Process evaluation of physical activity counselling with and without the use of mobile technology: a mixed methods study

Background: A monitoring-and-feedback tool was developed to stimulate physical activity by giving feedback on physical activity performance to patients and practice nurses. The tool consists of an activity monitor (accelerometer), wirelessly connected to a Smartphone and a web application. Use of this tool is combined with a behaviour change counselling protocol (the Self-management Support Programme) based on the Five A's model (Assess-Advise-Agree-Assist-Arrange).

Objectives: To examine the reach, implementation and satisfaction with the counselling protocol and the tool.

Design: A process evaluation was conducted in two intervention groups of a three-armed cluster randomised controlled trial, in which the counselling protocol was evaluated with (group 1, n = 65) and without (group 2, n = 66) the use of the tool using a mixed methods design.

Settings: Sixteen family practices in the South of the Netherlands. Participants: Practice nurses (n = 20) and their associated physically inactive patients (n = 131), diagnosed with Chronic Obstructive Pulmonary Disease or Type 2 Diabetes, aged between 40 and 70 years old, and having access to a computer with an Internet connection.

Methods: Semi structured interviews about the receipt of the intervention were conducted with the nurses and log files were kept regarding the consultations. After the intervention, questionnaires were presented to patients and nurses regarding compliance to and satisfaction with the interventions. Functioning and use of the tool were also evaluated by system and helpdesk logging.

Results: Eighty-six percent of patients (group 1: n = 57 and group 2: n = 56) and 90% of nurses (group 1: n = 10 and group 2: n = 9) responded to the questionnaires. The execution of the Self-management Support Programme was adequate; in 83% (group 1: n = 52, group 2: n = 57) of the patients, the number and planning of the consultations were carried out as intended. Eighty-eight percent (n = 50) of the patients in group 1 used the tool until the end of the intervention period. Technical problems occurred in 58% (n = 33). Participants from group 1 were significantly more positive: patients: chi(2)(2, N = 113) = 11.17, p = 0.004, and