Summary

Introduction

Alpha is a disease caused by pathogen X, that has spread throughout the United Kingdom that has been around for at least 1 year. Whilst Alpha’s symptoms are unique enough to be detected by a doctor, Alpha’s presence can be confirmed using an RT-PCR test. Applying the RT-PCR test to the genetic code of pathogen X returns a Cycle Threshold (Ct) value, a logarithmic scale that is related to how quickly antibodies can detect pathogen X. Currently, the mean Ct value for symptomatic cases is 30.20 in England, Wales, and Scotland. However, not all cases may have clear symptoms, so a value of 34.90 is currently set as the clinically significant critical Ct value for a positive test to occur. Given that there have been both local anecdotal reports of false negative tests (resultant from higher Ct values) and foreign reports of higher loads of X (resultant from lower Ct values), this report will aim to determine whether such a new variant exists within England, Scotland or Wales.

To do so, recent data showing the Ct values to 2 decimal places of people with the symptoms of Alpha in England, Scotland and Wales have been compared to a larger set of Ct values (also to 2 decimal places) from a year earlier, when no such variants existed. Overall, the aim is to determine whether new variants have arrived in England, Wales or Scotland by seeing if the mean Ct Value for symptomatic cases of 30.20 has changed. If the mean Ct value has decreased, then the viral load of pathogen X has increased, so more stringent steps would need to be taken to prevent transmission of Alpha. If the mean Ct Value has increased, the current critical value of 34.90 would likely need to be increased to decrease the chances of a false negative test (testing negative despite having symptoms of Alpha). In either case, treatment may need to be modified to accommodate for the new variant of Alpha.

Methods

The pre-variant data set, shown below, already follows the normal distribution so no assumptions need to be made to undergo a statistical hypothesis test.

Figure 1: This is a histogram showing the normal distribution of the Pre variant data.

Figure 1: This is a histogram showing the normal distribution of the Pre variant data.

Box plots of the other 3 data sets were generated to determine any outliers. These outliers, along with any Ct values below 0 (given this number must be positive) were removed from these data sets. This reduced the modified data set for England from 100 to 96 elements, for Scotland from 32 to 29 with Wales’ data set unchanged on 25 elements.

Before any statistical hypothesis tests on these 3 data sets can be performed, it must be checked to see if the data sets mentioned in the introduction can be approximated to a normal distribution. This is done using the Central Limit Theorem:

This states that if the samples within a data set are identically independently distributed with some mean µ and finite variance \(σ^2\), then the sample mean \(x̄\) ~ \(N(µ, σ^2/n)\) - approximately normally distributed with mean µ and variance \(σ^2/n\) if n (the number of samples in the data set) is large. Typically, this approximation requires that n is larger than 30 to hold.

In this case, all 4 data sets are assumed to be independent and identically distributed. On an individual basis, it should be assumed that all individual samples are independent of each other and equally distributed. From a geographic perspective, it should be assumed that all 3 regions are geographically distinct (results from one region do not affect the other regions) for geographic independence and that results from all 3 regions are also identically distributed.

Unfortunately, modifying the data sets means that neither the Data set for Wales nor Scotland have enough elements to satisfy that n is large. Regardless, two-tailed hypothesis tests at a confidence level of 10% were performed on all 4 data sets.

Boxplot:

A boxplot is a simple method graphically representing the distribution of data given. It consists of the lower and upper quartile, the median, the highest and lowest value that is not an outlier, and outliers (if any). We have used boxplots in order to graphically visualise the data for England, Scotland and Wales and identify any outliers. We have used a common definition that an outlier is any value that is :

  • either greater than \(Q_3\)+1.5(\(Q_3\)-\(Q_1\))
  • or less than \(Q_1\)-1.5(\(Q_3\)-\(Q_1\))

where \(Q_1\) is the lower quartile, \(Q_3\) is the upper quartile, and \(Q_3\)-\(Q_1\) is the interquartile range.

Hypothesis Test:

A hypothesis is a statement made about the value of a population parameter. You can test a hypothesis about a population by carrying out an experiment or taking a sample from the population. In order to carry out the test, two hypotheses need to be formed; the null hypothesis which is the hypothesis that you assume to be correct, and the alternative hypothesis which tells you about the parameter if your assumption is shown to be wrong.

To carry out a hypothesis test you assume the null hypothesis is true, then consider how likely the observed value of the test statistic was to occur. If this likelihood is less than a given threshold, called the significance level of the test, then we can reject the null hypothesis.

A critical region is a region of the probability distribution which, if the test statistic falls within it, would cause us to reject the null hypothesis. The critical value is the first value to fall inside of the critical region.

We can also test hypotheses about the mean of a normally distributed random variable by looking at the mean of a sample taken from the whole population. Thus, in this case, hypothesis testing is the appropriate method of testing whether there are new variants of alpha as we are able to use a two-tailed test to examine whether there are new variants of alpha by comparing the mean of each sample data of England, Scotland and Wales to the mean Ct value of 30.20.

Results

Figure 2: This is a histogram showing the normal distribution of the data from England.

Figure 2: This is a histogram showing the normal distribution of the data from England.

Figure 3: This is a boxplot showing the distribution of the data from England and any outliers the data has which are represent by hollow circles.

Figure 3: This is a boxplot showing the distribution of the data from England and any outliers the data has which are represent by hollow circles.

Figure 4: This is a histogram showing the normal distribution of the data from England after anomalous data has been removed.

Figure 4: This is a histogram showing the normal distribution of the data from England after anomalous data has been removed.

Figure 5: This is a histogram showing the normal distribution of the data from Scotland.

Figure 5: This is a histogram showing the normal distribution of the data from Scotland.

Figure 6: This is a boxplot showing the distribution of the data from Scotland and any outliers the data has which are represent by hollow circles.

Figure 6: This is a boxplot showing the distribution of the data from Scotland and any outliers the data has which are represent by hollow circles.

Figure 6: This is a histogram showing the normal distribution of the data from Scotland after anomalous data has been removed.

Figure 6: This is a histogram showing the normal distribution of the data from Scotland after anomalous data has been removed.

Figure 7: This is a histogram showing the normal distribution of the data from Wales.

Figure 7: This is a histogram showing the normal distribution of the data from Wales.

Figure 8: This is a boxplot showing the distribution of the data from Wales and any outliers the data has which are represent by hollow circles.

Figure 8: This is a boxplot showing the distribution of the data from Wales and any outliers the data has which are represent by hollow circles.

As shown in the boxplot of the Wales data, there are no anomalous data points. Thus the Wales data did not need to be cleaned and a modded graph of the Wales data was not necessary.

#Conclusion \(H_0\): µ = 30.20

\(H_1\): µ ≠ 30.20