How we measure continuity

Continuity has three realms, applying to multiple interactions over time:

Information - maintaining consistent patient records.

Management - a patient pathway is overseen within a managerial structure.

Relational - between an individual patient and the same clinician.

We are concerned with relational continuity, as it is this which has shown over numerous studies to have the most beneficial impact on patient and doctor satisfaction, improved outcomes and reduced resource use.

The most common statistical measure is UPC or “Usual Provider Continuity”.  The way to understand this is to consider a patient who has 10 consultations over a period of time. If 6 of them are with the same clinician, UPC is 6/10 or 60%.

A general practice has several doctors,  thousands of patients and consultations.  By a series of algorithms, we calculate the UPC for all patient/doctor/consultations in a time period, and combine to produce an overall number, achieved/potential % continuity. Single consultations are excluded.  Values are calculated separately for face to face and telephone consultations, though for research purposes they could be combined.  Records which are not true consultations, eg medication reviews, are excluded.

What is the right time period to use?

  • Longer might be better, to allow for multiple visits
  • Shorter gives more rapid feedback, especially important at times of change

We normally measure within a month, as a trade off between these two.

The figure is slightly higher than that produced over a longer period, but it is a good proxy for longer term continuity and by using the same measure consistently over time and between practices, we can make valid comparisons.  For research purposes, different time periods can be used.

“It’s very hard to improve if you don’t know how you are doing.”

We aim to make it easier to improve.  Continuity is one part of our Navigator suite.

Note:  we also use subjective measures of continuity, by surveys of clinicians (“Was continuity important in this consultation, and if so was it achieved?”) and surveys of patients (“Did you see your usual doctor?”).  These may be useful, but have drawbacks:

  • Expectations are subjective which makes comparisons unreliable
  • Collection takes time and cost, so tends to be done intermittently
  • Sampling can introduce bias in the results

More evidence, suggestions and resources:  the RCGP Continuity of Care Toolkit

An alternative means of calculating UPC is offered from Bristol Continuity Audit.