Int J Clin Pharm DOI 10.1007/s11096-014-0040-9
RESEARCH ARTICLE
The impact of hospitalization on potentially inappropriate prescribing in an acute medical geriatric division Dvora Frankenthal • Yaffa Lerman Yehuda Lerman
•
Received: 1 May 2014 / Accepted: 12 November 2014 Koninklijke Nederlandse Maatschappij ter bevordering der Pharmacie 2014
Abstract Background Screening Tool of Older Person’s Prescriptions (STOPP) and the Screening Tool to Alert doctors to Right Treatment (START) have been increasingly used to evaluate potentially inappropriate prescriptions (PIPs) and potentially prescription omissions (PPOs). The impact of hospitalization on PIPs/PPOs has not been investigated in depth. Objective To compare the prevalence of PIPs/PPOs in elderly patients on hospital admission and discharge and to identify associated risk factors. Setting An acute medical geriatric division of the Tel Aviv Medical Center (Israel). Method This retrospective cross-sectional study included patients admitted from 12/2011 to 12/2012 aged C65 years. Data from patients’ records included demographic details, diagnoses and medications at admission and discharge. STOPP/START criteria were applied to each patient’s record. Main outcome measure Prevalence of PIPs/PPOs on hospital admission and discharge. Results Three hundred patients were included (mean ± SD age 81.9 ± 7.2 years). Admission PIPs prevalence was 39.3 % (118 patients, 172 PIPs) and it increased to 46.0 % (138 patients, 209 PIPs) at discharge (P = 0.009). Admission PPOs prevalence was 41.0 % (123 patients, 153 PPOs) and it decreased to 28.3 % (85 patients, 99 PPOs) at discharge (P \ 0.001). Having at least one PIP/PPO at discharge but not at admission was associated with length of hospital stay (OR 1.02, 95 % CI 1.001–1.03). History of falls increased the risk of being a ‘‘new PIP patient’’ (OR 2.25, 95 % CI D. Frankenthal (&) Y. Lerman Y. Lerman School of Public Health, Sackler Faculty of Medicine, Tel Aviv University, POB 39040, Tel Aviv, Israel e-mail: [email protected] Y. Lerman Geriatric Division, Tel Aviv Sourasky Medical Center, 6 Weizmann St., Tel Aviv, Israel
1.03–4.9), whereas diabetes increased the risk of being a ‘‘new PPO patient’’ (OR 3.86, 95 % CI 1.2–12.5). Conclusion Hospitalization in a geriatric division resulted in an increase in PIPs and a decrease in PPOs. Strategies to reduce PIPs need to be implemented, especially for patients with longer hospital stay and a history of falls. Keywords Elderly Hospitalization Israel Medications START criteria STOPP criteria
Impacts on Practice •
•
•
In Israel, actions to promote appropriate prescribing in hospitals are needed given the increase in PIPs during hospitalization among elderly patients in geriatric wards. Enhanced monitoring at discharge involving the screening of medications for PIPs/PPOs is an essential part of the care for geriatric patients. To decrease PIPs and PPOs at discharge, special attention should be paid to the medications of patients with longer hospital stays, with a history of falls, and with diabetes.
Introduction Potentially inappropriate prescriptions (PIPs) in older people are becoming a global healthcare concern in an ageing population. A PIP is the use of medicines that pose more risk than benefit, particularly when there are safer alternatives [1]. Another aspect of PIPs in older people is the omission of indicated medications with proven efficacy in patients with an ordinary life expectancy [2]. PIPs can be
123
Int J Clin Pharm
detected by using explicit (criterion-based) prescribing indicators whose ultimate aim is to optimize prescription appropriateness and reduce negative outcomes, including preventable adverse drug effects [3]. Various criteria have been devised to identify PIPs in older people, one of the most recent being the Screening Tool of Older Person’s potentially inappropriate Prescriptions (STOPP) and the Screening Tool to Alert doctors to Right Treatment (START). The STOPP criteria focus upon avoiding the use of drugs that are potentially inappropriate in older people, and the START criteria aim to identify under-treatment or omissions in prescribing of medications among older adults [4]. The prevalence of patients with at least one PIP as identified by the STOPP criteria was reported as being between 26.7 and 77 % [5, 6] of hospitalized older patients, and the reported prevalence of potentially prescription omissions (PPOs) according to START criteria was between 41.9 and 65 % [7, 8]. The influence of hospitalization on the prescribing of drugs is difficult to assess. Some studies reported a decrease in PIPs after a hospital stay, whereas others showed an increase [5, 9–11]. The applicability of the STOPP/START criteria in Israeli hospitals is unknown, and no studies have thus far applied these screening tools in hospitalized older Israeli patients.
Study population Three-hundred patients were randomly selected for this study from all patients (N = 900) aged 65 years or older who had been admitted to the geriatric division over a 12-month period (12/2011–12/2012). Subjects who died during hospitalization were excluded. Data collection Data were collected by means of the electronic patient record. The patients’ data included demographic details, medical histories, reason for admission, history of falls, diagnoses and medications on admission and discharge, biochemistry results, length of hospital stay and cognitive disorder (known dementia and/or an impaired Mini Mental State Examination [MMSE] score of \24/30) [12]. A disabled patient was defined as functionally dependent for at least one item in the Katz scale [13]. Comorbidity was quantified by adding the scores assigned to specific diagnoses in the Charlson Comorbidity Index (CCI) [14]. Polypharmacy (defined by the Israeli Ministry of Health as [8 medications) pertained only to medications used chronically. The frequencies of PIPs and PPOs in the studied sample were evaluated by the study pharmacist using the STOPP/START criteria and they were calculated both at admission and at discharge.
Aim of the study
Statistical analysis
The aim of this study was to compare PIPs/PPOs in elderly patients at the time of admission and discharge from an acute medical geriatric division, and to identify the factors associated with having a PIP/PPO at hospital discharge.
Statistical analysis was performed using the Statistical Package of Social Sciences version 14.0 (Chicago, IL, USA). The PIP/PPO proportion was defined as the number of patients with at least one PIP/PPO according to the STOPP/START criteria. McNemar’s test was used for paired data to examine any change in prescribing in order to compare prescribing practices pre-and post-hospitalization using STOPP/START criteria, The Wilcoxon matched pairs signed ranks test was used to compare the number of medications prescribed to the patients at admission and at discharge. A ‘‘new PIP/PPO patient’’ was defined as having at least one PIP/PPO at discharge but none at admission. Multivariate analysis using a backward stepwise logistic regression was applied to examine the association between independent risk factors and overall PIPs/PPOs at discharge and being a ‘‘new PIP/PPO patient’’ at discharge. The following variables were entered in the logistic regression model: age, gender, number of medications at admission, CCI, functioning, cognitive state, history of falls, diabetes, osteoporosis and length of hospital stay. Significance was set at P \ 0.05.
Ethical approval The study protocol was reviewed and approved by the hospital’s institutional ethical review board.
Methods This retrospective cross-sectional study was conducted in a medical geriatric division of the Tel Aviv Medical Center, which is the second largest (1,300 beds) multidisciplinary academic medical center in Israel. The geriatric division comprises two wards with a total of 70 beds, and it treats and rehabilitates older patients with neurological and orthopedic conditions.
123
Int J Clin Pharm
(n = 150, 50.0 %), antidiabetics (n = 83, 27.6 %) and antidepressants (n = 75, 25.0 %).
Results Demographics
PIPs determined by STOPP criteria The 300 study participants had a mean ± SD age of 81.9 ± 7.2 years, and 62.3 % of them were females (Table 1). The causes for admission to the geriatric division were: falls (47.0 %), neurologic (19.3 %), orthopedic (17.0 %), infection (4.3 %), cardiovascular (3.6 %), gastrointestinal (3.0 %), urologic (2.6 %), respiratory (2.0 %) or miscellaneous (1.2 %). The length of hospital stay was 36.0 ± 21.5 days. Hypertension was by far the most common comorbidity (73.3 %), followed by diabetes mellitus (32.0 %), ischemic heart disease (23.6 %) and atrial fibrillation (19.0 %). There were 135 patients (45.0 %) who had a CCI score C2, and 101 (33.6 %) of the patients had some cognitive impairment. The mean number of medications used at admission was 6.8 ± 3.4, and polypharmacy ([8 medications) was recorded for 91 (30.3 %) patients. The five most commonly prescribed classes of drugs were: antithrombotic agents (n = 223, 74.3 %), antihypertensives (n = 209, 69.6 %), lipid-modifying agents
Table 1 Characteristics of the study population (N = 300) Variable
Mean ± SD (%)
Gender Female Age [years (mean ± SD)] 65–74
187 (62.3 %) 81.9 ± 7.2 51 (17.0 %)
75–84
128 (42.6 %)
C85
121 (40.3 %)
Type of residence Home Nursing home
273 (91.0 %) 24 (8.0 %)
Sheltered accommodation
3 (1.0 %)
CCI C 2
135 (45.0 %)
Number of medications
6.8 ± 3.4
Geriatric characteristics Polypharmacy ([8 drugs)
91 (30.3 %)
C1 fall within 3 mo pre-admission
141 (47.0 %)
Living alone
131 (43.6 %)
Need help with C1 ADLs
107 (35.6 %)
Cognitive disorder (MMSE \ 24)
101 (33.6 %)
Most frequent comorbidities Hypertension
220 (73.3 %)
Diabetes mellitus
96 (32.0 %)
Ischemic heart disease Atrial fibrillation
71 (23.6 %) 57 (19.0 %)
ADLs activities of daily living, CCI Charlson comorbidity index, MMSE mini-mental state examination, SD standard deviation
At admission, the prevalence of PIPs as defined by STOPP criteria was 39.3 % (118 patients, 172 PIPs) and it increased to 46.0 % (138 patients, 209 PIPs) at discharge from the hospital (P = 0.009; Table 2). The mean number of drugs increased during hospitalization, from 6.8 ± 3.4 at admission to 8.1 ± 3.3 at discharge (P \ 0.001). Polypharmacy also increased, from 30.6 % at admission to 44.6 % at discharge (P \ 0.001). The most commonly identified PIPs on admission and discharge (Table 2) were proton pump inhibitors for peptic ulcers at full therapeutic dosage for [8 weeks (14 % at admission and 15 % at discharge) and benzodiazepines for patients at risk of falls (9 and 15 %, respectively). The significant individual STOPP criteria which changed from admission to discharge were the use of benzodiazepines in patients with a history of falls, increasing from 9 to 15 % (P \ 0.001), the use of opiates in patients with a history of falls, increasing from 0 to 4.3 % (P \ 0.001), and the use of neuroleptic drugs in patients at risk of falls, increasing from 2.3 to 5.6 % (P = 0.013). The use of glibenclamide decreased from 3.3 % at admission to 0.3 % at discharge (P = 0.03). 40 (13.3 %) patients had no PIPs at admission but were discharged with at least one PIP, and 19 (6.3 %) patients were admitted with at least one PIP and were discharged with no PIPs. Multiple logistic regression analysis revealed that older age (OR 1.05, 95 % CI 1.01–1.08), the number of medications on admission (OR 1.17, 95 % CI 1.08–1.26), a history of falls (OR 1.96, 95 % CI 1.04–3.69) and osteoporosis (OR 1.98, 95 % CI 1.06–3.67) were associated with having PIPs at discharge. A longer hospital stay (OR 1.02, 95 % CI 1.001–1.03) and a history of falls (OR 2.25, 95 % CI 1.03–4.90) increased the risk of being a ‘‘new PIP patient’’, whereas a higher number of medications at admission lowered the risk of being a ‘‘new PIP patient’’ (OR 0.87, 95 % CI 0.78–0.97). Gender, cognitive state, function, CCI and diabetes were not predictors for PIPs. PPOs determined by START criteria At admission, the prevalence of PPOs as defined by START criteria was 41.0 % (123 patients, 153 PPOs), and it decreased to 28.3 % (85 patients, 99 PPOs) at discharge (P \ 0.001, Table 3). The most commonly identified PPOs on admission and discharge (Table 3) were the omission of calcium and vitamin D supplements in patients with known osteoporosis (8.6 and 4.6 %, respectively) and the omission of statins in patients with diabetes mellitus and coexisting
123
Int J Clin Pharm Table 2 Potentially inappropriate prescriptions (PIPs) identified by STOPP (300 patients) STOPP criteria
Admission N (%)
Discharge N (%)
Cardiovascular system Aspirin [ 150 mg/day
5 (1.6)
Aspirin with no history of coronary, cerebral or peripheral vascular symptoms or occlusive event
1 (0.3)
3 (1.0) 0
Non-cardioselective b-blocker with COPD
2 (0.6)
2 (0.6)
Use of diltiazem or verapamil in NYHA class 3/4 heart failure
2 (0.6)
2 (0.6)
Calcium channel blockers with chronic constipation
2 (0.6)
4 (1.3)
Loop diuretics as first-line monotherapy for hypertension
3 (1.0)
2 (0.6)
Warfarin for first uncomplicated DVT [6 months
1 (0.3)
0
Aspirin with concurrent bleeding disorder
0
1 (0.3)
b Blocker in combination with verapamil
0
1 (0.3)
CNS and psychotropic drugs Long-term long-acting benzodiazepines
7 (2.3)
9 (3.0)
Long-term neuroleptics as long-term hypnotics
2 (0.6)
8 (2.6)
Long-term neuroleptics in patients with parkinsonism
2 (0.6)
5 (1.6)
SSRIs in patients with history of clinically significant hyponatremia
7 (2.3)
5 (1.6)
TCA in patients with dementia
2 (0.6)
2 (0.6)
TCA in patients with cardiac conduction abnormalities
2 (0.6)
1 (0.3)
TCA in patients with constipation
4 (1.3)
3 (1.0)
TCA with an opiate or calcium channel blocker
3 (1.0)
Prolonged use of first-generation antihistamines
0
0 1 (0.3)
Gastrointestinal system PPIs for PUD at full therapeutic dosage for [8 weeks
42 (14.0)
45 (15.0)
Respiratory system Systemic corticosteroids instead of inhaled corticosteroids in patients with moderate-to-severe COPD
1 (0.3)
Ipratropium in patients with glaucoma
3 (1.0)
3 (1.0)
0
NSAIDs in patients with moderate or severe hypertension Colchicine for chronic gout where there is no contraindication
1 (0.3) 2 (0.6)
0 4 (1.3)
Long-term corticosteroids as monotherapy for RA or osteoarthritis
1 (0.3)
1 (0.3)
Musculoskeletal system
Urogenital system Antimuscarinic drugs in patients with dementia
4 (1.3)
3 (0.3)
a-Blocker in males with frequent incontinence
5 (1.6)
3 (0.3)
a-Blocker in patients with long term urinary catheter
3 (1.0)
4 (0.3)
Antimuscarinic drugs in patients with chronic glaucoma
0
1 (0.3)
Antimuscarinic drugs in patients with chronic constipation
0
2 (0.6)
Antimuscarinic drugs in patients with chronic prostatitsm
0
1 (0.3)
Endocrine system Glibenclamide in patients with type 2 diabetes mellitus
10(3.3)
1(0.3)a
b-Blockers in patients with diabetes mellitus and frequent hypoglycemic episodes
1 (0.3)
0
1 (0.3)
0
Analgesic drugs Use of long-term powerful opiates as first-line therapy for mild-to-moderate pain Regular opiates for more than 2 weeks in patients with constipation not taking laxatives Duplicate drug classes Any duplicate drug class prescription
0
1 (0.3)
4 (1.3)
6 (2.0)
Benzodiazepines
27 (9.0)
45 (15.0)b
Neuroleptic drugs
7 (2.3)
17 (5.6)a
15 (5.0)
10 (3.3)
Drugs that adversely affect fallers
Vasodilator drugs with persistent postural hypotension
123
Int J Clin Pharm Table 2 continued STOPP criteria Opiates
Admission N (%) 0 172 (PIPs)
Discharge N (%) 13 (4.3)b 209 (PIPs)
ACE angiotensin converting enzyme, CHF chronic heart failure, CNS central nervous system, COPD chronic obstructive pulmonary disease, DVT deep vein thrombosis, MI myocardial infarction, NSAIDs non steroidal anti-inflammatory drugs, PPIs proton pump inhibitors, PUD peptic ulcer disease, RA rheumatoid arthritis, SSRIs serotonin reuptake inhibitors, TCA tricyclic antidepressant a
P \ 0.05 (McNemar’s test)
b
P \ 0.001 (McNemar’s test)
cardiovascular risk factors (8.6 and 7.0 %, respectively). The significant individual START criteria which changed from admission to discharge were the omission of calcium and vitamin D supplements in patients with known osteoporosis, decreasing from 8.6 to 4.6 %, respectively (P = 0.004), the omission of aspirin or clopidogrel in patients with a documented history of atherosclerotic coronary, cerebral or peripheral vascular disease, decreasing from 5.3 to 2.6 %, respectively (P = 0.04), and the omission of statins for patients with a documented history of coronary, cerebral or peripheral vascular disease, decreasing from 3.0 to 1.0 %, respectively (P = 0.03). Fifty-one (17.0 %) patients had at least one PPO at admission and were discharged with no PPOs, and 13 (4.3 %) patients were admitted with no PPOs and were discharged with at least one PPO. Multiple logistic regression analysis revealed that older age (OR 1.05, 95 % CI 1.01–1.09), diabetes (OR 4.6, 95 % CI 2.57–8.23) and osteoporosis (OR 2.44, 95 % CI 1.30–4.58) were associated with having PPOs at discharge. Longer hospital stay (OR 1.02, 95 % CI 1.002–1.04) and diabetes (OR 3.86, 95 % CI 1.2–12.5) increased the risk of being a ‘‘new PPO patient’’. Gender, cognitive state, function, CCI, history of falls and number of medications at admission were not predictors for PPOs.
Discussion The current study results revealed a significant increase in the prevalence of PIPs, as defined by the STOPP criteria, and a significant decrease in PPOs, as defined by the START criteria, from the time of admission to the time of discharge. They also showed a significant increase in the mean number of drugs and polypharmacy from admission to discharge. The conclusions of studies that compared the prevalence of PIPs at admission and discharge are conflicting. Publications on PIPs during hospitalization differ in terms of the adapted criteria as well as in the characteristics of the hospital facilities, rendering them difficult for comparison.
Onatade et al. [5] reported a significant increase in the number of medications and polypharmacy from admission to discharge, and a reduction of PIPs according to the STOPP criteria on discharge from a specialist health and ageing unit in a hospital in England. Mansur et al. [9] reported a prevalence of PIPs of 43.5 % at admission and 44.4 % at discharge from an acute geriatric ward in a hospital in Israel using the 2003 Beers criteria. Bakken et al. [10] showed a significant increase of PIPs using the NORGEP criteria in an intermediate-care nursing home unit and in hospital wards in Norway. We believe that this is the first observational study which compared the prevalence of PPOs using the START criteria in this healthcare setting. The findings of this study revealed a 39.3 % prevalence of PIPs prior to admission to a medical geriatric division of a medical center according to STOPP criteria, and a 41.0 % prevalence of PPOs according to START criteria. The PIP/PPO rate as determined by the STOPP/ START criteria reported in the current study is consistent with those of previous studies of inappropriate prescribing in hospitalized patients according to STOPP/START criteria [7, 15]. The increase in PIPs in the current study was due to drugs that adversely affect fallers that had been newly administered to patients whose cause for admission was a fall. The significant individual STOPP criteria which changed from admission to discharge were the use of benzodiazepines, opiates and neuroleptics. These drugs had been previously reported to be the most prevalent potentially inappropriate medications that patients were taking during hospitalizations [5, 16, 17]. It is well-recognized that the incidence of psychological symptoms (depression, anxiety, agitation and insomnia) are often caused or exacerbated by medical conditions during hospitalization among geriatric patients [18]. For example, benzodiazepines and antipsychotics are frequently used because of delirium and sleep cycle alterations, which are common complications of critical illness, and opiates are added for pain associated with the reason for the index hospital admission. Even though these PIPs may be appropriate during hospitalization and their benefits being greater than the risks for the individual patient, the indications for their
123
Int J Clin Pharm Table 3 Potentially prescriptions omissions (PPOs) identified by START (300 patients) START criteria
Admission N (%)
Discharge N (%)
Cardiovascular system Warfarin in patients with chronic atrial fibrillation
6 (2.0)
Aspirin in patients with chronic atrial fibrillation for whom warfarin is contraindicated
1 (0.3)
1 (0.3)
16 (5.3)
8 (2.6)a
Aspirin or clopidogrel with a documented history of atherosclerotic coronary, cerebral or peripheral vascular disease with sinus rhythm
0
Antihypertensive therapy where systolic blood pressure [160 mmHg
7 (2.3)
3 (1.0)
Statin therapy in patients with a documented history of coronary, cerebral or peripheral vascular disease and who are independent for ADL and have a life expectancy [5 years
9 (3.0)
3 (1.0)a
ACE inhibitor in patients with CHF
11 (3.6)
14 (4.6)
ACE inhibitor following an acute MI
1 (0.3)
1 (0.3)
11 (3.6)
9 (3.0)
b-Blocker in patients with chronic stable angina Respiratory system Regular inhaled corticosteroids for patients with moderate-to-severe asthma or COPD
1 (0.3)
0
CNS 3 (1.0)
1 (0.3)
Antidepressant drug in patients with moderate/severe depressive symptoms for at least 3 months Gastrointestinal system
L-Dopamine
in patients with idiopathic Parkinson’s disease
1 (0.3)
1 (0.3)
PPIs in patients with severe GERD or peptic stricture requiring dilation
1 (0.3)
Fiber supplement for chronic symptomatic diverticular disease with constipation
2 (0.6)
1 (0.3)
26 (8.6)
14 (4.6)a
12 (4.0)
11 (3.6)
0
Musculoskeletal system Calcium and vitamin D supplement in patients with known osteoporosis Endocrine system Metformin in patients with type 2 diabetes w/wo metabolic syndrome ACE inhibitor or ARB in patients with diabetes and nephropathy
4 (1.3)
Antiplatelet therapy in patients with diabetes mellitus and coexisting major cardiovascular risk factors
15 (5.0)
Statin therapy in patients with diabetes mellitus and coexisting major cardiovascular risk factors
26 (8.6) 153 (PPOs)
0 11 (3.6) 21 (7.0) 99 (PPOs)
ACE angiotensin converting enzyme, ARB angiotensin receptor blockers, CHF chronic heart failure, CNS central nervous system, COPD chronic obstructive pulmonary disease, DVT deep vein thrombosis, GERD gastroesophageal acid reflux disease, MI myocardial infarction, PPIs proton pump inhibitors, RA rheumatoid arthritis, SSRIs serotonin reuptake inhibitors, TCA tricyclic antidepressant a
P \ 0.05 (McNemar’s test)
b
P \ 0.001 (McNemar’s test)
use are usually temporary. Failing to discontinue such medications before hospital discharge is potentially harmful in the long run [19]. Hospitalized older patients are generally more vulnerable than other home dwelling older people when it comes to receiving PIPs. This can be explained by the frail condition of the patients after being hospitalized for serious conditions along with being prescribed with multiple medications due to multiple disease conditions. Clinicians must determine which PIPs should be discontinued before hospital discharge and, if that is not possible because of medical conditions, they should recommend that the primary physician discontinue the medications after hospital discharge. It was reported that the lack of prompt follow-up care after a hospital stay may
123
exacerbate a variety of problems, one of which is suboptimal medication management [20]. The current study showed a decrease in the prevalence of PPOs as defined by START criteria from admission to discharge. The decrease in PPOs during hospitalization was due to the addition of beneficial medications which had not been provided to these patients before admission. It is much easier for physicians to start therapies than to stop them. The START criteria is based on general guidelines so that physicians are more familiar with noticing PPOs than PIPs. The significant individual START criteria which changed from admission to discharge were the prescription of calcium and vitamin D supplements for patients with known osteoporosis, and the prescription of statins and aspirin or
Int J Clin Pharm
clopidogrel for patients with a documented history of atherosclerotic coronary and cerebral or peripheral vascular disease. These PPOs were also the ones most commonly reported in other studies [21]. The omission of beneficial drugs may be partly responsible for admissions to the hospital. Dalleur et al. showed that PPOs were related to 38 of 302 admissions (12.6 %). Those authors reported that 25 % of the patients not receiving musculoskeletal drugs, such as calcium and vitamin D, had a PPO-related admission for a fall with a fracture [22]. We demonstrated that older age and osteoporosis were associated both with PIPs and PPOs at discharge. Additionally, a higher number of medications and a history of falls increased the risk of PIPs at discharge, whereas diabetes increased the risk of PPOs. These risk factors have also been reported in other studies as predictors for PIPs and PPOs [21, 23]. Notably, analysis of our data revealed that every additional day of hospitalization increased the risk of being a ‘‘new PIP/PPO patient’’ by 2 %. They also showed that a history of falls doubled the risk of being a ‘‘new PIP patient’’, and that diabetes increased the risk of being ‘‘a new PPO patient’’ by nearly fourfold. It follows, therefore, that in order to decrease PIPs/PPOs at discharge from the hospital, special attention should be paid to the medications of patients with longer hospital stays, patients with a history of falls and patients with diabetes. This study demonstrated risk factors associated with PIPs/PPOs during hospitalization. In practice, the length of hospital stay is mainly influenced by comorbidities, so we believe that careful medication review is important regardless of the length of hospital stay. The focus should be on an intervention for improving prescribing quality in the hospital, especially before discharge. Implementation of a medication review based on screening tools like STOPP/START criteria in older hospitalized patients is recommended. There is a role for a clinical pharmacist to conduct frequent medication reviews and interventions among older patients during hospitalization. There is also a need for additional interventional research in order to evaluate how inappropriate prescribing affects health outcomes in hospitalized older patients. This study has several limitations. First, it is limited to the experience of one large medical center. Second, having been performed on older patients admitted to acute geriatric wards and in a population with specific comorbidities, our results can not be generalized to the entire elderly population. Third, its retrospective design precluded the possibility for physicians who had initiated PIPs in our study population to explain their reasons for prescribing medications that the STOPP criteria lists as being inappropriate. Fourth, the process of identifying PIPs/PPOs by the study pharmacist was not validated.
Conclusion Hospitalization in a geriatric division resulted in an increase of PIPs and in a decrease in PPOs. Every additional day of hospitalization increased the risk of being a ‘‘new PIP/PPO patient’’. Having a history of falls doubled the risk of being a ‘‘new PIP patient’’, while diabetes increased the risk of being a ‘‘new PPO patient’’ by nearly fourfold. Enhanced monitoring at discharge involving the screening of medications for PIPS/PPOs appears to be an essential component for reducing the PIP/PPO risks for older adults after hospital discharge. Acknowledgments We would like to thank Mrs. Ilana Gelernter from the statistical laboratory in Tel Aviv University for her help in the statistical analysis. Funding
None.
Conflicts of interest
The authors declare no conflicts of interest.
References 1. Hamilton HJ, Gallagher PF, O’Mahony D. Inappropriate prescribing and adverse drug events in older people. BMC Geriatr. 2009;28(9):5. 2. Barry PJ, Gallagher P, Ryan C, O’mahony D. START (screening tool to alert doctors to the right treatment)—an evidence-based screening tool to detect prescribing omissions in elderly patients. Age Ageing. 2007;36:632–8. 3. O’Connor MN, Gallagher P, O’Mahony D. Inappropriate prescribing: criteria, detection and prevention. Drugs Aging. 2012;29(6):437–52. 4. Gallagher P, Ryan C, Byrne S, Kennedy J, O’Mahony D. STOPP (Screening Tool of Older Person’s Prescriptions) and START (Screening Tool to Alert doctors to Right Treatment). Consensus validation. Int J Clin Pharmacol Ther. 2008;46:72–83. 5. Onatade R, Auyeung V, Scutt G, Fernando J. Potentially inappropriate prescribing in patients on admission and discharge from an older peoples’ unit of an acute UK hospital. Drugs Aging. 2013;30:729–37. 6. Lang PO, Vogt-Ferrier N, Hasso Y, Le Saint L, Drame´ M, Zekry D, et al. Interdisciplinary geriatric and psychiatric care reduces potentially inappropriate prescribing in the hospital: interventional study in 150 acutely ill elderly patients with mental and somatic comorbid conditions. J Am Med Dir Assoc. 2012;13: 406.e1–7. 7. Liu CL, Peng LN, Chen YT, Lin MH, Liu LK, Chen LK. Potentially inappropriate prescribing (IP) for elderly medical inpatients in Taiwan: a hospital-based study. Arch Gerontol Geriatr. 2012;55:148–51. 8. Lang PO, Hasso Y, Drame´ M, Vogt-Ferrier N, Prudent M, Gold G, et al. Potentially inappropriate prescribing including under-use amongst older patients with cognitive or psychiatric co-morbidities. Age Ageing. 2010;39:373–81. 9. Mansur N, Weiss A, Beloosesky Y. Is there an association between inappropriate prescription drug use and adherence in discharged elderly patients? Ann Pharmacother. 2009;43:177–84. 10. Bakken MS, Ranhoff AH, Engeland A, Ruths S. Inappropriate prescribing for older people admitted to an intermediate-care nursing home unit and hospital wards. Scand J Prim Health Care. 2012;30:169–75.
123
Int J Clin Pharm 11. Laroche ML, Charmes JP, Nouaille Y, Fourrier A, Merle L. Impact of hospitalisation in an acute medical geriatric unit on potentially inappropriate medication use. Drugs Aging. 2006;23:49–59. 12. Folstein MF, Folstein SE, McHugh PR. ‘‘Mini-mental state’’. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189–98. 13. Katz S, Downs TD, Cash HR, Grotz RC. Progress in development of the index of ADL. Gerontologist. 1970;10:20–30. 14. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373–83. 15. Gallagher P, Omahony D. STOPP (Screening tool of older persons’ potentially inappropriate prescriptions): application to acutely ill elderly patients and comparison with Beers’ criteria. Age Ageing. 2008;37:673–9. 16. Wahab MS, Nyfort-Hansen K, Kowalski SR. Inappropriate prescribing in hospitalised Australian elderly as determined by the STOPP criteria. Int J Clin Pharm. 2012;34:855–62. 17. Gallagher P, Lang PO, Cherubini A, Topinkova´ E, Cruz-Jentoft A, Montero Errasquı´n B, et al. Prevalence of potentially inappropriate prescribing in an acutely ill population of older patients admitted to six European hospitals. Eur J Clin Pharmacol. 2011;67:1175–88.
123
18. Kane RL, Quslander JG, Abrass IB, Resnick B. Essentials of clinical geriatrics. 6th ed. New York: McGraw-Hill; 2009. 19. Morandi A, Vasilevskis E, Pandharipande PP, Girard TD, Solberg LM, Neal EB, et al. Inappropriate medication prescriptions in elderly adults surviving an intensive care unit hospitalization. J Am Geriatr Soc. 2013;61:1128–34. 20. Kanaan AO, Donovan JL, Duchin NP, Field TS, Tjia J, Cutrona SL, et al. Adverse drug events after hospital discharge in older adults: types, severity, and involvement of Beers criteria medications. J Am Geriatr Soc. 2013;61:1894–9. 21. Lam MP, Cheung BM. The use of STOPP/START criteria as a screening tool for assessing the appropriateness of medications in the elderly population. Expert Rev Clin Pharmacol. 2012;5:187–97. 22. Dalleur O, Spinewine A, Henrard S, Losseau C, Speybroeck N, Boland B. Inappropriate prescribing and related hospital admissions in frail older persons according to the STOPP and START criteria. Drugs Aging. 2012;29:829–37. 23. Hill-Taylor B, Sketris I, Hayden J, Byrne S, O’Sullivan D, Christie R. 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. J Clin Pharm Ther. 2013;38:360–72.
RESEARCH ARTICLE
The impact of hospitalization on potentially inappropriate prescribing in an acute medical geriatric division Dvora Frankenthal • Yaffa Lerman Yehuda Lerman
•
Received: 1 May 2014 / Accepted: 12 November 2014 Koninklijke Nederlandse Maatschappij ter bevordering der Pharmacie 2014
Abstract Background Screening Tool of Older Person’s Prescriptions (STOPP) and the Screening Tool to Alert doctors to Right Treatment (START) have been increasingly used to evaluate potentially inappropriate prescriptions (PIPs) and potentially prescription omissions (PPOs). The impact of hospitalization on PIPs/PPOs has not been investigated in depth. Objective To compare the prevalence of PIPs/PPOs in elderly patients on hospital admission and discharge and to identify associated risk factors. Setting An acute medical geriatric division of the Tel Aviv Medical Center (Israel). Method This retrospective cross-sectional study included patients admitted from 12/2011 to 12/2012 aged C65 years. Data from patients’ records included demographic details, diagnoses and medications at admission and discharge. STOPP/START criteria were applied to each patient’s record. Main outcome measure Prevalence of PIPs/PPOs on hospital admission and discharge. Results Three hundred patients were included (mean ± SD age 81.9 ± 7.2 years). Admission PIPs prevalence was 39.3 % (118 patients, 172 PIPs) and it increased to 46.0 % (138 patients, 209 PIPs) at discharge (P = 0.009). Admission PPOs prevalence was 41.0 % (123 patients, 153 PPOs) and it decreased to 28.3 % (85 patients, 99 PPOs) at discharge (P \ 0.001). Having at least one PIP/PPO at discharge but not at admission was associated with length of hospital stay (OR 1.02, 95 % CI 1.001–1.03). History of falls increased the risk of being a ‘‘new PIP patient’’ (OR 2.25, 95 % CI D. Frankenthal (&) Y. Lerman Y. Lerman School of Public Health, Sackler Faculty of Medicine, Tel Aviv University, POB 39040, Tel Aviv, Israel e-mail: [email protected] Y. Lerman Geriatric Division, Tel Aviv Sourasky Medical Center, 6 Weizmann St., Tel Aviv, Israel
1.03–4.9), whereas diabetes increased the risk of being a ‘‘new PPO patient’’ (OR 3.86, 95 % CI 1.2–12.5). Conclusion Hospitalization in a geriatric division resulted in an increase in PIPs and a decrease in PPOs. Strategies to reduce PIPs need to be implemented, especially for patients with longer hospital stay and a history of falls. Keywords Elderly Hospitalization Israel Medications START criteria STOPP criteria
Impacts on Practice •
•
•
In Israel, actions to promote appropriate prescribing in hospitals are needed given the increase in PIPs during hospitalization among elderly patients in geriatric wards. Enhanced monitoring at discharge involving the screening of medications for PIPs/PPOs is an essential part of the care for geriatric patients. To decrease PIPs and PPOs at discharge, special attention should be paid to the medications of patients with longer hospital stays, with a history of falls, and with diabetes.
Introduction Potentially inappropriate prescriptions (PIPs) in older people are becoming a global healthcare concern in an ageing population. A PIP is the use of medicines that pose more risk than benefit, particularly when there are safer alternatives [1]. Another aspect of PIPs in older people is the omission of indicated medications with proven efficacy in patients with an ordinary life expectancy [2]. PIPs can be
123
Int J Clin Pharm
detected by using explicit (criterion-based) prescribing indicators whose ultimate aim is to optimize prescription appropriateness and reduce negative outcomes, including preventable adverse drug effects [3]. Various criteria have been devised to identify PIPs in older people, one of the most recent being the Screening Tool of Older Person’s potentially inappropriate Prescriptions (STOPP) and the Screening Tool to Alert doctors to Right Treatment (START). The STOPP criteria focus upon avoiding the use of drugs that are potentially inappropriate in older people, and the START criteria aim to identify under-treatment or omissions in prescribing of medications among older adults [4]. The prevalence of patients with at least one PIP as identified by the STOPP criteria was reported as being between 26.7 and 77 % [5, 6] of hospitalized older patients, and the reported prevalence of potentially prescription omissions (PPOs) according to START criteria was between 41.9 and 65 % [7, 8]. The influence of hospitalization on the prescribing of drugs is difficult to assess. Some studies reported a decrease in PIPs after a hospital stay, whereas others showed an increase [5, 9–11]. The applicability of the STOPP/START criteria in Israeli hospitals is unknown, and no studies have thus far applied these screening tools in hospitalized older Israeli patients.
Study population Three-hundred patients were randomly selected for this study from all patients (N = 900) aged 65 years or older who had been admitted to the geriatric division over a 12-month period (12/2011–12/2012). Subjects who died during hospitalization were excluded. Data collection Data were collected by means of the electronic patient record. The patients’ data included demographic details, medical histories, reason for admission, history of falls, diagnoses and medications on admission and discharge, biochemistry results, length of hospital stay and cognitive disorder (known dementia and/or an impaired Mini Mental State Examination [MMSE] score of \24/30) [12]. A disabled patient was defined as functionally dependent for at least one item in the Katz scale [13]. Comorbidity was quantified by adding the scores assigned to specific diagnoses in the Charlson Comorbidity Index (CCI) [14]. Polypharmacy (defined by the Israeli Ministry of Health as [8 medications) pertained only to medications used chronically. The frequencies of PIPs and PPOs in the studied sample were evaluated by the study pharmacist using the STOPP/START criteria and they were calculated both at admission and at discharge.
Aim of the study
Statistical analysis
The aim of this study was to compare PIPs/PPOs in elderly patients at the time of admission and discharge from an acute medical geriatric division, and to identify the factors associated with having a PIP/PPO at hospital discharge.
Statistical analysis was performed using the Statistical Package of Social Sciences version 14.0 (Chicago, IL, USA). The PIP/PPO proportion was defined as the number of patients with at least one PIP/PPO according to the STOPP/START criteria. McNemar’s test was used for paired data to examine any change in prescribing in order to compare prescribing practices pre-and post-hospitalization using STOPP/START criteria, The Wilcoxon matched pairs signed ranks test was used to compare the number of medications prescribed to the patients at admission and at discharge. A ‘‘new PIP/PPO patient’’ was defined as having at least one PIP/PPO at discharge but none at admission. Multivariate analysis using a backward stepwise logistic regression was applied to examine the association between independent risk factors and overall PIPs/PPOs at discharge and being a ‘‘new PIP/PPO patient’’ at discharge. The following variables were entered in the logistic regression model: age, gender, number of medications at admission, CCI, functioning, cognitive state, history of falls, diabetes, osteoporosis and length of hospital stay. Significance was set at P \ 0.05.
Ethical approval The study protocol was reviewed and approved by the hospital’s institutional ethical review board.
Methods This retrospective cross-sectional study was conducted in a medical geriatric division of the Tel Aviv Medical Center, which is the second largest (1,300 beds) multidisciplinary academic medical center in Israel. The geriatric division comprises two wards with a total of 70 beds, and it treats and rehabilitates older patients with neurological and orthopedic conditions.
123
Int J Clin Pharm
(n = 150, 50.0 %), antidiabetics (n = 83, 27.6 %) and antidepressants (n = 75, 25.0 %).
Results Demographics
PIPs determined by STOPP criteria The 300 study participants had a mean ± SD age of 81.9 ± 7.2 years, and 62.3 % of them were females (Table 1). The causes for admission to the geriatric division were: falls (47.0 %), neurologic (19.3 %), orthopedic (17.0 %), infection (4.3 %), cardiovascular (3.6 %), gastrointestinal (3.0 %), urologic (2.6 %), respiratory (2.0 %) or miscellaneous (1.2 %). The length of hospital stay was 36.0 ± 21.5 days. Hypertension was by far the most common comorbidity (73.3 %), followed by diabetes mellitus (32.0 %), ischemic heart disease (23.6 %) and atrial fibrillation (19.0 %). There were 135 patients (45.0 %) who had a CCI score C2, and 101 (33.6 %) of the patients had some cognitive impairment. The mean number of medications used at admission was 6.8 ± 3.4, and polypharmacy ([8 medications) was recorded for 91 (30.3 %) patients. The five most commonly prescribed classes of drugs were: antithrombotic agents (n = 223, 74.3 %), antihypertensives (n = 209, 69.6 %), lipid-modifying agents
Table 1 Characteristics of the study population (N = 300) Variable
Mean ± SD (%)
Gender Female Age [years (mean ± SD)] 65–74
187 (62.3 %) 81.9 ± 7.2 51 (17.0 %)
75–84
128 (42.6 %)
C85
121 (40.3 %)
Type of residence Home Nursing home
273 (91.0 %) 24 (8.0 %)
Sheltered accommodation
3 (1.0 %)
CCI C 2
135 (45.0 %)
Number of medications
6.8 ± 3.4
Geriatric characteristics Polypharmacy ([8 drugs)
91 (30.3 %)
C1 fall within 3 mo pre-admission
141 (47.0 %)
Living alone
131 (43.6 %)
Need help with C1 ADLs
107 (35.6 %)
Cognitive disorder (MMSE \ 24)
101 (33.6 %)
Most frequent comorbidities Hypertension
220 (73.3 %)
Diabetes mellitus
96 (32.0 %)
Ischemic heart disease Atrial fibrillation
71 (23.6 %) 57 (19.0 %)
ADLs activities of daily living, CCI Charlson comorbidity index, MMSE mini-mental state examination, SD standard deviation
At admission, the prevalence of PIPs as defined by STOPP criteria was 39.3 % (118 patients, 172 PIPs) and it increased to 46.0 % (138 patients, 209 PIPs) at discharge from the hospital (P = 0.009; Table 2). The mean number of drugs increased during hospitalization, from 6.8 ± 3.4 at admission to 8.1 ± 3.3 at discharge (P \ 0.001). Polypharmacy also increased, from 30.6 % at admission to 44.6 % at discharge (P \ 0.001). The most commonly identified PIPs on admission and discharge (Table 2) were proton pump inhibitors for peptic ulcers at full therapeutic dosage for [8 weeks (14 % at admission and 15 % at discharge) and benzodiazepines for patients at risk of falls (9 and 15 %, respectively). The significant individual STOPP criteria which changed from admission to discharge were the use of benzodiazepines in patients with a history of falls, increasing from 9 to 15 % (P \ 0.001), the use of opiates in patients with a history of falls, increasing from 0 to 4.3 % (P \ 0.001), and the use of neuroleptic drugs in patients at risk of falls, increasing from 2.3 to 5.6 % (P = 0.013). The use of glibenclamide decreased from 3.3 % at admission to 0.3 % at discharge (P = 0.03). 40 (13.3 %) patients had no PIPs at admission but were discharged with at least one PIP, and 19 (6.3 %) patients were admitted with at least one PIP and were discharged with no PIPs. Multiple logistic regression analysis revealed that older age (OR 1.05, 95 % CI 1.01–1.08), the number of medications on admission (OR 1.17, 95 % CI 1.08–1.26), a history of falls (OR 1.96, 95 % CI 1.04–3.69) and osteoporosis (OR 1.98, 95 % CI 1.06–3.67) were associated with having PIPs at discharge. A longer hospital stay (OR 1.02, 95 % CI 1.001–1.03) and a history of falls (OR 2.25, 95 % CI 1.03–4.90) increased the risk of being a ‘‘new PIP patient’’, whereas a higher number of medications at admission lowered the risk of being a ‘‘new PIP patient’’ (OR 0.87, 95 % CI 0.78–0.97). Gender, cognitive state, function, CCI and diabetes were not predictors for PIPs. PPOs determined by START criteria At admission, the prevalence of PPOs as defined by START criteria was 41.0 % (123 patients, 153 PPOs), and it decreased to 28.3 % (85 patients, 99 PPOs) at discharge (P \ 0.001, Table 3). The most commonly identified PPOs on admission and discharge (Table 3) were the omission of calcium and vitamin D supplements in patients with known osteoporosis (8.6 and 4.6 %, respectively) and the omission of statins in patients with diabetes mellitus and coexisting
123
Int J Clin Pharm Table 2 Potentially inappropriate prescriptions (PIPs) identified by STOPP (300 patients) STOPP criteria
Admission N (%)
Discharge N (%)
Cardiovascular system Aspirin [ 150 mg/day
5 (1.6)
Aspirin with no history of coronary, cerebral or peripheral vascular symptoms or occlusive event
1 (0.3)
3 (1.0) 0
Non-cardioselective b-blocker with COPD
2 (0.6)
2 (0.6)
Use of diltiazem or verapamil in NYHA class 3/4 heart failure
2 (0.6)
2 (0.6)
Calcium channel blockers with chronic constipation
2 (0.6)
4 (1.3)
Loop diuretics as first-line monotherapy for hypertension
3 (1.0)
2 (0.6)
Warfarin for first uncomplicated DVT [6 months
1 (0.3)
0
Aspirin with concurrent bleeding disorder
0
1 (0.3)
b Blocker in combination with verapamil
0
1 (0.3)
CNS and psychotropic drugs Long-term long-acting benzodiazepines
7 (2.3)
9 (3.0)
Long-term neuroleptics as long-term hypnotics
2 (0.6)
8 (2.6)
Long-term neuroleptics in patients with parkinsonism
2 (0.6)
5 (1.6)
SSRIs in patients with history of clinically significant hyponatremia
7 (2.3)
5 (1.6)
TCA in patients with dementia
2 (0.6)
2 (0.6)
TCA in patients with cardiac conduction abnormalities
2 (0.6)
1 (0.3)
TCA in patients with constipation
4 (1.3)
3 (1.0)
TCA with an opiate or calcium channel blocker
3 (1.0)
Prolonged use of first-generation antihistamines
0
0 1 (0.3)
Gastrointestinal system PPIs for PUD at full therapeutic dosage for [8 weeks
42 (14.0)
45 (15.0)
Respiratory system Systemic corticosteroids instead of inhaled corticosteroids in patients with moderate-to-severe COPD
1 (0.3)
Ipratropium in patients with glaucoma
3 (1.0)
3 (1.0)
0
NSAIDs in patients with moderate or severe hypertension Colchicine for chronic gout where there is no contraindication
1 (0.3) 2 (0.6)
0 4 (1.3)
Long-term corticosteroids as monotherapy for RA or osteoarthritis
1 (0.3)
1 (0.3)
Musculoskeletal system
Urogenital system Antimuscarinic drugs in patients with dementia
4 (1.3)
3 (0.3)
a-Blocker in males with frequent incontinence
5 (1.6)
3 (0.3)
a-Blocker in patients with long term urinary catheter
3 (1.0)
4 (0.3)
Antimuscarinic drugs in patients with chronic glaucoma
0
1 (0.3)
Antimuscarinic drugs in patients with chronic constipation
0
2 (0.6)
Antimuscarinic drugs in patients with chronic prostatitsm
0
1 (0.3)
Endocrine system Glibenclamide in patients with type 2 diabetes mellitus
10(3.3)
1(0.3)a
b-Blockers in patients with diabetes mellitus and frequent hypoglycemic episodes
1 (0.3)
0
1 (0.3)
0
Analgesic drugs Use of long-term powerful opiates as first-line therapy for mild-to-moderate pain Regular opiates for more than 2 weeks in patients with constipation not taking laxatives Duplicate drug classes Any duplicate drug class prescription
0
1 (0.3)
4 (1.3)
6 (2.0)
Benzodiazepines
27 (9.0)
45 (15.0)b
Neuroleptic drugs
7 (2.3)
17 (5.6)a
15 (5.0)
10 (3.3)
Drugs that adversely affect fallers
Vasodilator drugs with persistent postural hypotension
123
Int J Clin Pharm Table 2 continued STOPP criteria Opiates
Admission N (%) 0 172 (PIPs)
Discharge N (%) 13 (4.3)b 209 (PIPs)
ACE angiotensin converting enzyme, CHF chronic heart failure, CNS central nervous system, COPD chronic obstructive pulmonary disease, DVT deep vein thrombosis, MI myocardial infarction, NSAIDs non steroidal anti-inflammatory drugs, PPIs proton pump inhibitors, PUD peptic ulcer disease, RA rheumatoid arthritis, SSRIs serotonin reuptake inhibitors, TCA tricyclic antidepressant a
P \ 0.05 (McNemar’s test)
b
P \ 0.001 (McNemar’s test)
cardiovascular risk factors (8.6 and 7.0 %, respectively). The significant individual START criteria which changed from admission to discharge were the omission of calcium and vitamin D supplements in patients with known osteoporosis, decreasing from 8.6 to 4.6 %, respectively (P = 0.004), the omission of aspirin or clopidogrel in patients with a documented history of atherosclerotic coronary, cerebral or peripheral vascular disease, decreasing from 5.3 to 2.6 %, respectively (P = 0.04), and the omission of statins for patients with a documented history of coronary, cerebral or peripheral vascular disease, decreasing from 3.0 to 1.0 %, respectively (P = 0.03). Fifty-one (17.0 %) patients had at least one PPO at admission and were discharged with no PPOs, and 13 (4.3 %) patients were admitted with no PPOs and were discharged with at least one PPO. Multiple logistic regression analysis revealed that older age (OR 1.05, 95 % CI 1.01–1.09), diabetes (OR 4.6, 95 % CI 2.57–8.23) and osteoporosis (OR 2.44, 95 % CI 1.30–4.58) were associated with having PPOs at discharge. Longer hospital stay (OR 1.02, 95 % CI 1.002–1.04) and diabetes (OR 3.86, 95 % CI 1.2–12.5) increased the risk of being a ‘‘new PPO patient’’. Gender, cognitive state, function, CCI, history of falls and number of medications at admission were not predictors for PPOs.
Discussion The current study results revealed a significant increase in the prevalence of PIPs, as defined by the STOPP criteria, and a significant decrease in PPOs, as defined by the START criteria, from the time of admission to the time of discharge. They also showed a significant increase in the mean number of drugs and polypharmacy from admission to discharge. The conclusions of studies that compared the prevalence of PIPs at admission and discharge are conflicting. Publications on PIPs during hospitalization differ in terms of the adapted criteria as well as in the characteristics of the hospital facilities, rendering them difficult for comparison.
Onatade et al. [5] reported a significant increase in the number of medications and polypharmacy from admission to discharge, and a reduction of PIPs according to the STOPP criteria on discharge from a specialist health and ageing unit in a hospital in England. Mansur et al. [9] reported a prevalence of PIPs of 43.5 % at admission and 44.4 % at discharge from an acute geriatric ward in a hospital in Israel using the 2003 Beers criteria. Bakken et al. [10] showed a significant increase of PIPs using the NORGEP criteria in an intermediate-care nursing home unit and in hospital wards in Norway. We believe that this is the first observational study which compared the prevalence of PPOs using the START criteria in this healthcare setting. The findings of this study revealed a 39.3 % prevalence of PIPs prior to admission to a medical geriatric division of a medical center according to STOPP criteria, and a 41.0 % prevalence of PPOs according to START criteria. The PIP/PPO rate as determined by the STOPP/ START criteria reported in the current study is consistent with those of previous studies of inappropriate prescribing in hospitalized patients according to STOPP/START criteria [7, 15]. The increase in PIPs in the current study was due to drugs that adversely affect fallers that had been newly administered to patients whose cause for admission was a fall. The significant individual STOPP criteria which changed from admission to discharge were the use of benzodiazepines, opiates and neuroleptics. These drugs had been previously reported to be the most prevalent potentially inappropriate medications that patients were taking during hospitalizations [5, 16, 17]. It is well-recognized that the incidence of psychological symptoms (depression, anxiety, agitation and insomnia) are often caused or exacerbated by medical conditions during hospitalization among geriatric patients [18]. For example, benzodiazepines and antipsychotics are frequently used because of delirium and sleep cycle alterations, which are common complications of critical illness, and opiates are added for pain associated with the reason for the index hospital admission. Even though these PIPs may be appropriate during hospitalization and their benefits being greater than the risks for the individual patient, the indications for their
123
Int J Clin Pharm Table 3 Potentially prescriptions omissions (PPOs) identified by START (300 patients) START criteria
Admission N (%)
Discharge N (%)
Cardiovascular system Warfarin in patients with chronic atrial fibrillation
6 (2.0)
Aspirin in patients with chronic atrial fibrillation for whom warfarin is contraindicated
1 (0.3)
1 (0.3)
16 (5.3)
8 (2.6)a
Aspirin or clopidogrel with a documented history of atherosclerotic coronary, cerebral or peripheral vascular disease with sinus rhythm
0
Antihypertensive therapy where systolic blood pressure [160 mmHg
7 (2.3)
3 (1.0)
Statin therapy in patients with a documented history of coronary, cerebral or peripheral vascular disease and who are independent for ADL and have a life expectancy [5 years
9 (3.0)
3 (1.0)a
ACE inhibitor in patients with CHF
11 (3.6)
14 (4.6)
ACE inhibitor following an acute MI
1 (0.3)
1 (0.3)
11 (3.6)
9 (3.0)
b-Blocker in patients with chronic stable angina Respiratory system Regular inhaled corticosteroids for patients with moderate-to-severe asthma or COPD
1 (0.3)
0
CNS 3 (1.0)
1 (0.3)
Antidepressant drug in patients with moderate/severe depressive symptoms for at least 3 months Gastrointestinal system
L-Dopamine
in patients with idiopathic Parkinson’s disease
1 (0.3)
1 (0.3)
PPIs in patients with severe GERD or peptic stricture requiring dilation
1 (0.3)
Fiber supplement for chronic symptomatic diverticular disease with constipation
2 (0.6)
1 (0.3)
26 (8.6)
14 (4.6)a
12 (4.0)
11 (3.6)
0
Musculoskeletal system Calcium and vitamin D supplement in patients with known osteoporosis Endocrine system Metformin in patients with type 2 diabetes w/wo metabolic syndrome ACE inhibitor or ARB in patients with diabetes and nephropathy
4 (1.3)
Antiplatelet therapy in patients with diabetes mellitus and coexisting major cardiovascular risk factors
15 (5.0)
Statin therapy in patients with diabetes mellitus and coexisting major cardiovascular risk factors
26 (8.6) 153 (PPOs)
0 11 (3.6) 21 (7.0) 99 (PPOs)
ACE angiotensin converting enzyme, ARB angiotensin receptor blockers, CHF chronic heart failure, CNS central nervous system, COPD chronic obstructive pulmonary disease, DVT deep vein thrombosis, GERD gastroesophageal acid reflux disease, MI myocardial infarction, PPIs proton pump inhibitors, RA rheumatoid arthritis, SSRIs serotonin reuptake inhibitors, TCA tricyclic antidepressant a
P \ 0.05 (McNemar’s test)
b
P \ 0.001 (McNemar’s test)
use are usually temporary. Failing to discontinue such medications before hospital discharge is potentially harmful in the long run [19]. Hospitalized older patients are generally more vulnerable than other home dwelling older people when it comes to receiving PIPs. This can be explained by the frail condition of the patients after being hospitalized for serious conditions along with being prescribed with multiple medications due to multiple disease conditions. Clinicians must determine which PIPs should be discontinued before hospital discharge and, if that is not possible because of medical conditions, they should recommend that the primary physician discontinue the medications after hospital discharge. It was reported that the lack of prompt follow-up care after a hospital stay may
123
exacerbate a variety of problems, one of which is suboptimal medication management [20]. The current study showed a decrease in the prevalence of PPOs as defined by START criteria from admission to discharge. The decrease in PPOs during hospitalization was due to the addition of beneficial medications which had not been provided to these patients before admission. It is much easier for physicians to start therapies than to stop them. The START criteria is based on general guidelines so that physicians are more familiar with noticing PPOs than PIPs. The significant individual START criteria which changed from admission to discharge were the prescription of calcium and vitamin D supplements for patients with known osteoporosis, and the prescription of statins and aspirin or
Int J Clin Pharm
clopidogrel for patients with a documented history of atherosclerotic coronary and cerebral or peripheral vascular disease. These PPOs were also the ones most commonly reported in other studies [21]. The omission of beneficial drugs may be partly responsible for admissions to the hospital. Dalleur et al. showed that PPOs were related to 38 of 302 admissions (12.6 %). Those authors reported that 25 % of the patients not receiving musculoskeletal drugs, such as calcium and vitamin D, had a PPO-related admission for a fall with a fracture [22]. We demonstrated that older age and osteoporosis were associated both with PIPs and PPOs at discharge. Additionally, a higher number of medications and a history of falls increased the risk of PIPs at discharge, whereas diabetes increased the risk of PPOs. These risk factors have also been reported in other studies as predictors for PIPs and PPOs [21, 23]. Notably, analysis of our data revealed that every additional day of hospitalization increased the risk of being a ‘‘new PIP/PPO patient’’ by 2 %. They also showed that a history of falls doubled the risk of being a ‘‘new PIP patient’’, and that diabetes increased the risk of being ‘‘a new PPO patient’’ by nearly fourfold. It follows, therefore, that in order to decrease PIPs/PPOs at discharge from the hospital, special attention should be paid to the medications of patients with longer hospital stays, patients with a history of falls and patients with diabetes. This study demonstrated risk factors associated with PIPs/PPOs during hospitalization. In practice, the length of hospital stay is mainly influenced by comorbidities, so we believe that careful medication review is important regardless of the length of hospital stay. The focus should be on an intervention for improving prescribing quality in the hospital, especially before discharge. Implementation of a medication review based on screening tools like STOPP/START criteria in older hospitalized patients is recommended. There is a role for a clinical pharmacist to conduct frequent medication reviews and interventions among older patients during hospitalization. There is also a need for additional interventional research in order to evaluate how inappropriate prescribing affects health outcomes in hospitalized older patients. This study has several limitations. First, it is limited to the experience of one large medical center. Second, having been performed on older patients admitted to acute geriatric wards and in a population with specific comorbidities, our results can not be generalized to the entire elderly population. Third, its retrospective design precluded the possibility for physicians who had initiated PIPs in our study population to explain their reasons for prescribing medications that the STOPP criteria lists as being inappropriate. Fourth, the process of identifying PIPs/PPOs by the study pharmacist was not validated.
Conclusion Hospitalization in a geriatric division resulted in an increase of PIPs and in a decrease in PPOs. Every additional day of hospitalization increased the risk of being a ‘‘new PIP/PPO patient’’. Having a history of falls doubled the risk of being a ‘‘new PIP patient’’, while diabetes increased the risk of being a ‘‘new PPO patient’’ by nearly fourfold. Enhanced monitoring at discharge involving the screening of medications for PIPS/PPOs appears to be an essential component for reducing the PIP/PPO risks for older adults after hospital discharge. Acknowledgments We would like to thank Mrs. Ilana Gelernter from the statistical laboratory in Tel Aviv University for her help in the statistical analysis. Funding
None.
Conflicts of interest
The authors declare no conflicts of interest.
References 1. Hamilton HJ, Gallagher PF, O’Mahony D. Inappropriate prescribing and adverse drug events in older people. BMC Geriatr. 2009;28(9):5. 2. Barry PJ, Gallagher P, Ryan C, O’mahony D. START (screening tool to alert doctors to the right treatment)—an evidence-based screening tool to detect prescribing omissions in elderly patients. Age Ageing. 2007;36:632–8. 3. O’Connor MN, Gallagher P, O’Mahony D. Inappropriate prescribing: criteria, detection and prevention. Drugs Aging. 2012;29(6):437–52. 4. Gallagher P, Ryan C, Byrne S, Kennedy J, O’Mahony D. STOPP (Screening Tool of Older Person’s Prescriptions) and START (Screening Tool to Alert doctors to Right Treatment). Consensus validation. Int J Clin Pharmacol Ther. 2008;46:72–83. 5. Onatade R, Auyeung V, Scutt G, Fernando J. Potentially inappropriate prescribing in patients on admission and discharge from an older peoples’ unit of an acute UK hospital. Drugs Aging. 2013;30:729–37. 6. Lang PO, Vogt-Ferrier N, Hasso Y, Le Saint L, Drame´ M, Zekry D, et al. Interdisciplinary geriatric and psychiatric care reduces potentially inappropriate prescribing in the hospital: interventional study in 150 acutely ill elderly patients with mental and somatic comorbid conditions. J Am Med Dir Assoc. 2012;13: 406.e1–7. 7. Liu CL, Peng LN, Chen YT, Lin MH, Liu LK, Chen LK. Potentially inappropriate prescribing (IP) for elderly medical inpatients in Taiwan: a hospital-based study. Arch Gerontol Geriatr. 2012;55:148–51. 8. Lang PO, Hasso Y, Drame´ M, Vogt-Ferrier N, Prudent M, Gold G, et al. Potentially inappropriate prescribing including under-use amongst older patients with cognitive or psychiatric co-morbidities. Age Ageing. 2010;39:373–81. 9. Mansur N, Weiss A, Beloosesky Y. Is there an association between inappropriate prescription drug use and adherence in discharged elderly patients? Ann Pharmacother. 2009;43:177–84. 10. Bakken MS, Ranhoff AH, Engeland A, Ruths S. Inappropriate prescribing for older people admitted to an intermediate-care nursing home unit and hospital wards. Scand J Prim Health Care. 2012;30:169–75.
123
Int J Clin Pharm 11. Laroche ML, Charmes JP, Nouaille Y, Fourrier A, Merle L. Impact of hospitalisation in an acute medical geriatric unit on potentially inappropriate medication use. Drugs Aging. 2006;23:49–59. 12. Folstein MF, Folstein SE, McHugh PR. ‘‘Mini-mental state’’. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189–98. 13. Katz S, Downs TD, Cash HR, Grotz RC. Progress in development of the index of ADL. Gerontologist. 1970;10:20–30. 14. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373–83. 15. Gallagher P, Omahony D. STOPP (Screening tool of older persons’ potentially inappropriate prescriptions): application to acutely ill elderly patients and comparison with Beers’ criteria. Age Ageing. 2008;37:673–9. 16. Wahab MS, Nyfort-Hansen K, Kowalski SR. Inappropriate prescribing in hospitalised Australian elderly as determined by the STOPP criteria. Int J Clin Pharm. 2012;34:855–62. 17. Gallagher P, Lang PO, Cherubini A, Topinkova´ E, Cruz-Jentoft A, Montero Errasquı´n B, et al. Prevalence of potentially inappropriate prescribing in an acutely ill population of older patients admitted to six European hospitals. Eur J Clin Pharmacol. 2011;67:1175–88.
123
18. Kane RL, Quslander JG, Abrass IB, Resnick B. Essentials of clinical geriatrics. 6th ed. New York: McGraw-Hill; 2009. 19. Morandi A, Vasilevskis E, Pandharipande PP, Girard TD, Solberg LM, Neal EB, et al. Inappropriate medication prescriptions in elderly adults surviving an intensive care unit hospitalization. J Am Geriatr Soc. 2013;61:1128–34. 20. Kanaan AO, Donovan JL, Duchin NP, Field TS, Tjia J, Cutrona SL, et al. Adverse drug events after hospital discharge in older adults: types, severity, and involvement of Beers criteria medications. J Am Geriatr Soc. 2013;61:1894–9. 21. Lam MP, Cheung BM. The use of STOPP/START criteria as a screening tool for assessing the appropriateness of medications in the elderly population. Expert Rev Clin Pharmacol. 2012;5:187–97. 22. Dalleur O, Spinewine A, Henrard S, Losseau C, Speybroeck N, Boland B. Inappropriate prescribing and related hospital admissions in frail older persons according to the STOPP and START criteria. Drugs Aging. 2012;29:829–37. 23. Hill-Taylor B, Sketris I, Hayden J, Byrne S, O’Sullivan D, Christie R. 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. J Clin Pharm Ther. 2013;38:360–72.
