Two Types of Process Measures

“Process measures are measures of whether an activity has been accomplished.  For example, process measures could be whether inventory checks were made or whether patients received evidence-based interventions.” The Improvement Guide, 2nd edition (2009), p. 96.

In my post on Process Measures (here), I gave an example of a diabetes care process:

There are two types of process measures the improvement team can propose, related to the activities of the care process.

1.      An adherence or fidelity measure.  Example:  for patients flagged as eligible for outreach, what fraction of the time does the care team follow the agreed-upon process?

2.      Patient-process interaction measure.  Examples:  (a) of patients with A1c >= 9% and with an appointment more than 30 days away, what fraction of patients do we reach to schedule?  (b) of patients reached for scheduling, what fraction of patients actually show up for the appointment?

Since a health care system plausibly controls adherence to internal work flows, a goal for a type 1 measure should be 100%. 

On the other hand, a type 2 measure depends on activities of both patients and the care process.   For example, “scheduled patients who show up for appointment” depends on staff following the process steps for scheduling the patient and the patients actually showing up in the clinic. 

In the absence of performance data from the process, it’s hard to know a priori what an achievable goal should be for a type 2 measure.  A goal of 100% will not necessarily make sense for this type of measure, at least in the near term.

The authors of The Improvement Guide usefully included an example of each type:  inventory check appears to be an activity under the control of an organization (type 1); patients receiving evidence-based interventions looks like a type 2 measure.

Implication for Management

In the diabetes care process, the proposed outcome measure is per cent of patients with A1c <= 9%.  If you can’t say whether staff follow the care process reliably, you have a poor foundation to improve the outcome.    You can’t distinguish the impact of poor fidelity to the care process from inadequacy of the process as factors that drive performance.   Adherence to the agreed-to care process eliminates a source of variation and simplifies detective work aimed at improvement. 

If you already use daily huddles to review the previous day's work and to look ahead to today's work, a type 1 process measure provides a natural and relevant measure for your daily Plan-Do-Study-Act cycles.