NURS 8201 Use of Regression Analysis in Clinical Practice

NURS 8201 Use of Regression Analysis in Clinical Practice

NURS 8201 Use of Regression Analysis in Clinical Practice

Regression analysis is one of the statistical models used in estimating the relationship between variables. The researcher has the ability to determine the effect that an independent variable has on the dependent variable (Willis & Riley, 2017). For example, an increase in one or more values on the independent variable would have an effect on the dependent variable. This paper examines regression analysis was used by an author including its weaknesses and strengths.

Article Summary

The article authored by Hatakeyama et al., (2019) aimed at finding the relationship between quality of clinical practice guideline (CPGs) and overall assessment scores. This study considered the previous studies that had been done and published between 2011 and 2015. These selected studies were subjected through an independent valuation using AGREE II. The author analyzed the results using a regression analysis. For instance, the analysis included the effect that the six domains and 23 items has on the overall assessment. The study collected a total of 206 CPGs and correlated all the domains to the items on the overall assessment to determine the strength of the relationship before taking the regression analysis on the proposed items.

Use of Regression on the Article

The author decided to subject domain 3, domain 4, domain 5, and domain 6 of the regression analysis. Domain three represented rigor of development, domain four was for clarity of presentation, domain five was for applicability and finally domain 6 was for editorial independence. The analysis was majoring on how these domains influence the overall assessment (Hatakeyama et al., 2019). The analysis showed that all the domains had a significant relationship with the overall assessment. The author also found that four different items on AGREE II, which were item 8, 15, 19 and 22 had an effect on overall assessment. The regression analysis showed that the change in one unit of the items above had a significant change on the overall assessment which in this case acted as the dependent variable (Hatakeyama et al., 2019). Therefore, the improvement of overall assessment dependent on the increase and decrease of the items that acted as independent variables in this case.

Other statistical analysis that could have been used in the study is ANOVA analysis because it shows the strength of the relationship between the items selected. Besides, it allows the researcher to determine the effect that each dependent variables have on each other and how the relationship between the dependent variables can influence the study (Fontaine et al., 2019). Use of ANOVA tests in this study could have strengthened and relayed more information on the collection of items that could have a great impact on the overall assessment.

Click here to ORDER an A++ paper from our Verified MASTERS and DOCTORATE WRITERS: NURS 8201 Use of Regression Analysis in Clinical Practice

The strength of the regression analysis is on the ability of the author to examine more than one dependent

NURS 8201 Use of Regression Analysis in Clinical Practice
NURS 8201 Use of Regression Analysis in Clinical Practice

variable. According to the study the author was interested in 22 items and their effect on overall assessment. The study is able to report on the influence of 22 items more easily as compared to other methods that could have been complex (Hatakeyama et al., 2019). Despite the strength that regression analysis has on the study, the method also has its weakness it lacks the ability to examine the relationship between the independent variables considered in the study.

Conclusion

Regression analysis is a powerful tool in assessing the relationship between dependent and independent variables. The author in the selected the study has the ability to evaluate which of the 22 items have a high or low effect on the overall assessment.

References

Fontaine, G., Cossette, S., Maheu-Cadotte, M. A., Deschênes, M. F., Rouleau, G., Lavallée, A., … & Mailhot, T. (2019). Effect of implementation interventions on nurses’ behaviour in clinical practice: a systematic review, meta-analysis and meta-regression protocol. Systematic reviews8(1), 1-10. https://doi.org/10.1186/s13643-019-1227-x

Hatakeyama, Y., Seto, K., Amin, R., Kitazawa, T., Fujita, S., Matsumoto, K., & Hasegawa, T. (2019). The structure of the quality of clinical practice guidelines with the items and overall assessment in AGREE II: a regression analysis. BMC health services research19(1), 1-8. https://doi.org/10.1186/s12913-019-4532-0

Willis, B. H., & Riley, R. D. (2017). Measuring the statistical validity of summary meta‐analysis and meta‐regression results for use in clinical practice. Statistics in medicine36(21), 3283-3301. https://doi.org/10.1002/sim.7372

Logistic Regression in Nursing Practice

The article I chose to examine and analyze this week is “Prediction of influenza vaccination outcome by neural networks and logistic regression.”

Tritica- Majnaric,L., Zekic-Susac, M., Sarlija, N., & Vitale, B. (2010).Prediction of influenza vaccination outcome by neural networks and logistic regression. Journal of Biomedical Informatics, 43(5), 774-781

Post your critical analysis of the article as outlined above

The critical issue with influenza vaccination is to foretell vaccine efficacy. The study presented in this article aimed to create a model to facilitate a credible prediction of the effect on influenza vaccination contingent upon valid medical data. A neural network approach was utilized, and its presentation was compared with using the logistic regression model The three neural network algorithms that were tested include multilayer perceptron, radial basis, and probabilistic in conjunction with parameter optimization and regularization techniques to create an influenza vaccination model that could be used for prediction purposes in the medical practice of primary health care physicians, where the vaccine is usually dispensed (Tricia- Majnaric, Zekic-Susac, Sarlija, & Vitale, 2010). The variety of input variables was determined from the model of the vaccine strain, which has been altered and which a poor influenza reaction is likely. The action of models was quantified by the standard hit rate of difference in vaccine outcomes. Sensitivity analysis was performed on the best model, and the importance of input variables was discussed (Tricia- Majnaric, Zekic-Susac, Sarlija, & Vitale, 2010).

Logistic regression was widely used for dissecting the multivariate data, including dichotomous reactions with this research study. Logistic regression is identified as an analysis relationship between multiple independent variables and a single dependent variable which yields a predictive equation (Polit, 2010). The three neural network algorithms along with the logistic regression model were implemented to deliver the influenza vaccination probability model and apply it for prediction purposes within practice outcomes. Another statistical method that could address the research problem discussed in this article is the analysis of variance (ANOVA) statistical technique. ANOVA entails analyzing and arranging differences among three or more groups being compared to draw inferences. From this research study, the three NN algorithms, multilayer perceptron (MLP), radial-basis function network (RBFN), and probabilistic network (PNN), were tested and analyzed to draw an inferential conclusion.

Propose potential remedies to address the weaknesses of each study.

A strength of this study was the degree of the sensitivity and specificity analysis that was appropriately expressed. The generalization ability of the models used by a 10-fold cross-validation procedure showed that the model attained by multilayer perception produced the highest average hit rate among neural network algorithms and also outperformed the logistic regression model about sensitivity and specificity discussed (Tricia- Majnaric, Zekic-Susac, Sarlija, & Vitale, 2010). The sensitivity analysis was also implemented on the best models, and the vitality of the input variables was evaluated. A weakness of this study was the small sample size. The study consisted of 90 patients out of 150 people who required the influenza vaccine between 2003-2004 in Croatia. A more significant sample number would have been beneficial in this dataset and incorporating other methods to discover a more successful model.

Analyze the importance of this study to evidence-based practice, the nursing profession, or society.

The importance of prevention and control of the influence epidemic is imperative to the nursing profession and health care society. Based on the results of this research study, it is indicated that both types of data, those related to previous influenza viruses exposure and those describing the health status of examinees, influence outcome values of performed predictive models and could be used as efficient predictors of the influenza vaccine efficacy discussed (Tricia- Majnaric, Zekic-Susac, Sarlija, & Vitale, 2010). This is important, especially in the elderly, as this population has been more affected by the influenza virus. The available vaccines are less effective in this age group. With new vaccine preparations and approaches, the effectiveness of influenza vaccines can be implemented through evidence-based practice.

References

Polit, D. F. (2010). Statistics and data analysis for nursing research (2nd ed.). Upper Saddle River, NJ: Pearson Education.

Tritica- Majnaric,L., Zekic-Susac, M., Sarlija, N., & Vitale, B. (2010).Prediction of influenza vaccination outcome by neural networks and logistic regression. Journal of Biomedical Informatics, 43(5), 774-781