Community-Based Care for the Management of Type 2 Diabetes: An Evidence-Based Analysis

Medical Advisory Secretariat

. 2009 Oct 1;9(23):1–40.

Community-Based Care for the Management of Type 2 Diabetes

An Evidence-Based Analysis

Medical Advisory Secretariat

PMCID: PMC3377524 PMID: 23074528

Executive Summary

In June 2008, the Medical Advisory Secretariat began work on the Diabetes Strategy Evidence Project, an evidence-based review of the literature surrounding strategies for successful management and treatment of diabetes. This project came about when the Health System Strategy Division at the Ministry of Health and Long-Term Care subsequently asked the secretariat to provide an evidentiary platform for the Ministry’s newly released Diabetes Strategy.

After an initial review of the strategy and consultation with experts, the secretariat identified five key areas in which evidence was needed. Evidence-based analyses have been prepared for each of these five areas: insulin pumps, behavioural interventions, bariatric surgery, home telemonitoring, and community based care. For each area, an economic analysis was completed where appropriate and is described in a separate report.

To review these titles within the Diabetes Strategy Evidence series, please visit the Medical Advisory Secretariat Web site, http://www.health.gov.on.ca/english/providers/program/mas/mas_about.html,

Diabetes Strategy Evidence Platform: Summary of Evidence-Based Analyses
Continuous Subcutaneous Insulin Infusion Pumps for Type 1 and Type 2 Adult Diabetics: An Evidence-Based Analysis
Behavioural Interventions for Type 2 Diabetes: An Evidence-Based Analysis
Bariatric Surgery for People with Diabetes and Morbid Obesity: An Evidence-Based Summary
Community-Based Care for the Management of Type 2 Diabetes: An Evidence-Based Analysis
Home Telemonitoring for Type 2 Diabetes: An Evidence-Based Analysis
Application of the Ontario Diabetes Economic Model (ODEM) to Determine the Cost-effectiveness and Budget Impact of Selected Type 2 Diabetes Interventions in Ontario

Objective

The objective of this report is to determine the efficacy of specialized multidisciplinary community care for the management of type 2 diabetes compared to usual care.

Clinical Need: Target Population and Condition

Diabetes (i.e. diabetes mellitus) is a highly prevalent chronic metabolic disorder that interferes with the body’s ability to produce or effectively use insulin. The majority (90%) of diabetes patients have type 2 diabetes. (1) Based on the United Kingdom Prospective Diabetes Study (UKPDS), intensive blood glucose and blood pressure control significantly reduce the risk of microvascular and macrovascular complications in type 2 diabetics. While many studies have documented that patients often do not meet the glycemic control targets specified by national and international guidelines, factors associated with glycemic control are less well studied, one of which is the provider(s) of care.

Multidisciplinary approaches to care may be particularly important for diabetes management. According guidelines from the Canadian Diabetes Association (CDA), the diabetes health care team should be multi-and interdisciplinary. Presently in Ontario, the core diabetes health care team consists of at least a family physician and/or diabetes specialist, and diabetes educators (registered nurse and registered dietician).

Increasing the role played by allied health care professionals in diabetes care and their collaboration with physicians may represent a more cost-effective option for diabetes management. Several systematic reviews and meta-analyses have examined multidisciplinary care programs, but these have either been limited to a specific component of multidisciplinary care (e.g. intensified education programs), or were conducted as part of a broader disease management program, of which not all were multidisciplinary in nature. Most reviews also do not clearly define the intervention(s) of interest, making the evaluation of such multidisciplinary community programs challenging.

Evidence-Based Analysis Methods

Research Questions

What is the evidence of efficacy of specialized multidisciplinary community care provided by at least a registered nurse, registered dietician and physician (primary care and/or specialist) for the management of type 2 diabetes compared to usual care? [Henceforth referred to as Model 1]
What is the evidence of efficacy of specialized multidisciplinary community care provided by at least a pharmacist and a primary care physician for the management of type 2 diabetes compared to usual care? [Henceforth referred to as Model 2]

Inclusion Criteria

English language full-reports
Published between January 1, 2000 and September 28, 2008
Randomized controlled trials (RCTs), systematic reviews and meta-analyses
Type 2 diabetic adult population (≥18 years of age)
Total sample size ≥30
Describe specialized multidisciplinary community care defined as ambulatory-based care provided by at least two health care disciplines (of which at least one must be a specialist in diabetes) with integrated communication between the care providers.
Compared to usual care (defined as health care provision by non-specialist(s) in diabetes, such as primary care providers; may include referral to other health care professionals/services as necessary)
≥6 months follow-up

Exclusion Criteria

Studies where discrete results on diabetes cannot be abstracted
Predominantly home-based interventions
Inpatient-based interventions

Outcomes of Interest

The primary outcomes for this review were glycosylated hemoglobin (rHbA1c) levels and systolic blood pressure (SBP).

Search Strategy

A literature search was performed on September 28, 2008 using OVID MEDLINE, MEDLINE In-Process and Other Non-Indexed Citations, EMBASE, the Cumulative Index to Nursing & Allied Health Literature (CINAHL), the Cochrane Library, and the International Agency for Health Technology Assessment (INAHTA) for studies published between January 1, 2000 and September 28, 2008. Abstracts were reviewed by a single reviewer and, for those studies meeting the eligibility criteria, full-text articles were obtained. Reference lists were also examined for any additional relevant studies not identified through the search. Articles with unknown eligibility were reviewed with a second clinical epidemiologist, then a group of epidemiologists until consensus was established. The quality of evidence was assessed as high, moderate, low or very low according to GRADE methodology.

Given the high clinical heterogeneity of the articles that met the inclusion criteria, specific models of specialized multidisciplinary community care were examined based on models of care that are currently being supported in Ontario, models of care that were commonly reported in the literature, as well as suggestions from an Expert Advisory Panel Meeting held on January 21, 2009.

Summary of Findings

The initial search yielded 2,116 unique citations, from which 22 RCTs trials and nine systematic reviews published were identified as meeting the eligibility criteria. Of these, five studies focused on care provided by at least a nurse, dietician, and physician (primary care and/or specialist) model of care (Model 1; see Table ES 1), while three studies focused on care provided by at least a pharmacist and primary care physician (Model 2; see Table ES 2).

Based on moderate quality evidence, specialized multidisciplinary community care Model 2 has demonstrated a statistically and clinically significant reduction in HbA1c of 1.0% compared with usual care. The effects of this model on SBP, however, are uncertain compared with usual care, based on very-low quality evidence. Specialized multidisciplinary community care Model 2 has demonstrated a statistically and clinically significant reduction in both HbA1c of 1.05% (based on high quality evidence) and SBP of 7.13 mm Hg (based on moderate quality evidence) compared to usual care. For both models, the evidence does not suggest a preferred setting of care delivery (i.e., primary care vs. hospital outpatient clinic vs. community clinic).

Table ES1: Summary of Results of Meta-Analyses of the Effects of Multidisciplinary Care Model 1.

Outcome	Estimate of effect* (95% CI)	Heterogeneity I² (p-value)	GRADE
Glycosylated Hemoglobin (HbA1c [%])	-1.00 [-1.27, -0.73]	4% (p=0.37)	Moderate-quality
Subgroup: Moderate-to-High Quality	-0.91 [-1.19, -0.62]	0% (p=0.74)
Systolic Blood Pressure (mm Hg)	-2.04 [-13.80, 9.72]	89% (p=0.002)	Very-low quality

Open in a new tab

Mean change from baseline to follow-up between intervention and control groups

Table ES2: Summary of Results of Meta-Analyses of the Effects of Multidisciplinary Care Model 2.

Outcome	Estimate of effect* (95% CI)	Heterogeneity I² (p-value)	GRADE
Glycosylated Hemoglobin (HbA1c [%])	-1.05 [-1.57, -0.52]	0% (p=0.75)	High-quality
Systolic Blood Pressure (mm Hg)	-7.13 [-11.78, -2.48]	46% (p=0.17)	Moderate quality

Open in a new tab

Mean change from baseline to follow-up between intervention and control groups

Background

To review these titles within the Diabetes Strategy Evidence series, please visit the Medical Advisory Secretariat Web site, http://www.health.gov.on.ca/english/providers/program/mas/mas_about.html,

Diabetes Strategy Evidence Platform: Summary of Evidence-Based Analyses
Continuous Subcutaneous Insulin Infusion Pumps for Type 1 and Type 2 Adult Diabetics: An Evidence-Based Analysis
Behavioural Interventions for Type 2 Diabetes: An Evidence-Based Analysis
Bariatric Surgery for People with Diabetes and Morbid Obesity: An Evidence-Based Summary
Community-Based Care for the Management of Type 2 Diabetes: An Evidence-Based Analysis
Home Telemonitoring for Type 2 Diabetes: An Evidence-Based Analysis
Application of the Ontario Diabetes Economic Model (ODEM) to Determine the Cost-effectiveness and Budget Impact of Selected Type 2 Diabetes Interventions in Ontario

Objective

The objective of this report is to determine the efficacy of specialized multidisciplinary community care for the management of type 2 diabetes compared to usual care.

Clinical Need and Target Population

Diabetes is a highly prevalent chronic metabolic disorder that interferes with the body’s ability to produce or effectively use insulin. The majority (90%) of diabetes patients have type 2 diabetes and in 2005, an estimated 8.8% of Ontario’s population had diabetes, representing more than 816,000 Ontarians. (1) Clinically, diabetes is the leading causes of blindness, end-stage renal disease, and non-traumatic amputation in Canadian adults and is a significant cause of cardiovascular complications, hypertension, stroke, cataracts, and glaucoma. (2) In 2000, the direct health care cost of diabetes was $1.76 billion, a total that’s projected to rise to $3.14 billion by 2016.

Based on the United Kingdom Prospective Diabetes Study (UKPDS), intensive blood glucose and blood pressure control lower the risk of microvascular and macrovascular complications in type 2 diabetics. Specifically, a 1% reduction in HbA1c has been associated with a 10% reduction in diabetes-related mortality and a 25% reduction in microvascular end-points. (3) Likewise, intensive blood pressure control is associated with a 32% reduction in risk of mortality from diabetes-associated conditions, two-thirds of which are cardiovascular diseases. (4) Furthermore, tight blood pressure control is associated with a 34% reduction in the risk of macrovascular disease (including myocardial infarction, sudden death, stroke, and peripheral vascular disease), a 44% reduction in the risk of stroke, and a 37% reduction in the risk of microvascular disease. (4)

Diabetes Management and Organization of Diabetes Care

Due to poor compliance with evidence-based recommendations for diabetes management regimens, diabetes and its complications have significantly added to the cost of primary health care and prolonged waiting times for treatment in emergency and surgery departments. (1) While many studies have documented that patients often do not meet the glycemic control targets specified by national and international guidelines, the factors associated with glycemic control are less well studied, one of which is the provider(s) of care. (5)

Multidisciplinary teams refer to “individuals from different disciplines who contribute specialized knowledge in non-hierarchical relationships and who act according to situational demands rather than traditional organizational roles.” (6) Such approaches to care may be particularly important for the management of diabetes and its associated risk factors. Ideal collaborative relationships among health care professionals enable cooperative problem-solving and decision-making that result in synergistic benefits to patient care. (6)

Currently, chronic disease management approaches supported by government involve an interdisciplinary approach to diabetes care and associated risk factor management. According to CDA guidelines, a diabetes health care team should be multi- and interdisciplinary and sustain effective communication with the health care system at large. (1) The core team should consist of at least a family physician and/or diabetes specialist, as well as diabetes educators (registered nurse and registered dietician) (1). It has been noted, however, that increasing the role of allied health care professionals in diabetes care and their collaboration with physicians may represent a more cost-effective option for diabetes management. (7) Multidisciplinary community care may also be a viable option for disease management due to access pressures and time constraints on primary care physicians to manage diabetes.

Several systematic reviews and meta-analyses have examined multidisciplinary care programs, but these have either been limited to a specific components of multidisciplinary care (e.g. intensified education programs), or were conducted as part of a broader disease management program, of which not all teams were multidisciplinary in nature. Furthermore, most reviews are qualitatively reported due to substantial clinical heterogeneity in the interventions being delivered and do not clearly define the intervention of interest, making results difficult to interpret.

Evidence-Based Analysis of Effectiveness

Research Questions

What is the evidence of efficacy of specialized multidisciplinary community care provided by at least a registered nurse, registered dietician, and a physician (primary care and/or diabetes specialist) for the management of type 2 diabetes compared to usual care? [Model 1]
What is the evidence of efficacy of specialized multidisciplinary community care provided by at least a pharmacist and a primary care physician for the management of type 2 diabetes compared to usual care? [Model 2]

Inclusion Criteria

English-language full-reports
Randomized controlled trials (RCTs), systematic reviews or meta-analyses
Published between January 1, 2000 to September 28, 2008
Patients with diabetes, where the majority (i.e., ≥ 80%) of the study population has type 2 diabetes
Adults ≥ 18 years of age
Total sample size of ≥ 30
Studies must describe a specialized multidisciplinary community care intervention, defined as:
Multidisciplinary (two or more health care disciplines)
At least one provider is a specialist in diabetes management
Ambulatory-based health care service provision
Integrated communication and care provision between health care providers
Comparator is usual care, defined as health care provision by non-specialist(s) in diabetes (such as primary care providers) and may include usual referral to other health care professionals or services as necessary
Report clinical outcome measures of glycosylated hemoglobin and/or blood pressure
Studies with a minimum follow-up of 6 months

Exclusion Criteria

Studies where discrete results on diabetes cannot be abstracted
Studies without a clearly defined multidisciplinary specialized community-based intervention
Predominantly home-based interventions
Inpatient-based interventions

Outcomes of Interest

Primary outcomes: glycosylated hemoglobin (HbA1c) levels and systolic blood pressure (SBP).

Search Strategy

A literature search was performed on September 28, 2008 using OVID MEDLINE, MEDLINE In-Process and Other Non-Indexed Citations, EMBASE, the Cumulative Index to Nursing & Allied Health Literature (CINAHL), the Cochrane Library, and the International Agency for Health Technology Assessment (INAHTA) for studies published between January 1, 2000 and September 28, 2008. Abstracts were reviewed by a single reviewer and, for those studies meeting the eligibility criteria, full-text articles were obtained. Reference lists were also examined for any additional relevant studies not identified through the search. Articles with an unknown eligibility were reviewed with a second clinical epidemiologist and then a group of epidemiologists until consensus was established.

Given the high clinical heterogeneity of the articles that met the inclusion criteria, specific models of specialized multidisciplinary community care were examined based on what was reported in the literature, models of care that are currently supported in Ontario, as well as suggestions from an Expert Advisory Panel Meeting held on January 21, 2009. The inclusion criteria were revised to examine specific models of care, as described in the research questions.

Statistical Analyses

Data on study population and intervention characteristics (including multidisciplinary team composition), clinical outcomes of glycemic control and blood pressure, and study design were extracted. Results for studies that reported baseline and final HbA1c (or within-group changes) and/or SBP values were meta-analyzed using a random-effects model.

Meta-analysis of pre-post continuous measurements values (such as HbA1c) presents statistical challenges as studies quite often report only baseline (pre) and final values (post) for intervention and control groups, without reporting between-group changes from baseline to final values. While the absolute difference between pre- and post- can be calculated (final value minus baseline value), the standard deviation of this intra-group difference, necessary for meta-analysis, is often lacking.

In order to account for this discrepancy, baseline values for HbA1c and SBP were meta-analyzed to determine if there were any differences in study populations at baseline. Next, both final values and the change from baseline to follow-up within the intervention and control groups were meta-analyzed. Standard deviations for the change from baseline to final values were generated by imputing varying correlation coefficients (0.25, 0.50, and 0.75) and observing their effect on summary estimates and statistical heterogeneity. The range of correlation coefficients used ensured a wide range of potential correlation coefficients for sensitivity testing. Smaller correlation coefficients (closer to 0) yield more conservative estimates, resulting in an increased standard deviation. This, in turn, generates wider confidence intervals around individual trial effect sizes and results in a slight decrease in the summary of effect size. Using smaller correlation coefficients also decreases statistical heterogeneity by widening confidence intervals. Imputation techniques have been historically shown to have little effect on the summary estimates and conclusions of a meta-analysis. (8) Therefore, all meta-analyses are reported using a correlation coefficient of 0.50.

Assessment of Quality of Evidence

The quality of evidence assigned to individual studies was determined using a modified CONSORT Statement Checklist for Randomized Controlled Trials. (9) The CONSORT Statement was adapted to include three additional quality measures: the adequacy of control group description, significant differential loss to follow-up between groups, and ≥30% study attrition. Individual study quality was defined based on total scores according to the CONSORT Statement checklist: very low (0 to < 40%), low (≥40 to < 60%), moderate (≥60 to < 80%), and high (≥80 to 100%).

The quality of the trials was assessed as high, moderate, low, or very low according to the GRADE Working Group criteria (10;11) and is presented in Table 3.
Quality refers to the criteria such as the adequacy of allocation concealment, blinding and follow-up.
Consistency refers to the similarity of estimates of effect across studies. If there are important and unexplained inconsistencies in the results, our confidence in the estimate of effect for that outcome decreases. Differences in the direction of effect, the magnitude of the difference in effect, and the significance of the differences guide the decision about whether important inconsistency exists.
Directness refers to the extent to which the interventions and outcome measures are similar to those of interest.

Table 3: GRADE Quality Assessment for Specialized Multidisciplinary Community Care for Management of Type 2 Diabetes.

Intervention*	# of Studies	Quality Assessment					Summary of Findings
		Quality Assessment					No. of Patients		Effect (Mean Difference [95% CI])*	Quality
		Design	Quality	Consistency	Directness	Other	Int*	Control	Effect (Mean Difference [95% CI])*	Quality
Outcome: Glycosylated Hemoglobin (HbA1c)
At least a RN, RD and MD	4	RCT High	Serious limitations† Moderate	Consistent Moderate	Direct Moderate	None Moderate	341	313	-1.00 [-1.27, -0.73]	Moderate
At least a pharmacist and PCP	2	RCT High	No serious limitations High	Consistent High	Direct High	None‡ High	148	134	-1.05 [-1.57, -0.52]	High
Outcome: Systolic Blood Pressure
At least a RN, RD and MD	2	RCT High	Serious limitations§ Moderate	Unexplained heterogeneity Low	Direct Low	Imprecise or sparse data Unlikely publication bias Very-low	133	197	-2.04 [-13.80, 9.74]	Very-low
At least a pharmacist and PCP		RCT High	Serious limitations¥ Moderate	Consistent Moderate	Direct Moderate	None Moderate	214	214	-7.13 [-11.78, -2.48]	Moderate

Open in a new tab

MD, primary care physician and/or diabetes specialist; PCP, primary care physician; RD, registered dietician; RN, registered nurse; CI, confidence interval; Int, intervention; RCT, randomized controlled trial.

^†

Unclear allocation concealment in 2 studies (12;13); potential for control group contamination in 1 study, where the same physician provided care to intervention and control groups (14); > 30% loss to follow-up in 1 study (12); not analyzed using intention-to-treat in 2 studies (12;15); frequency of testing of HbA1c amongst controls may have effected improvement in glycemic control in 1 study. (14)

^‡

Studies were powered to detect a change in HbA1c.

^§

Unclear allocation concealment in 1 study (16); not analyzed using intention-to-treat in 1 study (15); not powered to detect a change in blood pressure in both studies (15;16); no description of methods for obtaining blood pressure measurement in 1 study. (16)

^¥

All blood pressure outcome assessment were obtained by automated blood pressure monitors; however, blinding of outcome assessor in only 1 study (17); description of frequency and methods for obtaining blood pressure measurement in only 1 study (18), where an average of 5 measurements were taken 1 minute apart

As stated by the GRADE Working Group, the following definitions of quality were used in grading the quality of the evidence:

High	Further research is very unlikely to change confidence in the estimate of effect.
Moderate	Further research is likely to have an important impact on confidence in the estimate of effect and may change the estimate.
Low	Further research is very likely to have an important impact on confidence in the estimate of effect and is likely to change the estimate.
Very Low	Any estimate of effect is very uncertain

Open in a new tab

Results of Evidence-Based Analysis

Based on a systematic literature search of six electronic databases, 2,116 unique citations were identified (published between January 2000 and October 2008). Following the title and abstract review, 325 full-text articles were retrieved and reviewed for more detailed evaluation of study objectives and methodology to determine inclusion. Of these, 295 articles were excluded (154 for inappropriate intervention or control group, 128 because of study design or type of report, seven for inappropriate population, three for inadequate follow-up, and three for inappropriate outcomes). Of the remaining full-text studies reviewed, 22 RCTs were eligible for inclusion based on having at least two health care disciplines in the multidisciplinary team. Upon closer examination, however, only five RCTs involved specialized multidisciplinary community care provided by at least a registered nurse, registered dietician, and a physician (primary care and/or specialist), while three RCTs involved specialized multidisciplinary community care provided by at least a pharmacist and primary care physician.

Nine systematic literature reviews were finally identified (eight through systematic search and one through manual searching) that focused on concepts relating to multidisciplinary diabetes care. All the identified RCTs were categorized as Level 1 evidence (Table 4).

Table 4: Quality of Evidence of Included Studies*.

Study Design	Level of Evidence	Number of Eligible Studies
Study Design	Level of Evidence	Any Multidisciplinary Team	At least a Nurse, Dietician, and Physician	At least a Pharmacist and Primary Care Physician
Large RCT, systematic review of RCTs	1	31	5	3
Large RCT unpublished but reported to an international scientific meeting	1(g)	0	0	0
Small RCT	2	0	0	0
Small RCT unpublished but reported to an international scientific meeting	2(g)	0	0	0
Non-RCT with contemporaneous controls	3a	0	0	0
Non-RCT with historical controls	3b	0	0	0
Non-RCT presented at international conference	3(g)	0	0	0
Surveillance (database or register)	4a	0	0	0
Case series (multisite)	4b	0	0	0
Case series (single site)	4c	0	0	0
Retrospective review, modeling	4d	0	0	0
Case series presented at international conference	4(g)	0	0	0

Open in a new tab

^†

For each included study, levels of evidence were assigned according to a ranking system based on a hierarchy proposed by Goodman. (19) An additional designation “g” was added for preliminary reports of studies that have been presented at international scientific meetings. Non-RCT, clinical trial that is not randomized, e.g. a cohort study; RCT, randomized controlled trial.

A diagram of the literature search flow is presented in Appendix 2.

Summary of Existing Evidence: Systematic Reviews and Meta-Analyses

Nine systematic reviews were identified (eight through systematic literature search and one through manual-searching) that focused on concepts relating to multidisciplinary programs for diabetes care. Of these, seven were narrative systematic reviews, one was a meta-analysis, and one was a meta-regression analysis. A summary of the systematic reviews, including the search years, number of trials included, objective, and applicability to the present research questions is presented in Appendix 3. The majority of these reviews were not applicable to the present analysis as they were part of a broader disease management program, quality improvement strategies in diabetes management, case management, did not report a clinical outcome of glycosylated hemoglobin or systolic blood pressure, or did not restrict inclusion criteria to interventions that were specialized and multidisciplinary in nature. No systematic reviews were identified that examined specialized multidisciplinary care provided by at least a registered nurse, registered dietician and physician and none had the same inclusion and exclusion criteria as the current review.

A meta-analysis of 24 studies published between 1987 and 2001 was conducted by Knight and colleagues, which focused on broader disease management programs for diabetes. (20) Inclusion was not restricted to programs that were multidisciplinary in nature. Upon pooling the study results, diabetes disease management programs resulted in a 0.49% reduction in HbA1c (95% CI, -0.56 to -0.41%). The results of this meta-analysis are, however, not interpretable as there was significant clinical and statistical heterogeneity (p < 0.001 for test for homogeneity) amongst the disease management programs, with no attempt made for exploratory subgroup analysis.

A meta-regression of 66 studies published between 1966 and 2006 was conducted by Shojania et al., focusing on quality improvement strategies for the management of diabetes. (21) Most quality improvement strategies examined resulted in small to modest improvements in glycemic control. Two of the strategies that may have involved specialized multidisciplinary care resulted in more robust improvements in HbA1c. Specifically, team changes resulted in a reduction in HbA1c of -0.67% (95% CI, -0.91% to -0.43%) and case management in a reduction in HbA1c of -0.52% (95% CI, -0.73% to -0.31%). Yet although team changes included interventions involving the addition of a team member (i.e., shared care), or the use of multidisciplinary teams, interventions involving expansion or revision of professional roles were also included, such as nurses or pharmacists who played a more active role in medication management. The results of the team changes analysis may, therefore, not be directly applicable to the current review.

Summary of Findings

Of the 2,116 citations reviewed, 22 RCTs were eligible for inclusion based on having at least two health care disciplines in the multidisciplinary team (Table 6). Overall, there was substantial clinical heterogeneity across the 22 RCTs with respect to patient populations, the composition of the health care team, the various components of care being provided, and the outcomes reported. As such, two specific models of care were focused on in this analysis:

Table 6: Specialized Multidisciplinary Teams Identified through Broad Literature Search.

Study	Health Care Team*	Model of Care Examined
California Medi-Cal Diabetes Study Group, 2004 (14)	RN, RD, E, PCP, CDE	At least RN, RD, MD
Choe, et al, 2005 (22)	Pharm, PCP	At least Pharm, PCP
Gaede, et al, 2001 (13)	RN, RD, MD	At least RN, RD, MD
Gabbay, et al, 2006 (23)	dRN, PCP	NA
Gary, et al, 2003 (24)	RN (CDE in training), CHW, MD	NA
Groeneveld, et al, 2001 (16)	dRN (CDE), RD, PCP	At least RN, RD, MD
Hiss, et al, 2007 (25)	dRN, PCP	NA
Johansen, et al, 2007 (15)	dRN, RD, D, Physio	At least RN, RD, MD
Krein, et al, 2004 (26)	NP, PCP	NA
Litaker, et al, 2003 (27)	NP, PCP	NA
Maislos, et al, 2004 (12)	dRN (CDE), RD, D	At least RN, RD, MD
McLean, et al, 2008 (18)	Pharm, PCP, dRN	At least Pharm, PCP
McMurray, et al, 2002 (28)	dRD, MD (N, PCP, I, or E)	NA
O’Hare, et al, 2004 (29)	dRN, CHW, usual care	NA
Piette, et al, 2001 (30)	RN, PCP	NA
Rothman, et al, 2005 (17)	Pharm (CDE), PCP	At least Pharm, PCP
Shea, et al (31)	dRN, D, PCP	NA
Shibayama, et al, 2007 (32)	dRN (CDE), MD	NA
Smith, et al, 2004 (33)	dRN, PCP	NA
Soja, et al, 2007 (34)	RNs, MDs (I, C)	NA
Taylor, et al, 2003 (35)	dRN, MD	NA
Wolf, et al, 2004 (36)	RD, PCP	NA

Open in a new tab

C, cardiologist; CDE, certified diabetic educator; CHW, community health worker (non-health care professional); D, diabetologist; dRN, diabetes specialist nurse; dRD, diabetes specialist dietician; E, endocrinologist; I, internist; MD, physician (unspecified specialty); N, nephrologist; NP, nurse practitioner; PCP, primary care physician; Pharm, pharmacist; Physio, physiotherapist; RD, registered dietician; RN, registered nurse; NA, not applicable.

Model 1: care provided by a registered nurse, registered dietician and physician (primary care and/or specialist).

Model 2: care provided by a pharmacist and primary care physician.

Summaries of selected studies are presented in Table 7 (pages 19-20) and Table 8 (page 21), highlighting their patient demographic and design details as they relate to the two alternate models of care.

Table 7: Summary of Study Characteristics: Model 1.

Study, Design (N), Country	Inclusion Criteria^†	Intervention Group			Setting	Control^¶	Outcomes^#	Length of FU (Freq. of FU)**	Study Quality
Study, Design (N), Country	Inclusion Criteria^†	Care provider	Types of Interventions Delivered	Method of Care Delivery	Setting	Control^¶	Outcomes^#	Length of FU (Freq. of FU)**	Study Quality
MODEL OF CARE: 1. At least a registered nurse, registered dietician and physician (primary care and/or specialist)
California Medi-Cal Diabetes Study Group, 2004 (14) RCT (N = 335) USA	T2DM (>1y duration), age ≥18 y, HbA1c ≥7.5%	RN, RD, E, PCP, CDE	Diabetes education Diet counselling Pharmacotherapy management Exercise advice or training Promotion of self-care, behavioural modification or problem-solving skills Integration of multidisciplinary team with primary care Case management or care coordination Smoking cessation counselling Psychosocial counselling Blood glucose self-monitoring Program retention strategies Program discharge plan Foot care Weight management Prevention and management of retinopathy, nephropathy, hypertension, dyslipidemia & CVD	Clinic visits (Individual)	Community outpatient clinics (n=3)	Usual care (by PCP)	Primary: HbA1c	36 months (every 6 months)	High
Gaede, et al, 2001 (13) RCT (N = 149) Denmark	T2DM, age 45-65y	RN, RD, MD	Diabetes education (structured program) Diet counselling Pharmacotherapy management Exercise advice or training Promotion of self-care, behavioural modification or problem-solving skills Smoking cessation counselling	Clinic visits (Individual & Group)	Hospital outpatient clinic	Usual care (by PCP)	†† HbA1c, total-and HDL-cholesterol, triglycerides, body weight, current smokers, daily dietary intake, exercise, use of glucose- or lipid-lowering drugs	Mean: 3.8 years (every 3 months)	Moderate
Groeneveld, et al, 2001 (16) Cluster RCT N = 246 Netherlands	T2DM, age <76y, treated by a PCP	dRN (CDE), RD, PCP	Diabetes education Diet counselling Pharmacotherapy management Integration of multidisciplinary team with primary care	Clinic visits (Individual)	Primary carepractices (n=15)	Usual care (by PCP)	†† HbA1c, FBG, lipids, BP, weight	12 months (every 3 months)	Moderate
Johansen, et al, 2007 (15) RCT N = 106 Norway	T2DM, age 18-75 y, Caucasian, ≥ 1 CV risk factor	dRN, RD, D, Physio	Diabetes education (structured program) Diet counselling Pharmacotherapy management Exercise advice or training Remuneration for gymnasium membership	Clinic visits (Individual & Group)	Hospital outpatient clinic	Usual care (by PCP)	†† HbA1c; BP; FBG; total, HDL and LDL cholesterol; triglycerides; microalbumin-uria; leisure-time activity; HRQoL	24 months (every 3 months)	High
Maislos, et al, 2004 (12) Cluster RCT N = 82 Israel	T2DM, poorly controlled HbA1c (≥10%)	dRN (CDE), RD, D	Diabetes education (structured program) Diet counselling Pharmacotherapy management Exercise advice or training Promotion of self-care, behavioural modification or problem-solving skills Integration of multidisciplinary team with primary care Blood glucose self-monitoring	Clinic visits (Individual)	Primary carepractices (n=2)	Usual care (by PCP, RN)	†† HbA1c, compliance in attending clinic	6 months (as needed)	Low

Open in a new tab

RCT, randomized controlled trial

^†

BP, blood pressure; CV, cardiovascular; HbA1c, glycosylated hemoglobin; mo, months; PCP, primary care physician; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus; y, years

^‡

NA, not available (contacted study author; data unavailable); NR, not reported

^§

^‖

PCP, primary care physician

^¶

MD, physician; PCP, primary care physician; Pharm, pharmacist; RN, nurse

blood pressure; FBG, fasting blood glucose; HDL, high-density lipoprotein; HRQoL, health-related quality of life; LDL, low-density lipoprotein cholesterol

^**

FU, follow-up

^††

No specification of primary or secondary outcomes

Table 8: Summary of Study Characteristics: Model 2.

Study, Design (N), Country	Inclusion Criteria^†	Intervention Group			Setting	Control^¶	Outcomes^#	Length of FU (Freq. of FU)**	Study Quality
Study, Design (N), Country	Inclusion Criteria^†	Care provider	Types of Interventions Delivered	Method of Care Delivery	Setting	Control^¶	Outcomes^#	Length of FU (Freq. of FU)**	Study Quality
MODEL OF CARE: 2. At least a pharmacist and primary care physician
Choe, et al, 2005 (22) RCT (N = 80) USA	T2DM, HbA1c ≥8.0%, age ≤70 y [excluded patients with severe co-morbidity)	Pharm, PCP	Diabetes education Pharmacotherapy management (with prior approval by PCP) Promotion of self-care, behavioural modification or problem-solving skills Integration of pharmacist within primary care Case management or care coordination	Clinicvisits, telephone follow-up (Individual)	Primary care clinic	Usual care (by PCP)	Primary: HbA1c Secondary: process measures (LDL, retinal exam, urine microalbumin-uria screening, monofilament testing for neuropathy)	12 months (monthly)	High
McLean, et al, 2008 (18) RCT (N = 227) Canada	Diabetes, adults, BP >130/80 mm Hg on 2 screening visits 2 weeks apart	Pharm, PCP, dRN	Diabetes education (structured program) Pharmacotherapy management (with prior approval by PCP) Promotion of self-care, behavioural modification or problem-solving skills Integration of pharmacist within primary care CVD risk reduction counselling	Clinic visits (Individual)	Community	Usual care (by RN or Pharm) + minimal education	Primary: BP Secondary: BPtargets (≤130/80 mm Hg), anti-hypertensive drug therapy, angiotensin-converting enzyme inhibitor	6 months (every 6 weeks)	High
Rothman, et al, 2005 (17) RCT (N = 217) USA	T2DM, age ≥ 18y, HbA1c ≥8%, English-speaking, life expectancy 〉 6 months	Pharm (CDE), PCP	Structured diabetes education Pharmacotherapy management Promotion of self-care, behavioural modification or problem-solving skills Integration of pharmacist within primary care Case management or care coordination Clinical registry tracking for uncontrolled clinical outcomes	Clinic visits, telephone or in-person follow-up (Individual)	Primary care clinic	Usual care (by PCP)	Primary: BP, HbA1c, aspirin use at 6 and 12 months Secondary: diabetes knowledge, satisfaction, use of clinical services, adverse events	12 months (every 2-4 weeks)	High

Open in a new tab

RCT, randomized controlled trial;

^†

BP, blood pressure; CV, cardiovascular; HbA1c, glycosylated hemoglobin; mo, months; PCP, primary care physician; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus; y, years;

^‡

NA, not available (contacted study author; data unavailable); NR, not reported;

^§

CDE, certified diabetic educator; D, diabetologist; dRN, diabetes specialist nurse; E, endocrinologist; HCP, health care professional; MD, physician (unspecified specialty); PCP, primary care physician; Pharm, pharmacist; Physio, physiotherapist; RD, registered dietician; RN, registered nurse; PCP, primary care physician; MD, physician; PCP, primary care physician; Pharm, pharmacist; RN, nurse; blood pressure; FBG, fasting blood glucose; HDL, high-density lipoprotein; HRQoL, health-related quality of life; LDL, low-density lipoprotein cholesterol;

^**

FU, follow-up;

^††

No specification of primary or secondary outcomes.

Summary of Multidisciplinary Community Care Model 1

Summary of Participant Demographics across studies

A total of 918 study participants with type 2 diabetes were randomized and analyzed across the three studies examining specialized multidisciplinary community care provided by at least a nurse, dietician and physician (range: 82 to 335). Study demographics reported were:

Age range of study participants: 55 to 63 years
Percentage of female participants: reported in four studies with a mean of 54% and a range of 23% to 42.6%
BMI: reported in three studies with a range of 28.6 to 33.1 kg/m²
Baseline HbA1c: reported in four studies with a mean range of 7.5% to 12.9%
SBP: reported in three studies with a mean range of 130 to 149 mm Hg
Duration of diabetes: reported only four studies with a range of 3 to 12 years.
Total cholesterol: reported in four studies with a range of 5.0 to 6.2 mmol/L. Three studies also reported HDL cholesterol values, which ranged from 1.01 to 1.3 mmol/L.
Ethnicity: reported in one study in which more than 50% of the population was an ethnic minority (i.e. African American, Hispanic, or other).
Smoking status: reported in three studies in which 8% to 38% were self-reported current smokers.

The study demographics are summarized in Table 9 (pages 22-23)

Table 9: Summary of Participant Demographic Characteristics of Included Studies.

Study, N	Sex: Female (%)	Age (years)*	Ethnicity (%)	Duration of Diabetes (years)*	BMI (kg/m2)*	Baseline HbA1c (%)*	Baseline SBP (mm Hg)*	Total Cholesterol*	HDL Cholesterol*	Smoker Status (%)
MODEL OF CARE: 1. At least a registered nurse, registered dietician and physician (primary care and/or specialist)
California Medi-Cal Diabetes Study Group, 2004 (14) N = 335	Intervention: 72.6 Control: 70.9	Intervention: 57.0 ± 0.9 Control: 56.9 ± 1.0	Intervention: African-Am:16.1 Hispanic: 39.2 Caucasian: 34.9 Other: 9.7 Control: African-Am: 15.7 Hispanic: 38.4 Caucasian: 36.0 Other: 9.9	Intervention: 10.3±0.8 Control: 12.0±0.8	Intervention: 33.1±0.8 Control: 31.5±0.8	Mean ± SE: Intervention:9.6±0.1 Control:9.7±0.1	Mean ± SE: Intervention: 136 ± 2 Control: 134 ± 1	Mean ± SE: Intervention: 210.0±3.3 mg/dl Control: 212.1±3.7 mg/dl	Mean ± SE: Intervention:41.9 ± 1.0 mg/dl Control:43.0 ± 1.1 mg/dl	Intervention: 14.8 Control: 13.0
Gaede, et al, 2001 (13) N = 149	NR	55.1 ± 7.2	NR	Median (IQR): 6 (4-10)	NR	Intervention: 8.4±1.5 Control: 8.8±1.7	NR	Intervention: 5.4±1.0 mmol/L Control: 5.8±1.3 mmol/L	Intervention: 1.03±0.2 mmol/L Control: 1.01±0.3 mmol/L	Intervention: 38.4 Control: 34.2
Groeneveld, et al, 2001 (16) N = 246	Intervention: 65.9 Control: 53.5	Intervention: 62.7 ± 11 Control: 62.3 ± 10	NR	Intervention: 4.1 ± 3.7 Control: 4.6 ± 4.0	NR	NA	Intervention: 137±27 Control: 149±24	Intervention:6.2 ± 1.2 mmol/L Control:6.2 ± 1.3 mmol/L	NR	NR
Johansen, et al, 2007 (15) N = 106	Intervention: 28 Control: 23	Intervention: 59 ± 9 Control: 58 ± 11	NR	Median (IQR): Intervention: 4 (1-10) Control: 3 (1-12)	Median (min, max): Intervention: 30.6 (22.6, 484) Control: 28.6 (16.1, 42.3)	Intervention:7.5± 1.5 Control:7.6± 1.6	Intervention: 136 ± 16 Control: 130 ± 13	Intervention:5.0 ± 1.0 mmol/L Control:5.0 ± 0.9 mmol/L	Intervention:1.3 ± 0.4 mmol/L Control:1.3 ± 0.4 mmol/L	Intervention: 8 Control: 15
Maislos, et al, 2004 (12) N = 82	Intervention: 50 Control: 65	Intervention: 58 ± 14 Control: 63 ± 9	NR	NR	Intervention: 30.8 ± 3.6 Control: 30.8 ± 3.0	Intervention: 12.9 ± 3.4 Control: 12.6 ± 2.9	NR	NR	NR	NR
Choe, et al, 2005 (22) N = 80	Intervention: 51.2 Control: 53.9	Intervention: 52.2 ± 11.2 Control: 51.0 ± 9.0	Intervention: African-Am: 17.1 Caucasian: 80.5 Other: 2.4 Control: African-Am: 12.8 Caucasian: 71.8 Other: 5.1 Unknown: 10.3	NR	NR	Intervention:10.1± 1.8 Control:10.2± 1.7	NR	NR	NR	NR
McLean, et al, 2008 (18) N = 227	Intervention: 34.8 Control: 65.5	Intervention: 66.2 ± 11.3 Control: 63.7 ± 12.7	NR	NR	Intervention: 31.7 ± 6.0 Control: 31.6 ± 7.9	NR	Intervention: 142.5 ± 15.5 Control: 139.9 ± 11.9	NR	NR	Intervention: 9.6 Control: 10.77
Rothman, et al, 2005 (17) N = 217	Intervention: 56 Control: 56	Intervention: 54 ± 13 Control: 57 ± 11	Intervention: African-Am: 70 Control: African-Am: 59	Intervention:8 ± 9 Control: 9 ± 9	Intervention: 35 ± 9 Control: 34 ± 8	Intervention: 10.8 ± 2.1 Control: 10.7 ± 2.5	Intervention: 141.2 ± 21.8 Control: 137.4 ± 21.2	Intervention:217.0± 86.5 mg/dl Control:207.1± 64.1 mg/dl	NR	NR

Open in a new tab

All values summarized as mean ± SD unless otherwise specified

Study Characteristics and Setting

All studies included for examination were RCTs published between 2001 and 2007 with two being cluster-RCTs (Table 8). The studies were conducted in a variety of geographical locations, including Europe (three studies), the USA (one study), and Israel (one study); no Canadian studies were identified. The quality of the individual studies varied, with two studies being of high quality, two of moderate quality, and one of low quality. Differences in quality were predominantly attributable to inadequate descriptions of the randomization process, lack of allocation concealment, sample size calculations, and/or a lack of intention-to-treat analyses. All studies were conducted in a community outpatient setting, where two were specifically conducted in outpatient hospital clinics and two in primary care practices.

Intervention Characteristics of Diabetes Programs

Specialized Multidisciplinary Health Care Professional Team

The composition of the specialized multidisciplinary team varied across studies, differing in which professional was termed the ‘diabetes specialist’ and what allied health care providers complemented the core diabetes care team. Two studies involved a primary care physician as part of the multidisciplinary team. Three studies involved a certified diabetic educator (which was a diabetes specialist registered nurse in two instances). Three studies involved diabetes specialist physicians (two diabetologists, one endocrinologist), while one study supplemented the core diabetes health care team with a physiotherapist.

Interventional Characteristics Delivered within Diabetes Programs

All programs were multifaceted in nature, involving at least four interventional components in various combinations. At a minimum, all programs included diabetes education, diet counselling, and pharmacotherapy advice and management. In four of the diabetes programs, exercise advice and training was also provided by the multidisciplinary team. Three studies involved a structured education program, promotion of self-care, behaviour modification or problem-solving skills, and integration of the multidisciplinary team with primary care. Other components included in some programs were smoking cessation counselling, case management, and psychosocial counselling.

Method of Care Delivery and Length and Frequency of Follow-up

All studies involved delivery via patient clinic visits and two studies also used group care or education sessions as an adjunct to clinic visits. Length of follow-up ranged from 6 months to 3.8 years with three studies having a follow-up every 3 months and two having follow-ups as needed and every 6 months.

Comparator Groups

All studies involved comparing specialized multidisciplinary teams (at least a nurse, dietician and physician) compared to usual care provided by a primary care physician. In one trial, usual care was provided by both primary care physicians and nurses. In this instance, however, neither the nurses nor primary care physician were specialty trained in diabetes management (Table 8). (12)

Outcomes

All studies used HbA1c as an outcome, but only one study did so as their primary outcome of interest. Two studies reported SBP as a study outcome (other outcomes were displayed previously in Table 8).

Results: HbA1c

Four of the five studies were eligible for inclusion in the meta-analyses of HbA1c results (baseline HbA1c values were not reported in Groeneveld, et al. 2001). (16) Among these, there was no significant difference in the mean baseline HbA1c values [-0.03% (95% CI, -0.36, 0.29)], as shown in Figure 1a. All five trials did, however, report final mean HbA1c values (including Groeneveld, et al. 2001), as shown in Figure 1b. Based on the reported values, there was a significant reduction in HbA1c associated with care Model 1 compared to usual care [-0.94% (95% CI, -1.32, -0.56)] with moderate statistical heterogeneity (I²=65%). Figure 1c presents the mean change in HbA1c from baseline to follow-up between groups for Model 1 compared to usual care. Overall, care Model 1 resulted in a reduction in HbA1c of 1.0% (95% CI, -1.27, -0.73) compared to usual care, which is considered to be both statistically and clinically significant. Furthermore the statistical heterogeneity associated with this comparison was minimal (I²=4%). The estimate of effect did not vary greatly based on subgroup analysis of moderate-to-high-quality evidence, with Model 1 resulting in an overall reduction in HbA1c of 0.91 % (95% CI, -1.19, -0.62) (Figure 1d).

Results: Systolic Blood Pressure

Two of the five studies were eligible for inclusion in the meta-analyses of SBP. The mean baseline SBP values for these two studies are presented in Figure 2a and the final mean SBP values are presented in Figure 2b.

Overall, there was no significant difference in the baseline SBP values between the studies [-2.97 mm Hg (95% CI, -20.61, 14.67)]. Based on the reported values, there was also no difference in SBP associated with care Mode 1 compared to usual care [-4.93 mm Hg (95% CI, -10.80, 0.95)], with moderate statistical heterogeneity associated with the results (I²=64%).

Figure 2c presents the mean change in SBP from baseline to follow-up between groups for care Model 1 compared to usual care. Overall, the model had no effect on the mean change in SBP between groups (-2.04 mm Hg [95% CI, -13.80, 9.72]. However, because this is based on very low quality evidence (according to GRADE, Table 3), the estimate of effect is uncertain; further, there was high statistical heterogeneity associated with this comparison (I²=89%).

Summary of Multidisciplinary Community Care Model 2

Summary of Participant Demographics across studies

A total of 524 study participants with type 2 diabetes were randomized and analyzed across three studies examining specialized multidisciplinary community care provided by at least a pharmacist and primary care physician (range: 80 to 227). Study demographics reported were:

Age range of study participants: mean of 51 to 63.7 years
Percentage of female participants: mean of 53% and a range of 35% to 66%
BMI: reported in two studies with a range of 31.6 to 35.0 kg/m²
Baseline HbA1c: reported in two studies with a mean range of 10% to 11%
SBP: reported in two studies with a mean range of 137.4 to 142.5 mm Hg
Duration of diabetes: reported only in one study with a mean of approximately 8 years.
Total cholesterol: reported in one study with a mean of 207.1 to 217.0 mg/dL (no trials reported HDL cholesterol values)
Ethnicity: reported in two studies with Caucasians making up 80% of study participants in one study and African Americans making up more than 50% of study participants in the second
Smoking status: reported in one study with 9 to 11% of patients being self-reported current smokers.

The study demographics were summarized previously in Table 9 (pages 22-23).

Study Characteristics and Setting

The included studies were all RCTs conducted in North America and published between 2005 and 2008; one was a Canadian study. All three studies were also conducted in the community outpatient setting, with two specifically conducted in primary care practices and one in community pharmacies. According to the CONSORT Statement for Randomized Controlled Trials, the studies were all of high quality.

Intervention Characteristics of Diabetes Programs

Specialized Multidisciplinary Health Care Professional Team

In each of the studies, the composition of the specialized multidisciplinary team was fairly homogeneous, with each involving a core team consisting of a pharmacist and primary care physician. In one trial, two pharmacists were involved in the team, one of whom was a certified diabetic educator. (17) Another study involved a diabetes specialist registered nurse in addition to the core team. (18)

Interventional Characteristics Delivered within Diabetes Programs

All programs were multifaceted in nature, involving at least five interventional components in various combinations (see Table 8, page 21). All programs included diabetes education and counselling, integration of the pharmacist with primary care, pharmacotherapy advice and management, and the promotion of self-care, behaviour modification or problem-solving skills. Two studies involved a structured education program, case management, or a diabetes care coordinator. Other components of care included cardiovascular disease risk counselling and a tracking registry of patient outcomes.

Method of Care Delivery and Length and Frequency of Follow-up

All studies involved care delivered by individual patient clinic visits. In addition to clinic visits, two studies involved telephone follow-up. Length of follow-up ranged from 6 to 12 months, while the frequency of follow-up ranged from 2 to 6 weeks.

Comparator Groups

Two studies involved comparing specialized multidisciplinary community care provided by at least a pharmacist and primary care physician to usual care provided by a primary care physician. In the third study, usual care was provided by a registered nurse or pharmacist, plus minimal diabetes education in pamphlet form. (18) Neither the nurse nor pharmacist in this study had specialty training in diabetes management, nor did they work together to provide care as a multidisciplinary team.

Outcomes

Two studies used HbA1c as a primary outcome (17;22), while two used SBP (Rothman et al., 2005 used both metrics). (17;18) Other outcomes included diabetes process measures, achievement of blood pressure targets, aspirin and drug utilization, diabetes knowledge, diabetes satisfaction, use of clinical services, and adverse events.

Results: HbA1c

Two of the three studies were eligible for inclusion in the meta-analyses of HbA1c. As shown in Figure 3a, there was no significant difference in the mean baseline Hba1c values between studies [0.10% (95% CI, -0.40, 0.60)] for care Model 2. Following diabetes management with the model, patients achieved a significant reduction in HbA1c compared to usual care [-0.95% (95% CI, -1.43, -0.46)] with no statistical heterogeneity (I²=0%) (Figure 3b).

Figure 3c presents the mean change in HbA1c from baseline to follow-up between groups for studies comparing care Model 2 to usual care. Overall, the model resulted in a reduction in mean HbA1c of 1.05% (95% CI, -1.57, -0.52) compared to usual care, which is considered to be both statistically and clinically meaningful. No statistical heterogeneity associated with this comparison (I²=0%).

Results: Systolic Blood Pressure

Two of the three studies examined were eligible for inclusion in the meta-analysis of SBP. Figure 4a presents the baseline SBP values for studies applying care Model 2. Overall, there was no significant difference in the mean baseline SBP values between studies [2.96 mm Hg (95% CI, -0.17, 6.09)]. Figure 4b presents the final mean SBP values for both trials, which reported a significant reduction in SBP associated with care Model 2 compared to usual care [-6.10 mm Hg (95% CI, -11.41, -0.79)] with no statistical heterogeneity (I²=0.02%).

Figure 4c presents the mean change in SBP from baseline to follow-up between groups for care Model 2 and usual care. Overall, the model resulted in a significant and clinically meaningful reduction in mean SBP of 7.13 mm Hg (95% CI, -11.78, -2.48). Moderate statistical heterogeneity was associated with this comparison (I²=46%).

Conclusions

Model 1: Specialized multidisciplinary community care provided by at least a registered nurse, registered dietician and physician (primary care and/or specialist) for the management of type 2 diabetes:
- - Has demonstrated a statistically and clinically significant reduction in HbA1c compared to usual care based on moderate quality evidence.
- - Has demonstrated an uncertain estimate of effect on SBP compared to usual care based on very-low quality evidence.
Model 2: Specialized multidisciplinary community care provided by at least a pharmacist and primary care for the management of type 2 diabetes:
- - Has demonstrated a statistically and clinically significant reduction in HbA1c compared to usual care based on high quality evidence.
- - Has demonstrated a statistically and clinically significant reduction in SBP compared to usual care based on moderate quality evidence.
For both models, the evidence does not suggest a preferred setting for care delivery (i.e. primary care vs. hospital outpatient clinic vs. community clinic).

Appendices

Appendix 1: Search Strategies

Search date: September 28, 2008

Databases searched: OVID MEDLINE, MEDLINE In-Process and Other Non-Indexed Citations, EMBASE, Cochrane Library, CINAHL, INAHTA/CRD

Database: Ovid MEDLINE ® <1996 to September Week 3 2008>

exp Intermediate Care Facilities/ (223)
(intermedia* adj2 care).ti,ab. (514)
exp ambulatory care/ (15683)
exp Ambulatory Care Facilities/ (14875)
exp Outpatients/ (3629)
((outpatient* or ambulatory) adj2 (care* or service* or clinic* or facility or facilities)).ti,ab. (15858)
exp Patient Care Team/ (22124)
exp Nursing, Team/ (624)
exp Cooperative Behavior/ (12319)
exp Interprofessional Relations/ (20749)
exp “Delivery of Health Care, Integrated”/ (5240)
team*.ti,ab. (33586)
(multidisciplin$ or multi-disciplin$ or interdisciplin$ or inter-disciplin$ or collaborat$ or cooperat$ or co-operat$ or multi?special$).ti,ab. (92458)
(integrat$ or share or shared or sharing).ti,ab. (167984)
exp Community Health Services/ (181030)
exp Program Evaluation/ (30015)
exp “episode of care”/ (910)
exp Professional Role/ (35965)
exp Primary Health Care/ (34098)
exp “Continuity of Patient Care”/ (6191)
exp Disease Management/ (6014)
disease management program*.ti,ab. (794)
(patient care adj2 manage$).ti,ab. (245)
exp Case Management/ or exp Subacute Care/ (6515)
(care adj2 model*).ti,ab. (2957)
exp Program Development/ (11519)
or/1-26 (564400)
limit 27 to yr=“2000 - 2008” (423967)
limit 28 to (english language and humans) (318172)
limit 29 to (controlled clinical trial or meta analysis or randomized controlled trial) (14433)
exp Technology Assessment, Biomedical/ or exp Evidence-based Medicine/ (34042)
(health technology adj2 assess$).mp. [mp=title, original title, abstract, name of substance word, subject heading word] (617)
(meta analy$ or metaanaly$ or pooled analysis or (systematic$ adj2 review$)).mp. or (published studies or published literature or medline or embase or data synthesis or data extraction or cochrane).ab. (64322)
exp Random Allocation/ or random$.mp. [mp=title, original title, abstract, name of substance word, subject heading word] (367054)
exp Double-Blind Method/ (52682)
exp Control Groups/ (696)
exp Placebos/ (9167)
(RCT or placebo? or sham?).mp. [mp=title, original title, abstract, name of substance word, subject heading word] (93164)
or/30-38 (473324)
29 and 39 (38646)
(diabet* adj2 (program* or clinic* or center* or centre*)).mp. [mp=title, original title, abstract, name of substance word, subject heading word] (3130)
limit 41 to (english language and humans and yr=“2000 - 2008”) (2165)
42 and 39 (414)
exp Diabetes Mellitus, Type 2/ (37974)
((ketosis resistant or adult onset or slow onset or maturity onset or non?insulin dependent or stable or type 2 or type II) adj2 (diabet$ or DM)).mp. [mp=title, original title, abstract, name of substance word, subject heading word] (46955)
(t2dm or niddm).mp. [mp=title, original title, abstract, name of substance word, subject heading word] (4237)
46 or 45 or 44 (47508)
40 and 47 (783)
43 or 48 (1100)

Database: EMBASE <1980 to 2008 Week 39>

(intermedia* adj2 care).ti,ab. (631)
exp ambulatory care/ (12187)
exp Outpatient Department/ (9466)
exp outpatient care/ (12499)
((outpatient* or ambulatory) adj2 (care* or service* or clinic* or facility or facilities)).ti,ab. (20467)
exp TEAM NURSING/ (6)
exp Cooperation/ (13299)
exp TEAMWORK/ or team*.ti,ab. (41041)
exp Integrated Health Care System/ (231)
(multidisciplin$ or multi-disciplin$ or interdisciplin$ or inter-disciplin$ or collaborat$ or cooperat$ or co-operat$ or multi?special$).ti,ab. (116921)
(integrat$ or share or shared or sharing).ti,ab. (208598)
exp Case Management/ (454)
exp Rehabilitation Care/ (2739)
exp community care/ (23465)
exp Social Care/ (34975)
exp ambulatory care nursing/ (5)
exp primary health care/ (41469)
*Disease Management/ (254)
disease management program*.ti,ab. (869)
(patient care adj2 manage$).ti,ab. (196)
exp Program Development/ (753)
(care adj2 model*).ti,ab. (2336)
exp Health Program/ (53182)
or/1-23 (511612)
limit 24 to (human and english language and yr=“2000 - 2009”) (194121)
Randomized Controlled Trial/ (162835)
exp Randomization/ (26273)
exp RANDOM SAMPLE/ (1261)
exp Biomedical Technology Assessment/ or exp Evidence Based Medicine/ (292930)
(health technology adj2 assess$).mp. [mp=title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer name] (645)
(meta analy$ or metaanaly$ or pooled analysis or (systematic$ adj2 review$) or published studies or published literature or medline or embase or data synthesis or data extraction or cochrane).ti,ab. (61896)
Double Blind Procedure/ (70620)
exp Triple Blind Procedure/ (12)
exp Control Group/ (2245)
exp PLACEBO/ or placebo$.mp. or sham$.mp. [mp=title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer name] (207387)
(random$ or RCT).mp. [mp=title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer name] (420855)
(control$ adj2 clinical trial$).mp. [mp=title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer name] (279987)
or/26-37 (778561)
38 and 25 (36604)
exp Non Insulin Dependent Diabetes Mellitus/ (54933)
((ketosis resistant or adult onset or slow onset or maturity onset or non?insulin dependent or stable or type 2 or type II) adj2 (diabet$ or DM)).ti,ab. (38625)
(t2dm or niddm).ti,ab. (7266)
42 or 40 or 41 (62672)
39 and 43 (841)
(diabet* adj2 (program* or clinic* or center* or centre*)).ti,ab. (4401)
limit 45 to (human and english language and yr=“2000 - 2009”) (2068)
38 and 46 (567)
44 or 47 (1307)

Database: CINAHL/Pre-CINAHL

# Query Limiters/Expanders Last Run Via Results

S45 (S44 or S39) Search modes - Boolean/Phrase Interface - EBSCOhost