A multicentre study on the effects of self-administration of medication during hospitalization on medication safety, adherence, and patient satisfaction in Dutch hospitals.

In 2016 66.2% of the total Dutch population had one or more medicines provided
that were eligible for reimbursement by health insurance (1). Overall, the use
of prescription medication is increasing in the Netherlands (2). Moreover, the
retail of self-medication, e.g. pain medication, seems to be growing as well
(3). During hospitalization patients receive their medication from the hospital
pharmacy, including self-medication and vitamins. In general, this implies that
31% of home medication is substituted to hospital formulary medication at
admission (4). Only 57% of this substituted medication is re-initiated at
discharge, increasing the risk of medication errors, for example due to
duplication of medication (4). As part of standard care nurses are responsible
for medication administration in hospitalized patients. Within healthcare
patient-centered care is increasingly advocated. A concept of patient-centered
care is patient empowerment. In this concept care is not given according to the
old paternalistic model, e.g. medication administration by nurses, but there is
room for the patient*s autonomy and preferences. This may be achieved by
inpatient self-administration of medication (SAM) during hospital admission.
The concept of inpatient self-administration involves the use of the medication
boxes from home, dispensing newly started medication packaged in boxes as would
be dispensed in ambulatory care (instead of unit dose dispensing) and patient
education and coaching in using the medication the right way.

Medication administration errors (MAEs) occur daily in health care and can lead
to serious harm (5). A systematic review of 65 studies from 13 countries
reported a median error rate of 52 errors per 100 admissions (6). Improvement
of medication safety is a major concern to policymakers and health care
workers. SAM could be a way to reduce MAEs, due to the recognizability of
medication, no need for medication substitution to the hospitals preferences,
and coaching patients to take their medication properly. Furthermore, SAM may
lead to better adherence after hospital admission. Short after hospital
discharge, over 60% of patients have insufficient knowledge about their current
medication use due to medication changes during hospitalization, such as:
discontinuation of medication and dose changes (7). In addition, 55% of elderly
patients do not use their medication correct two to four weeks after hospital
discharge. Three months later non-adherence occurred in 70% of these patients
(8, 9). The misuse of medication, in turn, increases the risk of hospital
readmission and death (8, 9). Lastly, SAM gives patients empowerment which
could positively influence patients satisfaction during hospitalization.

A recent systematic review looked into the efficacy of implementing SAM during
hospitalization on outcomes related to patients, staff and institutions (10).
Furthermore, medicine knowledge, adherence and patient satisfaction were
studied after SAM was implemented, as well as the success of implementation of
SAM and satisfaction of the staff (10). Staff and institution outcomes were
defined as the success of implementation of SAM and satisfaction of the staff.
The review included 43 studies, none of which was conducted in the Netherlands.
The review showed a positive effect on knowledge of the medication name and
regimen by patients, but not on knowledge of side effects (10). Whether SAM
increases medication safety, was not studied. The effect of SAM on patient
adherence was inconclusive. But in general the study methodology was inadequate
for accurate measurement of adherence, e.g. counting pills or questionnaires
(11). Therefore more research is needed. The review showed that participation
in SAM schemes leads to increased satisfaction of patient and staff (10).
Patients advised that SAM should be the golden standard when hospitalized. The
success rate of SAM varied from 26 to 86% (10). This argues for implementing
SAM in Dutch healthcare.

Given the paucity of evidence as is illustrated by the review, we aim to study
the effects of SAM on medication safety, adherence, patient satisfaction, and
staff satisfaction.

The main objective of this study is:
* To determine the effect of inpatient self-administration of medication on the
number of medication administration errors during hospitalization
*
The secondary objectives of this study are:
* To determine the effect of inpatient self-administration of medication on the
severity of medication administration errors during hospitalization.
* To determine the effect of inpatient self-administration on medication
adherence after hospitalization
* To determine the effect of inpatient self-administration of medication on
patient satisfaction during hospitalization
* To determine the effect of inpatient self-administration of medication on
staff satisfaction during hospitalization

This study is a multicentre prospective quasi-experimental study with a
pre-post design.

During SAM medication is stocked at the patient*s bedside. The stock comprises
the medication the patient should receive based on the information documented
in the patient*s file. Patients are provided with information about the
medication that is documented in their files. This information could either be
shown on an electronic device or on paper, depending on what is standardly used
in the hospital. When medication is scheduled to be administered, patients
collect those form their own stock, administer, and document the administration
by themselves. Nurses check at every administration round whether patients
succeeded in the administration.
Each day, patients are qualified for SAM. For this, a checklist is used and
filed in the patient*s dossier. In the case, patients do not meet the criteria
of SAM, they will be excluded from SAM. In those situations, nurses will take
over the administration process and patients will receive standard care as
described previously.

The burden associated for patients participating with this study is limited
(only 13 minutes time investment for filling in questionnaires). Furthermoren,
during the post periode patient have to participate in SAM schemes. The burden
of SAM is classified as low because results of a recent questionnaire shows
that admitted patients have the urge to act in SAM schemes.The risk of SAM
during hospitalization is estimated as the risk patients are at home when using
medication. This risk may different from the risk a patient have during usual
care, when nurses are responsible for medication administration. If this risk
is equal, higher or lower is the subject of this study. We expect a better
medication safety for participating patients