6.3 Research Study Methodology, Literature Evaluation & Biostatistics

Evidence-based infection prevention relies on critical scientific evaluation and statistical literacy. The Infection Preventionist (IP) must analyze published research, assess methodological rigor, evaluate statistical significance versus clinical relevance, and apply biostatistical formulas to diagnostic testing and surveillance data.

The Hierarchy of Scientific Evidence

Scientific study designs possess varying degrees of protection against bias and confounding. The pyramid of evidence, ordered from highest to lowest strength of causality, guides clinical decision-making:

Core Biostatistical Concepts & Hypothesis Testing

Interpreting research literature requires a rigorous understanding of biostatistical hypothesis testing:

Hypotheses and Errors

• Null Hypothesis (H0): The baseline assumption that no true statistical difference, association, or effect exists between comparison groups.
• Alternative Hypothesis (Ha): The claim that a true difference or effect exists between groups.
• Type I Error (α / False Positive): Rejecting the null hypothesis when it is actually true (concluding a difference exists when there is none). The standard alpha threshold is set at α = 0.05.
• Type II Error (β / False Negative): Failing to reject the null hypothesis when it is actually false (failing to detect a true clinical difference).
• Statistical Power (1 - β): The probability of correctly rejecting a false null hypothesis (detecting a true difference if one exists). Standard power is targeted at 80% (0.80) or 90%, influenced by sample size, effect size, and significance level.

p-Values and Confidence Intervals

• p-Value: The probability of obtaining a test statistic at least as extreme as the observed result, assuming the null hypothesis is true. A p-value <0.05 indicates statistical significance, but does not measure the magnitude or clinical importance of the effect.
• 95% Confidence Interval (CI): A range of values constructed around a point estimate that has a 95% probability of containing the true population parameter.
  • For ratio measures (Relative Risk or Odds Ratio), if the 95% CI includes 1.0 (e.g., RR = 0.85, 95% CI: 0.65-1.12), the finding is not statistically significant at the α = 0.05 level.
  • For difference measures (Mean Difference), if the 95% CI includes 0.0, the finding is not statistically significant.
  • A narrower confidence interval indicates greater statistical precision, usually driven by a larger study sample size.

Diagnostic Testing Metrics & Calculations

IPs evaluate diagnostic and screening assays using four foundational biostatistical performance metrics:

Sensitivity and Specificity (Intrinsic Test Properties)

• Sensitivity: The proportion of individuals with the disease who test positive: $\text{Sensitivity} = \frac{\text{True Positives (TP)}}{\text{True Positives (TP)} + \text{False Negatives (FN)}}$ High sensitivity minimizes false negatives. A highly sensitive test is used for screening (SNOUT: Sensitive test Negative rules OUT).
• Specificity: The proportion of individuals without the disease who test negative: $\text{Specificity} = \frac{\text{True Negatives (TN)}}{\text{True Negatives (TN)} + \text{False Positives (FP)}}$ High specificity minimizes false positives. A highly specific test is used for confirmation (SPIN: Specific test Positive rules IN).

Predictive Values (Prevalence-Dependent Metrics)

• Positive Predictive Value (PPV): The probability that a patient with a positive test result actually has the disease: $\text{PPV} = \frac{\text{True Positives (TP)}}{\text{True Positives (TP)} + \text{False Positives (FP)}}$ PPV increases directly as disease prevalence increases in the population being tested.
• Negative Predictive Value (NPV): The probability that a patient with a negative test result actually is free of the disease: $\text{NPV} = \frac{\text{True Negatives (TN)}}{\text{True Negatives (TN)} + \text{False Negatives (FN)}}$ NPV decreases as disease prevalence increases.

Critical Appraisal of Infection Control Literature

When reviewing published studies to update institutional protocols, the IP conducts a critical appraisal evaluating:

1. Internal Validity: The degree to which the study design, execution, and analysis minimize bias and confounding within the sample.
   • Selection Bias: Non-random sampling or systematic differences between comparison groups.
   • Measurement / Information Bias: Misclassification of exposures or outcomes (e.g., unblinded assessors evaluating surgical site infections).
   • Confounding: A distortion caused by a third variable associated with both the exposure and outcome (controlled via restriction, matching, stratification, or multivariable logistic regression).
2. External Validity (Generalizability): The degree to which study findings apply to the IP’s specific healthcare setting, patient demographics, and operational workflows.