If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
PharmacoTherapy, -Epidemiology and -Economics – Groningen Research, Institute of Pharmacy, University of Groningen, Antonius Deusinglaan 1, 9713 AV Groningen, the Netherlands
Department of Clinical Pharmacy, Deventer Hospital, Nico Bolkesteinlaan 75, 7416 SE Deventer, the NetherlandsPharmacoTherapy, -Epidemiology and -Economics – Groningen Research, Institute of Pharmacy, University of Groningen, Antonius Deusinglaan 1, 9713 AV Groningen, the Netherlands
78% of older patients with cancer have potentially inappropriate/omitted medication
•
A comprehensive pharmacist-led medication review can optimize patients' treatment
•
Follow-up on recommendations from these medication reviews is high (73%)
•
Pharmacists' expert opinion contributes significantly to STOPP/START criteria
Abstract
Objectives
To determine the prevalence of Potentially Inappropriate Medication (PIMs) and Potentially Omitted Medication (POMs) in older patients with cancer.
Materials and Methods
In this prospective observational study (hospital) pharmacists conducted comprehensive medication reviews in older patients with cancer (aged ≥65 years) receiving parenteral chemotherapy and/or immunotherapy at the Deventer Hospital. PIMs and POMs were identified using the Screening Tool of Older Persons' potentially inappropriate Prescriptions (STOPP), the Screening Tool to Alert doctors to the Right Treatment (START), and pharmacists' expert opinion. Recommendations regarding PIMs and POMs were communicated to the patient's oncologist/haematologist and follow-up was measured. Associations between covariates and the prevalence of PIMs and POMs were statistically analysed.
Results
For the 150 patients included, 180 PIMs and 86 POMs were identified with a prevalence of 78%. Using pharmacists' expert opinion in addition to only STOPP/START criteria contributed to 49% of the PIMs and 23% of the POMs. A follow-up action was required in 73% of the 266 PIMs and POMs. Number of medicines and Charlson Comorbidity Index score were both associated with having at least one PIM and/or POM (p = .031 and p = .016, respectively).
Conclusion
The prevalence of PIMs and POMs and subsequent follow-up in older patients with cancer is high. A pharmacist-led comprehensive medication review is a good instrument to identify these PIMs and POMs and to optimize patients' treatment. A complete approach, including pharmacists' expert opinion, is recommended to identify all PIMs and POMs in clinical practice.
Ageing, multiple morbidities, and the use of multiple medicines make older patients a high-risk group for drug-related problems (DRPs). The diagnosis of cancer further increases this risk. Cancer treatment leads to the use of more medicines, multiple involved health care providers, and a higher disease burden. Frequent hospital visits and the associated transfer of information about medication use are additional risk factors for DRPs, which can lead to compromised cancer management plans. Since this population will continue to grow, addressing the appropriateness of medication use in this population will become even more important [
Evaluation of a pharmacist-led medication assessment used to identify prevalence of and associations with polypharmacy and potentially inappropriate medication use among ambulatory senior adults with cancer.
]. Different criteria are used to identify Potentially Inappropriate Medications (PIMs) and Potentially Omitted Medications (POMs). Potentially Inappropriate Medications are defined as medicines that are used by a patient, but are either unnecessary or do not have additional value, or can be optimized in their use. Potentially Omitted Medications refer to medicines that are not used by a patient, but adding them is clinically indicated and can be beneficial for the patient. In Europe, the Screening Tool of Older Persons' potentially inappropriate Prescriptions (STOPP) and the Screening Tool to Alert doctors to the Right Treatment (START) are most recommended to identify PIMs and POMs [
Application of the STOPP/START criteria: a systematic review of the prevalence of potentially inappropriate prescribing in older adults, and evidence of clinical, humanistic and economic impact.
]. However, using only these criteria does not lead to identification of all relevant PIMs and POMs and therefore a more comprehensive medication assessment is needed [
The Dutch multidisciplinary guideline ‘polypharmacy in the elderly’ recommends comprehensive medication reviews in patients aged ≥65 years with polypharmacy and having at least one predefined risk factor [
]. Oncology is not mentioned as a specific risk factor in this guideline and no Dutch study was found investigating PIMs and POMs and the impact of pharmacist-led comprehensive medication reviews in this population. In general, studies on the prevalence of PIMs and POMs in older patients with cancer have various limitations and methods and results differ highly [
Evaluation of a pharmacist-led medication assessment used to identify prevalence of and associations with polypharmacy and potentially inappropriate medication use among ambulatory senior adults with cancer.
Potentially inappropriate medication use and associated healthcare utilization and costs among older adults with colorectal, breast, and prostate cancers.
Therefore, this study aims to determine the prevalence of PIMs and POMs in older patients with cancer by conducting pharmacist-led comprehensive medication reviews. Secondary objectives are to examine subtypes of PIMs and POMs, to determine follow-up of PIMs and POMs, and to assess risk factors for PIMs and POMs.
2. Materials and Methods
In this prospective observational study, pharmacist-led comprehensive medication reviews were conducted in a multidisciplinary team with older patients with cancer between May 2018 and January 2019 at the Deventer Hospital (a middle-sized teaching hospital in The Netherlands). Patients aged ≥65 years, treated for cancer by a medical oncologist/haematologist, and receiving parenteral chemotherapy and/or immunotherapy at the day care unit were enrolled in this study. Patients at the start of therapy as well as patients who already started therapy were included.
Patients were asked to bring all their medication or a medication overview to the day care unit. While receiving chemotherapy or immunotherapy, a pharmacist or pharmacist in training interviewed the patient. The actual medication use, including non-prescription medicines, was verified with the patient (medication reconciliation) and problems with usage of medication were addressed using a questionnaire. Based on this information and the patient's medical records, PIMs and POMs were identified by the pharmacist using the revised STOPP/START criteria (2015) [
] and pharmacists' expert opinion. Expert opinion consisted of interpretation of medication surveillance signals, practical recommendations, and guideline adherence. Reviewing medication surveillance signals generated from the pharmacy information system is standard practice in Dutch hospital pharmacies. The pharmacists' expert opinion was part of the typical work and knowledge of a hospital pharmacist responsible for medication reconciliation and medication review. No specific framework, process, or list was used for the pharmacists' expert opinion. All identified PIMs and POMs and their corresponding recommendations were double-checked and if necessary complemented by a hospital pharmacist before communicating them to the patient's oncologist/haematologist. If there were discrepancies between the pharmacist and hospital pharmacist, the PIMs and POMs and their corresponding recommendations were based on consensus between the two. For each PIM/POM the oncologist/haematologist decided if a follow-up action was required. Two follow-up actions were possible: the recommendation was implemented by the oncologist/haematologist or the PIM/POM with corresponding recommendation was sent to the patient's general practitioner.
The prevalence of PIMs and POMs (percentage of patients with at least one PIM and/or POM) was determined for PIMs and POMs combined as well as separately. PIMs and POMs were further classified by the Anatomical Therapeutic Chemical (ATC) classification, by the classification used in the STOPP/START criteria, and by the classification used in the Systematic Tool to Reduce Inappropriate Prescribing (STRIP) method [
To determine the association of covariates with the prevalence of PIMs and POMs, the following information was collected for each patient: age, gender, number of different medicines, polypharmacy, use of a medication roll (medication pre-packaged per intake moment), cancer type, curative intent, and the Charlson Comorbidity Index (CCI) score. The number of different medicines included all medication used at home at the time of the interview, as well as the chemotherapy and/or immunotherapy, and accompanying supportive care agents. Polypharmacy was defined as concurrent use of five or more medicines for chronic use with a different ATC classification on ATC3-level, excluding medicines for dermal use (definition by the Dutch guideline ‘polypharmacy in the elderly’ [
]). Use of a medication roll was included as a measure for “self-management”. The patient's oncologist/haematologist indicated whether the cancer treatment was intended to be curative or not. Finally, the CCI score, determined by the classic scoring index by Charlson et al. [
], was included as a measure for vulnerability and was based on the patient's medical records.
Differences in these covariates for patients with and without PIMs/POMs were assessed using descriptive statistics (independent-samples t-test, Mann-Whitney U Test, Pearson's χ2 test, or Fisher's exact test). For factors significantly associated with the prevalence of PIMs and POMs (p-value <.05), univariate logistic regression followed by multivariate logistic regression was used to assess odds ratios (ORs) and 95% confidence intervals (CIs).
This study was assessed and approved by the Medical Ethics Committee of Isala Hospital as a non-interventional study. All patients signed a written consent form prior to participating in this study.
3. Results
For this study, 159 patients were approached to participate of which four patients refused participation and five patients had their appointment rescheduled until after the research period. The patients' characteristics of the 150 patients included in this study are depicted in Table 1. In total, these patients used 1656 medicines, with a mean of eleven medicines per patient (range 3–21). One hundred and forty-four patients (96%) used five or more medicines and 99 patients (66%) used ten or more medicines. When excluding the chemotherapy and/or immunotherapy regimen and accompanying supportive care agents at the day care unit, the mean number of medicines per patient was seven with 77% and 23% of the patients using five or more and ten or more medicines, respectively.
Table 1Patient characteristics.
n = 150
Age, years (median (IQR) [range])
72 (8) [65–90]
Gender (n (%))
Male
88 (59)
Female
62 (41)
Number of medicines (mean (SD) [range])
11.0 (3.8) [3–21]
Number of medicines without chemotherapy, immunotherapy and supportive care agents (mean (SD) [range])
A total of 180 PIMs and 86 POMs were identified. These 266 PIMs and POMs give a mean of 1.8 per patient (range 0–8). PIMs and POMs were prevalent in 117 (78%) of the patients. The prevalence of PIMs and POMs separately was 65% and 46%, respectively (Fig. 1).
The number of patients with no, 1, 2, 3, or ≥ 4 PIMs (separately), POMs (separately), and PIMs and/or POMs (combined). Percentages are calculated as part of the total (n = 150) per category. Abbreviations: PIM, Potentially Inappropriate Medication; POM, Potentially Omitted Medication. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Based on the ATC classification, the most common groups of medication for the 180 PIMs were proton pump inhibitors (PPIs) (19%), antihypertensive drugs (11%), benzodiazepine agonists (9%), analgesics (8%), alpha-adrenoreceptor antagonists (6%), and antidepressants (6%). Four PIMs (2%) concerned antineoplastic agents. The most common groups of medication for the 86 POMs were statins (40%), antihypertensive drugs (19%), and vitamin D (15%). Table 2 specifies the criteria used for identification of the PIMs and POMs.
Table 2Criteria used for identification of PIMs and POMs.
The total number of criteria used for identification of PIMs (186) exceeds the total number of PIMs (180) because 6 PIMs were identified using two criteria.
STOPP criteria
Total
95 (51)
A1. No evidence-based indication
42 (23)
A2. Usage longer than recommended
25 (13)
A3. Double medication
8 (4)
D5. Benzodiazepine ≥4 weeks
12 (6)
Other
8 (4)
Expert opinion
Total
91 (49)
Medicine not effective
24 (13)
Over treatment
15 (8)
(Potential) side effect
9 (5)
Contraindication or interaction
2 (1)
Incorrect dosage
15 (8)
Problem with usage
26 (14)
POMs
86 (100)
START criteria
Total
66 (77)
B4. Antihypertensives, high BP
7 (8)
B5. Statins, high cardiovascular risk
30 (35)
H2. Bisph/vitD/calc, chronic prednisone use
10 (12)
H3. VitD/calc, osteoporosis
6 (7)
H5. VitD/calc, home bound / fall incidents
4 (5)
J2. ACE-inhibitor, DM with kidney damage
5 (6)
Other
4 (5)
Expert opinion
Total
20 (23)
Under treatment
20 (23)
Abbreviations: ACE, angiotensin converting enzyme; bisph, bisphosphonate; BP, blood pressure; calc, calcium; DM, diabetes mellitus; PIM, potentially inappropriate medication; POM, potentially omitted medication; START, screening tool to alert doctors to the right treatment; STOPP, screening tool of older persons' potentially inappropriate prescriptions; vitD, vitamin D.
a The total number of criteria used for identification of PIMs (186) exceeds the total number of PIMs (180) because 6 PIMs were identified using two criteria.
For 195 (73%) of the 266 identified PIMs and POMs a follow-up action was required according to the oncologist/haematologist. PIMs required more frequently a follow-up action than POMs, 76% vs 67% respectively. For 39% of the PIMs and POMs requiring a follow-up action, this action was realized by the oncologist/haematologist. The distribution of follow-up actions is summarized in Fig. 2.
Follow-up actions of 180 PIMs (separately), 86 POMs (separately), and 266 PIMs and POMs (combined). Percentages are calculated as part of the total per category.Abbreviations: PIM, Potentially iIappropriate Medication; POM, Potentially Omitted Medication. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
PIMs and POMs with a follow-up action realized by the oncologist/haematologist predominantly concerned PPIs (PIMs), anti-infectives (PIMs), antineoplastic agents (PIMs), musculoskeletal medication (PIMs/POMs), and vitamin D (POMs). Potentially Inappropriate Medications and Potentially Omitted Medicationss that were sent most frequently to the general practitioner were alpha-adrenoreceptor antagonists (PIMs), respiratory medication (PIMs), antihypertensive drugs (PIMs/POMs), and statins (POMs). PIMs identified with STOPP criterion A3 ‘double medication’ or expert opinion ‘contraindication or interaction’ were always considered as requiring a follow-up action. Follow-up was also high for expert opinion ‘incorrect dosage’ and ‘problem with usage’ with a follow-up action required for 87% and 86% of the PIMs, respectively. Follow-up was the lowest for START criterion B5 ‘statins for patients with high cardiovascular risk’ with no follow-up action required for 43% of the POMs.
The number of medicines and the CCI score were associated with having at least one PIM and/or POM (Table 3). The other covariates were not statistically significant associated with the prevalence of PIMs and POMs. For an increase of one medicine, the odds of having at least one PIM and/or POM increased with 1.125. For an increase of one point in the CCI score, the odds of having at least one PIM and/or POM increased with 1.501. In multivariate logistic regression analysis both associations were no longer statistically significant. The Pearson correlation coefficient between the variables number of medicines and CCI score was 0.4.
Table 3Associations between covariates and prevalence of PIMs and POMs.
A high prevalence of PIMs and POMs (78%) was found in older patients with cancer by conducting pharmacist-led comprehensive medication reviews using both STOPP/START criteria and pharmacists' expert opinion.
The prevalence of PIMs in the current study is higher than in most previous studies. This might be due to a more thorough and complete approach for the comprehensive medication reviews. Most studies based their PIMs on the medical records rather than having an interview involving the patient. Furthermore, the current study used pharmacists' expert opinion in addition to the standardized STOPP/START criteria.
Studies that did not include a patient interview and only used standardized criteria (Beers or STOPP) found a prevalence of PIMs ranging from 16% to 57% [
Potentially inappropriate medication use and associated healthcare utilization and costs among older adults with colorectal, breast, and prostate cancers.
]. Fourteen percent of the PIMs in the current study regarded problems with usage, most of which were identified based on the interview with the patient. These PIMs were missed in these previous studies. In addition, the medication reconciliation with the patient attributes to a more complete overview of the actual medication use and therefore to potentially more PIMs. Reis et al. [
Evaluation of a pharmacist-led medication assessment used to identify prevalence of and associations with polypharmacy and potentially inappropriate medication use among ambulatory senior adults with cancer.
] found a prevalence of 48%, 51%, and 52%, respectively, when interviewing the patient or conducting a full comprehensive medication review. However, not all PIMs and POMs can be identified with a set of standardized criteria and therefore the knowledge and expertise of a pharmacist is necessary to attribute to these criteria. This is well shown in this study where half of the PIMs and a quarter of the POMs were identified by pharmacists' expert opinion.
To fully optimize patients' treatment, inappropriate medication should be addressed as well as omitted medication. Only two studies were found in which POMs were identified in older patients with cancer, with a prevalence of 34% and 98% [
] is largely attributed to omitted vaccinations, which is not applicable to the Dutch situation. In the Netherlands, older patients are annually offered an influenza vaccination and a pneumococcal vaccine is not included in the Dutch START criteria [
The high prevalence of PIMs on PPIs and benzodiazepine agonists and POMs on statins, antihypertensive drugs, and vitamin D are in line with several other studies in patients with cancer as well as patients without cancer [
Prevalence and predictors of inappropriate prescribing according to the screening tool of older People’s prescriptions and screening tool to alert to right treatment version2 criteria in older patients discharged from geriatric and internal medicine wards: a prospective observational multicenter study.
]. Only four PIMs concerned antineoplastic drugs. Because most PIMs involved regular medication, the problems identified in the oncology population may not be much different from other populations of older polypharmacy patients and therefore STOPP/START criteria seem well applicable. Associations with the prevalence of PIMs and POMs were found for the number of medicines and the CCI score, in line with previous studies [
Evaluation of a pharmacist-led medication assessment used to identify prevalence of and associations with polypharmacy and potentially inappropriate medication use among ambulatory senior adults with cancer.
]. However, in this study these variables were not very strongly associated and borderline significant. The significant associations were no longer present in the multivariate logistic regression analysis possibly due to a lack of power and the mild correlation (Pearson correlation coefficient 0.4) between CCI score and number of medicines. Because of the time investment needed, implementation in daily practice can be challenging. This study indicates that a specific focus on patients with more medicines and/or a higher CCI score could be considered. However, ORs were small, associations were not statistically significant in multivariate logistic regression analysis, and this study was not designed to determine which (sub)group of patients would benefit most from pharmacist-led comprehensive medication reviews. Future research could provide more insight on this subject.
Measuring follow-up further distinguishes this study from previous studies on PIMs and POMs in older patients with cancer. It was outside the scope of this study to assess actual changes in medication, additional laboratory measurements, or actions by the general practitioner, which could lead to an overestimation of the follow-up on PIMs and POMs. However, the follow-up percentage found in this study (73%) is in line with other studies that found action to be taken in 69%–82% of recommendations made by pharmacists [
]. For all STOPP/START criteria, a follow-up action was required for the majority of PIMs and POMs. Even for criteria that might seem less relevant in older patients with cancer (for example starting statins or vitamin D + calcium), more than half of the POMs required a follow-up action and were therefore considered clinically relevant by the oncologist/haematologist. This shows that the criteria, which were used, are relevant to this patient population.
Strengths of this study are the combination of a pharmacist-led comprehensive medication review and medication reconciliation with the patient, the incorporation of pharmacists' expert opinion, the identification of PIMs as well as POMs, and measuring the follow-up of recommendations. Limitations are that this is a single-institution study and only patients who received parenteral chemotherapy and/or immunotherapy were included. In addition, only the prevalence of PIMs and POMs was measured with the immediate follow-up, so long-term outcomes for patient and healthcare cannot be assessed.
In conclusion, PIMs and POMs are highly prevalent among older patients with cancer and a pharmacist-led comprehensive medication reviews is a good instrument to optimize patients' treatment. A complete approach, including pharmacists' expert opinion, is recommended to identify all PIMs and POMs.
Author Contributions
Conception and Design: FMAM van Loveren, IRF van Berlo – van de Laar, ALT Imholz, E van 't Riet, K Taxis, FGA Jansman.
Data Collection: FMAM van Loveren.
Analysis and Interpretation of Data: FMAM van Loveren, IRF van Berlo – van de Laar, E van 't Riet.
Manuscript Writing: FMAM van Loveren, IRF van Berlo – van de Laar.
Approval of Final Article: FMAM van Loveren, IRF van Berlo – van de Laar, ALT Imholz, E van 't Riet, K Taxis, FGA Jansman.
Disclosure
None.
Declaration of Competing Interest
None.
Acknowledgements
The authors thank pharmacists in training Carlijn van der Velde and Marjolein Niewenhuijse for their contribution to the comprehensive medication reviews and data collection.
References
Hamaker M.E.
Jonker J.M.
De Rooij S.E.
Vos A.G.
Smorenburg C.H.
Van Munster B.C.
Frailty screening methods for predicting outcome of a comprehensive geriatric assessment in elderly patients with cancer: a systematic review.
Evaluation of a pharmacist-led medication assessment used to identify prevalence of and associations with polypharmacy and potentially inappropriate medication use among ambulatory senior adults with cancer.
Application of the STOPP/START criteria: a systematic review of the prevalence of potentially inappropriate prescribing in older adults, and evidence of clinical, humanistic and economic impact.
Potentially inappropriate medication use and associated healthcare utilization and costs among older adults with colorectal, breast, and prostate cancers.
Prevalence and predictors of inappropriate prescribing according to the screening tool of older People’s prescriptions and screening tool to alert to right treatment version2 criteria in older patients discharged from geriatric and internal medicine wards: a prospective observational multicenter study.
30496-5&id=gr1.jpg){kind=link}
30496-5&id=gr2.jpg){kind=link}