Before we can analyze data, we need to know what kind of data we're looking at. In biostatistics, data comes in four main types, each with its own rules and uses. Getting this right is the first step to sound analysis.

Nominal data is about naming things. You can't say blood type A is "more" or "less" than B; they're just different. Ordinal data adds a layer of order. A pain score of 8 is worse than 4, but not necessarily twice as bad. The distance between the numbers isn't uniform.

Interval and ratio data are where we get more precise measurements. With interval data, the difference between 37°C and 38°C is the same as between 39°C and 40°C. But 0°C doesn't mean "no heat." Ratio data has a true zero. A weight of 0 kg means "no weight," and 100 kg is exactly twice as heavy as 50 kg. This distinction is crucial for choosing the right statistical tools.

Once we have our data, we often want to find a single number that represents the whole group. This is a measure of central tendency. The three most common are the mean, median, and mode.

The mean is simply the average. You add up all the values and divide by the number of values. It’s useful, but can be misleading if there are extreme outliers. For instance, the average recovery time for a group of patients could be skewed by one person who took much longer to recover.

The median is the middle value when the data is sorted. It’s the number that splits the dataset in half. Because it’s based on position, it isn't affected by extreme outliers. If you have an even number of data points, the median is the average of the two middle values.

The mode is the value that appears most often. It’s the only measure of central tendency you can use for nominal data. A dataset can have one mode, more than one mode (multimodal), or no mode at all.

Knowing the center isn't enough. We also need to know how spread out the data is. Are all the values clustered tightly around the mean, or are they all over the place? This is where measures of dispersion, like variance and standard deviation, come in.

Imagine two drugs that both lower blood pressure by an average of 10 mmHg. For one drug, every patient sees a drop of exactly 10 mmHg. For the other, some patients see a drop of 20 mmHg while others see no change. The average is the same, but the second drug is far less predictable. The spread matters.

Variance gives us a measure of spread, but its units are squared (e.g., mmHg²), which is hard to interpret. To fix this, we take its square root.

A small standard deviation means the data points are clustered closely around the mean, suggesting consistency. A large standard deviation means the data is widely scattered.

In clinical practice, we rely on tests to diagnose diseases. But no test is perfect. We use four key metrics to understand a test's performance: sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).

Sensitivity and specificity are characteristics of the test itself. They don't change based on how common the disease is.

• Sensitivity: How well the test identifies people who have the disease. It's the probability of a positive test, given the person is sick. Think of it as the "true positive rate."
• Specificity: How well the test identifies people who do not have the disease. It's the probability of a negative test, given the person is healthy. This is the "true negative rate."

Positive Predictive Value (PPV) and Negative Predictive Value (NPV) are more useful for interpreting a specific patient's result. These values do depend on the prevalence of the disease in the population.

• PPV: If a patient tests positive, what's the probability they actually have the disease?
• NPV: If a patient tests negative, what's the probability they are actually healthy?

Let's check your understanding of these foundational concepts.

These core concepts are the building blocks for understanding clinical research and making evidence-based decisions in pharmacotherapy.