Outcomes of a Medicare Part D telephone medication therapy management program
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
RESEARCH
Outcomes of a Medicare Part D
telephone medication therapy
management program
Leticia R. Moczygemba, Jamie C. Barner, and Evelyn R. Gabrillo
Received December 14, 2011, and in revised
Abstract form June 29, 2012. Accepted for publication
July 21, 2012.
Leticia R. Moczygemba, PharmD, PhD, is
Objective: To determine the impact of telephone medication therapy manage- Assistant Professor, School of Pharmacy,
ment (MTM) on medication- and health-related problems (MHRPs), medication ad- Virginia Commonwealth University, Rich-
herence, and total drug costs for Medicare Part D participants. mond. Jamie C. Barner, PhD, is Professor,
University of Texas, Austin. Evelyn R. Gabril-
Design: Quasiexperimental. lo, PharmD, is Clinical Pharmacist Specialist,
Setting: Regional Medicare Part D plan in Texas in 2007. Scott & White Health Plan Prescription Ser-
Participants: Medicare Part D beneficiaries who were MTM eligible. vices, Temple, TX.
Intervention: Pharmacist-provided telephone MTM consultation. Correspondence: Leticia R. Moczygemba,
Main outcome measures: Change in MHRPs, medication adherence, and total PharmD, PhD, School of Pharmacy, Virginia
Commonwealth University, PO Box 980533,
drug costs from baseline to 12-month follow-up. Richmond, VA 23298-0533. Fax: 804-628-
Results: The intervention (n = 60) and control (n = 60) groups were similar in age 3991. E-mail: lrmoczygemba@vcu.edu
(71.2 ± 7.5 years and 73.9 ± 8.0 years [mean ± SD], respectively), number of medica- Disclosure: The authors declare no con-
tions (13.0 ± 3.2 and 13.2 ± 3.4), chronic diseases (6.5 ± 2.3 and 7.0 ± 2.1), and medi- flicts of interest or financial interests in any
cation regimen complexity index (21.5 ± 7.8 and 22.8 ± 6.9). Men made up 51% of the product or service mentioned in this article,
including grants, employment, gifts, stock
intervention group and 28% of the control group (P = 0.009). MHRPs at baseline were holdings, or honoraria.
4.8 ± 2.7 in the intervention group and 9.2 ± 2.9 in the control group, with 2.2 ± 2.0
Funding: By the American Society of Health-
and 7.3 ± 3.0 at the 12-month follow-up. MHRPs decreased (P = 0.01) in the interven- System Pharmacists Foundation. Also sup-
tion group. We found no predictors of change in medication adherence. Drug costs ported in part by award no. KL2TR000057
decreased by $682 ± 2,141 in the intervention group and increased by $119 ± 1,763 from the National Center for Advancing
Translational Sciences. The contents of this
in the control group. A t test indicated that the cost difference was significant (P = article are solely the responsibility of the au-
0.03), but the adjusted regression analysis did not identify any significant predictors. thors and do not necessarily represent offi-
Conclusion: A telephone MTM program reduced MHRPs. Unadjusted cost com- cial views of the National Center for Advanc-
ing Translational Sciences or the National
parisons showed cost savings in the intervention group. Future research should focus Institutes of Health.
on understanding how telephone MTM affects medication adherence.
Acknowledgments: To Nishi Sarda Goel,
Keywords: Medication therapy management, Medicare Part D, telephone, phar- PharmD, for assistance with data collection
macists, medication-related problems. and to Paul Godley, PharmD; John Chad-
J Am Pharm Assoc. 2012;52:e144–e152. dick, BS, BSPharm; and the Scott & White
Health Plan medication therapy manage-
doi: 10.1331/JAPhA.2012.11258 ment providers and staff for their support of
this study.
Previous presentation: International Soci-
ety of Pharmacoeconomics and Outcomes
Research Annual Meeting, May 21-25, 2011,
Baltimore, MD.
e144 • JAPhA • 5 2 : 6 • N ov /D e c 2012 www.j aph a. or g Journal of the American Pharmacists AssociationPART D TELEPHONE MTM RESEARCH
T
he Medicare Modernization Act of 2003 mandated the multiple studies reporting improvements in clinical and eco-
provision of medication therapy management (MTM) nomic outcomes.3,4,11,15,22,23 MTM was also included in the health
for Medicare Part D beneficiaries with multiple chronic care reform legislation and has been recognized as an integral
diseases, multiple Part D drugs, and high prescription drug component of the medical home model by the Patient-Centered
costs.1 When Medicare Part D was implemented in 2006, MTM Primary Care Collaborative.24 Further, a recent report high-
was a new service for Medicare beneficiaries. However, MTM lighted the positive impact of MTM in a call for recognition of
was not a new concept for pharmacists, who have been manag- pharmacists as health care providers.25
ing medications for diseases such as diabetes, hypertension, All of the 2011 Medicare Part D MTM programs used the
dyslipidemia, and asthma and identifying and resolving medi- telephone as one way to deliver MTM20; therefore, understand-
cation-related problems for more than 20 years.2–17 Moreover, ing the impact of telephone MTM is important. Telephone MTM
the explicit recognition of pharmacists as MTM providers by programs for Medicare Part D patients have been reported to
the Centers for Medicare & Medicaid Services (CMS) highlight- improve efficacy and safety of medication regimens26 and re-
ed the pharmacist's role in medication management and served duce drug costs.26–28 Providing MTM via the telephone may be
as a catalyst for advancing pharmacist services.1 an effective alternative to face-to-face MTM services, particu-
Since 2003, the pharmacy profession has made consider- larly if pharmacy/pharmacist resources are limited or if patient
able progress in advancing MTM. In addition to defining MTM conditions such as poor physical functioning or transportation
and creating an MTM framework,18,19 pharmacists have been barriers prevent provision of face-to-face MTM. This study re-
proactive in the development of MTM programs. Program deliv- ports 12-month outcomes, with an emphasis on resolution of
ery has varied from face-to-face to telephone to a combination medication- and health-related problems (MHRPs), medication
of the two.20 MTM has been provided in a variety of settings, in- adherence, and total Part D drug costs from the Scott & White
cluding ambulatory care, community pharmacy, and safety net, Health Plan (SWHP) telephone MTM program for Medicare
and to both Medicare and non-Medicare populations.13,15,17,21 Part D patients.
Overall, the impact of MTM programs has been positive, with
Objectives
The objectives of this study were to (1) describe the type and
At a Glance number of MHRPs identified and resolved by pharmacists
during a telephone MTM consultation, (2) determine whether
Synopsis: This study looked at the delivery of patients receiving MTM services (intervention group) had
medication therapy management (MTM) services by improved resolution of MHRPs compared with patients who
telephone in a Medicare Part D plan in Texas, com- did not receive MTM services (control group), (3) determine
paring intervention participants with a control group whether the intervention group had improved medication ad-
that chose not to have MTM. Effectiveness of MTM was herence compared with the control group, and (4) determine
measured by looking at changes in medication- and whether the intervention group had decreased total Part D
health-related problems, medication adherence, and drug costs compared with the control group.
total drug costs from baseline to a 12-month follow-
up. Intervention participants were significantly more Methods
likely than the control group to have a resolution of MTM intervention
medication therapy problems, indicating that tele- In 2007, SWHP provided a telephone MTM program that was
phone MTM can be an effective way to deliver the ser- developed by the SWHP MTM coordinator in collaboration with
vice. In addition, participants in the intervention group physicians and other health professionals.29 The opt-in MTM
saw an overall decline in drug costs. program was provided internally to eligible beneficiaries by
Analysis: Part of this study was focused on out- SWHP pharmacists according to an established program pro-
comes central to patient health, such as medication tocol based on the MTM framework created by the American
problems or adherence. Validation of the effective- Pharmacists Association and National Association of Chain
ness of telephone MTM is clearly important for sev- Drug Stores Foundation.19 Because the pharmacists were al-
eral reasons, including patient convenience, patient ready employees of SWHP, they were not reimbursed sepa-
access, use of pharmacist resources, and the fact rately, as the payment for MTM was included as an administra-
that Medicare Part D, in requiring MTM, permits tive fee in the plan bid. The foundation of the MTM consultation
the telephone as one method of delivery. Insurance was a comprehensive medication therapy review conducted
companies, the government, and the health care field via telephone by the SWHP pharmacist. The review was indi-
in general are looking for ways to reduce costs, and vidualized based on the patient's needs. The medication review
as telephone MTM showed a capacity in this study to included assessing, identifying, and resolving medication ther-
contain drug costs, it is worth further investigation. apy problems related to appropriateness, effectiveness, adher-
Any method that can simultaneously benefit patients ence, cost, and safety. The categories for medication-related
and reduce costs should be explored. problems were cost-related/formulary interchange, potential
or actual drug interaction, drug needed but not prescribed,
Journal of the American Pharmacists Association www. japh a. or g N ov /D e c 2012 • 52:6 • JAPhA • e145RESEARCH PART D TELEPHONE MTM
dose, efficacy, schedule/duration, prescribed drug not needed, phone one time during the study period. Recruitment for the
adverse effects, medication adherence, therapeutic duplica- study began in May 2007 and lasted until January 2008. The in-
tion, and safety. Education needs regarding medications and/or tervention group consisted of patients who voluntarily enrolled
disease management were also assessed. A unique component in the SWHP MTM program and received at least one MTM con-
of the program was its emphasis on conducting a comprehen- sultation during the period from May 2007 to January 2008.
sive review of preventive care needs as well as medication- The control group consisted of patients who were eligible to
related problems. The U.S. Preventive Task Force Recommen- receive MTM but did not enroll in the program. To control for
dations were used to guide the preventive care assessment, selection bias and baseline differences, intervention and con-
which included a review of whether patients were up to date trol groups were matched by the number of chronic diseases
with immunizations (tetanus, influenza, and pneumococcal), and Part D drugs. Data collection in the intervention and con-
comprehensive metabolic profile, sigmoidoscopy/colonoscopy, trol groups occurred via retrospective review of the electronic
mammography/breast exam, depression screening, cholester- medical record (number of MHRPs) and secondary database
ol screening, liver function tests, thyroid-stimulating hormone analysis of prescription claims (medication adherence and
test, and dual x-ray absorptiometry scan. Diabetes preventive total Part D drug costs). The study was approved by the Uni-
care needs, which included making sure eye and foot exams versity of Texas, Virginia Commonwealth University, and SWHP
and urine creatinine labs were up to date, were also assessed. institutional review boards.
Follow-up was determined on a case-by-case basis and individ-
ualized based on a patient's needs. The MTM consultation was MHRPs: Intervention group
documented in a customized SWHP database created by the The pharmacist reviewed the patient's electronic medical re-
MTM coordinator and an information systems programmer.29 cord, which consisted of medical history, medications, physi-
Following the MTM consultation, patients were mailed a cian notes, laboratory reports and prescription claim records,
personal and portable medication record for self-management. to identify potential MHRPs to discuss with the patient (i.e.,
Patients were also mailed a medication action plan that con- preassessment). MHRPs were further assessed in the interven-
tained patient-tailored information for the patient to use in op- tion group during the MTM telephone consultation (i.e., assess-
timizing medication and health self-management. The action ment). During the telephone consultation, problems identified
plan also reinforced education provided during the MTM con- in the chart review were confirmed and additional MHRPs were
sultation by including educational materials when applicable. assessed. The pharmacist then made recommendations to the
Patients were encouraged to voluntarily share the medication patient for resolution of MHRPs. The personal medication re-
record and action plan with health care providers to enhance cord and medication action plan were mailed to each patient
continuity of care and help ensure that the provider was aware after the MTM consultation. When necessary, the patient fol-
of the patient's most current MHRPs. The interventions includ- lowed up with his/her physician for recommendations such as a
ed working with the patient and/or caregiver on patient-specif- medication change. The duration of the MTM consultation was
ic problems or communicating with physicians or other health an estimated 45 to 140 minutes for MTM preassessment, as-
care providers about the opportunities to resolve existing or sessment, and documentation, with an average of 75 minutes
potential MHRPs. Beneficiaries who had two or more chronic spent per patient.
diseases, were taking two or more Part D medications, and in- The number of problems was assessed at baseline and at
curred at least $1,000 in Part D costs per quarter were eligible the 12-month follow-up. The change in MHRPs was defined as
for SWHP's MTM program. Approximately 18,000 beneficiaries the difference in the number of MHRPs from baseline to fol-
were enrolled in SWHP's Medicare Part D Plan in 2007 and, low-up. The electronic medical record and prescription claims
of these, 1,999 were eligible for MTM and 123 beneficiaries were used to determine resolution of MHRPs. Only medication
received MTM services. therapy–related problems identified at baseline were evaluat-
ed at the follow-up. Recommendations that were not accepted
Study design and sample were considered as a failure to resolve the MHRP. A detailed
This study used a quasiexperimental design for comparing the description of the MHRPs has previously been reported.29,30
change in the number of MHRPs, change in medication adher-
ence, and change in total Part D drug costs between the inter- MHRPs: Control group
vention and control groups from baseline to a 12-month follow- The assessment of MHRPs in the control group was conducted
up. All SWHP Medicare Part D beneficiaries who received an retrospectively via review of the electronic medical record. The
MTM consultation from a SWHP pharmacist in 2007 (n = 123) identification of MHRPs was the same procedure used in iden-
were targeted and invited to participate in the study at the end tifying initial problems (i.e., preassessment) in the intervention
of the MTM consultation. In addition to meeting SWHP's MTM group. To ensure consistency in evaluating MHRPs in the in-
eligibility criteria, beneficiaries in the intervention group had tervention and control groups via chart review, the researcher
to verbally consent to participate in the study and consent to (also a pharmacist) was trained by the SWHP MTM coordinator
the use and disclosure of protected health information. Ben- who conducted MTM consultations in the intervention group.
eficiaries 90 years or older were excluded from the study due Because the control group did not receive a telephone inter-
to patient privacy concerns. MTM was conducted over the tele- vention with the SWHP pharmacist, only the following MHRP
e146 • JAPhA • 5 2 : 6 • N ov /D e c 2012 www.j aph a. or g Journal of the American Pharmacists AssociationPART D TELEPHONE MTM RESEARCH
subsets were evaluated in the intervention and control groups: dictors of change in MHRPs, medication adherence, and total
therapeutic duplication, cost/formulary interchange, dose, Medicare Part D drug costs from baseline to 12-month follow-
drug interactions, and preventive care needs. The number of up. The data for the variable change in Part D drug costs were
MHRPs was recorded and measured at baseline and at the normally distributed; therefore, transformation of the data was
12-month follow-up. not necessary. The a priori significance level was 0.05.
Medication adherence Results
Medication adherence was assessed in the intervention and Of the 123 beneficiaries who met the Medicare Part D criteria,
control groups by examining refill history using SWHP's pre- 95 also met the study inclusion criteria. Of those eligible for the
scription claims database. The medication possession ratio study, 60 were enrolled, resulting in a 63% enrollment rate.
(MPR) was used to measure adherence at baseline (12 months Reasons for declining to participate included preference and
preintervention) and at the 12-month follow-up (12 months breach of confidentiality.
postintervention). An average MPR was calculated. The follow-
ing formula was used to calculate the MPR for each medication Sociodemographic and health-related variables
at baseline and 12-month follow-up: sum of each day's supply Table 1 displays the sociodemographic and health-related
of prescription medications during the 12-month preinterven- characteristics of the intervention and control groups. The
tion period divided by that during the 12-month postinterven- mean (±SD) age of participants in the intervention group was
tion period.31–33 The MPRs then were summed, and finally, the 71.2 ± 7.5 years, and the majority were white (78.3%) and
MPR sum was divided by the number of medications to calcu- male (51.7%). Members of the intervention group were taking
late the average MPR for each study participant. The change in 13.0 ± 3.2 medications, had 6.5 ± 2.3 chronic diseases, and
medication adherence was measured as the difference in MPR had an MRCI of 21.5 (range 8–43). The three most common dis-
from baseline to 12-month follow-up. eases were hypertension (95.0%), dyslipidemia (77.0%), and
diabetes (55.0%). Control group participants were aged 73.9
Total Part D drug costs ± 8.0 years, and the majority were also white (91.7%). Only
Prescription claim records were used to obtain the total cost of 28.3% were male in the control group. Members of this group
all Part D drugs, which included the amount paid by the SWHP were taking 13.2 ± 3.4 medications, had 7.0 ± 2.1 chronic dis-
Part D program and the patient copay, filled during the study eases, and had an MRCI of 22.8 (range 9–42.5). The three most
period. The sum of the total Part D drug costs at baseline (12 common diseases in the control group also were hypertension
months preintervention) and follow-up (12 months postinter- (95.0%), dyslipidemia (86.7%), and diabetes (60.0%). With
vention) was calculated. The change in total Part D drug costs the exception of gender (P = 0.009), no significant baseline dif-
from baseline to follow-up was calculated from baseline to the ferences between the two groups were found.
12-month follow-up. All costs were adjusted for inflation to U.S.
January 2009 dollars. MHRPs
A total of 357 MHRPs were identified at baseline in the inter-
Independent variable and covariates vention group, and 62% (220 of 357) were considered resolved
The primary independent variable in this study was MTM use (i.e., recommendations accepted) at the 12-month follow-up.
(yes/no), and it distinguished the intervention and control Pharmacists identified 6.0 ± 2.9 MHRPs per patient at base-
groups. Sociodemographic covariates included age, gender, line, and this number decreased to 2.3 ± 2.0 MHRPs at the
and race. Health-related covariates were number of medica- 12-month follow-up, corresponding with an average of 3.7
tions, number of chronic diseases, and medication regimen problems resolved per patient. The most common problem
complexity. The medication regimen complexity index (MRCI)34 was cost related, with 85% of patients having at least one cost-
was used to determine medication regimen complexity in the related problem and 42% having three or more cost-related
intervention and control groups. The MRCI was calculated only problems. At least one (range 0–7) preventive care need was
for prescription medications in both groups. All sociodemo- identified in 78% of participants, and a need for disease man-
graphic and health-related variables were continuous, with the agement education was identified in 52% of the study partici-
exception of gender and race. pants in the intervention group. A potential drug interaction
was identified in 27% of participants. The most common type
Data analysis of pharmacist recommendation was medication related (51%),
Descriptive statistics were calculated for all study variables. followed by preventive care (28%), and education (21%).
Bivariate analyses (t tests for continuous variables and chi- For the subset of problems that was assessed in both
square tests for categorical variables) were performed to as- groups, the intervention group had 4.8 ± 2.7 MHRPs at base-
sess baseline differences between the intervention and control line and 2.2 ± 2.0 problems remained at the 12-month fol-
groups. Because some categories of race had less than five ob- low-up. In contrast, the control group had 9.2 ± 2.9 MHRPs
servations, race was collapsed into two categories: white and identified at baseline and 7.3 ± 3.0 problems remained at the
nonwhite, to avoid violating a chi-square assumption. Three 12-month follow-up. All of the participants in the control group
separate multiple regression models were used to identify pre- had at least one preventive care need identified, and approxi-
Journal of the American Pharmacists Association www. japh a. or g N ov /D e c 2012 • 52:6 • JAPhA • e147RESEARCH PART D TELEPHONE MTM
Table 1. Sociodemographic and health-related character- Table 2. Type and number of MHRPs for Medicare Part D
istics of Medicare Part D beneficiaries in the intervention beneficiaries in the intervention (n = 60) and control (n = 60)
and control groups groups at baseline
Intervention Control Intervention Controla
Characteristic Mean ± SD Mean ± SD No. (%) No. (%)
n 60 60 Medication-related problems
Age (years) 71.2 ± 7.5 73.9 ± 8.0 Cost-related/formulary interchange 133 (66.8) 178 (83.6)
No. of chronic conditions at Potential or actual drug interaction 25 (12.6) 21 (9.9)
baseline 6.5 ± 2.3 7.0 ± 2.1 Drug needed but not prescribed 11 (5.5) —
No. of medications at baseline 13.0 ± 3.2 13.2 ± 3.4 Dose 7 (3.5) —
Medication regimen complexitya 21.5 ± 7.5 22.8 ± 6.9 Efficacy 6 (3.0) 7 (3.3)
Gender, no. (%) Schedule/duration 4 (2.0) —
Men 31 (51.7) 17 (28.3) Prescribed drug not needed 3 (1.5) —
Women 29 (48.3) 43 (71.7) Adverse effects 3 (1.5) —
Race, no. (%)b Medication adherence 3 (1.5) —
White 47 (85.5) 55 (91.7) Therapeutic duplication 2 (1.0) 2 (0.9)
Nonwhite 8 (14.6) 5 (8.3) Safety 2 (1.0) 5 (2.3)
a
Medication regimen complexity was measured using the medication regimen Total 199 (99.9)b 213 (100.0)
complexity index.34
b
Race was missing for five participants in the intervention group. Preventive care problems
Tetanus immunization 21 (18.3) 40 (12.0)
mately 70% had five or more preventive care needs. At least Thyroid-stimulating hormone labora-
one cost-related problem was identified in 92% of participants, tory 14 (12.2) 13 (3.9)
and 55% had three or more cost-related problems. One-third Urine albumin–to–creatinine ratio
of participants in the control group had a potential drug inter- (patients with diabetes) 13 (11.3) 27 (8.1)
action identified during the review. The multiple regression in- Sigmoidoscopy/colonoscopy 12 (10.4) 32 (9.6)
dicated that the difference in MHRPs measured from baseline Diabetic eye exam 11 (9.6) 32 (9.6)
to 12-month follow-up was statistically significantly larger in Pneumococcal immunization 10 (8.7) 27 (8.1)
the intervention group (P = 0.01) compared with the control Diabetic foot exam 8 (7.0) 20 (6.0)
group after adjusting for the other variables in the model. No
Influenza immunization 6 (5.2) 54(16.3)
other significant predictors of problems were found. Table 2
Cholesterol panel 4 (3.5) 10 (3.0)
summarizes the type and number of problems identified in the
Mammography/breast exam 4 (3.5) 23 (6.9)
intervention and control groups, Table 3 describes the type and
number of pharmacist recommendations for the intervention Depression screening 4 (3.5) 23 (6.9)
group, and Table 4 reports the multiple regression results. Dual-energy X-ray absorptiometry
scan 3 (2.6) 22 (6.6)
Medication adherence and total Part D drug costs Comprehensive metabolic profile 3 (2.6) 3 (0.9)
MPR was similar in the intervention and control groups at Liver function test 2 (1.7) 6 (1.8)
baseline (0.53 ± 0.15 and 0.55 ± 0.18, respectively) and was Total 115 (100.1)b 332 (99.7)b
relatively unchanged (0.51 ± 0.18 and 0.57 ± 0.17) at the Education needs
12-month follow-up. The multiple regression analysis did not Diabetes management 21 (48.8) —
identify any significant predictors of change in medication ad- Medication related 8 (18.6) —
herence. The Part D drug costs in the intervention group were Over-the-counter medications 5 (11.6) —
$4,619 ± 1,746 at baseline and decreased to $3,938 ± 1,022 Other disease managementc 9 (20.9) —
at 12-month follow-up, resulting in a cost savings of $681 per Total 43 (99.9)b —
patient annually. In the control group, Part D drug costs were Total no. of MHRPs
$4,723 ± 2,520 at baseline and increased by $119 (to $4,842 All problems 357 —
± 3,405) at 12-month follow-up. A t test indicated a significant
Subset of problems 287 545
difference in the change in drug costs between the intervention
Abbreviation used: MHRP, medication- and health-related problem.
and control groups (P = 0.03), but when sociodemographic, a
Because the pharmacist did not have direct interaction with the control group, only
health-related, and use variables were controlled for in the a subset of problems were evaluated in the control group.
multiple regression analysis, no significant predictors of the
b
Percentages do not add up to 100% because of rounding.
c
Other disease management included education related to coronary obstructive
change in drug costs were seen. pulmonary disease, epilepsy, osteoporosis, restless leg syndrome, hypertension,
Sjogren’s syndrome, constipation, urinary frequency, and smoking cessation.
e148 • JAPhA • 5 2 : 6 • N ov /D e c 2012 www.j aph a. or g Journal of the American Pharmacists AssociationPART D TELEPHONE MTM RESEARCH
age of participants who met their low-density lipoprotein goals
Table 3. Type and number of pharmacist recommendations was similar in both groups (74% and 78%, respectively).35
for 60 Medicare Part D beneficiaries in the intervention Despite the benefits of telephone delivery, clinically complex
groupa participants with many medications may receive the most
No. (%) benefit from face-to-face services. An initial face-to-face visit
Medication-related recommendations with telephone follow-up visits could be an effective delivery
Patient assistance program or coupon 76 (41.3) model. As MTM evolves, studying and developing criteria that
Change drug 58 (31.5) MTM programs can use to determine participants who should
Laboratory monitoring for efficacy or safety 31 (16.8) receive telephone services, face-to-face services, or a combi-
Change dose 6 (3.3) nation of the two will be important. This distinction would help
Discontinue drug 5 (2.7) maximize MTM resources and meet patient needs.
Change schedule/duration 5 (2.7) Only 6% of eligible beneficiaries were enrolled in the
SWHP Part D MTM program, which is lower than previous re-
Add drug 3 (1.6)
ports of participation rates in Part D opt-in programs that re-
Total 184 (99.9)b
ported average participation ranging from 14% to 18%.38 This
Preventive care recommendations lack of participation is concerning because participants were
Tetanus immunization 21 (18.1) not required to pay an additional fee to participate in Part D
Thyroid-stimulating hormone laboratory 14 (12.1) MTM. The low uptake of MTM may have been a result of the
Urine albumin–to–creatinine ratio 14 (12.1) lack of familiarity with the program or program marketing.
Sigmoidoscopy/colonoscopy 12 (10.3) CMS now mandates an opt-out enrollment for Part D MTM pro-
Diabetic eye exam 11 (9.4) grams; however, patients must still choose to participate in the
Pneumococcal immunization 10 (8.6) program. Therefore, seeking feedback from patients to develop
Diabetic foot exam 8 (6.9) and test marketing tools may help improve patient uptake of
Influenza immunization 6 (5.2) services.
Cholesterol panel 4 (3.4)
MHRPs
Mammography/breast exam 4 (3.4)
Depression screening 4 (3.4) The most common type of problem identified by SWHP phar-
macists was related to cost, which differs from other studies
Dual-energy X-ray absorptiometry scan 3 (2.6)
that have identified a need for additional drug therapy13,15,22,23
Comprehensive metabolic profile 3 (2.6)
to be one of the most common medication-related problems.
Liver function test 2 (1.7) In our study, these findings were not surprising because phar-
Total 116 (99.8)b macists who provided MTM observed that many participants
Education recommendations were interested in learning about cost alternatives and how to
Total 75 (100.0) maximize their Part D benefit during their encounters. The in-
Total no. of recommendations 375 terest in cost savings and the Part D benefit may have been
Some problems had more than one recommendation.
a
due to the newness of the Medicare Part D program in 2007.
Percentages do not add up to 100% because of rounding.
b
However, these results coincide with other studies that re-
Discussion ported cost-related problems were commonly addressed dur-
This study demonstrated a significant difference in the number ing Part D pharmacist consultations. In fact, one study found
of problems resolved and, in the unadjusted analysis, an annual cost-related recommendations to be the second most common
drug cost savings for beneficiaries who participated in a tele- pharmacist intervention for a group of Part D beneficiaries who
phone MTM consultation. Given that all Part D plans are us- received MTM via the telephone.39 Another study evaluated the
ing the telephone as one method of MTM delivery,20 telephone potential impact of pharmacists reviewing Medicare Part D
services will likely continue to have a role in MTM. Benefits of beneficiaries’ drug regimens for potential therapeutic substitu-
telephone services include increased efficiency of use of phar- tions and determined that 54% (27 of 50 participants) had pos-
macy personnel35 and increased access for participants with sible therapeutic substitutions that could save an estimated
transportation barriers or who live in rural areas.36 In fact, one $1,300 per year in Part D drug costs.36 Another telephone MTM
pharmacist-managed dementia telephone clinic that initially program for Part D beneficiaries included a review of medica-
began as face-to-face changed the delivery method to meet tions for cost effectiveness as part of the protocol but did not
the needs of rural participants and those with transportation report how many medication changes resulting from cost were
burdens.36 SWHP MTM participants are reported to like the made.26 Similarly, one Part D plan used their formulary to opti-
convenience of the telephone,37 and this preference has been mize drug costs by making recommendations such as generic
noted by others who have participated in services provided by a substitutions when appropriate.28 Because cost appears to
pharmacist via the telephone.35,36 Further, one study compared be a concern for patients, pharmacists can work with them to
participants in a face-to-face (n = 78) and telephone (n = 79) identify potential cost savings alternatives and opportunities to
pharmacist-provided lipid clinic and found that the percent- avoid the coverage gap or enter the gap later in the year.40 Al-
Journal of the American Pharmacists Association www. japh a. or g N ov /D e c 2012 • 52:6 • JAPhA • e149RESEARCH PART D TELEPHONE MTM
Table 4. Multiple regression of variables predicting change in MHRPs from baseline to 12-month follow-up
Variable Parameter estimate (b) t P
Intercept 2.40 1.38 0.07
Intervention group (MTM use) 0.81 2.56 0.01a
Age –0.004 –0.19 0.85
Female –0.39 –1.20 0.23
Nonwhite 0.013 0.03 0.98
No. of medications 0.02 0.40 0.69
No. of chronic diseases –0.13 –1.62 0.11
Medication regimen complexity 0.02 0.96 0.34
Abbreviation used: MTM, medication therapy management.
F statistic = 2.11, df = 7, model P value = 0.0489, R2 = 0.12, adjusted R2 = 0.06.
a
Statistically significant at P < 0.05.
though the recent health care reform legislation has provisions may obtain medications from sources that are not able to be
to close the gap in prescription drug coverage for Part D ben- tracked by administrative claims, it would be useful to docu-
eficiaries, it will not be completely closed until 2020. During ment the source of each medication during the consultation.
MTM encounters, pharmacists can help patients navigate their Other studies that have evaluated telephone MTM programs for
Part D plans in order to maximize the benefits. Part D participants have not included medication adherence
Preventive care needs were commonly identified in both as an outcome measure. One of the goals of Part D MTM is
groups, although a low number of preventive care problems to improve medication adherence. Therefore, future research
were resolved in the intervention group. Given the emphasis on should focus on measuring medication adherence and consider
preventive care during the MTM consultation, the low uptake factors such as $4 generic medications when designing the
of pharmacist recommendations for preventive care problems study.
was disappointing. A telephone reminder may have improved Although the adjusted model with the addition of the co-
resolution. In fact, in 2010, CMS expanded MTM program re- variates did not reach statistical significance, the practical
quirements to include ongoing monitoring with reviews at least importance of saving $682 per patient per year ($57 monthly)
quarterly, which is likely to improve resolution of problems. should not be overlooked. In our study and others,26,28,36,39 phar-
Also, participants may not have had previous experience with macists addressed cost-related concerns during MTM consul-
a pharmacist involved in their preventive care. Nonetheless, tations, which may suggest that during Part D MTM consulta-
preventive care provides an opportunity to expand the phar- tions, emphasis should be given to reviewing medications for
macist's roles and responsibilities on health care teams. The cost-saving alternatives. This emphasis on cost is further sup-
Affordable Care Act expanded preventive care coverage for ported by cost savings in another study that reported a $40 per
Medicare beneficiaries to include an annual wellness visit that month cost savings for 12,196 Medicare Part D patients who
assesses preventive care needs according to the U.S. Preven- participated in a telephone MTM service in a community phar-
tive Services Task Force guidelines and recommendations. macy setting.27 In that study, cost savings were higher ($40 vs.
Physicians, physician assistants, nurse practitioners, and $29) in the telephone group than in the face-to-face group (n
clinical nurse specialists and other health professionals such = 9,140) in the same setting.27 These studies only considered
as pharmacists working under direct physician supervision can Part D drug costs. Other costs such as the impact of MTM on
bill for these services. overall health care costs and return on investment are impor-
tant to payer stakeholders and warrant further examination.
Medication adherence and total Part D drug costs
An MPR of 0.8 or more is commonly accepted as a measure of Limitations
good adherence.41,42 In this study, participants in the interven- This study reports outcomes from a regional Part D telephone
tion and control groups had an average MPR of about 0.5 at MTM program; therefore, these results may not be generaliz-
baseline and this remained stable at the 12-month follow-up. able to all Part D plans or plans that deliver face-to-face MTM.
We expected that adherence would increase in the interven- Despite this limitation, our results support telephone MTM
tion group following the MTM consultation. These results were as having a positive impact on medication-related problems,
also surprising considering that medication adherence was not which is important because all Part D plans are using the tele-
identified as a common problem during the MTM consultation. phone as one method of MTM delivery. The intervention did not
Participants in both groups may have been using $4 generic target a specific condition; therefore, evaluating condition-
programs at local pharmacies when applicable. Also, SWHP specific measures was beyond the scope of this study. As all
pharmacists recommended patient assistance programs dur- eligible beneficiaries must be offered the opportunity to par-
ing MTM encounters in the intervention group, which may have ticipate in Medicare Part D MTM programs, randomization was
affected the MPR in the follow-up period. Because patients not possible in this study. However, a control group was used
e150 • JAPhA • 5 2 : 6 • N ov /D e c 2012 www.j aph a. or g Journal of the American Pharmacists AssociationPART D TELEPHONE MTM RESEARCH
to assess the impact of the intervention on study outcomes. 7. Schumock GT, Butler MG, Meek PD, et al. Evidence of the eco-
Although matching was used to help address selection bias, nomic benefit of clinical pharmacy services: 1996–2000. Phar-
macotherapy. 2003;23:113–32.
those who opted into the MTM program may be more proactive
about their health and seeking preventive care services com- 8. Schnipper JL, Kirwin JL, Cotugno MC, et al. Role of pharmacist
pared with those who did not enroll in the program. Matura- counseling in preventing adverse drug events after hospital-
ization. Arch Intern Med. 2006;166:565–71.
tion bias is also a concern because participants in the inter-
vention group may have been more comfortable managing their 9. McDonough RP, Doucette WR. Drug therapy management: an
problems over time. Further, the retrospective identification empirical report of drug therapy problems, pharmacists’ inter-
ventions, and results of pharmacists’ actions. J Am Pharm As-
of MHRPs in the control group did not allow for confirmation
soc. 2003;43:511–18.
of the problems with participants. To address this limitation,
only a subset of problems was evaluated for the comparison 10. Hanlon JT, Weinberger M, Samsa GP, et al. A randomized,
controlled trial of a clinical pharmacist intervention to improve
between the intervention and control groups. However, the
inappropriate prescribing in elderly outpatients with poly-
number of problems identified in the control group may be
pharmacy. Am J Med. 1996;100:428–37.
overestimated, as some of the problems might have been po-
11. Smith SR, Catellier DJ, Conlisk EA, Upchurch GA. Effect on
tential instead of actual problems. We were not able to capture
health outcomes of a community-based medication therapy
medical record or prescription claim information for services
management program for seniors with limited incomes. Am J
external to the SWHP system, which is an inherent limitation Health Syst Pharm. 2006;63:372–9.
to using secondary and administrative data. In particular, this
12. Sturgess IK, McElnay JC, Hughes CM, Crealey G. Community
lack of information may have affected calculation of the MPR if
pharmacy based provision of pharmaceutical care to older pa-
participants were using $4 generics at local pharmacies. In ad- tients. Pharm World Sci. 2003;25:218–26.
dition participants may have received vaccinations at influenza
13. Doucette WR, McDonough RP, Klepser D, McCarthy R. Com-
clinics or pharmacies in their respective communities.
prehensive medication therapy management: identifying and
resolving drug-related issues in a community pharmacy. Clin
Conclusion Ther. 2005;27:1104–11.
A telephone MTM program decreased MHRPs in Medicare Part
14. Stebbins MR, Kaufman DJ, Lipton HL. The PRICE clinic for
D beneficiaries. Part D drug cost savings among the interven- low-income elderly: a managed care model for implement-
tion group were observed in the unadjusted analysis. Although ing pharmacist-directed services. J Manag Care Pharm.
preventive care recommendations had a low uptake in this 2005;11:333–41.
study, opportunities to expand the pharmacist's role in pre- 15. Isetts BJ, Schondelmeyer SW, Artz MB, et al. Clinical and
ventive care, such as those included in the health care reform economic outcomes of medication therapy management
legislation, should be explored. Further research is needed to services: the Minnesota experience. J Am Pharm Assoc.
quantify the impact of telephone MTM on medication adher- 2008;48:203–11.
ence. 16. Lewis NJ, Bugdalski-Stutrud C, Abate MA, et al. The medica-
References tion assessment program: comprehensive medication assess-
ments for persons taking multiple medications for chronic dis-
1. Centers for Medicare & Medicaid Services. Medicare prescrip-
eases. J Am Pharm Assoc. 2008;48:171–80.
tion drug benefit final rule. Accessed at www.cms.hhs.gov/
providerupdate/regs/CMS4068F.pdf, February 21, 2005. 17. Christensen DB, Roth M, Trygstad T, Byrd J. Evaluation of a pilot
medication therapy management project within the North Car-
2. Cranor CW, Bunting BA, Christensen DB. The Asheville Proj-
olina state health plan. J Am Pharm Assoc. 2007;47:471–83.
ect: long-term clinical and economic outcomes of a commu-
nity pharmacy diabetes care program. J Am Pharm Assoc. 18. Bluml BM. Definition of medication therapy management: de-
2003;43:173–84. velopment of professionwide consensus. J Am Pharm Assoc.
2005;45:566–72.
3. Bunting BA, Cranor CW. The Asheville Project: long-term
clinical, humanistic, and economic outcomes of a community- 19. American Pharmacists Association and National Associa-
based medication therapy management program for asthma. tion of Chain Drug Stores Foundation. Medication therapy
J Am Pharm Assoc. 2006;46:133–47. management in community pharmacy practice: core ele-
ments of an MTM service (version 1.0). J Am Pharm Assoc.
4. Bunting BA, Smith BH, Sutherland SE. The Asheville Project:
2005;45:573–9.
clinical and economic outcomes of a community-based long-
term medication therapy management program for hyperten- 20. United States Centers for Medicare & Medicaid Services. 2011
sion and dyslipidemia. J Am Pharm Assoc. 2008;48:23–31. Medicare Part D medication therapy management program
fact sheet. Accessed at www.cms.gov/Medicare/Prescription-
5. Bluml BM, McKenney JM, Cziraky MJ. Pharmaceutical care
Drug-Coverage/PrescriptionDrugCovContra/downloads//MT-
services and results in Project IMPACT: Hyperlipidemia. J Am
MFactSheet2011063011Final.pdf, May 1, 2012.
Pharm Assoc. 2000;40:157–65.
21. Moczygemba LR, Goode JV, Gatewood SB, et al. Integration
6. Goode JK, Swiger K, Bluml BM. Regional osteoporosis screen-
of collaborative medication therapy management in a safety
ing, referral, and monitoring program in community phar-
net patient-centered medical home. J Am Pharm Assoc.
macies: findings from Project IMPACT: Osteoporosis. J Am
2011;51:167–72.
Pharm Assoc. 2004;44:152–60.
Journal of the American Pharmacists Association www. japh a. or g N ov /D e c 2012 • 52:6 • JAPhA • e151RESEARCH PART D TELEPHONE MTM
22. Ramalho de Oliveira D, Brummel AR, Miller DB. Medication 32. Vik SA, Maxwell CJ, Hogan DB, et al. Assessing medication ad-
therapy management: 10 years of experience in a large inte- herence among older persons in community settings. Can J
grated health care system. J Manag Care Pharm. 2010;16:185– Clin Pharmacol. 2005;12;e152–64.
95. 33. MacLaughlin EJ, Raehl CL, Treadway AK, et al. Assessing
23. Smith M, Giuliano MR, Starkowski MP. In Connecticut: im- medication adherence in the elderly: which tools to use in clin-
proving medication management in primary care. Health Af- ical practice? Drugs Aging. 2005;22:231–55.
fairs (Millwood). 2011;4:646–54. 34. George J, Phun YT, Bailey MJ, et al. Development and valida-
24. Patient-Centered Primary Care Collaborative. The patient- tion of the medication regimen complexity index. Ann Phar-
centered medical home: integrating comprehensive medica- macother. 2004;38:1369–76.
tion management to optimize patient outcomes. Accessed at 35. Dolder NM, Dolder CR. Comparison of a pharmacist-managed
www.pcpcc.net/files/medmanagepub.pdf, August 4, 2011. lipid clinic: in-person versus telephone. J Am Pharm Assoc.
25. Giberson S, Yoder S, Lee MP. Improving patient and health 2010;50:375–8.
system outcomes through advanced pharmacy practice: a re- 36. Oderda L, Holman C, Nichols B, et al. Pharmacist-managed
port to the U.S. Surgeon General, Office of the Chief Pharma- telephone clinic review of antidementia medication effective-
cist. Washington, DC: U.S. Public Health Service; 2011. ness. Consult Pharm. 2011;26:264–73.
26. Pindolia VK, Stebelsky L, Romain TM, et al. Mitigation of medi- 37. Moczygemba LR, Barner JC, Brown CM, et al. Patient satisfac-
cation mishaps via medication therapy management. Ann tion with a pharmacist-provided telephone medication thera-
Pharmacother. 2009;43:611–20. py management service. Res Social Adm Pharm. 2010;6:143–
27. Winston S, Lin Y. Impact on drug cost and use of Medicare Part 54.
D of medication therapy management services in 2007. J Am 38. Academy of Managed Care Pharmacy. Sound medication
Pharm Assoc. 2009;49:813–20. therapy management programs: version 2.0 with validation
28. Fox D, Ried LD, Klein GE, et al. A medication therapy manage- study. J Manag Care Pharm. 2008;14(1 suppl B):s2–44.
ment program's impact on low-density lipoprotein cholesterol 39. Perera PN, Guy MC, Sweaney AM, et al. Evaluation of pre-
goal attainment in Medicare Part D patients with diabetes. J scriber responses to pharmacist recommendations communi-
Am Pharm Assoc. 2009;49:192–9. cated by fax in a medication therapy management program
29. Moczygemba LR, Barner JC, Gabrillo ER, Godley PJ. Devel- (MTMP). J Manag Care Pharm. 2011;17:345–54.
opment and implementation of a telephone medication ther- 40. Alston G, Hanrahan C. Can a pharmacist reduce annual costs
apy management program for Medicare beneficiaries. Am J for Medicare Part D enrollees? Consult Pharm. 2011;26:182–9.
Health Syst Pharm. 2008;65:1655–60.
41. Hansen RA, Kim MM, Song L, et al. Comparison of methods to
30. Moczygemba LR, Barner JC, Lawson KA, et al. The impact of assess medication adherence and classify nonadherence. Ann
telephone medication therapy management on medication Pharmacother. 2009;43:413–22.
and health-related problems, medication adherence, and
42. Karve S, Cleves MA, Helm M, et al. Good and poor adherence:
Medicare Part D drug costs: a 6-month follow-up. Am J Geriatr
optimal cut-point for adherence measures using administra-
Pharmacother. 2011;9:328–38.
tive claims data. Curr Med Res Opin. 2009;25:2303–10.
31. Grymonpre RE, Didur CD, Montgomery PR, Sitar DS. Pill count,
self-report, and pharmacy claims data to measure medication
adherence in the elderly. Ann Pharmacother. 1998;32:749–54.
e152 • JAPhA • 5 2 : 6 • N ov /D e c 2012 www.j aph a. or g Journal of the American Pharmacists AssociationYou can also read
