Towards the development of a personalised E-Health intervention for use in community pharmacies to analyse and improve medication non-adherence in people with type 2 diabetes mellitus, who are non-adherent to oral blood glucose and/or blood pressure lowering drugs.

Medication adherence is suboptimal in 10-56% of all people with type 2 diabetes
mellitus (T2DM). This non-adherence is associated with higher HbA1c levels,
blood pressure and cholesterol levels. Moreover, non-adherence can lead to an
increased number of hospitalisations, mortality rates and health care costs.
Multiple interventions have been developed to enhance medication adherence.
However, very few studies demonstrated an improvement of treatment outcomes and
even the most efficacious interventions only achieved modest effect sizes. An
explanation for these moderate effects can be that interventions are not
tailored to the needs and preferences of individual patients. Therefore, the
aim of this study is to develop and test a personalised intervention to improve
medication adherence in people with T2DM, who are non-adherent to oral blood
glucose and/or blood pressure lowering drugs.

The main objective of this study is to develop and investigate the (cost-)
effectiveness of a pharmacist-led personalised intervention, with E-Health
components, to analyse and improve medication adherence in people with T2DM,
who are non-adherent to oral blood glucose and/or blood pressure lowering drugs
compared to an already available general T2DM information platform. In
addition, the study has three secondary objectives:

1. To identify non-adherence profiles based on individual factors, barriers,
needs and preferences related to medication adherence of people with T2DM who
are non-adherent to oral blood glucose and/or blood pressure lowering drugs.

2. To perform a process analysis of the personalised intervention in order to
assess the extent to which the intervention has been performed according to the
study protocol and to gain insight into the barriers and facilitators of the
intervention program.

3. To develop and refine non-adherence algorithms after evaluation of the
personalised intervention in order to further improve the intervention for
future use.

A parallel-group randomised controlled trial is conducted in 40-50 (community)
pharmacies and adjoining practises in the Netherlands and the United Kingdom
(UK) (30-40 the Netherlands / 10 in the UK). A total of 300 participants will
be included (150 the Netherlands / 150 the UK) and the follow-up period of the
trial will be six months.

The trial consists of an intervention and control condition. The intervention
condition is a personalised intervention, which consists of four supporting
programs. Participants will receive supporting program(s) based on a predefined
non-adherence profile. The four supporting programs/non-adherence profiles are:
(I) knowledge and perceptions, (II) practical problems, (III) side effects, and
(IV) negative mood and beliefs. All of the supporting programs consist of one
or more interventions that are theoretically grounded or have previously been
shown to improve medication non-adherence. These interventions include: brief
messaging, medication schedule, reminding messaging, medication review,
medication dispensing systems, smart messaging, referral to a general
practitioner (GP) and a Self-guided Self Help application. The supporting
programs will be tailored to a participants* specific situation, needs and
preferences. Participants that are assigned to the control condition will have
access to an already available general T2DM information platform via their
smartphone/tablet/computer.

Overall, for our participants we do not foresee any specific risks for possible
damage and/or potential harm. Moreover, we do not include vulnerable
participants and therefore conclude that the conduct of the research involves a
negligible risk to human participants and is justified. We summarise the burden
and risk per study procedure. First, participants vhave a pharmacyappointment
at baseline (approximately 30 minutes per visit). In a subgroup of the pharmacy
consultations researchers will be present and audio recording will be made for
the proces evaluation of the study. Participants can indicate whether they do
or do not want this.

Second, at baseline general practice or specialist registry data will be
consulted to obtain a participants' blood pressure value and HbA1c level (with
a time window of three months before and one month after the time point). Also
at 6 months registry data will be consulted to obtain blood pressure and HbA1c
level (with a time window of two months before and two months after the time
point). By keeping the blood pressure and HbA1c measurements in line with
routine care (regular diabetes check-ups), the burden and risks for
participants are reduced.

Third, participants are asked to fill in questionnaires at baseline, and after
three and six months (30-45 minutes per fill-in moment and 1,5-2,25 hours in
total). These questionnaires are validated instruments that are often used in
health research. Participants in the intervention group will also receive a
telephone call whether they are satisfied with the interventions, after one
month (10 minutes). Furthermore, a double telephone pillcount will be conducted
at baseline and after 6 months in all participants (20-30 minunted per
measurement). The risk of the personalised intervention is minimal, since the
supporting programs are all theoretically grounded or have previously been
shown to improve medication non-adherence and no main risks have been
identified in these studies.