• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

Animal & Veterinary

  • Print
  • Share
  • E-mail

Effect on Resistance Following Changes in Antimicrobial Use

SLIDE 1: Effect on Resistance Following Changes in Antimicrobial Use

Marc Lipsitch, D.Phil.
Harvard School of Public Health


SLIDE 2: Rapid Response to Intervention in a Hospital

A graph


SLIDE 3: Controlling Erythromycin Resistance in Group A Streptococci, Finland

A graph


SLIDE 4: Transparency


SLIDE 5: Factors influencing rates of change in resistance after change in antimicrobial use

  • Fitness “cost” of resistance genes or mutations
    Influx of susceptible bacteria into the system
    Continuing selection by other antimicrobials
    Persistent reservoir of resistant strains


SLIDE 6: Implications for Thresholds

  • Resistance may change slowly following withdrawal
  • Therefore threshold should be well below serious health threat
    • at a level that can be maintained for some years
  • Other measures (besides withdrawal of offending agent) may be necessary to make resistance decline
    • Removal of reservoirs
    • Changes in other antimicrobial practices


SLIDE 8: Human health threshold (T(x))

  • Appropriate for foodborne human pathogens (Campylobacter, Salmonella)
  • Foodborne impact can be estimated using attributable risk for resistant cases
  • Prevalence of resistance plays confirmatory role
  • Any serious human illness unresponsive to standard treatment would be evidence of harm


SLIDE 9: Human health threshold (T(x)): 2

  • May serve as sentinel for harder-to-quantify effects in commensals
  • Relationship to microbiologic threshold is probably not linear, but this does not matter


SLIDE 10: Microbiologic (Meat) Threshold (T(x))

  • Appropriate for commensals where direct disease-food link unlikely to be provable in any specific infection (E coli, Enterococcus)
  • Goal should be to avoid adding to resistance of opportunistic pathogens to important therapies
  • Suggested level: prevalence of resistance exceeding that in normal human flora


SLIDE 11: Limitations of Thresholds

  • Reactive: Cannot undo creation of new resistant organisms / gene complexes
  • Withdrawal of drug may be slow to affect resistance levels


SLIDE 12: Incorporating New Information

  • Need continual monitoring of resistance levels in animal, human bacterial isolates
  • For pathogens, need periodic measurements of risk factors for infection (and resistant infection)
  • Monitoring of cross-resistance important as threshold is neared
  • Relevant new information includes changes in:
    • Human treatment recommendations
    • Eating habits or food preparation
    • Microbiological safety practices


SLIDE 13: Summary

  • Threshold concept is a sound way to address most problems of foodborne resistance
  • Responds only to identifiable risks
    • At the expense of possibly failing to prevent difficult-to-identify risks
  • Inability to precisely specify relationship of t(x) to T(x) should not impede threshold setting
  • Different threshold concepts for
    • Pathogens: Human health threshold
    • Commensals (opportunistic pathogens): Microbiologic threshold