physasst Posted December 22, 2012 Share Posted December 22, 2012 Interesting, could be some business interest in this.... http://www.ncbi.nlm.nih.gov/pubmed/23244257 Abstract OBJECTIVE: A physician assistant home care (PAHC) program providing house calls was initiated to decrease hospital readmission rates. We evaluated the 30-day readmission rates and diagnoses before and during PAHC to identify determinants of readmission and interventions to reduce readmissions. METHODS: Patients who underwent cardiac surgery were evaluated postoperatively for 13 months as pre-PAHC (control group) and 13 months with PAHC. Physician assistants made house calls on days 2 and 5 following hospital discharge for the PAHC group. Both groups were seen in the office postoperatively. We retrospectively reviewed the charts of 26 months of readmissions. Readmission rates for the control and PAHC groups were compared, as were the reasons for readmissions. Readmission diagnoses were categorized as infectious, cardiac, gastrointestinal, vascular, pulmonary, neurologic, and other. Also noted were the interventions made during the home visits. RESULTS: There were 361 patients (51%) in the control group and 340 patients (49%) in the PAHC group. Overall readmission rate for the control group was 16% (59 patients) and 12% (42 patients) for the PAHC group, a 25% reduction in the rate of readmissions (P = .161). The rate of infection-related readmissions was reduced from 44% (26 patients) to 19% (8 patients) (P = .010). Home interventions included adjustment of medications (90%), ordering of imaging studies (7%), and administering direct wound care (2%). CONCLUSIONS: The 30-day readmission rate was reduced by 25% in patients receiving PAHC visits. The most common home intervention was medication adjustment, most commonly to diuretic agents, medications for hypoglycemia, and antibiotics. Link to comment Share on other sites More sharing options...
MedicinePower Posted December 22, 2012 Share Posted December 22, 2012 Very interesting. I'd be interested to know how many patients the PAs were seeing a day, what they kept in their "black doctor's bag", if they drew labs or urine, etc. Link to comment Share on other sites More sharing options...
marktheshark89 Posted December 23, 2012 Share Posted December 23, 2012 Interesting study for sure. Its too bad that the overall results were not statistically significant. Link to comment Share on other sites More sharing options...
physasst Posted December 23, 2012 Author Share Posted December 23, 2012 Interesting study for sure. Its too bad that the overall results were not statistically significant. It's just because of the relatively low number of patients for readmission. The p value is reduced because as sample size decreases, the chance increases that we would obtain spurious correlations given that the null is true. 59 and 42 re-admissions aren't enough to eliminate spurious correlations. I wouldn't say it's bad, just that a larger study is needed to see if the results are reproducible....Still a good study for PAs...Better than one saying that there was an increase in admissions or something like that. Link to comment Share on other sites More sharing options...
MedicinePower Posted December 23, 2012 Share Posted December 23, 2012 P = .010 I am curious to know what the Ho (null hypothesis) was and what their H1 (alternative hypothesis) was, in addition to the alpha was. I assume they were using a µ for Ho that would be less than their current complication rate and the H1 would be ≥, but a P=.010 means you reject the null hypothesis which demonstrates a lack of statistical significance. I tutor statistics at the local college if you are wondering, in addition to taking the class and getting an A. :) Link to comment Share on other sites More sharing options...
physasst Posted December 23, 2012 Author Share Posted December 23, 2012 I am curious to know what the Ho (null hypothesis) was and what their H1 (alternative hypothesis) was, in addition to the alpha was. I assume they were using a µ for Ho that would be less than their current complication rate and the H1 would be ≥, but a P=.010 means you reject the null hypothesis which demonstrates a lack of statistical significance. I tutor statistics at the local college if you are wondering, in addition to taking the class and getting an A. :) I don't know what their hypothesis was, and it did reject the null, but likely due to the small sample size. With larger numbers I think you would see a stronger correlation and significant P. I think you are correct on their Ho and H1, but I haven't read the whole paper yet. Just thought it was an interesting paper on PA practice in CV surgery. I'm not sure what they set their alpha level at. Link to comment Share on other sites More sharing options...
MedicinePower Posted December 23, 2012 Share Posted December 23, 2012 I don't know what their hypothesis was, and it did reject the null, but likely due to the small sample size. With larger numbers I think you would see a stronger correlation and significant P. I think you are correct on their Ho and H1, but I haven't read the whole paper yet. Just thought it was an interesting paper on PA practice in CV surgery. I'm not sure what they set their alpha level at. Agreed, their n needs to be bigger. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.