Maths for Clinicians 8

• What is a case-control study?

  Compares exposure level in group with specific disease (cases) & in another group without the disease (controls)

  Observational study

  Uses self-report questionnaires, interviews & records
• Where can the control group participants be found?

  • Hospital (with other diseases)
  • Community-based controls (preferable as likely to be healthier)

  E.g, electoral register

  Controls similar to case patients (e.g, from same geographical area) but don't have the outcome (disease)
• What are the weaknesses of case-control studies?

  • Measurement & selection bias (recall & interviewer)
  • No causal relationship
  • Inefficient for rare exposures
  • Confounding variables
  • Can't calculate incidence rates
• What are the advantages of case-control studies?

  • Quick (no follow-up)
  • Efficient for rare outcomes
  • Can examine more than 1 exposure for 1 outcome
  • No loss to follow-up bias
• What are regression models?

  • Assess strength of association between exposure & outcome
  • Multiple regression does this after adjustment for confounders
• What are the main types of regression models?

  • Linear- continuous outcome- Regression coefficient (summary statistic)
  • Logistic- Binary outcome- Odds ratio (summary statistic)
• How do you calculate odds ratio (OR)?

  Odds in exposed group/Odds in unexposed group

  Value interpreted in same way as RR

  Approximates RR when outcome is rare