Document 15089

Journal of crohns and Cotitis (2007) 1, 77-81
ffi
a v a i l a b l e a t w w w . s c i e n c e d i r ec t . c o m
"lj!'s.iun."Direct
ELSEVIER
to treatment in inflammatorybowel
Non-adherence
diseasein CzechRepublic*
Petr Cervenyu'*,MartinBortlikb,Jiri Vlcek', AlesKubenau,MilanLuk6sb
u Department of Social ond Clinical Pharmacy, Faculty of Pharmocy in Hradec Kralove,
charles University,Prague,CzechRepublic
b tBDClinical Researchcenter, ISCARE
Lighthouse,CharlesUniversity,Progue, CzechRepublic
Received28 June 2007; accepted
KEYWORDS
Adherence:
UtcerativecoLitis;
Crohndisease;
Inflammatorybowet
disease;
Patientcomptiance
'16August2007
Abstract
Objective:To assessoveratt non-adherenceto the treatment among patients with Crohns
dj5ease(CD)and utcerativecotitis (UC).
bowetdisease(lBD)patientswere enroLledin the study
Potientsandmethods:396infLammatory
their non(2OOmates,196females,210 cD, 186UC)and futfilted the questionnaireto assess
adherentbehaviourduringthe treatment.The data wasanatysedusingfactor anatysis.
was reportedby 32%of patients.A 12%of patients
Results:overattintentionalnon-adherence
reducingwasreportedby
the treatment.Votuntarydose
reportedtheyat teastoncediscontinued
non-refillthe medicationin time. Therewere no
19%of patients.An 11%of patientsoccasionalty
differencesin intentionaladherencebetweenmatesandfemales,diseasetype, previousbowel
statuses.A 42%of patientsreportedunintentionaL
surgery maritaL,smokingand non-smoking
Factoranatysjsprovednon-adherentpatientsare more Liketyto havea higher
non-adherence.
( r=0.109,p=0.008).
activityof the disease
is reLativetyhigh amongIBDpatientsand a
The overattintentionalnon'adherence
Conclusions:
how to
gastroenterotogist's
attention shouldbe focusedon it. Our resuttsstimutatediscussion
improyeeducationof the patientswith inftammatoryboweldiseaseand accentimportanceof
the maintenancetherapyto them.
by EtsevierB.V.AtLrightsreserved.
Pubtished
@2007EuropeanCrohn'sandCotitisOrganisation.
1.lntroduction
i The study was fjnanciattysupportedby Ferring-L6iiva a. s.
* Correspondingauthor. Heyrovskeho120350005 Hradec Kra(ov6
CzechRepubuc,Tet.: +420 495 067 251; fax, +420 495 517 266
(P Cerveny).
E-moil oddress:[email protected]
of inftammatorybowet disease(lBD)
Pharmacotherapy
phasesof the
aswett asasymptomatic
targetssymptomatic
diseaseand may be effective in remissioninductionand
papersdocumented
a poor
A few publ,ished
maintenance.l
triatsettings.l-8
rateoutsidethe cl.inical.
adherence
1873-9946/5- see front matter @2OO7EuropeanCrohn'sand Cotitis Organisation.Pubtishedby EtsevierB.V Att rights reserved.
doi:10.1016/j.crohns.2007.08.002
78
P. Cervenyet at.
Patient'sadherenceto treatment is defined as the rate of
cooperation in foLtowingthe physician'sprescriptionsand
recommendations.e
A sufficient adherenceis one of the key
factorsof treatmentsuccess,
the pubtishedpapersdocumented
an increasedrisk of ftare-upamongnon-adherentpatients.3,6
Non-adherencetogicaLtymeans a lack of cooperationin
fotlowingthe physician's
prescriptionsand recommendations.e
ln spite of progressin the pharmacotherapyin lBD, many
patientsstitt ftare and non-adherence
to the treatment maVbe
the reasonof this.a
Non-adherenceto pharmacotherapyis nowadaystaken for
oneof the mostseriousprobtemsfor modernmedicineto face.e,10
The doctor patient retationship,treatment regimenand
other disease-retatedfactors ptay a key role in the adherence process.The treatment duration, severat adverse
effects of the medications,or symptom reduction, or even
disappearance,during the remission phase ptay atso it's
specific role.lo 12Besidesthis adherencerapjdty decreases
with the increasingnumber of prescribedmedicinesand is
atsoinverselyretated to the number of dosesper day.12,13
In our study we assessedrate and most frequent reasons
of non-adherenceto the pharmacotherapyof lBD.We useda
Table 1 Demographic
characteristics
of the examined
cohort(N=396)
Characteristics
N (%)
Mate
Femate
Averageoge
Min.
200(50.5)
196(49.51
39.1years
18
79
Education
Etementary
Trained
Highschoot
University
Status
Student
Working
Pensioner
Condition
Singte
Maried
Widow/widower
Divorced
Smokers
29 (7.3)
125(31.5)
187(47.2)
5 5( 1 3 . 9 )
18 (9.6)
271 (68.4)
87 (22.0)
139(35.2)
200(50.s)
1 1( 2 . 8 )
4 s ( 1 .15 1
78 (19.7)
Avercgeage of IBDdiagnostics
MAX.
Averagelength of IBD treatment
Min.
MAX.
Fami[ior occurrenceof IBD
lBD, inftammatorybowel disease.
9
79
7.4 years
1
41
63 cases(15.9)
Table2
Clinicatdata of the examinedcohort lN=396)
Diseasecharacteristics
N (%)
Crohn'sdisease
Activity
Remission
Chronicallyactiye
Ftareup
Localization
Terminaliteum
Colon
lleo-colitic
UpperGIT
Diseos belnviour
Non-stenosing/nori-perf
orating
Sten6in€
Moratir€
Itedicatidl6
Anincaltl ates
lmrrurE-fe5sa.tts
Topicatqtideroids
Systemt qtiderdds
tledbtix! free
2 1 0( 5 3 . 1 )
Ulcerative a&is
ActiYity
Refi|isin
CfYorta|ly ditE
Ftarc |.p
Lndifr*n
Proctitis
Pr.rh*!dti5
l€ft-ibd
bCgrie
hffitB
'brc
Ani
brrr| EgE*s
Tqidcrh:EBrie
5t$€[email protected]
tffibfe
148(70.51
50 (23.8)
1 2( 5 . 7 \
69 (32.8)
58 (27.7)
8 0( 3 8 . 1 )
3 (1.4)
1 1 3( 5 3 . 8 )
54125.7)
43 (20.s)
'188
81
65
34
9
186 (46.e)
131(70.4)
3e (21.0)
16 (8.6)
16 (8.6)
18 (9.61
e4(s0.s)
58(r1.3)
169
25
12
28
12
GI[,gffit-nquesh-ornaiesj-e
10 gastrEisticgc!
r.urg patientswith lBD,fotowed uo in
cefliers in the CzechRepubtic.
2. Patiefits and rethods
PatientswrJ- r€ J€€!-€ss of IBD(Crohn'sdiseaseand utcerative co{itbr- :bjo.€a -E in the gastroenterotogicatambutances,bEre e-r:!ca r :re study betweenAugust2005and
-re =-ra-riDed
February 2(Ecohort comprised of 396 pa:f,
gar€-:erotogicaL units from attover the
tients fruu
CzechReg.a-- .q rEtai€d characteristicsof the examined
-a6
cohort is distr4le r
1 and 2. Duringone ambutatory
visit of a gasrET=D'-ti.
patient was askedfor participation in tlE CJ€=[rrai.E srvey. The patients were briefty
infomed dDr ds-€.rce probtemsand the study'scharacter. ]lFt €e yEe.-.€d an anonymousquestionnaire,
where thet €a-rc eErs
their own experience with fottowing tlF EaFt€.r{ogist's
recommendationsand orders.llret E E:6-:-. 3-.a.rdnteed
that no DersonaL
data are
Non-adherenceto treatment in inftammatory bowel diseasein CzechRepubtic
79
duringthe treatmentwas admitted by 50 patients(12.6%).
The most frequent reasonswere: feeting wetl and therefore
supposingthere is no needto continuewith pharmacologicaI
treatment (54%)and adversedrug effects occurrence(16%).
In women pregnancyor breast feeding were atsoreasonsfor
intentionat treatment discontinuation(10%).
At least one intentionat dose reduction without gastroenterotogist'sconsentwas in the questionnairestated by 78
respondents(19.7%).The most frequent reasonswere: feeting wett (64.1%)and runningout of the medicationand
therefore a necessityto reducedosesin order to saveit titl
the next visit of a gastroenterotogisr(7.9%1. Adversedrug
effects occurrencewas againone of the frequent reasonsof
just as pregnancy
and breastfeeding
dosereducing(11.5%),
(3.8%).
in women
Forty five respondents(11.3%)admitted they minimatty
once did not refitl their medicationson time. The meantime
of being without the medications in these patients was
7 days.
Onty 9 patients (2.2%)stated they use their medications
onty when wetl feeting deteriorates and onty 2 patients
(0.5%)use it just prior to visitingthe IBDUnit. No one reported comptete disregardto the prescribedtreatment.
Generalty,at least one form of intentional non-adherence
was stated by 129 respondents(32.5%).
UnintentionaInon-adherence(occasionattyforgotten prescribeddail.ydoses)was reported by 169 patients (42.6%).
As for addjtional questions, 88 patients (22.2%)stated
they detected at least once an adverseeffect of the prescribed IBDtherapy drugs.
In patients with CD receiving aminosaticylatesthe nonadherencerate at any timepoint was 34%vs. 40%in patients
vs. 38%in patients on systemiccorti'
on immunosupressants
costeroidsvs. 31%in patientson topicat corticosteroids.The
differenceswere statisticattyinsignificant.
The non-adherencetended to increasein CDwith disease
activity (30%in remissionvs. 42%in chronicattyactive vs. 42%
in flare), howeverdid not reach statistical significance'
Simitarstatisticattynot significantdifferenceswere found
patients
with lJC. The overatl non-adherenceat any time
in
3. Statisticalanalysis
was 32%in patients on aminosaticytatesvs.40% in those on
vs, 39%on systemic corticosteroidsvs.
The data was processedby the statisticat software [email protected], immunosuDressants
'l'1.5.
42%on toDical corticosteroids.
For the characteristicsof the tested cohort,
version
The differences in activity in CD was not of statistical
descriDtivestatisticswere used.
significance(33%in remissionvs. 31%in chronicaltyactive vs.
The / (chi square) testing was performed to compare
25%in flare).
frequencies of demographicvariables and diseasecharac293 patients (73.9%)rate their treatment as quite efteristics between adherent and non'adherent groups.
44 (1'1.1%)take it for mitdty effective, 57 patients
processed
fective,
factor
anatysis,
by
the
The interview data were
(14.5%)were not abte to score it and onty 2 respondents
a method that enabtesthe reduction of a large body of data
(0.5%)evatuatedtheir treatment as completety ineffective.
into a few independent factors. The factor analysisatlows
25.7%of lhe Datientsseek additiona[information about the
the researcherto keep the maximumamountof information,
prescribed
drugs. As for disease,40.1%of the patients seek
factors
and
white finding retationships between derived
additionatinformation.
additional variabtes.la To corretate individuaL adherence
By means of 12 testing no statisticatty significant retafactors with other ctinical and demographicvariabtes, the
tionshiDsin adherencebetween the two diseases,genders,
Kendatt'sTau coefficient vatueswere used.
<0.05
smokersand non-smokers,marital status and other demostatisticatty
significant.
was considered
A p vatue of
graphicvariabteswere found.
alt tests were two-taited.
to be stated and that the responsesin no way witl inftuence
their further treatment options. Those who agreed with
participation were enrotled. Patients were not given advancednotice about the study before visiting the IBDUnit.
After agreement attending gastroenterologistprovided
information about the diseasetype (CDor UC),activity, and
tocatizationto the speciat protocot which was a part of the
questionnaire.The disease activity was simptified for the
purpose of statistical anatysisto a three-grade scate (remission,chronicattyactive, ftare). Subsequenttypatient was
giventhe questionnaireto complete it. After that, he seated
the questionnaire in the envetope and when leaving the
ambutance, he teft it at the designated place at the reception. After termination of the study the questionnaires
werecotlected from the gastrocentersandjointtyunsealedat
the Departmentof Socialand CtinicaIPharmacyat the Faculty
of Pharmacy.This ensuredfut[ anonymityto the patient.
Key questions focused on adherence, adapted from a
previouslypubtlshedpaper,l were inctuded with additionat
questions to comptete a 3o-item questionnaire. The key
questions on adherence targeted at treatment discontin'
uation, dose reduction, regutarity in using medicines and
refitling medicines. Additionat data was of demographic
character, knowLedgeon diseaseand medicinesand regis'
tered adversedrug effects (data not shown).
In our study,patientswere consideredto be non-adherent
if they stated to have at least one of the possibte nonadherent behavioursin the questionnaire:a) not using the
prescribedmedicationsat atu b) usingthe prescribedmedications onty prior to visiting a gastroenterotogist;c) using
the prescribedmedicationsonty when their feeting of wetl.being deteriorates; d) treatment discontinuation without
consent;e) dosereductionwithout
their gastroenterotogist's
consent;f) not refitting their medtheir gastroenterotogist's
ications on time.
Prior to the study [aunchthe questionnairewas validated
in the pitot group of IBDpatients, foLtowedup in the GastroenterotogicatCentre of the UniversityHospitatin Prague.s
4. Results
4.1. Factoranalysis
A totat of 396 IBDpatients participated in the questionnaire
survev. At least one period of treatment discontinuation
The 11 questions, crucial for adherence, were proceeded
usingthe factor anatysis.The four independentfactors were
P. Cervenyet at.
80
Table3 Contributionsof 1'l crucialinterviewquestionsto
the factors characterizingadherencein the cohort
(correlationcoefficientt)
tike diseasetype, previousbowet surgery
characteristics
werenot found'
gender,maritalor smokingstatuses,
5. Discussion
Factor
Regutarityin using
Actionafter omitted dose
Omitteddosesduringtast
month
Treatment
discontinuation
Dosereducing
Doseincreasjng
Refillingthe medicinesin
me
Not refittingduringlast
year
Efficacyevatuation
on disease
Knowtedge
on medicines
Knowtedge
0.822 0.121 -0.043 0.090
o.T)4 0.001 0.070 -0.003
0.725 0.106 0.008 0.213
0.130 0.094 -0.1'16 0.677
0.152 0.051 0.074 0.647
0.081 0.230 0.189 0.555
0.098 0.899 0.072 0.145
0.126 0.893 0.042 0.003
o.M2 0]32 -0.099 -0.333
-0.008 0.026 0.873 0.101
0.036 -0.049 0.856 0.099
identified (Tabte3). Thesefactors can be taken for certain
character traits of adherencein the examinedcohort'
1. A factor characterizinggeneral non-adherenceto treatment. A tow regutarjty in using medicinesand frequent
dosesomitting is remarkabte.lt is atso characterizedby
treatment discontinuation,dosesreduction and non"refittingthe medjcinesin time.
Z. A fa-ctor characterizing patient's depreciation to the
treatment. Non-refjtLingthe mediclnesin time and
irregular usingof them is considerabte'
3. A fictor characterizing patient's consistencyin treatment. lt is characterizedby a tack of treatment discontinuation and votuntary increasein dose, when wetl
feeting deteriorates. An active searchingfor additional
information on medicinesand diseaseis considerabte'
4. A factor characterizingintentional non-adherenceand a
tack of confidenceto the treatment' lrregularity in medicines using is considerabte,just as treatment disconti'
nuation and doses reducing. A very low rating of
treatment effectiveness,but an active seekingadditionat
information on diseaseand medications,are remarkabte'
The factors were subsequentlycorretatedwith additionat
questionnaire data (personat, demographic, disease,and
medication data), usingthe Kendatt'sTau coefficient vatues
,, and a signjficance
Level(p vatue).
Adversedrug effects occurrencepositivetycorretateswith
Factor3 (r=0.119,p=0.005)andatsowithFactor4(r=0 107'
p=0.011). Both factors are characterizedby active seeking
additionatinformation on diseaseand medicinesby patients'
Factor4, characterizingintentionat non-adherenceand a
tack of confidence to the treatment, is in a positive corretationwith diseaseactivity (T=0.109,p=0'008)' Patients
with a history of non-adherentbehaviourare of a higher risk
of ftare-up.
In addition, by means of factor anatysis, statisticatty
significant retationshipsbetween non-adherenceand other
lnsufficient adherenceto the pharmacotogicaItreatment of
IBD may be a reasonof it's faiLure.A complicacyand comptexity of a processof patient'sadherenceto the treatment
jt's research'e
unabtesusinga singtemethod for
papers
address this propubtished
previousty
A few
btems.l-8 ln case of studies using anatytical determination
of active substanceor a metaboLitein biotogical materiat,
Van Heesand Van Tongerenassessednon'adherenceto the
sutfasatazinetreatment upon 12%,2Other studies anatyti'
catty demonstratedthe absenceof mesataminein 12% and
7
in 13% of examinedpatient cohorts.
Usingpatient medication recordsfrom the pharmacy,the
wasstated at onty 40%'
oercen;ee
'Patients of full.yadherentpatients
were designatedas non-adherentif they used less
than a0%of the prescribeddosesduring the past 6 months'4
Sewitchet at. stated intentiona[ non-adherenceat 36'5%
of the examinedcohort.rShateet at. provedretrospectively
ingestionof tessthan 80%of pre5cribedmedicinesin 43%of
pa--tients.t35%of non-adherentpatients were identified in
the study of Lopezsan Roman./
Kaneet at. suggestmate genderand singtestatus are risk
In the studyof Shateet at', younger
factorsfor adherence.a
years'
Datients, patients educated beyond the age of 16
prescribed
a
being
emp(oyment,
patientshavingfutt'time
and
medications
no
other
taking
regimen,
three-times"a-Aay
identified
ferceiving their medication as ineffective, were
A higher degreeof intentionalnon-ad'
as non-aJherent.5
herencein the studyby Lopezsan Romanwasassociatedwith
ereater patient's dePressionand patient-physician discorpatients
iance, with patients having tonger'standinglBD,
themconsidered
who trusted their Physicianless and who
treatment"
their
about
setvesto be lessinformed
Cervenyet al. proved adherencetends to decreasewith
younger "ge of Patients and interestingty atso with higher
education ot patients. lt was also documentedthat adverse
drug effects impair patient's attitude to the treatment and
subitantiatty decreaseadherenceto the treatment'8
Our study faited to prove any statisticattysignificant retagender'
tionshipsbetween adherenceto the treatment and
and its
type
disease
previous
surgery
bowet
maritai status,
non-smokers'
and
smokers
and
locatization
pubDifferencesin methodotogiesof our and previousty
of
resutts'
comparison
tished studies unable more detailed
Anotherbarrier of comparisonis definition of non-adher'
ence, that is arbitrarity set in each study' Howevergeneratty'
baseduDonthe output of descriptivestatistics, we can conctude the ratio of intentionatty non-adherentpatients in our
published
cohort is simitar tike described in previousty
the
we
report
non-adherence,
unintentionat
DaDers.As for
ln
iaiio in our study to be smatler,than previoustypublished
nonthe
unintentionat
our study we did not focus further on
adherenie, as we assumedthis type of non-adherenceis not
as riskv as the intentional one that is of greater impticatjon
and needsto be addressed.T
By means of factor anatysis,we faiLed to confirm that
adherencerate dependson gender,diseasetyPe and other
demographicvariabtes.we report a very strong inftuenceof
Non-adherenceto treatment in inftammatorv bowet diseasein CzechRepubtic
adverse drug effects on adherence and, contrary to
previously pubLishedpapers which do not address this
issue,we found it adherenceimpairing. Intentionalnonadherent behaviour due to adverse drug effects was the
secondmost often reported reasonin the questionnaire.In
addition, adversedrug effects were independenttyprovenby
the factor analysis to affect a patient's confidence in
treatment. Adversedrug effects occurrencewas frequently
reported in patients activety searchingfor additional
information on diseaseand medications.ln caseof medicines
these information logicatty inctude those on adverse drug
effects, knowtedgeof which may impair adherenceto the
treatment. Patientsmay be more consciousof adversedrug
effects and therefore they may observethem on themsetves.
Adversedrug effects ofdrugs for IBDshoutdbe in the scopeof
and shoutdbe taken accountof.
a gastroenteroLogist
Our another very important finding is non-adherentpa'
tients are tikety to suffer from more severe disease.This
confirmsresuttsof previoustypubtishedstudiesadherenceto
the pharmacotherapyis important for maintainingremission
in oatientswith 18D.3,6,8
Resuttsfrom factor analysis suggestadherence to the
treatment in the examined cohort may be strongly inftuenced by patients knowtedgeon diseaseand pharmacotherapy.Thesefindingsopen a discussionof optimat educationof
patients on these topics by a gastroenterotogist.Particutar
attention shoutd be paid to information on adverse drug
effects, as atsoresuttsfrom factor anatysis.
Generatty,if there is a tack of a treatment success,a
gastroenterotogistshoutd think among others of possibte
non-adherent behaviour of a Datient. Patients shoutd be
intensivetyinstructed about the importanceof maintenance
therapy and adherenceto the treatment in order to achieve
the therapeutic goa[ - reducedfrequencyof retapse.
Our research resutts may be re(ated to the entire IBD
poputationin the CzechRepubtic.The examinedcohort was
formed by patients fottowed up in 10 gastroenterol.ogicat
units from attover the country and the number of cohort
subjectswas not inconsiderabte.
6. Conclusion
At least one form of intentional non-adherencewas documented in 32.5%of the examined cohort of 396 IBD patients. The intentional non-adherenceis characterized by
treatment discontinuationand dosesreduction, both withconsent.The most frequent reason
out a gastroenterologist's
was feeling wett. Adversedrug effects atso pl.ayimportant
rote in adherenceprocess.Another important form of nonadherent behaviourby patients is that of not refilting their
DrescriDtionson time.
Thefactor analysisdocumenteda higherdegreeof disease
activity among non-adherent patients. Adherence to the
treatment was also proven to be impajred by adversedrug
effects.
Adherenceto the treatment seemsto be inftuencedby a
Datient'sdiseaseand medication awareness,Gastroenterol-
81
ogistsshoutdeducate IBDpatients regardingthe importance
of maintenance therapy and adherence to the treatment
generarry.
Lowering of intentional non-adherencemay be one of
possibleways how to improve effectivenessof IBDtherapy.
Acknowledgements
To the members of The working Group for IBD in Czech
Repubtic,to have madethe study possibteto be doneat their
working ptaces: Pavet Kohout, MD, PhD, Prague; Otga
Shonove,MD, Cesk6 Budejovice; Jana KozLuhova,MD,
Plzen;LudekHrdl.icka,MD, Prague;KareLMarey,MD, Prague;
Lenka Nedbatova,MD, Leberec; Michat Konecne,MD, PhD,
Olomouc; Tomas Douda, MD, PhD, Hradec Kratove; Libor
Buzga,MD, Ostrava;MirostavSamek,MD, Most.
References
1. Sewitch MJ, AbrahamowiczM, Barkun A, et at. Patient nonadherenceto medication in inftammatory bowet disease.Arn
J Gastroenterol2003:.7.153544,
2. van HeesPA,vanTongerenJH. Comptianceto therapyin patients
on a maintenancedose of sutfasatazine.J CIin Gastroenterol
6.
Aug1982;4(4):333
3. RiteySA, Mani V GoodmanMJ, LucasS. Why do patients with
Gut Feb 1990;31(2):179-83.
utcerativecolitisreLapse?
4. KanesV CohenRD,AikensJE, HanauerSB. Prevatenceof nonadherencewith maintenancemesalaminein quiescentuLcera33.
Oct 2001;96(10):2929
tive coutis.AmJ Gostroenterol
5. ShateMJ, Ritey5A. Studiesof compliancewith detayed-retease
mesaLazinetherapy in patients with inftammatory bowet
disease.AIiment PharmacolTher Jul f 5 2003;I I {2): 191-8.
6. Kane5, Huo D, AikensJ, Hanauer5. Medicationnonadherence
and the outcomesof Datientswith quiescentutcerativecoLitis.
An J lted Jan2003;11411):3943.
7. Lopez San Roman A, Bermejo E Carrera E, Perez-AbadM,
Boixeda D. Adherence to treatment in inflammatory bowet
disease. Rev EJp E ferm Dig Apt 2005;97(4)i249-57
8. Cerveni P, Borttik M, KubinaA, vtdek J, LakatosPL, Lukei M.
Nonadherencein inftammatorybowel disease:resuttsof factor
analysis.lnflomm Bowel DtsMay 2007;30 [Epubaheadof print]
PMID:17538983.
9. Vermeire E, HearnshawH, Van RoyenB DenekensJ. Patient
adherenceto treatment: three decadesof research A compre_
hensivereview.J ciin Phorm Ther Oct 2001;2615\133142.
10, Mitter NH. Comptiancewith treatment regimens in chronic
asymptomaticdiseases.Am J Med Feb 17 1997)102(24)1439
11. Levy RL, FetdAD. Increasingpatient adherenceto gastroenterotogytreatment and prevention regimens.Am J Gastroenterol
42.
Jul 1999;94(7'1i1733
12. Ctaxton AJ, Cramer J, Pierce c. A systematic review of the
associationsbetween dose regimensand medication compti
ance. CIin Thet Aug2001;23(8):1296-310.
13. LevyG. Medicationnon-comptiance- when hard sciencemeets
soft science. lnt CongrSer2001:1220'.12533,
14. SheshkinD. Handbookof parametric and nonparametric
statisticalprocedures.3rd ed, BocaRaton,FL: CRCPress;2004.
`