APIC: Rate comparisons must be apple to apple

Use standardized methods that can be validated

The Association for Professionals in Infection Control and Epidemiology’s (APIC) position paper on disclosure of nosocomial infection rates includes key points summarized as follows:

  • The purpose of surveillance in each health care organization is to estimate whether its own observed nosocomial infection experience is better or worse than expected for the patient population being surveyed. This information may be used to improve the care of future patients. Infection surveillance strategies in each health care organization are best determined after assessing the types of patients served. Surveillance can address nosocomial infections or related processes that impact the highest risk patients, occur with high frequency, or may result in the most significant outcomes.
  • For each surveillance strategy to be valid, it must include:
  1. consistent surveillance intensity;
  2. application of standardized definitions of infections;
  3. method(s) to adjust for differences in patient related risk.
  • Using standard denominators, infection rates from specific sites can be calculated (e.g., surgical site infection rate in patients undergoing a designated operation, urinary tract infection rate in patients with indwelling catheters). For many site-specific infections, rates would be expressed as the number of infections (the numerator) divided by the number of device days or at-risk days during the surveillance period (the denominator), and standardized by multiplying by a constant (generally 1,000). Other infection rates should be expressed as a proportion of patients who become infected after exposure to a single event (e.g., a specific surgical procedure).
  • Selections of appropriate denominators, surveillance time periods, and formulas for rate calculations are key to developing valid surveillance systems. To make any comparison of infection surveillance data, the measures themselves must be consistent and standardized. This is true whether comparisons are made within individual facilities over time, or from one facility to another.
  • APIC agrees with the Centers for Disease Control and Prevention (CDC) and the American Hospital Association, which have stated that crude or overall infection rates, not adjusted for risk, are inappropriate for comparison and potentially misleading.
  • When nosocomial infection data must be released for review, the health care organization should ensure that data be accompanied by a detailed description of:
  1. the surveillance methodology (specifically, assurance that consistent methods were used both for finding infections and defining infections);
  2. whether there has been any adjustment for patient risk;
  3. the formula for the rate calculation;
  4. any limitations of data, the surveillance methods, or the mechanisms of comparison.
  • When requesting nosocomial infection data, an outside organization (e.g., payers, accrediting agencies, peer review organizations) should clarify how such data will be interpreted. Specifically, how will the infection experiences of different health care organizations be compared when standardized methods probably were not required or used?
  • APIC supports the concept that comparing patient outcomes and process data, including nosocomial infection experiences, may be an important aspect of quality assessment and performance improvement activities. Comparisons of patient data, including nosocomial infections, only can be accomplished by the development of standardized methods that can be validated. For the successful application of current and evolving surveillance strategies, health care organizations must ensure that expertise in surveillance methodology is available and utilized.1

Reference

1. Association for Professionals in Infection Control and Epidemiology. Position statement on release of nosocomial infection data. APIC News March/April 1998; 17(2).