4.3 Choosing a sample size calculator

To make life easier when designing a study there are several sample size calculators available online, including Epitools [Tip: hold Ctrl and click a link to open it in a new tab. (Hide tip)] and Sample Size Calculators. Which sample size calculator to choose depends on the objectives of a study. Many AMR studies aim to estimate a single proportion, such as the proportion of isolates that are resistant to a particular drug. Some also compare two proportions, for example the number of resistant isolates in cases in one region compared to another region. Similarly, AMU studies might aim to measure the total amount of AMU, or compare AMU between farms. There are appropriate sample size calculators available for both of these objectives.

In addition to specifying the desired precision, confidence level and power to calculate a sample size, you may also have to provide values for other parameters. For example, to estimate the sample size required for a study comparing two proportions, you need to provide an input value for the proportion in the baseline (or control) population, and an input value for the proportion in the comparison population. This is where it can be helpful to think about the minimum clinically meaningful difference you wish to detect.

Activity 6: Sample size calculations in practice

Timing: Allow about 30 minutes

Let’s use an online statistical calculator to calculate the sample size for a study comparing two proportions.

Imagine you wish to compare the prevalence of multidrug-resistantSalmonella isolates on pig farms in two neighbouring regions. You know from a previous study that the prevalence of multidrug resistance in Salmonella isolates in pigs from one of the two provinces is around 50%. You’re not sure about the equivalent figure in the second province, but you are concerned that it is higher than in the first province. You decide it is important to be able to detect a prevalence of multidrug resistance of at least 55% in the second region, to test your suspicions. You would like your study to have a confidence level of 95%, and a power of 90%.

Using this Epitools online calculator, calculate the minimum sample size required to detect a difference in multidrug resistance prevalence between the two provinces. Don’t change the default values for ‘ratio of sample sizes’, or ‘use of 1 or 2 tailed test’ (which should be set at 1, and 2-tailed respectively) – these concepts are beyond the scope of this module.

Hint: remember that a proportion is a decimal between 0 and 1, so 50% is expressed as 0.5. You usually need to enter proportions, not percentages, when calculating sample sizes.

Fill in your values in the calculator and click on ‘Submit’. Use the space below to make notes on what you have done, and on the outcomes of the calculation. Are you surprised by the size of the sample needed to answer your experimental question?

To use this interactive functionality a free OU account is required. Sign in or register.
Interactive feature not available in single page view (see it in standard view).

Discussion

The total required sample size is 4268 isolates. This means you would need to collect samples from at least 2134 pigs in each province. In practice, more pigs may need to be sampled than calculated, to compensate for potential issues such as loss of collected samples, issues with sample transport, failure to isolate Salmonella from some of the collected samples, expected bacterial prevalence (because not all animals will normally carry Salmonella – see details in Section 5.4) and so on. This type of consideration is often described as adjusting for the anticipated ‘dropout rate’ of the study. The size of the required sample is likely to have resource implications that need to be considered before the decision is made to proceed.

4.2 Statistical determinants of sample size

4.4 Adjusting statistical parameters to suit constraints