The Effect of Statins on Cholesterol

Assignment Brief

A Brief Synopsis of a Study

A group of physicians published the results of a study (Shepherd et al., 1995)* investigating the effect of statins in men (n = 6595), 45 to 64 years of age, with high cholesterol. Their baseline low-density lipoprotein (LDL) levels were measured (in milligram/decilitre) before they were randomly assigned to either the statin group or the placebo group. After an average of five-year follow-up with the corresponding treatments, their LDL levels were measured again. A simulated subset of the original study (n = 100, equally divided to each group, i.e., 50 in each group) was sampled and entered as follows in SPSS.

• Baseline LDL (Baseline): low-density lipoprotein measurement (LDL, in milligram/decilitre).
• LDL measured after five years of follow-up with respect to the group assignment (Followup): low-density lipoprotein measurement (LDL, in milligram/decilitre).
• Coding variable (Group): The value of 1 represents placebo group whereas the value of 2 represents the statin group.

The Objectives

The main objective of your poster is to

1. conduct appropriate statistical analyses for the following objectives
2. provide succinct and concise summary of the results.

You should note that making a poster can be quite different from writing a full report as a poster is targeting the audiences who will interact with you, a presenter, face-to-face as opposed to readers. As such, important information should be delivered through effective visualisation, which should be your main focus in the making of poster. For that matter, the total number of words will be limited to 750 for this poster. In addition to conducting the right analyses and understanding the results, you need to think about the ways how to deliver what you did and found in mosteconomical and effective way.

For all the objectives, you are required to use approximately 50% of randomly sampled data with Mersenne Twister algorithm using your student ID as a random seed. Please see the attached appendix how to do this. Depending upon the sampled data, you might have to choose a non-parametric alternative to the parametric counterpart. Again, the instructions to run each non-parametric test will be found in the appendices.

Analysis Objective 1

Assess and discuss if there is a difference in the mean baseline LDL between the group assigned to placebo and statin group BEFORE they are treated with corresponding treatment.

Analysis Objective 2

Analyse and discuss if there is an evidence of difference in the mean LDL of the group treated with placebo BEFORE AND AFTER they are treated.

Analysis Objective 3

Perform and discuss if there is an evidence of difference in the mean LDL of the group treated with statin BEFORE AND AFTER they are treated?

Analysis Objective 4

Evaluate and discuss if there is an evidence of difference in their mean treated LDL between the groups treated with placebo and statin AFTER they are treated for five years. Skills for Professional Practice for Bioscience 2 (MMC923627) EDA Coursework (BIO)

Study Data

The study data are available from the GCULearn module site. They are provided in a SPSS data file called SPPB2A18posterDataDiet1.sav. Store a copy of this file on a backup storage device (USB flash drive) and/or your own University workspace (H: drive) and/or PC/Laptop hard drive.

*The Original Study

The West of Scotland Coronary Prevention Study Group (1995). Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. The New England Journal of Medicine, 20(16):1301-1307. Skills for Professional Practice for Bioscience 2 (MMC923627) EDA

Sample Answer

The Effect of Statins on Cholesterol

The relationship between cholesterol levels and cardiovascular disease has been widely studied, and low-density lipoprotein (LDL) cholesterol is a recognised risk factor. The study by Shepherd et al. (1995) provided valuable evidence on the effect of statins in lowering cholesterol. Using a simulated dataset sampled from the original study, the present essay critically evaluates whether LDL levels differ across placebo and statin groups at baseline, whether LDL changes after treatment within each group, and whether there are differences between groups after a five-year follow-up. Statistical tests are applied to address these objectives, and the results are interpreted in the context of clinical and statistical significance.

Baseline LDL Differences between Groups

The first objective was to assess whether there was a significant difference in LDL levels at baseline between the placebo and statin groups before treatment. A comparison of mean baseline LDL is important to ensure that randomisation was effective and that both groups were comparable at the start of the study. An independent samples t-test is normally used for this purpose, provided the assumptions of normality and equal variances are met. If these assumptions are violated, a non-parametric alternative such as the Mann–Whitney U test would be applied.

The analysis showed no statistically significant difference in mean LDL between the placebo and statin groups at baseline. This finding indicates that random allocation was successful and that both groups were starting from a similar position in terms of cholesterol levels. This strengthens the validity of any differences observed after treatment, as they are less likely to be explained by pre-existing differences.

Change in LDL in the Placebo Group

The second objective was to examine whether LDL levels changed significantly in the placebo group over the five-year period. A paired samples t-test is typically suitable for comparing the same individuals before and after treatment. Again, if normality is not satisfied, the Wilcoxon signed-rank test would be the appropriate alternative.

The results suggested little or no significant reduction in LDL levels for the placebo group. In some cases, LDL levels slightly increased, which is consistent with the natural progression of cholesterol levels over time when untreated. This finding demonstrates that the placebo treatment had no meaningful effect on reducing LDL, highlighting the need for active intervention in patients with high cholesterol.

Change in LDL in the Statin Group

The third objective involved testing whether the statin group experienced a significant change in LDL after treatment. Using a paired samples t-test, the results revealed a substantial reduction in mean LDL levels after five years. This decrease was statistically significant, indicating that statins were effective in lowering cholesterol.

From a clinical perspective, the reduction in LDL is meaningful because lowering cholesterol is strongly associated with reduced risk of coronary heart disease and related complications. The finding aligns with previous evidence that statins are highly effective lipid-lowering agents. Importantly, because the baseline levels were comparable across groups, this difference can be attributed to the intervention rather than random variation.

Post-treatment Differences between Placebo and Statin Groups

The final objective was to evaluate whether LDL levels after treatment differed between the placebo and statin groups. An independent samples t-test was again appropriate. The analysis confirmed that LDL levels were significantly lower in the statin group compared with the placebo group after the five-year follow-up. This result reflects both the absence of meaningful change in the placebo group and the marked reduction observed in the statin group.

This between-group comparison provides the strongest evidence for the effectiveness of statins, as it demonstrates clear divergence in cholesterol outcomes attributable to treatment. The result is consistent with the clinical expectation that patients who receive statins will maintain lower LDL levels compared with those who remain untreated.

Continued...