient measures are reported for hospitals with more than 80 deaths; physician measures, which use the 20% Part B Medicare claims file, are available only for hospitals with at least 400 deaths. This suppression is done to preserve both patient confidentiality and statistical precision.
For population-based geographic rates, event measures based on a count of fewer than 11 patients are not displayed for reasons of patient confidentiality. Rates with fewer than 26 expected events are also suppressed because of a lack of statistical precision.
Benchmarking:
Difference between benchmark and comparison areas/hospitals.
The question being explored in the Benchmarking tool is: what if rates in the comparison areas/hospitals were equal to rates in the benchmarks? Comparison areas/hospitals are those that are being compared; Benchmarks are what they are being compared to. To see what the situation would be if your area or hospital's rates were in line with others or with national or state averages, choose your area or hospital as the comparison, and choose other areas/hospitals or averages as benchmarks. To see what the situation would be if several areas/hospitals practiced like another, or in line with an average, choose these areas/hospitals as the comparison areas and one benchmark.
Explanation of the surplus/deficit.
The last column in the Benchmarking table gives the number of events - dollars, number of days, visits, admissions, etc. - that would or would not have taken place in the comparison area(s) or hospital(s) if their rates had been equal to the benchmark(s). This value is calculated by subtracting the benchmark rate from the comparison rate, and multiplying this difference by the population of the comparison area.
For example, the 1,657 patients dying between 2001 and 2005 who were assigned to the UCLA Medical Center experienced 18.5 days in hospital during the last six months of life. By contrast, patients at UCSF Medical Center experienced 13.5 days in hospital. To calculate the surplus days during the last six months of life at UCLA compared to UCSF, the tool calculates the difference in the rates (18.5 - 13.5 = 5.0) and multiplies the difference by the number of patients assigned to UCLA (5.0 x 1,657) to conclude that there was a surplus of more than 8,000 days experienced by patients assigned to UCLA Medical Center compared to the UCSF benchmark.
Geographic Query Finder:
Desired hospital is not found.
Our hospital names come from the American Hospital Association's Annual Survey database. Recent hospital name changes may not be reflected in our data. Many hospitals are commonly known by names that are not their official names in that database. If you do not find the hospital you are looking for, choose the option "Find all hospitals in an HSA (or HRR)," enter your area, and look for your hospital on that list.
Our list of hospitals is not exhaustive. It includes only those hospitals that had at least 80 deaths among the fee-for-service Medicare population in the period from 2001 through 2005. Smaller hospitals, and hospitals with a significant proportion of managed care patients, may not appear. Hospitals that closed during the study period will not appear.
Desired ZIP code is not found.
Our ZIP code to HSA/HRR crosswalk is based on 2005 ZIP codes. If your desired ZIP code was recently created, it may not appear.
|
|
