Statistics for Health Care Professionals


Statisticsfor Health Care Professionals

Statisticsfor Health Care Professionals

Thechart is comprehensive, however it should include principalinvestigators. Therefore, the measurement in terms of deliverables isascertained. Ventricular assist devices (VAD) are high-priced andtechnically demanding, therefore, when the principal investigatorscolumn is included then the outcome measurement will be complete(Bersten et al., 2013, p. 302). In the columns, there is weaned andtransplanted consequently when the principal investigators are listedthen the metric of how the doctors deliver to reach the desiredoutcome is clear and known. The basis of VAD scorecard is to checkthe quality from the perspective of clinical outcomes. Consequently,it is important to include the principal investigators although it isin a broad sense.

Thebenchmark to be used in meeting the goals ought to be to them who aresuccessfully weaned also transplanted. This ought to be refined totake account of the time the patients stayed in the hospital infuture. The benchmark ought to be competitive because internal havebeen used over time so to gauge the deliverable against competitorswho do the same is needful. It is vital for the hospital to measureitself against the Health Care Effectiveness Data and InformationSets (HEDIS) therefore using a competitive benchmark is a step inthis direction.

Surgerytimes are varied although allocated time may be near same although itis clear that no surgery can take place in zero minutes. In fiftyoperations, the time is so varied, however what is clear is fiftypercent of the surgeries took less than hundred and thirty minuteswhile the other fifty percent took more than that. This shows datathat is upper ending that is called skewed to the right (Provost etal., 2011, p. 55). This is a more likely a problem because in dataanalysis it violates the assumptions underlying parametric tests(Munro, 2005, p. 64). What ought to be done is to transform the datato a near normal distribution center data set? It is true the visualof where the average is versus where the number is telling because itshows skewness that needs to be handled well to gauge the rightoutcome.

Inevery data collected increasing, the respondents always provide aclearer picture of the issue. The customer satisfaction report willhelp the twenty per cent, however it will not be appropriate to usewhen you are to pinpoint actually how to improve the eighty percentwho are doing well.

Inthe issuance of drugs, avoiding type 1 error over type 2 error ismore prudent because type 1 error effect is more visible. A politicalpolicy that adheres to avoiding type 1 error, in fact, impedes theprogress of marketing new drugs (Heimann, 2010, p. 64). The visibleof type 1 errors makes more open to attacks from interest groups,which politicians like to be therefore many regulations are formed inthe name of managing the industry although they have an adverseeffect on the industry. The matter here is that in type 2-errorvisibility is not there and eventually, when it occurs there are manyvariables that attributing the error to a person is not likely.

Itis a serious repercussion in the situation of blood or organ donationto make any type of error because the consequences probably will notbe reversible so gauging type 1 over type 2 is not applicable in ablanket manner to every situation.

Thethree tests that are used commonly in inferential statistics testingare known as a general linear model, which includes Analysis ofvariance (ANOVA), t-test, and Chi-square test. The difference betweenthe three tests is in t-test is utilized to determine if the averagescores between two groups are statistically different. The ANOVA testis used to test differences among many groups while Chi-square testis used to test the significance of frequencies and nominal data(Boslaugh, 2012, p. 65). A t-test is only used in two groups whereasthe other two go beyond that.

Thetwo tests in that scenario are excellent but perhaps it would be eyeopening to apply ANOVA too to compare if you get the same result asthe t-test because t-test is only used on the null hypothesis. Thisall points to running every test to be confident in the result towork on because all tests, in the long run, prove a vital aspect thatis crucial to progress.


Bersten,A.D &amp Soni N. (2013). Oh`sIntensive Care Manual.UK. Elsevier Health


Boslaugh,S. (2012). Statisticsin a Nutshell.USA. &quotO`Reilly Media, Inc.

Heimann,C.FL. (2010). AcceptableRisks: Politics, Policy, and Risky Technologies. USA.

Universityof Michigan Press.

Munro,B.H. (2005). StatisticalMethods for Health Care Research, Volume 1. Philadelphia.

LippincottWilliams &amp Wilkins.

Provost,L.P. &amp Murray S. (2011). TheHealth Care Data Guide: Learning from Data for

Improvement.San Francisco. John Wiley &amp Sons.