A study of adherence to antibiotic treatment in ambulatory respiratory infections

A study of adherence to antibiotic treatment in ambulatory respiratory infections
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institutionand sharing with colleagues.Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third partywebsites are prohibited.In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further informationregarding Elsevier’s archiving and manuscript policies areencouraged to visit:  Author's personal copy Astudy   of    adherence   to   antibiotic   treatmentinambulatoryrespiratoryinfections CarlLlor a, *,SilviaHerna´ndez a ,   CarolinaBayona b ,AnaMoragas a ,NuriaSierra a ,MartaHerna´ndez c ,MarcMiravitlles d a Primary   Care   Centre    Jaume   I,   c.Felip   Pedrell,   45–47,   43005   Tarragona,   Spain b Primary   Care   Centre   Valls   Urba` ,   Valls,Spain c Hospital    Joan    XXIII    of    Tarragona,   Tarragona,   Spain d Institut    d’Investigacions   Biome` diques    August    PiISunyer    (IDIBAPS),   Hospital   Clinic,   Barcelona,   Spain 1.   Introduction Medical   adherence   is   defined   asthe   extent   to   which   a   patient’staking   of    medication   is   consistent   with   medical   orhealth   advice. 1 Non-adherence   to   medications   is   particularly   important   inclinicalpractice.   Adherence   tomedications   has   long   been   a   concernbecause   it   often   affects   the   outcome   of    treatment.   In   a   review   of    63studies   over   a30-year   period,   theauthors   reported   that   if    thepatient   is   adherent,   the   oddsof    a   good   outcome   are   almost   three-fold   higher   than   for   those   who   are   non-adherent. 2 In   the   case   of infectious   diseases,   non-adherence   toantibiotics   might   also   lead   tothestoring   of    antibiotics   at   home,   which   induces   self-medication,leading   to   a   vicious   circle,   and   thereby   favoring   the   emergence   of bacterial   resistance. 3 Measuring   adherence   is   difficult   because   most   of    the   direct   andindirect   measures   available   have   limitations.   Since   their   introduc-tionin   1986,   microelectronic   devices   have   become   thegoldstandard   inadherence   research. 4 The   most   commonly   used   systemisthe   Medication   Event   Monitoring   System   (MEMS).   MEMSmedication   bottles   contain   a   microelectronic   chip   that   registersthe   date   and   time   of    opening   of    every   bottle.   Assuming   that   theopening   of    a   bottle   represents   theintake   of    medication,   MEMSprovides   a   detailed   profile   of    the   patient’s   adherence   behavior.   Forthis   reason,   MEMS   is   currently   regarded   as   the   gold   standard   forthe   measurement   of    adherence. 5–7 MEMS   have   been   used   tomonitor   adherence   mainly   with   long-term   medications,   and   in   thecase   of    infectious   diseases,   this   technology   has   particularly   beenusedto   track   medication   adherence   with   antiretroviral   agents   andwith   anti-tuberculosis   drugs.   However,   data   on   the   antibiotic-taking   behavior   inrespiratory   tract   infections   inthe   communityare   lacking. 7 With   the   use   of    MEMS   wepreviously   observed   thatadherence   toantibiotic   regimens   in   respiratory   infections   de-creased   with   an   increase   in   the   number   of    daily   doses. 8 Simple   questions   are   the   most   commonly   used   measures   of treatment   adherence   inmedical   consultation.   The   simplestquestion   isasking   if    the   patient   has   taken   the   treatment   asrequested.   Physicians   assume   that   patients   provide   honestanswers   and   we   usually   believe   their   responses.   However, International    Journal   of    Infectious   Diseases   17   (2013)   e168–e172 AR    T   I   CL    E   I   N   F   O  Articlehistory: Received   28   May   2012 Receivedinrevisedform18    July2012 Accepted   26   September   2012 Corresponding    Editor:   William   Cameron,Ottawa,   Canada Keywords: Medication   adherenceMedication   event   monitoring   systemAntibacterial   agentsRespiratory   tract   infectionsPharyngitis SU   M   M   A   RY  Objectives: To   assessthedifferenttypesofantibiotic-takingbehaviorandtocompareself-reportedwithobjectivelymeasuredadherenceto   antibioticregimensinrespiratoryinfections. Methods: ThiswasaprospectivestudyofpatientswithsuspectedbacterialpharyngitisandlowerrespiratorytractinfectionsrecruitedfromfiveprimarycareclinicsinCatalonia.Adherencetovariousantibioticregimens   was   assessedbytheMedicationEventMonitoringSystem(MEMS),   whichrecordedeveryopeningofthe   patient’sbottleoftablets,and   aself-reportedadherencequestion.The   outcomevariableswereantibiotic-takingadherence,correctdosing,and   timing   adherence. Results: Atotal   of428patientswereincludedintheanalysis.Five   typesofantibioticusebehaviorwereobserved:excellentadherence(130patients,30.4%),acceptableadherenceover   time(53;12.4%),decliningadherenceover   time(123;28.7%),non-adherenceto   correctdosing(108;25.2%),andunacceptableadherence(14;3.3%).Excellentadherencewas   significantlyassociatedwiththenumberof dailydosesofantibioticandantibioticduration.A   totalof254patientsreportednever   forgettingtotaketheantibiotic(59.3%),achievinganegativepredictivevalueof100%anda   positivepredictivevalueof 51.2%. Conclusions: Outpatientswithrespiratoryinfectionstreatedwithantibioticsshowedpoor   adherenceoutcomes.Self-reportedadherencewasremarkablyhigherthanthatobservedwiththe   useofMEMS   andfailedto   predicttruepatientadherence.  2012   InternationalSocietyfor   InfectiousDiseases.PublishedbyElsevierLtd.Allrightsreserved. *Corresponding   author.   Tel.:   +34   977247211;   fax:   +34   977248459. E-mail   address:   (C.Llor). Contents   lists   available   at   SciVerse   ScienceDirect InternationalJournalofInfectiousDiseases jou   rnalho   mepag   e:   w   ww.elsevier   .co   m/locate/ijid 1201-9712/$36.00   –   see   front   matter    2012   International   Society   for   Infectious   Diseases.   Published   by   Elsevier   Ltd.   All   rights   reserved.  Author's personal copy self-reportedquestionsmayoftenprovide   inflated   estimatesof adherencebehavior. 9 Theuseof    a   non-judgmental,   non-threat-ening   approachis   therefore   recommended,   precedingthequestionwitha   remark   suchas   the   following:‘‘Peopleoftenhavedifficulty   taking   theirpillsfor   onereason   oranother’’,beforeaskingif    the   patienthas   missed   anydose. 10 Theuseof    thisapproachdecreases   theoverestimationof    true   adherenceinchronic   disorders,   butthebenefit   of    thisin   acuteconditions   suchasrespiratory   tracts   infections   remains   unanswered.Inthecurrent   studywe   aimed   to   assess   thedifferent   typesof    antibioticusebehavior   among   patients   with   respiratory   tract   infectionsandto   compare   the   performance   of    a   self-reported   adherencequestionwith   objectively   measuredadherence   of    antibioticregimens   in   theseinfections. 2.   Methods We   performeda   prospective,observational   studyin   fivegeneral   medicine   outpatient   clinics   from2003   to2008   inCatalonia,Spain.   We   recruitedpatients   aged18yearsorolderpresenting   totheprimary   care   practicewith   uncomplicated,acute( < 7   days),   suspectedbacterialpharyngitis   andlowerrespiratorytract   infections.We   excluded   patients   whohadreceivedprevious   treatmentwith   antibiotics,thosewhopre-sentedcriteria   forhospitalization,   thosewith   anyconditionrequiringtheaidofother   persons   fordrugadministration,   andthose   with   hypersensitivity   toantibiotics.   Thepatients   weretreatedwith   differentantibiotic   regimens   previously   includedintheMEMS   (AardexGroupLtd,   Zug,Switzerland)   containers.Thephysiciansdecided   whichof    theseantibiotic   treatments   wastobeadministered.Before   the   initiation   of    the   study,   the   Spanish   health   authoritieswere   informed   about   its   characteristics   and   how   itwas   to   beconducted.   Spanish   legislation   at   the   time   of    the   study   determinedthat   institutional   review   board   approval   was   not   required   forobservational   studies.   However,   the   patients   gave   informedconsent   to   participate   ina   study   on   the   rational   use   of    antibiotics.They   were   provided   with   complete   information   about   thecharacteristics   of    the   study   and   their   participation,   but   were   notinformed   at   that   time   about   the   future   assessment   of    adherence   toavoid   bias   intheresults.   When   they   returned   to   the   clinic,   thephysician   collected   the   MEMS   container   and   self-reported   adher-ence   was   evaluated   by   means   of    the   following   question:   ‘‘Wealmostalways   forget   totake   all   of    the   pills,   didyou   ever   forget   totakeany?’’   Patients   were   fully   informed   about   the   results,   andpermission   was   requested   toinclude   these   data   anonymously   inthecurrent   report.   All   the   data   included   in   the   database   wereencoded   to   ensure   confidentiality.   The   data   contained   inthemicroprocessors   were   transferred   to   the   computer   and   processedwith   PowerView   program   v.   1.3.2.   (Aardex   Ltd).   Multiple   openingsof    the   container   within   aperiod   of    less   than15   min   were   notcounted.  2.1.Adherence    parameters Three   different   outcome   measures   were   taken   into   account:   (1)‘Taking   adherence’,   calculated   asthepercentage   of    times   thecontainer   was   opened   during   the   course   of    the   treatment,   relatedto   the   total   number   of    pills   included   in   the   container.   Good   takingadherence   was   considered   when   itwas   greater   than   80%.(2)‘Correct   dosing’,   calculated   asthe   number   of    days   on   which   thepatient   opened   the   container   at   least   the   prescribed   number   of times,   that   is,   at   least   three   times   for   those   assigned   tothe   threetimes-daily   antibiotics,   twice   for   patients   treated   with   twice-dailyregimens,   and   once   for   those   receiving   once-daily   antibioticcourses.   For   twice-   and   three   times-daily   regimens,   dosing   on   day   1may   berestricted   due   tothe   late   start   of    treatment   (after   visitingthephysician),   and   this   has   tobetaken   into   account.   Good   correctdosingwas   considered   when   itwas   greater   than   80%.   (3)   ‘Timingadherence’,   indicating   whether   the   opening   of    the   containercoincided   with   the   times   recommended:   intervals   of    8+4   h   duringat   least80%   of    the   threetimes-daily   courses   of    antibiotics,   12+6   hintervals   during   at   least   80%   of    the   antibiotic   course   for   the   twice-dailyantibiotics,   and   24+12   h   intervals   during   at   least   80%   of    theantibiotic   course   for   the   once-daily   antibiotics.Excellent   adherence   was   defined   when   these   three   adherenceoutcomes   were   good.  2.2.   Statistical   analysis Descriptive   statistics   were   used   todescribe   the   differentadherence   parameters   observed   inthis   study.   We   used   Chi-squaretests   to   compare   proportions.   The   sensitivity,   specificity,   andpositive   and   negative   predictive   values   of    the   self-reportedadherence   question   were   determined   with   a   two-way   contingencytable,using   the   adherence   parameters   provided   by   MEMS   asthegoldstandard.   Alogistic   regression   model   was   constructed   toidentifyvariables   significantly   and   independently   associated   withexcellent   adherence.   The   variables   were   included   inthe   model   if theywere   associated   with   a   high   score   with   a    p -value   of    < 0.10.Variables   were   eliminated   from   the   model   using   the   stepwiseautomatic   variable   screening   method,   the   alpha   thresholds   forinclusion   and   exclusion   being   set   at   0.20.   Statistical   significancewas   accepted   at    p <   0.05. 3.   Results A   total   of    481   patients   were   recruited.   The   self-reportedadherence   question   was   not   registered   for   37   patients.   Further-more,   seven   antibiotic   treatment   failures   were   observed   requiringachange   inantimicrobial   treatment,   and   the   adherence   questionwas   not   evaluated   inthese   cases.   Seven   patients   didnot   return   theMEMS   container   and   two   more   refused   to   give   consent   (Figure   1).Ofthe   428   patients   with   complete   information   available   foranalysis,   236   (55.1%)   received   antibiotics   three   times   daily,   151received   twice-daily   antibiotic   regimens,   and   the   remaining   41patients   received   once-daily   antibiotic   schedules.   The   differentantibiotics   used   are   described   inFigure   1.Atotal   of    251   patients(58.6%)   were   diagnosed   with   a   lower   respiratory   tract   infectionand   the   remaining   177   patients   were   diagnosed   with   suspectedbacterial   pharyngitis.   The   mean   age   of    all   thepatients   was   of 47.1  21.2years,   and   231   were   females   (54.0%). Atotal   of    265   patients   opened   the   vial   at   least   80%   of    the   times(61.9%),   146   presented   correct   dosing   adherence   (34.1%),   and   165achieved   good   timing   adherence   for   at   least   80%   of    the   antibioticcourse   (38.6%).   Five   patterns   of    antibiotic   taking   behavior   wereobserved   inthis   study:   130patients   (30.4%)   achieved   80%   of    all   theadherence   outcomes   and   therefore   presented   excellent   adherence.Another   53   patients   (12.4%)   missed   only   one   dose   for   achievingexcellent   adherence   and   presented   a   relatively   acceptable   adher-ence   during   the   antibiotic   course.   Atotalof    123   patients   (28.7%)presented   declining   adherence   over   time   with   good   correct   dosingatthe   beginning   of    the   antibiotic   course   followed   by   a   reduction   inthedaily   doses   along   the   remainder   of    the   course   until   the   end.Thirteen   of    these   patients   (10.6%)   abruptly   stopped   taking   thetablets   inthefirst   half    of    the   medication   course.   Atotal   of    108patients   (25.2%)   presented   non-adherence   to   consistent   correctdosingover   time   and   14   (3.3%)   presented   an   unacceptableadherence   pattern,   with   incorrect   dosing   and   a   further   decline.Theadherence   parameters   were   consistently   worse   with   threetimes-daily   antibiotic   regimens   and   better   with   once-daily   courses(  p   <   0.001)   (Table   1). C.   Llor    etal.    /    International    Journal   ofInfectious   Diseases   17    (2013)   e168–e172   e169  Author's personal copy Candidate   variables   included   in   the   multivariate   regressionanalysis   were   the   patient   characteristics   (age,   gender,   presence   of high   blood   pressure,   dyslipidemia,   ordiabetes   mellitus,   smokingstatus,   and   retired   or   not)   and   antibiotic-related   variables   (dailydoses,   duration,   and   presence   of    adverse   effects).   Excellentadherence   was   significantly   associated   with   thenumber   of    dailydoses   of    the   antibiotic   (odds   ratio   (OR)   0.22,   95%   confidenceinterval   (CI)   0.15–0.32)   and   antibiotic   duration   (OR    0.77,   95%   CI0.61–0.96).A   total   of    254   patients   answered   the   self-reported   adherencequestion   (good   self-reported   adherence)   negatively   (59.3%).   Theremaining   patients   answered   either   affirmatively   orelicited   anunclear   response.   A   total   of    304   patients   were   correctly   identifiedbythis   approach   (71.0%).   All   patients   presenting   excellentadherence   by   means   of    the   MEMS   determination   reported   neverforgetting   to   take   their   medications.   On   the   other   hand,   the   answerwasnegative   in   only   124   of    the   298patients   who   did   not   presentexcellent   adherence   (41.6%).   As   shown   inTable   2,   the   negativepredictive   value   of    the   question   was   100%,   with   a   positivepredictive   value   of    51.2%.   Table   3describes   the   different   responsesdepending   on   the   type   of    antibiotic   used. 4.Discussion This   study   compared   self-reporting   with   objectively   measuredmedication   adherence   using   an   evidence-based   cut-off    point   inthesame   study   population.   The   main   result   of    this   study   is   thatmedication   adherence   objectively   measured   by   MEMS   was   verypoor   since   only   30%   of    the   patients   presented   excellent   adherence.Furthermore,   the   use   of    self-reported   adherence   remarkablyoverestimated   the   true   adherence.Fiveadherencetypeswere   identified   in   this   studybymeansof theMEMS   method:excellentadherence,relativelyconsistentadherenceover   time   definedasthosepatientswho   missed   onlyonedose   forachievingexcellent   adherence,decliningadherenceovertime,non-adherence   to   correctdosing,andunacceptableadherence.In   a   qualitativesemi-structured   interview   studyof    46people,   Hawkings   et   al. 11 reported   sixdifferent   types   of    antibioticuserbehavior:   thosewho   alwaystookantibiotics   asprescribed,couldnot   takedosesbecauseof    work,   childcare,or   socialconstraints,frequentlyforgot   doses,believed   it   made   sensetostop   taking   antibiotics   astheystartedto   getbetter,   activelysoughttolimitantibiotic   usebecause   theybelievedtheir   ownbodiesbecameusedto   them   or   becauseantibioticsareunnatural,anddeliberatelyplannedto   stop   early   soastohaveanantibioticsupplyforself-use   inthefuturetoavoid   thechallenges   of consultingandobtaining   antibiotics   in   primarycare.   In   theHawkingsstudy,over   a   thirdof    the   respondents   reportedthattheyalways   tookantibiotics   asdirectedbytheclinician   orpharmacist.   The   results   of    the   present   studyclearly   indicatethatless   thana   thirdofthe   patients   tookthe   tablets   asrequested.Weusedthe   same   cut-off    point   recommended   bythe   previousauthors,i.e.,at   least   80%ofall   the   adherence   parametersevaluated. 12 However,   onlya   little   more   than   12%ofthepatientsnearlyachieved   thesegoalssincetheyonlymissed   onedose,   andin   all   thesecases   the   patients   statedthat   theywere   adherent   by a Patients were recommended to take two tablets   every 12 hours 481 Eligible paents(203 pharyngis and 278 LRTI)428   Paents included in the analysis(177 pharyngis and 251 LRTI)37 Paents were not asked about their adherence7   Paents failed and the anbioc was changed;   they were not asked7   Paents did not return the MEMS container2   Paents refused to give informed consent at   the follow-up visit41   Once-daily anbioc regimens: •   17   Levofloxacin 500 mg 10 tablets •   12Azithromycin 500 mg 3 tablets •   9   Moxifloxacin 400 mg 7 tablets •   3   Moxifloxacin 400 mg 5 tablets151   Twice-daily anbioc regimens: •   57Amoxicillin/clavulanate 1000/125   mg 28 tablets a •   73   Phenoxymethylpenicillin 400 mg 30 tablets a •   15   Cefditoren 200 mg 20 tablets •   6   Cefuroxime acetyl 500 mg 20 tablets 236Three mes-daily anbioc regimens • 129Amoxicillin/clavulanate 500/125   mg 24 tablets107 Amoxicillin 500 mg 24 tablets • Figure   1.   Flow   of    patients   through   the   study.  Table   1 Types   ofantibiotic-taking   behavior   depending   onthe   number   of    daily   doses   in   the   antibiotic   regimenOnce-daily   antibioticregimenTwice-dailyantibiotic   regimenThree   times-dailyantibiotic   regimenTotalExcellent   adherence   34   (82.9)   77   (51.0)   19   (8.1)   130   (30.4)Acceptableadherence   over   time   5   (12.2)   23   (15.2)   25   (10.6)   53   (12.4)Decliningadherence   over   time   2   (4.9)   34   (22.5)   87   (36.8)   123   (28.7)Non-adherence   to   consistent   correct   dosing   0(0)16   (10.6)   92   (39.0)   108   (25.2)Unacceptable   adherence   0(0)1   (0.7)   13   (5.5)   14   (3.3)Total41   151   236   428 C.Llor    etal.    /    International    Journal   ofInfectious   Diseases   17    (2013)   e168–e172 e170  Author's personal copy meansof    the   self-reportedadherencequestion.Thisdemon-stratesthat   the   rationale   for   choosing   a   cut-off    of    80%   to   defineadherencein   order   to   differentiatebetweenadherenceandnon-adherenceis   arbitrarilychosen,   and   despitecommonly   beingusedbyhealthprofessionals,   it   isnotshared   bypatientswho   havetotake   themedication.In   our   study,   more   thanhalfof    the   patients   with   non-adherenceto   consistently   correct   dosing   and   declining   adherence   overtime   admitted   to   have   forgotten   some   doses.   Incorrect   dosing   ismorelinked   to   unintentional   non-adherence,   since   it   is   influencedby   the   constraints   of    work,   child   care,   school,   and   simplyforgetting,   while   declining   adherence   over   time   is   more   associatedwith   intentional   non-adherence,   and   it   ismore   likely   that   thesepatients   are   not   aware   of    the   consequences   of    stopping   early.Despite   being   unintentional,   patients   who   systematically   forgot   totakea   pill   every   day   were   more   aware   of    being   non-adherent   thanthose   who   had   apriori   intentional   non-adherence,   since   nearly   80%ofthe   former   respondents   admitted   having   forgotten   totake   somedosesvs.   60%   of    the   latter   who   did   so.   Patients   who   stopped   takingantibiotics   as   they   started   to   get   better,   those   who   limited   their   usebecause   of    some   misbeliefs,   and   those   who   planned   tohave   anantibiotic   supply   at   home   are   supposed   tohave   excellentadherence   at   the   beginning   of    treatment   and   a   deterioratingadherence   after   some   days,   but   curiously   only   10%   of    these   patientsstopped   the   treatment   too   soon.   The   remaining   90%   of    thesepatients   actually   decreased   the   frequency   of    the   doses   after   aperiod   of    perfect   adherence.   This   probably   means   that   most   of    thepatients   with   declining   adherence   over   time   were   aware   thattaking   the   antibiotics   was   necessary   and   felt   guilty   about   stoppingtotake   them.The   number   of    adherence   types   is   likely   to   varywith   the   studypopulation   under   analysis.   In   studies   involving   long-term   condi-tions,   other   typologies   of    medication   use   behavior   have   beendetected,   such   asimproving   adherence   over   time.   For   example,Knafl   et   al. 13 identified   10   adherence   types   for   subjects   with   HIV   onantiretroviral   medications,   including   seven   relatively   consistent,onedeteriorating,   and   two   improving   adherence   types.   However,with   treatment   lengths   of    upto10   days,   such   asthescheduleaddressed   inour   study,   thenumber   of    medication   use   behaviors   ismuch   lower,   with   three   typologies   being   the   most   common   –oneof    good   adherence   and   two   basic   patterns   of    non-adherence.   Weonly   included   outpatients   with   relatively   benign   acute   conditionsand   this   fact   might   explain   why   somany   patients   failed   topresentexcellent   adherence   behavior.Another   conclusion   of    this   study   is   that   medication   adherencemeasured   by   the   self-reported   adherence   question   was   remark-ably   higher   than   that   objectively   measured   by   MEMS,   indicatingthatself-reporting   seems   tobeprone   tooverestimating   of    trueadherence.   To   our   knowledge   this   is   thefirst   time   that   a   self-reported   question   has   been   used   toreport   the   adherence   of patientsinacute   infectious   diseases.   The   main   explanation   thatmay   underlie   the   difference   between   self-reported   and   ‘true’adherence   is   that   patients   may   not   want   toadmit   that   they   arenon-adherent,   and   therefore   reported   adherence.   Self-reportedadherence   is   able   todetect   non-adherence   when   the   patientreports   forgetting   some   doses,   since   a   patient   who   admits   nothaving   forgotten   any   dose   can   beeither   adherent   ornon-adherentwith   respect   totiming   and   dosing.   Therefore   this   screeningquestion   has   little   value   inclinical   practice.There   were   some   limitations   to   this   study.   The   adherence   dataanalyzed   were   collected   electronically   using   MEMS   caps.   Capopenings   do   not   always   necessarily   correspond   to   actual   medica-tion-taking.   Patients   may   sometimes   have   removed   multiple   dosesatone   cap   opening   inorder   to   put   them   in   pill   boxes,   in   which   casethecap   openings   underestimate   actual   adherence.   Moreover,multiple   openings   of    the   container   within   a   period   of    less   than15   min   werenot   counted.   Neither   can   we   ensure   that   whenphenoxymethylpenicillin   and   the   new   pharmacokinetically   en-hanced   formulation   of    amoxicillin/clavulanate   wereadministered,the   patients   took   two   tablets   at   the   same   time.   The   diagnosis   wasclinical   and   therefore   itcannot   be   guaranteed   that   all   the   episodesincluded   were   actually   bacterial   infections;   however   this   couldhave   happened   equally   in   all   the   treatment   regimens   and   shouldnot   be   directly   related   to   theadherence   totreatment.   Nonetheless,webelieve   that   the   electronic   method   usedinthis   study,   the   largesample   studied,   and   the   fact   that   the   patients   were   not   informed   astothe   real   objective   of    the   study   until   thesecond   visit,undoubtedly   constitute   the   greatest   strengths   of    this   study.In   conclusion,   less   than   halfof    the   patients   treated   with   regularcourses   of    antibiotics   presented   excellent   or   acceptable   adherence.Approximately   one   in   four   patients   presented   non-adherence   tocorrect   dosing   and   approximately   one   infour   presented   decliningadherence   over   time.   The   adherence   outcomes   were   consistentlyand   significantly   worse   with   three   times-daily   antibiotic   schedules  Table   2 Validity   of    the   self-reported   adherence   questionResponse   to   the   self-reported   adherence   question a True   adherence   TotalExcellent   Not   excellentNegative   response   (good   self-reported   adherence)   130124   254Affirmative   or   unclear   response   (not   good   self-reported   adherence)   0174   174Total130298   428 a Wealmost   always   forget   to   take   all   of    the   pills,   did   you   ever   forget   to   take   any?Sensitivity:   130/130   =   100%;   specificity:   174/298   =   58.4%;   positive   predictive   value:   130/254   =51.2%;   negative   predictive   value:   174/174   =   100%.  Table   3 Response   to   the   self-reported   adherence   question   depending   on   the   antibiotic-taking   behaviorAntibiotic-taking   behavior   Response   to   the   self-reported   adherence   question a TotalNegative   response   Affirmative   orunclear   responseExcellentadherence   130   (100)   0   (0)   130Acceptable   adherence   over   time   53   (100)   0   (0)   53Declining   adherence   over   time   47   (38.2)   76   (61.8)   123Non-adherence   to   consistently   correct   dosing   23   (19.8)   85   (78.7)   108Unacceptable   adherence   1(7.1)   13   (92.9)   14Total   254   (59.3)   174   (40.7)   428 a Wealmost   always   forget   to   take   all   the   pills,   did   you   ever   forget   totake   any? C.Llor    etal.    /    International    Journal   ofInfectious   Diseases   17    (2013)   e168–e172   e171
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks