Sampling: Performance Enhancing Drugs

Purpose:  To address how the testing of athletes for the use steroids should be carried out.  This will include both who to sample (e.g. winner only, everyone, random selection...) and what to sample (e.g. hair, urine, blood...)

Learning Outcomes:

By the end of the assignment students will be able to:

• Assess the limitations of different sample types (e.g. urine, blood, hair)
• Define various sampling strategies
• Develop a rational sampling process

Types of Sampling Plans:

A range of sampling strategies exist and can be used to dealing with all manner of large samples ranging from large groups of individuals to lake water analysis.  Each strategy has advantages and disadvantages which must be considered in the selection of the strategy.  Summarized below are the most common sampling strategies employed in chemical analysis:

• Random Sampling: Despite the name, a random sampling strategy does require planning to execute properly, without a plan unintended biases may be introduced into the sampling strategy. For example of a random sample would be the division of a land plot into a uniform, numbered, grid, from which random sectors could be selected via a random number table/generator, for contaminant analysis (e.g. PCBs, heavy metals...).

• Selective (Judgmental) Sampling: This is at the opposite extreme of random sampling, and is done if you have prior information about the target. For example, if you wanted to evaluate the metal content in pennies you may not select coins that are corroded or choose coins from a specific mint date.

• Systematic Sampling: Sampling the target population at regular intervals in space or time. This is often considered to fall between the extremes of random and selective sampling. For example testing river water for pesticide content at a given point at set time intervals (e.g. every two hours).

• Systematic - Judgmental Sampling: This method is an example of the possible combinations of the above three methods (other combinations are also possible). This strategy can be employed when there is some prior knowledge about the system that can guide the sampling.  For example, if a waste site was leaching toxins into the soil, knowledge of the flow of water in the soil would direct the sampling to one side or the other of the waste site (Judgmental Sampling).  This selected region could then be spatially analyzed at regular intervals (Systematic Sampling).

• Stratified Sampling: The population may be divided into sub populations (groups) that are distinctly different (this might be size of sample, type of sample, depth of sample). Then, the overall sampling within the groups is randomly conducted and the samples are pooled. For example if soil samples are to be analyzed for lead content the soil sample can be sifted into different particle sizes (e.g >1 mm, 1 mm to 0.1 mm, < 0.1 mm). After the sub populations are established random samples are taken pooled from each sub population for analysis.

• Cluster sampling: is a sampling technique where the population is divided into groups or clusters and random samples are selected from the cluster for analysis. The main objective of cluster sampling is to reduce costs by increasing sampling efficiency. For example, this method could be used if you wished to test for lead exposure from water pipes in a city.  Clusters could be made from individual blocks, or neighborhoods, and from these clusters random individuals could be selected for testing.

Who to Test?

In the 1988 Summer Olympics, in the 100 m final a new world record was set for the 100 m dash at 9.79 seconds by the Canadian sprinter Ben Johnson.  However, this result was disqualified three days later when Johnson’s drug test revealed the presence of the synthetic steroid Stanozolol.  In fact, over the course of their careers, five of the ten finalists in the 1988 Olympic 100 m final tested positive for some form of steroid abuse.  With small field sizes (e.g. ten athletes) the testing of all athletes in the competition is not overly burdensome (in both cost and time) increasing the likelihood that an athlete who is using performance enhancing drugs will be detected.

However, with some events greater numbers of athletes are involved, such as football with rosters of almost 80 members or cycling with over 200 athletes per race, testing every athlete become prohibitive.  With the aim of identifying as many athletes who are doping as possible, while doing so in a reasonable amount of time and at a reasonable expense, how would you go about selecting the athletes to test in each of the scenarios below?

1. Presuming you have a large field of athletes competing in a single day event, such as a one-day cycling event (e.g., Paris-Roubaix), how would you go about sampling the athletes to determine if any had used performance enhancing drugs?

2. How would you change your sampling procedure if the cyclists were competing in a multi-stage race (e.g., the Tour de France) where the overall winner has the lowest cumulative time for all stages? This type of race has resulted in a few cyclists winning the Tour de France without ever having won a single stage of the race.

3. In team sports such as football where there is no clear winner among the individuals on a team, how would you go about selecting athletes to test for performance enhancing drugs?

4. The tables below are graphic representations of the final position (number) of competitors in a single day cycling race. In each table, select eight competitors who will be tested for performance enhancing drugs based on the sampling strategy listed above the table.

5. Which of the above sampling strategies is likely to be the most successful in identifying those athletes who are using performance-enhancing drugs in a single day cycling race?

6. If the numbers on the charts represented the jersey number for players on a baseball team, which method would be most successful in identifying those players using performance enhancing drugs?

7. Describe a situation where a systematic sampling protocol would be favorable over a random sampling protocol.

8. What concentration of a steroid in a sample from an athlete would be considered to be doping? Will it make a difference if the steroid is an endogenous (already exists in humans) or exogenous (does not exist in humans) compound?

9. How might doping with an endogenous steroid (e.g. testosterone) be identified?

What to Test?

The testing of human samples is not as simple as testing water samples from a river, one cannot readily aliquot a sample directly from the source.  Consequently, consideration must be given to how samples can be obtained as well as whether or not the analyte of interest will be found in those samples.  Traditionally, biological samples have included blood, urine and hair, as these samples can be obtained from an individual with little to no discomfort or health risk.

A secondary consideration to the sampling of biological organisms is the lifetime of the analyte in the organism.  Some analytes may be rapidly processed and evidence of their presence may be short lived, whereas others may persist for extended periods of time.  In some instances this may not be a problem, as the effect of the performance enhancer may only persist during the period of time when the drug is in the body; however, other drugs can have effects that last well beyond their lifetime in the body.

1. Which of the three sample types (hair, urine, blood) is likely to have the greatest variability in its composition when taken from an individual at different points in time? What factor(s) will have the greatest influence on this sample?

2. Which of the three sample types would allow for the determination of steroid use at the furthest point in time after the use of the drug?

3. How will the variability in the sample type identified above impact the ability to quantify testosterone or other steroids in the sample? What could be done to correct for this impact?