Advanced Biostatistics: Basic Concepts

• Course: Advanced Biostatistics & Research Methodology
• Resource Person: Ammara Arshad
• Date: 2 March, 2025
• For: MS Nutrition Science, Spring 2025, Department of Nutrition and Dietetics
• Lecture Topic: Introduction to Advanced Biostatistics
• Topic: Basic Concepts

Learning Objectives

• Understand fundamental statistical concepts in nutrition research
• Identify different types of data and their applications in health and nutrition sciences
• Explain probability distributions and their role in dietary and epidemiological studies

What is Biostatistics?

• Biostatistics Definition: application of statistical principles to biological, medical, and nutritional research
• Nutrition Importance: analyzes diet-disease relationships, supports evidence-based dietary recommendations, and aids in public health policy-making regarding nutrition interventions
• Examples in Nutrition: evaluating the effectiveness of dietary interventions in reducing obesity and analyzing nutrient intake patterns in different populations

Fundamental Statistical Concepts

• Population: entire group of interest, for example, all diabetic patients in a country
• Sample: subset of the population used for study, for example, 500 diabetic patients from a city
• Parameter: numerical summary of a population
• Statistic: numerical summary of a sample
• Descriptive: describing data we obtained from the chosen sample is known as descriptive statistics, for example, mean calorie intake
• Inferential: drawing conclusions about a population is known as inferential statistics, for example, predicting obesity risk based on diet

Descriptive Statistics vs. Inferential Statistics

Descriptive Statistics

• Goal: to summarize data obtained from a sample
• Focus: variables of a sample
• Presentation: graphics and charts
• Measures: central tendency (mean, median, mode) and variability (range, quartiles, variance, standard deviation, skewness)

Inferential Statistics

• Goal: to make generalizations to a larger population based on the sample data
• Focus: parameters of the population
• Presentation: probabilities, confidence intervals, and hypotheses
• Measures: hypothesis tests like Chi-square, ANOVA, t-tests; confidence intervals; regression analysis

Variable Types:

• Independent Variable: dietary intake
• Dependent Variable: BMI changes
• Confounding Variable: physical activity level

Types of Data in Nutrition Research

• Qualitative Data (Categorical):
• Nominal: No meaningful order, for example, blood type, dietary preferences like vegetarian vs. non-vegetarian
• Ordinal: Meaningful order but no exact difference, for example, food insecurity levels: mild, moderate, severe
• Quantitative Data (Numerical):
• Discrete: Countable numbers, for example, number of meals per day
• Continuous: Measured values, for example, blood glucose levels, daily caloric intake

Data Measurement Scales in Nutrition

• Nominal: classification without ranking (e.g., type of diet: vegan, keto, Mediterranean)
• Ordinal: ordered categories (e.g., severity of malnutrition: mild, moderate, severe)
• Interval: numeric values with no true zero (e.g., temperature in Celsius affecting food storage conditions)
• Ratio: numeric values with a true zero (e.g., weight, daily protein intake)

Probability Distributions in Nutrition Research

• Probability Definition: determining likelihood of health outcomes (e.g., probability of vitamin D deficiency in a population)

Common Distributions

• Normal Distribution: used for variables like BMI, blood pressure
• Binomial Distribution: used for categorical outcomes like presence/absence of obesity
• Poisson Distribution: used for rare events like severe malnutrition cases in a population
• Application: helps in designing dietary interventions and predicting disease risks.

Summary of Key Concepts

• Biostatistics is crucial in nutrition research for analyzing diet-health relationships
• Different types of data require appropriate statistical methods
• Probability distributions help in making dietary and epidemiological inferences

• Lecture Topic: Research Design & Methodology
• Introductory Lecture – Basic Concepts

Learning Objectives

• Differentiate between types of research study designs in nutrition
• Understand experimental vs. observational studies
• Explore systematic reviews and meta-analysis in dietary research

Introduction to Research Design

• Research design is a structured plan for collecting and analyzing data to answer scientific questions
• Importance: it helps in drawing accurate conclusions in dietary and health studies
• Key components: hypothesis, study population, sampling, data collection, and statistical analysis

Types of Research Designs

Observational Studies

• Cross-sectional: snapshot of a population's diet and health
• Case-control: retrospective analysis of diet-disease relationships
• Cohort: long-term dietary pattern tracking

Experimental Studies

• Randomized Controlled Trials (RCTs): assess effects of dietary interventions
• Clinical Trials: testing new nutritional therapies

Cross-Sectional Studies

• Assess dietary patterns and health outcomes at one time
• National surveys on sugar consumption and obesity rates are an example
• Advantage: cost-effective and quick
• Limitation: cannot establish causation

Case-Control Studies

• Compare those with a disease to those without, retrospectively.
• Comparing trans fat intake in heart disease patients vs. healthy individuals is an example
• Strength: efficient for studying rare diseases
• Limitation: recall bias, selection bias.

Cohort Studies

• Follow participants over time to assess risk factors
• Following individuals with high salt intake to study hypertension development is an example
• Strength: establishes temporal relationships
• Limitation: expensive, requires long-term follow-up

Randomized Controlled Trials (RCTs)

• Gold standard for assessing diet interventions
• Testing effects of intermittent fasting on metabolic health example
• Strength: controls confounding factors
• Limitation: expensive, ethical concerns

Systematic Reviews & Meta-Analysis

• Systematic Review: summarizes multiple studies on a topic
• Meta-Analysis: statistically combines the results of multiple studies
• Examining the effect of omega-3 fatty acids on heart health is an example

Summary of Key Concepts

• Research design choice affects study outcomes
• Observational studies help identify associations; experimental studies determine causality
• Systematic reviews provide strong evidence for dietary guidelines