Use of Computers in Data Collection, Data Analysis, and Epi Info

1. Use of Computers in Data Collection

A. Electronic Data Entry

• Computer-Assisted Personal Interview (CAPI): Interviewers enter responses directly into a laptop/tablet during face-to-face interviews, reducing transcription errors.
• Computer-Assisted Self-Interview (CASI): Respondents enter their own answers - useful for sensitive topics (e.g., sexual behavior, substance use).
• Computer-Assisted Telephone Interview (CATI): Used in large telephone surveys; software routes questions based on previous answers (skip logic).

B. Electronic Data Collection Tools

C. Advantages of Computer-Based Data Collection

• Eliminates transcription errors (data entered once)
• Skip patterns / validation rules enforced in real-time
• Immediate data availability for analysis
• GPS tagging for spatial data
• Reduced paper costs and storage needs
• Facilitates remote/cloud-based data submission

D. Surveillance and Routine Data

• Disease surveillance systems (e.g., IDSP in India, CDC's NNDSS in the US) use computer networks to aggregate case reports from peripheral health facilities.
• Vital registration systems (births, deaths) are increasingly computerized.

2. Use of Computers in Data Analysis

A. Types of Analysis Supported

B. Common Software Packages

C. Data Management Tasks Computers Handle

• Data cleaning (identifying outliers, missing values, duplicates)
• Data coding and recoding
• Database merging and linking
• Report generation and visualization (tables, graphs, charts)

3. Epi Info - In Detail

What is Epi Info?

Important update: The CDC officially discontinued product development and technical support for Epi Info in September 2025. The desktop version still runs on Windows 11 for existing users, but no new updates will be issued. (CDC Epi Info page)

Key Modules of Epi Info 7

1. Form Designer (MakeView)

• Used to create electronic questionnaires/data entry forms
• Supports text fields, checkboxes, drop-down lists, coded fields
• Can enforce skip logic, range checks, and required fields
• Supports multi-page and relational questionnaire designs

2. Enter (Data Entry Module)

• Allows data entry into forms created in Form Designer
• Built-in validation prevents out-of-range entries
• Data stored in a project database (.mdb or .prj format)

3. Classic Analysis / Visual Dashboard

• Core statistical analysis module
• Produces frequency tables, cross-tabulations, means, proportions
• Generates 2x2 tables with:
  • Odds Ratios (OR) with 95% CI
  • Risk Ratios (RR)
  • Risk Differences
  • Chi-square, Fisher's Exact tests
• Logistic regression (conditional and unconditional)
• Survival analysis (Kaplan-Meier, Cox proportional hazard)
• Complex survey data analysis
• Can read and write to Excel, CSV, dBase formats

4. StatCalc

• Stand-alone calculator for:
  • Sample size calculations (cohort, case-control, cross-sectional)
  • Chi-square test from 2x2 tables
  • Population-based surveys (margin of error)
  • Person-time incidence rates

5. Epi Map

• Links data to maps using shapefiles
• Creates choropleth maps to display geographic distribution of disease
• Useful for spatial epidemiology

6. Epi Report

• Generates formatted reports combining tables, graphs, and maps
• Can publish output as HTML or Word documents

Uses of Epi Info in Epidemiology

1. Outbreak investigation - Rapid creation of line-listing forms; attack rate calculations; epidemic curves
2. Cross-sectional surveys - Community health surveys with weighted analysis
3. Case-control and cohort studies - OR, RR, attributable risk computation
4. Disease surveillance systems - Small to mid-sized surveillance system development
5. Field epidemiology training - Standard tool in FETP (Field Epidemiology Training Programs) worldwide
6. Meta-analysis - Some versions supported pooled analysis
7. Clinical research - Data entry and analysis for small clinical studies

Advantages of Epi Info

• Free of cost (public domain - no license fee)
• Easy to learn - suitable for non-statisticians
• Integrates data entry + analysis in one package
• Available in 13 languages
• WHO and CDC endorsed
• Runs on Windows, with web/cloud and Android versions developed
• Produces publication-ready tables and graphs

Limitations of Epi Info

• Not suitable for very large datasets (limited scalability)
• Discontinued as of September 2025 (no future updates)
• Less flexible than R or STATA for complex analyses
• Limited support for advanced statistical modeling
• Replaced in many settings by R, REDCap, and Kobo Toolbox

Workflow in Epi Info (Step-by-Step)

1. Design form (Form Designer)
         ↓
2. Enter/import data (Enter module or import from Excel)
         ↓
3. Clean data (check frequencies, identify outliers)
         ↓
4. Analyze data (Analysis module - FREQ, TABLES, MEANS, REGRESS)
         ↓
5. Create maps (Epi Map)
         ↓
6. Generate reports (Epi Report / export to Word/HTML)

Key Commands in Epi Info Analysis Module

Summary Table