Hypothesis testing and statistical analyses for clinical research

Hypothesis testing and statistical analyses for clinical research

This is an assignment that focuses on the Hypothesis testing and statistical analyses for clinical research. The paper also explains the difference between null and alternate hypothesis.

Hypothesis testing and statistical analyses for clinical research

Assignment 2: Hypothesis testing and statistical analyses
This is the second of 3 formative assignments for Clinical Research Methods II, and counts 60% towards your semester mark.
Plagiarism: You must include a plagiarism declaration after your title page. Your assignment will automatically be submitted to Turnitin when you submit.  You will have a maximum of three submission attempts.
Formatting: Please use Arial, size 10 font with normal margins, 1.15 line spacing, UK or SA English and also justified paragraphs.  You must also include a title page clearly indicating the name of the assignment, your full name and student number and the course code.  Additionally, you must include a plagiarism declaration after your title page. Make sure you use the correct numbers to indicate your answers i.e. make sure they correspond to the question number. Where applicable, please use the Vancouver referencing style.

Questions:
1.      Part 1: Hypothesis testing (15 marks)

1.1.   Explain in your own words what the difference is between the null and alternate hypothesis, and why we say that research studies aim to disprove rather than prove a hypothesis.                                                                                                                                                             (5 marks)

1.2.   You would like to test whether a simple intervention, such as allocating an emergency stock of pre-numbered/pre-labelled patient folders, will improve patient throughput in the emergency department during a mass incident.  Turn this into a research question using the PICO format (list each of the PICO elements AND write out your research question in full) and then generate the null and alternate hypotheses that you will test.
(10 marks)

Hypothesis testing and statistical analyses for clinical research

2.      Part 2: Statistical analyses (65 marks)

2.1.   In your own words, explain the difference between parametric and non-parametric methods of data analysis. However, in your explanation make sure you describe what is meant by the term “parameter”, how it is different to “variables” and how this relates to choosing the most appropriate statistical test.                                                                                                 (8 marks)

2.2.   Describe the distinguishing characteristics of the normal/Gaussian distribution.      (5 marks)

2.3.   Explain how you differentiate the Independent variable from the Dependent variable. (2 marks)

2.4.   For each of the following research scenarios:
(i)               Firstly, list the independent variable, and indicate what type of data it is                   (2 marks)
(ii)              Secondly, list the dependent variable, and indicate what type of data it is                      (2 marks)
(iii)            Thirdly, use the table provided (Appendix 1) to look up the most appropriate statistical test                    (2 marks)
(iv)            Fourthly, explain in your own words what your chosen statistical test does?                  (2 marks)

2.4.1  Little is known about the association between the duration of prehospital CPR by emergency medical services and outcomes among paediatric patients with out-of-hospital cardiac arrests (OHCAs). This study included a large sample of paediatric patients with OHCAs before EMS arrival who were transported to medical institutions after resuscitation by bystanders or EMS personnel. You want to compare mean prehospital CPR duration (defined as the time from CPR initiation to the time of prehospital return of spontaneous circulation or to the time of hospital arrival) between patients with a favourable outcome at 1-month versus those with an unfavourable outcome at 1-month.
(Adapted from this paper: DOI: 10.1186/s13049-019-0658-7)

Hypothesis testing and statistical analyses for clinical research

2.4.2  The objective of this study was to identify differences in the diurnal timing of NSTEMI versus STEMI presentations in the emergency department.  Retrospective records of patients with myocardial infarctions presenting to an emergency department were retrieved for the years 2013-2017. NSTEMI and STEMI cases were identified based on diagnosis codes, and also arrival times to the emergency department were categorized into 4 time periods (12AM – 5:59AM, 6AM – 11:59AM, 12PM – 5:59PM, 6PM – 11:59PM)
(Adapted from this paper: https://doi.org/10.1016/j.annemergmed.2018.08.036)

2.4.3. You have a young colleague who started work with you 2 weeks ago. Also, you admire their eagerness and work ethic, and hope they won’t burn out too soon. You start to wonder whether measures of burnout differ by levels of experience. To measure levels of burnout, you will use a series of statements for which colleagues can choose a number on a 5-point Likert scale (ranging between strongly agree and strongly disagree), and you will compare these between colleagues who joined the service less than 2 years ago, those who have been in the service 2-5 years, and those who have been there more than 5 years.

2.4.4.  Consequently, you want to investigate the relationship between birth weight (in kgs) and body-mass index (BMI) in adolescence. You recruit a sample of 16-19 year olds and measure their current height and weight, and obtain their confirmed birth weight from patient records.

2.5.   Statistical testing is largely about helping us objectively decide whether the differences we observe in our datasets are real or due to chance.
2.5.1. Explain, in your own words, what a p-value indicates                                              (2 marks)
2.5.2. Also, what are some of the limitations of the p-value?                                                     (5 marks)
2.5.3. How do we address the issue of uncertainty when we present our results.       (3 marks)