Validation of a Generic Measure of Continuity of Care: When Patients Encounter Several Clinicians

We would like to compliment Haggerty et al. on their paper. Continuity of care is particularly important for patients with multimorbidity. This makes a strong plea for a generic instrument that measures continuity of care regardless of morbidity and across multiple care settings. The authors have made an important contribution to filling this knowledge gap.

As only few generic questionnaires exist, all with their own pros and cons, it is a pity that the authors did not refer to our previously developed generic measure, the Nijmegen Continuity Questionnaire (NCQ): we developed this questionnaire in 2011 [1] and tested it more recently on its psychometric properties [2]. The NCQ has already shown to be a reliable and valid generic questionnaire that measures continuity of care as a multidimensional concept (personal and team/cross-boundary continuity) from the patient's perspective across multiple care settings.

We identified interesting similarities between these generic questionnaires. Both instruments measure continuity of care from the perspective of the patient. Moreover, items from previous questionnaires on continuity of care inspire both instruments. And lastly, items on the communication and cooperation between care providers seem to be comparable between both questionnaires.

But besides similarities, we also identified differences. The instrument of Haggerty et al. mainly focuses on management continuity, which seems to be comparable with team/cross-boundary continuity in the NCQ. The items of their instrument are therefore based on studies of patient experiences when care is received from various care providers over time. The NCQ also measures another important aspect of continuity of care, namely having a personal care provider in every care setting with whom the patient can develop an ongoing relationship (personal/relational continuity). This dimension has shown to be important both for patients who contacted various care providers and for patients who only contacted one care provider.

Another difference between the two measures is the definition of continuity of care and its related discussion which items are relevant for measuring continuity of care. The instrument of Haggerty et al. includes items about the information received from care providers in terms of helpfulness for staying healthy and coping with minor complications. These items were not considered relevant for the NCQ, as another definition of continuity of care was used.

We found in our study that it was extremely difficult for patients to have an opinion about their care team, such as 'how well do you feel known by everyone on the care team'. A lot of patients showed difficulties with answering these kind of items, especially when they felt completely known by one care provider, but hardly by another. Therefore, we decided not to include items about the care team in our final version.

Overall, we believe that both instruments are useful in measuring continuity of care, especially for patients with multimorbidity and across multiple care settings. Which instrument to use should depend on the dimensions of continuity of care that researchers want to measure and their definition of continuity of care.

[1] Uijen AA, Schellevis FG, van den Bosch WJ, Mokkink HG, van Weel C, Schers HJ. Nijmegen Continuity Questionnaire: development and testing of a questionnaire that measures continuity of care. J Clin Epidemiol 2011; 64(12):1391-1399

[2] Uijen AA, Schers HJ, Schellevis FG, Mokkink HGA, Weel van C, Bosch van den WJHM. Measuring continuity of care: psychometric properties of the Nijmegen Continuity Questionnaire. Br J Gen Pract 2012; 62(600): 949-957

Competing interests:   None declared