<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>Medication Supply in Closed Loop Medication - Conceptual Understanding and Prerequisites</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Helge Ovesen</string-name>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Liv Johanne Wekre</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Ingvild Klevan</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>Central Norway Hospital Pharmacy Trust</institution>
        </aff>
        <aff id="aff1">
          <label>1</label>
          <institution>Trondheim Hospital Pharmacy</institution>
        </aff>
      </contrib-group>
      <pub-date>
        <year>2017</year>
      </pub-date>
      <fpage>28</fpage>
      <lpage>37</lpage>
      <abstract>
        <p>Closed loop medication (CLM) is an electronic supported medication management process. In addition to reducing errors, the approach provides better documentation, better traceability and better support for evaluating efficacy. Commonly today, medication information is registered in many standalone systems and manual transfer between these represents a risk of human error. To be able to discuss the CLM concept, a shared understanding is needed. In this paper, a definition of CLM is suggested and the concept is elaborated regarding prerequisites and dependencies. The definition of CLM was discussed in meetings with stakeholders regionally. Sales data from the hospital pharmacy in Trondheim (July-December 2016) were reviewed with respect to possible inclusion in a CLM. Dependencies, e.g. pharmaceutical formulation and suitability regarding repackaging into identifiable unit doses, was considered. The following definition of CLM is suggested: 'CLM is an electronically supported process for medication management where information from ordering to the point of administration is transferred seamlessly between different systems. Documentation during the process is done electronically and traceable. The CLM follows the single medication unit'. At St. Olav's hospital in the city of Trondheim, Norway, the local unit dose system repackages tablets, capsules, suppositories, ampoules, vials and more. The weekly output is about 45 000 units. Approximately 60% of total sales may be identifiable at the point of administration. For tablets and capsules, the number is about 80%. Only 14% of the medications are available as identifiable unit doses from the industry. These numbers strongly indicate that in order to reach the CLM ambitions, systems for repackaging of medications are needed.</p>
      </abstract>
      <kwd-group>
        <kwd>Closed loop medication</kwd>
        <kwd>unit dose medication</kwd>
        <kwd>ICT-assisted medication management</kwd>
        <kwd>patient safety</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>-</title>
      <p>The</p>
      <p>medication (management) cycle is often defined as the processes from
prescribing, via ordering, preparation and administration, to evaluation of (immediate)
effect. In most Norwegian hospitals, the processes in the medication cycle today are
supported by many standalone ICT and paper based systems, which deal with
medication information separately. Manual transfer of information between these
systems represents a risk of errors, hence compromising patient safety. Errors occur in
all the different steps of the cycle; 20% of the reported medication related errors in
Norwegian hospitals in 2014 were associated with the prescribing-phase, while
approximately 80% occurred during dispensing and administration of medications [1].</p>
      <p>The concept of closing the medication loop, as a course of action to reduce errors
during medication management, was brought forward in the late 1990s [2]. Closed
loop medication (CLM) is an electronic supported medication management process,
see Figure 1. This concept aims to ensure; right medication, in the right dose, by the
right route, at the right time and last but not least – to the right patient (the five R’s).
In addition to reducing errors, the approach aims to provide better documentation,
better traceability and a better support for evaluating efficacy. However, to be able to
discuss this concept across different hospital settings, a common basis for the
understanding of CLM is needed. A conceptual understanding is further elaborated in
the following paper, through examples from medication sales statistics from St.
Olav’s Hospital in Trondheim and a definition of CLM is suggested. Prerequisites and
dependencies for the concept are brought forward, in addition to the coherency
between functionalities in the systems and benefits thereof.
1.1</p>
    </sec>
    <sec id="sec-2">
      <title>Prerequisites and dependencies for CLM</title>
      <p>At a minimum, a CLM includes an electronic ICT system for prescribing and
administration of the medication, in addition to electronic identification of the
healthcare professional and the patient at the point of administration. In many cases, the
systems also include electronic system support for dispensing as well as ordering, i.e.
the supply link from hospital pharmacies is integrated into the loop.</p>
      <p>ICT systems for electronic prescribing in, e.g. an electronic medication record
(EMR), ensures consistent and defined medication information. This information is
reused in other ICT systems, along with administrative and clinical information about
the patient. A continuous and comprehensive medical chart solution potentially covers
all structured clinical patient information; hence the entire hospital uses the same data
about the patient when moving between wards and clinics. The ICT systems and user
interfaces are usually adjusted to meet the departments’ unique individual needs due
to information in the work flows.An electronic stock keeping ICT system for
medication supply can be the link between the medication module in EMR and the
production, logistics and accounting systems at the hospital pharmacy. Based on a
predefined primary assortment for the individual wards, the pharmacy re stocks the
medication store at the ward automatically. Scanning of the medication units by the
health-care professionals at both receiving and usage is a prerequisite for good
inventory management. In addition, by introducing digital systems for electronic
ordering, stock status information is presented to the doctor during the ordering phase
of the medication management chain.</p>
      <p>Electronic identifiable medications at the time of administration are another
prerequisite for CLM. This requires that each medication unit is identifiable, i.e. has a
unique ID/barcode. There are two alternative ways to produce identifiable
pharmaceuticals at unit level; unit doses in the original packaging from the industry or
repackaged unit doses from hospital pharmacies or other businesses [3]. ID-marked
unit doses can be supplied in two alternative ways from the hospital pharmacy; either
through deliveries to medication stocks at the ward, or as patient-specific unit doses.
The patient-specific doses are prescribed in the EMR to the pharmacy, labelled with
the patient ID and other relevant information about the medicine before they are
delivered to the ward.</p>
      <p>Accessible and readable barcodes on the medication package assure electronically
identifiable medications. Barcode labelling on the secondary packing of medication is
not a requirement for marketing authorisation in Norway, only a Nordic commodity
number (V.nr). Barcodes on secondary packing are suitable for identification control
in processes where complete packages are included, for example by handling
prescriptions at the pharmacy. Nevertheless, in CLM it is a necessity to have barcodes
at unit level upon administration of the medication. In Europe, more strict regulations
for barcode labelling of medications are now being introduced through the Falsified
Medicine Directive (FMD) [4]. This legislation is primarily introduced to prevent
counterfeiting of medications; hence, the requirement of standardised barcodes will
only apply to the outer packing. Barcodes on primary packages are however
introduced as a national demand through the procurement process of medications
(LIS).</p>
      <p>At the point of administration, verification of the medications should be done with
the patient present. Health-care personnel thus depend on having a mobile medication
store and scanning equipment at hand bedside. Mobile storing solutions/trolleys with
trays allocated for each patient, barcode readers and displays showing the actual
prescription could be used. The scanning equipment or barcode readers must identify
both the medication ID, the patient ID, in addition to the health personnel ID. The
latter could also be solved with identification by login to the EHR-system. This fulfils
the requirement of a record of who administered the medication, what was given and
at what time the administration took place.</p>
      <p>For medication that must be reconstituted before administration (e.g. powdered IV
medication) or dilution of solutions for injection or infusion, to be applicable into a
CLM, there should be a digital system for generating a new label or barcode that can
be used for identification of the ready to use medicine at the point of administration.
This applies to both when such preparations are being prepared at the hospital
pharmacy and at the hospital ward.</p>
      <p>The last prerequisite is a standardised and complete medication registry, which
will ensure a common understanding regarding medication information (including
barcode information) between different ICT systems in CLM. The Norwegian
Medicines Agency (NoMA) is the regulatory authority responsible for medication
information in Norway and provides this through the national Prescribing and
Expedition Support Register (FEST).</p>
    </sec>
    <sec id="sec-3">
      <title>1.2 Expected benefits of CLM</title>
      <p>
        Overall, it is expected that errors occurring during medication management will
reduce and patient safety improve when introducing CLM. Studies commonly show
that errors during the prescribing/ordering phase and the administration phase are
reduced [5–7]. The reduction of errors through patient identification by wrist-bands
with barcodes at the time of administration is in itself shown to be efficient, but the
outcome varies between different settings [
        <xref ref-type="bibr" rid="ref9">8, 9</xref>
        ]. Although several studies indirectly
show improved patient safety, further investigations on context and cost-benefit
analyses should be performed to support decision-makers in the prioritising processes
[
        <xref ref-type="bibr" rid="ref10 ref11 ref12 ref13">10–13</xref>
        ].
      </p>
      <p>Decision support connected to the single medication unit when ordering the
medication could give patient-specific notice on allergies or whether the dose was
adjusted according to blood test results. At the point of administration, the ICT system
can provide guidance/support regarding, e.g. the administration route of the
medication.</p>
    </sec>
    <sec id="sec-4">
      <title>Conceptual choices for medication supply at St. Olav’s Hospital/Trondheim</title>
    </sec>
    <sec id="sec-5">
      <title>Hospital Pharmacy</title>
      <p>In 2004, at St. Olav’s Hospital it was decided to introduce automated medication
supply and, with a time horizon, CLM. Consequently, the hospital pharmacy in
Trondheim started to repackage medication into unit doses in 2012. The stock keeping
and ordering system at St. Olav’s Hospital (Delta) was introduced at the same time as
the unit doses.</p>
      <p>The unit dose packaging system at the hospital pharmacy can repack tablets,
capsules, suppositories, ampoules and vials. In addition to the automated dispensing
and labelling of unit doses, pharmacy personnel manually barcode multi-dose
packages (e.g. mixtures and ointments) and other medications that are not suitable for
repackaging. This manual approach also assures traceability in the supply chain for
these medications. The unit dose facility at the hospital pharmacy in Trondheim has a
production flexibility that allows it to dispense medications for a day’s consumption
for the individual patient at a 750-bed hospital. The actual medications are then
collected on a plastic ring with a patient tag (Figure 2). This functionality is currently
not in production, awaiting introduction of electronic ordination through the
electronic health record (EHR) procurement-project (Helseplattformen). Meanwhile,
the unit doses (without patient labelling) are supplied to local storage at the hospital
wards and allocated into patient-specific trays in medication trolleys.</p>
      <p>Based on our knowledge regarding the current state of CLM, we have identified both
a need for a deeper conceptual understanding and a thorough investigation of the
prerequisites for potentially achieving the goal. Accordingly, the research questions
for this work were:
What is CLM?
To which extent could CLM be achieved based on the current supply model at St.
Olav’s Hospital?
The concept of CLM was discussed during meetings with stakeholders regionally.
The discussions were arranged in conjunction with two regional project meetings,
with staff from the pharmacy and the hospital, as well as technical resources. Relevant
literature, reports over the topic from other health regions and ongoing projects were
studied to give input to the regional discussion [2, 14]. Different views, about which
work processes and system support are included in a CLM, were taken into account.
A common conclusion was in the end summed by agreeing on a definition of CLM.</p>
    </sec>
    <sec id="sec-6">
      <title>Sales statistics from medication supply at St. Olav’s Hospital</title>
      <p>Sales statistics for medications from the hospital pharmacy in Trondheim to all
departments organised under St. Olav’s Hospital Health Trust in the period 1.3.2016–
31.8.2016 were retrieved from the internal pharmacy enterprise resource planning
(ERP)-systems, FarmaPro. The data were categorised and reviewed to give a
complete picture of the types of medications supplied to a large university hospital in
Norway, as well as to highlight the proportion of the total medication delivery that
could potentially be included in the CLM on the basis of the conceptual choice made.
Dependencies for the latter, e.g. pharmaceutical formulation and suitability for
repackaging into identifiable unit doses, were considered based on the types of
medicines already repacked into unit doses from the hospital pharmacy. To categorise
preparations, subheadings were linked to package type in the FEST register. These
package types are then merged into more general groups, e.g. all vials in one category,
and all blister packages in another category (See Table 1).</p>
    </sec>
    <sec id="sec-7">
      <title>Potential for single dose units from the pharmaceutical industry</title>
      <p>An estimate of the proportion of medications that can be delivered as unit doses from
the industry was also made. The available unit dose medicines (tablets and capsules)
were identified with package type ‘ENDOSE’ in FEST for each product. In addition,
medications ready to use as unit doses were considered single dose in this context.
This applies, for example, to Metronidazole Inf 5mg/ml 100ml infusion bottles. These
medications were also identified using known package types from FEST.
3</p>
      <sec id="sec-7-1">
        <title>Results and Discussion</title>
        <p>Based on the discussion with stakeholders regionally, the following definition of
CLM is suggested: ‘CLM is an electronically supported process for medication
management where information from ordering to the point of administration is
transferred seamlessly between different ICT systems. Documentation during the
process is done electronically and traceable. The CLM follows the single medication
unit’.</p>
        <p>This definition is not very different from the other definitions found, but an
important statement is added in the last sentence: ‘CLM follows the single medication
unit’ and not the whole list of medications. This means that one of the patients’
medications may be included in a CLM while others are not. When it comes to
Formulation
knowledge and clinical decision support in CLM, only support for correct
management of the single medication is included in the process-support. This means
that, e.g. interaction support is not included in a CLM.
Insulatard FlexPen®
Prefilled pen
Marcain-adrenalin inj.®</p>
        <p>Blister card is cut into
single units and
repackaged
Converted to bulk before
repackaging
Box (bulk) from
manufacturer
Box (bulk) from
manufacturer with
antibiotics/cytotoxic that
cannot be repacked into
unit dose
1ml and 2 ml
ampoules for
inhalations.</p>
        <p>Multi-dose inhalator
Single dose syringe
Multi-dose syringe
Multi-dose vials that
can be used for 24h</p>
        <p>Single dose ampoule
Morfin NAF 40mg/ml
10ml®
Inflectra® Powder for
infusion</p>
        <p>Multi-dose vials that can be
used for 24h</p>
        <p>Single use vials
Liniment
Ointment and
creams
Oral spray
Nasal spray
Nasal drops
Ear drops
Eye drops
Implants
Bottle
Bottle
Bottle
Bottle
Bottle
Bottle
Minims</p>
        <p>Locoid Crelo®
MDU</p>
        <p>BetnovatChinoform®
MDU
MDU
SDU
MDU
MDU
SDU
WP/SDU</p>
        <p>Nitrolingual®
Instanyl®
Flutide nasal®
Cilox®
Blocadren Depot® Multi-dose eye drops
Oxibuprocain Minims® Single use eye drops
Zoladex® implant Single dose</p>
        <p>Multi-dose liniment that
can be
shared between several
patients
Multi-dose creams that can
be
shared between several
patients
Multi-dose spray
Single dose
Single use nasal drops
Mixtures Bottle MDU Mycostatin® mixt. Multi-dose bottles
Suppositories SDU Voltaren® sup. Single dose
Vagitories SDU Dalacin® vag. Single dose
Skin patches WP Norspan® Depot patch Single dose
*SDU = single dose unit, MDU= multi-dose unit, WP= whole package
In the period from 1.3.2016–31.8.2016 (six months), a total of 2 001 671 single units
were produced at the facility in Trondheim Hospital Pharmacy, corresponding to
about 45 000 single dose units weekly. From this, 61% are potentially electronically
identifiable at the point of administration.</p>
        <p>Regarding tablets and capsules, the figure is 83%. For the same period, we could
identify only 14% of medications available as unit doses from the industry. The
proportion of these unit doses branded with barcodes is unknown.</p>
        <p>The remaining medicines which could not be electronically identifiable at unit
level at the point of administration are: ointments, creams, syrups and tablets/capsules
distributed in boxes. For this group of substances, additional labelling is a possible
solution to increase the availability of barcode branded medicines.</p>
        <p>Based on the analysis of sale statistics, we realise that CLM may be difficult or not
possible to establish for all groups of medications. This supports our definition of
CLM that it follows the single medication unit.</p>
        <p>To summarise, the above suggested definition opens for a qualitative approach to
the area and the following is made possible:
- Overview of the share of total medications delivered as identifiable and hence
eligible for CLM
- Comparison of CLM status between units and hospitals
- A more thorough understanding of the link between pharmacy dispensing
andCLM
- Measurable goals regarding CLM for a unit or hospital</p>
      </sec>
      <sec id="sec-7-2">
        <title>Summary and Conclusion</title>
        <p>Only a low proportion of medications were available as identifiable unit doses from
the pharmaceutical industry at the time of the study and with the defined
prerequisites; this is based on medications sold as unit doses with the correct package
type indicated in FEST. In real life, the number might be higher but not anywhere
close to the 80–90% that can be delivered based on preparation of unit doses in the
hospital pharmacy. These numbers strongly indicate that in order to reach the CLM
ambitions, systems for repackaging of medications are needed. Further investigation
regarding medication error status, cost-effectiveness and an optimised ICT-supported
work-flow should be done in the present context as a basis for future decision support.</p>
        <p>References
1.
2.
3.</p>
      </sec>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          <string-name>
            <surname>E</surname>
            ,
            <given-names>S.</given-names>
          </string-name>
          and
          <string-name>
            <surname>F. Ø</surname>
          </string-name>
          ,
          <article-title>Årsrapport 2014 for meldeordningen for uønskede hendelser i spesialisthelsetjenesten</article-title>
          .
          <year>2015</year>
          , Kunnskapssenteret.
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          <string-name>
            <surname>Organisations</surname>
          </string-name>
          .
          <year>1998</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          <article-title>Forskrift om tilvirkning av legemidler i apotek</article-title>
          , H.-o. Omsorgsdepertementet, Editor.
          <year>2001</year>
          : www.lovdata.no.
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          <string-name>
            <surname>DIRECTIVE</surname>
          </string-name>
          <year>2011</year>
          /
          <article-title>62/EU OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL</article-title>
          , E.
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          <string-name>
            <surname>Barber</surname>
            , N.,
            <given-names>T.</given-names>
          </string-name>
          <string-name>
            <surname>Cornford</surname>
            , and
            <given-names>E.</given-names>
          </string-name>
          <string-name>
            <surname>Klecun</surname>
          </string-name>
          ,
          <article-title>Qualitative evaluation of an electronic prescribing and administration system</article-title>
          .
          <source>Quality and Safety in Health Care</source>
          ,
          <year>2007</year>
          .
          <volume>16</volume>
          (
          <issue>4</issue>
          ): pp.
          <fpage>271</fpage>
          -
          <lpage>278</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          <string-name>
            <surname>Quality</surname>
          </string-name>
          and Safety in Health Care,
          <year>2007</year>
          .
          <volume>16</volume>
          (
          <issue>4</issue>
          ): pp.
          <fpage>279</fpage>
          -
          <lpage>284</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          <string-name>
            <surname>Williams</surname>
            ,
            <given-names>C.T.</given-names>
          </string-name>
          ,
          <article-title>Inside a closed-loop medication strategy</article-title>
          .
          <source>Nursing Management</source>
          ,
          <year>2004</year>
          . 35: pp.
          <fpage>8</fpage>
          -
          <lpage>9</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          <string-name>
            <surname>Khammarnia</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <given-names>A.</given-names>
            <surname>Kassani</surname>
          </string-name>
          , and
          <string-name>
            <given-names>M.</given-names>
            <surname>Eslahi</surname>
          </string-name>
          ,
          <article-title>The efficacy of patients' wristband bar-code on prevention of medical errors: a meta-analysis study</article-title>
          .
          <source>Appl Clin Inform</source>
          ,
          <year>2015</year>
          .
          <volume>6</volume>
          (
          <issue>4</issue>
          ): pp.
          <fpage>716</fpage>
          -
          <lpage>27</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          9.
          <string-name>
            <surname>Hassink</surname>
            ,
            <given-names>J.J.M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>M.M.P.M. Jansen</surname>
            , and
            <given-names>P.J.</given-names>
          </string-name>
          <string-name>
            <surname>Helmons</surname>
          </string-name>
          ,
          <article-title>Effects of bar code-assisted medication administration (BCMA) on frequency, type and severity of medication administration errors: a review of the literature</article-title>
          .
          <source>European Journal of Hospital Pharmacy: Science and Practice</source>
          ,
          <year>2012</year>
          .
          <volume>19</volume>
          (
          <issue>5</issue>
          ): pp.
          <fpage>489</fpage>
          -
          <lpage>494</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref10">
        <mixed-citation>
          10.
          <string-name>
            <surname>Risør</surname>
            ,
            <given-names>B.W.</given-names>
          </string-name>
          ,
          <string-name>
            <given-names>M.</given-names>
            <surname>Lisby</surname>
          </string-name>
          , and
          <string-name>
            <given-names>J.</given-names>
            <surname>Sørensen</surname>
          </string-name>
          ,
          <article-title>An automated medication system reduces errors in the medication administration process: results from a Danish hospital study</article-title>
          .
          <source>European Journal of Hospital Pharmacy</source>
          ,
          <year>2016</year>
          .
          <volume>23</volume>
          (
          <issue>4</issue>
          ): pp.
          <fpage>189</fpage>
          -
          <lpage>196</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref11">
        <mixed-citation>
          11.
          <string-name>
            <surname>Agrawal</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ,
          <article-title>Medication errors: prevention using information technology systems</article-title>
          .
          <source>British Journal of Clinical Pharmacology</source>
          ,
          <year>2009</year>
          .
          <volume>67</volume>
          (
          <issue>6</issue>
          ): pp.
          <fpage>681</fpage>
          -
          <lpage>686</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref12">
        <mixed-citation>
          12.
          <string-name>
            <surname>Nuckols</surname>
            ,
            <given-names>T.K.</given-names>
          </string-name>
          , et al.,
          <article-title>The effectiveness of computerised order entry at reducing preventable adverse drug events and medication errors in hospital settings: a systematic review and meta-analysis</article-title>
          .
          <source>Systematic Reviews</source>
          ,
          <year>2014</year>
          .
          <volume>3</volume>
          (
          <issue>1</issue>
          ): p.
          <fpage>56</fpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref13">
        <mixed-citation>
          13.
          <string-name>
            <surname>Ehteshami</surname>
            <given-names>A</given-names>
          </string-name>
          ,
          <string-name>
            <given-names>R.P.</given-names>
            ,
            <surname>Tavakoli</surname>
          </string-name>
          <string-name>
            <given-names>N</given-names>
            ,
            <surname>Kasaei</surname>
          </string-name>
          <string-name>
            <surname>M.</surname>
          </string-name>
          ,
          <article-title>The role of health information technology in reducing preventable medical errors and improving patient safety</article-title>
          .
          <source>Int J Health Syst Disaster Manage</source>
          <year>2013</year>
          (
          <volume>1</volume>
          ): pp.
          <fpage>195</fpage>
          -
          <lpage>199</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref14">
        <mixed-citation>
          <article-title>Veilder om legemiddelhåndtering</article-title>
          . Helse
          <string-name>
            <surname>Sør-Øst</surname>
            <given-names>RHF</given-names>
          </string-name>
          : https://sykehusapotekene.no/Documents/Veileder%20Legemiddelh%
          <article-title>C3%A5ndtering%20Hel se%20S%C3%B8r%C3%98st%202015.pdf</article-title>
          .
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>