Two Branches of Statistics
Statistics used in health information management fall broadly into two categories: descriptive and inferential. Understanding the distinction, and knowing when each applies, is fundamental for the RHIA exam analytics domain.
Descriptive Statistics
Descriptive statistics summarize and organize data without drawing conclusions beyond the dataset itself. They answer the question, what does the data show.
- Measures of central tendency, including mean, median, and mode.
- Measures of dispersion, including range, variance, and standard deviation.
- Frequency distributions, which show how often each value occurs.
- Percentages and rates, such as infection rate per one thousand patient days.
Example of Descriptive Statistics
Calculating the average length of stay for all patients discharged from a hospital last month is a descriptive statistic. It describes the sample without making claims about future patients or other hospitals.
Inferential Statistics
Inferential statistics use a sample of data to draw conclusions or make predictions about a larger population. They answer the question, what can we conclude beyond the data we have.
- Hypothesis testing, which determines whether an observed difference is statistically significant.
- Confidence intervals, which provide a range of plausible values for a population parameter.
- Regression analysis, which models relationships between variables to predict outcomes.
- Chi-square tests, which examine relationships between categorical variables.
Example of Inferential Statistics
Using a sample of patients from several hospitals to conclude that a new discharge protocol reduces readmission rates across the broader patient population is an example of inferential statistics, since the conclusion extends beyond the sample studied.
Why the Distinction Matters
Confusing descriptive and inferential statistics can lead to overstated conclusions. A hospital might correctly describe that its readmission rate decreased last quarter, a descriptive fact, but would need inferential statistics with appropriate hypothesis testing to claim the change was due to a specific intervention rather than random variation.
Application in Quality and Research
Quality improvement projects often begin with descriptive statistics to understand current performance, then move to inferential statistics when testing whether an intervention caused a measurable, statistically significant change. Research studies rely heavily on inferential methods to generalize findings to broader populations.
RHIA Exam Preparation
Review common descriptive and inferential techniques and practice identifying which category a given statistical scenario belongs to, since exam questions often present a research or quality scenario and ask you to classify the statistical method described.