p.523 Clinical indicators assess particular health structures, processes, and outcomes. They can be rate- or mean-based, providing a quantitative basis for quality improvement, or sentinel, identifying incidents of care that trigger further investigation. They can assess aspects of the structure, process, or outcome of health care. Furthermore, indicators can be generic measures that are relevant for most patients or disease-specific, expressing the quality of care for patients with specific diagnoses.

p.524 Indicators have been defined in several different ways:

1. As measures that assess a particular health care process or outcome.
2. As quantitative measures that can be used to monitor and evaluate the quality of important governance, management, clinical, and support functions that affect patient outcomes.
3. As measurement tools, screens, or flags that are used as guides to monitor, evaluate, and improve the quality of patient care, clinical support services, and organizational function that affect patient outcomes.

p.524 An ideal indicator would have the following key characteristics: (i) indicator is based on agreed definitions, and described exhaustively and exclusively; (ii) indicator is highly or optimally specific and sensitive, i.e. it detects few false positives and false negatives; (iii) indicator is valid and reliable; (iv) indicator discriminates well; (v) indicator relates to clearly identifiable events for the user (e.g. if meant for clinical providers, it is relevant to clinical practice); (vi) indicator permits useful comparisons; and (vii) indicator is evidence-based. Each indicator must be defined in detail, with explicit data specifications in order to be specific and sensitive.

p.524-525 A rate-based indicator uses data about events that are expected to occur with some frequency. These can be expressed as proportions or rates (proportions within a given time period), ratios, or mean values for a sample population. To permit comparisons among providers or trends over time, proportion- or rate-based indicators need both a numerator and a denominator specifying the population at risk for an event and the period of time over which the event may take place.
 A sentinel indicator identifies individual events or phenomena that are intrinsically undesirable, and always trigger further analysis and investigation. Each incident would trigger an investigation. Sentinel events represent the extreme of poor performance and they are generally used for risk management.