• Skip to secondary menu
  • Skip to main content
  • Skip to primary sidebar
  • My Store
  • Glossary
  • Home
  • About Me
  • Contact Me

Statistics By Jim

Making statistics intuitive

  • Graphs
  • Basics
  • Hypothesis Testing
  • Regression
  • ANOVA
  • Probability
  • Time Series
  • Fun

Retrospective Study: Definition & Examples

By Jim Frost 1 Comment

What is a Retrospective Study?

A retrospective study looks back in time and assesses events that have already occurred. The researchers already know the outcome for each subject when the project starts. Instead of recording data going forward as events happen, these studies use participant recollection and data that were previously recorded for reasons not relating to the project. These studies typically don’t follow patients into the future.

In retrospective designs, the researchers collect their data using existing records. Consequently, they can complete their assessment more quickly and inexpensively than prospective designs that must follow subjects over time and record the data under carefully controlled conditions. However, the data that a retrospective study uses might not have been measured consistently or accurately because they weren’t explicitly designed to be part of a study.

Image of a doctor performing a retrospective study.Researchers using a retrospective design must work with data that others measured for non-study reasons. Those data were not selected and assessed with the project’s requirements in mind. Additionally, some studies might ask the subjects to remember information and use other subjective assessments, introducing various biases.

The statistical analysis for a retrospective study is frequently the same as for prospective designs (looking forward). The main difference is that the project occurs after the outcomes are known rather than how researchers analyze the data.

Statisticians consider retrospective designs to be inferior to prospective methods because they tend to introduce more bias and confounding. Retrospective studies are observational studies by necessity because they assess past events and it is impossible to perform a randomized, controlled experiment with them. However, they can be quicker and cheaper to complete, making them a good choice for preliminary research.  Findings from a retrospective study can help inform a prospective experimental design. Learn more about Experimental Designs.

Retrospective Study Designs

Retrospective studies use various designs. While these designs differ in detail, they all tend to compare subjects with and without a condition and determine how they differ. Using the usual hypothesis tests, researchers can determine whether there are statistically significant relationships between subject variables (risk factors, personal characteristics, etc.) and the outcome of interest.

Cohort and case-control studies are standard retrospective designs. Let’s learn more about them!

Retrospective Cohort Study

This study design compares groups of subjects who are similar overall but differ in a particular characteristic, such as exposure to a risk factor. Because it is a retrospective study, the researchers find individuals where the outcomes are known when the project starts. Retrospective cohort studies frequently determine whether exposure to risk and protective factors affects an outcome.

In these projects, researchers use databases and medical records to identify patients and gather information about them. They can also ask subjects to recall their exposure over time. Then the researchers analyze the data to determine whether the risk factor correlates with the outcome of interest.

Suppose researchers hypothesize that exposure to a chemical increases skin cancer and conduct a retrospective cohort study. In that case, they can form a cohort based on a group commonly exposed to that chemical (e.g., a particular job). Then they access medical databases and records to collect their data. After identifying their subjects and obtaining the medical information, they can immediately analyze the data, comparing the outcomes for those with and without exposure.

Learn more about Cohort Studies.

Case-Control Studies

Case-control designs are generally retrospective studies. Like their cohort counterparts, case-control studies compare two groups of people, those with and without a condition. These designs both assess risk and protective factors.

Retrospective cohort and case-control studies are similar but generally have differing goals. Cohort designs typically assess known risk factors and how they affect outcomes at different times. Case-control studies evaluate a particular incident, and it is an exploratory design to identify potential risk factors.

For example, a case-control assessment might evaluate an episode of severe illness occurring after a company picnic to identify potential food culprits.

Learn more about Case-Control Studies.

Advantages of a Retrospective Study

A retrospective study tends to have the following advantages compared to a prospective design:

Cheaper: You don’t need a lab or equipment to measure information. Others did that for you!

Faster: The events have already occurred in a retrospective study—no need to wait for them to happen and then look for the differences between the groups.

Great for rare diseases: You can specifically look through a database for individuals with a rare disease or condition. In a prospective experiment, you need an immense sample size and hope enough of the rare outcomes occur for you to analyze.

Disadvantages of a Retrospective Study

Unfortunately, they tend to have the following disadvantages relating to a greater propensity for inaccuracies, inconsistencies, lack of controlled conditions, and bias:

  • A retrospective study uses data measured for other purposes.
  • Different people, procedures, and equipment might have recorded the data, leading to inconsistencies.
  • Measurements might have occurred under differing conditions.
  • Control variables might not be measured, leading to confounding.
  • Recall bias.

Reference

Dean R Hess, Retrospective Studies and Chart Reviews, Respiratory Care, October 2004, 49 (10) 1171-1174.

Share this:

  • Tweet

Related

Filed Under: Basics Tagged With: experimental design

Reader Interactions

Comments

  1. Joe Lombardi says

    November 7, 2022 at 8:26 am

    Coincidentally, I just read this Israeli retrospective cohort study regarding the incidence of myocarditis and pericarditis in unvaxxed post-COVID-19 patients: https://pubmed.ncbi.nlm.nih.gov/35456309/

    Good news for a change.

    Reply

Comments and Questions Cancel reply

Primary Sidebar

Meet Jim

I’ll help you intuitively understand statistics by focusing on concepts and using plain English so you can concentrate on understanding your results.

Read More...

Buy My Introduction to Statistics Book!

Cover of my Introduction to Statistics: An Intuitive Guide ebook.

Buy My Hypothesis Testing Book!

Cover image of my Hypothesis Testing: An Intuitive Guide ebook.

Buy My Regression Book!

Cover for my ebook, Regression Analysis: An Intuitive Guide for Using and Interpreting Linear Models.

Subscribe by Email

Enter your email address to receive notifications of new posts by email.

    I won't send you spam. Unsubscribe at any time.

    Follow Me

    • FacebookFacebook
    • RSS FeedRSS Feed
    • TwitterTwitter

    Top Posts

    • How to Interpret P-values and Coefficients in Regression Analysis
    • How To Interpret R-squared in Regression Analysis
    • Multicollinearity in Regression Analysis: Problems, Detection, and Solutions
    • Mean, Median, and Mode: Measures of Central Tendency
    • How to Find the P value: Process and Calculations
    • How to do t-Tests in Excel
    • Z-table
    • One-Tailed and Two-Tailed Hypothesis Tests Explained
    • Choosing the Correct Type of Regression Analysis
    • How to Interpret the F-test of Overall Significance in Regression Analysis

    Recent Posts

    • Slope Intercept Form of Linear Equations: A Guide
    • Population vs Sample: Uses and Examples
    • How to Calculate a Percentage
    • Control Chart: Uses, Example, and Types
    • Monte Carlo Simulation: Make Better Decisions
    • Principal Component Analysis Guide & Example

    Recent Comments

    • Jim Frost on Monte Carlo Simulation: Make Better Decisions
    • Gilberto on Monte Carlo Simulation: Make Better Decisions
    • Sultan Mahmood on Linear Regression Equation Explained
    • Sanjay Kumar P on What is the Mean and How to Find It: Definition & Formula
    • Dave on Control Variables: Definition, Uses & Examples

    Copyright © 2023 · Jim Frost · Privacy Policy