Polypharmacy in older adults is associated with a higher incidence of medication-related problems, such as pharmacological interactions, adverse reactions, and medication errors. This study aimed to identify and evaluate potentially inappropriate prescription of medications (PIP) using the screening tool of older persons’ prescriptions/screening tool to alert doctors to right treatment (STOPP/START) criteria in a representative community pharmacy in Trujillo, Peru. The population sample consisted of 158 subjects, with 66.5% between 65 and 69 years old, 53.8% male, and 77.9% receiving between 3 and 4 medications. Medical prescriptions were analyzed using the explicit STOPP/START version 2 criteria, revealing that 93.7% of patients presented STOPP criteria (to stop medication) and 53.8% presented START criteria (to start medication). The most frequently identified STOPP criterion was the gradual withdrawal of benzodiazepines, while the most frequently identified START criterion was the initiation of antihypertensive treatment in patients with systolic blood pressure > 160 mmHg and diastolic blood pressure > 90 mmHg. Principal component analysis showed that the high prevalence of PIP was significantly associated (p < 0.05) with the number of prescribed medications and as age decreases, the number of medical diagnoses also decreases. It is concluded that there is a high prevalence of PIP in the studied population, posing a greater challenge in older adults, highlighting the importance of addressing prescribing practices with greater involvement of community pharmacists.

medications, older adults, polypharmacy, potentially inappropriate medication

Globally, significant demographic changes have been observed in recent decades, characterized by an increase in life expectancy and shifts in physical and social factors. According to the World Health Organization (2022), it is estimated that by 2050, the population of individuals aged 60 years and older will increase from 12% to 22%, representing a near-doubling of this demographic. This rate of aging is more rapid than in previous centuries. In Peru, similar changes have occurred in the population structure, with an increase in the proportion of adults aged 60 years and older, reaching 13.3% in 2022 (Instituto Nacional de Estadística e Informática 2022).

This increase in the population of older adults is accompanied by a rise in demand for healthcare services, as this demographic is characterized by vulnerability, multimorbidity, and polypharmacy, resulting from age-related physiological changes (Heeren et al. 2021). Pharmacotherapy in this age group is complex due to the decline in organ function, affecting the pharmacokinetics and pharmacodynamics of drugs used to treat various conditions, primarily chronic diseases (Bopari and Korc-Grodzicki 2011; Rankin et al. 2018).

Polypharmacy is a common characteristic among older adults, defined as the concurrent use of five or more medications by an individual, with the use of ten or more medications considered excessive polypharmacy (Morin et al. 2018). Polypharmacy is associated with negative health outcomes, increasing the incidence of adverse effects, adherence issues, metabolic problems, and primarily drug interactions (Garin et al. 2021).

The trend toward polypharmacy in older adults leads to high rates of medication-related problems (MRPs), including medication errors and potentially inappropriate prescriptions (PIP) (Rivera Plaza 2018). Compared to younger patients, older adults are more vulnerable to PIPs due to age-related factors. PIPs are defined as situations where the risk of adverse effects outweighs the clinical benefit; when a medication is used at a frequency or duration exceeding recommended guidelines; when there is an increased risk of drug interaction; when there is duplication of active ingredients; or when beneficial medications are omitted to treat or prevent a health problem (Gavilán et al. 2006; Fajreldines et al. 2016).

Currently, tools and methods have been developed to assess the quality of medication prescription and use in older adults, which are classified into implicit and explicit methods (Barris Blundell 2015; O’mahony et al. 2015). Implicit methods rely on clinical judgment and assessment by healthcare professionals, taking into account a comprehensive view of the patient and whether prescriptions correspond to a need or indication. In Spain, the medication appropriation index (MAI) is the most widely accepted implicit method. Explicit methods, on the other hand, focus on analyzing medications using pre-defined criteria based on scientific databases or expert consensus. The key features of explicit methods are that they are reproducible, easy to use, and consume fewer resources compared to implicit methods, which are time-consuming and dependent on the health professional’s level of knowledge (Spinewine et al. 2007; Mud Castelló et al. 2013; Lopez-Rodriguez et al. 2020).

The most widely used explicit methods are the Beers criteria, the McLeod criteria, the START, and the STOPP. The Beers criteria were among the first to be published and have been updated over time; however, they have limitations, such as failing to consider certain pharmacological interactions, therapeutic duplication, and omission of prescribed medications (Galán Retamal et al. 2014; Pérez et al. 2018; Fick et al. 2019). In 2008, Gallagher et al. published the STOPP-START criteria, which describe common errors in treatment and omission of treatment, highlighting the efficacy and ease of application of this method. However, in 2014 an updated version was released due to the expansion of therapeutic evidence, considering 114 criteria – 80 STOPP criteria and 34 START criteria. (Gallagher et al. 2008; O’mahony et al. 2015; Farhat et al. 2021; Díaz Planelles et al. 2023).

It has been shown that the application of the STOPP/START criteria is effective in optimizing prescribing and reducing the risk of morbidity in geriatric patients (Hill-Taylor et al. 2016; Sallevelt et al. 2020; Delgado-Silveira et al. 2023). Most studies were conducted in the hospital setting, while community pharmacy-based studies applying these criteria to identify PIP arescarce. Therefore, in this study, we aimed to identify and evaluate PIP using the STOPP/START criteria in a model community pharmacy in Trujillo, Peru.

This study recruited a total of 158 elderly adults (85 men and 73 women) who visited the pharmacy establishment with a medical prescription and met the previously established criteria. The largest proportion of individuals were between 65 and 69 years old (66.5%), who are individuals who are still able to care for themselves and obtain their medications. Regarding the number of medications, 77.9% had between 3 and 4 medications in their prescriptions. Notably, 22.1% of elderly adults had 5 or more medications in their prescriptions (Table 1).

In Fig. 1A, it can be observed that the most frequent diagnoses were related to mental disorders (24%) and diseases of the nervous system (18%). It can be noted that the most frequent diseases in older adults are those related to the senescence of various organs and bodily systems. The pharmacological groups with the highest prescription rates were psychotherapeutics (18.3%), analgesics and palliative care (17.6%), and anticonvulsants (17.2%), as shown in Fig. 1B. According to Fig. 1C, the medication with the highest percentage of prescription was clonazepam (10.3%), followed by paracetamol associated with tramadol (3.4%) and fluoxetine (2.6%). The high consumption of benzodiazepines in the study population is noteworthy.

Figure 1. 

Classification of diagnoses in older adult patients (A), count of pharmacological groups (B), and main medications prescribed in the study population (C).

Table 2 shows the relationship between STOPP-START indicators and variables such as sex, age, and number of prescribed medications. It can be observed that there is no significant difference in the incidence of potentially inappropriate prescriptions (PIP) according to sex and age. However, it is noteworthy that there is a high number of PIP across all age ranges. Regarding the relationship with the number of medications, a statistically significant difference was observed (p < 0.05), i.e., polypharmacy is associated with the appearance of PIP.

Fig. 2 shows that 35.4% and 51.9% had between 1 and 3 STOPP and START indicators, respectively, while 33.5% had between 4 and 6 STOPP indicators. Only 6.3% of the study population did not present any STOPP indicators, and 46.2% did not present any START indicators.

Figure 2. 

Distribution of STOPP and START indicators according to the number of PIP.

The study provides details of the PIP in accordance with the STOPP/START criteria classification outlined in Table 3. A total of 733 STOPP prescriptions and 140 START prescriptions were found. 11.5% of the STOPP criteria indicate gradual discontinuation of benzodiazepine use when used for more than four weeks, posing a risk of accidents due to sedation, loss of balance, confusion, and falls. Additionally, 5.5% of the criteria recommend discontinuing NSAID use when indicated for a period exceeding three months for the treatment of osteoarthritis, as prolonged administration is associated with nephrotoxicity and alterations in the gastric mucosa. Regarding the START criteria, 14.3% indicate the use of antihypertensive medication in the presence of persistent hypertension with systolic pressure values > 160 mmHg and diastolic pressure values > 90 mmHg. Furthermore, 25% indicate a change in treatment when analgesics such as paracetamol, NSAIDs, or weak opioids are ineffective, suggesting initiation of treatment with opioid agonists. Similarly, 17.9% indicate the prescription of laxatives after regular opioid use.

Factorial analysis was performed using Principal Component Analysis (PCA) (Table 4 and Fig. 3); the six analyzed variables were regrouped into two components (p<0.05), the first component consisting of the number of prescribed medications, the number of STOPP criteria, and the number of START criteria. Meanwhile, the second component consisted of the variables age, sex, and diagnosis. Component 1 identified that the high prevalence of PIP is significantly associated with the number of prescribed medications, meaning that a higher number of prescribed medications increases the likelihood of PIP occurrence. In component 2, age was assigned a negative value, demonstrating that as age decreases, there is a lower number of diagnoses.

Figure 3. 

Study variables grouped into two components in rotated space, according to factor analysis. DX: diagnostics; NUMSTOP: number of STOPP criteria: NUMSTART: number of START criteria; NUMED: number of medications.

Polypharmacy is a public health issue present in patients with multiple chronic diseases and becomes one of the main causes of multimorbidity, especially in vulnerable older adults from a pharmacological context. This can be attributed to the use of potentially inappropriate medications and adverse drug reactions due to changes in physiological functions and pharmacokinetic and pharmacodynamic characteristics that occur with aging (Baré et al. 2020; Sánchez-Rodríguez et al. 2020).

In the present study, a high consumption of psychotropic, neurological, and pain medication was found, which is due to the fact that neurodegenerative problems are prominent in this age group (Tarawneh and Galvin 2010). The older adult population, due to polymorbidity, is prone to higher use of medications, which is reflected in the prescription trend in the analyzed recipes. In this regard, the probability of a doctor prescribing, for example, a benzodiazepine is high compared to the prescription of other pharmacological groups (Quinn et al. 2019). On the other hand, it is also true that in low- and middle-income countries there is a tendency towards excessive medication use (Albarqouni et al. 2023), such as in the scenario of the present study.

Benzodiazepines, which have shown high consumption in this study, are associated with confusion, dizziness, and risk of falls in older adults, mainly those with prolonged action. In that sense, polypharmacy that includes benzodiazepines along with other central depressants should be avoided because it potentiates the effects, increasing the risk of morbidity and mortality (Gerlach et al. 2017; Gómez et al. 2017; Niznik et al. 2022). Another of the most prescribed drugs was paracetamol alone or associated with tramadol, used to treat acute pain, chronic pain, and fever. Paracetamol is one of the most widely used analgesics in older adults and has a good safety profile. However, in this age group, there may be a decrease in clearance and volume of distribution, suggesting the need for dose adjustment. Additionally, continued consumption of this drug is related to a higher incidence of hepatotoxicity due to its metabolism by the enzyme CYP2E1, which generates N-acetyl-p-benzoquinone-imine (NAPQI), a toxic metabolite (Mian et al. 2018; Freo et al. 2021). In Peru, as well as in other countries, paracetamol is available over-the-counter at certain doses (Digemid 2024); however, guidance and counseling by the pharmacist in these circumstances is crucial to avoid potential cases of overdose and intoxication (Mor et al. 2023; Thrimawithana et al. 2024).

Initially, the determination of PPI in older adults using the STOPP-START criteria was performed in a European context, with STOPP prevalence ranging from 35% to 77% and START from 51% to 73% (O’Mahony 2020). However, recent studies have found PPI rates of 86% (Parodi López et al. 2022) and up to 91% (Marín-Garricho et al. 2022). In a South American context, in Brazil, PPI was found to be 95.4% (Pereira et al. 2019), similar to the present study. Both contexts reflect unfavorable scenarios of healthcare coverage for the geriatric population, particularly regarding the pharmacotherapy received, and call for improvements in healthcare systems for this vulnerable group, prone to polymedication and consequent unnecessary and avoidable adverse reactions.

In the OPERAM trial, a pharmacotherapy team decided to determine the frequency of STOPP/START criteria in a European hospital setting, finding that 99% of patients presented a STOPP or START criterion. Medications prescribed without evidence-based clinical indication, duplicate prescriptions, and the criterion with the highest prevalence were related to the use of benzodiazepines for more than four weeks (Sallevelt et al. 2022), as mentioned, the most frequent criterion found in our study. In Palestine, Abukhalil et al. conducted a multicenter retrospective observational study evaluating inappropriate prescriptions in elderly hospitalized. They identified that 66.8% of participants presented with PIP due to multiple comorbidities and polypharmacy; they also detailed a high frequency of START criteria at the cardiovascular system level, and the most frequent STOPP criterion was duplication of therapy (Abukhalil et al. 2022).

In Mexico and Brazil, potentially inappropriate prescriptions were also evaluated in hospitalized older adults. In the study conducted in Mexico, a PIP prevalence of 67% was found according to the STOPP/START criteria, and in Brazil, a high number of STOPP criteria corresponding to the central nervous system (24.8%) was identified. The START criteria that stand out are errors due to omission of medications (59.9%); however, when identified at hospital admission, the percentages were significantly reduced (García Orihuela et al. 2020; Saturno-Hernández et al. 2021). In Spain, Mud Castelló et al. (2014) detected potentially inadequate prescriptions in older adults in two community pharmacies. They found that 67% were polymedicated patients according to the Beers criteria, and the use of short-, intermediate-, and long-acting benzodiazepines was the most frequent, similar to our study. Regarding the STOPP/START criteria, the most frequent STOPP criterion was the administration of IBP for more than 8 weeks.

In Lima, Peru, prescriptions filled at pharmacies and drugstores were also evaluated according to the Beers criteria, finding that 69.2% of older adults had potentially inappropriate prescriptions due to the consumption of anxiolytics, with a higher incidence in women (61.6%) (Gonzales Foroca and Mamani Huaranga 2020). Additionally, the incidence of potentially inappropriate prescriptions for adults attending a pharmacy chain was analyzed using the Beers criterion, finding a 69.2% PIP, generally in the age range of 65 to 70 years and with a higher percentage in adult women (Morales and Flores 2022). These two studies use the Beers criteria to identify PIP, and it is necessary to mention that the Beers criteria compared to STOPP/START may detect fewer PIP (Kerliu et al., 2021). Most of the studies found identify PIP in the hospital setting; however, this study generates the need to apply explicit criteria to detect PIP using the STOPP START criteria in a community pharmacy context to ensure correct pharmacotherapy in the older adult population. Nevertheless, one of the main limitations in the country is the lack of coordination in the healthcare system and policies that incentivize pharmacists working in community pharmacies to perform additional activities beyond dispensing to optimize pharmacotherapy outcomes (Rodríguez-Tanta et al. 2023).

In the context of community pharmacy, a study was conducted in Bulgaria that identified PIP in a single pharmacy using the EU(7)-PIM List criteria, finding that 67% of the polymedicated population had at least one PIP (Milushewa et al., 2023), comparable to what we obtained in polymedicated patients (Table 2), although it is necessary to mention that the explicit criteria used are different and also that the STOPP-START criteria tend to detect more PIP (Monteiro et al. 2020).

The application of the STOPP START criteria in public and private health services would allow for a reduction in the incidence of adverse effects and optimization of treatments received by older adults in general. While the STOPP/START criteria constitute a good tool for identifying problems with medication in older adults, it is also true that, in a global context, these criteria may have limitations, as indicated by Verdoorn et al. (Verdoorn et al. 2015), who found that only 19% of medication-related problems (MRP) in users of Dutch community pharmacies were associated with the STOPP/START criteria. Although the Dutch context may differ from that of other countries, it reflects the need to continue associating explicit criteria with pharmaceutical care practices in the community.

Our study also reflects the need for clinical pharmacists in the community context, with the necessary competencies to identify and resolve, within their competence, potential problems associated with medications in older adult users who attend pharmacies, or, if necessary, refer or recommend patients to medical attention.

A limitation of the present study is that it was based on patients’ prescriptions, and complete access to the medical history of the participants was not available. Additionally, being a cross-sectional study, it was not possible to establish follow-up and have more precise exposure to medication; perhaps many of them are not entirely adherent to the medications they purchased at the pharmacy. Although we are detecting a high prevalence of PIP, the clinical relevance of these PIP is not being considered, which can be addressed in future multidisciplinary studies.

Polypharmacy is a common problem in the elderly due to the physiological changes they experience and is directly associated with the incidence of PIP. According to the criteria used in the study, a high prevalence of STOPP and START criteria was found in the analyzed prescriptions. Additionally, a significant association was identified through component analysis: with a higher number of prescribed medications, there is a higher probability of PIP occurrence. It is necessary for community pharmacies to implement systems for detecting PIPs in the geriatric population, which, along with other strategies, allow for the optimization of pharmacotherapy in older adult users. However, this constitutes a challenge for the healthcare systems of countries where there is a lack of better articulation of primary care processes.

Conceptualization (MBV, JCF), Methodology (MBV, GGH, JCF), Software (GGH, ACB, KRT, HVM), Validation (MBV, JCF, HVM), Formal analysis (MBV, GGH), Investigation (MBV, ACB, HVM, KRT), Resources (MBV, GGH, HVM, KRT, JCF), Data Curation (GGH, ACB, JCF), Writing – Original draft, Writing – Review and Editing (GGH, JCF, ACB), Visualization (ACB, KRT), Supervision (JCF), Project administration (MBV).

All authors have read and agreed to the published version of the manuscript.

No potential conflict of interest was reported by the authors.