Study Notes on Biostatistics and Epidemiology

Statistics in a Nutshell

Introduction to Biostatistics and Epidemiology

  • Presenter: Fernando Cánovas

  • Association: Bachelor’s Degree in Dentistry, UCAM (Universidad Católica de Murcia)

  • Fields Covered: Human Health, Animal Health, Environmental Health

Key Concepts in Statistics

  • Biostatistics: The branch of statistics that deals with data analysis in biological sciences and health.

  • Epidemiology: The study of how diseases affect the health and illness of populations.

  • Public Health: The practice of protecting and improving the health of people in a community.

  • One Health: A collaborative approach to understanding health issues that links human, animal, and environmental health.

Statistical Inference

  • Population: The entire group that you want to draw conclusions about.

  • Sample: A subset of the population used to estimate characteristics of the whole group.

  • Bar Inference: Drawing conclusions about a population from a bar chart representation.

  • Dichotomous Outcomes: Outcomes that fall into one of two categories (e.g., yes/no).

  • Biostatistical Variables: Variables measured in biostatistics that can yield different types of data (categorical, continuous, etc.).

  • Endpoints: Outcomes or results of a study that are used to measure the effectiveness of a treatment or intervention.

Data Representation

  • Histograms: A graphical representation of the distribution of numerical data.

  • Prevalence Question: An inquiry that seeks to determine how common a particular condition is within a population.

  • Categorical Data: Data that can be sorted into categories, such as gender or race.

  • Continuous Data: Data that can take any value within a range, such as blood pressure readings.

Relationships Between Variables

  • Correlation: The statistical technique used to measure and analyze the strength and direction of the relationship between two variables (denoted as Y and X).

    • Process to Study Correlation:

    1. Distribute Y across each subgroup of levels of X.

    2. Summarize each subgroup using appropriate statistics.

    3. Compare the distributions using statistical methods.

    4. If Y values are similar across groups of X, they are considered independent; otherwise, they are dependent.

  • P-value: The probability of observing a sample as extreme as the one observed, or more so, assuming that the null hypothesis (H0) is true.

Statistical Tests

Classification of Statistical Tests

  • Parametric Tests

    • Assumptions about the underlying parameters of the population from which samples are drawn.

    • Typical requirements include normal distribution and equal-interval scales.

  • Non-parametric Tests

    • Do not assume a specific population distribution.

    • Useful when data do not meet parametric test assumptions.

Assumptions of Statistical Tests

  • Dependent Variable: The variable being tested or measured (e.g., health outcomes).

  • Independent Variable: The grouping variable (e.g., sex, handedness).

  • Homogeneity of Variances: Assumption that different samples come from populations with equal variances (often tested using Levene’s test).

  • Normality: Assumption that the dependent variable is approximately normally distributed, often assessed via the Shapiro-Wilk test.

Example Studies

Example: Blood Pressure and Hypertension

  • Analysis of blood pressure (BP) levels in relation to hypertension (HT) categorized by sex and laterality (left-handed/right-handed).

  • Data Summary: BP readings by sex:

    • Women:

    • HT Status: Yes/No Distribution

    • Men:

    • HT Status: Yes/No Distribution

    • Proportions of hypertension differed significantly by sex, demonstrating a dependency on gender.

  • Graphical representations (charts/histograms) illustrate distributions and help visualize these relationships.

Proportionality in Study

  • The study aimed to compare hypertension rates across different demographics and assess their correlation with continuous health metrics.

Conclusion

  • The importance of understanding statistical relationships and their implications in health and medicine is essential for researchers and practitioners.

  • Knowledge of statistical methods aids in better decision-making through evidence-based practice and study design.