Economic outcomes of pharmacist-physician medication therapy management for polypharmacy elderly: A prospective, randomized, controlled trial

+

MODEL

Journal of the Formosan Medical Association (2017) xx, 1e9

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.jfma-online.com

Original Article

Economic outcomes of pharmacist-physician medication therapy management for polypharmacy elderly: A prospective, randomized, controlled trial Hsiang-Wen Lin a,b,c,*, Chih-Hsueh Lin d, Chin-Kai Chang e, Che-Yi Chou f, I-Wen Yu g, Cheng-Chieh Lin d,h, Tsai-Chung Li i, Chia-Ing Li j, Yow-Wen Hsieh a,b a School of Pharmacy and Graduate Institute, College of Pharmacy, China Medical University, Taichung, Taiwan b Department of Pharmacy, China Medical University Hospital, Taichung, Taiwan c Department of Pharmacy Systems, Outcomes & Policy, College of Pharmacy, University of Illinois at Chicago, Chicago, IL, USA d Department of Family Medicine, China Medical University Hospital, Taichung, Taiwan e Department of Rehabilitation, China Medical University Hospital, Taichung, Taiwan f Department of Internal Medicine, China Medical University Hospital, Taichung, Taiwan g Supra Integration and Incubation Center, Taipei, Taiwan h School of Medicine, College of Medicine, China Medical University, Taichung, Taiwan i Graduate Institute of Biostatistics, China Medical University, Taichung, Taiwan j Department of Medical Research, China Medical University Hospital, Taichung, Taiwan

Received 24 October 2014; received in revised form 24 April 2017; accepted 25 April 2017

KEYWORDS Aged; Economics; Medical; Medication therapy management; Pharmacists; Polypharmacy; Taiwan

Background/purpose: With an increasing geriatric population, the need for effective management of chronic conditions and medication use in the elderly is growing. Medication use in the elderly presents significant challenges due to changes in pharmacodynamic and pharmacokinetic profiles. We aimed to examine the impact of a collaborative physician-pharmacist medication therapy management (MTM) program for polypharmacy elderly patients. Methods: Elderly patients with multiple chronic conditions on polypharmacy were enrolled in this prospective, randomized, and controlled study over 16 months of implementation. The intervention group consisted of patients randomized to a collaborative pharmacist-physician MTM program. They were monitored continuously by a clinical pharmacist, while patients in the control group received only usual care with follow-up assessment. Primary outcome was

* Corresponding author. School of Pharmacy and Graduate Institute, College of Pharmacy, China Medical University (CMU), No. 91, HsuehShih Road, Taichung 40402, Taiwan E-mail address: [email protected] (H.-W. Lin). http://dx.doi.org/10.1016/j.jfma.2017.04.017 0929-6646/Copyright ª 2017, Formosan Medical Association. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Please cite this article in press as: Lin H-W, et al., Economic outcomes of pharmacist-physician medication therapy management for polypharmacy elderly: A prospective, randomized, controlled trial, Journal of the Formosan Medical Association (2017), http:// dx.doi.org/10.1016/j.jfma.2017.04.017

+

MODEL

2

H.-W. Lin et al. economic differences, measured in total medical expenditure. Secondary outcomes of clinical and humanistic effects were compared between the two groups. Results: The total number of enrolled patients was 87 and 91 in the MTM and usual groups, respectively. The difference-in-difference estimate on medical expenditure during the 16month implementation period was $3,758,373 New Taiwan Dollars ($127,015 US Dollars) less than the usually care group. Impact was also seen in humanistic outcomes while lipid profiles and mortality trended toward improvement. Conclusion: The pharmacist-physician collaborative MTM program for polypharmacy elderly had significant cost savings and improvement in humanistic measures, demonstrating the importance of clinical pharmacists and MTM programs for elderly patients in Taiwan. The results suggest the possibility of clinical benefits, but the study was not substantially powered to find a statistical difference. Copyright ª 2017, Formosan Medical Association. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/bync-nd/4.0/).

Introduction With a globally increasing geriatric population and rising costs of medication, the need for effective management of chronic conditions and appropriate medication use in the elderly is growing.1e3 The benefits of providing comprehensive patient care in the elderly via different approaches, especially through the coordinated efforts of an interdisciplinary team, have been demonstrated.4 “Polypharmacy” refers to the use of multiple medications by a patient, and has been defined in several ways.5 Some definitions use a numerical value and others use a consideration of quality.6 Examination of polypharmacy should assess the effectiveness of medication use and appropriateness of the prescription.7 Polypharmacy in the elderly is associated with multiple chronic diseases, more frequent contact with their prescribers, and increased risks of mortality.8e10 Health care providers must attentively survey medication use for the likelihood of medicationrelated problems (MRPs) and try to resolve these MRPs, especially in the elderly.11,12 It is helpful to engage pharmacists in comprehensive medication reviews to identify and craft appropriate solutions for MRPs, due to the limited time available to physicians in the outpatient settings.13 Physicians in Taiwan may see hundreds of patients in a single day of clinic and are less likely to spend time with their patients as physicians in the United States.14 There remain many challenges to implementing pharmacist cognitive services, including a lack of pharmacists’ time and a lack of recognition of ability by either the patients or the prescribing physicians.14 Appropriate interventions by pharmacists can reduce the complexity of a given medication regimen, improve patient compliance and quality of life, as well as provide a further reduction in morbidity, mortality, and/or health care costs for the elderly.4,15e17 Therefore, developing and establishing physicianpharmacist collaboration models to solve elderly patient’s MRPs in clinical practice settings is highly valuable.11 Elderly patients have benefited from medication therapy management (MTM) services in the United States.14,18,19 The core elements of MTM service include “conducting

medication therapy reviews,” “establishing personal medication records,” “planning medication-related actions,” and “providing intervention, referral, documentation and/or follow-up”.20 Little is known about the impact of MTM services on elderly outpatient populations outside of the United States due to variances in pharmacy practice environments, health care systems, and reimbursements from country to country.21 Although more than 92% of all health care facilities in Taiwan have been contracted with the National Health Insurance Administration (NHIA) since 2008, there is currently no official MTM program implemented in health care settings in Taiwan.22

Aim of the study A collaborative physician-pharmacist MTM service was established in an academic medical center in Taiwan with the goal of improving the quality of care for geriatric polypharmacy patients. This study aims to examine the economic and other outcomes of this collaborative MTM service.

Methods Setting This study was conducted at the outpatient clinics of China Medical University Hospital (CMUH), a university teaching hospital in Taichung, Taiwan. CMUH has more than 2000 beds and approximately 5000 outpatient visits per day. The average number of prescription items per outpatient at CMUH was 7.39 in 2011 compared to an average 5.05 throughout Taiwan.22 Most pharmacists in the outpatient pharmacy at CMUH perform the tasks of medication preparation, dispensing, and distribution. One full-time clinical pharmacist was trained to provide direct patient care services. The pharmacist implemented this collaborative physician-pharmacist MTM program in a stand-alone “Pharmaceutical Care Clinic”, which was located close to the other physician clinics in the outpatient units.

Please cite this article in press as: Lin H-W, et al., Economic outcomes of pharmacist-physician medication therapy management for polypharmacy elderly: A prospective, randomized, controlled trial, Journal of the Formosan Medical Association (2017), http:// dx.doi.org/10.1016/j.jfma.2017.04.017

+

MODEL

Outcomes of pharmacist-physician MTM Taiwan

Patients Elderly patients age 65 and older who had three or more chronic diseases (identified by diagnosed ICD9 codes), more than six prescription items, and had made more than four outpatients visits or visited two or more different specialties in CMUH during an assessment period from November 2007 to October 2008 were eligible for recruitment. Those with the following characteristics were excluded: cancer patients, organ transplant recipients, patients suffering from (a) psychological problem(s), and those who were unable to participate in the assessments of humanistic outcomes.

Study design A collaborative physician-pharmacist MTM team was created at CMUH. This prospective, randomized, and controlled intervention study was designed to examine the economic and other outcomes (i.e., clinical and humanistic effects) of such collaborative MTM service. A sample size of 350 patients per group was calculated as required to have at least 80% power to detect a 12% reduction on total medical expenditure at a two-sided 0.05 significance level and that at least 80 patients per group would be needed to save at least $100,000 TWD.30 Such savings would relate to the reduction of $113,250 TWD for all hospitalization expenses in a similar study conducted in India.25 Consequently, due to a limitation in human resources and time, 80 patients were expected to be recruited into either the intervention and control groups. After informed written consent was obtained from each participant, the participants were randomly assigned to either the MTM intervention group or the usual care (UC) control group using nontransparent and ordered envelopes that were generated using block randomization and were prepared before recruitment. Each block included four assignments in a random order, including two intervention subjects and two control subjects. A total of 50 blocks were planned to reach the required number of subjects. The prescribers, dispensing pharmacists in outpatient pharmacies, and researchers were blinded to the initial randomized assignments, but the research assistants and the responsible clinical pharmacist were not. The institutional ethics committee at CMUH formally approved this study protocol in advance and continuously reviewed throughout the whole study implementation (DMR97-IRB-152).

Intervention development and implementation This study was implemented over a 16 month period from November 2009 to March 2011, which included a 4 month enrollment period. Patients were followed for 12 months after enrollment. Comprehensive intervention elements were provided by the clinical pharmacist for patients in the MTM group, either through direct patient contact, physician contact, or comprehensive medical chart review. The interventions reflected the published literature on MTM service and focused on potentially inappropriate medications in the elderly.19,23e25 This included identifying MRPs, such as medication duplications, drug interactions, dosing for

3 renal and liver impairment, suspected adverse drug reaction (ADR), therapeutic drug monitoring, unnecessary medications, and inappropriate nonpharmacological managements. The clinical pharmacist in MTM service also provided face-to-face and telephone counseling to patients on health education and medication adherence. Cases with MRPs or care-related concerns were presented at bi-weekly meetings of the physician-pharmacist MTM team if an appropriate intervention could not be determined by the clinical pharmacist alone or a complex patient needed team review. The physician-pharmacist MTM team included two geriatricians, one cardiologist, one nephrologist, and one clinical pharmacist supervisor, in addition to the study’s responsible clinical pharmacist. The team attempted to propose resolutions for any difficult issues which were raised by the clinical pharmacist during the one-hour case discussion meeting. After either the clinical pharmacist alone or physicianpharmacist MTM team made a decision on the appropriate action to be taken, the clinical pharmacist would contact the patient’s prescribing physician before the next appointment or contact patients directly through in-person or telephone interactions. The prescribing physician was encouraged to follow the suggestions made by the clinical pharmacist or MTM team. However, the ultimate responsibility for the patients’ prescription remained with the prescribing physician. Such patient-centered discussions between the clinical pharmacist and each prescribing physician usually took less than 5 min for one discussion. Patients enrolled in the UC group did not receive these aforementioned interventions but may have received conventional services from the prescribing physicians, other pharmacists, and/or other health care providers. Elderly patients in either MTM group or UC group would receive at least one physician for their disease managements as traditional service. For instance, some physicians might aggressively provide patients with more specific recommendations of diet control or prescribe statin medications for primary or secondary preventions of strokes, coronary artery diseases or sudden death following the treatment guidelines in either groups, regardless involving pharmacists’ commends or not. Patients in both groups received follow-up interviews about their self-report economic, clinical and humanistic outcomes in the first, third, sixth and twelfth months, conducted by trained research assistants. All enrolled patients in both groups were offered small gifts after completing each follow-up assessment. A brochure about safe medication use for the elderly was distributed to all patients to avoid educational bias.

Outcome measures and data collection The ECHO (Economic, Clinical, and Humanistic Outcomes) model, which is commonly used to assess clinical practice improvement, was applied to evaluate the study intervention effectiveness.26 The primary endpoint was economic, measured as difference-in-difference (DID) estimate of total medical expenditures for the entire 16 month implementation period consumed in the outpatient departments, emergency rooms (ERs), and inpatient departments in

Please cite this article in press as: Lin H-W, et al., Economic outcomes of pharmacist-physician medication therapy management for polypharmacy elderly: A prospective, randomized, controlled trial, Journal of the Formosan Medical Association (2017), http:// dx.doi.org/10.1016/j.jfma.2017.04.017

+

MODEL

4 CMUH. The total expenditure was compared within each group, using the 6-month period prior to study implementation as the control. The DID model was chosen by the study designers because the sample was not randomized on economic measures, and differences in baseline spending between the two groups were likely to be present. The DID model allows for a program-level measure of change against each group’s own baseline for the program’s entire 16-month implementation. Economic differences on the patient level was measured in 12-month total medical expenses, as each patient was enrolled in the program for this time period. The clinical and humanistic measures were the secondary endpoints and compared on the recruitment date (baseline) and in the twelfth month. The closest available observations from the end of follow-up were used for patients that were lost to follow up. The clinical measures included the common laboratory data used to evaluate the progression of general chronic diseases for the elderly, such as lipid profiles (low-density lipoprotein [LDL], high-density lipoprotein [HDL]), Hemoglobin A1c (HbA1c), kidney function (urinary albumin: creatinine ratio [ACR], serum creatinine [CRE], glomerular filtration rate [GFR]), and functional status (Activity of Daily Living, using Barthel index and the instrumental activities of daily living [IADL], and Geriatric Depression Scale [GDS] short form for depression tendency). The total score of the Barthel index is 100, and higher scores represent better daily function. Total score of IADL is 24, and higher scores represent better instrumental daily activity. Positive scores above 5 on the GDS Short Form represent the necessity to conduct an indepth psychological assessment to differentiate the disease of major depression. We utilized the general health-related quality of life measures, Euroqol Quality of Life Scale for 5 domains (EQ5D) and three levels, Euroqol Quality of Life Visual Analogue Scale (EQ-5D VAS), and EQ-5D Index for humanistic outcomes.27 In addition to using the 5 domains scale and VAS to assess the patients’ health status, we adapted Japanese algorithm to calculate the EQ index accordingly. There is no official Taiwanese algorithm available for EQ index and the Japanese algorithm was found to be more preferable than the other Asian options.28,29 Changes in clinical and humanistic effects were compared to the extent of minimal clinical importance differences (MCID).

Data analysis Categorical variables at baseline and at 12 months were compared using chi-square or Fisher exact tests. The continuous variables were compared using students’ t test or Wilcoxon rank sum test. We mainly performed intention-totreat analysis for the economic measures. A per-protocol analysis was used for secondary clinical and humanistic outcomes measures. The effect size, in terms of standardized mean difference changes, for the clinical and humanistic outcomes was examined. A standardized mean difference of zero represents no difference, and increasing distance from zero shows either increasing or decreasing effect. In each test, statistical significance was considered to be an alpha of
0.05. Statistical analyses were performed using SPSS statistical Software 16.0 (SPSS Inc., Chicago, IL.).

H.-W. Lin et al.

Results Patient characteristics Of the 21,072 elderly patients who had ever visited CMUH during the assessment period of November 2007 to October 2008, 6155 patients were identified as eligible polypharmacy elderly patients (Figure 1). In total, 178 elderly patients, 87 in MTM group and 91 in the UC group were enrolled during November 2009 to March 2010. There were no statistically significant differences in demographic, clinical, or humanistic variables between the two groups at baseline and remained so after any loss to follow up (Table 1). Prior medical history by disease varied only by the proportion with diagnosed diabetes mellitus (46% vs 25%, p Z 0.004). However, the two groups had similar HbAlc. Their average Barthel index and IADL scores were about 93 and 19, respectively, which represent mild dependency. Patients in both groups rated the pain/ discomfort domain with problems higher than the other domains in the EQ-5D.

Economic, clinical and humanistic effects The difference between groups for baseline medical expenditure (i.e., MTM intervention group vs UC control group) in the 6 months prior to program implementation was þ$574,559 TWD. The difference between groups in medical expenditure during the 16-month implementation period was -$3,183,814 TWD. The DID estimate on medical expenditure among the MTM group was $3,758,373 TWD less than that in the UC group (Table 2). Patients in the MTM group had 12 month total expenditures of $2,594,405 TWD less than the patients in the UC group (Table 3). This represents a savings of $29,821 TWD per patient per year. Inpatient costs accounted for 63.2% of the total savings. The exchange rate of TWD to USD was 29.59 to 1 on March 31, 2011. There were statistically significant changes in CRE, GFR, GDS, Barthel index, IADL, EQ index, and EQ-VAS (see Table 3). The differences in LDL, HDL, ACR, GFR, GDS, IADL and EQ index were clinically positive or less-negative in the MTM group. The non-significant differences in HbA1C, CRE, Barthel index and EQ-VAS in the MTM group moved toward the negative directions. None of these values reached the MCID, except for a decrease of 4.09 points in the Barthel index and an increase of 0.216 points of EQ-index in the MTM group. The effect on mortality had an absolute difference of six (p value 0.06), although the study was not designed to examine this effect.

Discussion This study demonstrates that a collaborative physicianpharmacist MTM program targeting polypharmacy geriatric patients was truly cost saving. Non-clinically significant benefits on certain clinical and humanistic outcomes were also shown. The benefit-to-cost ratio for the MTM service program was 3.53:1, equal to DID program-level savings of $3,183,814 TWD vs. $900,000 TWD in clinical pharmacist

Please cite this article in press as: Lin H-W, et al., Economic outcomes of pharmacist-physician medication therapy management for polypharmacy elderly: A prospective, randomized, controlled trial, Journal of the Formosan Medical Association (2017), http:// dx.doi.org/10.1016/j.jfma.2017.04.017

+

MODEL

Outcomes of pharmacist-physician MTM Taiwan

5

Figure 1 Study design and patient enrollment flow chart. Abbreviations: CMUH Z China Medical University Hospital; PPMTM Z pharmacist-physician medication therapy management; ECHO Z economic, clinical and humanistic outcomes.

salary for 16-month implementation period. The benefit-tocost ratio for 12-month patient-level data was 3.42:1, calculated from a total expenditure reduction of $2,594,405 TWD vs. $759,375 TWD in clinical pharmacist salary during 12-month period. These results are slightly less than the findings obtained from the previous studies

that demonstrated savings of 4.0e4.81 for every dollar spent on clinical pharmacy and MTM services in the United States.33,34 Our study used a collaborative model and involved physicians, whose time was not included in the cost evaluation. Nevertheless, the impressive fact that 63.2% of the yearly savings demonstrated in this study were

Please cite this article in press as: Lin H-W, et al., Economic outcomes of pharmacist-physician medication therapy management for polypharmacy elderly: A prospective, randomized, controlled trial, Journal of the Formosan Medical Association (2017), http:// dx.doi.org/10.1016/j.jfma.2017.04.017

+

MODEL

6

H.-W. Lin et al. Table 1

Demographic, clinical and humanistic variables among enrollees at baseline. Usual care group (n Z 91)

Variables/Group

Age, y (mean  SD) 78.4  6.0 Male,% 64.8 With disease status With DM, % 25.3 With HTN,% 23.1 With Hyperlipidemia,% 3.3 With CVA,% 22 With IHD,% 31.9 With hepatic diseases,% 1.1 With renal diseases,% 6.6 With lung diseases,% 9.9 With cancer history,% 7.7 Clinical measures at baseline (n Z retrievable sample size for MTM/UC groups) LDL mg/dl (n Z 32/43) 94.28  22.86 HDL mg/dl (n Z 29/41) 42.18  13.91 HbAlc % (n Z 33/50) 7.47  1.58 ACR mg/mg (n Z 7/14) 157.61  345.56 CRE mg/dL (n Z 54/73) 1.13  1.04 60.25  26.63 GFR mL/min/1.73 m2 (n Z 75/70) GDS 3.92  3.55 75.6% GDS
5, %)a 93  15.5 Barthel index total scoreb 18  7.0 IADLc Humanistic measures at baseline EQ-5D domains (% respondents with any problem)d EQ_ Mobility 24.2% EQ_ Self care 14.3% EQ_ Usual activities 23.1% EQ_ Pain/discomfort 35.2% EQ Anxiety/Depression 14.4% EQ index (Japan) 0.819  0.17 EQ-5D VAS 66  14

MTM group (n Z 87)

P-Value

77.9  6.1 58.6

0.582 0.394

46 28.7 3.5 18.4 24.1 0 6.9 12.6 4.6

0.004 0.489 0.956 0.551 0.251 0.327 0.933 0.561 0.391

98.94  28.53 39.4  13.6 7.25  1.10 441.59  971.48 1.48  1.71 59.34  23.83 3.53  3.45 79% 93  16.7 19  6.0

0.450 0.407 0.456 0.467 0.185 0.829 0.459 0.578 1.000 0.309

22.5% 12.4% 18.0% 41.6% 18.0% 0.833  0.18 65  15

0.852 0.748 0.888 0.394 0.458 0.594 0.646

Abbreviations: ACR: urinary albumin: creatinine ratio; CVA: cerebral vascular accident; CRE: creatinine; DM: diabetes mellitus; EQ-5D: Euroqol Quality of Life Scale for 5 domains; EQ-5D VAS: Euroqol Quality of Life Scale for 5 domains visual analogue scale; GDS: Geriatric Depression Scale; GFR: glomerular filtration rate; HbA1c: hemoglobin A1c; HDL: high-density lipoprotein; HTN: hypertension; IADL: Instrumental activities of daily living; IHD: ischemic heart disease; LDL: low-density lipoprotein. a GDS score
5 infers no depression tendency. b Total score of Barthel index is 100 and a higher score represents better daily function. c Total score of IADL is 24 and a higher score represents better instrumental daily activity. d Using per-protocol analysis, EQ index (0e1) and EQ-5D VAS (0e100): a higher score represents better health status.

Table 2 Difference-in-difference estimate of the MTM program implementation effect on total medical expenditure in China Medical University Hospital. TWD

MTM group Usual care group Difference between groups

Change in total medical expenditure in CMUH 6 months before implementation

16 months of implementation

4,704,656 4,130,097 574,559

9,103,164 12,286,978 3,183,814

Difference between periods

4,398,508 8,156,881 3,758,373

Abbreviations: CMUH: China Medical University Hospital; MTM Z medication therapy management; TWD: Taiwan New Dollar.

inpatient savings. It suggests that the MTM program played an important role in preventing hospitalizations. Although the recruited patients might encounter the different severity of health statues with same or different diseases, there were no statistically significant differences in

demographic, clinical, or humanistic variables between the two groups at baseline, except the prior medical history of diabetes mellitus. Accordingly, it was assumed that the clinical conditions were comparable between the two groups in the enrollment period. After one year intervention and

Please cite this article in press as: Lin H-W, et al., Economic outcomes of pharmacist-physician medication therapy management for polypharmacy elderly: A prospective, randomized, controlled trial, Journal of the Formosan Medical Association (2017), http:// dx.doi.org/10.1016/j.jfma.2017.04.017

+

MODEL

Outcomes of pharmacist-physician MTM Taiwan Table 3

7

Economic, clinical, and humanistic outcome measures after 1-year intervention or last follow-up.

Variables/Group

Usual care group (n Z 91)

MTM group (n Z 87)

Mortality, n (%) 8 (8.8%) 2 (2.3%) Censored sample, n (%) 83 (91.2%) 85 (97.7%) Economic outcomes (TWD) of groupa Total OPD expenditure 4,757,752 4,207,388 Total ER expenditure 823,714 418,767 Total IPD expenditure 4,786,177 3,147,083 Total medical expenditure 10,367,643 7,773,238 Changes from baseline in clinical and humanistic outcomes Clinical outcomes (retrievable sample size, MTM/UC groups)b,c LDL [mg/dL] (n Z 32/43) 7.64  33.47 3.92  40.59 HDL [mg/dL] (n Z 29/41) 2.36  7.67 4.33  10.26 HbAlc [%] (n Z 33/50) 0.05  0.97 0.14  1.06 ACR (n Z 7/14) 43.04  72.77 159.76  578.71 CRE [mg/dL] (n Z 54/73) 0.03  1.40 0.16  1.03 GFR [mL/min] (n Z 75/70) 3.41  12.62 1.16  12.16 GDS (n Z 42/60) 0.63  2.79 0.98  3.40 GDS (5, %) (n Z 42/60) 2 (5%) 6 (10%) Barthel index (n Z 62/44) 1.94  6.55 4.09  13.95 IADL (n Z 60/44) 1.57  4.09 1.25  4.16 Humanistic outcomes (retrievable sample size, MTM/ UC groups)b,c EQ index (n Z 62/45) 0.01  0.18 0.216  0.16 EQ-VAS (n Z 53/38) 4.98  13.98 2.10  11.95

Difference/ effect size (standardized mean difference)b

P valuec

6 þ2

0.060

550,364 404,947 1,639,094 2,594,405

0.3064 0.2123 0.0878 0.4221 0.1083 0.1814 0.1106 4 0.2092 0.0777

0.0551 0.0592 0.0503 0.2185 0.0323 0.0277 0.0405 0.333 0.0391 0.0394

1.3147 0.2244

0.0464 0.0455

Abbreviations: ACR: urinary albumin: creatinine ratio; CRE: creatinine; ER: Emergency Room; EQ-5D: Euroqol Quality of Life Scale for 5 domains; EQ-5D VAS: Euroqol Quality of Life Scale for 5 domains visual analogue scale; GDS: Geriatric Depression Scale; GFR: glomerular filtration rate; HbA1c: hemoglobin A1c; HDL: high-density lipoprotein; IADL: Instrumental activities of daily living; IPD: Inpatient Department; LDL: low-density lipoprotein; OPD: Outpatient Department; MTM:medication therapy management; TWD: Taiwan New Dollar. a Economic outcomes compared the patient-level expenditure difference for 12 months enrollment. b Effect size (standardized mean difference of change) for those who have two points within the 1-year enrollment and implementation: (MTM group e UC group). c P value e Using per-protocol and para- & nonparametric (e.g., t test, ManneWhitney U test) analyses for the difference or effect size (standardized mean difference of change).

follow-up, the improvements in several clinical markers among the patients in the MTM group were mostly positive, but non-significant. Although CRE was significantly higher, interestingly GFR was also significantly higher in MTM group than in the UC group. The reduction of ACR, which was close to 160 mg/ml, in MTM group may reflect a better maintenance of kidney function for diabetic elderly patients. MTM patients encountered a non-significant reduction in LDL and rise in HDL (p values 0.0551 and 0.0591, respectively). Although there was a negative impact on HbA1C, the effect size was not significant, and the incremental HbA1C did not reach the MCID of 0.5%.35 A statistically significant decline of 4.09 points in the Barthel Index in the MTM group also reached the MCID, set at 2-points.36 There is currently no available algorithm for EQ index value in Taiwan, which lead the study designers to use the Japanese algorithm and take Thai and US patients’ EQ index values as comparisons.28 The mean EQ-index score for MTM group and UC group at baseline were all higher than that of Diabetic Thai patients using the Japanese algorithm (about 0.82 vs. 0.75), but the mean EQ VAS score was comparable to Thai peritoneal dialysis patients (65  0.26).28,29 Importantly, the average change in EQ-

index of 0.216 for the MTM group was greater than the estimated MCID of 0.033e0.074 as per previous research in the Medical Expenditure Panel Survey (MEPS) in the US, signifying improved humanistic outcomes.37 While the NHI covers more than 99% of the population in Taiwan, the elderly patients, who make up approximately 8% of the total population, account for nearly one-quarter of the total medical expenditure.31,32 The NHIA in Taiwan has made a concerted effort to constrain medical costs. The MTM service provided for elderly outpatients in this study is in alignment with these efforts. However, expansion of such collaborative physician-pharmacist MTM services requires the support of clinical practitioners, hospital administrations, the NHIA, as well as political support.21 Another 2006 study conducted for six months in an academic medical center in Taiwan also showed similar positive outcomes in solving medication related problems for the elderly.38 In that study, there were two part-time clinical pharmacists involved in a “Medication Safety Review Clinic” to help solve MPRs for patients, whereas this study used a full-time clinical pharmacist involved in the collaborative MTM service to provide pharmaceutical care and solve MRPs.

Please cite this article in press as: Lin H-W, et al., Economic outcomes of pharmacist-physician medication therapy management for polypharmacy elderly: A prospective, randomized, controlled trial, Journal of the Formosan Medical Association (2017), http:// dx.doi.org/10.1016/j.jfma.2017.04.017

+

MODEL

8

Limitations There are a number of limitations of this study. First, the incorporation of the ECHO model into MTM service may not be feasible for other researchers due to differences in pharmacist manpower, training, resources, and reimbursement. These factors may differ in other institutes and countries, which limits the external validity of this study. This study examined the only one well-trained clinical pharmacist involved in providing collaborative MTM services in only one hospital. Second, this study used a collaborative model and involved physicians, whose time was not included in the cost evaluation. We assumed physicians’ effect might be limited because the MTM service, with 1-h bi-weekly meetings for the physician-pharmacist MTM team, and the short discussions with the clinical pharmacist usually took less than 5 min for one discussion. All these events consumed a relatively small portion of time among involved physicians. In this case, the time of individual physician’s care toward each patient in either MTM intervention or UC control group were presumed to be same. Third, medical expenditures consumed outside of CMUH were not considered in the primary endpoint of economic assessment in this study. However, patients’ self-reports were used and there was no difference between the two groups’ outside expenses. Patients in the UC group selfreported $3066 TWD per patient for the out-of-pocket expenses in emergency rooms and hospital admissions, outpatient clinics, folk remedies, over-the counter medications, and dietary supplements within one year. Patients in the MTM group reported $3157 TWD per patient. In Taiwan, there are almost no procedural or economic restrictions on patients’ visiting multiple specialties across clinics in one medical institute or across various medical settings up to 2016. Merely 3.5% of all NHI beneficiaries in Taiwan visit only a single hospital for treatment. However, due to limitations in technology and availability of outside hospital electronic health records, this study made the assumption that enrolled patients were not using other hospitals. This patient behavior is equal across Taiwan and does not influence the generalizability of these results inside of Taiwan. However, differing patient habits in other countries may affect the generalizability of these results. Some bias or cross-over effect may exist between those patients in the UC group and those in the MTM group. Physicians at this hospital were not informed of patients’ group assignments. However, they were not procedurally blinded and may have taken care of patients for both groups. The clinical pharmacist’s interventions for MRPs toward the intervention group may have “coached” a change in physician attitude or practice. Although the principle investigator has made some efforts to audit the possible spill-over effects, the possibility of intervention bias is unavoidable. Last but not the least, the common laboratory values for the chronic diseases (e.g., CRE, GFR for kidney disease, GDS for depression, Barthel index for dementia, LDL or HDL for hyperlipidemia) were used as the surrogate indicators for disease severity in this study. This study was not powerful enough to detect significant differences in clinical and humanistic outcomes using intention-to-treat analysis.

H.-W. Lin et al. It could because the sample sizes used to evaluate the clinical effects were smaller due to the lack of available laboratory data for all patients, or no actual differences were observed because the laboratory data might not be comparable to their clinical conditions. Therefore, a perprotocol analysis was performed for the clinical and humanistic effects in this study. Further studies which are powered to find a difference may demonstrate risk reduction on morbidity and mortality.35

Conclusion This study demonstrates significant cost savings in a collaborative MTM service at an academic medical center in Taiwan. The use of a clinical pharmacist in a physicianpharmacist MTM program facilitated a high quality of care and had a positive impact on economic effect and some clinical and humanistic outcomes. The expansion of the MTM services to a larger population of vulnerable patients should be pursued and evaluated. This study reveals that the impacts of interdisciplinary care that have been shown in studies in the United States are also possible in Taiwan. Ideally, such a profound benefit-to-cost ratio of more than 3:1 should raise Taiwan’s NHIA’s awareness. Additional research is needed to examine to effects on clinical and humanistic outcomes. National Health Insurance in Taiwan and in other countries could consider funds to provide collaborative MTM and pharmaceutical care services with clinical pharmacists to save more medical expenses.

Funding This study was fully supported by Department of Health, Executive Yuan, Taiwan, R.O.C. (DOH098-TD-D-113-098011), and partially supported by the National Science Council (NSC 99-2320-B-039-031-MY3, NSC 102-2320-B-039-007), by China Medical University Hospital (DMR-99-140), and by Taiwan National Health Research Institutes (NHRI-EX105-10318PC; NHRI-EX106-10318P). The funding agencies had no role in the study implementation, analysis or interpretation of data, or preparation, review or approval of the manuscript.

Ethical approval The institutional ethics committee at China Medical University Hospital formally approved this study protocol in advance and continuously reviewed throughout the whole study implementation (DMR97-IRB-152).

Conflicts of interest The authors have declared that no competing interests exist.

Acknowledgements Dr. Cory Simonavice, PharmD, had an integral role and the revision and drafting of this manuscript. The authors would

Please cite this article in press as: Lin H-W, et al., Economic outcomes of pharmacist-physician medication therapy management for polypharmacy elderly: A prospective, randomized, controlled trial, Journal of the Formosan Medical Association (2017), http:// dx.doi.org/10.1016/j.jfma.2017.04.017

+

MODEL

Outcomes of pharmacist-physician MTM Taiwan also like to express their gratitude to the study participants, research assistants, prescribing physicians for their contribution of time, and Jenna Beall, Shih-Ying (Audrey) Hsu, Dan Lu, and Matthias C. Lu, and Dr. Chia-Hung Chou for their study assistance, comments and review of the manuscript.

References 1. National Institute on Aging, National Institutes of Health. U.S. Department of Health and Human services. Why population aging matters e a global perspective (NIH publication No. 07e6134). 2007. Available from: http://www.nia.nih.gov/ research/publication/why-population-aging-matters-globalperspective [Accessed 23 June 2012]. 2. Wolff JL, Starfield B, Anderson G. Prevalence, expenditures, and complications of multiple chronic conditions in the elderly. Arch Intern Med 2002;162:2269e76. 3. Budnitz DS, Lovegrove MC, Shehab N, et al. Emergency hospitalizations for adverse drug events in older Americans. N Engl J Med 2011;365:2002e12. 4. Boult C, Green AF, Boult LB, et al. Successful models of comprehensive care for older adults with chronic conditions: evidence for the Institute of Medicine’s “retooling for an aging America” report. J Am Geriatr Soc 2009;57:2328e37. 5. Fulton MM, Allen ER. Polypharmacy in the elderly: a literature review. J Am Acad Nurse Pract 2005;17:123e32. 6. Bushardt RL, Massey EB, Simpson TW, Arial JC, Simpson KN. Polypharmacy: misleading, but manageable. Clin Interv Aging 2008;3(2):383e9. 7. Kuijpers MA, van Marum RJ, Egberts AC, et al. Relationship between polypharmacy and underprescribing. Br J Clin Pharmacol 2008;65:130e3. 8. Preskorn SH, Silkey B, Shah R, et al. Complexity of medication use in the Veterans Affairs healthcare system: part I. Outpatient use in relation to age and number of prescribers. J Psychiatr Pract 2005;11:5e15. 9. Tsuji-Hayashi Y, Fukuhara S, Green J, et al. Use of prescribed drugs among older people in Japan: association with not having a regular physician. J Am Geriatr Soc 1999;47:1425e9. 10. Jyrkka J, Enlund H, Korhonen MJ, et al. Polypharmacy status as an indicator of mortality in an elderly population. Drugs Aging 2009;26:1039e48. 11. Simonson W, Feinberg JL. Medication-related problems in the elderly : defining the issues and identifying solutions. Drugs Aging 2005;22:559e69. 12. Steinman MA, Hanlon JT. Managing medications in clinically complex elders: “There’s got to be a happy medium”. JAMA 2010;304:1592e601. 13. Cutler DM, Everett W. Thinking outside the pillboxemedication adherence as a priority for health care reform. N Engl J Med 2010;362:1553e5. 14. Nkansah NT, Brewer JM, Connors R, et al. Clinical outcomes of patients with diabetes mellitus receiving medication management by pharmacists in an urban private physician practice. Am J Health Syst Pharm 2008;65:145e9. 15. Elliott RA. Reducing medication regimen complexity for older patients prior to discharge from hospital: feasibility and barriers. J Clin Pharm Ther 2012;37(6):637e42. 16. Reilly T, Barile D, Reuben S. Role of the pharmacist on a general medicine acute care for the elderly unit. Am J Geriatr Pharmacother 2012;10:95e100. 17. Midlo ¨v P, Bahrani L, Seyfali M, et al. The effect of medication reconciliation in elderly patients at hospital discharge. Int J Clin Pharm 2012;34:113e9. 18. Aspinall S, Sevick MA, Donohue J, et al. Medication errors in older adults: a review of recent publications. Am J Geriatr Pharmacother 2007;5:75e84.

9 19. Isetts BJ, Schondelmeyer SW, Artz MB, et al. Clinical and economic outcomes of medication therapy management services: the Minnesota experience. J Am Pharm Assoc 2003; 2008(48):203e11. 20. American Pharmacists Assocaition, National Association of Chain Drug Stores Foundation. Medication therapy management in pharmacy practice: core elements of an MTM service model. Version 2.0 March 2008. J Am Pharm Assoc 2003; 2008(48):341e53. 21. Farris KB, Fernandez-Llimos F, Benrimoj SI. Pharmaceutical care in community pharmacies: practice and research from around the world. Ann Pharmacother 2005;39:1539e41. 22. Bureau of National Health Insurance DOH, Executive Yuan. National Health Insurance in Taiwan; 2011. Available from: http://www.nhi.gov.tw/english/index.aspx [Accessed 2 June 2012]. 23. American Pharmacist Association. Medication therapy management service: a critical review. Final Report. American Pharmacist Association; 2005. Available from: http://www.lewin. com/publications/publication/148/ [Accessed 23 June 2012]. 24. Stefanacci RG, Cavallaro E, Beers MH, et al. Developing explicit positive beers criteria for preferred central nervous system medications in older adults. Consult Pharm 2009;24: 601e10. 25. Malhotra S, Karan RS, Pandhi P, et al. Drug related medical emergencies in the elderly: role of adverse drug reactions and non-compliance. Postgrad Med J 2001;77:703e7. 26. Gunter MJ. The role of the ECHO model in outcomes research and clinical practice improvement. Am J Manag Care 1999;5(4 Suppl):S217e24. 27. EuroQol Group. What is EQ-5D? Available from: http://www. euroqol.org/eq-5d/what-is-eq-5d.html [Accessed 1 September 2009]. 28. Sakthong P, Charoenvisuthiwongs R, Shabunthom R. A comparison of EQ-5D index scores using the UK, US, and Japan preference weights in a Thai sample with type 2 diabetes. Health Qual Life Outcomes 2008;6:71. 29. Sakthong P, Kasemsup V. Health utility measured with EQ-5D in Thai patients undergoing peritoneal dialysis. Value Health 2012;15:S79e84. 30. Torgerson DJ, Campbell MK. Economics notes: cost effectiveness calculations and sample size. BMJ 2000;321:697. 31. Bureau of National Health Insurance, Executive Yuan. Statistics & Surveys. Available from: http://www.nhi.gov.tw/English/ webdata/webdata.aspx?menuZ11&menu_idZ296&webdata_ idZ1942&WD_IDZ296 [Accessed 2 June 2012]. 32. Chen LW, Yip W, Chang MC, et al. The effects of Taiwan’s national health insurance on access and health status of the elderly. Health Econ 2007;16:223e42. 33. Chisholm MA, Vollenweider LJ, Mulloy LL, et al. Cost-benefit analysis of a clinical pharmacist-managed medication assistance program in a renal transplant clinic. Clin Transpl 2000; 14:304e7. 34. Perez A, Doloresco F, Hoffman JM, et al. ACCP: economic evaluations of clinical pharmacy services: 2001e2005. Pharmacotherapy 2009;29:128. 35. Bradley R, Kozura E, Buckle H, et al. Description of clinical risk factor changes during naturopathic care for type 2 diabetes. J Altern Complement 2009;15:633e8. 36. Hanlon JT, Schmader KE, Ruby CM, et al. Suboptimal prescribing in older inpatients and outpatients. J Am Geriatr Soc 2001;49:200e9. 37. Sullivan PW, Lawrence WF, Ghushchyan V. A national catalog of preference-based scores for chronic conditions in the United States. Med Care 2005;43:736e49. 38. Chan DC, Chen JH, Kuo HK, et al. Drug-related problems (DRPs) identified from geriatric medication safety review clinics. Arch Gerontol Geriatr 2012;54:168e74.

Please cite this article in press as: Lin H-W, et al., Economic outcomes of pharmacist-physician medication therapy management for polypharmacy elderly: A prospective, randomized, controlled trial, Journal of the Formosan Medical Association (2017), http:// dx.doi.org/10.1016/j.jfma.2017.04.017