Article Text

Download PDFPDF

Improvement of pharmacological treatments in nursing homes: medication review by consultant pharmacists
  1. Conxita Mestres1,
  2. Marta Hernandez1,
  3. Blanca Llagostera1,
  4. Montse Espier2,
  5. Marta Chandre2
  1. 1Pharmacy Service
  2. 2Nursing Home Heath Care Support Teams, Grup Mutuam, Barcelona, Spain
  1. Correspondence to Dr Conxita Mestres Miralles, Grup Mutuam, CSS Mutuam Güell, Mare de Deu de la Salut, 49, Barcelona 08024, Spain; conxita.mestres{at}


Objectives To establish a medication review programme in geriatric patients in nursing homes.

Methods 70 nursing homes, attended by 8 healthcare teams; two consultant pharmacists provide training and support. Prospective study, conducted between August and December 2013, consisting in patients medication profile review, using an algorithm developed by the Pharmacy Department. Drug-related problems (DRP) were communicated to physicians and documented using the Pharmaceutical Care Network Europe scheme. Medication appropriateness index (MAI) score was calculated before and after the pharmacist's interventions.

Results Medication profiles of 127 patients (mean age 84 years) were reviewed. 424 DRP were detected, the majority related to potential adverse drug events. Pharmacists’ interventions/recommendations were made mainly at prescription level. Lack of indication for the drug prescribed in the clinical record was a recurrent problem, thus indicating a somewhat poor quality in these records. Potentially inappropriate drug prescription was also a common problem (17.6%). Cardiovascular, alimentary tract and metabolism and nervous system drugs were those with more DRP. Degree of acceptance of interventions was 80%. The principal cause of non-acceptance was when the drug has been prescribed by a specialist. Number of drugs per patient dropped from 10.3 to 9.5, and the mean MAI score dropped from 5.4 to 1.3.

Conclusions The incorporation of consultant pharmacists at nursing home level has led us to know the quality of prescription and the DRP. We have established a multidisciplinary team that works for the improvement of drug use.


Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


According to data from the WHO, the world's population of people aged ≥60 years has doubled since 1980 and is forecast to reach two billion by 2050.1 With this global increase of older population, there is a continuous increase in demand for long-term care.

Long-term care facilities include different types of institutions that provide rehabilitative, restorative and ongoing skilled care to patients or residents, between them, nursing homes. People living in these institutions have a number of characteristics such us fragility, disabilities, a lot of comorbidities and a high prevalence of polypharmacy. Therefore, the optimisation of drug therapy has become an essential part of healthcare. Prescription for this population is complex, involving different clinicians (primary care and secondary care); therefore, inappropriate drug treatments have been found to be a common problem in these settings.2 ,3 The involvement of pharmacists (consultant or community pharmacists) in nursing homes, through the review of treatments and interventions/recommendations, has shown that their work could be beneficial. However, even though in the last years several studies in this field have been published, there are no conclusive results about the possible benefits in patients.4 This may be due to different reasons: different strategies have been used, none of them showing sound effectivity, heterogeneity of outcomes and difficulties to measure them (adverse drug events, hospital admissions, mortality, quality of life).5

All this was in accordance with our experience in the day-to-day work in our Pharmacy Department in the residential nursing homes managed by our group, where we have an increase of polymedicated residents with a higher burden of diseases. Moreover, the Department of Health of the Generalitat de Catalunya had established in its Health Plan 2011–20156 ,7 as a priority to focus on chronic patients and the management of their medication. Therefore, it was essential for our group to approach the problem of drug use in the nursing homes where we provide support. We decided to start a programme of pharmacists-led medication review of the drug treatments of our patients, with the aim to know the degree of problems related to medications and to establish a method in collaboration with physicians to optimise pharmacotherapy.



Mutuam is a group that provides (between other health services in geriatrics) healthcare support through 8 teams at 160 residential nursing homes in the city of Barcelona. Each team is formed by one physician and three nurses. Two consultant pharmacists provide support to these teams in all the areas related to drug use (drug selection, information, education). To perform this prospective study of medication review, we narrowed (due to our short staff of pharmacists) down to 70 nursing homes with a total of 3354 patients. The patients selected for the study (127) were those classified as Complex Chronic Patient (PCC—patients with multimorbidity or one disease that implies difficult clinical management) in the Health Plan,6 ,8 with the following criteria:

  1. The patient suffers from several diseases at the same time (multimorbidity) or suffers from only one, but with enough gravity.

  2. Their approach contains some of these elements:

    • high likelihood of crises and very poorly controlled symptoms;

    • evolution dynamics, variable and that requires continuous monitoring;

    • high use of hospital services or emergency visits;

    • progressive disease;

    • high use of resources and drugs and iatrogenic risk;

    • requirement of multidisciplinary approach;

    • environment of special uncertainty and doubt in the decisions of management;

    • need to activate and manage access to different devices and resources, often by preferential ways;

    • friability, functional loss, likely to acute decrease (functional or cognitive) or new onset of geriatric syndromes.

    • adverse psychosocial situations.

At the time of the study, the plan was at the beginning of its implementation; therefore, most residents that met these criteria in our institutions were not yet tagged. Patients with a very advanced illness were not included (palliative, last days of life).

The reviews included in this paper were performed between August and December 2013.

Intervention and data collection

Reviews were undertaken by pharmacists using an algorithm developed by the Pharmacy Service and approved by the Pharmacy Commission, according to criteria of efficacy, safety, efficiency and standard in geriatrics (figure 1).

Figure 1

Algorithm of drug treatment revision. ADE, adverse drug events.

Drug-related problems (DRP) detected by the pharmacists in the pharmacological profiles were communicated using encrypted email to physicians. Answers to pharmacists about the acceptance of recommendations/interventions performed were obtained through email, telephone or through meetings.

To perform the reviews, the pharmacists used the clinical station of the National Health System Database of Catalonia, where they can find data about sex, age, laboratory, patients' medical conditions, comorbidities, Barthel Index Score (indicative of daily activity performance) and Pfeiffer test (indicative of cognitive impairment) results (recorded by nursing staff) and medication profile.

Medication appropriateness, interactions, dosage in relation to hepatic and renal function and other items included in the algorithm were checked using the program CheckTheMeds database (

DRP and interventions were documented using the Detailed Classification of the Pharmaceutical Care Network Europe (PCNE) scheme V.5.00.9 This classification has four primary domains for problems (P1–P4), eight primary domains for causes (C1–C8), five primary domains for interventions (I0–I4) and four domains for the outcome of intervention (O0–O3). Each of these domains is more detailed in subdomains.

Data analysis

DRP were registered and analysed using Microsoft Excel V.2010. Drugs were classified using the Anatomical Therapeutic Chemical Classification (ATC) of the WHO.

The percentage of interventions/recommendations accepted was calculated, and the reasons of non-acceptance were recorded.

The effect of the medication review in the patient was calculated using the medication appropriateness index (MAI)10 and the mean number of drugs in the medication profile per patient; significant differences were calculated using the software G-Stat V.2.0.


A total of 127 pharmacological profiles (patients) were reviewed, whose characteristics are listed in table 1. The majority of patients were women, the mean age of the patients being 83.6 years. Barthel and Pfeiffer values show that the majority of these patients have a moderate-height degree of functional dependence and cognitive impairment.

Table 1

Baseline characteristics of patients included in the study

Drug-related problems

A total of 424 DRP were detected by the pharmacists, with a mean of 3.3 per patient. The DRPs are described in table 2 and categorised according to the PCNE classification V.6.2.

Table 2

Drug-related problems (n=434)* according to the Pharmaceutical Care Network Europe classification V.6.2

The major problems identified were related to the adverse drug event P2.1 (the majority were potential because they were detected and corrected before they were administered to the patient) 20%, P3.2 (unnecessary drug treatment) 42% and P1.4 (untreated indication) 20%.

Cause of DRP

The drug-related causes of these problems are described in table 3 and categorised according to the PCNE classification V.6.2. The majority of the causes (91%) were about drug selection, and the majority in this group was C1.1 (inappropriate drug (included contraindicated)), C1.2 (no indication for drug), C1.5 (indication for drug treatment not noticed) and C1.9 (new indication for drug treatment presented).

Table 3

Causes of drug-related problems on medication therapy and physicians' acceptance


All these problems generated 424 pharmacists’ interventions at prescription level (I1.3 and I1.4), both related with intervention proposed to the prescriber, which is the main objective of our study. The mean degree of acceptance was 80%. Table 3 shows the degree of acceptance by problem codes. Interventions directed to inform/correct drug interactions (C1.3) and preventive drug not prescribed (C1.8) are the problems with less acceptance. C1.6 (too many drugs prescribed for one indication) had only 0% acceptance, but it was present in only one case.

Drugs belonging to ATC codes N, C and A were those with more problems detected, but also those where the interventions had a higher degree of acceptance (table 4).

Table 4

Percentage of interventions acceptance by the Anatomical Therapeutic Chemical Classification (ATC) codes of drugs


To standardise the outcomes, we also followed the PCNE classification V.6.2. The problem was totally solved (O1.0) in 73% of the interventions, partially solved (O2.0) in 3%, no need or possibility to solve problem 16%, problem not solved due to lack of cooperation of prescriber (O3.2) 6% and interventions with unknown outcome 2% (O0.0).

The reasons for non-acceptance of the pharmacist's recommendations/interventions were reluctance of the physician to change a drug started by a specialist or to prioritise the stability of the patient at that moment, overriding the recommendation.

Before the pharmacist's interventions, the mean number of drugs prescribed to the patients was 10.5 (SD 3.49); after the interventions, it dropped to 9.6 (SD 3.54). This difference was statistically significant with a p value <0.0001.

Of the 132 recommendations of stopping a drug, 106 were accepted (80.3%); of the 60 recommendations of starting a drug, 34 were accepted (56.6%).

The time the pharmacists devoted to the review ranged between 15 and 40 min.

MAI was used to measure the inappropriate drug prescribed. The mean summed MAI score before intervention was 5.36, which dropped to 1.51 after the interventions for 127 patients (table 5). This study shows that 41% of the drugs prescribed, based on the MAI criteria, were inappropriate. The major category of inappropriate prescribing in this study was drug indication 53%, second category was drug–disease/conditions interactions 23%, third category was dose and duplication 8% each, and the rest were duration of therapy 3%, effectiveness 3% and drug–drug interactions 2%.

Table 5

Medication appropriateness index (MAI) scores (summed and per medication per patient) before and after the interventions


In the first place, the implementation of this study has led us to know the real situation of prescription quality in our nursing homes. Our results are similar to those obtained by other authors in reference to the degree of acceptation of the pharmacist's proposed changes (around 70%) and the kind of drugs usually used in DRPs, the alimentary tract and metabolism, Cardiovascular and nervous system being the main classes involved in the interventions.2 ,11

This programme has allowed us to introduce the consultant pharmacist and to increase the degree of their involvement with the healthcare team. Even though the participation of pharmacists in nursing homes is increasing or is well established in certain countries,4 ,12 ,13 there are very few such participations in Spain.

As expected, we have noted that the treatment reviews are time consuming and require skilled and prepared pharmacists as well as to have access to complete high-quality information sources. The fact that we have this medication review integrated in the entire pharmaceutical care programme of our organisation has important advantages as to have an already developed algorithm ready allows us to do the reviews in a structured way.

We have observed that one of our main causes of DRP is the lack of recording of health problems of patients in the clinical records (C1.2), originating at all levels of patient care (general practitioners, hospitals). These problems were mostly solved through pharmacists’ recommendations by our physicians, which helped complete the clinical records.

Another point is that many of the detected DRPs are in fact prescription errors (eg, potassium supplements in a low-dose furosemide; digoxin 0.25 μg/24 h 7 days a week; darunavir, ritonavir and etravirine administrated with calcium; vitamin B12 administrated intravenously every day for >1 month).

The MAI score has showed that interventions have produced an improvement in the pharmacological profile of our patients; however, we avoid comparing the score with other scores as it is known that variations in scores depend on the context.2 ,3

One of the limitations of our study is the impossibility to measure more tangible benefits in our patients derived from the pharmacist's interventions, as well as to determine the economic impact (assessment of the cost/savings). Our short staff and the low degree of patients tagged as chronic complex by primary care at the time of the study have also limited the number of medication profiles reviewed. As an initial programme of this kind in our settings, we have prioritised the implementation and establishment of a collaborative work between the pharmacists and other healthcare professionals (especially clinicians) on a regular basis.

Interprofessional work has had other benefits such as the introduction of guidelines to solve doubts of physicians and nurses, respond to information needs about drug-related issues, as well as to unite a common belief among clinicians that relates pharmacists’ work mainly with drug cost issues.


  • We have established a structured method for medication review by pharmacists in nursing homes.

  • High degree of acceptance of the pharmacist's interventions has been achieved.

  • We have a better knowledge of what the main DRPs in our patients are.

  • Pharmacists have been integrated to the team of physicians and nurses.

  • A very important point to consider is how to relate the time consumption of qualified pharmacists to undertake reviews in front of the difficulties to get the results in our patients.

  • We will be focused upon improving most DRPs and those that imply a high risk to the patients, as well as try to measure its impact on the quality of life. At the same time, we would also like to address another problem: patient safety in nursing homes (drug adverse effects and error prevention), where the importance of the pharmacist's implication has not been fully considered yet.

Key messages

  • What is already known on this subject

  • Drug use among residents in nursing homes is high and related to numerous problems and adverse effects.

  • Medication reviews performed by pharmacists improve the quality of prescription in nursing homes; however, the work of pharmacists in this field needs more expansion.

  • What this study adds

  • The inclusion of consultant pharmacists in the healthcare teams of nursing homes improves the detection of prescription errors.

  • Classification of interventions leads to a better understanding of the issues that need improvement and makes it easier to explain them to physicians.


We acknowledge the help of Dr Isabel Gonzalez Casafont.



  • Contributors Conceived idea: CM, MH and BLl, ME, MCh. Data collection: MH, BLl. Analysis of data, Discussion and Results: CM. MH, BLl. Writing of the paper and preparation for publication: CM, MH.

  • Competing interests None.

  • Provenance and peer review Commissioned; externally peer reviewed.