PHARMACOEPIDEMIOLOGYAND PRESCRIPTION

Changes in drug prescribing to Italian community-dwellingelderly people: the EPIFARM–Elderly Project 2000–2010

Carlotta Franchi & Mauro Tettamanti & Luca Pasina &

Codjo Djade Djignefa & Ida Fortino & Angela Bortolotti &Luca Merlino & Alessandro Nobili

Received: 13 August 2013 /Accepted: 26 November 2013 /Published online: 8 January 2014# Springer-Verlag Berlin Heidelberg 2014

AbstractPurpose To investigate the changes in the last decade (2000–2010) in drug prescribing among community-dwelling elderlypeople aged 65–94 years, in relation to age and sex.Methods We analyzed the data of nearly two million subjectsranging in age from 65 to 94 years recorded in the DrugAdministrative Database of the Lombardy Region (Italy) from2000 to 2010. Associations between drug use (at least onedrug, one chronic drug, polypharmacy or chronicpolypharmacy) and age, sex, and year of prescription wereanalyzed by logistic regression analysis. We also analyzeddifferences in changes linked to sex and age.Results Between 2000 and 2010, the prescriptions of at leastone drug or one chronic drug increased by 2 % (from 88.0 to90.3 %; p<0.0001) and 8 % (from 73.8 to 82.0 %; p<0.0001),respectively, while the mean number of packages/person/yearrose from 34.6 [standard deviation (SD) 32.4] to 48.5 (SD42.2). During this same period, there was a 10 % increase inthe prevalence of elderly people exposed to polypharmacy (≥5different active substances) (from 42.8 to 52.7 %; p<0.0001),and the prevalence of those exposed to chronic polypharmacy(≥5 different chronic drugs) doubled (from 14.9 to 28.5 %; p<

0.0001). Males were less frequently treated than females, ex-cept for chronic polypharmacy. People aged ≥80 years showedthe largest increase in all prescribing patterns. Drug consump-tion in ATC groups A, H, and N (women) and in B and C (men)increased most, with the greatest absolute differences occurringin the consumption of proton pump inhibitors (31.1 %), plateletaggregation inhibitors (30.1 %), and statins (23.8 %).Conclusion Prescriptions to community-dwelling elderlypeople have increased substantially during the last 10 years.Although thismight indicate an improvement in care, the largeincrease in the number of elderly people exposed topolypharmacy and chronic polypharmacy should be carefullyanalyzed in terms of quality of care, patient safety, and costs.

Keywords Drug utilization . Polypharmacy . Trend . Elderly

Introduction

The elderly population is increasing throughout the world. InItaly, people aged ≥65 years currently account for 21 % of theoverall population, and it is estimated this proportion willreach 32 % in 2043. [1–3] In the last 10 years, the scenarioof drug therapies and guidelines for treating chronic diseaseshas changed [4–6]. The availability of new drugs, increaseduse of generics and a more active approach to the treatment ofelderly people [7, 8], supported by new diagnostic tools andguidelines that suggest lower thresholds for starting drug[9–13], have given physicians more opportunity—and morepropensity—to start or addmedications to treatment regimens.Multimorbidity [14–16], high rates of hospitalization [17, 18]and an increasing number of specialists who treat every singlechronic disease are other factors that may have contributed tothe increase in drug prescription rates.

However, reports on drug prescription changes are scarceand differ in the presentation of the data and the settings and

Electronic supplementary material The online version of this article(doi:10.1007/s00228-013-1621-6) contains supplementary material,which is available to authorized users.

C. Franchi (*) : L. Pasina : C. D. Djignefa :A. NobiliLaboratory for Quality Assessment of Geriatric Therapies andServices, IRCCS-Istituto di Ricerche Farmacologiche “MarioNegri”, Via Giuseppe La Masa, 19, 20156 Milan, Italye-mail: carlotta.franchi@marionegri.it

M. TettamantiLaboratory of Geriatric Neuropsychiatry, IRCCS-Istituto di RicercheFarmacologiche Mario Negri, Milan, Italy

I. Fortino :A. Bortolotti : L. MerlinoRegional Health Ministry, Lombardy Region, Milan, Italy

Eur J Clin Pharmacol (2014) 70:437–443DOI 10.1007/s00228-013-1621-6

populations analyzed. A Swedish study in the county ofHolland [19] found that between 1988 and 2002 the total drugexposure of the county’s population nearly doubled. In aregister-based study [20], between 2005 and 2009 the preva-lence of polypharmacy (≥5 drugs) increased by 8 %, and theprevalence of excessive polypharmacy (≥10 drugs) increasedby nearly 10 %.

Although data on drug prescribing and expenditure indifferent countries have been available as sales statistics forspecific drugs and populations, no study has specifically fo-cused on the elderly. Specifically, the Italian setting and thetrends in recent years have nog yet been explored. In this studywe investigated the changes in the last decade (2000–2010) indrug prescribing among community-dwelling elderly peopleaged 65–94 years in relation to age and sex.

Methods

Data source and study population

This study is a part of a large pharmaco–epidemiologicalcollaborative project on drug prescription for elderly peopleliving in the Lombardy Region, the EPIFARM–Elderly Project(Progetto EPIdemiologia del FARMaco nell’Anziano). Dataon drug prescribing were obtained from the Drug Administra-tive Database of the Lombardy Region, Northern Italy, whichstores all prescriptions covered by the Italian National HealthSystem (NHS) issued to the nearly ten million individualsliving in the region (around 16 % of the Italian population).The structure of this database, routinely updated for adminis-trative and reimbursement purposes, has been described indetail elsewhere [21, 22]. Briefly, each prescription recordcontains information on the drug dispensed and the patient’sdata. Only drugs provided free of charge by the NHS and filledat the pharmacy are stored. All data used in this study weremanaged according to current Italian laws on privacy, and eachperson was identified by an anonymous code.

We selected all residents of the Lombardy Region aged 65–94 years between 1 January 2000 and 31 December 2010.Individuals who died, were institutionalized or were aged ≥95years in the index year were excluded. People aged ≥95 yearswere excluded due to problems in tracing drugs dispensed bynursing homes. However, in a previous study [23], we esti-mated these individuals accounted for approximately 1 % ofthe overall elderly population of the Lombardy Region.

For each year we calculated the prescription prevalence asthe proportion of all subjects who received at least one drugand as the proportion by age and sex. Chronic drug exposurewas defined as the prescription of at least four packages of adrug of the same active substance, polypharmacy as the pre-scription of five or more different active substances, andchronic polypharmacy as the prescription of five or more

different chronic drugs. Co-prescription was defined as theprescription of a drug belonging to a main ATC group in thesame year in which another drug belonging to a different mainATC group is prescribed.

All drugs were classified according to the Anatomic Ther-apeutic Classification (ATC) system. [24] We analyzed ATCmain anatomical groups (the 1st level) and its subgroups (the2nd and 4th levels) looking for differences between years.We report data for the first year of observation (2000),for the last year a complete dataset was available(2010), and for the year in the middle (2005). Reportedchanges are in absolute terms (percentage).

Statistical analysis

We divided age groups into quinquennials, although decenniawere used in the graphs in order not to cause clutter. Referencegroups were females and the 65- to 69-year age group. Dif-ferences in drug exposure between 2000 and 2010 wereevaluated by multivariable logistic analyses adjusting for ageand sex (model 1). Interactions between prescription preva-lence, sex and age were included in a full model (model 2).Since about 33 % of people were present in both the first andlast years, a clustered sandwich estimator was used to correctfor non-independence of data. To control for the effects of ageand sex on the differences over time, we reported the results ofthe analyses after standardization on the population livingin Lombardy Region in 2005 with very similar results(differences of <0.5 %).

Results

The general characteristics of community-dwelling elderlypeople living in the Lombardy Region in 2000, 2005, and2010 are reported in Table 1. Women were more prevalent(around 60 %) and people aged 65–74 years old were thelargest group (2000–2010: 58.9–53.4 %).

Prescribing patterns from 2000 to 2010, overall,and in relation to age and sex

The prevalence of elderly people who received at least onedrug prescription rose 2 % from 2000 to 2010, while thosereceiving at least one chronic drug increased by nearly 8 %(Table 1). The prevalence of elderly people with no drug, onlyone drug, or two to four drugs fell over time from 12.1 to9.7 %, from 10.7 to 7.9 %, and from 34.4 % to 29.7 %,respectively. The increase in prevalence of elderly peopletreated with at least one drug was similar for females (from89.1 to 91.1 %) and males (from 86.2 to 89.4 %), while theprevalence according to age rose from 86.1 to 87.9 % in those

438 Eur J Clin Pharmacol (2014) 70:437–443

aged 65–74 years, from 90.7 to 93.4 % in those aged 75–84 years, and from 90.5 to 92.7 % in those aged 85–94 years.

The prevalence of elderly people exposed to polypharmacyincreased by 10 % between 2000 and 2010 (from 42.8 to52.7 %). This was more evident in females (from 44.0 to53.4 %) and subjects aged 75–84 (from 49.0 to 60.8 %) and85–94 (from 51.3 to 63.5 %) years. Finally, the prevalence ofelderly people exposed to chronic polypharmacy doubled inthe period considered (Table 1).

Prescribing patterns from 2000 to 2010 by ATC classification

Electronic SupplementaryMaterial (ESM) Table 1s shows thechanges in drug prescribing to community-dwelling elderlypeople from 2000 to 2010 according to the first level of ATCclassification. Although cardiovascular drugs ©) were themost prescribed drug class over time, the prevalence of pre-scriptions for alimentary tract and metabolism drugs (A),

drugs for blood and blood-forming organs (B), and nervoussystem (N) drugs rose markedly from 2000 to 2010.

The main changes in drug prescribing among the first levelof ATC classification involved drugs belonging to classes Band C for males and A, H, and N for females (ESM Fig. 1s).No differences were found in relation to age. In 2000 cardio-vascular system drugs (C) were co-prescribed mainlywith A, B, M, and N ATC groups. This was confirmedin 2010, at which time the prevalence of these co-prescriptions had increased (A/C: from 29.6 to 41.8 %;B/C: from 21.0 to 36.2 %; M/C: from 28.1 to 28.6 %;N/C: from 7.4 to 19.8 %). ESM Table 2s shows the topten of the changes at the second level of the ATCclassification. Proton pomp inhibitors (A02BC), plateletaggregation inhibitors excluding heparin (B01AC), andstatins (C10AA) showed the biggest increases, namely,31.1, 30.1 and 23.8 %, respectively. Among the nervoussystem drugs, selective serotonin reuptake inhibitors(SSRIs) (N06AB) showed the highest increase (8.7 %).

Table 1 Main details of elderly people aged 65–94 years living in the Lombardy Region in 2000, 2005, and 2010

Variables 2000 2005 2010

n % n % n %

Total population of Lombardy Region 9,256,127 100 9,719,718 100 10,155,949 100

Elderly population (65–94 years)a 1,557,560 17.0 1,776,297 18.3 1,930,855 19.0

Age (years), mean (±SD) 75.1 (± 6.9) 75.3 (±6.9) 76.1 (±6.8)

Female 941,106 60.4 1,044,253 58.8 1,113,400 57.7

Male 616,454 39.6 732,044 41.2 817,455 42.3

Age groups (years)

65–69 495,296 31.8 557,080 31.4 508,881 26.4

70–74 421,355 27.1 455,344 25.6 521,894 27.0

75–79 325,846 20.9 363,954 20.5 403,308 20.9

80–84 154,535 9.9 250,608 14.1 289,106 15.0

85–89 119,700 7.7 97,249 5.5 161,950 8.4

90–94 40,828 2.6 52,062 2.9 45,716 2.4

No drug 187,764 12.0 197,409 11.1 185,683 9.7

At least one chronic drugb 1,148,741 73.8 1,365,099 76.8 1,581,059 82.0

Polypharmacy (≥5 drugs)c 666,125 42.8 822,930 46.3 1,018,413 52.7

Chronic polypharmacy (≥5 chronic drugs)d 231,672 14.9 352,638 19.8 551,170 28.5

Total prescriptions to the elderly 21,439,913 31,013,858 43,877,720

Total packages to the elderly 47,376,354 61,132,221 84,729,502

Total active substances to the elderly 979 1,157 1,142

Number of prescriptions/person/year, mean (±SD) 13.9 (±14.8) 17.6 (± 17.8) 22.8 (± 21.8)

Number of packages/person/year, mean (±SD) 34.6 (±32.4) 38. 7 (± 35.8) 48.5 (± 42.2)

Number of active substances/person/year, mean (±SD) 4.6 (±4.0) 4.2 (± 5.0) 4.4 (± 5.6)

SD, Standard deviationa Excluding elderly people living in nursing homes or diedbAt least four packages of a drug with the same active substancec Five or more different active substancesd Five or more different chronic drugs (at least four packages of the same active substance)

Eur J Clin Pharmacol (2014) 70:437–443 439

Univariable, multivariable, and interaction analyses

Table 2 summarizes the univariable, multivariable and interac-tion analyses. In comparison to 2000, community-dwellingelderly people living in the Lombardy Region in 2010 had asignificantly higher risk of being treated with at least one drug.This risk was even greater in the case of chronic treatment.Males were in general less treated than females (one drug, onechronic drug or polypharmacy) but, as shown by the interac-tions analysis, their drug consumption increased more than forfemales. Conversely, males had more frequently chronicpolypharmacy in 2000 and 2010 (OR 1.33; 95%CI 1.32–1.34).

Prescription prevalence reached its maximum in patientsaged 80–84 years with the prescribing pattern of one drug orone chronic drug and in those aged 85–89 years with theprescribing pattern of chronic polypharmacy. Individuals inthe age class 80–84 years showed the maximum increase inprescription prevalence for at least one drug/one chronic drug,

and those in the older age classes (≥80–84 years) showed themaximum increase in prescription prevalence forpolypharmacy. Finally, in chronic polypharmacy, the increasecontinued to rise with advancing age.

Discussion

To our knowledge only a few studies have recently investi-gated the changes in prescribing trends in the elderly duringthis last decade. The results of our study indicate that theprevalence of drug prescriptions issued to community-dwelling elderly people aged 65–94 years increased from2000 to 2010 for both overall use of drugs (at least one drug)and chronic treatments (at least 4 packages of the same activesubstance) and polypharmacy (≥5 different active substances),independently of the age structure of the elderly population.Although males were always less exposed than females to

Table 2 Univariable, multivariable and interaction analyses results

Variables At least one drug At least one chronic druga Polypharmacyb Chronic polypharmacyc

Univariable analyses

Year

2000 Reference Reference Reference Reference

2010 1.29 (1.28–1.30) 1.61 (1.60–1.62) 1.49 (1.48–1.50) 2.28 (2.27–2.30)

Multivariable analyses (model 1)

Year

2010 1.24 (1.23–1.25) 1.54 (1.53–1.55) 1.44 (1.43–1.45) 2.19 (2.18–2.20)

Sex

Female Reference Reference Reference Reference

Male 0.83 (0.83–0.84) 0.97(0.96–0.97) 0.97 (0.97–0.98) 1.33 (1.32–1.34)

Age group (years)

65–69 Reference Reference Reference Reference

70–74 1.44 (1.43–1.45) 1.48 (1.47–1.49) 1.42 (1.42-1.43) 1.53 (1.52–1.54)

75–79 1.93 (1.92–1.95) 2.07 (2.05–2.08) 1.93 (1.92–1.94) 2.24 (2.23–2.26)

80–84 2.24 (2.21–2.27) 2.50 (2.47–2.52) 2.31 (2.29–2.32) 2.81 (2.79–2.84)

85–89 2.00 (1.97–2.03) 2.40 (2.37–2.43) 2.32 (2.30–2.34) 2.96 (2.93–2.99)

90–94 1.54 (1.51–1.58) 1.99 (1.96–2.03) 2.11 (2.09–2.15) 2.76 (2.71–2.80)

Interaction between year and sex or age (model 2)

Males 1.07 (1.06–1.09) 1.06 (1.05–1.07) 1.07 (1.06–1.08) 1.00 (0.99–1.01)

Age 70–74 years 1.10 (1.08–1.11) 1.08 (1.07–1.10) 1.06 (1.05–1.07) 1.02 (1.01–1.04)

Age 75–79 years 1.28 (1.25–1.30) 1.28 (1.26–1.30) 1.22 (1.20–1.23) 1.14 (1.12–1.16)

Age 80–84 years 1.33 (1.30–1.37) 1.37 (1.35–1.40) 1.30 (1.28–1.32) 1.23 (1.20–1.25)

Age 85–89 years 1.21 (1.17–1.24) 1.31 (1.28–1.34) 1.30 (1.28–1.32) 1.28 (1.26–1.31)

Age 90–94 years 1.16 (1.11–1.22) 1.29 (1.25–1.34) 1.30 (1.26–1.34) 1.32 (1.28–1.36)

OR, Odds ratio; CI, confidence interval

Data are presented as the odds ratio with the 95 % confidence interval given in parenthesis

Model 1: Logistic regression analysis adjusted for age and sex. Model 2: Model 1 plus interaction between year and sex or agea At least four packages of a drug with the same active substanceb Five or more different active substancesc Five or more different chronic drugs (at least four packages of the same active substances)

440 Eur J Clin Pharmacol (2014) 70:437–443

drugs (except for chronic polypharmacy), the changes be-tween 2000 and 2010 in prescribing prevalence in the malesubjects were larger than those in the female subjects. A Swedishstudy that compared the patterns of drug use during a 15-yearperiod (1988–2002) for all ages (from 0 to ≥80 years) found thatdrug exposure for the entire study population doubled during thestudy period [19] and that the number of prescriptions for thoseaged ≥60 years increased by 133 %. Another study found thatboth the crude number of prescription claims and prescriptionrates in adults aged ≥65 years increased dramatically over aperiod of 10 years (1997–2006) in Ontario, with the greatestchanges occurring in females aged ≥85 years [25].

To study whether there was an association between age classand increased drug exposure, we used a multivariable analysisand found an age-related increasing risk to receive drugs that wasparticularly evident for chronic drugs and chronic polypharmacy.In patients exposed to at least one drug or to at least one chronicdrug the maximum increase was seen in the age group of 80–84years. In a previous study on prescription data for the 2005, wefound that age was the most important predictor of exposure tochronic therapies and polypharmacy [18]. In the present study, inwhich we standardized the prevalence of the elderly populationaccording to the population residing in the Lombardy Region in2005, the results obtained were almost identical, indicating thatthe observed changes in the prevalence of drug exposure be-tween 2000 and 2010 cannot be explained by a change in the agestructure of the population.A possible explanation of the increasein the number of drug prescriptions might be the increase in theprevalence of diagnoses of chronic diseases and multimorbidityin these individuals, which could have induced physicians toprescribe more drugs according to the specific guidelines foreach single chronic disease. Another explanation of the increasesin drug prescribing across our 10-year study period might be theimplementation of new guidelines for the treatment of manychronic diseases (e.g., diabetes, hypertension, osteoporosis) thatlowered the thresholds for starting drug treatments. This latterphenomenon is evidenced by the increase in prescriptions for thedrug classes of the cardiovascular system (C), blood and blood-forming organs (B), alimentary tract and metabolism (A) andcentral nervous system (N). The small reduction in prescribingfor musculoskeletal drugs (M) might be related to safety issues(particularly the increased risk of cardiovascular events, gastro-intestinal bleedings, and impaired renal function) for coxibs andother non steroidal anti-inflammatory drugs. The largest in-creases in prescribing drugs in the therapeutic subgroups between2000 and 2010 were for the lipid-modifying agents (C10),followed by antithrombotic agents (B01), and agents acting onthe renin–angiotensin system (C09). Silwer et al. [19] reportedthe largest changes in B01, from approximately zero to 43prescriptions per 100 inhabitants, while C09 prescriptions in-creased tenfold, calcium channel blockers (C08) prescriptionstripled, and C10 prescriptions increased from approximately zeroto 14 defined daily dose per 100 inhabitants. In another study,

Bajcar et al. [25] found that all classes of medication showed anincrease in average annual claims per person, with the top tenbeingmedications for osteoporosis (+1.965%), lipid-lowering (+572 %), thyroid replacements (+200 %), psychotropics (+198 %), cardiovascular (+191 %), diabetes (+188 %), gastroin-testinal (+155 %), narcotics/analgesics (+78 %), and corticoste-roids (+44 %).

In our study analgesics (particularly opioids—data not shown)and psychoanaleptics (particularly antidepressants—data notshown) showed the highest increase, occupying the sixth andseven positions, respectively, among the top ten therapeuticsubgroups (3rd ATC level). These changes might be related tochanges in the Italian NHS reimbursement policy. One studyshowed an increased prevalence in antidepressant (AD) use from2000 to 2007, especially for SSRIs [26], that could be explainedby the fact that the Italian NHS started to fully reimburse SSRIsfrom 2001 but also by the persistence of the treatment and thebroader indications for antidepressants [26].

Despite these increases in many different drug therapies, thereare still only few clinical trials routinely enrolling elderly patients,specifically those aged ≥75 years. Moreover, clinical guidelinesbased on the same trials contribute to the problem by focusing onthe benefits of individual medications on specific diseases, with-out considering the risks and the frequent lack of compliancewith treatments for multiple coexisting conditions [27].

Doctors are now faced with the contradiction of being ableto treat patients with new therapeutic agents on the marketwhile at the same time having to assess and control an in-creasing number of potential harmful drug–drug interactions,adverse drug reactions, and inappropriate use of medications,all factors with important implications for quality of care,safety, and cost [28].

We also found that although community-dwelling elderlypeople living in the Lombardy Region showed an increase inthe number of prescription claims and packages per person,the number of active substances per person per year remainedstable over time. This could be partly explained by newformulations of some drugs during the study, but more prob-ably by a change in chronic treatment since chronic prescrip-tion prevalence increased by 8 % compared to 2 % for non-chronic treatment. The 10 % increase in the prevalence ofpolypharmacy between 2000 and 2010 and the doubling ofthe prevalence of chronic polypharmacy supports this hypoth-esis. Further analyses will help clarify the reason for thischange in chronic treatment over the years.

Limitations

Although we analyzed a large regional dataset that includeddata on nearly two million people aged ≥65 years, this studyhas some limitations. First, our findings may not be fullyrepresentative of other regions. However, data from the Na-tional Drug Monitoring Center of the Italian Drug Agency

Eur J Clin Pharmacol (2014) 70:437–443 441

shows that the prescribing profile for the elderly in the Lom-bardy Region is comparable to that at the national level.Second, common to many studies based on administrativeprescription databases, there is a lack of information on thediseases and the indication for which the drugs are prescribed.However, drug prescribing has been used in many studies as aproxy for identifying comorbidity and disease-specific riskfactors, as well as for evaluating the burden of diseases inthe elderly population [29, 30]. Third, there is a lack of data ondrugs not reimbursed by the NHS and over-the-counter med-ications (not including drugs commonly used, such as benzo-diazepines, peripheral vasodilators, vitamins, laxatives, andseasonal drugs). This could result in an under-estimation ofpolypharmacy and of changes in drug use. Fourth, informa-tion on effective drug intake is absent. However, at least forchronic drugs, the periodic prescription may serve as anindirect indicator of regular drug intake. Finally, some chang-es in drug prescribing during the study periodmight have beeninfluenced by regulatory decisions on safety and efficacyimplemented by the Italian Drug Agency, such as those rela-tive to coxibs, non-steroidal anti-inflammatory drugs, andantipsychotics, to mention only the most common.

Conclusions

The results of this study show that exposure to drug therapies(overall use, chronic, and polypharmacy) of community-dwelling elderly people living in the Lombardy Region(Italy) rose significantly from 2000 to 2010. These changesmight have been influenced by the presence of new drugs andby increases in the diagnosis of chronic disease andmultimorbidity which could have induced physicians to pre-scribe more drugs, following new specific guidelines andlower cutoffs for starting treatment. These developments wereparticularly evident for exposure to chronic drugs and chronicpolypharmacy and for patients aged ≥75 years.

These new trends have important implications in terms ofmonitoring costs and the appropriateness of drug prescribingin these patients, and on the need to educate physicians andcreate interdisciplinary teams involving pharmacists, nurses,and social workers, for periodic critical reassessment of drugprofiles of elderly patients exposed to polypharmacy

Competing interests None.

Funding This study was supported by grants from the Region HealthMinistry of the Lombardy Region (Progetto ‘EPIdemiologia dei FARM-aci’ – EPIFARM).

Ethics approval All data were managed according to the current Italianlaw on privacy.

References

1. Istituto nazionale di statistica (ISTAT). Il futuro demografico del Paese.Previsioni generali della popolazione residente al 2065. Available at:http://www.istat.it/it/archivio/104818. Accessed 20 Nov 2012

2. Istituto nazionale di statistica (ISTAT). Natalità e fecondità dellapopolazione residente. Available at: http://www.istat.it/it/files/2012/11/nati-2011.pdf?title=Natalità+e+fecondità+della+popolazione+-+14%2Fnov%25. Accessed 20 Nov 2012

3. Nie JX, Wang L, Tracy CS, Moineddin R, Upshur RE (2008)Health care service utilization among the elderly: findingsfrom the Study to Understand the Chronic ConditionExperience of the Elderly and the Disabled (SUCCEED pro-ject). J Eval Clin Pract 14:1044–1049

4. Boyd CM, Darer J, Boult C, Fried LP, Boult L, Wu AW (2005)Clinical practice guidelines and quality of care for older patients withmultiple comorbid diseases: implications for pay for performance.JAMA 294(6):716–724

5. American Geriatrics Society Expert Panel on the Care of OlderAdults with Multimorbidity. Guiding principles for the care of olderadults with multimorbidity: an approach for clinicians. Available at:http://www.americangeriatrics.org/annual_meeting/2012_meeting_handouts/amhandouts050512/guiding_principles050512. Accessed30 May 2012

6. Fried TR, Tinetti ME, Iannone L (2011) Primary care clinicians’experiences with treatment decision making for older persons withmultiple conditions. Arch Intern Med 171(1):75–80

7. Gurwitz JH (2004) Polypharmacy: a new paradigm for quality drugtherapy in the elderly? Arch Intern Med 164(18):1957–1959

8. Rosenthal MB, Berndt ER, Donohue JM, Frank RG, Epstein AM(2002) Promotion of prescription drugs to consumers. N Engl J Med346:498–505

9. Chobanian AV, Bakris GL, Black HR et al (2003) The seventh reportof the Joint National Committee on Prevention, Detection,Evaluation and Treatment of High Blood Pressure: the JNC 7 report.JAMA 289:2560–2572

10. Masoudi FA, Krumholz HM (2003) Polypharmacy and comorbidityin heart failure. BMJ 327:513–514

11. Lange RA, Hillis LD (2004) Antiplatelet therapy for ischemic heartdisease. N Engl J Med 350:277–280

12. Garber AJ (2003) Benefits of combination therapy of insulin and oralhypoglycemic agents. Arch Intern Med 163:1781–1782

13. Tariot PN, Farlow MR, Grossberg GT, Graham SM, McDonald S,Gergel I (2004) Memantine treatment in patients with moderate tosevere Alzheimer disease already receiving donepezil: a randomizedcontrolled trial. JAMA 291:317–324

14. AndersonGF (2005)Medicare and chronic conditions. N Engl JMed353:305–309

15. WehlingM (2009)Multimorbidity and polypharmacy: how to reducethe harmful drug load and yet add needed drugs in the elderly?Proposal of a new drug classification: fit for the aged. J Am GeriatrSoc 57(3):560–561

16. Townsed A, Hunt K,Wyke S (2003) Managingmultiple morbidity inmid-life: a qualitative study of attitudes to drug use. BMJ 327:837

17. Corsonello A, Pedone C, Corica F, Incalzi RA (2007) Polypharmacyin elderly patients at discharge from the acute care hospital. Ther ClinRisk Manag 3(1):197–203

18. Nobili A, Marengoni A, Tettamanti M, Salerno F, Pasina L, FranchiC, Iorio A, Marcucci M, Corrao S, Licata G, Mannucci PM (2011)Association between clusters of diseases and polypharmacy in hos-pitalized elderly patients: results from the REPOSI study. Eur J IntMed 22(6):597–602. doi:10.1016/j.ejim.2011.08.029

19. Silwer L, Lundborg CS (2005) Patterns of drug use during a 15 yearperiod: data from a Swedish county, 1988–2002. PharmacoepidemiolDrug Saf 14(11):813–820

442 Eur J Clin Pharmacol (2014) 70:437–443

20. Hovstadius B, Petersson G (2013) The impact of increasingpolypharmacy on prescribed drug expenditure-a register-based studyin Sweden 2005–2009. Health Policy 109(2):166–174

21. Franchi C, Sequi M, Bonati M, Nobili A, Pasina L, Bortolotti A, FortinoI, Merlino L, Clavenna A (2011) Differences in outpatient antibioticprescription in Italy’s Lombardy region. Infection 39:299–308

22. Franchi C, Tettamanti M, Marengoni A, Bonometti F, Pasina L,Cortesi L, Fortino I, Bortolotti A, Merlino L, Lucca U, Riva E,Nobili A (2012) Changes in trend of antipsychotics prescription inpatients treated with cholinesterase inhibitors after warnings fromItalian Medicines Agency. Results from the EPIFARM-ElderlyProject. Eur Neuropsychopharmacol 22(8):569–577

23. Nobili A, Franchi C, Pasina L, Tettamanti M, Baviera M, Monesi L,Roncaglioni C, Riva E, Lucca U, Bortolotti A, Fortino I, Merlino L(2011) Drug utilization and polypharmacy in an Italian elderly pop-ulation: the EPIFARM-elderly project. Pharmacoepidemiol Drug Saf20:488–496

24. The Anatomical Therapeutic Chemical (ATC) classification. Aboutthe ATC/DDD system. WHO Collaborating Centre for DrugStatistics Methodology. Available at: http://www.whocc.no/atcddd/.Accessed 20 Nov 2012

25. Bajcar JM, Wang L, Moineddin R, Nie JX, Tracy CS, Upshur RE(2010) From pharmaco-therapy to pharmaco-prevention: trends inprescribing to older adults in Ontario, Canada, 1997–2006. BMCFam Pract 11:75

26. Parabiaghi A, Franchi C, Tettamanti M, Barbato A, D’Avanzo B,Fortino I, Bortolotti A, Merlino L, Nobili A (2011) Antidepressantsutilization among elderly in Lombardy from 2000 to 2007: dispens-ing trends and appropriateness. Eur J Clin Pharmacol 67(10):1077–1083

27. Tinetti ME, Bogardus ST Jr, Agostini JV (2004) Potential pitfalls ofdisease-specific guidelines for patients with multiple conditions. NEngl J Med 351(27):2870–2874

28. Califf RM (2009) Rational use of medications: if Canada can’t do it.CMAJ 181(1–2):15–16

29. Malone DC, Billups SJ, Valuck RJ, Carter BL (1999) Developmentof a chronic disease indicator score using a Veterans Affairs MedicalCenter medication database. IMPROVE Investigators. J ClinEpidemiol 52(6):551–557

30. Brilleman SL, Salisbury C (2013) Comparing measures ofmultimorbidity to predict outcomes in primary care: a cross-sectional study. Fam Pract 30(2):172–178

Eur J Clin Pharmacol (2014) 70:437–443 443

Reproduced with permission of the copyright owner. Further reproduction prohibited withoutpermission.