jmcp.2016.22.5.493 Rup

S Y S T E M AT I C R E V I E W
Pharmacist Interventions in the Management of Type 2 Diabetes

Mellitus: A Systematic Review of Randomized Controlled Trials
Sarah Pousinho, PharmD, MSc; Manuel Morgado, PharmD, PhD;
Amílcar Falcão, PharmD, PhD; and Gilberto Alves, PharmD, PhD
ABSTRACT out of 14 studies. Of note, in all 6 studies that estimated the 10-year CHD
BACKGROUND: Diabetes mellitus is a major health problem that is grow- risk among study patients, a greater improvement in the intervention group
ing rapidly worldwide. A collaborative and integrated team approach in versus the control group was found. In addition, pharmacist interventions
which pharmacists can play a pivotal role should be sought when managing also had a positive impact on medication adherence and HRQoL in most
patients with diabetes. studies that ascertained these outcomes. Finally, although only 3 studies
conducted a cost-effectiveness analysis, pharmacist interventions proved to
OBJECTIVE: To identify and summarize the main outcomes of pharmacist be cost-effective.
interventions in the management of type 2 diabetes.
CONCLUSIONS: The findings from this review clearly support the involve-
METHODS: PubMed, Cochrane Central Register of Controlled Trials, and ment of pharmacists as members of health care teams in the management
Web of Science were searched for randomized controlled trials evaluating of patients with type 2 diabetes.
the effectiveness of any pharmacist intervention directed at patients with
type 2 diabetes in comparison with usual care. Outcome measures of inter- J Manag Care Spec Pharm. 2016;22(5):493-515
est included glycosylated hemoglobin (A1c), blood glucose, blood pressure, Copyright © 2016, Academy of Managed Care Pharmacy. All rights reserved.
lipid profile, body mass index (BMI), 10-year coronary heart disease (CHD)
risk, medication adherence, health-related quality of life (HRQoL), and eco-
nomic outcomes. The risk of bias in included studies was assessed using
What is already known about this subject
the Cochrane risk of bias tool.
RESULTS: Thirty-six studies were included in this systematic review, • Type 2 diabetes is a serious and highly prevalent metabolic dis-
involving 5,761 participants. The studies evaluated the effects of several order that imposes unacceptable high costs on many countries
pharmacist interventions carried out in various countries and in differ- around the world.
ent health care facilities, such as community pharmacies, primary care • Regardless of the arsenal of therapeutic options currently avail-
clinics, and hospitals. The number of studies reporting each outcome of able to tackle this disease, metabolic control still remains subop-
interest varied. A1c was evaluated in 26 studies, of which 24 reported a
timal among patients with type 2 diabetes.
greater reduction in this outcome in the intervention group compared with
• The management of diabetes requires close collaboration between
the control group, with the difference in change between groups rang-
ing from -0.18% to -2.1%. Eighteen studies assessed change in systolic the patient and a multidisciplinary health care team, in which
blood pressure, of which 17 studies reported a greater improvement in this pharmacists may also take part by providing pharmaceutical care
outcome in the intervention group, with the difference in change between programs.
groups varying between -3.3 mmHg and -23.05 mmHg. For diastolic blood
pressure, a greater effect was also observed in the intervention group in What this study adds
14 out of 15 studies, with the difference in change between groups vary-
• Overall, the randomized controlled trials included in this review
ing between -0.21 mmHg and -9.1 mmHg. Thirteen studies described
total cholesterol as an outcome measure, of which 10 reported a greater demonstrated that pharmacist interventions resulted in greater
improvement in this outcome in the intervention group, with the difference improvements in many outcomes when compared with usual
in change between groups ranging from +18.95 mg dL-1 to -32.48 mg dL-1. care, such as in glycosylated hemoglobin, blood glucose, blood
With regard to low-density lipoprotein cholesterol, a greater reduction in pressure, lipid profile, medication adherence, and health-related
this parameter in the intervention group was documented in 12 out of 15 quality of life.
studies, with the difference in change between groups varying between • This review highlights the need for more economic analysis stud-
+7.35 mg dL-1 and -30 mg dL-1. Similarly, favorable data were reported on
ies in order to evaluate the cost-effectiveness of pharmacist inter-
high-density lipoprotein cholesterol in the intervention group in 9 out of 12
studies that assessed this outcome, with the difference in change between ventions and for further research into the intervention elements
groups ranging from -5.8 mg dL-1 to +11 mg dL-1. Data on triglycerides were that contribute the most to the observed effects.
also reported in 12 studies, of which 9 reported a greater reduction in • This review encourages the construction of standard, well-vali-
triglycerides levels in the intervention group, with the difference in change dated tools to ascertain certain outcomes, so that data from differ-
between groups varying between +12 mg dL-1 and -62 mg dL-1. Overall, a ent studies concerning these outcomes can be analyzed together.
beneficial effect on BMI was also described in the intervention group in 12
www.jmcp.org Vol. 22, No. 5 May 2016 JMCP Journal of Managed Care & Specialty Pharmacy 493
Pharmacist Interventions in the Management of Type 2 Diabetes Mellitus: A Systematic Review of Randomized Controlled Trials
T
ype 2 diabetes mellitus is a chronic disease that, if left management, in which the patient must play an active role
uncontrolled, may cause microvascular and macrovas- along with a multidisciplinary health care team.18 In this con-
cular complications in the long term, which are the main text, pharmacists can also contribute positively to diabetes
causes of increased morbidity and mortality and decreased management by providing pharmaceutical care programs,
health-related quality of life among patients.1-5 Unfortunately, which involve working closely with the patient and other
diabetes—particularly type 2 diabetes—is reaching epidemic health care professionals in designing, implementing, and
proportions as its prevalence increases at an alarming rate in monitoring therapeutic plans to achieve specific outcomes that
developed and developing countries.6 Indeed, it was estimated will improve patient quality of life.19 Because of the ongoing
that 415 million people globally suffered from diabetes in 2015, relationships with other health care professionals, pharmacists
and this number is predicted to increase to 642 million people can also serve as a “bridge” between the patients and these
by 2040.6 health care professionals, thereby ensuring continuity of care,
Also of concern is that diabetes and associated complica- which is essential in the management of chronic diseases such
tions impose an increasing and huge economic burden on as diabetes.
national health care systems worldwide. The global health In the past decade or so, there has been a growing body of
expenditure on diabetes was estimated to be at least $673 literature assessing the effectiveness of pharmacist involvement
billion U.S. dollars (USD) in 2015, representing 11.6% of the in the management of diabetic patients in various settings.
world’s total health expenditure, and this amount is expected Nevertheless, only a few systematic reviews have been pub-
to exceed $802 billion USD in 2040.6 For these reasons, diabe- lished on this topic, and some of them evaluated pharmacist
tes is one of the most worrying health problems of the twenty- interventions in patients with type 1 diabetes as well as with
first century and requires immediate attention. type 2 diabetes.20,21 Moreover, some of these reviews focused
Despite the existence of effective therapies and the demon- on a limited number of outcomes,20-22 included other types of
strated benefits derived from tight control of blood glucose and studies (e.g., cohort studies) in addition to randomized con-
other cardiovascular risk factors, such as blood pressure and trolled trials,21,23 excluded some types of pharmacist interven-
serum lipids,7-9 evidence indicates that the achievement of rec- tions,21 and considered a small number of studies.22,23
ommended targets for these factors remains suboptimal among In order to address these shortcomings, the purpose of this
patients with type 2 diabetes.10-12 Lack of adherence to treat- systematic review is to give a global and comprehensive review
ment and other recommendations might explain these find- of the effectiveness of pharmacist interventions in the man-
ings, given that more than 50% of chronically treated patients agement of type 2 diabetes specifically, focusing on clinical,
do not follow the recommended lifestyle changes or do not take humanistic, and economic outcomes and including only the
the prescribed pharmacotherapy.13 The factors that contribute most robust studies, that is, randomized controlled trials, with-
to low levels of adherence include complex treatment regimens, out putting restrictions on the type of pharmacist interventions.
medication side effects, poor patient-provider communication,
patient financial resources and beliefs, psychiatric disorders, ■■  Methods
and memory impairment.14,15 Search Strategy and Inclusion Criteria
In order to address the current challenges of achieving Three electronic databases (PubMed, Cochrane Central Register
therapeutic goals among the diabetes population, new models of Controlled Trials, and Web of Science) were searched from
of health care delivery should be developed and implemented. inception to January 2015. The PubMed search strategy served
Because of their expertise in pharmacotherapy and their acces- as a reference for the development of search strategies for the
sibility in the community, pharmacists are able to build strong remaining databases. The search terms used included medi-
relationships with patients and become a reliable source of cal subject headings and text terms combined with Boolean
information. Thus, pharmacists are in an ideal position to operators. The detailed search strategy used for each database
provide patient education and monitor and promote adherence is provided in the Appendix (available in online article).
to self-care and therapeutic regimens, which have a positive Studies were included in this review if they were randomized
impact on achieving therapeutic outcomes in diabetes.16,17 controlled trials or cluster-randomized controlled trials evaluat-
In addition, because of their extended scientific and techni- ing the effectiveness of interventions delivered only or mainly
cal knowledge, pharmacists are especially alerted to certain by pharmacists and directed at patients with type 2 diabetes in
aspects, such as the occurrence of adverse drug reactions and comparison with usual care. Studies that took place in a commu-
interactions, and specific features associated with aging and nity pharmacy and in outpatient primary care and hospital set-
comorbidities. tings were also included. Studies were included if they reported
Because of the complex nature of diabetes, and as recom- 1 or more of the following outcomes: glycosylated hemoglobin
mended by the American Diabetes Association, a collab- (A1c); blood glucose (fasting, postprandial, or random); blood
orative and integrated team approach should be sought for its pressure; lipid profile (total cholesterol, low-density lipoprotein
494 Journal of Managed Care & Specialty Pharmacy JMCP May 2016 Vol. 22, No. 5 www.jmcp.org
FIGURE 1 Flowchart of Study Selection Process

Identification
Records identified through

Duplicates removed
database searching
(n = 212)
(n = 707)
Screening
Records screened based on titles

Records excluded
and abstracts
(n = 436)
(n = 495)
Eligibility
Full-text articles assessed for eligibility No access to full-text articles

(n = 59) (n = 4)
Full-text articles excluded with reasons

Included
Studies included
(n = 36) Not RCT (n = 5)
Not type 2 diabetes specific population (n = 2)
Outcome data not reported in sufficient detail (n = 1)
Duplicate publications (n = 2)
Review (n = 1)
Study protocol (n = 1)
Conference abstracts (n = 7)
RCT = randomized controlled trial.
cholesterol [LDL], high-density lipoprotein cholesterol [HDL], independently checked the data extracted. No blinding regard-
and triglycerides); body mass index (BMI); 10-year coronary ing the journal or author was done. The data extracted from
heart disease (CHD) risk, medication adherence, health-related each study included authors, publication year, study design,
quality of life (HRQol), or economic outcomes. setting and country where the study took place, sample size,
No limitation regarding language or publication year was patient age and gender, follow-up duration, details of phar-
imposed. Conference abstracts were not included. macist interventions and usual care, inclusion and exclusion
criteria, and study outcomes.
Study Selection The results for the outcome measures included in this review
Two reviewers independently screened all titles and abstracts were summarized as change from baseline to final follow-up in
retrieved from the electronic databases using the prespeci- intervention and control groups. When not reported, the dif-
fied inclusion criteria. Then, the full text of each potentially ference in change between groups was calculated (change from
eligible article was obtained and screened independently by 2 baseline in the intervention group minus change from baseline
reviewers to further assess its suitability for inclusion in this in the control group).
review. Any disagreement was resolved through discussion. In order to facilitate data visualization, bar charts were built
for the following outcomes: A1c, blood glucose, blood pres-
Data Extraction and Synthesis sure, and lipid profile. Since it was not feasible to include all
A single reviewer extracted data from included studies using studies in these charts, 7 to 9 studies with different follow-up
a standardized form (Microsoft Excel format, Microsoft durations and in which various types of interventions were
Corporation, Redmond, WA).24 Subsequently, another reviewer carried out in different countries and health care facilities were
selected for each outcome in order to reflect the variability of and encompassed 1 or more of the following: counseling and
the studies included in this systematic review. In the process of education on diabetes, medication, lifestyle modification, and
selecting studies to be included in the charts, priority was also self-monitoring; reinforcement of medication adherence or
given to studies with larger sample sizes. complications screening; provision of materials such as educa-
tional leaflets and pill boxes; medication review; identification
Risk of Bias Assessment and resolution of drug-related problems; discussions with the
Two reviewers independently assessed the risk of bias in primary care provider regarding pharmacotherapy; adjustment
included studies using the Cochrane risk of bias tool.25 Given of pharmacotherapy; and referrals to other health care profes-
the nature of the interventions studied, participants and the sionals. Two studies mentioned motivational interviews as a
personnel delivering the intervention could not be blinded. technique used to deliver advice to patients.43,53 In most stud-
Therefore, the criteria relative to blinding of participants and ies, the control group received usual care from medical and
personnel were not considered. So, the risk of bias in each nursing staff and/or community pharmacists, depending on the
study was assessed according to the following criteria: suitabil- study setting. In 8 studies, pharmacist interventions were pro-
ity of random sequence generation, concealment of allocation, vided to the control group at the end of the study.44,46,47,52,55-57,59
blinding of outcome assessors, completeness of outcome data, Globally, the included studies involved 5,761 participants. The
selective outcome reporting, and other sources of bias. Each duration of follow-up ranged from 45 days to 24 months. A
risk of bias item was rated as “low risk” if it was unlikely that detailed description of the characteristics of included studies is
a bias would seriously alter the results; “unclear” if it was likely
presented in Table 1.
that a bias would raise some doubt about the results; or “high
risk” if it was likely that a bias would seriously alter the results. Study Risk of Bias
Any disagreement was resolved through discussion. The risk of bias varied among the included 36 studies
(Figure 2). In half (50.0%) of the studies, the allocation
■■  Results
sequence was sufficiently generated. The allocation sequence
Study Selection
was concealed, and outcome assessors were blinded in only
Searching the electronic databases yielded 707 citations
a few studies (13.9% and 2.8%, respectively). In most studies
(Figure 1). After screening titles and abstracts, 59 citations
(97.3%), there was or might have been a risk of bias because of
potentially met the inclusion criteria. Of these, it was not pos-
selective outcome reporting. Only 11 studies (30.6%) reported
sible to retrieve 4 full-text articles, so they were not included.
outcome data completely, and 13 studies (36.1%) were free of
Five studies were excluded because they were not random-
other source of bias.
ized controlled trials. Two studies were also excluded because
the pharmacist interventions were not directed specifically at
Study Outcomes
patients with type 2 diabetes. Another study was also excluded
A1c and Blood Glucose. A1c was considered as an outcome
because the outcome data were not reported in sufficient detail.
measure in 26 studies (Table 2). A1c mean value decreased in
Additionally, 2 duplicate publications, 1 review and 1 study
the intervention group during the follow-up period in all studies,
protocol, and 7 conference abstracts were excluded because
these types of publication were not suitable for inclusion in and 24 studies reported a greater improvement in this outcome
this review. In total, 36 studies met the inclusion criteria and in the intervention group compared with the control group,
were included in this systematic review. Additionally, another 3 some of which are shown in Figure 3A. Of note, in 11 stud-
study reports, found among the databases search results, were ies, the reduction in A1c in the intervention group was greater
also obtained and used to extract data, since they contained than that recorded in the control group by approximately 1%
relevant information for this review. or more.28,31,32,35,41,48-50,54,57,60 In another study, A1c mean value
decreased in the intervention group by 1.7%, while there was
Study Characteristics a 0.1% increase in the control group.51 Nevertheless, 2 studies
Among the included studies, 2 were cluster-randomized presented an imbalance between the baseline A1c mean values
controlled trials in which the participating pharmacies were of both groups.54,60 The difference in A1c change from baseline to
randomly assigned to either the intervention group or the final follow-up between the intervention group and the control
control group.26,27 Eleven studies were conducted in North group ranged from -0.18% to -2.1%. Fifteen studies reported
America,28-38 2 in South America,39,40 5 in Europe,27,41-44 1 in a statistically significant difference in change between the 2
Africa,45 15 in Asia,46-60 and 2 in Australia.26,61 The settings in groups (Table 2).26-28,31,32,34,35,37-39,42,49,53,60,61 However, in 1 study,
which the studies took place included community pharma- there was a significant difference between the baseline A1c mean
cies, primary care clinics or health centers, and hospitals. values of both groups, and the appropriate statistical analysis to
Pharmacist interventions varied across the included studies adjust for this baseline difference was not conducted.60
TABLE 1 Characteristics of Included Studies

No. of Patients (n)a
Lost to Follow-up (n)
Setting/ Age in Years (Mean [SD]) Follow-up
Author, Year Country Gender (%) Duration Pharmacist Intervention Control Outcome Measures
Adepu et al. Community No. of patients (IG/CG): 35/35 6 monthsb Counseling on disease, drugs, and Counseling and Random CBG, KAP regard-
200747 pharmacies/ Lost to follow-up (IG/CG): 3/7 lifestyle modification; provision of information leaflets ing diabetes, DQoL
India Age (IG/CG): 51.45 [12.27]/53.77 [10.35] information leaflet highlighting the at end of study
Gender (IG/CG): 74.3%/62.9% male disease and lifestyle modifications
Adepu and Ari Medicine No. of patients: 240 3 months Education regarding disease, medica- Education at final CBG, medication adherence,
2010 46 outpatient Lost to follow-up: 13 tion, and lifestyle modification follow-up visit KAP regarding diabetes
department at Age: 57
tertiary care Gender: 50.7% male
teaching
hospital/India
Adibe et al. Tertiary hospi- No. of patients (IG/CG): 110/110 12 months Pharmaceutical care including educa- Usual care offered HRQoL, cost-utility
201345 tals/ Nigeria Lost to follow-up (IG/CG): 11/17 tion on diabetes, self-monitoring, by hospitals analysis66
Age (IG/CG): 52.4 [7.6]/52.8 [8.2] medication, lifestyle modification,
Gender (IG/CG): 44.6%/40.0% male counseling, and effective interaction
with health providers
Al Mazroui Military hospi- No. of patients (IG/CG): 120/120 12 months Pharmaceutical care including discus- Usual care from A1c, FBG, SBP, DBP, TC,
et al. tal outpatient Lost to follow-up (IG/CG): 3/3 sions with PCP regarding drug ther- medical and LDL, HDL, TG, BMI,
2009 48 clinic/United Age (IG/CG): 48.7 [8.2]/49.9 [8.3] apy; education on disease and medi- nursing staff 10-year CHD risk, medica-
Arab Emirates Gender (IG/CG): 70.0%/68.3% male cation; and advice on self-monitoring, tion knowledge, medication
physical exercise, diet, medication adherence, adherence to
adherence, and smoking cessation lifestyle advice, HRQoL
Ali et al. Community No. of patients (IG/CG): 25/23 12 months Pharmaceutical care including medi- Usual service from A1c, BG, SBP, DBP, TC,
201241 pharmacies/ Lost to follow-up (IG/CG): 2/0 cine use review; education regarding general practitio- LDL, HDL, TG, BMI,
United Age (IG/CG): 66.4 [12.7]/66.8 [10.2] diabetes and its complications; and ner, practice nurse, DQoL, HRQoL, diabetes
Kingdom Gender (IG/CG): 43.5%/56.5% male counselling on lifestyle modification and community knowledge, SIMS, beliefs
with referral to other health care pro- pharmacy about medicines, others
fessional when appropriate
Chan et al. Diabetes clinic No. of patients (IG/CG): 51/54 9 months Pharmaceutical care program address- Routine medical
A1c,d SBP, DBP, TC, LDL,d
201249 at public hos- Lost to follow-up (IG/CG): 0/0 ing medication adherence, knowledge care HDL, TG, BMI, ACR, CHD
pital/ Hong Age (IG/CG): 63.2 [9.5]/61.7 [11.2] and beliefs, skills, perceived health risk,d stroke risk,d medica-
Kong Gender (IG/CG): 58.8%/51.9% male and cognitive functions, and identifi- tion knowledge,d medica-
cation of DRPs tion adherence,d cost-effec-
tiveness analysis
Choe et al. Primary care No. of patients (IG/CG): 41/39 24 months Evaluation and modification of phar- Regular care, A1c,d rates of LDL
200528 internal medi- Lost to follow-up (IG/CG): 5/10 macotherapy; education on diabetes including regular measurement,d retinal
cine clinic/ Age (IG/CG): 52.2 [11.2]/51.0 [9.0] self-management; and reinforcement follow-up visits examination,d urine
USA Gender (IG/CG): 48.8%/46.1% male of diabetes complications screening with PCP microalbumin screening,
processes monofilament testing for
neuropathy,d and A1c
measurement
Chung et al. Teaching hos- No. of patients (IG/CG): 120/121 12 months Pharmaceutical care including medi- Standard pharmacy A1c, FBG, medication
201450 pital/Malaysia Lost to follow-up: NR cation review; resolution of DRPs; and services adherence
Age (IG/CG): 59.7 [9.5]/58.5 [8.3] education on diabetes, hypertension,
Gender (IG/CG): 58.3%/53.7% female hyperlipidemia, and medications,
emphasizing the importance of medi-
cation adherence
Clifford et al. NR/Australia No. of patients (IG/CG): 99/99 12 months Pharmaceutical care including coun- Usual care, includ- A1c,d FBG,d SBP,d DBP,d
200561 Lost to follow-up (IG/CG): 7/11 seling on medication and lifestyle; ing reinforcement TC, HDL, TG, BMI,d ACR,
Age (IG/CG): 70.5 [7.1]/70.3 [8.3] provision of educational material; and of lifestyle issues 10-year CHD risk, stroke
Gender (IG/CG): 47.8%/56.8% male clinical, biochemical, and medication- risk, physical activity,
related data sent regularly to PCP medication use
Cohen et al. Veterans No. of patients (IG/CG): 53/50 6 months Part of a multidisciplinary diabetes Standard primary A1c, SBP,d LDL, HRQoL,
201129 Affairs Medical Lost to follow-up (IG/CG): 5/2 specific healthy lifestyle education care perceived competence,
Center/USA Age (IG/CG): 69.8 [10.7]/67.2 [9.4] intervention and behavioral and phar- adherence to self-care
Gender (IG/CG): 100.0%/96.0% male macologic interventions, including activities
medication changes
Farsaei et al. Outpatient No. of patients (IG/CG): 87/87 3 months Education on medications, adherence, Visits to PCP every A1c, FBG
201151 clinic/Iran Lost to follow-up: NR and self-management and provision of 3 months
Age (IG/CG): 53.4 [9.8]/52.9 [8.5] a diabetes dairy log and pill box
Gender (IG/CG): 63.2%/68.2% female
Fornos et al. Community No. of patients (IG/CG): 58/56 13 months Pharmacotherapy follow-up program Usual care A1c,d FBG,d SBP,d DBP, TC,d
2006 42 pharmacies/ Lost to follow-up (IG/CG): 2/0 consisting of detection and resolution LDL, HDL, TG, ACR, BMI,
Spain Age (IG/CG): 62.4 [10.5]/64.9 [10.9] of DRPs and diabetes education diabetes and medication
Gender (IG/CG): 57.1%/57.1% female knowledge,d DRPs,d drug
knowledge,d,65 medication
adherenced,65
Ghosh et al. Medicine No. of patients: 22 45 days Counseling on disease, medication, Counseling and FBG, PPBG
201052 department at Lost to follow-up: 4 lifestyle modification, self-monitoring, information leaflets
medical Age: NR and self-care and provision of infor- at the end of the
hospital/India Gender: 72.0% male mation leaflets study
TABLE 1 Characteristics of Included Studies (continued)

Grant et al. Community No. of patients (IG/CG): 118/114 3 months Assessment of medication use; educa- NR Medication adherence,
200330 health center/ Lost to follow-up (IG/CG): 8/13 tion regarding medication; arrange- adherence to diet and
USA Age (IG/CG): 63.3 [12.7]/64.9 [12.1] ments for social services or nutrition exercise program, adher-
Gender (IG/CG): 52.0%/51.0% female consultation; and information ence barriers, medication
forwarded to PCP discrepancies
Jaber et al. Internal medi- No. of patients (IG/CG): 23/22 4 months Pharmaceutical care including edu- Standard medical A1c,d FBG, SBP, DBP, TC,
199631 cine outpatient Lost to follow-up (IG/CG): 6/0 cation on diabetes; counseling on care from PCP LDL, HDL, TG, SCr, CrCl,
clinic/USA Age (IG/CG): 59 [12]/65 [12] medication, instructions on dietary ACR, body weight, HRQoL
Gender (IG/CG): 70.6%/68.2% female regulation, exercise and home blood
glucose monitoring; and evaluation
and adjustment of diabetes medication
Jacobs et al. Ambulatory No. of patients (IG/CG): 195/201 12 months Targeted physical assessment; edu- Usual care A1c,d SBP, DBP, LDL, BMI,
201232 general inter- Lost to follow-up (IG/CG): 22/24 cation on diabetes; laboratory tests medication use, microvas-
nal medicine Age (IG/CG): 62.7 [10.8]/63.0 [11.2] ordering; pharmacotherapy review, cular parameters screeningd
clinic/USA Gender (IG/CG): 68.0%/55.0% male modification and monitorization;
facilitation of self-monitoring of blood
glucose; and reinforcement of dietary
guidelines and exercise
Jarab et al. Outpatient No. of patients (IG/CG): 85/86 6 months Pharmaceutical care including edu- Usual care from A1c,d FBG,d SBP,d DBP,d
201253 diabetes clinic Lost to follow-up (IG/CG): 8/7 cation on diabetes, medication and medical and TC,d LDL,d HDL, TG,d
at teaching Age (IG/CG): 63.4 [10.1]/65.3 [9.2] lifestyle changes; review of prescribed nursing staff BMI, medication adher-
hospital/ Gender (IG/CG): 42.4%/44.2% female treatment; referral to a smoking cessa- ence, adherence to self-care
Jordan tion program; and provision of book- activities
let about diabetes medications and
lifestyle changes
Kjeldsen et al. Community No. of patients (IG/CG): 41/125 6 months Screening for nonadherence; medica- NR BG, SBP,d HRQoL, disease
201543 pharmacies/ Lost to follow-up (IG/CG): 4/23 tion review; patient education regard- knowledge,d medication
Denmark Age (IG/CG): 63.4 [7.8]/62.1 [10.2] ing metabolic syndrome; coaching; adherence, patient satisfac-
Gender (IG/CG): 59.5%/62.4% male support to structure and remember tion with service, others
medication intake; feedback to PCP;
and referral to other health services
Krass et al. Community No. of patients (IG/CG): 176/159 6 months Diabetes service including review of Usual care A1c,d BG, SBP, DBP, TC,
200726 pharmacies/ Lost to follow-up (IG/CG): 27/19 blood glucose self-monitoring; educa- TG, BMI, HRQoL d
Australia Age: 62 [11] tion on disease, medication, and life-
Gender: 51.0% male style; adherence support and detection
of DRPs; and referrals to PCP
Mahwi and Diabetic No. of patients (IG/CG): 65/65 3 months Pharmaceutical care, including educa- Usual medical care A1c, FBG, medication
Obied center/Iraq Lost to follow-up (IG/CG): 3/4 tion sessions adherence, DRPs
201354 Age (IG/CG): 52.0 [7.86]/53.4 [10.81]
Malathy et al. Multispecialty No. of patients (IG/CG): 137/70 3 months Counseling on diabetes and complica- Counseling and PPBG, TC, LDL, HDL, TG,
201155 hospitals and Lost to follow-up: NR tions, blood glucose control, lifestyle information leaflets KAP regarding diabetes
diabetic clinic/ Age (IG/CG): 52.07 [9.47]/51.02 [9.83] changes and foot care, and provision at end of study
India Gender (IG/CG): 71.5%/ 34.3% female of information leaflets about diabetes
and lifestyle changes
Mehuys et al. Community No. of patients (IG/CG): 153/135 24 monthsc Education on diabetes and its com- Usual pharmacist A1c,d FBG, medica-
201127 pharmacies/ Lost to follow-up (IG/CG): 5/3 plications, medication and healthy care tion adherence, diabetes
Belgium Age (IG/CG): 63.0/62.3 lifestyles, facilitation of medication knowledge,d adherence to
Gender (IG/CG): 51.0%/53.7% male adherence, and reminders about self-care activitiesd
annual eye and foot examinations
Mourão et al. Primary health No. of patients (IG/CG): 65/64 6 months Pharmaceutical care, including Usual health care A1c,d FBG,d SBP,d DBP, TC,d
201339 care units/ Lost to follow-up (IG/CG): 12/9 identification of DRPs; proposals for LDL,d HDL,d TG,d BMI,
Brazil Age (IG/CG): 60.0 [10.2]/61.3 [9.9] pharmacotherapy changes forwarded medication use,d DRPs
Gender (IG/CG): 68.0%/66.0% female to PCP; and education on diabetes,
nonpharmacological issues, and phar-
macological treatments
Odegard et al. Medicine No. of patients (IG/CG): 43/43 12 months Development and implementation of Normal care from A1c, medication appro-
200533 clinics/USA Lost to follow-up (IG/CG): 4/7 diabetes care plan and regular com- PCP priateness, medication
Age (IG/CG): 51.6 [11.6]/51.9 [10.4] munication with patient and PCP on adherence
Gender (IG/CG): 48.0%/38.0% female patient’s diabetes care progress
Plaster et al. Community No. of patients (IG/CG): 38/36 6 months Pharmaceutical care program, includ- Standard care FBG, SBP, DBP, TC, LDL,
201240 health center/ Lost to follow-up (IG/CG): 4/7 ing promotion of correct use of medi- HDL, TG, WC, body
Brazil Age: NR cation, provision of educative instruc- weight, 10-year CHD risk,
Gender (IG/CG): 70.6%/60.0% female tional material about lifestyle modifi- negative outcomes associ-
cations, and interaction with PCP ated with medication
Ramanath and General No. of patients (IG/CG): 57/56 3 months Educational materials and counseling Educational materi- FBG, PPBG, medication
Santhosh medicine Lost to follow-up (IG/CG): 5/8 als and counseling adherence, KAP regarding
201156 department at Age: NR at last follow-up diabetes, HRQoL
tertiary care Gender (IG/CG): 67.3%/50.0% male
hospital/India
TABLE 1 Characteristics of Included Studies (continued)

Rothman et al. General inter- No. of patients (IG/CG): 112/105 12 months Education, counseling, and Usual care from A1c,d SBP,d DBP,d TC, body
200534 nal medicine Lost to follow-up (IG/CG): 13/10 medication management PCP weight, aspirin and statin
practice/USA Age (IG/CG): 54 [13]/57 [11] use, diabetes knowledge,
Gender (IG/CG): 56.0%/56.0% female diabetes treatment satisfac-
tion, use of clinical services,
adverse events
Sarkadi and Community No. of patients (IG/CG): 39/38 24 months Educational program regarding Educational A1c
Rosenqvist pharmacies/ Lost to follow-up (IG/CG): 6/7 disease, lifestyle changes, and self- program at end
2004 44 Sweden Age (IG/CG): 66.4 [7.9]/66.5 [10.7] monitoring and provision of of study
Gender: NR educational materials
Scott et al. Community No. of patients (IG/CG): 76/73 9 months Education on diabetes, medication, Standard diabetes A1c,d SBP,d DBP, LDL,d
200635 health center/ Lost to follow-up (IG/CG): 5/4 lifestyle adjustments, self-monitoring, care HDL, BMI, body weight,
USA Age: NR and self-care; medication review; and DQoL, aspirin use,
Gender (IG/CG): 57.9%/64.4% female pharmacotherapy recommendations influenza vaccination
made to PCP
Simpson et al. Primary care No. of patients (IG/CG): 131/129 12 months Medication assessments, limited his- A1c, SBP,d DBP,d TC, LDL,
Usual care from
201136 clinics/Canada Lost to follow-up (IG/CG): 21/16 tory and physical examinations, and HDL, 10-year CHD risk,d
primary care team
Age (IG/CG): 58.8 [11.1]/59.4 [12.1] guideline-concordant recommenda- antihypertensive medica-
Gender (IG/CG): 56.5%/58.1% female tions discussed with PCP to optimize tion changes, number of
medication management health care-related con-
tacts, cost-effectiveness
analysis67
Sriram et al. General No. of patients (IG/CG): 60/60 8 monthsb Pharmaceutical care, including educa- Pharmaceutical A1c, FBG, BMI, DQoL,
201157 medicine Lost to follow-up: NR tion on diabetes, counseling on medi- care at end of study satisfaction with diabetes
department at Age (IG/CG): 53.65 [2.38]/57.98 [2.62] cation and lifestyle modification, and treatment
tertiary care Gender (IG/CG): 50.0%/50.0% female provision of an information leaflet,
teaching hos- a diabetic diet chart, and a diabetic
pital/India diary
Suppapitiporn Endocrine No. of patients (IG/CG): 180/180 6 months IG divided into 4 groups: all groups Conventional A1c, FBG
et al. clinic at hospi- Lost to follow-up: NR received diabetes drug counseling; treatment
200558 tal/Thailand Age (IG/CG): 61.4 [10.6]/59.9 [11.5] additionally, 1 group received a diabe-
Gender (IG/CG): 67.2%/64.4% female tes booklet, 1 group received special
medication containers, and 1 group
received both
Taveira et al. Veteran Affairs No. of patients (IG/CG): 64/54 4 months Part of multidisciplinary education Usual care from A1c,d SBP, DBP,d LDL,
201038 Medical Lost to follow-up (IG/CG): 6/3 intervention regarding self-care PCP non-HDL, BMI, tobacco
Center/USA Age (IG/CG): 62.2 [10.3]/66.8 [10.2] behaviors and behavioral and phar- use, adherence to self-care
Gender (IG/CG): 91.4%/100.0% male macological interventions, including behaviors
medication changes
Taveira et al. Veteran Affairs No. of patients (IG/CG): 44/44 6 months Part of a multidisciplinary educa- Standard care from A1c,d SBP, LDL, non-HDL,
201137 Medical Lost to follow-up (IG/CG): 0/2 tion intervention regarding self-care PCP 10-year CHD risk, tobacco
Center/USA Age (IG/CG): 60.2 [9.3]/61.4 [9.9] behaviors and behavioral and phar- use, adherence to self-
Gender (IG/CG): 100.0%/95.5% male macological interventions, including care behaviors, perceived
medication changes competence, depression
symptoms
Venkatesan Community No. of patients (IG/CG): 19/20 8 months Counselling about diabetes, lifestyle Counseling at end FBG, BMI, diabetes care
et al. pharmacies/ Lost to follow-up (IG/CG): 0/NR modifications, and medication and of study profile, diabetes knowledge
201259 India Age (IG/CG): 51.47 [9.99]/57.05 [12.06] provision of educational material
Wishah et al. Outpatient No. of patients (IG/CG): 52/54 6 months Pharmaceutical care, including opti- Usual care from A1c,d FBG,d TC, LDL, HDL,
2014 60 diabetes clinic Lost to follow-up (IG/CG): 2/3 mization of drug therapy; education medical and TG, BMI, diabetes knowl-
at teaching Age (IG/CG): 52.9 [9.6]/53.2 [11.2] and counseling on diabetes and medi- nursing staff edge, medication adher-
hospital/ Gender (IG/CG): 61.5%/51.9% female cation; enhancement of adherence to ence, adherence to self-care
Jordan medication and self-care activities; activities
and provision of educational leaflet
and brochures
a
Allocated to intervention group and control group.
b
Study duration.
c
24-month follow-up for A1c and FBG; 6-month follow-up for the other outcomes.
d
The study reported a statistically significant difference in change between groups.
A1c = glycosylated hemoglobin; ACR = albumin-to-creatinine ratio; BG = blood glucose; BMI = body mass index; CBG = capillary blood glucose; CG = control group; CHD = coronary heart disease;
CrCl = creatinine clearance; DBP = diastolic blood pressure; DQoL = diabetes quality of life; DRPs = drug-related problems; FBG= fasting blood glucose; HDL = high-density lipoprotein cholesterol;
HRQoL = health-related quality of life; IG = intervention group; KAP = knowledge, attitude, and practice; LDL = low-density lipoprotein cholesterol; NR = not reported; PCP = primary care pro-
vider; PPBG = postprandial blood glucose; SBP = systolic blood pressure; SCr = serum creatinine; SD = standard deviation; SIMS = satisfaction with information received about medicines; TC = total
cholesterol; TG = triglycerides; WC = waist circumference.
FIGURE 2 Risk of Bias in Included Studies Presented as Percentage Across All Studies
Random sequence generation
Allocation concealment
Blinding of outcomes assessment Low risk

Unclear
Incomplete outcome data High risk
Selective reporting
Other sources of bias

0 10 20 30 40 50 60 70 80 90 100
Percentage
Regarding blood glucose, 23 studies reported this parameter intervention group and the control group (Table 2).34,36,38,53,61
as an outcome measure (Table 2). In all studies, there was a The difference in change between the 2 groups ranged from
decrease in blood glucose (fasting, postprandial, or random) -0.21 mmHg to -9.1 mmHg.
in the intervention group from baseline to final follow-up, and Lipid Profile. Thirteen studies described total cholesterol as an
20 studies reported a greater improvement in this outcome in outcome measure (Table 3). In all of these studies, there was a
the intervention group when compared with the control group. reduction in the intervention group from baseline to final follow-
Nevertheless, the difference in change (ranged from -5.9 mg dL-1 up, and 10 studies reported a greater improvement in this out-
to -66.87 mg dL-1) between both groups was reported as sta- come in the intervention group in comparison with the control
tistically significant in only 5 studies (Table 2).39,42,53,60,61 The group. However, the difference in change (ranged from +18.95
greater reduction in fasting blood glucose in the intervention mg dL-1 to -32.48 mg dL-1) between the 2 groups was reported as
group in comparison with the control group is shown in Figure statistically significant in only 3 studies (Table 3).39,42,53
3B for some studies. Regarding LDL cholesterol, 15 studies reported data on
Blood Pressure. Eighteen studies evaluated the change in sys- this outcome, and all of them demonstrated a decrease in the
tolic blood pressure during the course of the study (Table 2). In intervention group from baseline to final follow-up (Table 3).
all of them, there was a reduction in mean systolic blood pres- Twelve studies reported a greater reduction in this outcome in
sure in the intervention group from baseline to final follow-up, the intervention group compared with the control group. For
and 17 studies reported a greater improvement in this outcome this parameter, the difference in change between the groups
in the intervention group compared with the control group, ranged from +7.35 mg dL-1 to -30 mg dL-1 and was reported as
some of which are shown in Figure 4A. For systolic blood statistically significant in only 4 studies (Table 3).35,39,49,53
pressure, the difference in change between the groups ranged Among the 12 studies that reported HDL cholesterol as an
from -3.3 mmHg to -23.05 mmHg and was shown to be sta- outcome measure (Table 3), 10 studies described an increase
tistically significant in only 9 studies (Table 2).29,34-36,39,42,43,53,61 in the intervention group from baseline to final follow-
Nevertheless, in 1 of these studies, there was a difference in up,35,36,39-41,48,49,55,60,61 1 study observed a decrease,53 and in
baseline mean values between the intervention and the control another study this parameter remained constant.42 Nine studies
groups, and the appropriate statistical analysis to adjust for this reported a greater increase in this outcome in the intervention
baseline difference was not conducted.39 group when compared with the control group. Nevertheless,
As for diastolic blood pressure, 15 studies reported data the difference in change between both groups was shown to be
on this outcome (Table 2). In all studies, mean diastolic blood statistically significant in only 1 study (P = 0.020).39 The differ-
pressure decreased in the intervention group during the follow- ence in change between the 2 groups ranged from -5.8 mg dL-1
up period, and 14 studies reported a greater reduction in this to +11 mg dL-1.
outcome in the intervention group in comparison with the con- Finally, 12 studies reported data on triglycerides (Table 3).
trol group, some of which are shown in Figure 4B. However, Eleven of them described a reduction in the intervention group
only a third of the studies revealed a statistically significant dif- from baseline to final follow-up, while in 1 study, there was
ference in change from baseline to final follow-up between the an increase in the triglycerides mean value.41 Nine studies
TABLE 2 Studies with Results for A1c, Blood Glucose, and Blood Pressure
A1c (%) Blood Glucose (mg dL-1) Blood Pressure (mmHg)
Difference Difference Difference
in Change in Change in Change
Change from Baseline to Between Change from Baseline to Between Change from Baseline Between
Author, Year Final Follow-up Groups a Final Follow-up Groupsa to Final Follow-up Groups a
Adepu et al. Random capillary blood glucose
200747 NR – IG: 198.31 to 142 (P < 0.001) -53.71b NR –
CG: 173.6 to 171 (P > 0.05)
Adepu and Ari Capillary blood glucose
2010 46 NR – IG: significant decrease (P = 0.001) – NR –
CG: nonsignificant decrease (P = 0.798)
Al Mazroui IG: 8.5 to 6.9 (P < 0.001) Fasting blood glucose SBP IG: 131.4 to 127.2 (P < 0.001)
-3.7b
et al. CG: 8.4 to 8.3b IG: 194.94 to 140.04 CG: 132.6 to 132.1b
-1.5b -40.86b
200948 CG: 184.68 to 170.64 DBP IG: 85.2 to 76.3 (P < 0.001)
-9.1b
CG: 83.9 to 84.1b
Ali et al. IG: 8.2 to 6.6 (P < 0.001) IG: 158.40 to 123.84 (P < 0.001) SBP IG: 146.26 to 126.17 (P < 0.001)
-23.05b
201241 CG: 8.1 to 7.5 (P = 0.033) CG: 171.54 to 162.72 (P = 0.097) CG: 136.22 to 139.17 (P = 0.450)
-1.01b -25.74b
DBP IG: 87.13 to 81.04 (P = 0.010)
-2.13b
CG: 85.65 to 81.7 (P = 0.090)
Chan et al. IG: 9.7 to 8.13 SBP IG: 141 to 134.5 -3.3
201249 CG: 9.5 to 9.1 -1.17 CG: 138 to 134.8 (P = 0.34)
NR –
(P < 0.001) DBP IG: 75 to 72.2 -2.1
CG: 74 to 73.3 (P = 0.23)
Choe et al. IG: 10.1 to 8.0 -1.2
NR – NR –
200528 CG: 10.2 to 9.3 (P = 0.03)
Chung et al. IG: 9.6 to 8.2 (P < 0.001) Fasting blood glucose
201450 CG: 9.5 to 9.3 (P = 0.265) -1.2b IG: 169.2 to 135.0 (P < 0.001) -43.2b NR –
CG: 165.6 to 174.6 (P = 0.257)
Clifford et al. IG: 7.5 to 7.0 Fasting blood glucose SBP IG: 157 to 143 -7
200561 CG: 7.1 to 7.1 -0.5 IG: 158.4 to 144.0 -21.6 CG: 156 to 149 (P = 0.024)
(P = 0.002) CG: 145.8 to 153.0 (P < 0.001) DBP IG: 77 to 72 -3
CG: 77 to 75 (P = 0.043)
Cohen et al. IG: 7.8 to 7.39 -0.21 SBP IG: 136.1 to 126.91 -8.39
NR –
201129 CG: 8.1 to 7.9 (NS)b CG: 136.1 to 135.3 (Sig.)b
Farsaei et al. IG: 9.3 to 7.5 (P < 0.001) Fasting blood glucose
201151 CG: 8.9 to 9.0 (P = 0.317) -1.8b IG: 176.6 to 145.8 (P < 0.001) -26.3b NR –
CG: 170.4 to 165.9 (P = 0.528)
Fornos et al. IG: 8.4 to 7.9 (P < 0.001) Fasting blood glucose SBP IG: 143 to 135 (P < 0.001) -10
2006 42 CG: 7.8 to 8.5 (P < 0.001) -1.2 IG: 172 to 153 (P < 0.001) -27 CG: 148 to 150 (P = 0.577) (P < 0.001)
(P < 0.001) CG: 160 to 168 (P = 0.042) (P <0.001) DBP IG: 80.2 to 78.2 (P = 0.050) -1.9
CG: 82.2 to 82.1 (P = 0.686) (NS)b
Ghosh et al. Fasting blood glucose
201052 IG: 227 to 180 (P < 0.05) -15b
CG: 227 to 195b
NR – NR –
Postprandial blood glucose
IG: 291 to 194 (P < 0.01) -35b
CG: 291 to 229b
Jaber et al. IG: 11.5 to 9.2 (P = 0.015) Fasting blood glucose SBP IG: 147 to 140 (P=0.07)
199631 CG: 12.2 to 12.1b -2.1 IG: 199.8 to 153.0 (P = 0.003) -14.4 CG: NR
–
(P = 0.021) CG: 228.6 to 198.0b (NS)b DBP IG: 88 to 82 (P = 0.07)
CG: NR
Jacobs et al. IG: 9.5 to 7.7 SBP IG: 142.5 to 132.5
-10.6b
201232 CG: 9.2 to 8.4 -1.0 CG: 134.8 to 135.4
NR –
(P < 0.05) DBP IG: 79.4 to 72.0
-6.7b
CG: 78.3 to 77.6
Jarab et al. IG: 8.5 to 7.7 Fasting blood glucose SBP IG: 132 to 126.2 -6.9
201253 CG: 8.4 to 8.5 -0.9 IG: 225.0 to 183.6 -57.6 CG: 134 to 135.1 (P = 0.035)
(P = 0.019) CG: 210.6 to 226.8 (P = 0.014) DBP IG: 85 to 77.9 -8.9
CG: 85 to 86.8 (P = 0.026)
Kjeldsen et al. IG: nonsignificant decrease SBP IG: 138 to 131.3 -5.3
NR – –
201543 CG: NR CG: 139 to 137.6 (P = 0.02)
Krass et al. IG: 8.9 to 7.9 IG: 169.2 to 153.0 (P < 0.001) SBP IG: 135 to 133 -4.8
200726 CG: 8.3 to 8.0 -0.7 CG: NR CG: 133 to 135 (P = 0.06)
–
(P < 0.01) DBP IG: 79 to 77 -1.1
CG: 77 to 76 (P = 0.52)
TABLE 2 Studies with Results for A1c, Blood Glucose, and Blood Pressure (continued)
A1c (%) Blood Glucose (mg dL-1) Blood Pressure (mmHg)
Change from Baseline to Between Change from Baseline to Between Change from Baseline to Between
Author, Year Final Follow-up Groups a Final Follow-up Groupsa Final Follow-up Groupsa
Mahwi and IG: 11.53 to 9.2 (P < 0.001) Fasting blood glucose
Obied CG: 9.97 to 9.5 (P = 0.341) -1.86b IG: 249.4 to 196.4 (P = 0.001) -37.3b NR –
201354 CG: 211.1 to 195.4 (P = 0.196)
Malathy et al. Postprandial blood glucose
201155 NR – IG: 237.0 to 204.47 (P < 0.001) -14.76b NR –
CG: 229.17 to 211.4b
Mehuys et al. IG: 7.7 to 7.1 (P < 0.001) c
Fasting blood glucosec
-0.5 -5.9
201127 CG: 7.3 to 7.2 (P = 0.162)c IG: 154.1 to 138.8 (P < 0.001) NR –
(P = 0.009) (P = 0.193)
CG: 153.9 to 145.8 (P = 0.004)
Mourão et al. IG: 9.9 to 9.3 Fasting blood glucose SBP IG: 152.9 to 140.8 -9.2
201339 CG: 9.5 to 10.2 -1.3 IG: 177.7 to 156.3 -34.8 CG: 140.4 to 137.5 (P = 0.013)
(P = 0.001) CG: 174.4 to 187.8 (P = 0.007) DBP IG: 85.1 to 82.1 -0.5
CG: 82.9 to 80.4 (P = 0.809)
Odegard et al. IG: 10.2 to 8.2 –
NR – NR –
200533 CG: 10.6 to NR (P = 0.61)
Plaster et al. Fasting blood glucose SBP IG: 138 to 131
-10b
201240 IG: 198 to 130 (P < 0.01) CG: 138 to 141
NR – CG: 181 to 173b -60b
DBP IG: 80 to 78
-3b
CG: 80 to 81
Ramanath and Fasting blood glucose
Santhosh IG: 151.13 to 132.50 -26.34b
201156 CG: 138.50 to 146.21
NR – NR –
Postprandial blood glucose
IG: 219.69 to 178.87 -41.59b
CG: 200.42 to 201.19
Rothman et al. IG: 11.0 to 8.5 SBP IG: 140 to 133 -9
200534 CG: 11.0 to 9.4 -0.8 CG: 137 to 139 (P = 0.008)
NR –
(P = 0.05) DBP IG: 82 to 78 -5
CG: 80 to 81 (P = 0.02)
Sarkadi and IG: decreased significantly
Rosenqvist by 0.4%
– NR – NR –
2004 44 CG: increased
nonsignificantly
Scott et al. IG: 8.8 to 7.08 (P = 0.003) SBP IG: 130.0 to 126.6 -5.5
200635 CG: 8.7 to 8.0 (P < 0.05) -1.0 CG: 130.7 to 132.8 (P = 0.023)
NR –
(P < 0.05) DBP IG: 79.3 to 75.9
-2.0b
CG: 79.6 to 78.2
Simpson et al. IG: 7.5 to 7.35 SBP IG: 130.4 to 123.0 (P < 0.001) -4.9
201136 CG: 7.3 to 7.33 -0.18 CG: 128.3 to 125.8 (P = 0.06) (P = 0.01)
NR –
(NS)b DBP IG: 74.4 to 72.1 (P < 0.05) -2.9
CG: 73.9 to 74.5b (P < 0.05)
Sriram et al. IG: 8.44 to 6.73 (P < 0.01) Fasting blood glucose
201157 CG: 9.03 to 8.31 (P > 0.05) -0.99b IG: 195.57 to 107.25 (P < 0.01) -51.89b NR –
CG: 186.00 to 149.57 (P > 0.05)
Suppapitiporn IG: 8.16 to 7.91 Fasting blood glucose
et al. CG: 8.01 to 8.80 -1.04b IG: 152.36 to 145.20 -16.16b NR –
200558 CG: 150.16 to 159.16
Taveira et al. IG: 8.1 to 7.2 (P<0.05) SBP IG: 131.1 to 123.8 (P < 0.05) -5.6
201038 CG: 7.9 to 7.9b -0.9 CG: 137.2 to 135.5b (NS)b
NR –
(P < 0.05) DBP IG: 74.4 to 67.9 (P < 0.05) -7.5
CG: 74.2 to 75.2b (P < 0.05)
Taveira et al. IG: 8.3 to 7.4 (P < 0.05) -0.9 SBP IG: 130.6 to 123.4 (P < 0.05) -8.9
NR –
201137 CG: 8.5 to 8.4b (P < 0.05) CG: 125.2 to 127.0b (NS)b
Venkatesan Fasting blood glucose
et al. NR – IG: 155.58 to 108.10 -66.87b NR –
201259 CG: 150.30 to 169.70
Wishah et al. IG: 8.9 to 7.2 Fasting blood glucose
-1.4 -51.4
2014 60 CG: 8.2 to 7.9 IG: 180.2 to 126.9 NR –
(P < 0.05) (P < 0.05)
CG: 159.6 to 158.0
a
Negative values indicate IG had greater decrease; positive values indicate CG had greater decrease.
b
P value not reported.
c
Over the 6-month follow-up period.
A1c = glycosylated hemoglobin; CG = control group; DBP = diastolic blood pressure; IG = intervention group; NR = not reported; NS = not significant; SBP = systolic blood pressure; Sig. = significant.
FIGURE 3 Effect of Pharmacist Interventions on A1c and Fasting Blood Glucose Compared with Control Group
A. A1c B. Fasting Blood Glucose
Control Group Intervention Group Control Group Intervention Group
P < 0.001 Chan et al., 201249 NR Al Mazroui et al., 200948

P = 0.03 Choe et al., 200528 NR Chung et al., 201450
NR Farsaei et al., 2011 51
NR Farsaei et al., 201151
P < 0.05 Jacobs at al., 201232
P = 0.014 Jarab et al., 201253
P < 0.01 Krass et al., 200726
NR Mahwi and Obied, 201354
NR Mahwi and Obied, 201354
Scott et al., 200635 NR Sriram et al., 201157
P < 0.05
NS Simpson et al., 201136 NR Suppapitiporn et at., 200558
NR Suppapitiporn et al., 200558 NR Venkatesan et al., 201259
-3 -2 -1 0 1 2 3 -100 -75 -50 -25 0 25 50 75 100

Change in A1c (%) Change in fasting blood glucose (mg dL-1)
from baseline to final follow-up from baseline to final follow-up
A1c = glycosylated hemoglobin; NR = P value not reported; NS = not statistically significant.
reported a greater improvement in this outcome in the inter- for the Hong Kong population also includes the duration of
vention group in comparison with the control group. Yet, the diabetes,49 while the Framingham prediction model considers
difference in change between the 2 groups was reported as the presence or absence of diabetes as a variable.63 In addi-
statistically significant in only 2 studies (Table 3).39,53 This dif- tion to the duration of diabetes, the UKPDS risk engine also
ference ranged from +12 mg dL-1 to -62 mg dL-1. incorporates the A1c mean values.62 Regarding blood pressure,
For all 4 parameters of the lipid profile, the greater improve- the UKPDS risk engine and the British National Formulary
ment observed in the intervention group compared with the prediction charts include systolic blood pressure,62,64 while
control group is shown in Figure 5A-D for some studies. the Framingham prediction model integrates systolic and
diastolic blood pressure.63 As for the lipid profile, the UKPDS
Body Mass Index. Fourteen studies described BMI as an out- risk engine and the British National Formulary prediction
come measure (Table 3). In all but 2 studies, mean BMI decreased charts consider the total cholesterol/HDL cholesterol ratio as
in the intervention group from baseline to final follow-up. a variable,62,64 while the Framingham prediction model only
These studies also reported a greater reduction in this group includes HDL cholesterol,63 and the equation validated for the
in comparison with the control group.26,35,38,41,42,48,49,53,57,59-61 Hong Kong population incorporates non-HDL cholesterol.49 In
Nevertheless, only 1 study revealed a statistically significant addition to the previously mentioned variables, the UKPDS risk
difference in change between the intervention group and the engine also considers ethnicity as a variable,62 and the equation
control group (P = 0.005).61 The difference in change between validated for the Hong Kong population integrates glomerular
the 2 groups ranged from +0.4 kg m-2 to -2.77 kg m-2. filtration rate and urinary albumin to creatinine ratio.49
10-Year CHD Risk. CHD risk was predicted among study All 6 studies recorded a decrease in CHD risk in the inter-
participants in 6 studies (Table 3). The method used to esti- vention group from baseline to final follow-up and reported a
mate this risk varied between studies. Three studies used greater improvement in this group compared with the control
the United Kingdom Prospective Diabetes Study (UKPDS) group. In the 3 studies that used the UKPDS risk engine,
risk engine,36,37,61 1 study used the Framingham prediction the 10-year CHD risk decreased 4.8%, 2.7%, and 4.9%,
method,40 1 study used the British National Formulary pre- respectively, in the intervention group during the follow-up
diction charts and the Framingham prediction method,48 and period. In the 2 studies that used the Framingham prediction
1 study used an equation specifically validated for the Hong method, there was a decrease of 2.9% and 8.0%, respectively,
Kong population.49 All 4 models incorporated the variables in the intervention group throughout the follow-up period. In
age, sex, and smoking status.49,62-64 The equation validated comparison with the control group, the difference in change
FIGURE 4 Effect of Pharmacist Interventions on Systolic Blood Pressure and Diastolic Blood Pressure
Compared with Control Group
A. Systolic Blood Pressure B. Diastolic Blood Pressure
P = 0.34 Chan et al., 201249 P = 0.23 Chan et al., 201249
P < 0.001 Fornos et al., 200642 NS Fornos et al., 200642
NR Jacobs et al., 201232 NR Jacobs et al., 201232
P = 0.035 Jarab et al., 201253 P = 0.026 Jarab et al., 201253
P = 0.06 Krass et al., 200726 P = 0.52 Krass et al., 200726
P = 0.023 Scott et al., 200635 NR Scott et al., 200635
P = 0.01 Simpson et al., 201136 P < 0.05 Simpson et al., 201136
NS Taveira et al., 201038 P < 0.05 Taveira et al., 201038
-15 -10 -5 0 5 10 15 -10.0 -7.5 -5.0 -2.5 0 2.5 5.0 7.5 10.0
Change in systolic blood pressure (mmHg) Change in diastolic blood pressure (mmHg)
NR = P value not reported; NS = not statistically significant.
between the 2 groups was reported as statistically significant and because of the difference in detail of the reported data, it
in only 2 studies (Table 3).36,49 Because the studies used dif- is not possible to define a range for the difference in change
ferent methods to assess this risk, it is not possible to define a between the groups across all studies.
range for the difference in change between the groups across Health-Related Quality of Life. Eleven studies considered
all studies. However, among the studies that used the UKPDS HRQoL as an outcome measure (Table 4). Various tools were
risk engine, the difference in change between the groups used to assess this outcome. Seven studies used generic
ranged from -1.5% to -5.1%, and in the 2 studies that used the tools (e.g., the 36-Item Short Form Health Survey and the
Framingham prediction method, this difference was -3.0% and EuroQoL-5 Dimension questionnaire), which can be applied
-12.0%, respectively. in different health conditions and diseases.26,29,31,43,45,48,56 Three
Medication Adherence. Medication adherence was evaluated studies used tools specifically developed for diabetes (e.g., the
in 13 studies (Table 4). The methods used to measure this Audit of Diabetes-Dependent Quality of Life questionnaire and
outcome among study participants varied between studies. the Diabetes Quality of Life questionnaire),35,47,57 and 1 study
Self-reported adherence was used as the only method in almost used generic and diabetes-specific tools.41
all studies, while 2 studies used pill count or prescription refill Most studies reported an improvement in HRQoL (overall
rate in combination with self-reported adherence.27,54 or subdomain scores) in the intervention group from baseline
Eleven studies revealed an improvement in medication to final follow-up, which was greater than that observed in
adherence in the intervention group from baseline to final fol- the control group. In the studies that used the EuroQoL-5
low-up. In 2 studies that used the same method for the assess- Dimension questionnaire to assess HRQoL among study
ment of medication adherence (Morisky Medication Adherence patients, the effect was mixed. There was an improvement of
Scale), the proportion of nonadherent patients decreased 45.5% 0.06 in the intervention group during the follow-up period
and 58.4%, respectively, in the intervention group during the in 1 study,43 while there was a decrease of 0.04 in the other
follow-up period.53,54 In 8 studies, a greater improvement in study.26 Only 1 study reported a statistically significant differ-
medication adherence was observed in the intervention group ence in change between the intervention group and the control
when compared with the control group, but only 2 studies group (P = 0.02).26 Because of the wide variability of tools used
reported a statistically significant difference (Table 4).49,65 Given to assess this outcome and because of the difference in detail
the wide variability of methods used to measure this outcome in reporting overall or subdomains scores, it is not possible to
TABLE 3 Studies with Results for Lipid Profile, Body Mass Index, and 10-Year CHD Risk
Lipid Profile (mg dL-1) Body Mass Index (kg m-2) 10-Year CHD Risk
Author, Change from Baseline to Between Change from Baseline to Between Change from Baseline to Between
Year Final Follow-up Groupsa Final Follow-up Groupsa Final Follow-up Groupsa
Al Mazroui TC IG: 203.40 to 172.85 -32.48b
Framingham risk prediction score
et al. CG: 203.79 to 205.72
IG: 10.6% to 7.7% (P < 0.001)
2009 48 LDL IG: 137.28 to 117.56 -24.75b -3.0%b
CG: 11.4% to 11.5% (P > 0.05)
CG: 134.57 to 139.60 IG: 28.34 to 27.29 (P < 0.005)
-1.06 b
BNF risk prediction
HDL IG: 46.40 to 51.04 (P < 0.05) 4.25b CG: 27.98 to 27.99 (P > 0.05)
CG: 46.02 to 46.40b (% of patients at low risk) 28.2%b,c
IG: 63.3% to 85.5%
TG IG: 141.71 to 110.71 -47.83 b
CG: 65.0% to 59.0%
CG: 137.28 to 154.11
Ali et al. TC IG: 160.48 to 159.32 (P = 0.811) 18.95b
201241 CG: 141.53 to 121.42 (P < 0.001)
LDL IG: 90.87 to 76.18 (P = 0.009) 7.35b
CG: 69.99 to 48.34 (P < 0.001) IG: 30.84 to 26.98 (P < 0.001)
-2.77b NR –
HDL IG: 46.02 to 56.46 (P = 0.001) 8.12b CG: 29.82 to 28.73 (P = 0.059)
CG: 46.40 to 48.34 (P = 0.575)
TG IG: 119.57 to 134.63 (P = 0.140) -15.05b
CG: 127.54 to 157.65 (P = 0.940)
Chan et al. TC IG: 170.15 to 155.07 -12.37
201249 CG: 182.14 to 179.43 (P = 0.08)
LDL IG: 101.32 to 87.39 -12.76 Score obtained from a validated CHD risk
CG: 107.12 to 105.96 (P = 0.026) IG: 25.2 to 25.04 -0.23 equation for Hong Kong population -0.11
HDL IG: 42.15 to 42.304 0.078 CG: 26.2 to 26.27 (P = 0.24) IG: 2.16 to 2.05 (P < 0.001)
CG: 44.47 to 44.548 (P = 0.93) CG: 2.17 to 2.17
TG IG: 154.11 to 134.63 0.00
CG: 168.28 to 148.80 (P = 0.99)
Clifford et al. TC IG: 193.4 to 181.7 -3.9
200561 CG: 189.5 to 181.7 (P = 0.14)
UKPDS risk engine score
HDL IG: 46.02 to 47.18 1.93 IG: 30.0 to 29.4 -0.7
IG: 25.1% to 20.3% (P = 0.002) -5.1%b
CG: 46.02 to 45.24 (P = 0.07) CG: 30.0 to 30.1 (P = 0.005)
CG: 26.1% to 26.4% (P = 0.17)
TG IG: 150.6 to 97.4 -53.1
CG: 141.7 to 141.7 (P = 0.09)
Cohen et al. LDL IG: 96.1 to 86.7 2.13
NR – NR –
201129 CG: 110.7 to 99.17 (NS)b
Fornos et al. TC IG: 222 to 202 (P < 0.001) -19
2006 42 CG: 218 to 217 (P=0.716) (P = 0.0054)
LDL IG: 141 to 126 (P < 0.001) -12
CG: 136 to 133 (P = 0.488) (NS)b IG: 31.0 to 30.1 (P = 0.018) -0.6
NR –
HDL IG: 48.5 to 48.5 (P = 0.981) 0.2 CG: 31.7 to 31.4 (P = 0.650) (NS)b
CG: 49.9 to 49.7 (P = 0.887) (NS)b
TG IG: 167 to 138 (P = 0.181) -32
CG: 168 to 171 (P = 0.229) (NS)b
Jaber et al. No significant changes within
– NR – NR –
199631 or between IG and CG
Jacobs et al. LDL IG: 121.5 to 93.7 -17.8b IG: 32.8 to 33.2
0.6b NR –
201232 CG: 115.1 to 105.1 CG: 31.8 to 31.6
Jarab et al. TC IG: 181.7 to 154.7 -30.9
201253 CG: 181.7 to 185.6 (P = 0.040)
LDL IG: 81.2 to 58.0 -23.2
CG: 85.1 to 85.1 (P = 0.031) IG: 32.4 to 31.9 -0.9
NR –
HDL IG: 50.3 to 44.5 -5.8 CG: 32.8 to 33.2 (P = 0.189)
CG: 50.3 to 50.3 (P = 0.728)
TG IG: 168.3 to 124.0 -62.0
CG: 177.1 to 194.9 (P = 0.017)
Krass et al. TC IG: 189.5 to 181.7 0.0
200726 CG: 189.5 to 181.7 (P = 0.85) IG: 31.4 to 31.1 -0.2
NR –
TG IG: 177.1 to 159.4 -17.7 CG: 31.3 to 31.1 (P = 0.37)
CG: 159.4 to 150.6 (P = 0.39)
Malathy et al. TC IG: 206.2 to 185.7 (P < 0.001) -9.2b
201155 CG: 202.8 to 191.5 (P < 0.05)
LDL IG: 141.12 to 120.9 (P < 0.001) -9.16b
CG: 138.26 to 127.2 (P < 0.01)
NR – NR –
HDL IG: 34.9 to 36.6 (P < 0.05) 1.0b
CG: 33.4 to 34.1b
TG IG: 150.9 to 140.6 (P<0.001) -3.1b
CG: 155.7 to 148.5b
TABLE 3 Studies with Results for Lipid Profile, Body Mass Index, and 10-Year CHD Risk (continued)
Lipid Profile (mg dL-1) Body Mass Index (kg m-2) 10-Year CHD Risk
Author, Change from Baseline to Between Change from Baseline to Between Change from Baseline to Between
Year Final Follow-up Groupsa Final Follow-up Groupsa Final Follow-up Groupsa
Mourão et al. TC
IG: 216.3 to 189.3 -27.3
201339 CG: 207.5 to 207.8 (P = 0.008)
LDL IG: 128.9 to 105.9 -23.5
CG: 123.0 to 123.5 (P = 0.026) IG: 30.3 to 30.4 0.4
NR –
HDL IG: 51.8 to 53.5 4.5 CG: 30.3 to 30.0 (P = 0.106)
CG: 53.4 to 50.6 (P = 0.020)
TG IG: 171.2 to 152.2 -34.0
CG: 162.4 to 177.4 (P = 0.007)
Plaster et al. TC IG: 205 to 182 (P < 0.01) -24b
201240 CG: 209 to 210b
LDL IG: 149 to 111 (P < 0.01) -30b
CG: 140 to 132b
Framingham risk prediction score
Men IG: 40 to 47 (P < 0.05) 11b
NR – IG: 22% to 14% (P < 0.01) -12%b
HDL CG: 44 to 40b
CG: 22% to 26% (P < 0.05)
Women IG: 42 to 49 (P < 0.05) 11 b
HDL CG: 45 to 41b

TG IG: 186 to 165 12b
CG: 207 to 174
Rothman et al. TC IG: 213 to 186 -15
NR – NR –
200534 CG: 201 to 189 (NS)b
Scott et al. LDL IG: 116.1 to 96.7 -11.2
200635 CG: 120.5 to 112.3 (P = 0.012) IG: 36.4 to 36.0
-0.2b NR –
HDL IG: 41.3 to 42.9 0.7b CG: 35.9 to 35.7
CG: 41.5 to 42.4
Simpson et al. TC IG: 170.53 to 161.64 (P < 0.05) -5.41
201136 CG: 168.99 to 165.51b (NS)b
LDL IG: 93.58 to 84.69 (P < 0.05) -5.41
CG: 93.19 to 89.33b (NS)b -1.5%
NR – IG: 19.5% to 16.8% (P<0.001)
HDL IG: 44.47 to 44.86 -0.77 (P = 0.005)
CG: 21.0% to 19.8% (P=0.06)
CG: 44.47 to 45.24 (NS)b
TG IG: 168.28 to 160.31 -15.94
CG: 154.11 to 162.08 (NS)b
Sriram et al. IG: 25.01 to 23.16 (P < 0.05)
NR – -1.94b NR –
201157 CG: 24.66 to 24.75 (P > 0.05)
Taveira et al. LDL IG: 92.8 to 81.5 (P < 0.05) -3.5
201038 CG: 92.8 to 85.0 (P < 0.05) (NS)b IG: 33.8 to 33.73 -0.08
NR –
Non- IG: 133.1 to 113.8 (P < 0.05) -13.7 CG: 31.7 to 31.71 (P = 0.11)
HDL CG: 125.3 to 119.7b (NS)b
Taveira et al. LDL IG: 101.0 to 92.5 (P < 0.05) -1.0
201137 CG: 101.5 to 93.9b (NS)b -2.6%
NR – IG: 20.6% to 15.7% (P < 0.05)
Non- IG: 137.8 to 126.6 (P < 0.05) 4.5 (NS)b
CG: 22.7% to 20.4%b
HDL CG: 149.6 to 133.9 (P < 0.05) (NS)b
Venkatesan
IG: 25.09 to 22.99
et al. NR – -0.67b NR –
CG: 25.81 to 24.38
201259
Wishah et al. TC IG: 184.6 to 165.2 -7.5b
2014 60 CG: 185.5 to 173.6
LDL IG: 114.2 to 100.9 -9.3b
CG: 109.3 to 105.3 IG: 31.1 to 30.6 -1.0
NR –
HDL IG: 40.3 to 43.6 2.2b CG: 29.3 to 29.8 (P =0.11)

CG: 44.0 to 45.1

TG IG: 209.8 to 169.5 -37.0b
CG: 155.9 to 152.6
a
Negative values indicate IG had greater decrease; positive values indicate CG had greater decrease. For HDL: positive values indicate IG had greater increase; negative values indicate CG had
greater increase.
b
c
Positive values indicate IG had greater decrease in 10-year CHD risk; negative values indicate CG had greater decrease in 10-year CHD risk.
BNF = British National Formulary; CG = control group; CHD = coronary heart disease; HDL = high-density lipoprotein cholesterol; IG = intervention group; LDL = low-density lipoprotein cholesterol;
NR = not reported; NS = not significant; TC = total cholesterol; TG = triglycerides; UKPDS = United Kingdom Prospective Diabetes Study.
FIGURE 5 Effect of Pharmacist Interventions on Total Cholesterol, Low-Density Lipoprotein Cholesterol,

High-Density Lipoprotein Cholesterol, and Triglycerides Compared with Control Group
A. Total Cholesterol B. Low-Density Lipoprotein Cholesterol
NR Al Mazroui et al., 200948 P = 0.026 Chan et al., 201249
P = 0.08 Chan et al., 201249 NS Fornos et al., 200642
NR Jacobs et al., 201232

P = 0.0054 Fornos et al., 200642
P = 0.031 Jarab et al., 201253
P = 0.04 Jarab et al., 201253
NR Malathy et al., 201155
NR Malathy et al., 2011 55
P = 0.026 Mourão et al., 201339
P = 0.008 Mourão et al., 201339 Scott et al., 200635
P = 0.012
NS Simpson et al., 201136 NS Simpson et al., 201136
-40 -30 -20 -10 0 10 20 30 40 -40 -30 -20 -10 0 10 20 30 40

Change in total cholesterol (mg dL-1) Change in low-density lipoprotein cholesterol (mg dL-1)
C. High-Density Lipoprotein Cholesterol D. Triglycerides
NR Al Mazroui et al., 200948 NR Al Mazroui et al., 200948
P = 0.93 Chan et al., 201249 P = 0.99 Chan et al., 201249

NS Fornos et al., 200642
NS Fornos et al., 200642
P = 0.728 Jarab et al., 2012 53
P = 0.017 Jarab et al., 201253

NR Malathy et al., 201155
P = 0.39 Krass et al., 200726
P = 0.020 Mourão et al., 201339
NR Scott et al., 200635 NR Malathy et al., 201155
NS Simpson et al., 201136 NS Simpson et al., 201136
-8 -6 -4 -2 0 2 4 6 8 10 12 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60

Change in high-density lipoprotein cholesterol (mg dL-1) Change in triglycerides (mg dL-1)
NR = P value not reported; NS = not statistically significant.
define a range for the difference in change between the groups risk reduction and the direct cost of time spent by the phar-
across all studies. However, in the 2 studies that used the macist in counseling and associated administrative work.49 The
EuroQoL-5 Dimension questionnaire, the difference in change estimated potential saving in costs was $5,086.30 USD per
between both groups was 0.057 and -0.02, respectively. patient.49 Adibe et al. (2013) conducted a cost-utility analysis of
Economic Outcomes. Three studies conducted an economic the pharmaceutical care intervention implemented.66 The total
analysis. Chan et al. (2012) estimated the cost-effectiveness cost per patient per year was $326.00 USD for the control group
of the pharmacist care program being studied based on CHD and $394.00 USD for the intervention group (P = 0.1009).66 In
TABLE 4 Studies with Results for Medication Adherence and Health-Related Quality of Life
Medication Adherence Health-Related Quality of Life
Difference Difference
in Change in Change
Author, Change from Baseline to Between Change from Baseline to Between
Year Final Follow-up Groupsa Final Follow-up Groupsa
Adepu et al. ADDQoL score
200747 NR – IG: significant improvement (P < 0.001) –
CG: nonsignificant changeb
Adepu and Ari Brief medication questionnaire score
2010 46 IG: 0.73 to 0.88 (P < 0.001) 0.59b NR –
CG: 1.11 to 0.67 (P = 0.021)
Adibe et al. HUI3 total score
201345 NR – IG: 0.61 to 0.86 0.22b
CG: 0.63 to 0.64
Al Mazroui et al. Proportion of nonadherent patientsc Improvement in all SF-36 domains in IG, for example:
2009 48 IG: 48.3% to 21.4% general health score
-10.3%b,d 7.2b
CG: 49.1% to 32.5% IG: 67.8 to 77.6
CG: 66.6 to 69.2
Ali et al. SF-36 total score
201241 IG: 65.61 to 79.09 (P < 0.001) 17.63b
CG: 70.04 to 66.53 (P = 0.145)
NR –
DQoL total score
IG: 29.81 to 23.48 (P = 0.001) -3.68b,e
CG: 30.52 to 27.87 (P = 0.323)
Chan et al. Proportion of doses taken
20.5%
201249 IG: 73.6% to 96.1% NR –
(P < 0.001)
CG: 82.1% to 84.1%
Chung et al. Malaysian medication adherence scalef
201450 IG: 70.0% to 75.0% 10.8%b NR –
CG: 64.5% to 58.7%
Cohen et al. SF-36 score
201129 No significant changes in either
NR – physical health 3.6b
mental health -0.3b
domains in IG
Fornos et al. Modification of the Morisky-Green questionnaireg
-0.5d
200665 IG: 0.6 to 0.2 (P < 0.001) NR –
(P < 0.001)
CG: 0.8 to 0.9 (P = 0.195)
Grant et al. No. of adherent days out of the past 7 days
0.0
200330 IG: 6.7 to 6.8 NR –
(P = 0.8)
CG: 6.9 to 7.0
Jaber et al. Health status questionnaire score
NR – –
199631 No significant changes in any domain within or between IG and CG
Jarab et al. Morisky Medication Adherence Scale h
201253 IG: 74.1% to 28.6% -39.2%b,d NR –

CG: 70.9% to 64.6%
Kjeldsen et al. Three self-reported adherence measures:
201543 Behavior-related nonadherence (nonintentional)h
IG: 68.3% to NR
CG: 61.1% to NR
Behavior-related nonadherence (intentional, self-regulation)h EQ-5D total score
IG: 22.5% to NR – IG: 0.782 to 0.842 0.057b
CG: 26.8% to NR CG: 0.812 to 0.815
Behavior-related nonadherence (intentional, effect-related) h
IG: 19.5% to NR
CG: 16.5 % to NR
No significant changes within or between IG and CG
Krass et al. EQ-5D score:
200726 Utility score
IG: 0.8 to 0.8 -0.02
NR – CG: 0.8 to 0.8 (P = 0.07)
Health state scale score
IG: 66.3 to 71.6 4.2
CG: 72.2 to 73.3 (P=0.02)
Mahwi and Obied Pill count and Morisky Medication Adherence Scaleh,i
201354 IG: 77.4% to 19.0% (P < 0.05) – NR –
CG: NR
Mehuys et al. Prescription refill ratej
201127 IG: 99.7%
CG: 94.7%
5.2%b NR –
Self-reported adherencef,k
IG: 59.9% to 61.9%
CG:64.2% to 61.0%
TABLE 4 Studies with Results for Medication Adherence and Health-Related Quality of Life (continued)
Medication Adherence Health-Related Quality of Life
Difference Difference
in Change in Change
Author, Change from Baseline to Between Change from Baseline to Between
Year Final Follow-up Groupsa Final Follow-up Groupsa
Odegard et al. Self-reported adherence (2-question recall technique)l
200533 IG: 56% to NR
CG: 35% to NR
– NR –
The intervention had no effect on improving adherence
during the study period. CG reported better adherence
throughout the study.
Ramanath and Morisky Medication Adherence Scale WHOQOL-BREF total score
Santhosh IG: significant improvement – IG: 39.58 to 43.57 (P < 0.05) 7.39b
201156 CG: nonsignificant improvement CG: 40.78 to 37.38b
Scott et al. DQoL total score
200635 NR – IG: 262.0 to 286.4 9.6b
CG: 232.5 to 247.3
Sriram et al. ADDQoL total score
201157 NR – IG: -2.156 to -1.410 (P < 0.01) 0.821b
CG: -1.899 to -1.974 (P > 0.05)
Wishah et al. Morisky Medication Adherence Scale score
2014 60 IG: 12.7 to 15.8 3.8b NR –
CG: 13.6 to 12.9
a
Positive values indicate IG had greater increase; negative values indicate CG had greater increase.
b
c
Patients who reported forgetting doses and intentionally missing or taking extra doses were classified as nonadherent.
d
Negative values indicate IG had greater increase in medication adherence; positive values indicate CG had greater increase in medication adherence.
e
Negative values indicate IG had greater increase in HRQol; positive values indicate CG had greater increase in HRQoL.
f
Proportion of adherent patients.
g
Number of wrong answers per diabetes medication.
h
Proportion of nonadherent patients.
i
Data regarding Morisky Medication Adherence Scale only.
j
Adherence during the study course.
k
Assessed by asking patients “How often do you not take your oral hypoglycaemic medication as prescribed?”
l
Proportion of patients having difficulty in remembering to take medications as prescribed.
ADDQoL = audit of diabetes-dependent quality of life; CG = control group; DQoL = diabetes quality of life; EQ-5D = EuroQoL-5 dimension; HUI3 = Health Utilities Index Mark 3;
IG = intervention group; NR = not reported; SF-36 = Short Form 36; WHOQOL-BREF = World Health Organization Quality of Life-BREF.
addition, quality-adjusted life-year (QALY) per patient per year ■■  Discussion
was 0.64 for the control group and 0.76 for the intervention group This systematic review examined randomized controlled tri-
(P < 0.0001).66 Thus, the authors found that the intervention led als evaluating the effectiveness of pharmacist interventions in
to an incremental cost of $69.00 USD and an incremental effect the management of patients with type 2 diabetes. It included
of 0.12 QALY gained, with an associated incremental cost-utility 36 studies involving 5,761 participants. The studies were con-
ratio of $571.00 USD per QALY gained, which showed that the ducted in various countries and took place in different health
intervention was very cost-effective.66 Simpson et al. (2015) also care facilities.
conducted a cost-effectiveness analysis for the pharmacist inter- Evidence from the included studies suggests that pharma-
cist interventions directed at patients with type 2 diabetes can
vention being studied.67 The authors found that the total cost per
have a positive impact on clinical outcomes, as demonstrated
patient per year was $190.00 Canadian Dollars (CAD; $144.02
by the reduction in A1c, blood glucose, blood pressure, and
USD) lower in the intervention group compared with the con-
BMI and by the improvement in the lipid profile observed in
trol group and that the intervention group had a 0.26% greater
the intervention group during the follow-up period in almost
reduction in the annualized risk of cardiovascular event in com- all studies. When compared with the control group, the effect
parison with the control group.67 The cost-effectiveness analysis of pharmacist interventions on these outcomes was shown to
showed that at a societal willingness-to-pay of $4,000.00 CAD be greater in the intervention group in most studies, some of
($3,025.95 USD) per 1% reduction in annual cardiovascular risk, which demonstrated that this difference in change between
the probability that the intervention was cost-effective compared both groups was statistically significant. The failure to reach
with usual care reached 95%.67 a statistically significant difference in all studies may be
explained by several factors, such as small sample size, short diabetes, and its presence is associated with increased mortal-
follow-up duration, cross-contamination between patients in ity, as revealed in a meta-analysis conducted by the Emerging
the intervention group and those in the control group, dif- Risk Factors Collaboration group (2010).68 Also important is
ference in the statistical tests used to perform the statistical the beneficial effect of pharmacist interventions on A1c and
analysis (paired-samples or independent-samples tests), and blood glucose seen in many studies, given that improving gly-
presence of a statistical difference between the baseline values cemic control is a key strategy to reduce the risk of microvas-
of both study groups. cular complications associated with diabetes.8 Also concerning
A1c, the difference in its decrease of 1% or greater between the
Metabolic Control
intervention group and the control group observed in several
Our findings are in accordance with those of other systematic
studies is clinically relevant. Indeed, it was previously dem-
reviews on the effectiveness of pharmacist interventions in the
onstrated by Stratton et al. (2000) that a 1% A1c reduction is
management of diabetes. Wubben et al. (2008) showed that in
linked to an estimated 14% reduction of the risk of myocardial
15 out of 18 studies there was a greater improvement in A1c in
infarction, an estimated 12% reduction of the risk of stroke,
the intervention group compared with the control group and
the difference in change between both groups ranged from and an estimated 16% reduction of the risk of heart failure.1
+0.2 to -2.1%21—this range is almost identical to the range Thus, the improvement observed in A1c in the intervention
defined for A1c in the present review. Regarding blood pres- group in some studies may also contribute to the prevention of
sure, the systematic review conducted by Santschi et al. (2012) macrovascular complications.
revealed that in comparison with the control group, 7 out of 12
(58.3%) studies demonstrated a statistically significant greater Adherence and HRQol
reduction in systolic blood pressure in the intervention group, Regarding medication adherence, in most of the included stud-
and 3 out of 9 (33.3%) studies reported a statistically signifi- ies, the direction of the effect was in favor of the pharmacist
cant greater decrease in diastolic blood pressure20 —these pro- interventions, which is similar to what was found in 2 previ-
portions are also similar to those found in the present review. ously published systematic reviews that analyzed the effect of
As for lipid profiles, Wubben et al. reported that most studies pharmacist interventions on this outcome.22,23 These findings
found decreases in LDL cholesterol and triglycerides but did suggest that pharmacists, through their interventions, may
not find statistically significant differences in change between play an essential role in enhancing adherence to prescribed
the intervention group and the control group,21 which is in medications among patients with type 2 diabetes, which, in
accordance with what is reported in the present review. turn, may have a beneficial effect on treatment outcomes. In
fact, in some studies that evaluated this outcome, the increase
Cardiovascular Control in medication adherence observed in the intervention group
The effects verified in some of the clinical outcomes previously during the follow-up period was accompanied by an improve-
mentioned might, in turn, have contributed to a decrease in ment in other outcomes, such as A1c, blood pressure, and lipid
the 10-year CHD risk, since the models used to assess this risk profile. However, it should be borne in mind that the method
incorporate some of these outcomes as variables (e.g., systolic most frequently used to measure this outcome (self-reported
blood pressure and LDL cholesterol). In fact, in the studies adherence) might overestimate adherence.69,70
that estimated the 10-year CHD risk, the decrease observed
As for HRQoL, most of the included studies demonstrated
in the intervention group during the follow-up period was
an improvement in overall or subdomain scores among the
accompanied by an improvement in those outcomes that were
patients of the intervention groups. The lack of significant
analyzed in this systematic review and that are also considered
improvements observed in some studies might be because
as variables in the different risk assessment models. However,
the evidence regarding this long-term outcome is limited by the there is no tool for measuring quality of life that is specifically
small number of studies that predicted this risk among study designed for use in pharmaceutical care, and the existing
participants. This was also noted in another systematic review tools might not have enough sensitivity to detect the subtle
evaluating the effects of pharmacist interventions on patients changes on HRQoL that may result from pharmaceutical
with diabetes, in which only 1 out of 21 studies estimated this care.71 Compared with a previous systematic review evaluating
outcome.21 the effectiveness of pharmacist interventions in type 2 dia-
The positive effects detected in some cardiovascular risk betes that also included HRQoL as an outcome measure, the
factors, such as those observed in blood pressure and lipid findings from the present review are more powerful, since it
profile in the intervention group in several studies, are of great included 11 studies that reported data on this outcome, while
importance because cardiovascular disease is about twice more the other review only included 1 study,23 which is among our
frequent in diabetic patients compared with people without included studies.
Cost-effectiveness interventions on these outcomes across all studies. Fifth, since

The evidence regarding the cost-effectiveness of pharmacist the method most frequently used to ascertain medication
interventions was limited. In fact, only 3 studies conducted adherence was self-reported adherence, there might have been
an economic analysis. This is in accordance with the findings an overestimation of the effect of pharmacist interventions on
from the other 2 systematic reviews that evaluated the effects this outcome. Therefore, future studies should consider the
of pharmacist interventions on patients with diabetes, in which use of more accurate methods in the assessment of medica-
a small proportion of studies (2 out of 21 studies and 1 out tion adherence (e.g., drug assays of blood or urine), in order to
of 8 studies, respectively) assessed the cost-effectiveness of provide more precise evidence regarding the effectiveness of
the interventions under study.21,23 However, given the current pharmacist intervention on this outcome among patients with
resource restraints in the health care systems, and in order to type 2 diabetes.
inform policymakers and influence their decisions towards Finally, although, overall, the findings from this review are
widespread implementation of pharmacist interventions on encouraging, it should be borne in mind that the included
the management of type 2 diabetes, cost-utility studies proving studies presented some weaknesses concerning methodologi-
the cost-effectiveness of such interventions are of the utmost cal quality, as assessed by the Cochrane risk of bias tool.
importance. Indeed, a comprehensive evaluation of phar- Of note, some types of risk of bias (e.g., allocation conceal-
maceutical services has to consider clinical and humanistic ment and blinding of outcome assessment) were classified as
outcomes, as well as economic outcomes (ECHO approach).72 “unclear” in a large proportion of studies, given that there was
The pharmacist interventions being studied encompassed no sufficient information in the publications to classify them
a variety of activities related to nonpharmacological and phar- precisely. Thus, future studies should take this limitation into
macological issues, such as education on lifestyle modification consideration and provide information that is as complete and
and medication review. The type of interventions delivered by detailed as possible in their publications, so that their risk of
the pharmacists in each study reflects in part the difference in bias can be assessed more accurately.
roles played by pharmacists within the health care systems in
different countries. Because pharmacist interventions in the ■■  Conclusions
included studies comprised different elements, it is difficult to This systematic review provides evidence that pharmacist
pinpoint which elements contributed the most for the observed interventions can have a positive influence on metabolic con-
effects. Thus, further research is needed to compare the effect trol, medication adherence, and HRQol of patients with type 2
of the different intervention elements on outcome measures diabetes. These promising findings support the involvement of
in order to identify the most effective and practical approach. pharmacists as members of a health care team in the manage-
ment of this patient population.
Limitations A considerable amount of data has been gathered in this
This review presents some limitations that are discussed as fol- review concerning various outcomes, which may make it pos-
lows. First, some included studies did not report their outcome sible to conduct a meta-analysis in the near future, by selecting
data in sufficient detail so that the data could be used in our data on some outcomes from a more homogeneous subset of
data synthesis, and the respective authors were not contacted studies included in this review.
in an attempt to obtain the missing data. However, in order Future studies evaluating the effectiveness of pharma-
to prevent this from happening in future systematic reviews, cist interventions in the management of type 2 diabetes
study authors should prepare their publications according to should include clinical, humanistic, and economic outcomes.
reporting guidelines, such as the Consolidated Standards of Additionally, when assessing some particular outcomes (e.g.,
Reporting Trials (CONSORT) Statement.73 Second, this review medication adherence) among study participants, studies
included studies that took place in different settings and that should preferably use the most accurate methods in order to
evaluated the effectiveness of different types of interventions, provide more reliable evidence regarding the effect of the inter-
which increased the level of heterogeneity of the results. Third, vention being studied on those outcomes.
the evidence regarding some of the outcomes under study was Further research is also needed to evaluate which interven-
limited by the small number of included studies that reported tion elements contribute the most for the observed effects. It
them (e.g., 10-year CHD risk and economic outcomes). Fourth, would also be of great value to concentrate efforts on building
for some outcomes, such as medication adherence and HRQoL, standard, well-validated tools to ascertain certain outcomes,
the included studies used several tools for assessment, which such as HRQoL, so that data from different studies concerning
prevented us from directly comparing the effect of pharmacist these outcomes can be analyzed together.
9. UK Prospective Diabetes Study Group. Tight blood pressure control and

Authors risk of macrovascular and microvascular complications in type 2 diabetes:
UKPDS 38. BMJ. 1998;317(7160):703-13.
SARAH POUSINHO, PharmD, MSc, CICS-UBI−Health Sciences
Research Centre, University of Beira Interior, Covilhã, Portugal. 10. Charpentier G, Genes N, Vaur L, et al. Control of diabetes and cardio-
MANUEL MORGADO, PharmD, PhD, CICS-UBI-Health Sciences vascular risk factors in patients with type 2 diabetes: a nationwide French
survey. Diabetes Metab. 2003;29(2 Pt 1):152-58.
Research Centre, University of Beira Interior, and Pharmaceutical
Services of Hospital Centre of Cova da Beira, Covilhã, Portugal. 11. Eliasson B, Cederholm J, Nilsson P, Gudbjornsdottir S; Steering
AMÍLCAR FALCÃO, PharmD, PhD, CNC−Centre for Neuroscience Committee of the Swedish National Diabetes Register. The gap between
and Cell Biology, and Laboratory of Pharmacology, Faculty of guidelines and reality: type 2 diabetes in a National Diabetes Register
1996-2003. Diabet Med. 2005;22(10):1420-26.
Pharmacy, University of Coimbra, Coimbra, Portugal. GILBERTO
ALVES, PharmD, PhD, CICS-UBI−Health Sciences Research 12. Kemp TM, Barr EL, Zimmet PZ, et al. Glucose, lipid, and blood pressure
Centre, University of Beira Interior, Covilhã, Portugal, and CNC- control in Australian adults with type 2 diabetes: the 1999-2000 AusDiab.
Centre for Neuroscience and Cell Biology, University of Coimbra, Diabetes Care. 2005;28(6):1490-92.
Coimbra, Portugal. 13. Debussche X. Is adherence a relevant issue in the self-management
education of diabetes? A mixed narrative review. Diabetes Metab Syndr Obes.
AUTHOR CORRESPONDENCE: Gilberto Alves, PharmD, PhD, 2014;7:357-67.
Faculty of Health Sciences, University of Beira Interior, CICS-UBI- 14. Nam S, Chesla C, Stotts NA, Kroon L, Janson SL. Barriers to diabetes
Health Sciences Research Centre, University of Beira Interior, management: patient and provider factors. Diabetes Res Clin Pract.
Av. Infante D. Henrique, 6200-506, Covilhã, Portugal. 2011;93(1):1-9.
Tel.: +351 275329002; Fax: +351 275329099; 15. Vermeire E, Hearnshaw H, Van Royen P, Denekens J. Patient adherence
E-mail: gilberto@fcsaude.ubi.pt. to treatment: three decades of research. A comprehensive review. J Clin
Pharm Ther. 2001;26(5):331-42.
16. American College of Clinical Pharmacy, Hume AL, Kirwin J, et al.
Improving care transitions: current practice and future opportunities for
pharmacists. Pharmacotherapy. 2012;32(11):e326-37.
DISCLOSURES
17. Nichols-English GJ, Provost M, Koompalum D, Chen H, Athar M.
This systematic review was not funded. The authors declare that they have no Strategies for pharmacists in the implementation of diabetes mellitus man-
conflicts of interest. agement programs: new roles in primary and collaborative care. Dis Manag
Concept and study design were created by Pousinho, Morgado, and Alves.
Health Outcomes. 2002;10(12):783-803.
Pousinho took the lead in data collection, along with Alves, and data inter-
pretation was performed by Pousinho, Falcão, and Alves. The manuscript 18. American Diabetes Association. Standards of medical care in diabetes−
was primarily written by Pousinho, along with Alves, and revised by Alves, 2015. Diabetes Care. 2015;38(Suppl 1):S1-93. Available at: http://care.diabe-
Morgado, and Falcão. tesjournals.org/content/38/Supplement_1. Accessed March 16, 2016.
19. Hepler CD, Strand LM. Opportunities and responsibilities in pharma-
REFERENCES ceutical care. Am J Hosp Pharm. 1990;47(3):533-43.
1. Stratton IM, Adler AI, Neil HA, et al. Association of glycaemia with mac- 20. Santschi V, Chiolero A, Paradis G, Colosimo AL, Burnand B. Pharmacist
rovascular and microvascular complications of type 2 diabetes (UKPDS 35): interventions to improve cardiovascular disease risk factors in diabetes:
prospective observational study. BMJ. 2000;321(7258):405-12. a systematic review and meta-analysis of randomized controlled trials.
Diabetes Care. 2012;35(12):2706-17.
2. Wu Y, Ding Y, Tanaka Y, Zhang W. Risk factors contributing to type 2
diabetes and recent advances in the treatment and prevention. Int J Med Sci. 21. Wubben DP, Vivian EM. Effects of pharmacist outpatient interventions
2014;11(11):1185-200. on adults with diabetes mellitus: a systematic review. Pharmacotherapy.
2008;28(4):421-36.
3. Huang ES, Brown SE, Ewigman BG, Foley EC, Meltzer DO. Patient
perceptions of quality of life with diabetes-related complications and treat- 22. Antoine SL, Pieper D, Mathes T, Eikermann M. Improving the adherence
ments. Diabetes Care. 2007;30(10):2478-83. of type 2 diabetes mellitus patients with pharmacy care: a systematic review
4. Solli O, Stavem K, Kristiansen IS. Health-related quality of life in diabetes: of randomized controlled trials. BMC Endocr Disord. 2014;14:53.
the associations of complications with EQ-5D scores. Health Qual Life 23. Omran D, Guirguis LM, Simpson SH. Systematic review of pharmacist
Outcomes. 2010;8:18. interventions to improve adherence to oral antidiabetic medications in peo-
5. Wexler DJ, Grant RW, Wittenberg E, et al. Correlates of health-related ple with type 2 diabetes. Can J Diabetes. 2012;36(5):292-99.
quality of life in type 2 diabetes. Diabetologia. 2006;49(7):1489-97. 24. Cochrane Consumers and Communication Review Group. Data extrac-
6. International Diabetes Federation. IDF Diabetes Atlas. 7th ed. Brussels, tion template. 2015. Available at: http://cccrg.cochrane.org/author-resources.
Belgium: International Diabetes Federation. 2015. Available at: http://www. Accessed March 16, 2016.
idf.org/diabetesatlas. Accessed March 14, 2016.
25. Higgins JP, Altman DG, Gotzsche PC, et al. The Cochrane Collaboration’s
7. Collins R, Armitage J, Parish S, Sleigh P, Peto R; Heart Protection Study tool for assessing risk of bias in randomised trials. BMJ. 2011;343:d5928.
Collaborative Group. MRC/BHF Heart Protection Study of cholesterol-low-
ering with simvastatin in 5963 people with diabetes: a randomised placebo- 26. Krass I, Armour CL, Mitchell B, et al. The Pharmacy Diabetes Care
controlled trial. Lancet. 2003;361(9374):2005-16. Program: assessment of a community pharmacy diabetes service model in
Australia. Diabet Med. 2007;24(6):677-83.
8. UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose
control with sulphonylureas or insulin compared with conventional treat- 27. Mehuys E, Van Bortel L, De Bolle L, et al. Effectiveness of a community
ment and risk of complications in patients with type 2 diabetes (UKPDS 33). pharmacist intervention in diabetes care: a randomized controlled trial.
Lancet. 1998;352(9131):837-53. J Clin Pharm Ther. 2011;36(5):602-13.
28. Choe HM, Mitrovich S, Dubay D, Hayward RA, Krein SL, Vijan S. 47. Adepu R, Rasheed A, Nagavi B. Effect of patient counseling on quality of
Proactive case management of high-risk patients with type 2 diabetes life in type-2 diabetes mellitus patients in two selected South Indian com-
mellitus by a clinical pharmacist: a randomized controlled trial. Am J munity pharmacies: a study. Indian J Pharm Sci. 2007;69(4):519-24.
Manag Care. 2005;11(4):253-60.
48. Al Mazroui NR, Kamal MM, Ghabash NM, Yacout TA, Kole PL,
29. Cohen LB, Taveira TH, Khatana SA, Dooley AG, Pirraglia PA, Wu WC. McElnay JC. Influence of pharmaceutical care on health outcomes in patients
Pharmacist-led shared medical appointments for multiple cardiovascular with type 2 diabetes mellitus. Br J Clin Pharmacol. 2009;67(5):547-57.
risk reduction in patients with type 2 diabetes. Diabetes Educ. 2011;37(6):
49. Chan CW, Siu SC, Wong CK, Lee VW. A pharmacist care program:
801-12.
positive impact on cardiac risk in patients with type 2 diabetes. J Cardiovasc
30. Grant RW, Devita NG, Singer DE, Meigs JB. Improving adherence Pharmacol Ther. 2012;17(1):57-64.
and reducing medication discrepancies in patients with diabetes. Ann
50. Chung WW, Chua SS, Lai PS, Chan SP. Effects of a pharmaceutical care
Pharmacother. 2003;37(7-8):962-69.
model on medication adherence and glycemic control of people with type 2
31. Jaber LA, Halapy H, Fernet M, Tummalapalli S, Diwakaran H. Evaluation diabetes. Patient Prefer Adherence. 2014;8:1185-94.
of a pharmaceutical care model on diabetes management. Ann Pharmacother.
51. Farsaei S, Sabzghabaee AM, Zargarzadeh AH, Amini M. Effect of
1996;30(3):238-43.
pharmacist-led patient education on glycemic control of type 2 diabetics:
32. Jacobs M, Sherry PS, Taylor LM, Amato M, Tataronis GR, Cushing G. a randomized controlled trial. J Res Med Sci. 2011;16(1):43-49.
Pharmacist Assisted Medication Program Enhancing the Regulation of
52. Ghosh S, Rajvanshi AK, Kishun S. Assessment the influence of patient
Diabetes (PAMPERED) study. J Am Pharm Assoc (2003). 2012;52(5):613-21.
counseling on quality of life in type-II diabetes mellitus patients. Int J
33. Odegard PS, Goo A, Hummel J, Williams KL, Gray SL. Caring for poorly Pharma Bio Sci. 2010;1(3):1-6.
controlled diabetes mellitus: a randomized pharmacist intervention. Ann
Pharmacother. 2005;39(3):433-40. 53. Jarab AS, Alqudah SG, Mukattash TL, Shattat G, Al-Qirim T.
Randomized controlled trial of clinical pharmacy management of patients
34. Rothman RL, Malone R, Bryant B, et al. A randomized trial of a primary with type 2 diabetes in an outpatient diabetes clinic in Jordan. J Manag
care-based disease management program to improve cardiovascular risk Care Pharm. 2012;18(7):516-26. Available at: http://www.jmcp.org/doi/
factors and glycated hemoglobin levels in patients with diabetes. Am J Med. abs/10.18553/jmcp.2012.18.7.516.
2005;118(3):276-84.
54. Mahwi TO, Obied KA. Role of the pharmaceutical care in the man-
35. Scott DM, Boyd ST, Stephan M, Augustine SC, Reardon TP. Outcomes of agement of patients with type 2 diabetes mellitus. Int J Pharm Sci Res.
pharmacist-managed diabetes care services in a community health center. 2013;4(4):1363-69.
Am J Health Syst Pharm. 2006;63(21):2116-22.
55. Malathy R, Narmadha M, Ramesh S, Alvin JM, Dinesh BN. Effect of
36. Simpson SH, Majumdar SR, Tsuyuki RT, Lewanczuk RZ, Spooner R, a diabetes counseling programme on knowledge, attitude and practice
Johnson JA. Effect of adding pharmacists to primary care teams on blood among diabetic patients in Erode district of South India. J Young Pharm.
pressure control in patients with type 2 diabetes: a randomized controlled 2011;3(1):65-72.
trial. Diabetes Care. 2011;34(1):20-26.
56. Ramanath KV, Santhosh YL. Impact of clinical pharmacist provided
37. Taveira TH, Dooley AG, Cohen LB, Khatana SA, Wu WC. Pharmacist-led patient education on QOL outcome in type II diabetes mellitus in rural
group medical appointments for the management of type 2 diabetes with population. Asian J Pharm Clin Res. 2011;4(4):15-20.
comorbid depression in older adults. Ann Pharmacother. 2011;45(11):1346-55.
57. Sriram S, Chack LE, Ramasamy R, Ghasemi A, Ravi TK, Sabzghabaee
38. Taveira TH, Friedmann PD, Cohen LB, et al. Pharmacist-led group medi- AM. Impact of pharmaceutical care on quality of life in patients with type 2
cal appointment model in type 2 diabetes. Diabetes Educ. 2010;36(1):109-17. diabetes mellitus. J Res Med Sci. 2011;16(Suppl 1):S412-18.
39. Mourao AO, Ferreira WR, Martins MA, et al. Pharmaceutical care pro- 58. Suppapitiporn S, Chindavijak B, Onsanit S. Effect of diabetes drug
gram for type 2 diabetes patients in Brazil: a randomised controlled trial. counseling by pharmacist, diabetic disease booklet and special medication
Int J Clin Pharm. 2013;35(1):79-86. containers on glycemic control of type 2 diabetes mellitus: a randomized
40. Plaster CP, Melo DT, Boldt V, et al. Reduction of cardiovascular risk controlled trial. J Med Assoc Thai. 2005;88(Suppl 4):S134-41.
in patients with metabolic syndrome in a community health center after a 59. Venkatesan R, Devi AS, Parasuraman S, Sriram S. Role of community
pharmaceutical care program of pharmacotherapy follow-up. Braz J Pharm pharmacists in improving knowledge and glycemic control of type 2
Sci. 2012;48(3):435-46. diabetes. Perspect Clin Res. 2012;3(1):26-31.
41. Ali M, Schifano F, Robinson P, et al. Impact of community pharmacy 60. Wishah RA, Al-Khawaldeh OA, Albsoul AM. Impact of pharmaceutical
diabetes monitoring and education programme on diabetes management: care interventions on glycemic control and other health-related clinical out-
a randomized controlled study. Diabet Med. 2012;29(9):e326-33. comes in patients with type 2 diabetes: randomized controlled trial. Diabetes
42. Fornos JA, Andres NF, Andres JC, Guerra MM, Egea B. A pharmaco- Metab Syndr. 2015;9(4):271-76.
therapy follow-up program in patients with type-2 diabetes in community 61. Clifford RM, Davis WA, Batty KT, Davis TM. Effect of a pharmaceuti-
pharmacies in Spain. Pharm World Sci. 2006;28(2):65-72. cal care program on vascular risk factors in type 2 diabetes: the Fremantle
43. Kjeldsen LJ, Bjerrum L, Dam P, Larsen BO, Rossing C, Søndergaard B. Diabetes Study. Diabetes Care. 2005;28(4):771-76.
Safe and effective use of medicines for patients with type 2 diabetes: a ran- 62. Stevens RJ, Kothari V, Adler AI, Stratton IM; United Kingdom
domized controlled trial of two interventions delivered by local pharmacies. Prospective Diabetes Study Group. The UKPDS risk engine: a model for
Res Social Adm Pharm. 2015;11(1):47-62. the risk of coronary heart disease in Type II diabetes (UKPDS 56). Clin Sci
44. Sarkadi A, Rosenqvist U. Experience-based group education in type 2 dia- (Lond). 2001;101(6):671-79.
betes: a randomised controlled trial. Patient Educ Couns. 2004;53(3):291-98. 63. Wilson PW, D’Agostino RB, Levy D, Belanger AM, Silbershatz H,
45. Adibe MO, Ukwe CV, Aguwa CN. The impact of pharmaceutical care Kannel WB. Prediction of coronary heart disease using risk factor
intervention on the quality of life of Nigerian patients receiving treatment categories. Circulation. 1998;97(18):1837-47.
for type 2 diabetes. Value Health Reg Issues. 2013;2(2):240-47. 64. Wood D, De Backer G, Faergeman O, Graham I, Mancia G, Pyorala K.
46. Adepu R, Ari SM. Influence of structured patient education on therapeu- Prevention of coronary heart disease in clinical practice: recommendations
tic outcomes in diabetes and hypertensive patients. Asian J Pharm Clin Res. of the Second Joint Task Force of European and other Societies on Coronary
2010;3(3):174-78. Prevention. Atherosclerosis. 1998;140(2):199-270.
65. Rodríguez NFA, Pérez JAF, Iglesias JCA. Valoración del conocimiento/ 69. Gonzalez JS, Schneider HE, Wexler DJ, et al. Validity of medica-
cumplimiento en un programa de seguimiento farmacoterapéutico en diabé- tion adherence self-reports in adults with type 2 diabetes. Diabetes Care.
ticos tipo 2 en farmacia comunitaria: estudio aleatorizado. Pharm Care Esp. 2013;36(4):831-37.
2007;9(1):2-9.
70. Osterberg L, Blaschke T. Adherence to medication. N Engl J Med.
66. Adibe MO, Aguwa CN, Ukwe CV. Cost-utility analysis of pharmaceutical 2005;353(5):487-97.
care intervention versus usual care in management of Nigerian patients with
type 2 diabetes. Value Health Reg Issues. 2013;2(2):189-98. 71. Kheir NM, van Mil JW, Shaw JP, Sheridan JL. Health-related quality of
life measurement in pharmaceutical care. Targeting an outcome that matters.
67. Simpson SH, Lier DA, Majumdar SR, et al. Cost-effectiveness analysis of Pharm World Sci. 2004;26(3):125-28.
adding pharmacists to primary care teams to reduce cardiovascular risk in
patients with type 2 diabetes: results from a randomized controlled trial. 72. Kozma CM, Reeder CE, Schulz RM. Economic, clinical, and humanistic
Diabet Med. 2015;32(7):899-906. outcomes: a planning model for pharmacoeconomic research. Clin Ther.
1993;15(6):1121-32.
68. Emerging Risk Factors Collaboration, Sarwar N, Gao P, et al. Diabetes
mellitus, fasting blood glucose concentration, and risk of vascular dis- 73. Schulz KF, Altman DG, Moher D; CONSORT Group. CONSORT 2010
ease: a collaborative meta-analysis of 102 prospective studies. Lancet. statement: updated guidelines for reporting parallel group randomised trials.
2010;375(9733):2215-22. BMJ. 2010;340:c332.
APPENDIX Search Strategy

PubMed
1. Diabetes mellitus, type 2[mh] (87.255)
2. Type 2 diabetes (113.885)
3. T2DM (7.760)
4. Non insulin dependent diabetes mellitus (106.016)
5. NIDDM (96.382)
6. 1 OR 2 OR 3 OR 4 OR 5 (125.207)
7. Pharmaceutical services[mh] (50.073)
8. Pharmaceutical care (69.689)
9. Clinical pharmacy (44.165)
10. Community pharmacy (15.465)
11. Pharmacist* (25.127)
12. 7 OR 8 OR 9 OR 10 OR 11 (124.573)
13. Randomized controlled trial[pt] (380.924)
14. Random* AND Control* (663.250)
15. 13 OR 14 (663.264)
16. 6 AND 12 AND 15 (278)
Cochrane Central Register of Controlled Trials
1. [mh “Diabetes mellitus, type 2”] (9.054)
2. Type 2 diabetes (19.141)
3. T2DM (1.225)
4. Non insulin dependent diabetes mellitus (5.281)
5. NIDDM (1.093)
6. 1 OR 2 OR 3 OR 4 OR 5 (20.135)
7. [mh “Pharmaceutical services”] (1.322)
8. Pharmaceutical care (3.551)
9. Clinical pharmacy (7.797)
10. Community pharmacy (1.336)
11. Pharmacist* (2.769)
12. 7 OR 8 OR 9 OR 10 OR 11 (12.439)
13. 6 AND 12 (254)
Web of Science
1. ts = Type 2 diabetes (351.632)
2. ts = T2DM (19.524)
3. ts = Non insulin dependent diabetes mellitus (29.057)
4. ts = NIDDM (24.822)
5. 1 OR 2 OR 3 OR 4 (375.234)
6. ts = Pharmaceutical services (23.487)
7. ts = Pharmaceutical care (39.352)
8. ts = Clinical pharmacy (19.468)
9. ts = Community pharmacy (15.019)
10. ts = Pharmacist* (57.991)
11. 6 OR 7 OR 8 OR 9 OR 10 (118.778)
12. ts = (Random* AND Control*) (1.260.847)
13. 5 AND 11 AND 12 (175)

jmcp.2016.22.5.493 Rup

Uploaded by

Copyright:

Available Formats

jmcp.2016.22.5.493 Rup

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

jmcp.2016.22.5.493 Rup

Uploaded by

Copyright:

Available Formats

S Y S T E M AT I C R E V I E W

Pharmacist Interventions in the Management of Type 2 Diabetes

FIGURE 1 Flowchart of Study Selection Process

Records identified through

Records screened based on titles

Full-text articles assessed for eligibility No access to full-text articles

Full-text articles excluded with reasons

RCT = randomized controlled trial.

TABLE 1 Characteristics of Included Studies

TABLE 1 Characteristics of Included Studies (continued)

TABLE 1 Characteristics of Included Studies (continued)

Random sequence generation

Blinding of outcomes assessment Low risk

Other sources of bias

A. A1c B. Fasting Blood Glucose

Control Group Intervention Group Control Group Intervention Group

P < 0.001 Chan et al., 201249 NR Al Mazroui et al., 200948

NR Suppapitiporn et al., 200558 NR Venkatesan et al., 201259

-3 -2 -1 0 1 2 3 -100 -75 -50 -25 0 25 50 75 100

A1c = glycosylated hemoglobin; NR = P value not reported; NS = not statistically significant.

A. Systolic Blood Pressure B. Diastolic Blood Pressure

Control Group Intervention Group Control Group Intervention Group

P = 0.34 Chan et al., 201249 P = 0.23 Chan et al., 201249

P < 0.001 Fornos et al., 200642 NS Fornos et al., 200642

NR Jacobs et al., 201232 NR Jacobs et al., 201232

P = 0.035 Jarab et al., 201253 P = 0.026 Jarab et al., 201253

P = 0.06 Krass et al., 200726 P = 0.52 Krass et al., 200726

P = 0.023 Scott et al., 200635 NR Scott et al., 200635

P = 0.01 Simpson et al., 201136 P < 0.05 Simpson et al., 201136

NS Taveira et al., 201038 P < 0.05 Taveira et al., 201038

NR = P value not reported; NS = not statistically significant.

HDL CG: 45 to 41b

CG: 44.0 to 45.1

FIGURE 5 Effect of Pharmacist Interventions on Total Cholesterol, Low-Density Lipoprotein Cholesterol,

A. Total Cholesterol B. Low-Density Lipoprotein Cholesterol

Control Group Intervention Group Control Group Intervention Group

NR Al Mazroui et al., 200948 P = 0.026 Chan et al., 201249

P = 0.08 Chan et al., 201249 NS Fornos et al., 200642

NR Jacobs et al., 201232

-40 -30 -20 -10 0 10 20 30 40 -40 -30 -20 -10 0 10 20 30 40

C. High-Density Lipoprotein Cholesterol D. Triglycerides

Control Group Intervention Group Control Group Intervention Group

NR Al Mazroui et al., 200948 NR Al Mazroui et al., 200948

P = 0.93 Chan et al., 201249 P = 0.99 Chan et al., 201249

P = 0.017 Jarab et al., 201253

NR Scott et al., 200635 NR Malathy et al., 201155

NS Simpson et al., 201136 NS Simpson et al., 201136

-8 -6 -4 -2 0 2 4 6 8 10 12 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60

NR = P value not reported; NS = not statistically significant.

201253 IG: 74.1% to 28.6% -39.2%b,d NR –

Cost-effectiveness interventions on these outcomes across all studies. Fifth, since

9. UK Prospective Diabetes Study Group. Tight blood pressure control and

APPENDIX Search Strategy

You might also like