Measuring Clinical Effectiveness: SAMPLE RESULTS part 2
A Comparison of Actual and Expected Treatment Rates
Because the CEI involved multiple hospitals, the data were pooled and used for comparison with treatment patterns at individual hospitals. These comparisons were adjusted to the case mix for each hospital to allow the individual hospital to be compared with the pooled data. To accomplish this, the CEI project team used the pooled data to develop expected utilization rates for each CEI-DRG based on averages for all of the hospitals. These expected rates were applied to the numbers of patients by the CEI-DRG at a hospital to calculate an expected number of patients. This number was compared with the actual number to determine whether the rate of utilization at a hospital was above or below the group as a whole (the pooled hospital data).
As an example, a question that routinely arises relates to the proper prophylaxis for patients at risk for DVT. Table 8 demonstrates this analysis. canadian cialis online
According to the pooled data, 700 of a total of 1,500 discharged patients, or 47%, received no drugs to prevent DVT. In the hospital being analyzed, that percentage was much higher—170 of 300 discharged patients, or 57%.
Table 8 Sample Utilization Benchmarking
|
Pooled Data |
Hospital |
||||||
|
DRG Name |
Total No. of Discharges | Discharges with No Antithrombosis Drugs | Percentage with No Antithrombosis Drugs (Pd) | Total No. of
Discharges (ndh) |
Discharges with No Antithrombosis Drugs |
Percentage with No Antithrombosis Drugs |
Expected Discharges with No Antithrombosis
(Pd* ndh) |
| CEI-DRG A CEI-DRG B
CEI-DRG C |
500 500 500 | 100 250 350 | 20% 50% 70% |
40 100 160 |
6
49 115 |
15%
49% 72% |
8 50 112 |
| 1,500 |
700 |
47% |
300 |
170 | 57% |
170 |
|
| Index (actual/expected) |
1.00 |
||||||
The CEI project team approached the question of the proper prophylaxis by first calculating the percentage of patients in each CEI-DRG who did not receive any anti-thrombotic drugs. This percentage varied considerably, from almost 0% for patients undergoing coronary bypass procedures to nearly 100% for children who were admitted with simple pneumonia.
In Table 8, the percentages ranged from 20% to 70%. Multiplying the pooled percentage in a CEI-DRG (Pd) by the number of hospital discharges in the CEI-DRG (ndh) resulted in an expected number of patients (edh) who would not have received any antithrombotic therapy if the physicians at that hospital had treated patients the same way as the group as a whole:
e = n * P
Summing the expected numbers across all CEI-DRGs produced a total expected number for the hospital. This figure was then compared with the actual number. In Table 8, the expected number was 170 and the actual number was 170. This means that even though the overall hospital rate appeared too high (57% vs. 47%), the hospital treatment patterns were actually in line with the group as a whole, after case mix was taken into account.
eriacta tablets
The ratio of the actual number to the expected number can be used to calculate an index, which can then be compared across multiple utilization measures to obtain a broad picture of a hospital in comparison with the group as a whole. If the RAD = Radiology (diagnostic and therapeutic); REH = Rehabilitation; RSP = Respiratory Therapy;TPY = Other Therapeutic Services.
Clinical Outcomes
The CEI was designed to allow the measurement of both clinical and financial outcomes along with their corresponding utilization patterns. Examples of some outcomes that can be developed from the CEI data set include:
- readmission rates.
- in-hospital mortality.
- complication rates.
- use of drugs to reverse the effects of a study drug (i.e., indication of an overdose).
buy prescription drugs online
The same analytical approach that was used for the utilization measures can also be used for the outcome measures. For example, complication rates can be calculated by CEI-DRG for the group as a whole and for each hospital. These values can be used to calculate the expected numbers of complications compared with the actual numbers at each hospital.
