Selection of medicines in Chilean public hospitals: an exploratory study.

2.5 MB·English

by  Collao, Juan F

Collaoetal.BMCHealthServicesResearch2013,13:10
http://www.biomedcentral.com/1472-6963/13/10
RESEARCH ARTICLE Open Access
Selection of medicines in Chilean public hospitals:
an exploratory study
Juan F Collao1,2*, Felicity Smith2 and Nick Barber2
Abstract
Background: There is a growing interest in high income countriesto control expenditure on medicines by
improving the rationale for theirselection.However, in middle income countries with differing priorities and needs,
littleattention has been paid to this issue. In this paper we explore the policies and processes for theselection and
useof medicines in a group of hospitals inChile,a middle income country which has recently joined the OECD.
Methods: A combination ofqualitative and quantitative methods was used. A national survey questionnaire was
distributed to investigatetherole and operationof PTCs (Pharmacy and Therapeutics Committees). Interviews were
conducted with key actors in theselection of medicines in large urban public hospitals.
Results: The national survey had an overall response rate of42% (83 out of 196), whilst 7 out of14 hospitals
participated in thequalitative study. High complexity hospitals are large urban hospitals; allof which claim to have
a working PTC.The pharmacy offices are mainly involved indispensing medicines with littleinvolvementin
clinical duties.
The interviews conducted suggest that the formulary ofall the hospitals visited is no more than a stock list. PTCs
are unable to influence the prescribing practices ofdoctors.Members do not feel prepared to challenge the
opinions of specialists requestinga certain drug, and decisions are based primarily on costs. The inclusion of
medicines inthe clinical practice ofhospitals is as a result of doctors bypassing thePTC and requestingthe
purchase ofexceptionalitems,some of which are included intheformulary if they are widely used.
Conclusions: There is an urgent need to develop medicine policiesin hospitals inChile. The procedures used to
purchase medicines need to be revised. Central guidance for PTCscould help ensure a more rational useof
medicines. PTCs need to be empowered to design formularieswhich cover all the clinical needs of doctors, training
members in theanalysis ofscientificevidence beyond their own specialities. An influential PTC can take the
appropriate measuresand design workablepolicies to enforce a cost effective-useof resources.
Background Drug and Therapeutics Committees (DTC) or Pharmacy
Expenditure on medicines is one of the largest costs in andTherapeuticsCommittees(PTC).
healthcare. Consequently, countries worldwide are in- At the hospital level, the presence of these commit-
creasingly trying to control the use of resources spent tees has been described in many high income countries,
on medicines. The World Health Organization (WHO) where DTCs are considered to improve the rationality of
recommends the use of formulary systems to improve decisions [2-17]. Furthermore, the role of pharmacists
rationality in the use of drugs at regional, district and and the presence of pharmacy services in hospitals have
local levels [1]. These formulary systems should be de- been associated with a more rational use of medicines
signedandupdatedbymultidisciplinarycommitteescalled [18,19].
Given the differing nature of health problems faced by
rich and poor countries, there are key differences in the
objectives of hospital DTCs when selecting medicines.
*Correspondence:[email protected]
1DepartamentodeCienciasFarmacéuticas,FacultaddeFarmacia,Universidad On the one hand, decision-makers in low income coun-
deValparaíso,GranBretaña1093,Valparaíso2360102,Chile tries need to focus on selecting medicines to save as
2DepartmentofPracticeandPolicy,UCLSchoolofPharmacy,Tavistock many lives as possible, in which case the WHO list of
Square,London,WC1H9JP,UK
©2013Collaoetal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative
CommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,and
reproductioninanymedium,providedtheoriginalworkisproperlycited.
Collaoetal.BMCHealthServicesResearch2013,13:10 Page2of11
http://www.biomedcentral.com/1472-6963/13/10
essential medicines is extremely useful. On the other programme called GES (meaning Explicit Guarantees in
hand, committees of high income countries place em- Health).Inthisprogramme,bothprivateandpublicinsur-
phasis not only on saving lives, but also on improving ance systems are required to fully cover all costs in the
the quality of life for their chronically ill patients and treatmentofadefinedlistof69diseases(atnocosttothe
promoting cost-effectiveness in prescribing, especially patient), such as different types of cancer, HIV, diabetes
regarding the use of newer drugs not listed as essential and cystic fibrosis. The treatment of these diseases must
bytheWHO. be in accordance with clinical guidelines designed by a
The objectives of hospital DTCs in middle income panelofexperts.
countriesareprobablysomewherebetweenthoseof high
and low income countries, with health needs requiring a
PharmaceuticalpoliciesinChile
migration from a live-saving essential pharmacotherapy,
For a medicine to be commercialized in Chile, the manu-
to one that not only saves lives, but also improves the
facturer needs to register their product with the National
quality of life of their populations. Thus the way that
Institute for Health (ISP). The ISP is not a specialised
DTCs work in middle income countries could be dir-
medicines agency, but the agency in charge of overseeing
ectly related to the extent to which the population can
the safety of every product that is consumed by humans
access medicines that are not on the WHO’s essential
and animals. At the time of the study, generic medicines
list.
were only required to present clinical studies performed
SuchisthecaseinChile,anuppermiddleincomecoun-
for the original drug. Currently the ISP isimplementinga
try according to the World Bank, which joined the Or-
National Agency for Medicines (ANAMED) and generic
ganization for Economic Cooperation and Development
drugswillberequiredtoprovebioequivalenceinorderto
(OECD) in2010. Membership of the OECD suggests that
begrantedmarketauthorizationinChile.
the country will engage in the process of developing poli-
Chile has a long tradition of pharmaceutical policies.
ciestoimprovethequalityoflifeofitspopulation.Never-
In fact, the first National Formulary of Medicines was
theless, information about Chilean hospitals is virtually
developed in Chile between 1965 and 1970, and later
non-existentandthereforea studyofdecisionsaboutme-
adopted by the WHO in 1977 [21]. Unfortunately, bet-
dicines in Chile is a good opportunity to improve our
ween 1973 and 2004 no pharmaceutical policies were in
understandingofhowdecisionsaremadeinhospitalsina
force, with the commercialization and use of medicines
middle income country engaged in the process of migra-
regulated almost exclusively by market laws. Currently,
tionmentionedabove.
the National Formulary of Medicines is simply a list of
medicines that have been granted market authorization
TheChileanHealthSystem
in Chile and does not include any treatment guidelines
Since 1980, three health systems have coexisted in Chile;
forthose drugs.
namely, a public system serving around 70% of the po-
The National Medicines Policy published in 2004
pulation, a private system serving approximately 27%,
(i.e. 4 years before the study) establishes that all hos-
and the health system for the armed forces for around
pitals, regardless of their complexity, must have a Phar-
3% of Chilean citizens [20]. The most important provi-
macy and Therapeutics Committee in charge of decisions
der of secondary care is the public network of hospitals
aboutmedicines.ThePTCshouldbechairedbythemed-
which covers most towns and cities in Chile; while large
icaldirector,thesecretaryofthe committeeshouldbethe
highly complex private hospitals are only found in the
headofthePharmacyDepartment,andallclinicaldepart-
biggest cities. Public hospitals are classified by the Min-
mentheadsshouldbepermanentmembers.
istry of Health as high and low complexity hospitals.
Thisstudyexploresthepoliciesandprocesses involved
There are no official criteria for classifying hospitals as
in the use of medicines in Chilean public hospitals and
high or low complexity according to defined parameters.
aims to explore the extent to which decisions concern-
Instead, the classification seems to be based on how the
ing the formulary of medicines are rational, and how
network operates: low complexity hospitals aim to im-
these decisions could be improved. Moreover, the study
prove geographical coverage, while high complexity hos-
may also help decision makers to develop policies and
pitals are regional or reference centres to which low
procedures for the selection and use of medicines in
complexity hospitals refer patients who require a more
othermiddleincome countries.
complexlevelofcare.
Thehealthreform Researchquestions
A health reform was in progress at the time of the
research, which aimed to ensure coverage for a list of 1.Whatare the characteristics ofChileanPublic
catastrophic diseases by progressively implementing a hospitalsinterms ofthenumberof beds,theduties
Collaoetal.BMCHealthServicesResearch2013,13:10 Page3of11
http://www.biomedcentral.com/1472-6963/13/10
ofthepharmacyoffice,the formulary systems,and Data processing and analysis The responses to the
W
theexistenceofmultidisciplinary committees? questionnaire were entered and analysed using SPSS 18
2.Whatare the centralandlocalpoliciesand softwareanddescriptiveprocedures.
proceduresforthe selection ofmedicinesinChilean
public hospitals? Ethics statement The survey requested information
3.Whatare theformaland informalwaysbywhichthe about the hospital and did not include any questions
useofadrugnotpreviously usediscommenced in which could involve patient data. The survey was then
Chilean publichospitals? submitted to the Chilean Ministry of Health for ethics
approvalwhichwasdeemedunnecessary.
Methods
Qualitativestudy
Due to the virtual absence of published information
Overview
about Chilean public hospitals, a combination of quanti-
The results of the questionnaire suggest that high com-
tative and qualitative methods was used to address these
plexity hospitals are more likely than low complexity
questions. A two-stage study was performed, comprising
hospitalstoinnovateintheirpharmacotherapy,andhence
a survey followed by semi-structured interviews with
be more likely to deal with new medicines. Therefore, we
purposively selected informants. By using these comple-
conductedaqualitativestudyinhighcomplexityhospitals
mentary methods it was possible to gather structured
inorder toanswerresearchquestions 2 and 3.FromMay
data concerning the aspects of Chilean hospitals that are
2008toAugust2008,interviewswerecarriedoutwiththe
described in the literature as factors that influence the
chief of the Department of Pharmaceutical Policy at the
innovativeness of institutions [8,10,22,23], and it was
Ministry of Health, members of PTCs, doctors who had
also possible to explore relevant issues and processes in
applied for new medicines to be added to the formulary,
context.
nurses, and clinicians working in the hospital (Additional
file1).
Nationwidesurvey
Overview Sampling strategy, procedure and recruitment Dif-
Apostalsurveywasdesignedinordertoidentify charac- ferent sampling strategies were used for the selection of
teristics of Chilean Public hospitals. The questionnaire cities, hospitals and interviewees inside hospitals. Firstly,
requested information on hospital size, complexity, and for practical purposes only three cities could be selected
theexistenceofaPTC[8,10,22-28](Additional file1). for the study. The cities were: Santiago, the capital city
located in the geographical centre of the country with
approximately 6 million inhabitants; Valparaíso located
Sampling strategy and procedure The survey was a on the coast 150 km from Santiago with a population of
population study, with the questionnaire sent to all 196 nearly one million; and Coquimbo, 450 km north of
hospitalsthatappearontheMinistryofHealth’swebsite. Santiago with less than half a million inhabitants. Thus,
these urban hospitals represent different regions of the
country where specialist doctors tend to be concentra-
Developmentofinstruments Thefirstdraftoftheques-
ted. Secondly, fourteen hospitals in these cities were se-
tionnaire was piloted with five hospital pharmacists, se-
lected using purposive sampling aiming to include at
lected using a snowballing technique following initial
least one of the different types of high complexity hospi-
contact with a hospital pharmacist previously known to
tals present in those selected cities (i.e. child, cancer,
a member of the research team. After corrections were
psychiatric, and general hospitals). Thirdly, a snowbal-
made, the second draft was piloted and discussed with a
ling strategy was employed to select hospital staff, start-
further snowball sample of hospital pharmacists until no
ing with the chief pharmacists who then identified other
newissuesemerged.
members of the PTC. Doctors who had applied for a
drug to be admitted to the formulary were identified
Data collection In January 2008, the questionnaire with from the minutes of PTC meetings. The last two doctors
an explanatory letter from the research team, a letter who had applied for a drug were contacted through the
from the Ministry of Health encouraging response, and chiefpharmacist.
pre-paid envelopes to return the questionnaire, were sent Clinicians were identified by convenience sampling, in
to all hospitals addressed to “The person in charge of the which the researcher approached the first doctor found
pharmacy office”. After two months, a second question- after walking into the wards and asking them to parti-
nairewassenttonon-respondentswithapersonalisedlet- cipate in the project (all wards of the hospital were
terencouragingresponsetothequestionnaire. included). Interviews with doctors were continued until
Collaoetal.BMCHealthServicesResearch2013,13:10 Page4of11
http://www.biomedcentral.com/1472-6963/13/10
saturationofcodeswasreached(i.e.nonewissuesemer- the different codes and then re-classified into sub-codes
ging), or in the cases of two hospitals (where recruiting untilnofurtherissuesemerged.
doctors was difficult) until the planned visiting time for
thehospitalwasover. Ethics statement The interview schedules did not
contain any questions involving individual patient’s data.
Nevertheless, we submitted the questions to the Ethics
Development of instruments Specific interview sche-
Committee of one of the participant hospitals; this com-
dules were designed for pharmacists, members of the
mittee established that no ethics approval was necessary
PTC, drug applicants, clinicians and nurses. The inter-
forthestudy.
view schedules included open questions about national
and local policies regarding the use of medicines, formal
Results
and informal procedures for using medicines not inclu-
Nationwidesurvey
ded in the formulary, the purchase of medicines, and
Characteristicsofsamplesandresponserates
abouthowthepharmacyofficeoverseestheuseofmedi-
83 hospitals responded to our questionnaire, represent-
cinesinthehospital.
ing 48% of all high complexity hospitals and 40% of low
complexity hospitals (42% overall). Geographically, the
Data collection The interviews were carried out bet-
survey obtained responses from 14 out of the 15 admin-
ween April and July 2008 by a Chilean pharmacist and
istrative regions of the country, which represents a good
researcher with no previous working experience in hos-
picture of the different realities of public hospitals in
pital pharmacy in Chile, hence with no prejudgement
Chile.
about the possible answers to the interviews. The inter-
viewer, trained in relevant research methods, maintained
Chileanpublichospitals
the principles of the qualitative enquiry while perfor-
Respondent hospitals ranged from basic undifferentiated
ming the interviews, limiting himself to encouraging
rural facilities with as few as six beds, to urban hospitals
interviewees to give more information about topics that
with a maximum of 720 beds. All the high complexity
emerged during interviews. Interviews were audio recor-
hospitals are located in urban areas (i.e. more than
ded with the consent of the interviewee; when permis-
10,000 inhabitants), and 54% of the low complexity hos-
sionwasnot granted, written notes weretaken.
pitals are in urban areas. Table 1 summarizes the find-
Minutes of PTC’s meetings held during the year 2007
ings of the national survey regarding the characteristics
were reviewed to identify medicines that had been in-
ofChileanhospitals.
cluded inthe formulary duringthatyear.
Thehospitalpharmacies
Dataprocessingandanalysis Recordedinterviewswere A pharmacist is in charge of the pharmacy office in all
transcribed verbatim and then analysed in their original but one high complexity hospital. Conversely, only 48%
W
language using the MAXQDA qualitative analysis soft- of low complexity hospitals have a pharmacist in charge.
ware. The analysis was carried out using a thematic ap- On average, high complexity hospitals have 3.9 pharma-
proach identifying policies and procedures relating to cists (full-time equivalents) per hospital, while low com-
the inclusion of drugs in the clinical practice of the hos- plexity hospitals have 0.5. Moreover, the number of
pitals. Subsequently, policies were sub classified into na- pharmacists per 100 beds is 1.16 for high complexity
tional and local policies, and procedures into formal and and0.75for lowcomplexityhospitals.
informal procedures. The analysis was carried out using Thepharmacyofficesarefoundtobeconcernedmain-
an approach based on the Grounded Theory, coding ly with the dispensing of medicines and medical supplies,
the data into different hierarchies as suggested by the with the manufacturing of extemporaneous medicines,
data itself. These hierarchies were the result of an it- and the preparation of parenteral nutrition. Pharmacists
erative process in which accounts were classified into are mostly involved in administrative work, although on
Table1SummaryofcharacteristicsofChileanhospitals
Highcomplexityhospitals Lowcomplexityhospitals
N Min Max Mean N Min Max Mean
Numberofbeds 28 116 720 340.6 51 6 317 63.8
Numberofclinicalunits 27 4 25 10.5 53 0 9 3.6
Pharmacistsper100beds 27 0.47 2.47 1.16 51 0 3.33 0.75
Expenditureperbed/year2007(US$) 26 1478 26891 6815 46 121 17928 2915
Collaoetal.BMCHealthServicesResearch2013,13:10 Page5of11
http://www.biomedcentral.com/1472-6963/13/10
Table2CompositionofPTCsinChileanpublichospitals
Complexity Doctors Pharmacists Nurses/midwives Dentists Commercial/ Other Nodegree
level administrativeprofessionals professionals
High 100%(28/28) 100%(28/28) 82%(23/28) 18%(5/28) 60%(17/28) 14%(4/28) 0%(0/28)
Low 100%(45/45) 56%(25/45) 84%(38/45) 44%(20/45) 33%(15/45) 22%(10/45) 53%(24/45)
Total 100%(73/73) 73%(53/73) 84%(61/73) 34%(25/73) 30%(22/73) 19%(14/73) 33%(24/73)
ThefiguresrepresentthepercentageofPTCshavingoneormorepermanentmemberswiththeindicatedprofession.
theirowninitiative,somepharmacistsareinvolvedinclin- ofPTCs,sixapplicantsfordrugs,fivenursesand41clini-
ical duties such as monitoring the use of antibiotics and cianswereinterviewed.
therapiesincriticalcareunits.
Medicinesbudget
PharmacyandTherapeuticsCommittees The budget for medicines was calculated according to
According to the survey, all the high complexity hospi- the historical expenditure plus an additional 6 to 7%
tals and 89% of the low complexity hospitals have a of annual growth. Although some pharmacists reported
PTC. In terms of how long these committees have been that the budget has increased as a result of the GES
established, 83% of high complexity hospitals have a programme, all of them agreed that the budget is insuf-
committee actively working for 5 or more years, while ficient as hospitals are facing an uncontrolled increase
only 39% of low complexity hospitals have a committee in debt.
that has been working for five or more years. The pro-
fessions of those working in the PTCs of hospitals that Policiesontheselectionofmedicines
responded to the national survey are summarized in
Table 2. Additionally, the frequency of meetings agreed Pharmacy and Therapeutics Committee At the time
by PTCs and the reported number of drug applications of the study, hospitals were periodically audited to find
discussedinatypicalPTCmeetingareshowninTables3 out if they had a PTC by the Sub-secretariat of Health
and4respectively. Network of the Ministry of Health. Nevertheless, no
evaluation into how these committees actually function
Drug expenditure was undertaken at the time. Indeed, pharmacists and
The annual expenditure on medicines in Chilean hospi- membersofthe committeeseemedtoonlybeawarethat
tals was estimated based on the responses to the survey there is an obligation to have a PTC through those au-
and expressed as expenditure per bed. For high com- dits rather than through the Department of Pharmaceu-
plexityhospitals theaverageexpenditure per bed wasUS ticalPolicies.
$6,815 during the year 2007, while for low complexity All of the hospitals visited during the study have a
hospitals the average was US$2,914. The overall expend- working PTC that has been functioning for at least five
itureperbed wasUS$4,233. years.Thesehospitalsuseaformstatingthegenericname,
the number of potential patients, and the scientific evi-
Qualitativestudy dence,asrecommendedbythenationalpolicy.
Responserates
Only seven out of the 14 hospitals originally invited, The formulary Every hospital is required to have a list
agreed to participate in the study; namely, three regional of the medicines that are used to treat patients, which
hospitals, one cancer hospital, one children’s hospital translatedliterallyintoEnglishwouldbe calleda“Thera-
and two general hospitals in the capital city. Altogether, peutic Arsenal”. If a medicine is on the list, the hospital
the head of the Department of Pharmaceutical Policies at is obliged by law to have a permanent stock of it and
the Ministry of Health, seven pharmacists, nine members couldbesuedbyapatientifadoctorprescribesthedrug
Table3AgreedfrequencyofmeetingsofPTCs
Highcomplexityhospitals Lowcomplexityhospitals
Frequencyofagreedmeetings N Percentage(validresponses) N Percentage(validresponses)
Bi-monthly 5 17.2 2 4.3
Monthly 12 41.4 8 17.0
Quarterly 10 34.5 22 46.8
Bi-annually 1 3.4 12 25.5
Other 1 3.4 3 6.4
Collaoetal.BMCHealthServicesResearch2013,13:10 Page6of11
http://www.biomedcentral.com/1472-6963/13/10
Table4ReportednumberofdrugapplicationsdiscussedinatypicalmeetingofthePTC
Highcomplexityhospitals Lowcomplexityhospitals
Numberofdrugapplicationsdiscussedpermeeting Frequency Percentage(validresponses) Frequency Percentage(validresponses)
Noapplicationdiscussedyet 0 0 5 10.6
Oneorless 10 35.7 14 29.8
Twoorthree 13 46.5 18 38.3
Fourorfive 3 10.7 6 12.8
Morethanfive 2 7.1 4 8.5
and it is not available in the hospital. However, accord- make an application, applicants are required to fill in the
ing to pharmacists this information is not common form, submit it to the head of the clinical department
knowledge and patients are not normally informed of for authorisation (signature) and submit it to the phar-
this legal requirement. In addition to GES guidelines, macy office. In terms of the types of evidence that are
only one of the seven participant hospitals had designed usually submitted with the application form, the consen-
clinical guidelines for the treatment of their patients, but sus is that applicants do not provide sufficient evidence
these guidelines were written by each clinical depart- tosupporttheapplication.
ment and not by the PTC. Interviewees admitted that
theseguidelinesareseldomused. “...well,Ican’tlie toyou,theevidence isnormally
weak,veryweak...”(MemberofthePTC)
The GES programme There is a different policy for
medicines that are included in the GES programme. Moreover, some pharmacists claimed that in certain
The clinical guidelines for each disease included in the cases doctors support their applications by writing the
programme recommend specific treatments to hospitals; phrase“becauseIneed it”ontheform.
however, the paying party (FONASA in the case of pub- In two of the hospitals the committee’s secretary is re-
licly insured patients) only reimburse medicines that are sponsible for providing the evidence to support the ap-
mentioned in the clinical guidelines, making the inclu- plication. We asked applicants and pharmacists about
sionofthesemedicinesintheformularymandatory. the source of the information that is usually presented
to the committee; the most common answer was “scien-
FormalPracticesrelatingtotheuseofmedicines tificpublications”but withoutnaminganyspecificpubli-
cation. Only one applicant mentioned databases such as
The Pharmacy and Therapeutics committee The se- the Cochrane Collaboration and Web of Knowledge.
lection of medicines in the hospitals we visited is carried Interestingly, all applicants and pharmacists mentioned
out,accordingtoourinterviewees,byamultidisciplinary the drug company’s representative as a source of scien-
committeereferredtoasthePharmacyandTherapeutics tificarticles.
Committee. Invariably, the chair of this committee is the
hospital director and the secretary is the chief pharma- “...andrepresentatives,theyalwaysbringpapers,
cist. The heads of the various clinical departments are theyare veryuseful...”(Chiefpharmacist)
permanent members of the committee, with the excep-
tion of one hospital where the committee is formed by The decision-making process In all the visited hospi-
the chair, the secretary, a clinical pharmacologist and the tals, after the application is received by the committee’s
head nurse. secretary, the application is put on the next meeting’s
agenda to be presented to the members. In the subse-
“...weusedtoinclude allthe headsofthe clinical quent meeting,itsinclusion ontheTAisdiscussed,once
departments,buttheywereonthe committeeonly members have had the chance to study the application.
toaskformedicines,theydidn’treally contribute During the meeting where the application is discussed,
tothe debate...”(Chiefpharmacist) the application is sometimes presented by the applicant
and sometimes by the secretary of the committee. Fol-
The application process There was no consensus as to lowingadiscussionoftheevidencesubmitted,adecision
who is authorised to apply for medicines to be added to is reached by consensus. In one of the hospitals, inter-
the formulary, some interviewees said that only doctors viewees mentionedtheuseofaformulatoweighthedif-
are authorised, while others said that any health profes- ferent aspects of the application such as cost, number of
sional can apply. This lack of consensus was generalised, patients to be treated and the drug’s cost-effectiveness
with statements for both sides made in all hospitals. To ratio.
Collaoetal.BMCHealthServicesResearch2013,13:10 Page7of11
http://www.biomedcentral.com/1472-6963/13/10
When asked about the discussion preceding the deci- “...Formeit’s also easier tobuymedicinesthrough
sion, all the interviewees agreed that it is mainly centred theexceptional purchase,theyarenormallymore
on costs; in particular, members of the PTC pointed out expensive drugs and ifweinclude themintheTA
that when the drug in question is out with their special- everybodywill startprescribing them...the
ity, they do not feel they can contribute an authoritative hospitalcan’t affordthat...”(Chief pharmacist)
opinion tothediscussion,hence thefocus oncosts.
It is important to point out that this is the result of
Perceptions of the PTC TheopinionofthePTCamong the PTC including medicines in the formulary but with-
members of the committee is very positive. They high- out any restrictions on their use. Once a new medicine
lightedthattheexistenceofthecommitteeaddsformality isontheTA,anydoctorcan prescribe it.
and transparency to the process of selecting medicines. Moreover, due to the extra work load involved in pro-
However, the opinion of clinicians is less clear; when cessing many requests for a particular drug, in all the
askedaboutthelocalpolicytoincorporateadrugintothe visited hospitals the committee decided to formally in-
TA,only16outofthe41intervieweddoctorswereaware clude the medicine in the formulary without a formal
of a local policy involving a PTC. Out of the 16 doctors application for the drug, or by asking doctors to submit
who werefamiliarwiththe policy,only sevencould name a formal application. In these cases the drug is included
the functions of the PTC. In addition, all of the doctors without a discussion of the evidence on the use of the
who were unaware of the policy thought that the policy drug. Some doctors and nurses raised the issue of the
forincludingamedicineintheTAwasto“askforitinthe limited scope of medicines used in public hospitals,
pharmacy”. which in their opinion leaves them no alternative but to
In addition, there was some criticism as to how the request the appropriate drug using the exceptional pur-
PTC works, with doctors mainly criticising the bureau- chaseprotocol.
cracy ofthesystem;whilepharmacistscomplainedabout
the extra work load that the committee entails for them, “...theTherapeutic Arsenal ofpublichospitalsis
aswellasthe slowness ofthe process. really limited,there areonly basicdrugsoutdated
longago...”(H1.21 Clin1)
The exceptional purchase procedure In parallel to the
formal application process for medicines not included
in the TA, hospitals have a formal procedure through Informalpracticesrelatingtotheuseofmedicines
which doctors can request the purchase of a medicine
not included in the TA. Doctors need to fill in a form The private purchase of medicines The most common
requesting the drug and explaining their reasons for the informal practice mentioned by physicians in the hospi-
request and submit the form to the pharmacy office. tals in this study, is the external purchase of drugs. In
This request is then sent to the medical sub-director this practice, the doctor issues a prescription for the pa-
who decides whether ornot thepharmacyoffice willbuy tient or his/her relatives to buy one or more medicines
it. According to doctors and pharmacists, the entire pro- in a community pharmacy, for use in the hospital. Even
cedure takes just one day or less for the doctor to obtain though this practice implies an agreement between doc-
the requested drug. All chief pharmacists agreed that tor and patient, there was no mention of any informed
this procedure is overused by doctors, and recognised consent.
that medical sub-directors rarely reject an application It is not clear how common this practice is, since dif-
forexceptionalpurchases. ferent kinds of interviewees had different opinions on
the subject. On the one hand, 27 out of the 41 physi-
“...wehadeight applicationsfordrugslastyear cians interviewed in hospitals mentioned this practice as
(2007)...Forexceptionalpurchase?Wellmore a common way for them to prescribe drugs not listed in
than2,000...waymore...Ijustcan’tcountthem” the formulary. On the other hand, pharmacists and PTC
(Chiefpharmacist) membersinsisteditisrare.
Nurses reported perceived frequencies of private pur-
Moreover, the number of requests for some medicines chases ranging from none to 20% of all prescriptions in
has grown steadily over time, overtly becoming part of their wards. Moreover, nurses pointed out that private
the medicines normally used in the clinical practice of purchases are usually for one of three reasons; namely,
the hospital without approval by the committee, even specific drugs not included in theTA, a temporary stock
though chief pharmacists are not keen to include them shortage of the drug in the hospital, and doctors simply
in the Therapeutic Arsenal because they create an eco- wanting to prescribe drugs of “better quality than those
nomic burden forthehospital. used by the hospital”. Furthermore, in four cases, the
Collaoetal.BMCHealthServicesResearch2013,13:10 Page8of11
http://www.biomedcentral.com/1472-6963/13/10
nursesreported that the deciding factor as to whether or marked resistance of individuals to participate was ob-
not patients receive private prescriptions, is if the doctor served during the fieldwork. Although the study does not
thinks the patient can afford to buy the drug or not (see identify the reasons for this resistance, we believe that in-
Figure1). cluding qualitative methods in the design is particularly
appropriatetoanswerourresearchquestions,insteadofa
The use of free prescription samples Similarly, the use questionnaireprovidinglimitedperspectives.
of free samples to treat patients in hospitals was fre- Despite this, we think that some biases could be pre-
quently mentioned in the interviews by clinicians and sent in the results of the qualitative study. Indeed, only
nurses, but pharmacists seem to be unaware of the use seven out of the 14 hospitals agreed to participate in the
of free samples to treat patients. Doctors gather medical qualitative study, with pharmacists of those hospitals
samples, probably from drug representatives visiting the expressing a high opinion of their PTCs. High complex-
hospital, and treat patients that cannot afford to pay for ityhospitalsthatrespondedtothequestionnaireaccount
the medicines themselves with samples provided by phar- for almosthalfthe total numberofthose hospitals, which
maceuticalcompanies.Nevertheless,itwasnotpossibleto raises the question of whether the absence of a working
establishwhether this iscommon practicefrom the inter- PTCmightbethereasonforsomehospitalsnotrespond-
views. In one of the visited hospitals, we frequently ob- ing.Thus, thereis a possibility thathospitalswithless ac-
servedrepresentativesofpharmaceuticalcompaniesinthe tivePTCscouldbeunderrepresentedinthisstudy.
hospital corridors giving doctors promotional plastic bags
withunknowncontents.
Policies
Discussion The national policy for the use of a PTC is clear and
This study presents, for the first time, information on acknowledged by hospitals. Nevertheless, it seems that
both the size and the complexity of Chilean hospitals the policy is often viewed by hospitals as an obligation
and the control of medicines within them. The results of imposed on them as part of the administrative changes
the qualitative study are, to the best of our knowledge, currently taking place in the public sector, rather than a
the first attempt to widen our knowledge on how de- useful tool to improve the rationality of decisions. In the
cisions about medicines are made in public hospitals local context, policies on the use of drugs are not pre-
of a middle income country like Chile. Nevertheless, sent or not enforced in practice, with an absence of clin-
conducting research in a health system where people ical guidelines to support doctors’ decisions to prescribe
are not used to research was not without problems; a medicines included in the formulary. In fact, with the
Figure1RoutestoincludeadrugintheclinicalpracticeofChileanpublichospitals.Threeroutesareavailablefordoctorstoprescribea
drugnotincludedintheformularyofChileanhospitals.(1)DoctorscanapplytothePTCforadrugthroughthepharmacyoffice.(2)Theycan
requesttheexceptionalpurchaseofamedicine.(3)Iftheythinkthepatientcanaffordthedrug,theycanaskthemtobuythedrugprivatelyto
beusedinthehospital.Thenumberofrequestsforexceptionalpurchasesgrewgraduallyovertimeforsomedrugs;toavoidthepaperwork
associatedwithprocessingtoomanyrequests,thepharmacyofficeasksthePTCfortheinclusionofthisdrugintotheformulary(4).
Collaoetal.BMCHealthServicesResearch2013,13:10 Page9of11
http://www.biomedcentral.com/1472-6963/13/10
exceptionofexpensiveantibiotics,noprescribingrestric- Arsenal; taking the appropriate measures to avoid this
tionswereidentified. practicewillresultinamorerationaluseofmedicines.
Moreover, it is important to improve the communica-
Theformulary tion between the PTC, clinicians and the pharmacy
The formulary used in Chilean hospitals seems to be no office. At the time of the study, the use of medicines
more than a stock list. According to our results, the so appeared to be a linear process in which medicines were
called Therapeutic Arsenal does not include clinical selected, prescribed, dispensed, and then administered;
guidelines to orientate doctors in the use of medicines but there was no feedback about the outcomes achieved
within hospitals. Moreover, the obligation to keep a per- with newly included medicines. Thus, to achieve a more
manent stock of the medicines that appear in the TA rational use, there should be continuous communication
might be discouraging hospitals from updating their between the PTC, the Pharmacy office and clinicians
TAs, but this could be overcome by including medicines about theuseofdrugs(seeFigure2).
along with appropriate policies for their use. In addition,
it is also possible that Therapeutics Arsenals do not ThePharmacyandTherapeuticsCommittee
cover the whole spectrum of medicines needed to treat Jenkings and Barber describe the lack of skills of DTC
patientsinthosehospitals,obligingdoctorstoseekmed- members to analyse and contextualise scientific informa-
icines outsidethe hospital. tion [29]. Coincidentally, members of the studied PTCs
Even when the reasons for prescribing medicines that self-reported that the evidence discussed in PTC ses-
are not included in the Therapeutic Arsenal are unclear, sionsisoftendifficulttoapply to theChilean reality,and
the lack of obligation to prove the bioequivalence of they do not feel able to challenge the opinions of specia-
generic medicines in Chile may be relevant. With hospi- lists in areas other than their own. This is clearly a bar-
tals using mostly generic drugs, the distrust doctors rier to making more rational decisions in Chilean
show towards these types of drugs is an important issue committees, highlighting the need to train members in
that can partially explain why doctors prescribe drugs the appraisal and analysis of scientific information be-
not included intheTAofpublichospitals. yondthescopeoftheirspecialities.
Additionally, PTCs in Chile need to develop formulary There is a positive perception among members of the
systems which can effectively cover all the pharmaco- PTC about having a committee in charge of decisions to
therapy needed in hospitals, avoiding the exceptional include a drug in the TA. However, it seems that for the
purchase of medicines. Moreover, it is necessary to im- group of hospitals under study, the PTC does not have
prove our understanding of doctors’ reasons for prescri- enough authority to influence prescribing in their own
bing medicines that are not included in the Therapeutic hospitals,leavinginnovationintheuseofdrugs(i.e.using
Selection
Acquisition Outcomes
PTC
Pharmacy
Prescription Administration
Dispensation
Figure2Information-flowdiagram.Insteadofalinearprocess,theuseofmedicinesinChileanhospitalsshouldbecontinuouswhere
informationoneachstageoftheuseofmedicinesissharedbythePTC,thepharmacyofficeandclinicalservices,providingcontinuousfeedback
abouttheuseofmedicinesincludedintheTherapeuticArsenal.
Collaoetal.BMCHealthServicesResearch2013,13:10 Page10of11
http://www.biomedcentral.com/1472-6963/13/10
drugsnotpreviouslyusedinthehospital)tothecriteriaof quality and to circumvent abuses of the exceptional pur-
individual doctors or groups of doctors. This lack of au- chasing system. In general, the lack of regulation in the
thority could be the result of poor argumentation about Chilean market for medicines may be having a negative
reasonstoincludeornotamedicineintheformularyand impact on the access to good quality medicines for the
the lack of communication about the decisions of health majority ofChileancitizens.
professionalsworkinginthehospital,underminingthele- More research is needed to better understand how me-
gitimacyofthecommittee’sdecisions. dicines are used in Chilean hospitals; for instance, it is
Thisneedforlegitimacy canbeimprovedwithanethi- fundamental to establish doctors’ reasons for prescribing
cal approach called the ‘Accountability for Reasonable- medicinesthatarenotincludedintheformulary.Onlyby
ness [30]; this approach was put forward to give greater understanding how medicines are used in Chilean hos-
legitimacy to health technology decisions. This approach pitals, can the appropriate policies be put in place to im-
advises that decisions must be effectively communicated proverationalityintheuseofmedicines.
to doctors, stating the rationale for the decision and gi- The results of thisstudy can serveas a basis for further
ving the opportunity to appeal. By using an ethical ap- research into how middle income countries can change
proach such as the Accountability for Reasonableness, their priorities, means and needs in terms of healthcare
decisions made by the PTC can gain greater legitimacy, and the way medicines are used. Such research can be
gradually building the authority needed to influence pre- usedtodevelop appropriatepoliciesfor anoptimal use of
scribinginthegroupof hospitals included inthisstudy. medicinesandtoimprovehealthoutcomes.
Moreover, several authors suggest different strategies
to improve the decision-making process for including a Additional file
drug in the formulary. In this respect, Tan et al. suggest
that institutions should clearly state a hierarchy of prior- Additionalfile1:Appendix.TheappendixshowstheEnglish
ities before any discussion on including a drug is made translationofthequestionnaireusedinthenationwidesurveyandthe
interviewschedulesusedduringthequalitativestudy.
[31], and Schiff et al., propose a Formulary Drug Appli-
cation Tool, which they suggest might help DTC mem-
bers to evaluate claims about drugs made during the Competinginterests
Theauthorsdeclarenocompetinginterest.
application process, giving a framework to improve the
levelofthediscussion[32]. Authors’contributions
AlloftheinterviewsandanalysiswerecarriedoutbyJFC.FScollaborated
withmethodologicalguidance;NBwithmethodologicalguidanceandthe
TheGESprogramme
analysisofinterviews.Allauthorsreadandapprovedthefinalmanuscript.
Medicines included in the GES programme are normally
drugs not included in the TA of the hospitals under Acknowledgements
study; in practice, they are imposed on hospitals, contra- TheauthorsthanktheChileanMinistryofEducationthroughitsMECESUP
programmeUVA0207forfinanciallysupportingthisstudy.
dicting what the National Policy of Medicines of 2004
states regarding the selection of medicines. Nevertheless, Received:28June2012Accepted:26December2012
sincedecisionstoincludeadrugintotheTAarenormally Published:7January2013
basedalmostexclusivelyoncosts,itishighlyprobablethat
References
hospitals would not include those drugs in their TAs if
1. HollowayK,GreenT:Drugandtherapeuticscommittees:Apracticalguide.
theywerenotobligedto.Thus,theobligationof hospitals Geneva,Switzerland:WHO;2003.
to include GES drugs could be improving the access to 2. FijnR,BrouwersJR,KnaapRJ,DeJong-VanDenBergLT:Drugand
Therapeutics(D&T)committeesinDutchhospitals:anation-wide
moreinnovativemedicinesinthosehospitals.
surveyofstructure,activities,anddrugselectionprocedures.BrJClin
Pharmacol1999,48(2):239–246.
Conclusions 3. FijnR,BrouwersJR,TimmerJW,deJong-vandenBergLT:Rational
pharmacotherapyandclinicalpracticeguidelines.Theoriesand
The national policy for selecting medicines in public
perspectivesonimplementingpharmacotherapeutictreatment
hospitals is, as recommended by the WHO, through lo- guidelines.PharmWorldSci2000,22(4):152–158.
cal multidisciplinary committees; however, there is a 4. FijnR,denBergLTdJ-v,BrouwersJR:RationalpharmacotherapyinThe
Netherlands:formularymanagementinDutchhospitals.PharmWorldSci
need to improve the ability of committees to influence 1999,21(2):74–79.
prescribing towards evidence based practice. More cen- 5. MannebachMA,AscioneFJ,GaitherCA,BagozziRP,CohenIA,RyanML:
tral guidance on the selection of new medicines could Activities,functions,andstructureofpharmacyandtherapeutics
committeesinlargeteachinghospitals.AmJHealthSystPharm1999,
help committees to make more rational decisions, and 56(7):622–628.
to design policies and protocols for the use of medicines 6. SegalR,PathakDS:Formularydecisionmaking:identifyingfactors
basedonthebest availableevidence. thatinfluenceP&Tcommitteedrugevaluations.HospFormul1988,
23(2):174–178.
There is also a need to revise the mechanisms for pur-
7. ThurmannPA,HarderS,SteioffA:Structureandactivitiesofhospitaldrug
chasing medicines in public hospitals to improve their committeesinGermany.EurJClinPharmacol1997,52(6):429–435.