Drug-Related Problems Identified in a Sample of Portuguese Institutionalised Elderly Patients and Pharmacists’ Interventions to Improve Safety and Effectiveness of Medicines
Drugs - Real World Outcomes
Drug-Related Problems Identified in a Sample of Portuguese Institutionalised Elderly Patients and Pharmacists' Interventions to Improve Safety and Effectiveness of Medicines
Filipa Alves da Costa 0 1 2 3 4 5 6 7 8
Lu´ısa Silvestre 0 1 2 3 4 5 6 7 8
Catarina Periquito 0 1 2 3 4 5 6 7 8
Clara Carneiro 0 1 2 3 4 5 6 7 8
Pedro Oliveira 0 1 2 3 4 5 6 7 8
Ana Isabel Fernandes 0 1 2 3 4 5 6 7 8
Patr´ıcia Cavaco-Silva 0 1 2 3 4 5 6 7 8
0 Catarina Periquito
1 Lu ́ısa Silvestre
2 & Filipa Alves da Costa
3 ISCSEM, Campus Universita ́rio, Quinta da Granja Monte de Caparica , 2829-511 Caparica , Portugal
4 Registo Oncolo ́gico Regional Sul (ROR-Sul), Instituto Portugueˆs de Oncologia de Lisboa Francisco Gentil , E.P.E, Rua Professor Lima Basto, 1099-023 Lisboa , Portugal
5 Centro de Investigac ̧a ̃o Interdisciplinar Egas Moniz (CiiEM) , Almada , Portugal
6 Ana Isabel Fernandes
7 Pedro Oliveira
8 Clara Carneiro
Background Currently, people live longer but often with poor quality of life. The decrease in healthy life-years is partly attributable to the institution of polypharmacy to treat various comorbidities. Objectives The objectives of the study were to determine the prevalence and nature of drug-related problems (DRPs) in polypharmacy elderly patients residing in nursing homes and to test the acceptability of a pharmacist's intervention. Methods An exposure cohort was constituted in three Portuguese nursing homes, where all polypharmacy (five or
more medicines) elderly patients (C65 years of age) were
analysed and then a random stratified sample was extracted
to be subject to an intervention. Clinical and therapeutic
data were collected and analysed for DRPs and classified
according to the II Granada Consensus, by a pharmacist-led
team. The intervention was the formulation of a
pharmacist’s recommendations to prescribers addressing clinically
relevant DRPs, along with suggestions for therapy changes.
Results The initial sample included 126 elderly patients
taking 1332 medicines, where 2109 DRPs were identified.
The exposure cohort included 63 patients, with comparable
baseline data (p [ 0.005). Manifest DRPs occurred in
31.7 % of the intervention group (mainly quantitative
ineffectiveness–DRP 4), whereas potential DRPs were
identified in 100 % of patients (mainly non-quantitative
unsafe–DRP 5). Amongst the DRPs identified, 584
(56.7 %) were reported to prescribers (all types of DRPs)
and 113 (11 %) to nurses (only non-quantitative
ineffectiveness–DRP 3). A total of 539 pharmacist
recommendations were presented to physicians, corresponding to 62
letters sent by mail, each including an average of 8.7
recommendations to solve DRPs present in intervention group
(IG) patients. There was a high non-response rate (n = 34
letters; 54.8 %; containing 367 pharmacist
recommendations; 68.1 %) and amongst recommendations receiving
feedback, only 8.7 % of pharmacist recommendations
made were accepted (n = 15). Positive responses were
significantly associated with a lower number of
recommendations made, whereas a higher number of
recommendations increased the odds of no response (p \ 0.001).
Conclusion A pharmacist-led medication review proved
useful in identifying DRPs in elderly polypharmacy
nursing home residents. Stronger bonds must be developed
between healthcare professionals to increase patient safety
in the vulnerable institutionalised elderly population.
The Granada Classification for drug-related
problems proved useful in guiding a pharmacist-led
medication review in elderly patients.
Nearly three manifest drug-related problems were
detected in polypharmacy elderly patients, which
could have been prevented.
Collaboration between pharmacists, nurses and
doctors can improve the rational use of medicines.
Effective communication with physicians may be
fostered by reducing the number of pharmacist
For the past decades, people have been living longer.
Portugal is currently one of Europe’s leading countries in
terms of life expectancy with an age index registered in
2014 of 138.6 % [
]. This demographic change is
attributable to various factors, where health technology
plays an important role. Medicines are undoubtedly an
effective tool to promote better health, provided they are
rationally used. The current challenge for healthcare
professionals has shifted from increasing people’s life to
achieving a better quality of life. The fact that Portugal’s
data on healthy life-years after age 65 years do not match
the country ranking in terms of longevity is a cause of
concern. Focus has centred on the use of potentially
inappropriate medication (PIM) in elderly patients, which
is in fact a type of drug-related problem (DRP). Various
tools have been developed to identify PIM and to alert
doctors to the most appropriate treatment, with the ultimate
goal of achieving a more rational pharmacotherapy and
better patient outcomes [
Pharmacist-led medication review is an area, within
pharmaceutical care, currently attracting much attention,
which can be performed using explicit or implicit criteria.
Most of the interventions made in nursing homes use
explicit criteria. These are much simpler to use when a type
1 medication review is undertaken [
], i.e. when all
information is based exclusively on patient medical records
and no information is obtained from the patient. Literature
describing interventions in institutionalised elderly
patients, where a medication review focusses on the
identification of DRP, is very scarce and the few published
studies have used different DRP classifications, making
comparisons very difficult [
]. In Portugal, only three
studies have been found evaluating DRPs in nursing
homes, two of them focussing on the detection of PIM,
either using the STOPP/START criteria [
] or Beers
]. The third study focussed on the use of the
medication regimen complexity index, which measures the
number of medicines and dosages, but not the quality of
pharmacotherapy or its outcome [
]. Internationally, there
are various papers published using different criteria for
identifying PIM, but the Beers criteria are the most used
worldwide. The START/STOPP criteria have been gaining
prominence in Europe [
] and there are reports using the
McLeod criteria [
], Priscus list [
] and even nationally
developed classifications [
]. However, few reports of
pharmacist-led medication reviews focus on the
identification of DRPs, and these have either used the
Pharmaceutical Care Network Europe classification [
] or the
DOCUMENT classification [
]. The Dader method has
been used in nursing homes [
] but, to our knowledge, no
paper has yet described the nature of DRP detected using
the Granada classification to guide pharmacist-led
medication reviews [
]. Therefore, this study aimed at
determining the prevalence of manifest and potential DRPs
in polypharmacy elderly patients residing in nursing homes
and to describe the nature of DRPs using the Granada
classification. Furthermore, the study aimed to test the
acceptability of a pharmacist’s intervention intending to
solve the DRPs detected.
2.1 Study Design
An exposure cohort was constituted in four Portuguese
nursing homes, where all polypharmacy (five or more
medicines) elderly patients (C65 years of age) were
analysed and then a random stratified sample was extracted to
be subject to an intervention. These were compared with
the remaining, meeting the inclusion criteria, to assess
2.2 Sample of Nursing Homes
Four nursing homes were invited to participate in the study,
two located in Alentejo and another two in Lisboa and the
Vale do Tejo region. Nursing homes were chosen using
geographical convenience sampling. Data collection was
primarily ensured by two trainee pharmacists, and each of
them was responsible for two facilities. The location was
chosen according to the place where they were practicing,
as data collection was undertaken during their
pre-registration training. Eligibility criteria were also defined for the
nursing homes, including completeness of fundamental
variables (primary diagnosis and prescribed therapy) in the
2.3 Sample of Patients
Eligible patients were polypharmacy (taking five or more
medicines at the time of study) elderly patients (C65 years
2.4 Study Period
The study started in October 2013. The interventions were
carried out during August and September 2014 and results
evaluated in October 2014.
2.5 Collected Data
All data analysed were retrieved from medical records and
included sociodemographic information and clinical and
therapeutic data (the latter checked in medication charts).
Clinical data collected included: medical diagnoses,
laboratory tests and measurements of clinical biomarkers, and
Medicines prescribed were classified according to the
ATC (Anatomic Therapeutic Chemical) classification
proposed by the World Health Organization [
Patients’ medical conditions, negative outcomes and
medicines used were analysed using information contained
in the medical records and in medication charts.
DRPs were identified by two trainee pharmacists, who
created a database with raw data and their findings. This
database was subsequently reviewed by members of the
research team specialised in pharmaceutical care, involving
two clinical pharmacists and a physician. One of the
reviewers (clinical pharmacist) checked all records, whilst
the other two checked only records where there was
DRP were searched in all patients and, when present,
classified according to the II Granada Consensus [
which divides DRPs into three main categories: necessity,
effectiveness and safety (Fig. 1).
A negative outcome, defined as a manifest DRP, was
considered when the medical record included information
that allowed the researchers to verify that the problem had
actually occurred, such as a raise in blood pressure
(ineffectiveness) or the experience of an adverse drug
reaction (safety). The remaining DRPs were considered as
The intervention consisted of the prioritisation of DRPs, a
report of those identified and considered clinically relevant
in the intervention group to prescribers and nurses, with the
provision of a pharmacist’s recommendation. This
pharmacist’s recommendation when headed for physicians
could include suggestions of therapy changes if
appropriate. Each recommendation could address one or various
DRPs, as the pharmacist’s recommendations could solve or
prevent more than one DRP at the same time. The language
used in the letters sent to physicians did not include the
mention of DRPs in an effort to make it more meaningful
to the target audience.
2.7 Outcome Measures
For all patients, the outcome measures were quantification
and qualification of identified DRPs and their classification
as potential or manifest. For the intervention group, process
measures also included the number of DRPs reported and
the number of pharmacist recommendations made to
physicians; the proportion of pharmacist recommendations
accepted was used as an outcome measure; Fig. 2 depicts
the study schema.
Drug taken presents a (potential or
manifest) problem (DRP)
The drug is
motives (DRP 3)
The drug is
motives (DRP 4)
The drug is
motives (DRP 5)
The drug is
motives (DRP 6)
Fig. 1 Schematic of the II Granada Classification for drug-related problems (DRP)
Patients residing in the nursing homes
Polypharmacy ( 5 medications) elderly (
65 years) (n=161)
Patients with complete medical records
Analysis for nature and prevalence of
DRP and for predictors of DRP (n=126)
Patients’ Inclusion criteria
Facilities’ Inclusion criteria
Random extraction clustered by
Comparison group (n=63)
Used to assess representativeness of
intervention group (table 1)
Intervention group (n=63)
Pharmacist analysed medication
and detected 1002 DRPs
Manifest DRPs and clinically
relevant potential DRPs prioritized
and reported (n=697; 69.6%)
Pharmacists’ Recommendations to
physicians (n=539), written in (n=62 letters)
addressing 584 DRPs*
Pharmacists’ recommendations to nurses
(n=63) addressing 113 DRPs
Response obtained to 28 letter (45.2%),
addressing 172 pharmacist’s
No feedback obtained to 34 letters (54.8%),
addressing 367 pharmacist’s
Positive response to 10 letters (35.7%; at least
one recommendation); acceptance of 15
pharmacist’s recommendations (8.7% from 172)
Negative response to 18 letters (64.3%);
declining all pharmacist’s recommendations
Data analysis was performed using the IBM software SPSS
(IBM SPSS Statistics for Windows, IBM Corp, Armonk,
NY. Released 2012), version 21.0 and the main variables
characterised using univariate analysis (relative and
absolute frequencies for categorical variables and central
tendency and dispersion measures for continuous variables).
Bivariate analysis was used to answer hypotheses
previously established. Because sample distribution was not
normal (Kolmogorov–Smirnov test), mostly
non-parametric tests were used and included the Chi-square test,
Spearman’s correlation coefficient and Mann–Whitney or
Kruskal–Wallis tests, according to the type of variables. A
confidence interval of 95 % was considered for all tests.
All four facilities invited accepted to participate in the
study, totalling 225 elderly patients, of whom 161 patients
met the inclusion criteria (C65 years of age, taking five or
more medications). However, one facility, from Alentejo,
was subsequently excluded because of data
incompleteness, making the final group 126 polypharmacy elderly
patients from three facilities in Portugal.
3.1 Frequency and Nature of DRPs Identified (All
The sample of 126 patients, taking a total of 1332
medicines, comprised mainly female individuals (n = 87;
69 %), with a mean age of 84.8 years [standard deviation
(SD) = 6.1], median of four registered comorbidities and
being treated with over ten medicines (median = 10;
mean = 10.6; SD = 4.2). A median of 14.5 DRPs
(SD = 8.4) were detected per patient, considering both
potential and manifest.
Patients were randomly assigned to the intervention
group using stratified sampling (n = 63), and their
characteristics compared with the control group (n = 63) to
assess representativeness (Table 1). Groups were shown to
be comparable at baseline.
All 1332 medicines being taken by all polypharmacy
patients (n = 126) were classified according to the ATC
classification. Figure 3 depicts the pharmacotherapeutic
classes more frequently involved in polypharmacy. The
three leading groups found accounted for 75 % of the drugs
prescribed: central nervous system (30.5 %),
cardiovascular system (24.5 %) and gastrointestinal tract (19.8 %).
All patients’ demographic and medical characteristics
were explored as potential determinants for polypharmacy.
Data indicate that the number of comorbidities was
moderately but significantly correlated with the number of
prescribed medicines (Table 2).
All patients (n = 126) were assessed for potential or
manifest DRP where a total of 2109 DRPs were identified.
Every patient had at least one DRP. The most commonly
found DRP belonged to the safety domain (n = 922;
43.7 %), followed by necessity (n = 727; 34.5 %) and last
by effectiveness (n = 460; 21.8 %). Considering the second
level of this classification, the order found was: unsafe drug
(non-quantitative, DRP 5) (n = 741; 35.1 %), unnecessary
treatment (DRP 2) (n = 581; 27.6 %), ineffective drug
(nonquantitative, DRP 3) (n = 387; 18.4 %), unsafe drug
(quantitative, DRP 6) (n = 181; 8.6 %), need for additional
treatment (DRP 1) (n = 146; 6.9 %) and last ineffective
drug (quantitative, DRP 4) (n = 73; 3.5 %).
Potential predictors of the occurrence of DRP were
explored but only the number of comorbidities
(Spearman’s r = 0.412; p \ 0.001) and the number of prescribed
medicines (Spearman’s r = 0.766; p \ 0.001) were found
to be directly and significantly correlated, the latter
exhibiting a strong correlation (Fig. 4).
Most of the identified DRPs were potential (n = 2064;
97.9 %) and only 45 were manifest (2.1 %). With regard to
manifest DRPs, the majority was classified as DRP 4 (n = 27;
60 %), followed by DRP 6 (n = 9; 20 %), DRP 1 (n = 5;
11.1 %), DRP 3 (n = 3; 6.7 %) and DRP 5 (n = 1; 2.2 %).
3.2 Intervention Group
Considering only the intervention group (n = 63), a total
of 1002 DRPs were identified, of which, 697 (69.6 %)
DRPs were reported. In total, 584 (83.8 %) DRPs were
reported to the physician and 113 (16.2 %) were reported
SD standard deviation, M mean, MD median
Sex [MD; M (SD)]
Age [Pearson’s correlation]
No. of comorbidities [Spearman’s correlation]
SD standard deviation, M mean, MD median
No. of prescribed medicines
Fig. 3 Drug classes more
frequently involved in
polypharmacy (five or more
medicines) in 126 elderly
(C65 years of age)
Fig. 4 Correlation between the number of prescribed medicines per
patient and the presence of total drug-related problems (DRP,
manifest and potential) in 126 polypharmacy (five or more
medicines), elderly (C65 years of age), institutionalised Portuguese patients
to nursing staff. Reports to nurses were mainly based on
changing administration times. Pharmacist
recommendations to physicians resulted from the analysis of the 584
DRPs detected and included 539 suggestions to solve or
prevent them, which could comprise dosage changes,
suspension or addition of medicines, and request of additional
A total of 62 letters addressing 539 recommendations
were made to physicians regarding the 63 polypharmacy
elderly patients in the intervention group. Each letter
reported 9.2 DRPs on average and contained 8.7
pharmacist recommendations (2–19). Considering all the
pharmacist’s recommendations sent (n = 539), response was
only obtained to 172 of these (31.9 %). From responses
obtained, only 15 pharmacist recommendations were
accepted (8.7 %).
Analysing the number of recommendations made in the
letters where a positive response was obtained, it is
interesting to note that there is a significant association between
these two variables (Table 3).
Although the sample addressed in this study was achieved
using a convenience strategy valuing geographical
location, the sample characteristics were comparable to the
reference population in the same regions, including gender
distribution. The main difference referred to the age
structure, as in the reference population, the larger age
class included those aged from 65 to 69 years and
diminished progressively every 5 years, whereas the study
sample observed the opposite direction because the larger
group was represented by those aged over 85 years [
This fact is not surprising because the sample group used
originated from nursing homes. Additionally, clinical and
therapeutic data were also comparable because the
medicines most consumed in the study sample were those
used in the most prevalent chronic conditions reported in
the National Health Enquiry [
Each elderly patient presented with a median of four
comorbidities registered in the medical file, around half of
the comorbidities reported elsewhere [
patient records are a possible explanation for this finding,
which is consistent with the literature .
The initial sample was prescribed a mean of 11
medicines per day, varying between 5 and 18, which is also
similar to results published previously [
11, 15, 16, 24, 27,
A direct correlation between the number of
comorbidities and the number of prescribed medicines (r = 0.490;
p \ 0.001) was found, as described before [
]. This is not
surprising because, as more health conditions arise, the
higher the need for additional therapy. However, no
relation was found between the number of medicines and age,
contradicting previous findings [
The number of DRPs identified in this study was
considerably higher than what was expected based on previous
15, 26, 27, 30, 31
]. However, it should be noted
that most of these studies did not differentiate between
potential and manifest DRPs and did not use the Granada
classification, thus making comparison impossible. The
fact that very few studies were found using this
classification in nursing homes may be related to a publication
bias because this classification is originally Spanish, and
subsequently validated in Portuguese, whereas others more
frequently found arise from English-speaking countries.
Although the III Granada Consensus has been published for
some time, we opted to use the II Consensus because this is
the one adopted nationally, and the only one validated for
use in Portugal [
19, 20, 32
The more frequently found DRP was DRP 5 (35.1 %),
consistent with what has been described as ‘‘risk of adverse
drug reactions’’ [
] and as ‘‘potential interactions’’ [
The second more frequent DRP was DRP 2, similar to what
has been reported as ‘‘medicine with unclear indication’’
]. However, it should be noted that there was an
enormous difference between potential and manifest DRPs,
with the latter representing only 2 % of all DRPs detected.
When analysing the types of DRPs, it becomes clear that an
important proportion is attributable to DRP 5, indicating
that most of the potential interactions detected by
pharmacists are unlikely to happen. However, it could also be
that they had not yet happened at the time, and to check it
we would need a longer follow-up. This has been reported
elsewhere  and thus the prioritised interventions
focussed on manifest DRP or those considered clinically
significant, which represented 70 % of the total DRPs
The possibility of information bias may not be
disregarded as some patient files may not be totally updated or
complete, as previously described [
]. This bias, if
present, will have particular influence on DRP 2 detected,
which may not be real but rather a result of missing
information. In regard to DRP 5, it should be noted that
there was no record found for notifications made to the
pharmacovigilance system, which indicates that these were
either not made, which is in line with the low level of
reporting practiced in Portugal [
], or simply not recorded
in the patient file.
The pharmacist recommendations made to physicians
sought to address clinically relevant DRPs and included
suggestions that could solve various DRPs by therapy
changes—including the addition of new medicines, the
suspension of others and dosage change—and/or additional
monitoring, which is quite similar to what has been
reported elsewhere [
16, 26, 31
Results shown indicate there were two main drawbacks
in this study. The first was undoubtedly the high proportion
of pharmacist recommendations for which no feedback was
obtained. This could be a result of logistic problems or lack
of cooperation between different healthcare providers. Our
experience in this study strongly reinforces the latter, as
most non-response cases arose from one of the facilities
where this culture is still lacking. The second drawback of
the study was the low proportion of accepted
recommendations (B10 %), even if only respondents are considered.
However, two considerations are warranted. First, if we
consider the unit of analysis as the letters sent to
physicians, rather than the recommendations, and considering
acceptance of at least one recommendation per letter, this
value would rise to 35.7 %, suggesting this proportion of
physicians may be more open to interdisciplinary
collaboration, which may be seen as more encouraging for
pharmacists willing to engage in such tasks. Second, and
perhaps more important, is the fact that our data suggest
there is a reverse association between the number of
pharmacist recommendations directed at physicians and the
odds of success, i.e. the higher the number of
recommendations, the lower the response rate. This finding has
important implications for practice as it suggests that for
pharmacists to be successful they should be more selective
when providing recommendations. Although our data may
not be extrapolated, our experience suggests that a
maximum of five recommendations should be made to increase
the odds of acceptance (p = 0.008).
Ideally, the follow-up of these patients should have been
prolonged to ensure that DRPs solved did not result in new
DRPs, which needed additional interventions. Furthermore,
more robust outcome measures would be needed to
evaluate if DRPs solved resulted in a better health and quality
of life for the patient, which is the ultimate goal of
pharmaceutical care. However, it should be stressed that this
study is innovative by using a simple classification, with
vast use in primary care worldwide, particularly in
Portuguese- and Spanish-speaking countries, but with so far
little research published applying it in nursing homes
residents. To our knowledge, there was only one publication
describing the use of the Dader method in a similar sample,
but which did not fully characterise the prevalence or type
of DRP found [
Pharmacist-led medication review proved useful in
identifying DRPs in polypharmacy elderly patients. Closer
collaboration between healthcare professionals is needed to
improve the acceptability of pharmacist interventions and
consequently the proportion of problems solved. Moreover,
it seems that, to be successful, the pharmacist should be
more selective in the recommendations made.
Compliance with Ethical Standards
Ethical approval The study complied with all clinical research
requirements in accordance with the Declaration of Helsinki. The use
of patients’ medical records was authorised by the clinical director of
the nursing home upon signing of an informed consent form. Patients
data were collected anonymously using their process code. No patient
or facility is here identified. Ethics committee approval was obtained
from ‘‘Comissa˜o de E´ tica para a Investigac¸a˜o nas A´ reas de Sau´de
Humana e Bem-Estar da Universidade de E´ vora’’.
Patient consent There was no informed consent collected from the
patient as the patient was not contacted and all information was
extracted anonymously from medical records. Additionally,
interventions were tested as a measure of quality improvement approved
by the facilities.
Funding No sources of funding were used to assist in the
preparation of this study. All indirect costs involved in the study were
entirely supported by the researchers.
Conflicts of interest Filipa Alves da Costa, Luisa Silvestre,
Catarina Periquito, Clara Carneiro, Pedro Oliveira, Ana Isabel Fernandes
and Patricia Cavaco-Silva have no affiliations that are directly
relevant to the content of this study.
Open Access This article is distributed under the terms of the
Creative Commons Attribution-NonCommercial 4.0 International
License (http://creativecommons.org/licenses/by-nc/4.0/), which
permits any noncommercial use, distribution, and reproduction in any
medium, provided you give appropriate credit to the original
author(s) and the source, provide a link to the Creative Commons
license, and indicate if changes were made.
1. PORDATA. Indicadores de envelhecimento em Portugal. Available at: http://www.pordata.pt/Portugal/Indicadores?de?envelhe cimento- 526 . Accessed 10 Oct 2014 .
2. American Geriatrics Society 2012 Beers Criteria Update Expert Panel. American Geriatrics Society updated Beers Criteria for potentially inappropriate medication use in older adults . J Am Geriatr Soc . 2012 ; 60 ( 4 ): 616 - 31 .
3. Gallagher P , Ryan C , Byrne S , Kennedy J , O'Mahony D. STOPP (Screening Toolo of Oldfer Person's Prescriptions) and START (Screening Tool to Alert doctors to Right Tretament) Consensus Validation . Int J Clin Pharmacol Ther . 2008 ; 46 ( 2 ): 72 - 83 .
4. Hanlon JT , Schmader KE , Samsa GP . A method for assessing drug therapy appropriateness . J Clin Epidemiol . 1992 ; 45 : 1045 - 51 .
5. Pharmaceutical Care Network Europe. PCNE guidelines for retrospective medication review in pharmacy V0, the official PCNE definition 2012 . Available from: http://www.pcne. org. Accessed 15 Sept 2013 .
6. van Mil JWF , Westerlund LOT , Hersberger KE , Schaefer MA . Drug-related problem classification systems . Ann Pharmacother. 2004 ; 38 : 859 - 67 .
7. Silva C , Ramalho C , Luz I , Monteiro J , Fresco P . Drug-related problems in institutionalized, polymedicated elderly patients: opportunities for pharmacist intervention . Int J Clin Pharm . 2015 ; 37 ( 2 ): 327 - 34 .
8. de Oliveira Martins S , Soares MA , van Mil JWF , Cabrita J . Inappropriate drug use by Portuguese elderly outpatients: effect of the Beers criteria update . Pharm World Sci . 2006 ; 28 ( 5 ): 296 - 301 .
9. Advinha AM , de Oliveira-Martins S , Mateus V , Pajote SG , Lopes MJ . Medication regimen complexity in institutionalized elderly people in an aging society . Int J Clin Pharm . 2014 ; 36 ( 4 ): 750 - 6 .
10. Frankenthal D , Lerman Y , Kalendaryev E , Lerman Y. Intervention with the screening tool of older persons potentially inappropriate prescriptions/screening tool to alert doctors to right treatment criteria in elderly residents of a chronic geriatric facility: a randomized clinical trial . J Am Geriatr Soc . 2014 ; 62 ( 9 ): 1658 - 65 .
11. Stafford AC , Alswayan MS , Tenni PC . Inappropriate prescribing in older residents of Australian care homes . J Clin Pharm Ther . 2011 ; 36 ( 1 ): 33 - 44 .
12. Lochner S , Kirch W , Schindler C. Managing hypertension among nursing-home residents and community-dwelling elderly in Germany: a comparative pharmacoepidemiological study . Eur J Clin Pharmacol . 2012 ; 68 ( 5 ): 867 - 75 .
13. Milos V , Rekman E , Bondesson A , Eriksson T , Westerlund T , Midlov P. Improving the quality of pharmacotherapy in elderly primary care patients through medication reviews: a randomised controlled study . Drugs Aging . 2013 ; 30 ( 4 ): 235 - 46 .
14. Chang CB , Lai HY , Yang SY , Wu RS , Liu HC , Hsu HY , Hwang SJ , Chang DC . Patient- and clinic visit-related factors associated with potentially inappropriate medication use among older home healthcare service recipients . PLoS One . 2014 ; 9 ( 4 ): e94350 .
15. Davidsson M , Vibe OE , Ruths S , et al. A multidisciplinary approach to improve drug therapy in nursing homes . J Multidiscip Healthc . 2011 ; 4 : 9 - 13 .
16. Brulhart MI , Wermeille JP . Multidisciplinary medication review: evaluation of a pharmaceutical care model for nursing homes . Int J Clin Pharm . 2011 ; 33 ( 3 ): 549 - 57 .
17. Gheewala PA , Peterson GM , Curtain CM , Nishtala PS , Hannan PJ , Castelino RL . Impact of the pharmacist medication review services on drug-related problems and potentially inappropriate prescribing of renally cleared medications in residents of aged care facilities . Drugs Aging . 2014 ; 31 ( 11 ): 825 - 35 .
18. Oliveira MPF , Novaes MRCG . Drug-related problems in institutionalized elderly in Brasilia, Brazil . Biomed Aging Pathol . 2011 ; 1 ( 3 ): 179 - 84 .
19. Committee Consensus . Segundo Consenso de Granada sobre Problemas Relacionados con Medicamentos. Ars Pharmaceutica . 2002 ; 43 ( 3-4 ): 175 - 84 .
20. Santos H , Igle´sias P, Ferna´ ndez -Llimo´s F , Faus MJ , Rodrigues LM . Segundo Consenso de Granada sobre Problemas Relacionados com Medicamentos. Traduc¸a˜o intercultural de Espanhol para Portugueˆs (europeu) . Acta Med Port . 2004 ; 17 : 59 - 66 .
21. WHO Collaborating Centre for Drug Statistics Methodology. Guidelines for ATC classification and DDD assignment 2014 . Oslo. ISBN 978-82-8082-596-4.
22. INE and INSA ( 2009 ). Inque´rito Nacional de Sau´de 2005 / 2006 . Lisboa-Portugal. Available at http://www.insa.pt/sites/INSA/ Portugues/Publicacoes/Outros/Documents/Epidemiologia/INS_ 05_ 06 .pdf. Accessed 22 Sept 2014 .
23. Chan DC , Chen JH , Kuo HK , We CJ , Lu IS , Chiu LS , Wu SC . Drug-related problems (DRPs) identified from geriatric medication safety review clinics . Arch Gerontol Geriatr . 2012 ; 54 ( 1 ): 168 - 74 .
24. Stafford AC , Tenni PC , Peterson GM , Jackson SL , Hejlesen A , Villesen C , Rasmussen M . Drug-related problems identified in medication reviews by Australian pharmacists . Pharm World Sci . 2009 ; 31 ( 2 ): 216 - 23 .
25. Brown EL , Raue PJ , Mlodzianowski AE , Meyers BS , Greenberg RL , Bruce ML . Transition to home care: quality of mental health, pharmacy, and medical history information . Int J Psychiatry Med . 2006 ; 36 ( 3 ): 339 - 49 .
26. Halvorsen KH , Ruths S , Granas AG , et al. Multidisciplinary intervention to identify and resolve drug-related problems in Norwegian nursing homes . Scand J Prim Health Care . 2010 ; 28 ( 2 ): 82 - 8 .
27. Nishtala P , McLachlan AJ , Bell JS , Chen TF . A retrospective study of drug-related problems in Australian aged care homes: medication reviews involving pharmacists and general practitioners . J Eval Clin Pract . 2011 ; 17 ( 1 ): 97 - 103 .
28. Olesen C , Harbig P , Buus KM , Barat I , Damsgaard EM . Impact of pharmaceutical care on adherence, hospitalisations and mortality in elderly patients . Int J Clin Pharm . 2014 ; 36 ( 1 ): 163 - 71 .
29. Ramanath KV , Nedumballi S. Assessment of medication-related problems in geriatric patients of a rural tertiary care hospital . J Young Pharm . 2012 ; 4 ( 4 ): 273 - 8 .
30. Finkers F , Maring JG , Boersma F , et al. A study of medication reviews to identify drug-related problems of polypharmacy patients in the Dutch nursing home setting . J Clin Pharm Ther . 2007 ; 32 ( 5 ): 469 - 76 .
31. Alldred DP , Zermansky AG , Petty DR , Raynor DK , Freemantle N , Eastaugh J , Bowie P. Clinical medication review by a pharmacist of elderly people living in care homes: pharmacist interventions . Int J Pharm Pract . 2007 ; 15 ( 2 ): 93 - 9 .
32. Consensus Committee . Tercer Consenso de Granada sobre Problemas Relacionados con Medicamentos (PRM) y Resultados Negativos asociados a la Medicacion (RNM) . Ars Pharm . 2007 ; 48 ( 1 ): 5 - 17 .
33. Ruths S , Straand J , Nygaard HA . Multidisciplinary medication review in nursing home residents: what are the most significant drug-related problems? The Bergen District Nursing Home (BEDNURS) Study . Qual Saf Health Care . 2003 ; 12 ( 3 ): 176 - 80 .
34. Jankel CA , Fitterman LK . Epidemiology of drug-drug interactions as a cause of hospital admissions . Drug Saf . 1993 ; 9 ( 1 ): 51 - 9 .
35. Azevedo A , Domingues B , Pimenta da Rocha R , Eiriz R , Xavier V . Quality assessment of long-term prescription records: QMP Study . Acta Med Port . 2014 ; 27 ( 5 ): 561 - 7 .
36. Batel-Marques F , Alves C , Penedones A , Dias P , Martins A , Santiago LM , Fontes-Ribeiro C , Caramona M , Macedo T. Pharmacovigilance in Portugal: activity of the Central Pharmacovigilance Unit . Acta Med Port . 2015 ; 28 ( 2 ): 222 - 32 .