Using and interpreting statistics - Essay Example

Order 204474 INTRODUCTION Using and Interpreting Statistics Statistics and biostatistics enabled medical practitioners to take apart authenticity from randomness and fiction as it steered organized reasoning, evaluation and decision-making, a vital component in the scientific sphere of medicine (Jenicek 2002, 26). Coupled to this are the advances in informatics' which made possible the treatment of huge quantity of data by highly developed and multifaceted analytical methods, refined samplings, or superior medical data linkage (Jenicek 2002, 26). Materials and Methods Informational data from a study associated with childhood diabetes, informational data from a heart and lung surgery, and informational data from transfusion orthopaedic surgery were collected, tabulated and statistically treated with 95% confidence interval, risk ratio, probability, chi square, mean, standard deviation, standard error, regression Results Question 1 No category have a value of 4 or 13.33 percent gene A present while Yes category have a value of 14 or 46.67 percent. This means that there are minimal numbers of diabetics with gene A. The Sensitivity study of gene A is equal to 0.22222, which means that respondents with gene A and true gene A rate is quite low. Specificity is equal to 0.38095. This is almost one third of the probability of without gene A in health and true negative rate. The findings reveal that there is a 30% or 0.30 probability of gene A. This is also called the risk of gene A or the estimated probability. The probability of the exposed cases is equal to 0.4667 while the probability of the exposed in control is equal to 0.1333. The overall or total exposed has a value of 0.3000. With 95% confidence, the interval 1.410185 to 27.15574 contains the unknown mean . a) Based on the given informational data, the study design used is randomized controlled trial design. This is because it involved the random allocation of different interventions (or treatments) to subjects . b) The risk of childhood diabetes for individuals with gene A compared to those without gene A can be calculated using proportion statistics treatment of the data on Table 2. c) Yes, the data suggest evidence to support Ho: gene "A" is associated with childhood diabetes because 14 or 46.67 percent though indicating minimal numbers of diabetics with gene A, the Sensitivity study of gene A is equal to 0.22222. Thus, despite the low number of respondents with gene A its Specificity is equal to 0.38095. This is almost one third of the probability of without gene A in health and true negative rate. The findings reveal that there is a 30% or 0.30 probability of gene A in childhood. This is also called the risk of gene A or the estimated probability. d) This study aimed to find out if gene "A" is associated with childhood diabetes. The findings revealed diabetics under the No category have a value of 4 which is 13.33 percent gene A present while the Yes category have a value of 14 which is 46.67 percent. This means that there are minimal numbers of diabetics with gene A. The Sensitivity of the study of gene A is equal to 0.22222, which means that the patient with gene A and true gene A rate is quite low. Specificity is equal to 0.38095. This is almost one third of the probability of without gene A in health and true negative rate. The findings reveal that there is a 30% or 0.30 probability of gene A. This is also called the risk of gene A or the estimated probability. The probability of the exposed cases is equal to 0.4667 while the probability of the exposed control is equal to 0.1333. The total of exposed respondents has a value of 0.3000. With 95% confidence, the interval 1.410185 to 27.15574 contains the unknown mean . Question 2 Within four years of hearts and lungs transplantations, a total of 138 organs were received. Out of this, 97 hearts were transplanted or 70.29 percent, and 41 Lungs or 29.71 percent. This means that there were more hearts transplanted than lungs. Negative skew or skewed to the left is indicated by mean value of 46.48454, which is lower than median (50) and median is lower than mode (59). This also means a longer left tail, extreme low scores with low frequency. This implies that there were more adult recipients of heart transplant than young adults. The distribution of the lungs organ in age at transplant was group into smallest and largest. The grouping showed that it skewed to the left or negatively skewed. The mean 49.56098 is lower than median 53 and median is lower than mode 59. This implies that there was more adults' recipient of lungs transplant than the young adult. The summary of the distribution of the hearts and lungs organ in age at transplant reveals that majority of the age at transplant of Hearts and lungs organ were lying on largest, which means that there were more adults who needed organ transplant than young adults. It can be gleaned from the table that there were 86 male who received heart organ or 88.66 percent while female just received 11 heart organs or 11.34 percent, while for lungs organ 22 male received the organ or 53.66 percent have lung transplant while 19 female or 46.34 percent have lung transplant. This means that majority of the heart and lungs organ transplant were received by male. The table shows that the creatinine level at heart transplant was group into smallest and largest. The grouping shows that it is skewed to the right or positively skewed. The mean (114.5362) is greater than median (107) and median is greater than mode (68). This means that the level of creatinine was lying on smallest group. This implies that creatinine at heart transplant was maintained at approximately 14 to 26 mg/kg/day well within normal level indicative of regular excretory kidney functions. It ca be gleaned from the table that the creatinine level during lungs transplant was grouped into smallest and largest. The grouping shows that it is skewed to the right or positively skewed. The mean (90.65625) is greater than median (87) and median is greater than mode (67). This means that the level of creatinine was lying on smallest group. This implies that creatinine at lungs transplant was well within normal level or approximately 14 to 26 mg/kg/day well within normal level indicative of regular excretory kidney functions. It can be seen in the table that there were sixty nine (69) heart organs received, found no complication after surgery or approximately seventy one percent (71.13%) while twenty eight (28) heart organ or twenty eight point eighty seven percent (28.87%) were found to have complications after surgery. And in terms of lungs organ received there were thirty two (32) or seventy eight percent (78%) were found no complication after surgery. However there were 9 lungs organ has a complication after surgery or approximately twenty one point ninety five percent (21.95%). This means that two third (2/3)of the heart and lungs organ received after surgery have no complications, however, one third (1/3) had complication. This means that there is ratio of survival was two is to one (2:1). The Lungs mean value of 0.2195122 is inside the range of the confidence interval of 0.0928145 and 0.3462098.This means that we can accept the null hypothesis. On the other hand, the hearts mean value of 0.2886598 is inside the range of the confidence interval of 0.1984832 and 0.3788364. This means that we can accept the null hypothesis. Question 3 a) There were nine hundred seventy one (971) patients undergoing primary hip replacement required allogeneic transfusion. b) There were 188 patients who required both autologous and allogeneic transfusion. This can be proven by the overall rate of allogeneic transfusion which was nine percent (9%) and in overall rate of autologous transfusion which was seventy five point five percent (75.5%) in patients who pre-donated. c) Yes, the 9.8gL-1 difference is statistically significant, however, there was no standard of error, so, standard deviation can not be calculated, subsequently, 95% confidence interval can not be computed. d) It is clear that the N was only 2041 which is supposedly 2042 from table 2 page 89 or it can be 2045 which is from the abstract page 87(2233-188). Based from table 3 the error arose when the N value of No ABD( 850 + 103 + 1005 + 72 = 2030) were summed up, and when the N of ABD were summed up, the total was 187 (121 + 10 + 56). e) Yes it was because in simple linear regression the requirement is only two predictors, thus, in this case considering the number of predictors, multiple regression modeling is fit. f) In any stepwise procedure, first, one must not forget the significant test at every level. For example, five percent (5%) or 0.05 probability of rejecting one null hypothesis is larger than five percent (5%). In order to prevent including variables that do not have a say to the discriminatory power of the model in the population, spell out a smaller significance level. Most applications however have all variables considered as having discriminatory power notwithstanding size. When choosing for an appropriate model, lookout alongside estimating more parameters than can be reliably estimated with a given sample size. While univariate analysis is an examination with two groups of independent variable and one dependent variable using t test and analysis of variance. g) The highest probability of transfusion was in revision hip which has a ratio of 5.4 while there were 2.2 risk ratios on primary hip, and on Knee, 1 risk ratio. h) The patient with age 77 has greater risk with a value of 1.8 compared to age 67 which has a risk ratio or probability of transfusion value of 1. i) When the coefficient is greater the risk ratio is also greater. This means that they are directly proportional. j) Validation Set Predicted probability N % transfusion 0.10 0.17 0.04 0.20 0.36 0.17 0.30 0.22 0.26 0.40 0.03 0.04 0.50 0.08 0.15 0.60 0.03 0.07 0.70 0.04 0.10 0.80 0.02 0.07 0.90 0.02 0.07 1.00 0.01 0.03 1.10 1.00 1.00 k) In a rating scale of 1 to 10, I would say 1 or dreadful because figures are not consistent making the research unreliable. For example in the abstract, the total number of samples 2233 is not consistent with the given figures on table 3. This should not happen because as a researcher, one should be familiar with accuracy and precision on reflecting correct figures as data collected for analysis to avoid further bias in estimations other than the level of error chosen to be tolerated Conclusions and Discussions Based on the findings of the study, the researcher came up with the following conclusions: 1.) There are minimal numbers of diabetics with gene A. The Sensitivity of the study of gene A is equal to 0.22222, which means that the patient with gene A and true gene A rate is quite low. Specificity is equal to 0.38095. This is almost one third of the probability of without gene A in health and true negative rate. The findings reveal that there is a 30% or 0.30 probability of gene A. This is also called the risk of gene A or the estimated probability. The probability of the exposed cases is equal to 0.4667 while the probability of the exposed control is equal to 0.1333. The total of exposed respondents has a value of 0.3000. With 95% confidence, the interval 1.410185 to 27.15574 contains the unknown mean . 2.) It can be seen in the table that there were sixty nine (69) heart organs received, found no complication after surgery or approximately seventy one percent (71.13%) while twenty eight (28) heart organ or twenty eight point eighty seven percent (28.87%) were found to have complications after surgery. And in terms of lungs organ received there were thirty two (32) or seventy eight percent (78%) were found no complication after surgery. However there were 9 lungs organ has a complication after surgery or approximately twenty one point ninety five percent (21.95%). This means that two third (2/3)of the heart and lungs organ received after surgery have no complications, however, one third (1/3) had complication. This means that there is ratio of survival was two is to one (2:1). The Lungs mean value of 0.2195122 is inside the range of the confidence interval of 0.0928145 and 0.3462098.This means that we can accept the null hypothesis. On the other hand, the hearts mean value of 0.2886598 is inside the range of the confidence interval of 0.1984832 and 0.3788364. This means that we can accept the null hypothesis. c) In a rating scale of 1 to 10, I would say 1 or dreadful because figures are not consistent making the research unreliable. Works Cited Anderson, RP, Jin, R, Grunkemeier. Understanding logistic regression analysis in clinical reports: an introduction. Ann Thorac Surg 75,753-757. Feagan et al. 2001. Transfusion Practice in Elective Orthopaedic Surgery. London Clinical Trials Research Group, The John P. Robarts Research Institute, London, Ontario, Canada. feagan@lctrg.com 11, 87-95. Jenicek, Milos. 2002. Foundations of Evidence-Based Medicine. New York: Parthenon Publishing. Loong, T-W. 2003. Understanding Sensitivity and Specificity with the Right Side of the Brain, BMJ: 327 716-719. Md, Martin Silink, Kaichi Kida Md Phd, and Arlan L Rosenbloom Md, eds. 2003. Type 2 Diabetes in Childhood and Adolescence: A Global Perspective. London: Martin Dunitz. . Read More