49
ISBN 82-8082-039-6 IN 0000-2069 Tittel: Introduction to Drug Utilization Research Opplag: 2 500 Design: Grete Søimer Foto: Photo Alto/Laurent Hamels Trykk: Nordberg Aksidenstrykkeri AS Introduction to Drug Utilization Research 2003 2003 Introduction to Drug Utilization Research

Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

  • Upload
    others

  • View
    1

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

ISBN 82-8082-039-6IN 0000-2069

Tittel: Introduction to Drug UtilizationResearchOpplag: 2 500Design: Grete SøimerFoto: Photo Alto/Laurent HamelsTrykk: Nordberg Aksidenstrykkeri AS

Intro

du

ction

to D

rug

Utilizatio

n R

esearch 2003

2003

Introduction to DrugUtilization Research

Page 2: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

World Health Organization

WHO International Working Group for Drug Statistics Methodology

WHO Collaborating Centre for Drug Statistics Methodology

WHO Collaborating Centre for Drug Utilization Research and Clinical

Pharmacological Services

Introduction toDrug Utilization Research

Page 3: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

WHO Library Cataloguing-in-Publication Data

Introduction to drug utilization research / WHO International Working Group for DrugStatistics Methodology, WHO Collaborating Centre for Drug Statistics Methodology,WHO Collaborating Centre for Drug Utilization Research and Clinical PharmacologicalServices.

1. Drug utilization 2. Research 3. Manuals I.WHO International Working Group forDrug Statistics Methodology II.WHO Collaborating Centre for DrugStatistics Methodology III.WHO Collaborating Centre for Drug Utilization Researchand Clinical Pharmacological Services

ISBN 92 4 156234 X (NLM classification: WB 330)

© World Health Organization 2003All rights reserved. Publications of the World Health Organization can be obtained from Marketing and Dissemination, World Health Organization, 20 Avenue Appia,1211 Geneva 27, Switzerland (tel: +41 22 791 2476; fax: +41 22 791 4857; email: [email protected]). Requests for permission to reproduce or translate WHOpublications – whether for sale or for noncommercial distribution – should be addressedto Publications, at the above address (fax: +41 22 791 4806; email:[email protected]). The designations employed and the presentation of the material in this publication do notimply the expression of any opinion whatsoever on the part of the World HealthOrganization concerning the legal status of any country, territory, city or area or of itsauthorities, or concerning the delimitation of its frontiers or boundaries. Dotted lines onmaps represent approximate border lines for which there may not yet be full agreement.

The mention of specific companies or of certain manufacturers’ products does not implythat they are endorsed or recommended by the World Health Organization in preference toothers of a similar nature that are not mentioned. Errors and omissions excepted, the namesof proprietary products are distinguished by initial capital letters.

The World Health Organization does not warrant that the information contained in thispublication is complete and correct and shall not be liable for any damages incurred as aresult of its use.

Printed in Oslo, Norway, 2003

2

Page 4: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

3

Page 5: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

Preface: Drug utilization research - the early work ...........................................................................6

Chapter 1: What is drug utilization research and why is it needed? ...............................................81.1 Definition and domains............................................................................................................81.2 Why drug utilization research? ................................................................................................9

1.2.1 Description of drug use patterns ....................................................................................91.2.2 Early signals of irrational use of drugs .......................................................................101.2.3 Interventions to improve drug use - follow-up ............................................................101.2.4 Quality control of drug use...........................................................................................10

1.3 Drug utilization studies and drug policy decisions................................................................111.4 General reading......................................................................................................................12

Chapter 2: Types of drug use information ........................................................................................132.1 Drug-based information .........................................................................................................13

2.1.1 Level of drug use aggregation .....................................................................................132.1.2 Indication .....................................................................................................................132.1.3 Prescribed daily doses .................................................................................................14

2.2 Problem or encounter-based information ..............................................................................152.3 Patient information.................................................................................................................162.4 Prescriber information ...........................................................................................................162.5 Types of drug utilization study ..............................................................................................172.6 Drug costs ..............................................................................................................................172.7 General reading......................................................................................................................182.8 Exercises ................................................................................................................................19

Chapter 3: Sources of data on drug utilization ................................................................................203.1 Large databases ......................................................................................................................203.2 Data from drug regulatory agencies ......................................................................................203.3 Supplier (distribution) data ....................................................................................................203.4 Practice setting data ...............................................................................................................21

3.4.1 Prescribing data ............................................................................................................213.4.2 Dispensing data ............................................................................................................223.4.3 Aggregate data .............................................................................................................223.4.4 Over-the-counter and pharmacist-prescribed drugs ....................................................223.4.5 Telephone and Internet prescribing .............................................................................22

3.5 Community setting data .........................................................................................................233.6 Drug use evaluation ...............................................................................................................233.7 General reading......................................................................................................................243.8 Exercises ................................................................................................................................24

Chapter 4: Economic aspects of drug use (pharmacoeconomy) ....................................................264.1 Introduction ............................................................................................................................264.2 Cost-minimization analysis....................................................................................................264.3 Cost-effectiveness analysis ....................................................................................................264.4 Cost-utility analysis ...............................................................................................................274.5 Cost-benefit analysis ..............................................................................................................274.6 General reading......................................................................................................................284.7 Exercises ................................................................................................................................28

Contents

Page 6: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

Chapter 5: Drug classification systems ..............................................................................................335.1 Different classification systems .............................................................................................335.2 The ATC classification system...............................................................................................335.3 Ambivalence towards an international classification system ................................................355.4 Implementation of the ATC/DDD methodology ...................................................................365.5 General reading......................................................................................................................365.6 Exercises ................................................................................................................................37

Chapter 6: Drug utilization metrics and their applications ............................................................386.1 The concept of the defined daily dose (DDD).......................................................................386.2 Prescribed daily dose and consumed daily dose....................................................................396.3 Other units for presentation of volume..................................................................................396.4 Cost ......................................................................................................................................396.5 General reading .....................................................................................................................406.6 Exercises ...............................................................................................................................41

Chapter 7: Solutions to the exercises .................................................................................................74

Acknowledgements...............................................................................................................................84

Page 7: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

6

The development of drug utilization researchwas sparked by initiatives taken in NorthernEurope and the United Kingdom in the mid-1960s (1, 2). The pioneering work of ArthurEngel in Sweden and Pieter Siderius in Holland(3) alerted many investigators to the importanceof comparing drug use between different count-ries and regions. Their demonstration of theremarkable differences in the sales of antibioticsin six European countries between 1966 and1967 inspired WHO to organize its first meetingon «Drug consumption» in Oslo in 1969 (4).This led to the constitution of the WHOEuropean Drug Utilization Research Group(DURG).

The pioneers of this research understood that acorrect interpretation of data on drug utilizationrequires investigations at the patient level. It

became clear that we need to know the answersto the following questions:• why drugs are prescribed;• who the prescribers are;• for whom the prescribers prescribe;• whether patients take their medicines correctly;• what the benefits and risks of the drugs are.

The ultimate goal of drug utilization researchmust be to assess whether drug therapy is rationalor not. To reach this goal, methods for auditingdrug therapy towards rationality are necessary.

The early work did not permit detailed compa-risons of the drug utilization data obtained fromdifferent countries because the source and formof the information varied between them. Toovercome this difficulty, researchers in NorthernIreland (United Kingdom), Norway and Sweden

Preface: Drug utilization research - the early work

Figure 1 Utilization of insulin and oral antidiabetic drugs in seven European countries from 1971-1980 expressedin defined daily doses (DDDs) per 1000 inhabitants per day. For comparison the prescribed daily doses (PDD)per 1000 inhabitants per day of oral antidiabetic drugs are given for Northern Ireland (UK) and Sweden for 1980(indicated with an asterisk).

Page 8: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

7

developed a new unit of measurement, initiallycalled the agreed daily dose (5) and later thedefined daily dose (DDD) (6). This unit wasdefined as the average maintenance dose ofthe drug when used on its major indication inadults. The first study used antidiabetic drugsas an example: it was found that the sum of theDDDs of insulin and oral antidiabetic drugs(about 20 DDDs per1000 inhabitants per day)roughly corresponded to the morbidity due todiabetes after correction for the number of pati-ents treated with dietary regimens alone. Amongthe first countries to adopt the DDD methodo-logy was the former Czechoslovakia (7) and thefirst comprehensive national list of DDDs waspublished in Norway in 1975 (8). Anotherimportant methodological advance was the adop-tion of the uniform anatomical therapeutic che-mical (ATC) classification of drugs (see chapter5.2). The use of standardized methodology allo-wed meaningful comparisons of drug use in different countries to be made (Fig. 1).

Drug utilization research developed quicklyduring the following 30 years and soon became arespectable subject for consideration at inter-national congresses in pharmacology, pharmacyand epidemiology. Particularly rapid develop-ments were seen in Australia (9) and LatinAmerica (10). The number of English-languagepapers on the subject listed in the Cumulativeindex medicus rose from 20 in 1973 (when theterm «drug utilization« first appeared) to 87 in1980, 167 in 1990, and 486 in 2000.

History has taught us that successful researchin drug utilization requires multidisciplinary col-laboration between clinicians, clinical pharmaco-logists, pharmacists and epidemiologists.Without the support of the prescribers, this rese-arch effort will fail to reach its goal of facilita-ting the rational use of drugs.

References

1. Wade O. Drug utilization studies - the firstattempts. Plenary lecture. In: Sjöqvist F,Agenäs I. eds. Drug utilization studies: implica-tions for medical care. Acta Medica

Scandinavica, 1984, Suppl. 683:7-9.

2. Dukes MNG. Development from Crooks to thenineties. In: Auditing Drug Therapy.Approaches towards rationality at reasonablecosts. Stockholm, Swedish PharmaceuticalPress, 1992.

3. Engel A, Siderius P. The consumption of drugs.Report on a study 1966-1967. Copenhagen,WHO Regional Office for Europe, 1968 (EURO3101).

4. Consumption of drugs. Report on a symposiumin Oslo 1969. Copenhagen, WHO RegionalOffice for Europe, 1970 (EURO 3102).

5. Bergman U, et al. The measurement of drugconsumption. Drugs for diabetes in NorthernIreland, Norway, and Sweden. EuropeanJournal of Clinical Pharmacology, 1975,8:83-89.

6. Bergman U et al., eds. Studies in drug utilizati-on. Methods and applications. Copenhagen,WHO Regional Office for Europe, 1979 (WHORegional Publications, European Series No. 8).

7. Stika L et al. Organization of data collection inCzechoslovakia. In: Bergman U et al., eds.Studies in drug utilization. Methods and appli-cations. WHO Regional Office for Europe,Copenhagen, 1979 (WHO Regional PublicationsEuropean Series No. 8) pp.125-136.

8. Baksaas Aasen I et al. Drug dose statistics, listof defined daily doses for drugs registered inNorway. Oslo, Norsk Medicinal Depot, 1975.

9. Hall RC. Drug use in Australia. In: Sjöqvist F,Agenäs I, eds. Drug utilization studies:Implications for medical care. Acta MedicaScandinavica, 1983, Suppl. XXX:79-80.

10. Drug Utilization Research Group, LatinAmerica. Multicenter study on self-medicationand self-prescription in six Latin Americancountries. Clinical Pharmacology andTherapeutics, 1997, 61:488-493.

11. Bergman U, Sjöqvist F. Measurement of drugutilization in Sweden: methodological and clini-cal implications. Acta Medica Scandinavica,1984, Suppl 683:15-22.

Page 9: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

1.1. Definition and domains

Drug utilization; pharmacoepidemiology; pharmacosurveillance; pharmacovigilance

• Drug utilization research was defined byWHO in 1977 as «the marketing, distribution,prescription, and use of drugs in a society, withspecial emphasis on the resulting medical, socialand economic consequences». Since then, anumber of other terms have come into use and itis important to understand the interrelationshipsof the different domains. • Epidemiology has been defined as «the studyof the distribution and determinants of health-related states and events in the population, andthe application of this study to control of healthproblems».• Pharmacoepidemiology applies epidemiologi-cal methods to studies of the clinical use ofdrugs in populations. A modern definition ofpharmacoepidemiology is: «the study of the useand effects/side-effects of drugs in large numbersof people with the purpose of supporting therational and cost-effective use of drugs in thepopulation thereby improving health outcomes».• Pharmacosurveillance and pharmacovigi-lance are terms used to refer to the monitoringof drug safety, for example, by means of spon-taneous adverse-effect reporting systems, case-control and cohort studies.

Pharmacoepidemiology may be drug-oriented,emphasizing the safety and effectiveness of indi-vidual drugs or groups of drugs, or utilization-oriented aiming to improve the quality of drugtherapy through pedagogic (educational) inter-vention. Drug utilization research may also bedivided into descriptive and analytical studies.The emphasis of the former has been to describepatterns of drug utilization and to identify pro-blems deserving more detailed studies.Analytical studies try to link data on drug utili-zation to figures on morbidity, outcome of treat-ment and quality of care with the ultimate goalof assessing whether drug therapy is rational or

not. Sophisticated utilization-oriented pharma-coepidemiology may focus on the drug (e.g.dose-effect and concentration-effect relationships),the prescriber (e.g. quality indices of the prescrip-tion), or the patient (e.g. selection of drug anddose, and comparisons of kidney function, drugmetabolic phenotype/genotype, age, etc.).

Drug utilization research is thus an essentialpart of pharmacoepidemiology as it describes theextent, nature and determinants of drug exposure.Over time, the distinction between these twoterms has become less sharp, and they are some-times used interchangeably. However, whiledrug utilization studies often employ varioussources of information that focus on drugs (e.g.aggregate data from wholesale and prescriptionregisters) the term epidemiology implies definedpopulations in which drug use can be expressedin terms of incidence and prevalence (see chapter 1.2.1).

Together, drug utilization research and pharma-coepidemiology may provide insights into the fol-lowing aspects of drug use and drug prescribing.

• Pattern of use: This covers the extent and pro-files of drug use and the trends in drug use andcosts over time.• Quality of use: This is determined using auditsto compare actual use to national prescriptionguidelines or local drug formularies.1 Indices ofquality of drug use may include the choice ofdrug (compliance with recommended assort-ment), drug cost (compliance with budgetaryrecommendations), drug dosage (awareness ofinter-individual variations in dose requirementsand age-dependence), awareness of drug inter-actions and adverse drug reactions, and the pro-portion of patients who are aware of or unawareof the costs and benefits of the treatment.• Determinants of use: These include user cha-racteristics (e.g. sociodemographic parametersand attitudes towards drugs), prescriber characte-ristics (e.g. speciality, education and factorsinfluencing therapeutic decisions) and drug cha-racteristics (e.g. therapeutic properties and affor-dability).

8

Chapter 1: What is drug utilization research and why is it needed?

[ ]

1 An audit in drug use was defined by Crooks (1979) as an examination of the way in which drugs are used in clinical practice carried out at intervalsfrequent enough to maintain a generally accepted standard of prescribing.

Page 10: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

• Outcomes of use: These are the health outcomes (i.e. the benefits and adverse effects) andthe economic consequences.

The initial focus of pharmacoepidemiology wason the safety of individual drug products (phar-macosurveillance), but it now also includes stu-dies of their beneficial effects. The driving forcebehind this development was a growing awa-reness that the health outcomes of drug use inthe rigorous setting of randomized clinical trialsare not necessarily the same as the health outco-mes of drug use in everyday practice. The clini-cal trials needed to obtain marketing authori-zation for new drugs involve limited numbers ofcarefully selected patients, who are treated andfollowed-up for a relatively short time in strictlycontrolled conditions. As a result, such trials donot accurately reflect how drug use will affecthealth outcomes in everyday practice under every-day circumstances. Pharmacoepidemiologicalstudies often make useful contributions to ourknowledge about effectiveness and safety, because,unlike clinical trials, they assess drug effects inlarge, heterogeneous populations of patients overlonger periods.

Drug utilization research also provides insightinto the efficiency of drug use, i.e. whether acertain drug therapy provides value for moneyand the results of such research can be used tohelp to set priorities for the rational allocation ofhealth care budgets.

1.2 Why drug utilization research?

Description of drug use pattern; early signals of irrational use of drugs; interventions to improve drug use; quality control cycle; continuous quality improvement

The principal aim of drug utilization research isto facilitate the rational use of drugs in popu-lations. For the individual patient, the rationaluse of a drug implies the prescription of a well-documented drug at an optimal dose, togetherwith the correct information, at an affordable

price. Without a knowledge of how drugs arebeing prescribed and used, it is difficult to initiatea discussion on rational drug use or to suggestmeasures to improve prescribing habits.Information on the past performance of prescri-bers is the linchpin of any auditing system.

Drug utilization research in itself does notnecessarily provide answers, but it contributes torational drug use in important ways as describedbelow.

1.2.1 Description of drug use patternsDrug utilization research can increase our under-standing of how drugs are being used as follows.

• It can be used to estimate the numbers of pati-ents exposed to specified drugs within a giventime period. Such estimates may either refer toall drug users, regardless of when they started touse the drug (prevalence), or focus on patientswho started to use the drug within the selectedperiod (incidence). • It can describe the extent of use at a certainmoment and/or in a certain area (e.g. in a coun-try, region, community or hospital). Such des-criptions are most meaningful when they formpart of a continuous evaluation system, i.e. whenthe patterns are followed over time and trends indrug use can be discerned.• Researchers can estimate (e.g. on the basis ofepidemiological data on a disease) to what extentdrugs are properly used, overused or underused.• It can determine the pattern or profile of druguse and the extent to which alternative drugs arebeing used to treat particular conditions.• It can be used to compare the observed patternsof drug use for the treatment of a certain diseasewith current recommendations or guidelines. • It can be used in the application of quality indi-cators to patterns of drug utilization. An exam-ple is the so-called DU90% (drug utilization90%), a further development of the «top-ten»list.

The DU90% segment reflects the number ofdrugs that account for 90% of drug prescriptions

9

Page 11: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

and the adherence to local or national prescrip-tion guidelines in this segment. This generalindicator can be applied at different levels (e.g.individual prescriber, group of prescribers, hos-pitals, region or county) to obtain a rough esti-mate of the quality of prescribing. • Drug utilization data can be fed back to pre-scribers. This is particularly useful when thedrug prescribing by a particular individual canbe compared with some form of «gold standard»or best practice, and with the average prescrip-tions in the relevant country, region or area. • The number of case reports about a drug pro-blem or adverse effects can be related to thenumber of patients exposed to the drug to assessthe potential magnitude of the problem. If it ispossible to detect that the reaction is more com-mon in a certain age group, in certain conditionsor at a given dose level, improving the informa-tion on indications, contraindications and appropriate dosages may be sufficient to ensuresafer use and avoid withdrawal of the drug fromthe market.

1.2.2 Early signals of irrational useof drugsDrug utilization research may generate hypothe-ses that set the agenda for further investigationsas outlined below, and thus avoid prolonged irra-tional use of drugs.• Drug utilization patterns and costs betweendifferent regions or at different times may becompared. Hypotheses can be generated to formthe basis for investigations of the reasons for,and health implications of, the differencesfound. Geographical differences and changes indrug use over time may have medical, social andeconomic implications both for the individualpatient and for society, and should therefore beidentified, explained and, when necessary cor-rected.• The observed patterns of drug use can be com-pared with the current recommendations andguidelines for the treatment of a certain disease.Hypotheses can then be generated to determinewhether discrepancies represent less than opti-

mal practice, whether pedagogic interventions(education) are required or whether the guide-lines should be reviewed in the light of actualpractice. These hypotheses should apply to bothunder use and over use of drugs.

1.2.3 Interventions to improve drug use - follow-upDrug utilization research undertaken in the follo-wing ways may enable us to assess whetherinterventions intended to improve drug use havehad the desired impact.• The effects of measures taken to ameliorateundesirable patterns of drug use (e.g. provisionof regional or local formularies, informationcampaigns and regulatory policies) should bemonitored and evaluated. The researchersshould bear in mind that prescribers may haveswitched to other drugs that are equally undesi-rable. These potential alternative drugs shouldbe included in the survey to assess the fullimpact of the measure.• The impact of regulatory changes or changesin insurance or reimbursement systems shouldbe assessed using a broad survey. This is neces-sary because the total cost to society may remainthe same or may even increase if more expen-sive drugs are used as alternatives.• The extent to which the promotional activitiesof the pharmaceutical industry and the educatio-nal activities of the society affect the patterns ofdrug use should be assessed.

1.2.4 Quality control of drug useDrug use should be controlled according to aquality control cycle that offers a systematicframework for continuous quality improve-ment. The components of such a cycle are illu-strated on the next page.

After step 4, the cycle begins again with newanalyses, the setting of new targets, and so on.

The quality control cycle can be applied atmany levels, ranging from local or regional dis-cussion groups consisting of physicians, clinicalpharmacologists or pharmacists to national and

10

Page 12: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

international initiatives. An important techniquethat can be used in conjunction with this cycleis benchmarking. By comparing drug utiliza-tion data from different localities, it is oftenpossible to detect substantial differences thatrequire further evaluation, which may then leadto the identification and promotion of best prac-tice. Such comparisons will be accurate andtruthful provided that the data are collected andaggregated in a standardized, uniform way (seechapter 5).

1.3 Drug utilization studies anddrug policy decisions Many of the questions asked in drug utilizationresearch and the answers obtained are importantfor initiating and modifying a rational drug poli-cy at both national and local levels. Two suc-cessful examples of the use of such research aregiven below.

Drug use in EstoniaAn important reason for undertaking studies ofdrug use in Estonia after its independence wasthe need to make decisions on drug policy. Atthe time, no information was available in thecountry on which drugs were used (sold), or onthe quantities and there was therefore no rationalefor regulating the drug market. Moreover, in theabsence of any feedback system it was impos-sible to gauge the impact of possible futureinterventions. A national drug classification system was therefore developed for Estonia, anda reporting system from wholesalers, based onthis classification, was implemented, checked

and validated from 1992-1994. Since then,annual reviews of drug utilization have beenused to provide background information fordecisions on regulatory and reimbursement poli-cies in Estonia; two examples are describedbelow.

If physicians have high rates of inappropriateprescribing, drug regulatory authorities canrequire educational intervention or imposerestrictions on specific drugs or on practitioners.In Estonia, it was decided to stop the import anduse of some hazardous products, such as phena-cetine, older sulphonamides and pyrazolones,after clarifying and explaining the reasons forthis in the national Drug information bulletin,which is distributed free by the drug regulatoryauthority to all prescribers in Estonia.

In planning the reimbursement policies, thetotal volume of drug use in DDDs was monito-red carefully. During the 1990s, the use of pre-scription-only medicines measured as number ofDDDs per capita was less than one third of thatreported from the Nordic countries. This provedto be the result of under-treatment of certainchronic diseases (i.e. hypertension and schizo-phrenia), and therefore the decision was to incre-ase the availability and use of cardiovascular andneuroleptic drugs. Thus, the national drug usesurveys in Estonia have been used to monitor theimpact of drug regulatory activities as well as tofollow the increase in drug expenditure.Because data on drug use are only part of thebackground material relevant to the discussionsand decisions on therapeutic strategies - at boththe local and national levels - it is difficult to

11

Step 1. Plan. Analyse current situation toestablish a plan for improvment (e.g. analysecurrent prescription patterns of individualprescribers, groups of prescribers, or healthfacilities).

Step 4. Act Revise plan or implement planon large scale (e.g. guide national imple-mentation of plan).

Step 3. Check Check to see if expectedresults are obtained (e.g. evaluate whetherprescription patterns really improve).

Step 2. Do. Implement the plan on a smallscale (e.g. provide feedback on possibleoveruse, underuse or drug misuse of individual drugs or therapeutic groups).

Page 13: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

evaluate the specific influence of drug utilizationresearch on developments in drug policies. It is,however, reasonable to assume that such studieshave contributed to a more rational use of drugsin Estonia.1

Drug use in Latin AmericaThe second example is the successful work withinthe Latin American DURG, in association withthe WHO Collaborating Centre of Pharmaco-epidemiology in Barcelona, Spain.

In September 1991, health professionals fromSpain and eight Latin American countries met inBarcelona for the «First Meeting of LatinAmerican Groups for Drug Epidemiology». Itwas made clear that in most of the countriestaking part, data on drug utilization were scarceand fragmentary. Some national drug regulatoryauthorities had no access to either quantitative orqualitative data on drug consumption and reali-zed that information on patterns of drug utili-zation would be useful for designing drug policyand educational programmes about drugs.

It was agreed at this meeting to set up a Latin American network (later called DURG-LA),with the following aims:– to promote drug utilization research in Latin

American countries;– to exchange experiences and information

between the participating groups; – to use the knowledge generated to give techni-

cal advice to drug regulatory authorities and toguide teaching of pharmacology;

– to write and disseminate information aimed at improving drug use, and

– to participate in the training of health pro-fessionals in pharmacoepidemiology and thera-peutics.

Seven further DURG-LA meetings have beenheld over the subsequent ten years to promotedrug utilization research. Part of the initial coregroup participated in a first multicentre study insix Latin American countries to examine self-medication and self-prescription. The study wascarried out in a sample of pharmacies from diffe-

rent social-class districts in the catchment areasof 11 health centres.1

1.4 General readingBergman U et al. Drug utilization 90% - a sim-ple method for assessing the quality of drug pre-scribing. European Journal of ClinicalPharmacology, 1998, 54:113-118.

Crooks J. Methods of audit in drug use. In:Duchene-Marulla ZP, ed. Advances in pharma-cology and therapeutics. Proceedings of 7thInternational Congress of Pharmacology, Paris,1978. Oxford, Pergamon Press, 1979:189-195.

Diogène E et al. The Cuban experience infocusing pharmaceuticals policy to health popu-lation needs: initial results of the NationalPharmacoepidemiology Network (1996-2001).European Journal of Clinical Pharmacology,2002, in press.

Drug Utilization Research Group, LatinAmerica. Multicenter study on self-medicationand self-prescription in six Latin Americancountries. Clinical Pharmacology andTherapeutics, 1997, 61:488-493.

Dukes MNG, ed. Drug Utilization Studies:Methods and Uses. Copenhagen, WHO RegionalOffice for Europe, 1993 (WHO RegionalPublications European Series No. 45)

Einarson TR, Bergman U, Wiholm BE.Principles and practice of pharmacoepidemiolo-gy. In: Avery’s Drug Treatment, 4th ed. AdisInternational:371-392.

Figueras A et al. Health need, drug registrationand control in less developed countries - thePeruvian case. Pharmacoepidemiology and DrugSafety, 2001, 10:1-2.

Laporte JR, Porta M, Capella D. Drug utiliza-tion studies: a tool for determining the effecti-veness of drug use. British Journal of ClinicalPharmacology, 1983, 16:301-304.

McGavock H. Handbook of drug use researchmethodology 1st ed. Newcastle upon Tyne,United Kingdom Drug Utilization ResearchGroup, 2000.

Strom BL. Pharmacoepidemiology, 3rd ed.New York, J Wiley, 2000.

12

1 The information about DURG-LA was provided in a personal communication by Dr Albert Figueras and Professor Joan-Ramon Laporte,Barcelona, Spain.

Page 14: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

13

Chapter 2: Types of drug use information

Different types of drug use information arerequired depending on the problem being exami-ned. These include information about the overalluse of drugs, drug groups, individual genericcompounds or specific products. Often, infor-mation about the condition being treated, thepatient and the prescriber is also required. Inaddition, data on drug costs will be important inensuring that drugs are used efficiently and eco-nomically. These types of drug information aredescribed in detail below, together with exam-ples to illustrate the ways in which the informa-tion can be used to promote the rational use ofdrugs.

2.1 Drug-based informationKnowledge of the trends in total drug use maybe useful, but more detailed information invol-ving aggregation of data on drug use at variouslevels, and information on indications, doses anddosage regimens is usually necessary to answerclinically important questions.

2.1.1 Level of drug use aggregationThe level at which data on drug use are aggrega-ted will depend on the question being asked.For example, the question might concern therelative use of drug groups in the treatment ofhypertension. It would then be appropriate toaggregate data on diuretics, beta-blockers andangiotensin-converting enzyme (ACE) inhibitors,etc. If, however, the question concerns the rela-tive use of beta-blockers in hypertension, data atthe substance (generic drug) level would be nee-ded. Information on the relative scale of use ofindividual products will sometimes be required,for example to find the market leader or toassess the relative use of generic versus brandedor innovator products. Information down to thelevel of dose strength will be necessary, forexample, to determine whether there is a trendtowards use of higher strengths of antibiotics, orto determine the relative use of strengths of anti-depressants to assess whether they are beingused at effective doses.

2.1.2 IndicationFor drugs with multiple indications, it will usuallybe important to divide data on use according toindication to allow a correct interpretation of theoverall trends. An example is the relative use ofdrug groups in treating hypertension. The over-all data might suggest that the relative use ofdiuretics is comparable to that of ACE inhibitorsand higher than the use of calcium channel bloc-kers (column A in Table 1). However, analysisof the data according to indication may revealthat 75% of ACE inhibitors are used to treathypertension whereas only 43% of diuretics areused for this indication (most of the high-ceilingdiuretics used are for treating heart failure). Thepicture that emerges of the use of the two druggroups in the treatment of heart failure is mar-kedly altered when use according to indication istaken into account (column B of Table 1).

Table 1 Relative use of drug groups in the treatmentof hypertension in Australia in 1998a

Drug group Ab Bc Cd

ACE inhibitors (C09A) 31.80 36.6 34.8

Calcium channel blockers

(C08C) 24.50 28 26.7

Diuretics (C03) 29.60 19.4 15.9

Beta-blockers

(C07AA, C07AB) 11.20 11.5 15.7

ATII antagonists

(C09CA) 3.00 4.6 6.9

Source: Australian Drug UtilizationSubcommittee and BEACH Survey April-December 1998, Sydney University, GPSCU1999.

a Values are the use of the drug group expressed as a percentage of the total use for these drug groups.

b Based on total use.c Adjusted for the percentage of total use of each group for the

treatment of hypertension.d Relative prescribing of these drug groups in hypertension

community practice patient encounters.

Page 15: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

Another example of a situation in which theindication is important is antibiotic utilization.In determining whether the use of a particularantibiotic, for example, amoxicillin, is rational, itwill usually be necessary to know what infecti-ons or problems it is being used to treat. Itwould therefore be necessary to break down dataon amoxicillin use into indications and comparethese uses with the appropriate guidelines. If itwere found that there was substantial use ofamoxicillin to treat acute sore throat, for exam-ple, this finding would indicate a problem thatneeded to be addressed. This is because a nar-row-spectrum agent (or no drug) would be amore appropriate treatment for a sore throat, andif amoxicillin is used to treat mononucleosis,which can present as a sore throat, there is ahigh incidence of rash.

2.1.3 Prescribed daily doses The prescribed daily dose (PDD) is the averagedaily dose prescribed, as obtained from a repre-sentative sample of prescriptions. The use ofDDD per 1000 inhabitants per day allows aggre-gation of data across drug groups and compari-sons between countries, regions and health faci-lities. However, the DDD metric may not reflectthe actual PDDs, and this needs to be considered

when making such comparisons. The PDDs dif-fer between countries and ethnic groups, andeven between areas or health facilities within thesame country. The PDD will also often differfor different indications of the same drug, so itwill sometimes be necessary to reach this levelof detail to interpret overall use data.

Data on the use of tricyclic antidepressantsand selective serotonin reuptake inhibitors(SSRIs) in Australia are shown in Table 2 asboth DDDs and prescription volumes.

The two metrics give different results for therelative use of the two groups of antidepressantdrugs because of the different relationship bet-ween the PDDs and the DDDs for the two druggroups. On average, the PDD is lower than theDDD for the tricyclics and higher for the SSRIs.In this case, knowledge of the PDDs is necessaryfor clinical interpretation of the data.

The DDD per 1000 inhabitants per day isoften used to derive a rough estimate of the pre-valence of use in the population being studied,and for chronic diseases it may even be used toassess the prevalence of a disease when the drugis prescribed for a single indication. Such esti-mates are valid only if the DDDs and the PDDsare similar.

14

Prescription % of total DDD/1000 % of total DDD/1000 volume prescription population/day population/day(millions)a volume

Tricyclics(N06AA) 3.53 48.82 8.40 28.09

SSRI(N06AB) 3.09 42.74 17.20 57.53

Moclobemide(N06AG02) 0.61 8.44 4.30 14.38

Total 7.23 100.00 29.90 100.00

Table 2 Relative use of antidepressants in Australia in 1998

Source: Australian Drug Utilization Subcommittee, Department of Health & Aged Care, Commonwealth ofAustralia, http://www.health.gov.au:80/haf/docs/asm.htm

Page 16: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

2.2 Problem or encounter-basedinformation

Reason for the encounter (the problem); drug treatment versus non-drug treatment; other problems managed; severity of the pro-blem managed; new or continuing presenta-tion; duration of consultation; medications prescribed for the problem; how the medica-tions were supplied; other medications pre-scribed

Rather than asking how a particular group ofdrugs is used, it may be useful to address thequestion of how a particular problem (e.g. sorethroat, hypertension or gastric ulcer) is managed.The different types of information that may berequired are listed in the box above.

As an example, consider how problem-basedinformation about the management of hyperten-sion might be used. Initially, concordance withguidelines for drug treatment or non-drug mana-gement of blood pressure and other risk factorsmight be assessed. Where drug treatment isused, the proportion of patients treated with eachof the drug groups gives an overall picture ofmanagement (column C of Table 1). This ismore direct information on how hypertension ismanaged than that provided by assessing theoverall use of the different drug groups as dis-cussed above. In the example shown in Table 1,the data in columns B and C are reasonably con-

sistent. This consistency between data using twodifferent approaches (i.e. drug and problem-based) gives confidence in the result.

Other questions that might be addressed usinga problem-based approach include the following:

• Does the severity of hypertension influence thechoice of single or combination therapy?• Is the management of newly-presenting pati-ents different to that of patients already receivingtreatment?• Are there likely to be any drug interactionswith co-prescribed treatments?• Is the choice of drug influenced by evidence-based outcome data?

For some diseases it may be important to studythe relative use of drug treatment and other the-rapeutic approaches to map out and understandpharmacotherapeutic traditions and other thera-peutic approaches. As an example, drug utiliza-tion research in Estonia has shown that there wasa reciprocal relationship between the use of hor-monal contraceptives and the abortion ratefrom1989-1997 (Fig. 2).

Another example was the excessive use of ulcersurgery in Estonia compared to Sweden duringthe Soviet era. This was because of the difficul-ties of obtaining modern anti-ulcer drugs inEstonia at that time (Fig. 3).

15

Figure 2 Abortion rate and use of hormonal contraceptives in Estonia in 1989-1997.Source: Kiivet R. Drug utilization studies as support to decisions in drug policy in Estonia. [MD Thesis]Stockholm, Karolinska Institutet, 1999.

Use

of h

orm

onal

cont

race

ptiv

es(D

DD/1

000/

day

Nr o

f abo

rtion

s(p

er 1

0 00

0 po

pula

tion

Page 17: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

2.3 Patient information

Age; gender; ethnicity; co-morbidities; knowledge; beliefs and perceptions

Information on demographic factors and otherdetails about the patient will often be useful.For example, the age distribution of patients maybe of critical importance, to assess the likelihoodof severe adverse effects with nonsteroidal anti-inflammatory drugs (NSAIDs), or whether thedrug is being used to treat patients in an agegroup different from that in which the clinicaltrials were performed. The co-morbidities of thepatient group may be important in determiningthe choice of treatment and predicting possibleadverse effects. For instance, in the manage-ment of hypertension, beta-blockers should notbe used to treat patients with asthma, and ACEinhibitors are the preferred treatment in patientswith heart failure.

Qualitative information relating to the know-ledge, beliefs and perceptions of patients andtheir attitudes to drugs will be important in somecases, for example in assessing the pressures putby patients on their doctors to prescribe antibio-tics, or in designing consumer information andeducation programmes.

2.4 Prescriber information

Demographic information - age, gender,medical school, years in practice; type of practice (e.g. specialist or family, rural or urban); practice size; patient mix; knowledgeabout drugs; factors driving prescribing behaviour

The prescriber plays a critical role in determi-ning drug use. Claims have even been made thatthe differences between doctors are greater thanthose between patients and that variations indrug prescribing behaviour often lack rationalexplanations. Dissecting the factors that deter-mine prescribing behaviour is therefore oftencentral to understanding how and why drugs areprescribed. Some questions that might beaddressed using prescriber information includethe following:• Are prescribing profiles influenced by the

prescriber’s medical education?• Do the prescribing profiles of specialists differ

from those of family practitioners?• Does the age or gender of the prescriber influ-

ence the prescribing profile?• Are there differences in prescribing behaviour

between urban and rural practices or between small and large practices? Do these variations indicate a need to target education to particular sectors?

16

0

20

40

60

80

1993 19950

4

8

12

16

20

anti-ulcer drugs, Estonia surgery, Estonia

anti-ulcer drugs, Stockholm surgery, Stockholm

Figure 3 Treatment of ulcer disease in Estonia and in Stockholm County 1993-1995.Source: Kiivet R. Drug utilization studies as support to decisions in drug policy in Estonia.[MD Thesis] Stockholm, Karolinska Institutet, 1999.

[ ]

Page 18: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

• Who are those prescribers who rapidly adoptrecently released drugs?• In assessing rational use of medicines by apractitioner, has the practice mix been taken intoaccount?• Can the factors that determine and change pre-scribing behaviour be identified?

2.5 Types of drug utilizationstudyDrug utilization studies can be targeted towardsany of the following links in the drug-use chain:

– the systems and structures surrounding drug use (e.g. how drugs are ordered, delivered andadministered in a hospital or health care facility);– the processes of drug use (e.g. what drugs are used and how they are used and does their use comply with the relevant criteria, guidelines or restrictions); and– the outcomes of drug use (e.g. efficacy,adverse drug reactions and the use of resour-ces such as drugs, laboratory tests, hospital beds or procedures).

Cross-sectional studiesCross-sectional data provide a «snapshot» ofdrug use at a particular time (e.g. over a year, amonth or a day). Such studies might be used formaking comparisons with similar data collectedover the same period in a different country,health facility or ward, and could be drug-, pro-blem-, indication, prescriber- or patient-based. Alternatively, a cross-sectional study can be car-ried out before and after an educational or otherintervention. Studies can simply measure druguse, or can be criterion-based to assess drug usein relation to guidelines or restrictions.

Longitudinal studiesPublic health authorities are often interested intrends in drug use, and longitudinal data arerequired for this purpose. Drug-based longi-tudinal data can be on total drug use as obtainedthrough a claims database, or the data may bebased on a statistically valid sample of pharma-cies or medical practices. Longitudinal data are

often obtained from repeated cross-sectional sur-veys (e.g. IMS (Intercontinental MedicalStatistics) practice-based data are of this type).Data collection is continuous, but the practitio-ners surveyed, and therefore the patients, arecontinually changing. Such data give informa-tion about overall trends, but not about prescri-bing trends for individual practitioners or practices.

Continuous longitudinal studiesIn some cases continuous longitudinal data at theindividual practitioner and patient level can beobtained. Claims databases are often able tofollow individual patients using a unique (butanonymous) identifier. These data can provideinformation about concordance with treatmentbased on the period between prescriptions, co-prescribing, duration of treatment, PDDs and soon. As electronic prescribing becomes morecommon, databases are being developed to pro-vide continuous longitudinal data comprisingfull medical and prescribing information at theindividual patient level. Such databases are verypowerful, and can address a range of issuesincluding reasons for changes in therapy, adverseeffects and health outcomes.

2.6 Drug costs

Total drug costs; cost per prescription; cost per treatment day, month or year; cost per defined daily dose (DDD); cost per prescri-bed daily dose (PDD); cost as a proportion of gross national product; cost as a propor-tion of total health costs; cost as a propor-tion of average income; net cost per health outcome (cost-effectiveness ratio); net cost per quality adjusted life-year (cost-utility- ratio)

Data on drug costs will always be important inmanaging policy related to drug supply, pricingand use. Numerous cost metrics can be used andsome of these are shown in the box above. Forexample, the cost per DDD can usually be usedto compare the costs of two formulations of the

17

Page 19: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

same drug. However, it is usually inappropriateto use this metric to compare the costs of diffe-rent drugs or drug groups as the relationship bet-ween DDD and PDD may vary.

Estimates of the costs at various levels andusing data aggregated in various ways will berequired, depending on the circumstances andthe perspective taken. A government perspectivemight require information on drug costs and costoffsets to government to be collected, whereas asocietal perspective would require both govern-ment and non-government (private sector) costsand cost offsets to be determined. A patient per-spective will be appropriate if questions aboutaffordability and accessibility are being asked.Costs may be determined at government, healthfacility, hospital, health maintenance organiza-tion or other levels within the health sector.

Costs will often need to be broken downaccording to drug group or therapeutic area todetermine, for example, the reason for an increasein drug costs. For instance, the introduction ofnew, expensive anti-cancer agents may be foundto be driving the increases in drug costs in a hos-pital. Changes in drug costs can result fromchanges in prescription volumes, quantity perprescription or in the average cost per prescripti-on. For example, most countries have experien-

ced a marked increase in the cost of anti-psycho-tic drugs over the last 5-10 years; the data on useand cost for these drugs in Australia are illustra-ted in Fig. 4.

In Australia, there has been little increase in theoverall volume of use of antipsychotic drugs,and the cost increase has been driven by thetransfer from the cheap ‘classical’ agents to themuch more expensive ‘atypical’ drugs such asclozapine, olanzapine and risperidone resultingin an increase in the average cost per prescrip-tion. In contrast, both the prescription volume ofantidepressant drugs and the average cost perprescription have increased over the same period,due to an ‘add-on’ prescribing effect of the moreexpensive SSRIs.

2.7 General readingEinarsson TR, Bergman U, Wiholm BE.Principles and practice of pharmacoepidemiology.In: Speight TM, Holford NH, eds. Avery’s DrugTreatment. Place, Adis International, 1999:371-392.

Lee D, Bergman U. Studies of drug utilization.In: Strom B. ed. Pharmacoepidemiology, 3rd ed.Chichester, J Wiley, 2000:463-481.

18

0

2

4

6

8

10

1990 1991 1992 1993 1994 1995 1996 1997 19980

20

40

60

80

100community usegovt cost with clozapine

use

cost

Figure 4 Antipsychotic drugs - use and cost trends in Australia

DD

s/10

00/D

ay$ m

illion

Page 20: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

2.8 Exercises3. Amoxicillin

You note that amoxicillin use expressed as DDDs per1000 population per day has increased over the last two years. What types of drug utilization data would you need to evaluate the possible reasons for this?

4. Antidepressant useThe use of antidepressant drugs (in DDDs per 1000 population per day) and their costs have been increasing for at least the last five years.What types of data would you need to deter-mine the reasons for the change and whether ithas resulted in positive or negative health out-comes?

19

Page 21: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

Drug-use chain; large databases; other sour-ces; drug use evaluation; pharmacoeconomics

The drug-use chain includes the processes ofdrug acquisition, storage, distribution, prescri-bing, patient compliance and the review of out-come of treatment. Each of these events is animportant aspect of drug utilization, and mostcountries have regulations to cover these aspects.Data are collected, or are available, at national,regional and local health facility or householdlevel and may be derived from quantitative orqualitative studies. Quantitative data may beused to describe the present situation and thetrends in drug prescribing and drug use at vari-ous levels of the health care system.Quantitative data may be routinely collected dataor obtained from surveys. Qualitative studiesassess the appropriateness of drug utilization andgenerally link prescribing data to reasons (indi-cations) for prescribing. Such studies have beenreferred to as «drug utilization review» or «drugutilization evaluation». The process is one of a«therapeutic audit» based on defined criteria andis intended to improve the quality of therapeuticcare. There is an increasing interest in the evalu-ation of the economic impact of clinical care andmedical technology. This has evolved into a dis-cipline dedicated to the study of how pharma-cotherapeutic methods influence resource utili-zation in health care known as pharmacoeco-nomics (see chapter 4).

The sources of drug utilization data vary fromcountry to country depending on the level ofsophistication of record keeping, data collection,analysis and reporting and the operational consi-derations of the health care system.

3.1 Large databasesThe increasing interest in efficient use of healthcare resources has resulted in the establishmentof computer databases for studies on drug utili-zation. Some of the databases can generate sta-tistics for patterns of drug utilization and adversedrug reactions. Data may be collected on drugsales, drug movement at various levels of the

drug distribution chain, pharmaceutical andmedical billing or samples of prescriptions. Thedatabases may be international, national or localin scope. They may be diagnosis-linked or non-diagnosis-linked. Diagnosis-linked data enabledrug use to be analysed according to patient cha-racteristics, therapeutic groups, diseases or con-ditions and, in the best of cases, clinical out-come. A useful analysis requires an understand-ing of the sources and organization of the data.

3.2 Data from drug regulatoryagencies

Drug registration; drug importation

Drug regulatory agencies have the legal respon-sibility of ensuring the availability of safe, effi-cacious and good-quality drugs in their country.They are thus the repositories of data on whichdrugs have been registered for use, withdrawn orbanned within a country. Regulatory agenciesalso have inspection and enforcement functions,and are responsible for supervising the importa-tion of drugs and for the issuance of permits fordrug registration.

It is possible, therefore, to obtain data on thenumber of drugs registered in a country fromsuch agencies. Where the agency issues importpermits and supervises drug importation, data onproduct type (i.e. generic or branded), volume,port of origin, country of manufacture, batchnumber and expiry date may be collected.Where the data reflect total national imports,estimates of quantities of drugs in circulationcan be obtained for defined periods and for various therapeutic groups

It may be difficult to obtain true estimates ifdocumentation is incomplete and not all trans-actions are recorded. Information on smuggledgoods or goods entering the country through ille-gal routes will not be captured by these data.

3.3 Supplier (distribution) dataDrug importation; local manufacture; cus-toms service

20

Chapter 3: Sources of data on drug utilization

[ ]

[ ]

[ ]

Page 22: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

Data on suppliers may be obtained from drugimporters, wholesalers or local manufacturers.In countries where permits or licences arerequired from drug regulatory authorities andministries of health before importation of drugs,data may be available from such sources.Customs services, in the process of clearingimports from the ports of entry, may collect dataon drugs. However, the codes used by customsservices are not detailed enough to capture allrelevant information. National agencies respon-sible for the collection of excise duty can alsoprovide information on the volume of productionand on distribution of drugs from local manufac-turers.

Data from these sources can generally be usedto describe total quantities of specific drugs ordrug groups, origins of supplies and type (i.e.branded or generic).

In the absence of a national mechanism for thedirect capture of data on drug production orimportation, wholesalers become an importantsource of information on drug acquisition. Suchdata are reliable insofar as wholesalers are theonly legal entity able to import drugs. In somecountries, medical, dental and veterinary practiti-oners, as well as pharmacists, can import phar-maceutical products. It is usually very difficultto collect comprehensive data from such sourceseven if there are regulatory requirements aboutsubmitting reports. Public sector procurementpractices, however, have reasonable documen-tation but provide data only on that sector.

Practice setting data

Prescribing data; dispensing data; drug use indicators; facility data (aggregate)

Data from health facilities may be used to evalu-ate specific aspects of health provision and druguse and to generate indicators that provide infor-mation on prescribing habits and aspects of pati-ent care. These indicators can be used to deter-mine where drug use problems exist, provide amechanism for monitoring and supervision and

motivate health care providers to adhere to esta-blished health care standards.

3.4.1 Prescribing dataPrescribing data are usually extracted from out-patient and inpatient prescription forms. Suchdata may be easily retrieved where records arecomputerized and computerized data also facili-tate trend analysis. In the absence of electronicdatabases, prescribing data are usually extractedfrom patient records or from patient interceptstudies or retrieved at dispensing points.

Information that may be obtained from pre-scriptions includes patient demography, drugname, dosage form, strength, dose, frequency ofadministration and duration of treatment. Wherediagnoses are noted on prescriptions, and parti-cularly for inpatient prescription, it is possible tolink drug use to indications. Trends in utilizationfor specific drugs and diseases can also be esta-blished. As an example, inpatient data may pro-vide a link to empirical treatment of infections asopposed to treatment based on microbiologicalassessment. This may be achieved by extractingrelevant data from the patient records, but requi-res that the records be of good quality.

Prescriptions are a good source of informationfor determining some of the indicators of druguse recommended by WHO including the:

– average number of drugs per prescription (encounter);

– percentage of drugs prescribed by generic name;

– percentage of encounters resulting in prescrip-tion of an antibiotic;

– percentage of encounters resulting in prescrip-tion of an injection;

– percentage of drugs prescribed from essential drugs list or formulary, and

– average drug cost per encounter.

Prescribing data allow the determination of thePDD which may differ from the DDD. Whilethe DDD is based on the dosages approved instandard product characteristics with clinical out-come data from controlled clinical trials, the

21

[ ]

Page 23: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

PDD is variable and dependent on factors suchas severity of illness, body weight, interethnicdifferences in drug metabolism and the prescri-bing culture of the health provider. Using DDDsenables comparison to be made between druggroups as the influences of prescribing cultureand available dosage strengths are eliminated.

In some countries, it is a legal requirementthat prescriptions dispensed by pharmacies anddrug outlets are kept for a minimum period beforedisposal. Where these regulations are adheredto, prescription data may be obtainable frompharmacies. However, in many developingcountries the rule is not generally followed. Incountries where computerized records of prescri-bing data are kept, they may be readily retriev-able depending on the depth of the database.

3.4.2 Dispensing data Drug dispensing is a process that ends with aclient leaving a drug outlet with a defined quan-tity of medication(s) and instructions on how touse it (them). The quantity of drugs dispenseddepends on their availability. Thus informationavailable from dispensers may include:– drug(s) prescribed;– dose(s) prescribed;– average number of items per prescription;– percentage of items prescribed that were actu -

ally supplied (an indicator of availability);– percentage of drugs adequately labelled;– quantity of medications dispensed; and – cost of each item or prescription.

These data may be obtained from records kept atthe drug outlet either in electronic or manualform.

3.4.3 Aggregate dataA number of data sources within the health faci-lity or hospital setting can provide aggregatedata on drug utilization. These sources includeprocurement records, warehouse drug records,pharmacy stock and dispensing records, medica-tion error records, adverse drug reaction recordsand patient medical records. These data sources

can be used to obtain information on variousaspects of drug use including:– the cost of individual drugs and classes of

drug;– the most frequently or infrequently used drugs;– the most expensive drugs;– the per capita consumption of specific pro-

ducts;– comparisons of two or more drugs used for the

same indication;– the prevalence of adverse drug reactions;– the prevalence of medication errors; and– the percentage of the budget spent on specific

drugs or classes of drug.

Aggregate data are often useful for comparingthe utilization of a particular drug to that ofother drugs and to utilization in other hospitals,regions or countries.

3.4.4 Over-the-counter and pharmacist-prescribed drugsPharmacists and other drug outlet managers mayprescribe over-the-counter preparations or phar-macist-prepared drugs that do not require pre-scription by a physician. Data on such medica-tions may be difficult to obtain especially inenvironments with weak drug regulation andpoor record keeping, but when such informationis available from stock or dispensing records, itbroadens the understanding of drug utilizationpatterns.

3.4.5. Telephone and Internet prescribingPhysicians in certain countries may prescribeover the telephone. Prescribing and dispensingusing the Internet also occurs, especially in deve-loped countries. Most Internet prescriptions arefor nutritional supplements and herbal preparati-ons. However, as exemplified by sildenafil(Viagra®), other medicines are also increasinglybeing sold on the Internet. Innovative ways haveto be devised to collect information on this typeof transaction.

22

Page 24: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

3.5 Community setting dataHousehold survey; compliance (adherence to treatment); drug utilization

The drugs available in households have eitherbeen prescribed or dispensed at health facilities,purchased at a pharmacy (with or without a pre-scription) or are over-the-counter medications.The drugs may be for the treatment of a currentillness or are left over from a previous illness. Itis not uncommon for patients to adhere poorly tothe instructions given for taking their dispensedmedicines. Thus dispensing data and utilizationdata may not be equivalent because they havenot been corrected for non-compliance.

Drug utilization by outpatients is best assessedby performing household surveys, counting left-over pills or using special devices that allowelectronic counting of the number of times a par-ticular drug is administered. Drug utilization byinpatients can be determined by reviewing treat-ment sheets or orders.

For both outpatients and inpatients, the dataon the utilization of a particular drug can beaggregated for a defined population in DDDs.Using DDDs has the advantage of allowing com-parison for example between inpatients and out-patients. Data on various dosage forms andgeneric equivalents of the same medication canalso be aggregated.

3.6 Drug use evaluationDrugs and therapeutic committee; prospec-tive evaluation; retrospective evaluation; criteria setting

Drug use evaluation, sometimes referred to asdrug utilization review, is a system of continu-ous, systematic, criteria-based drug evaluationthat ensures the appropriate use of drugs. It is amethod of obtaining information to identify pro-blems related to drug use and if properly develo-ped, it also provides a means of correcting theproblem and thereby contributes to rational drugtherapy.

Drug use evaluation can assess the actual pro-cess of administration or dispensing of a medica-tion (including appropriate indications, drugselection, dose, route of administration, durationof treatment and drug interactions) and also theoutcomes of treatment (e.g. cured disease condi-tions or decreased levels of a clinical parameter).The objectives of drug use evaluation include:

– ensuring that drug therapy meets current stan--dards of care

– controlling drug cost;– preventing problems related to medication;– evaluating the effectiveness of drug therapy; and – identification of areas of practice that require

further education of practitioners.

The problems to be addressed by drug use evalu-ation may be identified from any of the data des-cribed in section 3.4 (including prescription indi-cators, dispensing data and aggregate data). Themain source of data for drug use evaluation isthe patient records. An identifiable authoritativegroup, such as the drugs and therapeutic com-mittee, usually carries out reviews of drug use ina hospital or health facility. This group has theresponsibility for drawing up the guidelines, cri-teria, indicators and thresholds for the evaluati-on. Drug use evaluation may be based on datacollected prospectively (as the drug is being dis-pensed or administered) or retrospectively (basedon chart reviews or other data sources).

• Typical criteria reviewed in prospective studiesinclude the following– indications;– drug selection;– doses prescribed;– dosage form and route of administration;– duration of therapy;– costs;– therapeutic duplication;– quantity dispensed;– contraindications;– therapeutic outcome– adverse drug reactions; and– drug interactions.

23

[ ]

Page 25: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

• In retrospective studies, the criteria reviewed include:

– evaluation of indications;– monitoring use of high-cost medicines;– comparison of prescribing between physicians;– cost to patient;– adverse drug reactions; and– drug interactions.

It is possible to incorporate some of the abovecriteria into databases thus allowing drug expertsto evaluate any items that do not meet establis-hed criteria. For meaningful results to be obtai-ned from drug use evaluation a reasonable num-ber of records need to be assessed. A minimumof 50 to 75 records per health care facility isconsidered adequate. However, the number ofrecords sampled would depend on the size of thefacility and the number of prescribers.

3.7 General readingHow to investigate drug use in health facilities:Selected drug use indicators. Geneva,WorldHealth Organization, 1993 (unpublished docu-ment WHO/DAP/93.1; available on request fromDepartment of Essential Drugs and MedicinesPolicy, World Health Organization, 1211Geneva, 27, Switzerland).

3.8 ExercisesExamine the sources of data listed in theWorksheet. Imagine that you want to learnabout the utilization of antibiotics in your coun-try. In the spaces provided in the right-handcolumns of the worksheet, write down (1) whatkinds of useful data you might gather from each

source that could help you understand the situa-tion, and (2) some possible advantages and/orlimitations of each of the sources of data youhave listed.

When evaluating the advantages andlimitations of the data, consider the ans-wers to the following questions:

• How relevant are the data for learning about antibiotics?

• How easy is it to collect these types of data in your country?

• How much will it cost to collect and process the data and how long will it take?

• How reliable are these data?

For example, from data from previoussurveys, we might obtain the followinguseful information: historical utilizationrates by facility or geographical area, andpossibly utilization by type of antibiotic,health problem or age. The advantagesof using historical survey data are thatthey have already been collected andcarry no additional cost. However, theirlimitations include not being able to con-trol exactly which data have been collec-ted or from where, not knowing whethercurrent practices reflect those of the past,and having no patient-specific or provi-der-specific information. It would alsousually not be possible to find informa-tion on dosing of antibiotics.

24

Page 26: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

Worksheet for section 3.8 Sources of data on drug utilization

Data source Type of information available Advantages and limitations

Drug import records

Drug supply to health

facility

Orders and/or

delivery receipts

Previous reports of

surveys

Pharmacy stock cards/

pharmacy ledger book

Pharmacy sales

receipts

Large hospital or

insurance databases

Private

drug outlet sales

records

Community

or household surveys

Drug manufacturing

records

Other sources

25

Page 27: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

Pharmacoeconomics; types of cost; cost-mini-mization analysis; cost-effectivness analysis; cost-benefit analysis; cost-utility analysis

4.1 IntroductionDrug costs per se are important, as they account fora substantial part of the total cost of health care -typically 10-15% in developed countries and up to30-40% in some developing countries. However,drug costs usually need to be interpreted in the con-text of the overall (net) costs to the health system.Drugs cost money to buy, but their use may alsosave costs in other areas. For example, the purchaseof one specific type of drug may lead to reductionsin the following:– use of other drugs;– the number of patients requiring hospitalization

or in the length of stay in hospital;– the number of doctor visits required;– administration and laboratory costs compared

with those incurred by using another drug to treatthe same condition. Assessing the true cost to a health system of

using a specific drug will therefore require the costof acquisition of the drug to be balanced againstboth any cost savings resulting from the use of thatdrug and the extra health benefits it may produce.On the other hand, costs may arise from adversedrug reactions both in the short- and particularly thelong-term.

Assessing the value for money of using a drugrequires the extra health benefits achieved to beweighed against the extra net cost. This compari-son is usually expressed as an incremental cost-effectiveness ratio (ICER) which is the net incre-mental cost (costs minus cost offsets) of gaining anincremental health benefit over another therapy.

Concerns about the cost of medical care in gene-ral, and pharmaceuticals in particular, are currentlybeing expressed by all health systems. There is ageneral focus on providing quality care within limi-ted financial resources. Decision-makers are increa-singly dependent on clinical economic data to guidepolicy formulation and implementation. Some ofthe concepts used in making such decisions include:cost-minimization, cost-effectiveness,cost-benefit, and cost-utility.

4.2 Cost-minimization analysisCost-minimization analysis is a method of calcu-lating drug costs to project the least costly drugor therapeutic modality. Cost minimization alsoreflects the cost of preparing and administering adose. This method of cost evaluation is the oneused most often in evaluating the cost of a speci-fic drug. Cost minimization can only be used tocompare two products that have been shown tobe equivalent in dose and therapeutic effect.Therefore, this method is most useful for compa-ring generic and therapeutic equivalents or «metoo» drugs. In many cases, there is no reliableequivalence between two products and if thera-peutic equivalence cannot be demonstrated, thencost-minimization analysis is inappropriate.

If a new therapy were no safer or more effecti-ve than an existing therapy (i.e. there is no incre-mental benefit), it would normally justify thesame price as the existing therapy. An examplewould be the introduction of a new ACE inhibi-tor with essentially the same properties as exis-ting members of the class; the price would beequivalent to that of the existing drug(s). This isoften not as simple as it may seem, as it requiressound trial-based information on the doses of thetwo drugs required for equivalent efficacy. Analternative is to use the PDDs for the two drugsin the marketplace to determine the relative pri-ces. This is a pragmatic approach, but assumesthat the two drugs are actually used at equiva-lently effective doses, and this may not alwaysbe the case.

4.1 Cost-effectiveness analysis Cost-effectiveness analysis involves a more com-prehensive look at drug costs. While cost ismeasured in monetary terms, effectiveness isdetermined independently and may be measuredin terms of a clinical outcome such as number oflives saved, complications prevented or diseasescured.

Cost-effectiveness analysis thus measures theincremental cost of achieving an incrementalhealth benefit expressed as a particular healthoutcome that varies according to the indicationfor the drug. Examples of ICERs using this

26

Chapter 4: Economic aspects of druguse (pharmacoeconomy)

Page 28: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

approach are:– the cost per extra patient achieving a 10 mm

Hg fall in blood pressure;– the cost per extra asthmatic patient achieving

a reduction in oral corticosteroid use– the cost per extra episode of febrile neutro-

penia avoided; or – the cost per extra acute rejection episode avo-

ided in patients with kidney transplants.

It is often difficult to make judgements about therelative value for money across a range of druggroups and health outcomes such as those in theexamples given above.

4.4 Cost-utility analysis Cost-utility analysis is used to determine cost interms of utilities, especially quantity and qualityof life. This type of analysis is controversialbecause it is difficult to put a value on health sta-tus or on an improvement in health status as per-ceived by different individuals or societies.Unlike cost-benefit analysis, cost-utility analysisis used to compare two different drugs or proce-dures whose benefits may be different.

Cost-utility analysis expresses the value formoney in terms of a single type of health out-come. The ICER in this case is usuallyexpressed as the incremental cost to gain anextra quality-adjusted life-year (QALY). Thisapproach incorporates both increases in survivaltime (extra life-years) and changes in quality oflife (with or without increased survival) into onemeasure. An increased quality of life isexpressed as a utility value on a scale of 0 (dead)to one (perfect quality of life). An increasedduration of life of one year (without change inquality of life), or an increase in quality of lifefrom 0.5 to 0.7 utility units for five years, wouldboth result in a gain of one QALY. This allowsfor easy comparison across different types ofhealth outcome, but still requires value judge-ments to be made about increases in the qualityof life (utility) associated with different healthoutcomes. The use of incremental cost-utility

ratios enables the cost of achieving a health benefitby treatment with a drug to be assessed againstsimilar ratios calculated for other health inter-ventions (e.g. surgery or screening by mammo-graphy). It therefore provides a broader contextin which to make judgements about the value formoney of using a particular drug.

4.5 Cost-benefit analysisCost-benefit analysis is used to value both incre-mental costs and outcomes in monetary termsand therefore allows a direct calculation of thenet monetary cost of achieving a health outcome.A gain in life-years (survival) may be regardedas the cost of the productive value to society ofthat life-year using, for example, the averagewage. The methods for valuing gains in qualityof life include techniques such as willingness-to-pay, where the amount that individuals would bewilling to pay for a quality-of-life benefit isassessed. However, the techniques used to valuehealth outcomes in monetary terms remainsomewhat controversial, with the result that cost-benefit analysis is so far not widely used in phar-macoeconomic analyses.

Economic analyses such as those describedabove may be trial based or modelled. A trialbased analysis uses the incremental benefits anduse of resources in a clinical trial to calculate anICER, but this may not be as relevant to the useof the drug as it would be in the marketplace. Amodelled analysis is used to apply the benefitsand use of resources to a local clinical situation,and to extend the time frame beyond that of aclinical trial. This is particularly importantwhere the benefits of treatment may not be reali-zed until some time in the future. Two examplesare the avoidance of liver cancer or transplantati-on for patients with hepatitis C and the prolong-ation of life for hypertensive patients. Short-term surrogate outcome measures (clearance ofvirus and lowering of blood pressure, respective-ly) are used in clinical trials, and need to betranslated by modelling into the longer-term out-comes, which are more relevant to patients andpolicy-makers.

27

Page 29: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

4.6 General readingHow to investigate drug use in health facilities:Selected drug use indicators. Geneva,WorldHealth Organization, 1993 (unpublished docu-ment WHO/DAP/93.1; available on request fromthe Department of Essential Drugs andMedicines Policy, World Health Organization,1211 Geneva, 27, Switzerland).

Schulman KA et al. Pharmacoeconomics:Economic evaluation of pharmaceuticals. In:Strom B. ed. Pharmacoepidemiology, 3rd ed.Chichester, J Wiley, 2000.

4.7 Exercises1. Comparison of antihypertensivesYou are considering the use of a new alpha-anta-gonist for the treatment of hypertension. It isused once daily and you are told that it has beentested in trials against enalapril and losartan andit has been found to lower blood pressure to asimilar extent to these agents. You already haveprazosin on your subsidy list but the producersinform you that they have not carried out trialsagainst prazosin. The approximate costs for amonth’s supply of the existing drugs are prazosin$18, enalapril $28 and losartan $35. Beta-bloc-kers and thiazide diuretics are also on your sub-sidy list at a cost of about $8 for a month’s sup-ply but no trials of the new agent have been car-ried out against them. How would you approachthe pricing of the new alpha-antagonist?

2. Thrombolytics for acute myocardialinfarction (a purely hypothetical exercise)Congratulations! You have just been appointedas a member of the formulary committee of alarge teaching hospital in Sydney, Australia. Akey item on the agenda for the next meeting is aproposal to implement a management protocolfor the treatment of acute myocardial infarction.You have been asked to evaluate the availableevidence and advise to the committee as towhich of two available drugs represents the morecost-effective choice.

The draft clinical management protocol cur-rently proposes the use of the (hypothetical)thrombolytic drug Thrombase. A new drug,

Klotgon, has recently been brought to yourattention. The two drugs have been compared ina large randomized trial in which the primaryoutcome of mortality was measured 30 daysafter randomization.

Outcomes in 100 patientsNo treatment 15 deathsThrombase 10 deathsKlotgon 7 deathsDrug cost per patientThrombase $ 200Klotgon $ 1000

You are also aware that the average survival timefollowing non-fatal myocardial infarction iseight years.

Please answer the following questions. Beprepared to present your findings to the largegroup.a. If the hospital budget were unlimited, and if

1000 patients were to be treated, how many lives could be saved if patients were treated with Thrombase, compared with no treatment?How many could be saved with Klotgon, com-pared with no treatment?

b. If the hospital’s budget for purchasing throm-bolytics were $200 000, how many patients could be treated, and how many lives could besaved with each of the drugs, compared with no treatment at all?

c. What is the incremental cost per life saved, foreach of the thrombolytic agents, compared with no active treatment?

d. What are the incremental cost-effectiveness ratios (ICERs), expressed as the incremental cost per life-year gained, for each of the thrombolytic agents, compared with no active treatment?

e. What is the ICER for Klotgon compared to Thrombase?

f. What will you recommend to the formulary committee?

28

Page 30: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

3.Unfractionated heparin versus low-molecular-weight heparinBecause of your valuable contribution to thedevelopment of a cost-effective treatment proto-col for acute myocardial infarction, you havebeen retained as a member of the formularycommittee of the above-mentioned hospital. Anagenda item for consideration at your committee’snext meeting concerns a recommendation, froma very pleasant pharmaceutical company repre-sentative, that you replace unfractionated heparinwith a low- molecular-weight heparin in themanagement of patients with unstable coronaryartery disease. She very kindly gives you a sum-mary of some data from a clinical trial publishedin the New England Journal of Medicine. Theoutcomes were reported 30 days after randomi-zation.

You decide to investigate the costs of acquiringand monitoring treatment with the two drugs andnote the following:

Item Low-molecular- Unfractionated

weight heparin heparin

Monthly drug costs ($) $72.20 $27.09

Monthly cost none 5 tests/patient of

monitoring

anticoagulant effect @ $12.40 per test

Please answer the following questions. Be pre-pared to present your findings to a large group.a. Calculate the relative risk of the combined

(triple) end-point in patients who received low-molecular-weight heparin compared with those who received unfractionated heparin.

b. Calculate the risk difference and the number of patients who need to be treated to prevent asingle event with low-molecular-weight hepa-rin compared with unfractionated heparin.

c. Calculate the ICER for the main clinical outcome with low-molecular-weight heparin,compared with unfractionated heparin using drug costs only.

d. Recalculate the ICER for the main clinical outcome with low-molecular-weight heparin,compared with unfractionated heparin includ-ing the costs of monitoring treatment with heparin.

4. Celecoxib versus diclofenacA representative from a very supportive pharma-ceutical company addressed your medical staffduring a Saturday seminar last week. She gavean interesting presentation on the comparativesafety of some well-known anti-inflammatorypreparations. At the formulary meeting thisweek, the head of your rheumatology departmentis planning to propose adding celecoxib, a COX-

29

Outcome Low-molecular- Unfractionated P-valueweight heparin heparin

Combined risk of death 318/1607 364/1564 0.016

AMI1 or unstable angina (19.8%) (23.3%)

Percutaneous 236/1607 293/1564 0.002

revascularization ($ 1390 per (14.7%) (18.7%)

procedure)

Major bleeds 102/1569 107/1528 0.57

(6.5%) (7.0%)

Minor bleeds 188/1580 110/1528 <0.001(11.9%) (7.2%)

1Acute myocardial infarction

Page 31: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

2 inhibitor, to the hospital formulary in place ofNSAIDs. He intends to argue that the hospitalwill save a lot of money by avoiding the compli-cations associated with NSAIDs such as pepticulcers. This item was placed on the agenda of

the committee as a late submission, so you deci-de to review the evidence and prepare yourselffor the discussion. You find the following resultsof a clinical trial reported in the Lancet.

30

Mean (SD) arthritis assessment results at week 24

Primary assessments Celecoxib Diclofenac

Baseline Week 24 Baseline Week 24

Physician’s assessmenta 2.9 (0.7) 2.6 (0.8) 3.0 (0.8) 2.6 (0.8)

Patient’s assessmenta 3.0 (0.8) 2.7 (0.9) 3.1 (0.8) 2.8 (0.9)

No. of tender/painful joints 20.3 (14.4) 14.5 (14.1) 21.7 (14.4) 16.4 (14.7)

No. of swollen joints 14.9 (10.2) 10.7 (10.1) 14.3 (9.9) 10.4 (10.0)

Frequency of peptic ulceration and related complications

Celecoxib Diclofenac P-value(n = 212) (n = 218)

Patients in whom erosion, ulcer or both were detected

Gastric 38 (18%) 74 (34%) <0.001Duodenal 11 (5%) 23 (11%) <0.009

Ulcer incidence by Helicobacter pylori status

Positive serological test 7/93 (8%) 19/87 (22%) NSNegative serological test 1/97 (1%) 10/100 (10%) NS

Ulcer frequency by concomitant corticosteroid use

Corticosteroid use 2/80 (3%) 12/102 (12%) NSNo corticosteroid use 6/132 (5%) 21/116 (18%) NSNS: not significant.

The following adverse event data were also reported.

aIndependent assessments, graded from 1 (very good: symptom-free with no limitation of normal activities) to 5 (very poor: very severe symptoms thatare intolerable, and inability to carry out all normal activities).

Page 32: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

From your research you also know that:

• One per cent of patients with endoscopic damage are hospitalized with gastrointestinal bleeding.

• The cost of hospitalization for gastrointestinal bleeding is $1434/patient.

• Ten per cent of patients admitted with gastroin-testinal bleeding die.

• The cost of celecoxib for 60 x 100 mg tablets is $50.

• The usual dose of celecoxib is 200 mg bd.

• The cost of diclofenac is $11.60 for 50 x 50 mg tablets; $14.35 for 100 x 25 mg tablets.

• The usual dose of diclofenac is 50 mg-75 mg bd.• Answer the following questions. Be prepared

to present your findings to a large group.

a.Calculate the relative risk for peptic (i.e. gas-tric or duodenal) ulcers in the patients who received celecoxib compared with those who received the NSAID diclofenac.

b.Calculate the risk difference and the number of patients who have to be treated to prevent a single event with celecoxib, as compared with the NSAID.

c. Calculate the ICER for the main clinical out-come with celecoxib, compared with the NSAID, using drug costs only.

d. Recalculate the ICER for the main clinical outcome with celecoxib, compared with the NSAID, including the costs of treatment of gastrointestinal bleeding.

5.Oral montelukast versus an inhaled steroidA community-driven asthma awareness grouphas donated 10 cartons of montelukast tablets foradults treated in your hospital’s asthma clinic.They feel strongly that this product should bemade available since, according to the medicaladviser of the sponsoring company, it is muchmore effective and much easier to use than theusual «puffers». As this product is fairly new,the formulary committee has been asked tocomment on its effectiveness. Since the asthmaunit will prepare a submission for including it inthe hospital formulary after the supply of dona-ted drugs is exhausted, you have been asked tocomment on the comparative cost-effectivenessof the drug. You begin your assessment by con-sidering the following results at 22 weeks afterthe baseline assessment.

31

End-point Placebo Beclometasone Montelukast

Percentage change *FEV1 0.7 [–2.3, 3.7] 13.1 [10.1, 16.2] 7.4 [4.6, 10.1]

Change in daytime

asthma symptom score –0.17 [–0.3, –0.05] –0.62 [–0.75, 0.49] –0.41 [–5.3, –0.29]

Percentage change in total daily

β-agonist use 0.0 [–8.3, 8.3] –40.0 [–48.5, –31.5] –23.9 [–31.4, –16.5]

Change in morning PEFR [l/min] 0.8 [–7.1, 8.6] 39.1 [31.0, 47.1] 23.8 [16.6, 30.9]

Change in evening PEFR [l/min] 0.3 [–7.3, 8.0] 32.1 [24.2, 39.9] 20.8 [13.8, 27.8]

Change in nocturnal awakenings –0.5 [–0.9, –0.1] –2.4 [–2.8, –2.0] –1.7 [–2.07, 1.3]

[nights per week]

Change in eosinophil count

[cells x 103/µl] –0.02 [–0.07, 0.03] –0.07 [–0.12, –0.02] –0.08 [–0.12, –0.03]

Percentage of patients with

asthma attacks 27.3 10.1 15.6

Values are mean [95% CI]. *FEV1 = forced expiratory volume in one second; PEFR = peak expiratory flow rate

Page 33: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

The costs of the two drugs are:

– Beclametasone: Australian $26 for 28 days oftreatment;

– Montelukast: Australian $70 for 28 days of treatment.

Please answer the following questions. Be pre-pared to present your findings to a large group.

a.You wish to compare montelukast with beclo-metasone. Which outcome(s) will you use for the comparison? Why?

b.Calculate the ICER for the main clinical outcome.

c.Which is the better drug? Why?

32

Page 34: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

A drug classification system represents a com-mon language for describing the drug assortmentin a country or region and is a prerequisite fornational and international comparisons of drugutilization data, which have to be collected andaggregated in a uniform way. Access to stan-dardized and validated information on drug useis essential to allow audits of patterns of drugutilization, to identify problems in drug use, toinitiate educational or other interventions and tomonitor the outcomes of these interventions.The main purpose of having an internationalstandard is to be able to compare data betweencountries. A recent example is the internationalfocus on creating comparable systems for moni-toring cross-national patterns of antibacterial uti-lization to aid work against bacterial resistance.

5.1 Different classification systems

ATC classification; AT classification; EPhMRA; IMS; WHO Collaborating Centre for Drug Statistics Methodology

Drugs can be classified in different ways accor-ding to:– their mode of action;– their indications; or– their chemical structure.

Each classification system will have its advanta-ges and limitations and its usefulness willdepend on the purpose, the setting used and theuser’s knowledge of the methodology.Comparisons between countries may require aclassification system different from that neededfor a local comparison (e.g. between differentwards in a hospital). Of the various systems pro-posed over the years, only two have survived toattain a dominant position in drug utilizationresearch worldwide. These are the «AnatomicalTherapeutic» (AT) classification developed by theEuropean Pharmaceutical Market ResearchAssociation (EPhMRA) and the «AnatomicalTherapeutic Chemical» (ATC) classification deve-loped by Norwegian researchers. These systems

were originally based on the same main princi-ples. In the EPhMRA system, drugs are classi-fied in groups at three or four different levels.The ATC classification system modifies andextends the EPhMRA system to include a thera-peutic/pharmacological/chemical subgroup asthe fourth level and the chemical substance asthe fifth level (see, for example, the classifi-cation of glibenclamide in the box below).

The ATC classification is also the basis for theclassification of adverse drug reactions used bythe WHO Collaborating Centre for InternationalDrug Monitoring in Uppsala, Sweden(www.who-umc.org).

The main purpose of the ATC classification is asa tool for presenting drug utilization statisticsand it is recommended by WHO for use in inter-national comparisons. The EPhMRA classificationsystem is used worldwide by IMS for providingmarket research statistics to the pharmaceuticalindustry. It should be emphasized that the manytechnical differences between the EPhMRA classification and the ATC classification mean thatdata prepared using the two classification systemsare not directly comparable.

In 1996, WHO recognized the need to developthe ATC/DDD system from a European to aninternational standard in drug utilization studies.The European WHO Collaborating Centre forDrug Statistics Methodology in Oslo, Norway,which is responsible for coordinating the use ofthe methodology, was then linked to WHOHeadquarters in Geneva. This was intended toassist WHO in its efforts to ensure universalaccess to essential drugs and to stimulate ratio-nal use of drugs particularly in developing coun-tries.

5.2The ATC classification systemStructure; coding principles; therapeutic use; pharmaceutical formulations; strengths

The ATC classification system divides the drugsinto different groups according to the organ or

33

Chapter 5: Drug classification systems

[ ]

Page 35: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

system on which they act and according to theirchemical, pharmacological and therapeutic prop-erties.

Drugs are classified in groups at five differentlevels. The drugs are divided into 14 maingroups (first level), with two therapeutic/pharma-cological subgroups (second and third levels).The fourth level is a therapeutic/pharmacologi-cal/chemical subgroup and the fifth level is thechemical substance. The second, third andfourth levels are often used to identify pharma-cological subgroups when these are consideredto be more appropriate than therapeutic or chem-ical subgroups.

The complete classification of glibenclamide(see box below) illustrates the structure of thecode.

Thus, in the ATC system all plain glibenclamidepreparations are given the code A10B B01.

Medicinal products are classified according tothe main therapeutic use of their main activeingredient, on the basic principle of assigningonly one ATC code for each pharmaceutical for-mulation (i.e. similar ingredients, strength andpharmaceutical form).

A medicinal product can be given more than oneATC code if it is available in two or morestrengths or formulations with clearly differenttherapeutic uses. Two examples of this are givenbelow:

• Sex hormones in certain dosage forms orstrengths are used only in the treatment of cancerand are thus classified under L02 - Endocrinetherapy. The other dosage forms and strengthsare classified under G03 - Sex hormones andmodulators of the genital system.• Bromocriptine is available in differentstrengths. The low-dose tablets are used as pro-lactin inhibitors and are classified in G02 - Othergynaecologicals. Bromocriptine tablets in higherstrengths are used to treat Parkinson disease andare classified in N04 - Anti-Parkinson drugs.

Different formulations with different indicationsmay also be given separate ATC codes, forexample prednisolone is given several ATCcodes because of the different uses of the diffe-rent formulations (see box below).

The ATC system is not strictly a therapeutic clas-sification system. At all ATC levels, ATC codescan be assigned according to the pharmacologi-cal properties of the product. Subdivision on thebasis of mechanism of action will understanda-bly be rather broad, since a very detailed classifi-

34 A Alimentary tract and metabolism(first level, main anatomical group)

A10 Drugs used in diabetes(second level, main therapeutic group)

A10B Oral blood-glucose-lowering drugs (third level, therapeutic /pharmacological subgroup)

A10B B Sulfonamides, urea derivatives(fourth level, chemical/therapeutic/pharmacological subgroup)

A10B B01 Glibenclamide(fifth level, subgroup for chemical substance)

A07E A01 Intestinal anti-inflammatory agents (enemas and rectal foams)

C05A A04 Antihaemorrhoidals for topical use rectal suppositories)

D07A A03 Dermatological preparations (creams, ointments, lotions)

H02A B06 Corticosteroids for systemic use (tablets, injections)

R01A D02 Nasal decongestants (nasal spray, drops) S01B A04Ophthalmologicals (eye drops)

S02B A03 Otologicals (ear drops)

Page 36: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

cation of this kind would result in having onlyone substance per subgroup, which is better avo-ided (e.g. in the case of antidepressants). SomeATC groups are subdivided into both chemicaland pharmacological groups (e.g. ATC groupJ05A - Agents affecting the virus directly). If anew substance fits in both a chemical and phar-macological fourth level, the pharmacologicalgroup is normally chosen.

Substances classified as having the same ATCfourth level should not be considered as pharma-cotherapeutically equivalent since the profiles fortheir mode of action, therapeutic effects, druginteractions and adverse drug reactions may differ.

As the drugs available and their uses are cont-inuously changing and expanding, regular revisi-ons of the ATC system are necessary. An impor-tant principle is to keep the number of alterationsto a minimum. Before alterations are made, anypotential difficulties arising for the users of theATC system are considered and related to thebenefits that would be achieved by the alteration.Changes to the ATC classification would bemade when the main use of a drug had clearlychanged, and when new groups are required toaccommodate new substances or to improve thespecificity of the groupings.

Because the ATC system separates drugs intogroups at five levels (described above), statisticson drug utilization grouped at the five differentlevels can be provided. The information avai-lable ranges from figures showing total use of alldrug products classified e.g. in main group C -Cardiovascular system (first level), to figures forthe different subgroups (i.e. second, third andfourth level) to figures for the use of the separatesubstances.

More detailed information can be obtained at thelower (i.e. the fourth and fifth) levels. The hig-her levels are used if comparison of drug groupsis the aim of a study (see Fig. 5). This gives abetter overview and trends in drug use related todifferent therapeutic areas can easily be identified.

5.3 Ambivalence towards aninternational classification systemAll international standards demand compromisesand a drug classification system is no exceptionto this rule. Drugs may be used for two or moreequally important indications, and the main the-rapeutic use of a drug may differ from one coun-try to another. This will often result in several

35

Figure 5 Total sales of drugs used in cardiovascular disorders in Norway 1990-2001. ATC/DDD version 2002

0

10

20

30

40

50

60

70

80

ACE-inbitors +angiotensin II-antagonists (C09)

Serum lipid reducingagents (C10)

Calsium channelblockers (C08)

Diuretics (C03)

Beta blocking agents(C07)

DD

D/1

000

inha

b.da

y

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

Page 37: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

possible alternatives for classification, and adecision has to be made regarding the main use.Countries using a drug in a different way fromthat indicated by the ATC classification may notwish to adopt the ATC classification but prefer todevelop national classification systems.However, national traditions have to be weighedagainst the opportunity to introduce a methodo-logy that permits valid international comparisonsof drug utilization. Indeed, there are now manyexamples where an enthusiastic application ofthe ATC/DDD methodology has been instrumen-tal in stimulating national research in drug utili-zation and in developing an efficient drug con-trol system.

5.4 Implementation of theATC/DDD methodology

National drug register; dynamic system; different versions

As soon as the decision to introduce theATC/DDD methodology is taken, it is essentialto realize that its proper use inevitably includesan important and time-consuming first step.Each product has to be connected to the appro-priate ATC code and DDD (see chapter 6). Thelinkage between the national drug register andATC/DDDs has to be ascertained by personswith proper knowledge of the methodology.Experience has shown that in many countries,health authorities, health policy-makers and rese-archers have not always allocated adequateresources to this important initial step. Anotherproblem is that some users seem to be unawarethat the ATC/DDD methodology is a dynamicsystem to which changes are made continually.This has resulted in several different versions ofthe ATC/DDD system being used at the sametime, sometimes even within the same country.

It is important to realize that adopting theATC/DDD classification of drugs requiresresources and the necessary competence to carryout the work of allocating ATC codes to the pro-ducts. If possible, this work should be done on a

national basis to secure consistent use of themethodology within a country. As described inthe general introduction, the same substance mayhave several different ATC codes depending onthe application form and, to some extent, eventhe strength. For combination products, specificguidelines have been established for allocatingATC codes. Allocating DDDs to the productsnecessitates many of the same considerations asthe allocation of the ATC code. However, inorder to link the drug list with sales figures orprescription figures to obtain drug utilization sta-tistics, it is necessary to make appropriate calcu-lations such as the number of DDDs per drugpackage.

Finally, a given country will nearly alwayshave medicines and combination products forwhich no ATC codes or DDDs exist. In thesecases, it is important to consult the WHOCollaborating Centre for Drug StatisticsMethodology in Oslo and request new ATCcodes and DDDs. Once ATC codes and DDDshave been linked to the national drug lists, it isnecessary to update the drug list regularly inaccordance with the annual updates of theATC/DDD system.

The publication Guidelines for ATCClassification and DDD Assignment (seeGeneral reading) gives the information necessaryfor allocating ATC codes and DDDs at a nationalor local level. All officially assigned ATC codesand DDDs are listed in the ATC Index withDDDs (see General reading), a publication thatis also available in electronic format and is upda-ted every year. Training courses in theATC/DDD methodology are arranged annuallyin Norway and from time to time in other coun-tries. Further information is available on theweb site of the WHO Collaborating Centre forDrug Statistics Methodology athttp://www.whocc.no.

5.5 General readingGuidelines for ATC classification and DDDAssignment. Oslo, Norway, WHO CollaboratingCentre for Drug Statistics Methodology, 2003.

36[ ]

Page 38: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

ATC Index with DDDs. Oslo, Norway, WHOCollaborating Centre for Drug StatisticsMethodology, 2003.

Capellà D. Descriptive tools and analysis. In:Dukes MNG ed. Drug utilization studies, met-hods and uses. Copenhagen, WHO RegionalOffice for Europe, 1993 (WHO RegionalPublications, European Series, No. 45), 55-78.

Rønning M et al. Different versions of theanatomical therapeutic chemical classificationsystem and the defined daily dose - are drugutilization data comparable? European Journalof Clinical Pharmacology, 2000, 56:723-727.

Rägo L. Estonian regulatory affairs.Regulatory Affairs Journal. 1996, 7:567-573.

5.6 Exercises1. «Neurol» is a major tranquillizer belonging to

the butyrophenone group of antipsychotics. The only ATC code for this substance at pre-sent is in N01A X. The parenteral formulati-ons of Neurol are used for various indications e.g. in anaesthesia, as antiemetics and in the

treatment of schizophrenia. The oral formula-tions of Neurol are, however, used mainly in the treatment of schizophrenia.

Discuss whether it would be appropriate to assign an additional ATC code in N05 for oral formulations of Neurol.

2. Lisuride has been assigned two codes in the ATC classification system:G02CB02 (Prolactin inhibitors) and N02CA07 (Antimigraine preparations).Lisuride preparations in high strengths (e.g. tablets of 0.2 mg) are classified in G02CB. The recommended dose range for prolactin inhibition is 0.1-0.2 µg x 3. Low-strength preparations (e.g. tablets of 25 µg) are classi-fied in N02CA. The recommended dose rangefor treatment of migraine is 25 mg x 3.

Lisuride is also used for the treatment of parkinsonism. The recommended dose range for this indication is 0.2-0.6 mg daily.

Discuss whether it would be appropriate to assign an additional ATC code for lisuride in N04.

37

Page 39: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

6.1. The concept of the defineddaily dose (DDD)

Definition; DDDs per1000 inhibitants per day; DDDs per 100 bed-days; DDDs per inhibitant per year

The historical development of the concept of thedefined daily dose (DDD) and its early applicati-ons are described in the Preface.

The DDD is the assumed average maintenancedose per day for a drug used for its main indi-cation in adults.

It should be emphasized that the DDD is a unitof measurement and does not necessarily corres-pond to the recommended or prescribed dailydose (PDD). Doses for individual patients andpatient groups will often differ from the DDD asthey must be based on individual characteristics(e.g. age and weight) and pharmacokinetic consi-derations.

The DDD is often a compromise based on areview of the available information about dosesused in various countries. The DDD may evenbe a dose that is seldom prescribed, because it isan average of two or more commonly used dosesizes.

Drug utilization figures should ideally be pre-sented as numbers of DDDs per 1000 inhabitantsper day or, when drug use by inpatients is consi-dered, as DDDs per 100 bed-days. For antiinfec-tives (or other drugs normally used for shortperiods), it is often considered to be most appro-priate to present the figures as numbers of DDDsper inhabitant per year.

These terms are explained below.

DDDs per 1000 inhabitants per daySales or prescription data presented in DDDs per1000 inhabitants per day may provide a roughestimate of the proportion of the study popula-tion treated daily with a particular drug or groupof drugs. As an example, the figure 10 DDDs

per 1000 inhabitants per day indicates that 1% ofthe population on average might receive a certaindrug or group of drugs daily. This estimate ismost useful for chronically used drugs whenthere is good agreement between the averageprescribed daily dose (see below) and the DDD.It may also be important to consider the size ofthe population used as the denominator. Usuallythe general utilization is calculated for the totalpopulation including all age groups, but somedrug groups have very limited use among peoplebelow the age of 45 years. To correct for differ-ences in utilization due to differing age struc-tures between countries, simple age adjustmentscan be made by using the number of inhabitantsin the relevant age group as the denominator.

DDDs per 100 bed-daysThe DDDs per 100 bed-days may be appliedwhen drug use by inpatients is considered. Thedefinition of a bed-day may differ between hos-pitals or countries, and bed-days should beadjusted for occupancy rate. The same definitionshould be used when performing comparativestudies. As an example, 70 DDDs per 100 bed-days of hypnotics provide an estimate of the the-rapeutic intensity and suggests that 70% of theinpatients might receive a DDD of a hypnoticevery day. This unit is quite useful for bench-marking in hospitals.

DDDs per inhabitant per yearThe DDDs per inhabitant per year may give anestimate of the average number of days forwhich each inhabitant is treated annually. Forexample, an estimate of five DDDs per inhabi-tant per year indicates that the utilization is equi-valent to the treatment of every inhabitant with afive-day course during a certain year.Alternatively, if the standard treatment period isknown, the total number of DDDs can be calcu-lated as the number of treatment courses, and thenumber of treatment courses can then be relatedto the total population.

38

Chapter 6: Drug utilization metrics and their applications

Page 40: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

6.2 Prescribed daily dose and con-sumed daily doseThe prescribed daily dose (PDD) is defined asthe average dose prescribed according to a repre-sentative sample of prescriptions. The PDD canbe determined from studies of prescriptions ormedical or pharmacy records. It is important torelate the PDD to the diagnosis on which thedosage is based. The PDD will give the averagedaily amount of a drug that is actually pre-scribed. When there is a substantial discrepancybetween the PDD and the defined daily dose(DDD), it is important to take this into considera-tion when evaluating and interpreting drug utiliza-tion figures, particularly in terms of morbidity.

For drugs where the recommended dosage dif-fers from one indication to another (e.g. theantipsychotics), it is important to link the diag-nosis to the PDD. Pharmacoepidemiologicalinformation (e.g. on sex, age and whether thera-py is mono- or combined) is also important inorder to interpret a PDD.

The PDD can vary according to both the ill-ness treated and national therapeutic traditions.For instance, for the anti-infectives, PDDs varyaccording to the severity of the infection treated.The PDDs also vary substantially between dif-ferent countries, for example, PDDs are oftenlower in Asian than in Caucasian populations.The fact that PDDs may differ from one countryto another should always be considered whenmaking international comparisons.

It should be noted that the PDD does not nec-essarily reflect actual drug utilization. Someprescribed medications are not dispensed, andthe patient does not always take all the medica-tions that are dispensed. Specially designedstudies including patient interviews are requiredto measure actual drug intake at the patient level(i.e. the consumed daily dose).

6.3 Other units for presentationof volumeCommon physical units (e.g. grams, kilogramsand litres), numbers of packages or tablets andnumbers of prescriptions are also used for quan-

tifying drug utilization, but have certain disad-vantages (see below). These units can be appliedonly when the use of a single drug or of well-defined products is evaluated. Problems arise,however, when the utilization of whole druggroups is considered.

Grams of active ingredientIf utilization is given in terms of grams of activeingredients, drugs with low potency will accountfor a larger fraction of the total than drugs withhigh potency. Combined products may also con-tain different amounts of active ingredients fromplain products, and this difference will not bereflected in the figures.

Number of tabletsCounting numbers of tablets does not reflect thevariations in strengths of tablets, with the resultthat low-strength preparations contribute relati-vely more than high-strength preparations to thetotal numbers. Also, short-acting preparationswill often contribute more than long-acting pre-parations.

Numbers of prescriptionsNumbers of prescriptions do not accuratelyreflect total use, unless total quantities of drugsper prescription are also considered. However,counting of prescriptions is valuable in measu-ring the frequency of prescriptions and in evalua-ting the clinical use of drugs (e.g. diagnosis anddosages used).

Although they are useful in making nationalcomparisons it should be noted that none ofthese volume units is usually applicable in cross-national comparisons, as was pointed out duringthe 1969 WHO symposium in Oslo.

6.4 CostDrug use can be expressed in terms of costs (e.g.national currency). Cost figures are suitable foran overall analysis of expenditure on drugs.International comparisons based on cost parame-ters can be misleading and have limited value inthe evaluation of drug use. Price differencesbetween alternative preparations and different

39

Page 41: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

national cost levels make the evaluation difficult.Long-term studies are also difficult due to fluc-tuations in currency and changes in prices.When cost data are used, an increase in the useof cheaper drugs may have little influence on thetotal level of expenditure on drugs, while a shiftto more expensive drugs is more readily noticed.

The trends in drug use measured in cost maytherefore look very different from the same druguse measured in DDDs. As an example, the totaldrug use in Norway from 1987-1999 measuredin cost (Euros) and in DDDs is shown in Figs 6and 7.

6.5 General readingConsumption of drugs. Report of a symposiumin Oslo, 1969. Copenhagen, WHO RegionalOffice for Europe, 1970 (EURO 3102).

Studies in drug utilization: methods and appli-cations. Copenhagen, WHO Regional Office forEurope 1979 (Regional Publications EuropeanSeries No.8).

Bergman U et al. Auditing hospital drug uti-lization by means of defined daily doses per bed-day. A methodological study. European Journalof Clinical Pharmacology, 1980, 17:183-187.

40

0

200

400

600

800

1000

1200

1400

1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999

0

500

1000

1500

2000

2500

3000

1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999

Figure 6 Total sales of drugs in Norway in millions of Euros 1987-1999

Figure 7 Total sales of drugs in Norway in millions of DDDs 1987-1999

mill

Eur

o pa

rmac

y re

tail

pric

em

ill D

DD

Page 42: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

Baksaas I. Patterns in drug utilization - nationaland international aspects: antihypertensive drugs.Acta Medica Scandinavica, 1984, suppl.683:59-66.

Lee D, Bergman U. Studies of drug utiliza-tion. In: Strom B ed. Pharmacoepidemiology,3rd ed. Chichester, J Wiley, 2000:463-481.

6.6 Exercises1. Assign DDDs for the two antibacterials

below according to the following doserecommendations.Substance A: 500 mg on first day, then 250mg daily; duration of treatment 14 days.Substance B: 500 mg on first day, then 250mg daily; duration of treatment five days.

2. The DDD for budesonide inhalation powderwas changed from 0.3 mg to 0.8 mg in 1991.The following sales figures from Norway forbudesonide inhalation powder are found intwo different books on drug statistics.

1990 9.6 DDDs /1000 inhabitants/day (DDD = 0.3 mg)

1994 11.6 DDDs /1000 inhabitants/day (DDD = 0.8 mg)

Comment on the comparability of these figu-res. Discuss how to best present the salesfigures from these two years in the same article.

3. Annual sales figures given in millions of DDDs are:

Substance A Substance B1988 1.7 21.61996 9.1 9.9

Total number of inhabitants: 4 million

Calculate total number of four-day courses ofsubstance A sold per year and the equivalent number of courses per inhabitant.

Calculate total number of eight-day courses ofsubstance B sold per year and the equivalent number of courses per inhabitant.

41

Page 43: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

Chaper 2 - Types of drug use information

2.1. AmoxicillinThe use of a drug expressed as DDD/1000 population/day is derived by calculating the overallamount of a drug being used over a specified period of time (e.g. a year) and dividing this bythe DDD multiplied by the population and the number of days in the period.

DDD/1000 population/day = Amount used in 1 year (mg)*1000DDD(mg) x population x 365(days)

The amount used is a function of the number of prescriptions, the number of tablets or capsulesper prescription and the dose size of the tablets or capsules.

An understanding of the above allows the following hypotheses about the reasons for increa-sed use to be generated and tested.

Hypothesis 1

The number of prescriptions per year has increased. The information needed to test this hypothesis would be prescription numbers per year adjustedfor population changes over the study period. Remember that the DDD/1000 population/day iscorrected for population changes. Another way of addressing this hypothesis for a drug mainlyused acutely in short courses would be to obtain data expressed as the number of amoxicillintreatment courses/1000 population/year.

If the prescription rate has increased, questions could be asked about the reasons for this.

Have the indications changed? This would require data over time on the indications for which amoxicillin is used. Has there been increased promotion for example, to introduce a new brand? This would requirea survey of promotional materials over time.

Hypothesis 2

The amount of amoxicillin per treatment course has increased. This might be the result of an increase in the average length of the course and/or an increase inthe average PDD. The first possibility could be addressed by a survey of prescribers to find outabout the length of treatment courses, or a survey of prescriptions to calculate the duration oftreatment by dividing the PDD by the total quantity prescribed. To obtain the PDD, a prescrip-tion survey would be required, either designed for this purpose or making use of data from ong-oing surveys such as those conducted by IMS.

2.2. Antidepressant useBoth use and cost have been noted to increase over time. The types of data required to deter-mine the reasons for increased use are similar to those suggested in answer to question 1 withsome differences. In this case, the data on the use of all the antidepressants have been aggrega-ted so the data on the use of the individual agents and groups (TCAs, SSRIs and monoamineoxidase (MAO) inhibitors) will need to be disaggregated. In looking at the relative use ofdrugs or drug groups, it may be necessary to use both DDD/1000 population/day and prescrip-

42

Chapter 7: Solutions to the exercises

Page 44: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

tion numbers. To interpret the data fully, it may be necessary to determine the PDDs for eachdrug or drug group. If the relationship between DDD and PDD differs for the different drugs,the trends based on DDD/1000 population/day may be misleading. For example, the use of theSSRIs and moclobemide may have increased, while only a small decrease in TCA use hasoccurred. This would have increased total antidepressant use with a multiplier effect on cost asthe SSRIs and moclobemide are patented and much more expensive than the older drugs. Ifthis is what has occurred a number of questions can be asked.

• Is the incidence or prevalence of depression in the community increasing?• Is there increased awareness of depression by doctors and patients resulting in a

higher proportion of patients with depression being treated?• If so, is this the result of government educational initiatives, or pharmaceutical

promotion aimed at case-finding and enhancement of compliance.• Are there changes in the doses being used or in the duration of treatment?• Has there been a change in the indications for which antidepressant drugs are

used? • For example, has there been an increase in their use for the treatment of

obsessive–compulsive disorder, panic attacks or chronic pain?

Different types of data will be required to answer some of these questions and special surveyswill have to be designed and carried out. Some information (on indications, dose and durationof treatment) may be available from ongoing prescriber surveys carried out either by academicunits or by commercial sources such as IMS. Data on the incidence and prevalence of depres-sion may be available from government disease registries or similar sources. Qualitative studi-es may need to be designed and carried out to determine for example, the degree of awarenessof depression as a problem and the sources that have been used to obtain information aboutdepression and its treatment.

Cost is a function of price and volume. The issue of volume has been addressed above. Afull assessment of the reasons for cost increases will require information on the price trends forthe drugs over time.

Questions about changes in utilization of a drug or drug group over time require a number ofdifferent types and sources of data.

Chaper 4 - Economic aspects of drug use (pharmacoeconomy)

4.1. Comparison of antihypertensivesThe goal of treating hypertension in terms of health outcomes is to prolong life by preventingcardiovascular events and target organ damage. This is achieved by lowering blood pressure toa range where absolute cardiovascular risk is essentially reduced to the population level. Thereduction in blood pressure is a surrogate outcome measure, but is accepted by most regulatoryauthorities for registration purposes. All the drug groups lower blood pressure to approximate-ly the same extent. Outcome studies are available for diuretics, beta-blockers and for the ACEinhibitors, but not for the alpha-antagonists. In terms of subsidy listing, a principle should bethat, to achieve a price premium, a new drug should have demonstrated an increased benefit interms of health outcomes.

The company argues that this is a new innovative treatment that has been shown to be equiva-

43

Page 45: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

lent to losartan and they therefore demand a price equivalent to the A2 antagonists.You reply that this is just another alpha-antagonist and it should therefore be compared with

prazosin. The company states that there are no head-to-head trials of the new agent against prazosin

(they have not done any trials and have no intention of doing so) and therefore no evidence onwhich to base a comparison. They argue the new agent should be compared with the ACEinhibitors and A2 antagonists where there are good comparative data.

You reply that the lack of data comparing the new agent with prazosin is their problem, andthat if they want a higher price they should do the studies to demonstrate a health outcomebenefit over prazosin. Indeed, you wonder why prazosin has a price premium over the diure-tics and beta-blockers and whether this should be reviewed to determine whether the higherprice is justified.

The company now argues that the new innovative drug has a longer half-life than prazosin sothat it can be administered once a day compared to twice a day for prazosin. It would thereforeimprove compliance which is a very important consideration in treating hypertension.

You reply that the company has not demonstrated that the once-daily dose leads to improvedcompliance or health outcomes and there is little evidence to support this supposition. A smallpremium might be considered for the extra convenience for patients who are taking a life-longtreatment when they are essentially without symptoms.

The company decides not to proceed with the marketing of the new drug.Who is right in this story? What price would you offer for this drug? Are you concerned that itwon’t be available?

4.2. Thrombolytics for acute myocardial infarction

a. Of 1000 patients treated with a placebo,150 will die.

Of 1000 patients treated with Drug A (Thrombase), 100 will die, therefore 50 lives will be saved.

Of 1000 patients treated with Drug B (Klotgon) 70 will die, therefore 80 lives will be saved.

b. Treatment with Thrombase

If the budget is $200 000 and the cost of treatment is $200 per patient ($2000/$200), 1000 pati-ents could be treated and 50 lives saved (see question 1 above).

Treatment with Klotgon

If the budget is $200 000 and the cost of treatment is $1000 per patient ($200 000/$1000), 200patients could be treated and 80 x 200/1000 = 16 lives could be saved.

c. If 1000 patients are treated with Thrombase, 50 lives are saved.

ICER (Thrombase versus placebo for 1000 patiens)= (1000 x $200-1000 x $0)50 lives saved

= $200 000 = $4000 per life saved50

If 1000 patients are treated with Klotgon, 80 lives are saved.

44

Page 46: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

ICER (Klotgon versus placebo for 1000 patients)

= (1000 x $1000 – 1000 x $0) = $1 000 000 = $12 500 per life saved.80 lives saved 80

d. If 1000 patients are treated with Thrombase, 50 lives are saved. Assuming an increase insurvival time of eight years per patient, 50 x 8 = 400 life-years are gained.

ICER (Thrombase versus placebo for 1000 patients) = (1000 x $200 – 1000 x $0) = $200 000 = $500 per life-year gained.

400 life-years 400

If 1000 patients are treated with Klotgon, 80 lives are saved. Assuming an increase in survivaltime of eight years per patient, 80 x 8 = 640 life-years are gained.

ICER (Klotgon versus placebo for 1000 patients)

= (1000 x $1000 – 1000 x $0) = $1 000 000 = $1 562.50 per life-year gained.640 life-years 640

e. If 1000 patients are treated with Thrombase, 50 lives are saved; if 1000 patients are treatedwith Klotgon, 80 lives are saved; therefore, 30 lives are saved by treatment with Klotgon rather than Thrombase. Assuming an increase in survival time of eight years per patient, 30 x 8 = 240 life-years are gained.

ICER (Klotgon versus Thrombase for 1000 patients)

= (1000 x $1000 – 1000 x $200) = $800 000 = $3 333 per life-year gained.240 life-years 240

4.3. Unfractionated heparin versus low-molecular-weight heparina. Relative risk = 19.8% / 23.3% = 0.85.b. Risk difference = 19.8% / 23.3% = 3.5%

Number of patients who needed to be treated = 1/0.035 = 29 patients.

c. ICER (1000 patients) = (1000 x 72.20) – (1000 x 27.09) = $45 1103.5% x 1000 35

= $1 288.86 per event avoided.

d.. ICER (1000 patients) = (1000 x $72.20) – (1000 x ($27.09 + 5 x $12.40)) = –$16 890

(1000 x 23.3%) – (1000 x 19.8%) 35

Low-molecular-weight heparin is dominant. It is both cheaper and more effective than unfrac-

45

Page 47: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

tionated heparin when monitoring costs are included.

When a drug is dominant, it is not appropriate to calculate an ICER, as this can produce spuri-ous results. Why do you think this is?

4.4. Celecoxib versus diclofenaca. Relative risk = ((38 + 11) / 212) / ((74 + 23) / 218) = 23% / 44% = 0.52b. Risk difference = 23% – 44% = –21%

Number of patients who have to be treated to prevent a single event = 1/0.21 = 5 patients.c. Dose of celecoxib = 400 mg/day. One pack contains sufficient drugs for 15 days of treat-

ment. The duration of treatment is 24 weeks = 168 days. Therefore, 168/15 = 11.2 packs are required at a cost of 11.2 x$50 = $560 per patient.

Dose of diclofenac = 100-150 mg/day. Assume a conservative dose of 100 mg/day.

One pack contains sufficient drugs for 25 days of treatment. The duration of treatment is 168days. Therefore, 168 / 25 = 6.72 packs are required at a cost of 6.72 x $11.60 = $77.95 perpatient.

ICER (1000 patients) = (1000 x $560) – (1000 x $77.95) = $482 050 440 – 230 210

= $2 295.48 per ulcer avoided.

d. Incremental cost per ulcer or erosion avoided

ICER (1000 patients)= (1000 x $560 + 1000 x 23% x 1% x $1434) – (1000 x $77.95 + 1000 x 44% x 1% x $1434)

(1000 x 0.44) – (1000 x 0.23)

= $479 038.60 = $2281.14 per ulcer or erosion avoided.210

Incremental cost per hospitalization avoided

ICER (1000 patients)= (1000 x $560 + 1000 x 23% x 1% x $1434) – (1000 x $77.95 + 1000 x 44% x 1% x $1434)

(1000 x 0.44 x 0.01) – (1000 x 0.23 x 0.01)

= $479 038.60 = $228 113.20 per hospitalization avoided.2.1

Incremental cost per death avoided

ICER (1000 patients)

= (1000 x $560 + 1000 x 23% x 1% x $1434) – (1000 x $77.95 + 1000 x 44% x 1% x $1434)(1000 x 0.44 x 0.01 x 0.1) – (1000 x 0.23 x 0.01 x 0.1)

46

Page 48: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

= $479 038.60 = $2 281 136.20 per death avoided.0.21

4.5. Oral montelukast versus an inhaled steroida/b. There is no “right” answer to this question. What do you think?c. Beclometasone is both cheaper and more effective than montelukast. Therefore

beclometasone is dominant.

Chapter 5: Drug classification systems

1. It is appropriate to assign an additional ATC code in N05A (Antipsychotics) for oral for-mulations of «Neurol», because the main indications for the parenteral and oral formula-tions differ. A medicinal product can be given more than one ATC code if it is available in two or more strengths or formulations with clearly different therapeutic uses (see Guidelines for ATC classification and DDD assignment. Oslo, Norway, WHO Collaborating Centre for Drug Statistics Methodology, version 2003.)

2. It is not appropriate to assign an additional ATC code in N04 (Anti-Parkinson drugs) for lisuride because the dosages overlap with those used in prolactin inhibition.

Chapter 6: Drug utilization metrics and their applications

1. Substance A: 250 mgSubstance B: 300 mg

2. To make the sales figures comparable, it is important to recalculate the figures to reflect the same DDD version. The most recent DDD version should always be used (i.e. 0.8 mg for budesonide inhalation powder). Recalculation of the 1990 sales figure with the updated DDD gives 3.6 DDDs/1000 inhabitants/day.

3. Four-day courses, substance A: 1988: 0.43 million courses; 0.1 courses/inhabitant1996: 2.3 million courses; 0.57 courses/inhabitant

Eight days courses, substance B: 1988: 2.7 million courses; 0.68 courses/inhabitant1996: 1.2 million courses; 0.31 courses/inhabitant

47

Page 49: Introduction to Drug Utilization Research · 6 The development of drug utilization research was sparked by initiatives taken in Northern Europe and the United Kingdom in the mid-1960s

This manual has been prepared by the WHOCollaborating Centre for Drug StatisticsMethodology and the WHO CollaboratingCentre for Drug Utilization Research andClinical Pharmacological Services. Particularthanks are extended to all the members andobservers of the International Working Group onDrug Statistics Methodology and the staff of theWHO Collaborating Centre for Drug StatisticsMethodology and in particular to Professor DonBirkett, Professor Peter de Smet, ProfessorDavid Ofori-Adjei, Dr Ingrid Trolin, ProfessorUlf Bergman, Hanne Strøm, Bente Tange Harbøand Marit Rønning. A special thank you toProfessor Folke Sjöqvist for his excellent contri-bution to the editing of the manuscript.

List of abbreviations

ACE inhibitors: angiotensin-converting enzymeinhibitors

AT: anatomical therapeutic (classification)

ATC: anatomical therapeutic chemical (classifi-cation)

CEA: cost-effectiveness analysis

CMA: cost-minimization analysis ICER: incre-mental cost-effectiveness ratio

CUA: cost-utility analysis DURG: WHOEuropean Drug Utilization Research Group

DDD: defined daily dose

DU90%: drug utilization 90%

EPhMRA: European Pharmaceutical MarketResearch Organization

IMS: International Medical Statistics

MAO: monoamine oxidase

NSAIDs: nonsteroidal anti-inflammatory drugs

QUALY: quality-adjusted life-year

SSRIs: selective serotonin reuptake inhibitors

TCAs: Tricyclic antidepressants

48

Acknowledgements