educational group visits for the management of chronic health conditions: a systematic review

Patient Education and Counseling 95 (2014) 3–29

Review

Educational group visits for the management of chronic healthconditions: A systematic review

Ana R. Quinones a,b,*, Jeannette Richardson a, Michele Freeman a, Rochelle Fu a,b,Maya E. O’Neil a,c, Makalapua Motu’apuaka a, Devan Kansagara a,d

a Portland VA Medical Center, Health Services Research & Development, Portland, USAb Oregon Health & Science University, Public Health & Preventive Medicine, Portland, USAc Oregon Health & Science University, Psychiatry, Portland, USAd General Internal Medicine, Oregon Health & Science University, Portland, USA

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2. Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2.1. Data sources and search strategy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2.2. Eligibility criteria and abstraction methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2.3. Data extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2.4. Data synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

3.1. Arthritis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

3.2. History of falls. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

3.3. Asthma, COPD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

3.4. Hypertension, CHF/CAD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

3.5. Diabetes mellitus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

3.6. Multiple chronic conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

A R T I C L E I N F O

Article history:

Received 19 June 2013

Received in revised form 18 December 2013

Accepted 23 December 2013

Keywords:

Group visits

Chronic disease

Health education

Chronic disease self-management

A B S T R A C T

Objective: Review the effectiveness of group visits (appointments of multiple patients) on quality of life,

function, self-efficacy, utilization, and biophysical outcomes in randomized controlled trials of patients

with chronic conditions.

Methods: We searched MEDLINE1, Cochrane, CINAHL, and PsycINFO to January 2013 for English-

language trials of educational group visits led by non-prescribing facilitators (e.g., peer educators).

Results: We report on 80 arthritis/falls (n = 22), asthma/COPD (n = 10), CHF/hypertension (n = 12),

diabetes (n = 29), multiple conditions (n = 4), and pain (n = 4) studies. We found moderate evidence of

improved short-term self-efficacy in patients with arthritis (10 studies) and diabetes (10 studies). We

found no consistent evidence of improved quality of life; however a moderately strong body of evidence

suggests peer-led community-based programs might improve quality of life and utilization in patients

with multiple chronic conditions. Meta-analyses found short- (14 studies; mean change HbA1c = �0.27,

CI = �0.44, 0.11) and long-term (10 studies; mean change HbA1c = �0.23, CI = �0.44, �0.02) glycemic

improvement.

Conclusions: Group visits may improve self-efficacy and glycemic control. There was little consistent

evidence of improved quality of life, functional status, or utilization.

Practice implications: Group visits represent a reasonable alternative for educating patients with chronic

illness, though varied participation/retention suggests they should not be the sole alternative.

Published by Elsevier Ireland Ltd.

Contents lists available at ScienceDirect

Patient Education and Counseling

jo ur n al h o mep ag e: w ww .e lsev ier . co m / loc ate /p ated u co u

* Corresponding author at: Oregon Health & Science University, 3181 SW Sam Jackson Park Road, CB669, Portland, OR 97239-3098, USA. Tel.: +1 503 494 5889;

fax: +1 503 494 4981.

E-mail address: [email protected] (A.R. Quinones).

0738-3991/$ – see front matter . Published by Elsevier Ireland Ltd.

http://dx.doi.org/10.1016/j.pec.2013.12.021

http://crossmark.crossref.org/dialog/?doi=10.1016/j.pec.2013.12.021&domain=pdf

http://crossmark.crossref.org/dialog/?doi=10.1016/j.pec.2013.12.021&domain=pdf


mailto:[email protected]

http://www.sciencedirect.com/science/journal/07383991

www.elsevier.com/locate/pateducou


A.R. Quinones et al. / Patient Education and Counseling 95 (2014) 3–294

3.7. Chronic pain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

3.8. Effects of patient characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

3.9. Group visits compared to other active interventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

4. Discussion and conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

4.1. Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

4.2. Future research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

4.3. Practice implications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

1. Introduction

Health systems, payers, and providers are faced with thechallenges of preparing large numbers of patients to self-managethe day-to-day aspects of chronic illness care. As care deliveryrapidly evolves away from fee-for-service models and traditionalface-to-face physician encounters, group visits—appointmentswhere multiple patients are seen simultaneously—may offer analternative means to provide chronic disease education and self-management skills training to more patients, and with fewerresources. Education delivered to groups of patients by non-prescribing health professionals and facilitators has gainedtraction because of their potential to efficiently engage patientsin chronic illness self-management [1,2].

The Patient Protection and Affordable Care Act of 2010incentivizes the formation of health care innovations—such asthe Patient-Centered Medical Home—that build on a strongprimary care foundation to provide timely access, and acoordinated, regular source of care for patients [3]. Incorporatedinto these new care models, group visits may be a way to deliverhigh quality health education while prioritizing improved healthoutcomes and better patient care experiences. Although group visitinterventions have been used with greater frequency over the pastfew decades, there is little consensus as to whether, and for whom,group visits are an effective tool. Given differences in programstructure and content, the broad array of chronic conditions inwhich group visit interventions have been studied, and the lackclarity on effectiveness, we conducted a systematic review toexamine the effects of group visits on patient self-efficacy, qualityof life, health outcomes, and health care utilization in patients withchronic illness (Appendix 1). We also sought to examine whethereffects of group visit interventions depended on patient or programcharacteristics.

2. Methods

2.1. Data sources and search strategy

We conducted the original search February, 2012 and theupdated search January, 2013 of multiple databases (MEDLINE1

via PubMed1, Cochrane Register of Controlled trials, CINAHL(EBSCO), PsycINFO) using terms for non-prescribing practitionersand group visits, including but not limited to terms for groupeducation, group program(me), group session(s) (Appendix 2). Weobtained additional articles from reference lists of pertinentstudies and by consulting experts.

2.2. Eligibility criteria and abstraction methods

We included English-language randomized controlled trials ofeducational group visits led by non-prescribing facilitators forpatients with common chronic conditions: arthritis, history offalls, asthma, chronic obstructive pulmonary disease (COPD),hypertension, congestive heart failure (CHF)/coronary arterial

disease (CAD), diabetes mellitus, multiple co-occurring chronicconditions and chronic pain. We excluded group visit studies ifany portion of the intervention focused on individual-leveltreatment plans or prescription changes (i.e., shared medicalappointments), or focused exclusively on support groups or groupexercise classes. Existing Cochrane reviews of group exercisesummarize effectiveness of these interventions and represent asystematic evaluation of that literature [4,5]. In addition, a recentCochrane publication reviews exercise and medication manage-ment interventions designed to reduce fall incidence amongcommunity-living older adults [6]. We excluded diabetes studiespublished before 1998 because the overall approach to adultdiabetes care was qualitatively different after publication of theUK Prospective Diabetes Study, thereby rendering older studiesless applicable [7]. Reviewers trained in critical analysis ofliterature assessed whether abstracts met inclusion criteria anddual-reviewed full-text articles for inclusion and quality (Appen-dices 3 and 4).

2.3. Data extraction

One research team member abstracted each included study fordata on design, setting, population, findings, intervention struc-ture, comparator(s), participation, and attrition (Table 2). Wedefined content delivered as follows: (1) self-managementeducation: in addition to providing disease-specific informationto patients, these programs teach patients self-management skills,such as goal-setting and contracting, and build skills to reinterpretsymptoms. These interventions are often characterized bymotivational interviewing and cognitive behavioral therapytechniques; (2) didactic education: content is informational andformat is usually lecture-based (e.g., pathophysiology of disease);(3) experiential education: instruction based on demonstrations(e.g., cooking).

We found a breadth of measured and reported outcomecategories and variation in outcome metric validity, rendering afull synthesis of all outcomes infeasible and uninformative(Appendices 5 and 6). We chose, instead, to focus on distal healthoutcomes measuring quality of life and functional status becausethese are likely to be important to patients and could conceivablybe affected by group visit interventions. We included utilizationand medication adherence outcomes when reported, though weanticipated that fewer studies would be powered to examine theseoutcomes. We also examined intermediate outcome metrics,focusing specifically on biophysical markers such as HbA1c, and onself-efficacy/patient activation measures. Self-efficacy refers topersonal beliefs in one’s ability to self-manage illness. In thisreview, we used the term broadly to include patient activation,coping skills, or illness beliefs. We chose to examine this set ofoutcomes because there are validated tools [8,9], and these metricswere commonly reported. Furthermore, there is a link, bothconceptually and empirically, between the knowledge, skills, andattitude changes acquired during an educational intervention andintermediate health outcomes [10].

A.R. Quinones et al. / Patient Education and Counseling 95 (2014) 3–29 5

Two reviewers independently assessed the quality of each trialaccording to standard criteria [11]. Disagreements were recon-ciled by discussion, or if necessary, by a third team member.Individual studies were rated as ‘‘good,’’ ‘‘fair’’ or ‘‘poor’’(Appendix 4). We assessed overall quality of evidence using amethod developed by the Grading of Recommendations, Assess-ment, Development, and Evaluation (GRADE) Working Group[12], which considers the consistency, coherence, and applicabil-ity of a body of evidence, as well as the internal validity ofindividual studies.

2.4. Data synthesis

In compiling data tables (Tables 2–4), we prioritized well-validated scales. If studies report full scales as well as subscales, wereport full scales only. If studies did not report any outcome inthese categories, or report ad-hoc/non-validated measures, theirfindings were summarized narratively. We also present results onpatient characteristics that modify group visit effectiveness forstudies that report these analyses. Finally, we consider usual careas well as head-to-head comparators in trials with more than oneactive arm (Tables 2–4).

We conducted meta-analyses of the mean difference in HbA1cchange because we identified HbA1c as a clinically importantmarker for diabetes patients that is plausibly amenable to changein the short- (0–3 months) and medium-term (4–6 months). Weobtained a pooled estimate of relative risk using a random effectsmodel [13]. To determine whether the effects of group visits weremodified by intervention characteristics, we conducted sub-analyses according to study quality and intervention duration.Statistical heterogeneity was assessed by Cochran’s Q test and I2

statistic [14]. To examine publication bias, we used funnel plotsand Egger’s test to assess small study effects [15]. We alsoconducted multivariate meta-regression analyses to determinewhether duration, study quality, or publication year had anybearing on meta-analytic results. We followed PRISMA reportingguidelines for the meta-analyses of HbA1c. All analyses wereperformed using Stata 10.0.

3. Results

We included 84 publications of 80 studies focused on educationfor the management of prioritized chronic conditions (Fig. 1). Ingeneral, group visits were associated with short- and medium-term improvements in self-efficacy; however, fewer studiesexamined longer-term outcomes. There was little evidence thatinterventions improved quality of life, functional status, orutilization outcomes (Table 1).

3.1. Arthritis

Seventeen publications from the US, Europe, and Australiaevaluated the effectiveness of group visits that included self-management education (eleven studies), didactic education (eightstudies), and experiential education approaches (six studies) [16–32]. Studies varied widely in intervention structure, content,duration, and comparison group.

Seven of ten studies found group visit interventions improvedshort- and medium-term self-efficacy. Six of these studies foundbenefit for group visits delivering self-management education.Only one poor-quality study assessed outcomes beyond 12 months[32]. Despite the improvements seen in self-efficacy, only two ofeleven studies found improvements in quality of life relatedmeasures such as disability [28] and depression [18]. One US studyfound a self-management education intervention was associatedwith reduced physician visits [28], but this finding was not

confirmed in five other studies in Europe and Australia[17,18,20,21,27].

Eight studies compared two active interventions. Many of thesestudies compared interventions with more than one characteristicthat differed (i.e., different educational content and differentnumber of sessions), making it difficult to assess which interven-tion components are associated with observed effects. One studycompared self-management to didactic education in interventionswith the same number of sessions and found no difference inoutcomes between them [23]. Another study found that theinclusion of significant others along with patients in a self-management education intervention was associated with lowerself-efficacy than the intervention delivered to patients alone [30].Finally, one study found similar effects from a mail-deliveredindividualized self-management education program and an in-person group self-management education intervention [29].

Overall, there is a moderately strong body of evidence thatgroup self-management education interventions can improveshort- and medium-term self-efficacy in patients with arthritis,but they have little effect on quality of life or utilizationoutcomes.

3.2. History of falls

Four studies from the US, Canada, and Australia examineeffectiveness of educational group visits in patients with a historyof falls or at-risk for falling [33–36]. Two studies found groupdidactic education improved self-efficacy over the short-term [33],while another study which included a ‘‘booster’’ session at threemonths found improved long-term self-efficacy [34]. One studyfound improved timed-up-and-go physical performance [34],while another study found the intervention did not improvetimed-up-and-go when patients were simultaneously tasked withcognitive activities [36]. Only one of three studies found areduction in fall events [34], and no studies found improvedquality of life. Overall, didactic and experiential falls preventiontraining may improve patient self-efficacy and reduce the risk offalls, though the strength of this evidence is low because of thesmall number and inconsistencies among studies.

3.3. Asthma, COPD

Five studies conducted in the US or Australia examined theeffects of group visits compared with usual care in patients withasthma [37–41]. The group interventions involved didacticeducation in four studies [38–41] and self-management educationin one study [37]. Decreased utilization was observed in twostudies [37,40], and improvements in quality of life were noted intwo studies [37,38]. However, the studies were limited by selectionbias and other methodological issues.

Five studies of group visits in COPD patients were conducted ina variety of settings: Northern Ireland [42], the UK [43], theNetherlands [44], France [45], and a VA Medical Center in the US[46]. Three studies compared didactic education combined withexercise training to didactic education alone [43,44] or to usualcare [45]. Two other studies examined the effects of self-management education compared with didactic education [46],usual care [42], or individual support [42]. Better exercise capacitywas observed in the studies that combined exercise training withdidactic education, as compared with usual care [45] or withdidactic education alone [43,44].

Overall, a small body of fair-to-good quality evidence suggeststhat didactic and experiential group visits may be associated withsmall improvements. There is little methodologically soundevidence examining the impact of group visits in patients withasthma.

62 Citations i denti fie d fro m reference li sts of review articles, and manual search es for recent, unpublished or ongo ing stud ies

599 Pote ntial ly rele vant a rticle s i denti fie d for fur ther review

Total excluded articles = 51 5Non-Engli sh language = 17 No primar y data and not a n SR of primary stud ies = 56 Study po pulation not in s cope = 5 4 Does not meet interventio n criteria = 158 Not an RCT = 68 Outcome measures not in scope = 8 Full text not accessible = 29 Sha red medical a ppoint ment le d by a prescribing provider = 64 Use for cont extual purposes or referenc e mining = 39 Duplicate publication of same stu dy = 10 Pre-1998 DM study = 12

84 articles of 80 pri mar y studies of grou p visit s focu sing on e ducation for t he management of chronic conditions in adults

3,521 Citations i denti fie d for review of title and abstrac t

3,583 Potential ly relevant citations identified for further review

2,984 Citation s excluded due to lack of relevance in title or abst rac t

2,406 Citation s identified fro m electronic database sea rches after de duplication: 219 7 fro m M EDLI NE® via PubMed 78 fro m Coch rane library 44 fro m Ps ycINFO 87 fro m CINAH L

1,096 Citati ons derived fr om “shared medical appointment ” (SM A) lite rature search

Asthma/CO PD = 10

Chronic pain = 4

Hypertension/congestive heart failure = 12

Arthritis or history of falls = 21

Diabetes mellitus = 29

Multiple disease groups = 4

19 Studies used for SMA review, excluded

Fig. 1. Literature flow.


3.4. Hypertension, CHF/CAD

Our search identified three fair-quality studies of group visitsconducted in patients with heart failure in the Netherlands and theUS, published in four reports [47–50]. Seven studies examined theeffects of group visits on blood pressure in patients withhypertension [51–57]. The studies were conducted in a range ofinternational settings—Iran, the US, Thailand, Argentina, andItaly—and study quality varied widely. Three studies used self-management education techniques [52,55,56] and three studiesused didactic education [51–53,57] in comparison with usual care.One trial compared self-management education directly withdidactic education [54] Reductions in BP were noted in all threeself-management education studies [52,55,56] and in one didacticeducation study [51]. In the trial comparing self-managementeducation directly with didactic education, there were nosignificant reductions in BP found in either group at three months.However, significantly more self-management education patientshad controlled BP, defined as the proportion of patients with mean24-hour BP <140/90 mm Hg, compared with didactic education(70% vs. 44%, p = 0.04).

Overall, there were few studies of group visits in heartfailure patients, and their findings on self-efficacy, quality of

life, and biophysical measures were largely neutral. Groupself-management education interventions in patients withhypertension have reported improvements in BP control inshort-term and long-term studies, but the overall strength ofevidence is low.

3.5. Diabetes mellitus

We included 31 publications (29 studies) of group visits forpatients with diabetes [58–88] conduced in the US, Europe, Brazil,and Iran. We conducted meta-analyses of 17 studies comparing theeffects of group visits to usual care on HbA1c. Most of these trialswere conducted on large numbers of patients (Figs. 2 and 3). Loss tofollow-up was a concern when interpreting meta-analytic results.However, the majority of the 17 meta-analyzed trials (11 of 17)reported appropriate techniques to handle study attrition tominimize bias due to loss to follow-up (Appendix 4).

Overall, in 14 studies, group visits reduced HbA1c slightly moreover six months of follow-up than usual care, though there wassignificant heterogeneity which should temper confidence inthese results (mean change HbA1c = �0.27%; CI = �0.44, �0.11;I2 = 67.1%). At least part of the heterogeneity seemed to beassociated with study quality: the two good quality studies found

Overall (I-squared = 67 .1%, p = 0.00 0)

Stee d, 2005

Sharifira d, 201 2

Scain, 2009Lorig, 200 9

Khunti, 2012

Rygg, 2012

Group sessions <= 3 months

Lujan, 2007Miller, 20 02

Rosal, 2011

De Greef, 2011

Ande rson, 2005

Group sessions >3 monthsToobert, 20 11a

Brown, 20 02

Study

Sub total (I-squa red = 38.4%, p = 0.165 )

Sub total (I-squa red = 70.0%, p = 0.001 )

Phili s-Tsimika s, 20 11Poor

Poor

FairFair

Good

Fair

FairFair

Fair

Good

Poor

Fair

Poor

Quality

Poor

-0.27 (-0.44 , -0.11 )

-0.37 (-0.98, 0.24)

-0.32 (-0.51 , -0.13 )

-0.40 (-0.87, 0.07)0.06 (-0.18, 0.31 )

0.02 (-0.14, 0.18 )

-0.10 (-0.40, 0.20)

-0.75 (-1.20 , -0.30 )-0.50 (-0.78, -0.22 )

-0.53 (-0.92, -0.14 )

0.01 (-0.60, 0.62 )

0.21 (-0.31, 0.73 )

-0.40 (-0.83, 0.03)

-1.42 (-2.27, -0.57 )

Meandifference (95% CI)

-0.49 (-0.80 , -0.18 )

-0.20 (-0.38, -0.01 )

-0.10 (-0.72, 0.52)

-0.27 (-0.44 , -0.11 )

-0.37 (-0.98, 0.24)

-0.32 (-0.51 , -0.13 )

-0.40 (-0.87, 0.07)0.06 (-0.18, 0.31 )

0.02 (-0.14, 0.18 )

-0.10 (-0.40, 0.20)

-0.75 (-1.20 , -0.30 )-0.50 (-0.78, -0.22 )

-0.53 (-0.92, -0.14 )

0.01 (-0.60, 0.62 )

0.21 (-0.31, 0.73 )

-0.40 (-0.83, 0.03)

-1.42 (-2.27, -0.57 )

difference (95% CI)

-0.49 (-0.80 , -0.18 )

-0.20 (-0.38, -0.01 )

-0.10 (-0.72, 0.52)

0-2 -1.5 -1 -.5 0 .5 1

Favors treatmen t Favors control

(N=67)(N=824)(N=98)(N=150)(N=345 )(N=104)(N=146)(N=239)(N=97)

(N=280 )(N=252 )(N=256 )

(N=127 )(N=207 )

Fig. 2. Effect of group visits on HbA1c compared to usual care at �6 month follow-up, by duration of intervention.


no short-term improvements in HbA1c (mean changeHbA1c = 0.02; CI = �0.14, 0.17; I2 = 0.0%). Group visits lastingmore than three months had a greater effect on HbA1c improve-ment than those of shorter duration, but the quality of these longerduration intervention studies was also lower. We found

Overall ( I-squared = 56.9%, p = 0.013 )

Schilling er, 2009

Hornsten, 2008

Subt otal (I-squared = 53.8%, p = 0.043 )

Khunti, 201 2

Phili s-Tsim ika s, 201 1

Brown, 200 2

Subt otal (I-squared = 33.8%, p = 0.221 )

Rosal, 2011

Study

Group session s > 3 mon ths

Scain, 2009

Rygg, 2012

Group sessions <=3 month s

Toobe rt, 2011

Adolfsson , 200 7

Qualit y

Fair

Fair

Goo d

Poor

Poor

Fair

Fair

Fair

Fair

Fair

-2 -1.5 -1 -.5 Favors tr eatme nt

(N=82 4)

(N=10 4)(N=14 6)

(N= 101)

(N= 104)

(N= 339)(N= 252)

(N=280)(N= 256)

(N= 207)

Fig. 3. Effect of group visits compared to usual care on HbA1

similar effects on HbA1c at 7–12 months in the 10 studieswith longer-term follow-up. Funnel plot analyses showed noevidence of publication bias for 6 month outcomes (Egger biascoefficient = �1.62, CI = �3.73, 0.48), but some evidence ofpublication bias for 12 month outcomes (Egger bias

-0.23 (-0.44, -0.02 )

differen ce (95% CI )

0.20 (-0.25, 0.65)

-0.94 (-1.60 , -0.28 )

-0.33 (-0.63, -0.03 )

0.05 (-0.10, 0.20)

-0.60 (-1.34, 0. 14)

-0.79 (-1.40 , -0.18 )

-0.05 (-0.26, 0. 16)

-0.25 (-0.72, 0. 22)

Mean

-0.40 (-0.92, 0. 12)

-0.10 (-0.45, 0. 25)

0.00 (-0.48, 0.48)

-0.30 (-0.80, 0. 20)

-0.23 (-0.44, -0.02 )

differen ce (95% CI )

0.20 (-0.25, 0.65)

-0.94 (-1.60 , -0.28 )

-0.33 (-0.63, -0.03 )

0.05 (-0.10, 0.20)

-0.60 (-1.34, 0. 14)

-0.79 (-1.40 , -0.18 )

-0.05 (-0.26, 0. 16)

-0.2 5 (-0.72, 0. 22)

-0.40 (-0.92, 0. 12)

-0.10 (-0.45, 0. 25)

0.00 (-0.48, 0.48)

-0.30 (-0.80, 0. 20)

00 .5 1

Favors co ntrol

c at 7–12 month follow-up, by duration of intervention.


coefficient = �2.14, CI = �3.62, �0.66). Multivariate meta-regres-sion models showed that none of the covariates examined—duration, study quality, or publication year—were independentlyassociated with changes in HbA1c.

Five studies found improvements in self-efficacy, four of thesefinding positive effects beyond six months of follow-up. Perhapsnot surprisingly, the four studies focused on broader self-management rather than didactic education [66,69,76,80]. Despitefinding that some interventions may improve self-efficacy, therewas little evidence that group visits improved quality of life overthe short- or long-term. Few studies reported or were powered toevaluate utilization.

Eleven studies compared a group visits to one or more activeinterventions. Three of these studies found that interventionsfocused on self-management were associated with greaterimprovements in glycemic control than didactic approaches,though there were multiple other differences in the comparedinterventions, making it difficult to draw firm conclusions[63,68,87]. Two studies compared group to individual education.One fair-quality study found that an automated, telephone-based,self-management intervention performed similarly to an in-persongroup self-management intervention [80].

Overall, we found group visits in patients with diabetes mayhave modest effects on glycemic control over the short- and long-term, but the strength of evidence supporting this conclusion islow mostly because of inconsistencies across studies andmethodological weaknesses of the studies finding the mostpositive effects. Interventions focused on self-managementeducation were associated with improved self-efficacy and illnessbelief scores over the short- and long-term. However, there was noconsistent evidence that group visits improved quality of life.

3.6. Multiple chronic conditions

Four studies evaluated the Chronic Disease Self-ManagementProgram [89] in populations with multiple chronic conditions [90–93] in the US, China, and the Netherlands. Overall, the peer-led,community-based program was associated with medium-termimprovements in self-efficacy, health status, and utilization; andthese effects may persist long-term. These findings were based onmoderately strong evidence from two large US trials, but were notreplicated in other countries, and likely apply most to patientsengaged enough in care to agree to attend a multi-week course.

3.7. Chronic pain

Four studies evaluated group visits compared to usual care[94,95], educational reading materials [96], or individual treat-ment [97] in patients with chronic pain in the US and Europe.Though many findings were not statistically significant from thecomparison, some results favored group interventions. Overall, asmall body of literature suggests group-based self-managementeducation interventions may improve pain coping skills at leastover the short-term, though the strength of this evidence is lowbecause there were few studies and the methodological quality ofone of the studies finding benefit was poor.

3.8. Effects of patient characteristics

Relatively few studies examined how patient characteristicsmodified intervention effects. Overall, studies found little differencein group visit effectiveness according to patient demographic andsocioeconomic characteristics. However, among studies of arthritisand history of falls, two studies found that obese patients tended torespond to experiential (aerobic exercise) group visits more thanparticipants with lower BMI on self-reported disability [22] and falls

[34]. Among hypertension and heart failure studies, one study foundpatients with higher education and better cognitive status showedgreater short-term improvements in cardiac-specific quality of life[50]. One chronic pain study noted that group visit effectiveness wasmodified by agency-orientation, with high agency-oriented parti-cipants experiencing improvements in pain [98].

3.9. Group visits compared to other active interventions

Overall, in five studies, group visits focused on self-manage-ment were more effective than sessions limited to didacticeducation [25,54,68,71,99], however, in four of these five studies,the intervention arms differed considerably (e.g., having non-equivalent number of sessions), limiting the strength of thisconclusion. Studies that compared group to individual educationvisits found mixed results on a variety of outcomes, with noappreciable differences found in three studies [37,62,75], positiveeffects in four studies [35,63,87,97], and improvements forindividual relative to didactic education in one study [82]. Twostudies compared the effects of group self-management educationand mailed self-management programs, and found no differencesin self-efficacy, pain, and functional status outcomes [29,80].

4. Discussion and conclusion

We found 80 trials examining the effects of group visitinterventions across a variety of chronic illnesses. In general,many group visit interventions appear to be able to improve short-and medium-term patient self-efficacy (Table 1). There was littleconsistent, fair-to-good quality evidence that they improvedquality of life or utilization, though there was good evidence thatpeer-led, community-based self-management education programscould improve health status and utilization for patients withmultiple chronic conditions. Diabetes group visits were likelyassociated with small short-term improvements in glycemiccontrol—the longer-term effects are largely unknown since thevast majority of studies focused on short-term effects. Participa-tion rates varied widely across studies, but were often low (Tables3 and 4, Appendix 4). Findings from included studies are likelymost applicable to patients who are easy to contact, have time toparticipate, and have enough motivation to enroll in the program.

Chronic illness education is a highly complex endeavor withinterventions varying in their intended purpose, content, leader-ship, intensity, and format. Studies comparing two or more activeinterventions could help elucidate whether or not there are certainintervention factors associated with better outcomes. There werefew studies directly comparing didactic education approaches toself-management approaches, though, not surprisingly, moststudies finding improvements in self-efficacy focused on thelatter. Studies that compared group to individual approachesappear to have similar effects.

Learning and mastering chronic illness self-management is atime-consuming process. Theoretically, we might expect theduration of an intervention to be associated with its effectiveness,but we found it difficult to confirm this hypothesis. For example,we found greater improvement in glycemic control for interven-tions lasting longer than 3 months. However, the interventions oflonger duration were also of lower methodological quality.Unfortunately, we found few studies examining the effects of a‘‘booster’’ session.

It is unclear why the group visit literature has not found aconsistent effect on health, utilization, or quality of life outcomesdespite the logical inference that improved self-efficacy shouldlead to better disease control and outcomes. It is possible thatintervention or follow-up duration was inadequate. It is alsopossible that—in an era promoting clinical guidelines and quality


improvement—it is increasingly difficult to demonstrate incre-mental benefit because usual care has also improved. A recent trialof intensive diabetes treatment found few effects on healthoutcomes in part—as the authors speculate—because treatment inthe usual care group was quite good [100].

We found little cost-effectiveness data to guide decision-making of widespread investment in educational group visitmodalities. However, there is likely to be great variation in costsdepending on personnel, duration, and frequency considerations.Whether group visit expenditures are warranted may also dependon how highly proximate outcome measures—such as self-efficacy—are valued by patients and health systems.

On the other hand, peer-led, community-based self-manage-ment programs—such as the Chronic Disease Self-ManagementProgram—are low-cost and associated with improved self-efficacy, health status, and utilization. However, such programsdo not provide core skills and information patients might need tohelp self-manage illness (e.g., glucose self-monitoring). It is alsonot clear how much the community-based nature of theintervention matters. Offering the programs in communitycenters may make it easier for patients to participate on anongoing basis. It would be useful to use qualitative methods toshed more light on such issues.

Although we did not find evidence of direct harms associatedwith group visits, there are theoretic concerns, especially given thelack of robust evidence for long-term health outcomes. Forinstance, travel and participation time associated with groupvisits may preclude inclusion of patients with inflexible workschedules, and may be prohibitive for frail, older participants.

4.1. Limitations

In setting out to conduct this systematic review of group visitinterventions led by non-prescribing facilitators, a chief limitationis comparability of studies given vast heterogeneity and complex-ity of intervention content and outcomes examined. Althoughthere have been many published studies testing group visiteffectiveness, we found few with similar characteristics to beexplicitly compared. In addition, the sheer number of outcomesreported precluded reporting all outcomes. We pre-specifiedoutcomes that were commonly reported, clinically important, ormeasured self-efficacy since this was the broad intended effect ofthese interventions.

Table 1Summary of the evidence, by clinical area, outcome, and length of follow-up.

Condition Outcome Effecta GRADE

classificationb

Arthritis Self-efficacy

�6 months (+) Moderate

>6 months (+/�) Moderate

QOL/functional status

�6 months (�) Moderate

>6 months (�) Moderate

Physical performance

�6 months (0) –

>6 months (0) –

Utilization

�6 months (�) High

>6 months (�) High

Still, there may be important outcomes not captured in thisreview. Most notably, we did not consider knowledge improve-ment outcomes. Many studies reported various knowledge out-comes, but few were standardized, making any comparison acrossstudies difficult. Moreover, the clinical importance of short-termknowledge gains is debatable if they do not translate into gains inself-efficacy, health outcomes, or quality of life.

Additionally, we found good quality trials testing effectivenessof multicomponent interventions that included both, group andindividual elements. Unfortunately, these trials were not includedin our review because independent effects of group visits could notbe evaluated. Finally, we exclude non-English publications, thoughthe bias this exclusion may have introduced is likely to beminimized by searching multiple databases, bibliographies of keystudies, and communicating with experts.

4.2. Future research

Additional comparative studies evaluating alternatives foreducating patients are needed. Studies showing that mailed andphone-based self-management education programs were aseffective as in-person group visits are interesting and point toalternative educational modalities that may appeal to patientswith time or geographic constraints. Newer technologies such asmobile platforms and video-based interventions should also beevaluated as a means for delivering group education. The evidencebase would also benefit from a more standardized approach tooutcome reporting.

4.3. Practice implications

Group visits can improve short-term self-efficacy, but appear tohave little effect on quality of life and other health outcomes.Group visits may modestly reduce HbA1c in the short-term forpatients with diabetes. A peer-led, community-based self-man-agement program appears to improve health and utilizationoutcomes for patients with various chronic illnesses. Group visitsmay be as effective as individual visits and may represent areasonable alternative for educating patients with chronic illness,though varied and sometimes low participation and retentionsuggest they should not be the sole alternative.

Comment

Based on 10 studies, 7 of which were fair- to good-quality. Inconsistent

results among some studies limit the evidence base. Two large fair-

quality studies (n = 812; n = 544) have positive results; one good-quality

study (n = 34) and one fair-quality study (n = 218) have null findings.

Inconsistent results and infrequent reporting of long-term findings

among 5 studies.

Based on mainly fair-quality studies (7 of 10 studies). Variety of

instruments and subscales limit evidence base.

No significant findings in the 6 studies that examined long-term effects.

No evidence. Trials for group visits for patients with arthritis did not

report performance measures in the near- or long-term.

No significant effects on physician visits found in the near- or longer-

term in 6 studies.

Table 1 (Continued )


classificationb

Comment

History of falls Self-efficacy

�6 months (+) Low Based on only 1 small fair-quality trial (n = 83) with low participation.

>6 months (+) Moderate Based on 1 large good-quality trial (n = 310). Of note, this trial included a

booster session 3 months after intervention completion.


�6 months (�) Moderate Based on 1 small fair-quality trial (n = 83) with low participation.

>6 months (�) Moderate Based on 1 large good-quality trial (n = 310) that included a booster

session 3 months post-intervention.


�6 months (�) Moderate Based on 1 small fair-quality trial (n = 83) with low participation.

>6 months (+) Moderate Based on 1 large fair-quality trial (n = 454) with high participation and

low loss to follow-up.

Fall events

�6 months (�) Low Based on only 1 small poor-quality trial (n = 45) with unclear reporting

of findings.

>6 months (+/�) Moderate Based on 2 trials, one good-quality trial (n = 310) detailing positive

results and one fair-quality trial (n = 454) with null findings.

Asthma Self-efficacy

�6 months (�) Low Based on only 1 poor-quality trial (n = 125).

>6 months (0) – No evidence. Findings for self-efficacy not reported in the long-term.


�6 months (�) Low Based on only 1 poor-quality trial (n = 125).

>6 months (+) Low Based on only 1 large fair-quality trial (n = 323) with poor participation.


�6 months (0) – No evidence. Trials for patients with asthma did not report performance

outcomes.>6 months (0) –

Utilization

�6 months (+) Moderate Decreased ED visit findings based on 1 large fair-quality trial (n = 241)

with low participation.

>6 months (+/�) Moderate Limited by inconsistent results and poor participation. Based on 2 large

fair-quality studies (n = 323; n = 241) with low participation.

COPD Self-efficacy

�6 months (0) – No evidence. Trials on patients with COPD did not examine self-efficacy

scales.>6 months (0) –


�6 months (0) – No evidence.

>6 months (�) Moderate Based on 1 small good-quality trial (n = 45).


�6 months (�) Moderate Based on only 1 good-quality trial (n = 238) limited by poor

participation.

>6 months (+/�) Moderate Inconsistent findings from 2 good-quality trials of varying size (n = 238;

n = 45). The small trial reported improvements in walking distance

while the findings from the large trial were null.

Utilization


>6 months (�) Moderate Based on 1 small good-quality trial (n = 45).

Hypertension Self-efficacy


>6 months (0) –



>6 months (0) –

Biophysical

�6 months (+/�) Moderate Based on 4 trials of varying quality and size (n = 52; n = 96; n = 98;

n = 574). Inconsistent results, however improvements in blood pressure

found in the large good-quality trial (n = 574).

>6 months (+/�) Moderate Infrequent reporting of long-term effects. Based on 2 trials of varying

quality (poor, n = 292; good, n = 574). Good-quality trial found no long

term significant blood pressure changes.

Utilization


>6 months (0) –

CHF/CAD Self-efficacy

�6 months (�) Moderate Based on 2 large good- (n = 317) and fair-quality trials (n = 250). Null

findings in the near- and long-term, however both studies are limited

by poor participation and high loss to follow-up.

>6 months (�) Moderate


�6 months (�) Moderate Based on 1 large good-quality trial (n = 317) and one small fair quality

trial (n = 62) of Veterans. Limited by low study participation.

>6 months (�) Moderate Only 1 large good-quality trial (n = 317) examined quality of life

outcomes in the longer term.

Biophysical

�6 months (�) Moderate Only 1 large good-quality trial (n = 317) examined BMI in the near- and

longer-term, with no significant changes detected post intervention.>6 months (�) Moderate




classificationb

Comment

Diabetes mellitus Self-efficacy

�6 months (+) Moderate Based on 10 studies of varying quality reporting on self-efficacy. All

except one (fair-quality, n = 146) reported improvements in self-

efficacy, including a large good-quality trial (n = 824). Heterogeneity in

instruments reported limits evidence.

>6 months (+/�) Moderate Inconsistent findings reported in the long-term. Large good-quality trial

(n = 824) reports improvements continue into the longer-term.


�6 months (�) Moderate Based on 7 studies of varying quality reporting on a variety of

instruments.

>6 months (�) Moderate Based on 5 studies of varying quality reporting on a variety of

instruments.

Biophysical (HbA1c)

�6 months (+/�) Moderate 14 studies of varying quality reported small improvements in HbA1c in

the near-term. However, two good-quality studies (n = 67; n = 824)

found no near-term improvements in HbA1c.

>6 months (+/�) Moderate 10 studies of varying quality reported small improvements in HbA1c in

the longer-term. Again, one good-quality study (n = 824) reported no

benefit in glycemic control for the intervention group.

Utilization

�6 months (�) Moderate 3 trials of varying quality (poor-quality, n = 54; fair-quality, n = 345; fair

quality n = 146) found no significant change in clinician visits, ED visits,

or days hospitalized.

>6 months (�) Low Only 1 trial of fair-quality (n = 146) reported null findings on change in

clinician visits.

Multiple chronic

conditions

Self-efficacy

�6 months (+/�) Moderate Based on 4 large trials of mostly fair quality and one poor-quality trial

(n = 952; n = 551; n = 954; poor-quality n = 136). Findings limited by

inconsistent outcomes and variety of instruments reported.

>6 months (+) Moderate Based on only 1 large fair-quality trial (n = 551) reporting improvement

in self-efficacy sustaining into the longer-term.


�6 months (+) Moderate Based on 1 large fair-quality (n = 952) and 1 poor-quality trial (n = 136)

with inconsistent findings. The large fair-quality study reported

improvements.

>6 months (0) – No evidence. Longer-term outcomes not reported.

Utilization

�6 months (+/�) Moderate Based on 3 large fair-quality trials (n = 952; n = 551; n = 954). Null

findings for change in physician visits, and inconsistent findings for

hospital stays/days and ED visits.

>6 months (+/�) Moderate Based on 1 large fair-quality trial (n = 551) that found reduction in ED

visits, but no change for physician visits and hospital days.

Chronic pain Self-efficacy

�6 months (+/�) Low Based on 3 studies of fair- and good-quality (n = 45; n = 131; n = 156).

Limited evidence due to number of instruments reported and

inconsistent findings. Good-quality trial (n = 156) reported null

findings.

>6 months (0) – No evidence. Longer-term outcomes not reported.


�6 months (+/�) Moderate Based on 3 studies of varying quality (n = 45; n = 174; n = 156). Large trial

(n = 156) reported positive findings. Limited evidence stemming from

number of instruments reported.

>6 months (+) Low Based on only 1 poor-quality trial (n = 174) that reported improvement

in pain but did not report appropriate statistical test results (p-values).

Utilization

�6 months (+) Low Based on only 1 poor-quality trial (n = 174) that reported a reduction in

physician visits. The study did not report appropriate statistical test

results (p-values) in the longer-term.

>6 months (+) Low

Notes: GRADE = grades of recommendation, assessment, development, and evaluation; QOL = quality of life; CHF = congestive heart failure; CAD = coronary artery disease;

BMI = body mass index; ED = emergency department.a Effect: (+) benefit; (�) harm; (�) mixed findings/no effect; (0) no evidence.b GRADE classification: high = further research is very unlikely to change our confidence on the estimate of effect; moderate = further research is likely to have an important

impact on our confidence in the estimate of effect and may change the estimate; low = further research is very likely to have an important impact on our confidence in the

estimate of effect and is likely to change the estimate; very low = any estimate of effect is very uncertain.


Table 2Characteristics of group visit interventions focusing on education for the management of chronic disease.

Study Sample size

Setting

Program name,

if applicable

Demographics:

Mean age

% male

% minority Mean

disease duration

GV structure:

# Visits, frequency

Duration

Group size

GV content:

SME (self-

management)

DE (didactic)

EE (experiential)

GV leaders:

Number of leaders

Profession type

Comparator

Arthritis

Ackerman,

2012 [17]

N = 120

Australia

ASMP

65.1 yrs

40%

Race NR

Duration NR

6 weekly (2 h) sessions

1.5 months

4–21 patients

SME 2 leaders

Peer leader, health

professional

Usual care

(information book)

Barlow,

2000 [18]

N = 544

UK

ASMP

58.1 yrs

16%

4% nonwhite

11 yrs with arthritis


1.5 months

�10 patients

SME 2 leaders

Peer leaders

Usual care

Breedland,

2011 [19]

N = 34

Netherlands

FIT

48 yrs

29%

Race NR

9.7 yrs with RA

8 weekly (1 h)

education

16 semi-weekly (1.5 h)

exercise

2 months

Group size NR

DE, EE 5 team members

Psychologist, PT, OT,

dietitian, social worker

Usual care

Buszewicz,

2006 [20]

N = 812

UK

ASMP

68.6 yrs

37%

0.5% Caribbean black

Duration NR

6 weekly (2.5 h)

sessions

1.5 months

12–18 patients

SME, EE NR Usual care

(information book)

Ettinger,

1997 [22]

N = 439

US

FAST

69 yrs

30%

26% black

Duration NR

3 monthly (1.5 h)

sessions

18 biweekly and

monthly calls

18 months

10–15 patients

DE 2 leaders

Exercise leader, nurse

Group exercise arms:

GV2: 36 (1 h) aerobic

GV3: 36 (1 h) resistance

Class sizes 10–15

Freeman,

2002 [23]

N = 54

UK

51.4 yrs

15%

Race NR

4.5 months with RA


1 month

Group size NR

GV1: SME

GV2: DE

3 team members

Physiotherapist,

rheumatologist,

psychologist

GV2

Giraudet-Le

Quintrec,

2007 [24]

N = 208

France

54.8 yrs

14.1%

Race NR

13.1 yrs with RA


1 (4 h) booster after 6

months

2 months

8–10 patients

DE, EE 10 team members

Rheumatologist, rehab.

specialist, dietitian,

social assist., nurses,

PTs, and OTs

Usual care+:

Two information

leaflets written by

research team

Hammond,

1999 [16]

N = 35

UK

55.2 yrs

17%

Race NR

9.8 yrs with RA


Optional home visit 2

wks post

1 month

4–8 patients + spouses

invited

SME 1 leader

Rheumatology OT

Usual care

Hammond,

2008 [99]

N = 167

UK

LMAP

55.4 yrs

35%

Race NR

7.4 yrs with RA

GV1:

9 sessions (2.5 h) over 9

mo

12 months

6–10 patients

GV2:

5 (2 h) sessions

1.25 months

8–12 patients

GV1: SME, EE

GV2: DE, EE

3 leaders

Rheumatology OT,

community OT,

rheumatology PT

GV2

Hewlett,

2011 [25]

N = 127

UK

59.2 yrs

27%

Race NR

14 yrs with RA

GV1:


1 booster session (wk

14)

1.5 months

4–9 patients

GV2:

1 (1 h) session

Delivered by RA nurse

GV1: SME

GV2: DE

2 leaders

Clinical psychologist,

specialist OT

GV2

Kaplan,

1981 [26]

N = 34

US

48.2 yrs

0%

9% nonwhite

Duration NR

GV1:

1 (2.5 h) education

session

12 weekly (1–2 h)

counseling

4 months

GV2:

1 (2.5 h) education

session

Group size NR

GV1: DE,

counseling

GV2: DE

2 leaders

Patient counselor,

psychiatrist

GV2

Lorig,

1985 [27]

N = 286

US

ASMP

67.4 yrs

17%

3% nonwhite

Duration NR

6 sessions (2 h) over 4

months

4 months

15–20 patients + family

SME 2 leaders

Trained peer leaders

Usual care



Study Sample size

Setting

Program name,

if applicable

Demographics:

Mean age

% male

% minority Mean

disease duration

GV structure:

# Visits, frequency

Duration

Group size

GV content:

SME (self-

management)

DE (didactic)

EE (experiential)

GV leaders:

Number of leaders

Profession type

Comparator

Lorig,

1999 [28]

N = 331

US

ASMP

62.5 yrs

16%

100% Latino

Duration NR

6 sessions (2 h) over 6

weeks

1.5 months

10–15 patients and

family

SME Lay leaders Usual care

Lorig,

2004 [29]

N = 341

US

ASMP

65.2 yrs

25%

10% nonwhite

Duration NR


1.5 months

Group size NR

SME 2 leaders

Trained peer leaders

SMART group: mailed

individual self-

management program

Riemsma,

2003 [30]

N = 218

Netherlands

56.4 yrs

38%

Race NR

11.7 yrs with RA


3 (2 h) booster sessions

1.25 months

8 patients � spouses

GV1: SME, EE

(patients only)

GV2: SME, EE

(spouses included)

2 leaders

RA nurse, nurse

GV2, and

Usual care+: self-help

guide

Sevick,

2009 [31]

N = 316

US

ADAPT

69 yrs

28%

24% nonwhite

Duration NR

GV1:

3� month, months 1–4

Biweekly, months 5–6

Monthly, months 7–18

18 months

GV2:

GV1 structure +

3�/week grp exercise,

months 1–4

Group sizes NR

GV1: DE

GV2: DE, exercise

NR GV2, and

Healthy lifestyle group:

Monthly (1 h) DE GV,

months 1–3; monthly

phone contact, months

4–5; bimonthly phone

contact months 6–18

Taal,

1993 [32]

N = 75

Netherlands

49.6 yrs

20%

Race NR

4.3 yrs with RA


1.25 months

6–8 patients

SME, EE 2 leaders

RA nurse,

physiotherapist, or

social worker

Usual care+:

individual referral to

physiotherapist

History of falls

Arnold,

2010 [33]

N = 83

Canada

74.5 yrs

29%

Race NR

7.6 yrs with hip pain

GV1:

22 semiweekly (1.5 h)

sessions

2.75 months

GV2:

22 semiweekly (0.75 h)

sessions

2.75 months

Group sizes NR

GV1: DE, EE,

aquatic exercise

GV2: EE, aquatic

exercise

2 leaders

Aquatic fitness

instructor, PT

Usual care, and GV2

Clemson,

2004 [34]

N = 310

Australia

Stepping On

78.4 yrs

26%

Race NR

Duration NR

7 (2 h) sessions over 7

weeks

1 (1.5 h) booster (after

3mo)

1.75 months

12 patients

DE, EE OT with geriatrics

experience, team of

content experts for

educational areas

Usual care+:

�2 home social visits

from OT student

instructed not to

discuss falls or falls

prevention

Ryan,

1996 [35]

N = 45

US

78 yrs

0%

66% black

Duration NR

1 (1 h) session

1 day

7–8 women

DE 1 leader

Nurse

Individual visit, and

Usual care+: Health

promotion session with

no falls prevention info

Shumway-Cook,

2007 [36]

N = 454

US

75.6 yrs

23%

4% nonwhite

Duration NR

6 monthly (1 h)

sessions

6 months

Group size NR

DE, exercise 1 leader

Nurse

Usual care

(two CDC informational

brochures)

Asthma

Wilson,

1993 [37]

N = 323

US

NR 4 weekly sessions

1 month

6–8 patients

SME 1 leader

Nurse educator

3 comparators:

1) individual education

2) usual care with

workbook

3) usual care with no

supplemental

education

Abdulwadud,

1999 [38]

N = 125

Australia

Australian Asthma

Management

Program

Mean age 45.6

40% male

Race NR

Duration NR

3 weekly sessions

3 weeks

Up to 13 patients

DE 1 leader

Nurse educator

Usual care

Allen,

1995 [39]

N = 116

Australia

Mean age 40

46% male

Race NR

Duration NR

4 weekly sessions

4 weeks

10–12 patients

DE 2 leaders

Asthma educators

Usual care

Bolton,

1991 [40]

N = 241

US

Mean age 38

34% male

67% non-white

Duration NR

3 sessions

Duration NR

6–10 patients

DE 1 leader

Nurse educator

Usual care



Study Sample size

Setting

Program name,

if applicable

Demographics:

Mean age

% male

% minority Mean

disease duration

GV structure:

# Visits, frequency

Duration

Group size

GV content:

SME (self-

management)

DE (didactic)

EE (experiential)

GV leaders:

Number of leaders

Profession type

Comparator

Snyder,

1987 [41]

N = 79

US

Wheezers

Anonymous

Mean age 28

45% male

Race NR

Duration NR

2 sessions, NOS

Duration NR

8–12 patients

DE 1 leader

Respiratory therapist

Usual care

COPD

Wilson,

2008 [42]

N = 91

Northern Ireland

Mean age 61

48% male

Race NR

Duration NR

Current smokers

5 weekly sessions

5 weeks total

N per session NR

SME 1 leader

Respiratory Nurse

Specialist

Usual care (n = 35),

Individual support

(n = 27)

Kunik,

2008 [46]

N = 238

US VAMC

Mean age 66

96% male

16% Black

3% Hispanic

8 weekly sessions

8 weeks

Up to 10 patients

SME: CBT 1 leader

Psychology intern or

post-doctoral fellow

with CBT experience

DE group education

Bestall,

2003 [43]

N = 66

UK

Mean age 69

51% male

Race NR

Duration NR

16 DE bi-weekly

sessions, 8 weeks total

(both groups), followed

by 10 EE monthly

sessions, 1 year total

(exercise group only)

N per session NR

DE + EE: exercise NR DE group education

Effing,

2011 [44]

N = 159

Netherlands

COPE-active

Mean age 63

58% male

Race NR

Duration NR

35% smokers

DE: 4 weekly sessions/1

month total; 5 patients

EE: 2–3 times/week, 11

months total; 2–3

patients

DE + EE: exercise 2 leaders

Respiratory nurse

Physiotherapist

DE group education

Ninot,

2011 [45]

N = 45

France

Mean age 63

84% male

Race NR

Duration NR

26% smokers

8 sessions, 2� week

4 weeks total

DE + EE: exercise 2 leaders

DE led by health

professional, EE led by

exercise trainer

Usual care

CHF/CAD

Smeulders,

2010 [49,50]

N = 317

Netherlands

CDSMP

Mean age 67 73% male

Race NR

Duration NR

6 weekly sessions

6 weeks total

6–12 patients

SME 2 leaders

Cardiac nurse specialist

CHF patient peer leader

Usual care

Chang,

2005 [48]

N = 62

US VAMC

Mean age 69

% male NR

17% non-white

Duration NR

15 weekly sessions

15 weeks total

Group size NR

DE Experts on medical,

pharmaceutical,

lifestyle, nutrition, and

psychosocial issues

Usual care

Moore,

2006 [47]

N = 250

US

CHANGE

Mean age 62

17% black

2% non-white, NOS

Duration NR

5 sessions: 3 weekly

followed by 2 monthly

3 months total

6–8 patients

SME 1 leader

Cardiac nurse

Usual care

Hypertension

Baghianimogha-

dam [57]

N = 150

Iran

Mean age 57.9

39% male

Race NR

Duration 6.77 yr

Frequency NR

2 months total

Group size NR

DE + EE 1 leader

Health education

researcher

Usual care

Nessman,

1980 [51]

N = 52

US VAMC

Mean age 55

10% black

16% Mexican-American

Duration NR

8 weekly sessions

8 weeks total

Group size NR

DE 2 leaders

Nurse, psychologist

Informational control

(audiotape)

Rujiwatthanakorn,

2011 [52]

N = 96

Thailand

Mean age 61

40% male

Race NR

Duration NR

3 sessions

8 weeks total

6–7 patients

Duration NR

SME 1 leader

Nurse

Usual care

Balcazar,

2009 [53]

N = 98

US

Mean age 53

21% male

100% Mexican-

American, 87% born in

Mexico

Duration NR

4 sessions at weeks 1, 2,

3, 8

8 weeks total

15–20 patients

DE 2 leaders

Promotoras (Mexican-

American community

health workers)


Figar,

2006 [54]

N = 60

Argentina

PEM

Mean age 69

57% male

Duration NR

4 weekly sessions

4 weeks

10 patients

SME Physicians with

experience in HTN

education/

management

DE

Scala,

2008 [55]

N = 292

Italy

Mean age 62

42% male

Race NR

Duration NR

3 sessions

4 months total

4–5 patients

SME 1 leader

Moderator, tutor

assistants




Study Sample size

Setting

Program name,

if applicable

Demographics:

Mean age

% male

% minority Mean

disease duration

GV structure:

# Visits, frequency

Duration

Group size

GV content:

SME (self-

management)

DE (didactic)

EE (experiential)

GV leaders:

Number of leaders

Profession type

Comparator

Svetkey,

2009 [56]

N = 574

US

Mean age 60.5

39% male

37% black

1% Hispanic

Duration NR

20 weekly sessions

6 months total

10–15 patients

SME 2 leaders

Behavioral

interventionist,

assistants (community

health advisors)

Usual care

Diabetes mellitus

Adolfsson,

2007 [58]

N = 101

Sweden

63.1 yrs

54%

Minority NR

6.6 yrs with DM

4 (2.5 h) sessions

1 booster (2.5 h) within

7 months

5–8 patients

DE 7 physicians and 12

diabetes specialist

nurses

Usual care

Anderson,

2005 [59]

N = 239

US

61 yrs

18%

96% minority

8.5 yrs with DM


1.5 months

Group size NR

SME Certified diabetes

educators

Usual care

Brown,

2002 [60]

N = 256

US

The Starr County

Border Health

Initiative

54 yrs

36%

Race NR

7.85 yrs with DM

12 weekly, 12

biweekly, 3 monthly

(2 h) sessions

12 months

Group size NR

DE, EE Bilingual Mexican

American nurses,

dietitians, local

community workers

Usual care

Brown,

2005 [61]

N = 216

US

The Starr County

Border Health

Initiative

49.6 yrs

40%

Race NR

5.1 yrs with DM

GV1:

3 weekly, 12 biweekly,

3 monthly (2 h)

sessions

12 months

Group size NR

GV2:


3 support @ 3, 6, and 12

months

8 patients

DE, EE Bilingual Mexican

American nurses,

dietitians, local

community workers

GV2

De Greef,

2011 [62]

N = 67

Belgium

67.4 yrs

70.1%

Minority NR

64.5% diagnosed <5 yr

3 (1.5 h) sessions every

3wks

3 months

Group size NR

SME Clinical psychologist Usual care;

individual visit arm:

3 (15 min) visits with

similar content to GV

Deakin,

2006a [63]

N = 314

UK

61.6 yrs

Gender NR

Race NR

6.7 yrs with DM


1.5 months

16 patients (mean)

SME 1 diabetes research

dietitian/educator

Usual care+:

diabetes education and

review with individual

appointments with a

dietitian (30 min),

practice nurse (15 min)

and physician (10 min)

Dejesus,

2009 [64]

N = 54

US

76% aged 60+

48%

Race NR

Duration NR

1 session

7 patients

DE Diabetes nurse

educator

Usual care

Hornsten,

2008 [65]

N = 104

Sweden

63 yrs

54%

Race NR

All diagnosed � 2yrs


months

9 months

5–8 patients

SME Diabetes nurses Usual care

Khunti,

2012 [66]

Davies,

2008 [67]

N = 824

UK

DESMOND

59.5 yrs

55% male

6% minority

Duration NR

1 (6 h) session

1 day or 2 half-days

Group size NR

SME Healthcare professional Usual care+:

(resources to provide

equivalent contact time

as intervention)

Kulzer,

2007a [68]

N = 193

Germany

Mean age 55.6

50.3% male

Race NR

Mean duration 6.6 yrs

GV1: 4 DE sessions

GV2: 12 SME sessions

GV3: 6 SME sessions + 6

IV

Duration NR

Group size 6–10

GV1: DE

GV2: SME

Health psychologist Self-management

education - 6 90 min

group lessons and 6

90 min individual

lessons

Lorig,

2009 [69]

N = 345

US

DSMP

66.55 yrs

35.7%

32.7% minority

Duration NR

6 weekly (2.5 h)

sessions

1.5 months

10–15 patients

SME Peer leaders Usual care

Lujan,

2007 [70]

N = 150

US

58 yrs

20%

100% Mexican origin

Duration NR


2 months

6 patients (English

class)

23 patients (Spanish

class)

DE 2 leaders

Promotoras, nurses,

dietitians, social

workers

Usual care

(2 pamphlets)



Study Sample size

Setting

Program name,

if applicable

Demographics:

Mean age

% male

% minority Mean

disease duration

GV structure:

# Visits, frequency

Duration

Group size

GV content:

SME (self-

management)

DE (didactic)

EE (experiential)

GV leaders:

Number of leaders

Profession type

Comparator

Melkus,

2010a [71]

N = 109

UK

46 yrs

0%

100% minority

Duration NR

11 weekly (1–2 h)

sessions

3 months

Group size NR

SME Nurse practitioner Culturally neutral

group DE (10 weekly

sessions)

Miller,

2002 [72]

N = 98

US

72.5 yrs

47%

17% black

7.2 yrs

GV1:

10 weekly (1.5–2 h)

sessions

2.5 months

Group size NR

GV2:

Offered 6 (2 h) sessions

Group size NR

GV1: DE, EE

GV2: DE

Dietitian GV2 (participants were

mailed printed material

if they did not attend

the group session)

Philis-Tsimikas,

2011 [73]

N = 207

US

Project Dulce

50.7 years

29% male

Minority NR

Duration NR


8 monthly support

groups

10 months

Group size NR

DE Trained peer educator Usual care

Raji,

2002a [74]

N = 106

US

VAMC

Mean age 60 yrs

99% male

Race NR

Duration NR

4 daily sessions

4 sequential days

4–6 patients

DE Physician, nurse,

nutritionist,

pharmacist, exercise

physiologist, social

worker, and diabetes

educator

2 comparators: passive

education and no-

intervention

Rickheim,

2002a [75]

N = 170

US

Mean age 52.5

34% male

Race 7% non-white?

Duration 0.9 yrs

4 sessions (at 0, 2 wks,

3mo, 6 mo)

6 months total

DE A diabetes nurse

specialist (RN) and

diabetes nutrition

specialist (RD)

Individual education

sessions

Rosal,

2011 [76]

N = 252

US

Latinos en Control

83.7% aged 45+

23.4%

87.7% minority

31.3% diagnosed <5 yr

12 weekly + 8 monthly

First session (1 h)

individual Remaining

(2.5 h) group

11 months

Group size NR

SME, EE Nutritionist or health

educator and lay leader

or 3 supervised lay

leaders

Usual care

Rygg,

2012 [77]

N = 146

Norway

66 yrs

55%

0%

5 yrs with DM

3 biweekly (5 h)

sessions

1.25 months

8–10 patients

DE, EE Diabetes nurses; also

included physician,

physiotherapist,

nutritionist, and lay

person

Usual care

Sarkadi,

2004a [78]

N = 77

Sweden

Mean age 66

% male NR

Race NR

Duration 5.9 yrs

treatment; 2.6 yrs

control

12 monthly sessions

1 year total

Group size NR

DE Pharmacists trained to

be facilitators, and a

nurse specialist

Usual care

Scain,

2009 [79]

N = 104

Brazil

59 yrs

47%

9.4% black

10.5 yrs


1 months

8–10 patients

DE NR Usual care

Schillinger,

2009 [80]

N = 339

US

IDEALL

56.1 years

41% male

92.3% minority

Duration NR

9 monthly (1.5 h)

sessions

9 months

6–10 patients

SME 2 leaders

Physician and

language-concordant

health educator

Usual care;

automated telephone

self-management

support group (39

weekly, automated

calls over 9 months,

nurse phone follow-up)

Sharifirad,

2012 [81]

N = 97

Iran

BASNEF

67.05 yrs

35%

Minority NR

14 yrs with DM

4 (70 min) sessions

1 month

Group size NR

DE Physician, specialist of

endocrine disorder,

diabetes nurse, and

nutritionist.

Usual care

Sperl-Hillen,

2011a [82]

N = 623

US

IDEA

Mean age 61.8

50.6% male

22.1% Hispanic

5.5% Black

Duration 11.7 yrs

4 weekly sessions

4 weeks total

1–10 patients (mean 5)

DE Nurses and dietitians

trained to facilitate GE

sessions

3 individual education

sessions at 1-month

intervals

Steed,

2005 [83]

N = 127

UK

UCL-DSMP

59.8 yrs

71.2% male

51% minority

10.8 years

5 weekly (2.5 h)

sessions

1 booster (2.5 h) @ 3

months

1.25 months

Group size NR

SME Diabetes specialist

nurses and dietitians

Usual care



Study Sample size

Setting

Program name,

if applicable

Demographics:

Mean age

% male

% minority Mean

disease duration

GV structure:

# Visits, frequency

Duration

Group size

GV content:

SME (self-

management)

DE (didactic)

EE (experiential)

GV leaders:

Number of leaders

Profession type

Comparator

Surwit,

2002a [84]

N = 108

US

Mean age 57.4

58.3% male

8.3% Black

1% Asian

5 weekly sessions

5 weeks total

Group size NR

EE NR DE group visits

Toobert,

2011 [85]

N = 280

US

!

Viva Bien!

57.11 yrs

0%

100% minority

10.4 yrs with DM

2.5-day retreat + 36

weekly and biweekly

sessions

12 months

Group size NR

DE, EE Bilingual physician,

dietitian, exercise

instructor, bilingual

facilitator

Usual care

Weinger,

2011a [87]

N = 222

US

52.5 yrs

49.5%

10.3% minority

17.2 yrs with DM


wks

1.5 months

Group size NR

SME Certified diabetes

educator

Unlimited access to

individual DM nurse

and dietitian visits

Zapotoczky,

2001a [88]

N = 34

Austria

Mean age 62 yrs

36% male

12 monthly sessions

1 year total

18 patients

DE Clinical dietitian All subjects received 4-

wk group education.

Controls received usual

care with no further

group education.

Multiple chronic conditions

Lorig,

1999 [93]

N = 952a US CDSMP Mean age 65

35% male

9.7% non-white

Duration NR (heart

disease, lung disease,

arthritis, and stroke)

7 weekly sessions

7 weeks total

10–15 patients

SME 2 trained peer leaders Usual care

Lorig,

2003 [92]

N = 551

US

CDSMP (Spanish)

Mean age 57

21% male

Race NR

Duration NR

6 weekly sessions

6 weeks total

10–15 patients

SME 2 trained peer leaders Usual care

Fu, 2003 [91] N = 954

China

CDSMP

Mean age 64

29% male

Race NR

Duration NR

7 weekly sessions

7 weeks total

Group size NR

SME 2 trained peer

volunteer leaders

Usual care

Elzen,

2007 [90]

N = 136

Netherlands

CDSMP

Mean age 68

37% male

Race NR

Duration NR

6 weekly sessions

6 weeks

10–13 patients

SME 2 psychologists or 1

psychologist plus peer

leader

Usual care

Chronic pain

Ersek,

2003 [96]

N = 45

US

Age 81.9

Gender 13%

Race 84.71% Caucasian

Duration NR

7 weekly sessions

8 weeks

3–8 patients

SME 2 leaders

Doctoral-level health

providers

Receipt of an

educational booklet on

pain

Gustavsson,

2010 [97]

N = 156

Sweden

PASS

Age 45.7

Gender 11%

Race NR

Duration NR

7 weekly sessions

7 weeks + 1 booster at

week 20

Group size NR

SME, EE 1 leader

Physical therapists

Individual physical

therapy sessions

Haugli, 2000,

2003 [94,98]

N = 174

Norway

Age 43.08

Gender 2.27%

Race NR

Duration 9.89 years

12 every-other-week

sessions

9 months (including a

summer break)

6–10 patients

SME 2 leaders

Nurses, physicians

physiotherapists

Usual care

Vlaeyen,

1996 [95]

N = 131

Netherlands

Age 44

Gender 12%

Race NR

Duration 10.2 years

12 sessions

6 weeks

Maximum of 6 patients

GV1: SME, EE

GV2: DE, EE

Rehabilitation staff,

psychologist

Usual care

a Not included in meta-analysis.


Table 3Findings from studies comparing group visits to usual care.

Study

Sample size

Outcome (range

or units, as applicable)

Findings by time periodaUnless otherwise noted, results represent the difference in mean change from baseline (Treatment – Control,

95% CI), or the mean (SD) change from baseline comparing Treatment vs. Control.

GV duration

(months);

# sessions

% Participationb

/% Loss to

follow-upc

Study

quality

0–3 mo 4–6 mo 7–12 mo 13+ mo

ArthritisSelf-efficacy

Ackerman, 2012 [17]

N = 120

heiQ – Skill and technique

acquisition (range 1–6)

0.18 (–0.03 to 0.39) NR 0.07 (�0.18 to 0.32) NR 1.5/6 25/22 Poor

Barlow, 2000 [18]

N = 544

ASES (pain, 5–50 "= better) NR 2.65 (0.85 to 4.44) NR NR 1.5/6 NR/22 Fair

Breedland, 2011 [19]

N = 34

ASES (pain + other symptoms) Mean change (SD), T vs.

C: 0.42 (0.71) vs. 0.28

(0.85)

NR NR NR 2/24 NR/6 Good

ASES (function) Mean change (SD), T vs.

C: 0.29 (0.57) vs. 0.10

(0.38)

NR NR NR

Buszewicz, 2006 [20]

N = 812

ASES (pain) NR 1.63 (0.83 to 2.43) 0.98 (0.07 to 1.89) NR 1.5/6 30/24 Fair

Giraudet-Le Quintrec,

2007 [24]

N = 208

AHI (coping) NR NR �1.22 (5.55) vs. 0.22

(3.81), p = 0.0319

NR 2 (+ booster at 4)/9 18/9 Fair

Hammond, 1999 [16]

N = 35

ASES Median (IQR), T vs. C:

5.40 (3.63 to 6.70) vs.

5.50 (4.10 to 7.55)

NR NR NR 1/4 NR/31 Fair

Lorig, 1985 [27]

N = 286

Knowledge + self-management

scale (0–10)

NR 1.39 (2.00) vs. 0.10

(1.48) p<0.0001

NR NR 4/6 NA/16 Fair

Lorig, 1999 [28]

N = 331

ASES self-efficacy (0–10,

"= better)

NR 1.1 (2.2) vs.�0.04 (2.2),

p = 0.0001

NR NR 1.5/6 NR/17 Poor

Riemsma, 2003 [30]

N = 218

ASES (pain) 0.0 (0.7) vs. 0.1 (0.8),

p>0.1

0.3 (0.7) vs. 0.0 (0.8),

p>0.1

0.3 (0.7) vs. 0.1 (0.8),

p = 0.06

NR 1.25 + booster @ 3,

6, 9/8

26/17 Fair

ASES (function) 0.0 (0.5) vs. 0.0 (0.5),

p>0.1

0.0 (0.5) vs. 0.0 (0.6),

p>0.1

0.0 (0.7) vs. 0.0 (0.7),

p>0.1

NR

Taal, 1993 [32]

N = 75

ASES (pain) 0.43 vs. 0.11, p>0.05 0.40 vs. 0.24, p>0.05 NR 0.33 vs. 0.15, p>0.05 1.25/5 54/24 Poor

ASES (function) 0.17 vs. �0.13, p<0.05 0.07 vs. �0.02, p>0.05 NR 0.17 vs. �0.06, p<0.05

Quality of life/functional status

Ackerman, 2012 [17]

N = 120

AQoL (�0.04 to 1.0) �0.04 (�0.10 to 0.01) NR �0.02 (�0.09 to 0.05) NR 1.5/6 25/22 Fair

Barlow, 2000 [18]

N = 544

HADS (depression, 0–21

#= better)

NR �0.86 (99% CI �1.46 to

�0.26)

NR NR 1.5/6 NR/22 Fair

Breedland,

2011 [19]

N = 34

Dutch AIMS2 (physical) �0.68 (1.00) vs. �0.14

(0.49), p = 0.07


Dutch AIMS2 (psychological) �0.34 (1.11) vs. 0.08

(1.37), p = 0.4

NR NR NR

Dutch AIMS2 (social) �0.48 (1.90) vs. �0.88

(2.03), p = 0.6

NR NR NR

Buszewicz,

2006 [20]

N = 812

SF-36 (physical) NR 0.22 (�1.50 to 1.94) 0.33 (�1.31 to 1.98) NR 1.5/6 30/24 Fair

Giraudet-Le Quintrec,

2007 [24]

N = 208

EMIR (physical) NR NR 0.06 (5.69) vs. 0.33

(5.06), p = 0.7211

NR 2 + booster @ 4/9 18/9 Fair

EMIR (symptomatic) NR NR �1.00 (3.34) vs. �0.02

(3.13), p = 0.0311

NR

EMIR (psychological) NR NR �0.87 (3.34) vs. �0.38

(3.60), p = 0.3174

NR

EMIR (social) NR NR �0.26 (2.24) vs. �0.21

(2.35), p = 0.8684

NR

EMIR (work) NR NR 0.14 (2.57) vs. �0.06

(2.62), p = 0.6453

NR

A.R

. Q

uin

on

es et

al.

/ P

atien

t E

du

catio

n a

nd

Co

un

seling

95

(20

14

) 3

–2

91

8

Hammond, 1999 [16]

N = 35

HAQ (function) Median (IQR), T vs. C:

1.44 (1.09 to 1.91) vs.

1.50 (0.78 to 1.81)

NR NR NR 1/4 NR/31 Fair

Lorig, 1985 [27]

N = 286

HAQ (disability 0–3) NR �0.06 (0.34) vs. �0.02

(0.24), p = 0.45


Lorig, 1999 [28]

N = 331

HAQ (disability 0–3 #= better) NR �0.10 (0.49) vs. 0.0

(0.41), p = 0.03


Patel, 2009 [21]

N = 812

QALYs �0.006 (�0.01 to

�0.00)

NR �0.01 (�0.03 to 0.01) NR 1.5/6 30/24 Fair

Riemsma, 2003 [30]

N = 218

Dutch AIMS2

(function)

�0.2 (0.7) vs.�0.1 (0.8),

p>0.1

0.0 (0.8) vs. �0.1 (0.9),

p>0.1

�0.1 (0.8) vs. 0.0 (1.1),

p>0.1

NR 1.25 + booster @ 3,

6, 9/8

26/17 Fair

Taal, 1993 [32] N = 75 Dutch AIMS

(physical activities 0–10)

0.08 vs. �0.56, p>0.05 �0.64 vs. �0.56,

p>0.05

NR �0.16 vs. �0.48,

p>0.05

1.25/5 54/24 Poor

Biophysical and performance measures

Breedland, 2011 [19]

N = 34

VO2 max (mL/kg/min) 3.82 (3.86) vs. �0.44

(2.21), p = 0.002


Utilization

Ackerman, 2012 [17]

N = 120

MD visits 1 (0–10) vs. 1 (0–4),

p = 0.73

NR 1 (0–6) vs. 1 (0–10),

p = 0.40

NR 1.5/6 25/22 Fair

Barlow, 2000 [18]

N = 544

MD visits NR 0.20 (99% CI �0.30 to

0.70)


Lorig, 1985 [27]

N = 286

MD visits NR �0.46 (1.98) vs. �0.26

(3.07), p = 0.61


Lorig, 1999 [28]

N = 331

MD visits NR �0.06 (2.9) vs. �0.24

(1.7), p = 0.16


Patel, 2009 [21] Outpatient visits NR 0.17 (0.6) vs. 0.21 (0.6) 0.28 (0.7) vs. 0.26 (0.9) NR 1.5/6 30/24 Fair

History of fallsSelf-efficacy

Arnold, 2010 [33]

N = 83

ABC (falls efficacy) 5.8 (12.4) vs. 2.4 (10.7) NR NR NR 2.75/22 55/23 Fair

Clemson, 2004 [34]

N = 310

Mobility Efficacy Scale (MES) NR NR NR 4.28 (�8.40 to �0.54) 1.75 + booster @ 3/8 NA/15 Good


Arnold, 2010 [33]

N = 83

AIMS2 (BBSm) 1.0 (3.5) vs. 0.2 (2.3) NR NR NR 2.75/22 55/23 Fair

Clemson, 2004 [34]

N = 310

SF-36 (physical) NR NR NR 0.70 (�2.94 to 1.88) 1.75 + booster @ 3/8 NA/15 Good


Arnold, 2010 [33]

N = 83

TUG (dual task) �2.3 (5.1) vs. 0.2 (4.3) NR NR NR 2.75/22 55/23 Fair

Clemson, 2004 [34]

N = 310

Fall events NR NR NR 0.69 (0.50 to 0.96) 1.75 + booster @ 3/8 NA/15 Good

Ryan, 1996 [35]

N = 45

Fall events Unclear NR NR NR 1 day/1 NR/NR Poor

Shumway-Cook, 2007

[36]

N = 454

Fall events (incidence rates per

person-year)

NR NR 1.33 (0.14) vs. 1.77

(0.28), p = 0.15

NR 6/6 88/5 Fair

TUG NR NR �0.7 (�1.2 to �0.2) NR

AsthmaSelf-efficacy

Abdulwadud, 1999 [38]

N = 125

Asthma Attitudes and Beliefs

Questionnaire

NR Change in % of patients

who would take action

before asthma

deteriorated �6 vs. 9

NR NR 0.7/3 71/38 Poor


Abdulwadud, 1999 [38]

N = 125

AQLQ (maximum score 10) NR �0.26 (�1.3 to 0.77) NR NR 0.7/3 71/38 Poor

A.R

. Q

uin

on

es et

al.

/ P

atien

t E

du

catio

n a

nd

Co

un

seling

95

(20

14

) 3

–2

9

19


Study

Sample size

Outcome (range




GV duration

(months);

# sessions

% Participationb

/% Loss to

follow-upc

Study

quality

0–3 mo 4–6 mo 7–12 mo 13+ mo

Wilson, 1993 [37]

N = 130

Asthma bother scale NR NR % of patients who were

less bothered after 1 yr

55 vs. 25 (p = 0.028)

NR 3–4/4 56/14 Fair

Utilization

Wilson, 1993 [37]

N = 130

Acute visits NR NR Mean (SE) change in

annualized rate 0.93

(0.42) vs. 0.93 (0.35)

(p = ns)

Mean (SE) change in

annualized rate 2.04

(0.33) vs. 1.01 (0.33)

(p<0.05)

1/4 56/14 Fair

Bolton, 1991 [40]

N = 241

ER visits NR Mean (SD) per 100

persons 68 (110) vs.

220 (530), p = 0.003

Mean (SD) per 100

persons 69 (130) vs. 98

(210), p = 0.42

NR NR/3 45/7 Fair

COPDQuality of life/functional status

Wilson, 2008 [42]

N = 91

Smoking cessation NR NR Complete or

intermittent (% of

patients) 10 vs. 6,

p = 0.7

NR 1.2/5 60/NR Fair

Ninot, 2011 [45]

N = 45

SGRQ NR NR median �7.6 (p<0.01)

vs. �4.7 (p = 0.33)

NR 1/8 NA/16 Good


Ninot, 2011 [45]

N = 45

6MWD NR NR 30 (p<0.01) vs. 12.5

(p = 0.52)

NR 1/8 NA/16 Good

Utilization

Ninot, 2011 [45]

N = 45

Days in hospital for COPD

admission

NR NR 0.0 (p = 0.64) vs. 0.0

(p = 0.50)

NR 1/8 NA/16 Good

CHF/CADSelf-efficacy

Smeulders, 2010

[49,50] N = 317

GSES (16–80) Adjusted mixed effects,

mean difference (95%

CI) �1.32 (�3.05 to

0.42)

Adjusted mixed effects,


CI) �0.22 (�2.10 to

1.66)



CI) �1.59 (�3.59 to

0.40)

NR 1.5/6 44/16 Good

Cardiac self-efficacy: KCCQ (0–

36)



CI) 0.01 (�1.42 to 1.45)



CI) 1.13 (�0.48 to 2.73)



CI) �0.93 (�2.50 to

0.63)

NR

Cognitive Symptom

Management (0–25)



CI) �2.28 (�3.23 to

�1.32)



CI) �0.84 (�1.86 to

0.18)



CI) �0.73 (�1.89 to

0.43)

NR

Moore, 2006 [47]

N = 250

Index of Self-Regulation (range

not reported)

37.3 (5.3) v 37.5 (5.6) NR NR 36.8 (5.7) vs. 36.9 (5.5) 3/5 50/19 Fair


A.R

. Q

uin

on

es et

al.

/ P

atien

t E

du

catio

n a

nd

Co

un

seling

95

(20

14

) 3

–2

92

0

Smeulders, 2010

[49,50]

N = 317

Cardiac-specific QOL (0–100) Adjusted mixed effects,


CI) �4.10 (�6.95 to

�1.25)



CI) �3.13 (�6.29 to

0.02)



CI) �2.96 (�6.68 to

0.75)

NR 1.5/6 44/16 Good

HADS–Anxiety (0–21) Adjusted mixed effects,


CI) 0.26 (�0.45 to 0.97)



CI) �0.12 (�0.83 to

0.59)



CI) �0.50 (�1.27 to

0.28)

NR

HADS Depression (0–21) Adjusted mixed effects,


CI) 0.28 (�0.94 to 0.38)



CI) �0.24 (�0.94 to

0.45)



CI) �0.35 (�1.16 to

0.47)

NR

Chang, 2005 [48]

N = 62

Minnesota Living with Heart

Failure Questionnaire, Disease-

specific QoL, physical health

NR 0.8 (6.3) vs. 0.9 (12.2) NR NR 3.75/15 17/13 Fair

HypertensionBiophysical

Nessman, 1980 [51]

N = 52

DBP % of subjects attaining

8-week values of

�90 mm Hg (T v C) 92.3

vs. 53.8; p<0.01

% of subjects attaining

8-week values of

�90 mm Hg (T v C) 96.0

vs. 61.9, p<0.01

NR NR 2/4 36/0 Poor

Rujiwatthan-akorn,

2011 [52]

N = 96

SBP and DBP SBP: �23.59 vs. �5.94,

p = 0.001 DBP:�6.75 vs.

1–49, p = 0.035

NR NR NR 2/3 70/12 Poor

Balcazar, 2009 [53]

N = 98

BP Normal: SBP <120 and DBP

<80 mmHg

Increase in % of subjects

15 vs. �16

NR NR NR 2/4 NR/0 Poor

Scala, 2008 [55]

N = 292

SBP and DBP NR NR NR SBP: 138.4 (2.5) vs.

146.4 (2.7), p<0.001

DBP: 87.3 (3.7) vs. 89.8

(6.3), p<0.001

4/3 NR/42 Poor

Svetkey, 2009 [56]

N = 574

SBP, main effect for the patient

intervention, SBP change in

mmHg

NR SBP main effect: �2.6

(�4.4 to �0.7) Mean

(SD) SBP: �7.1 (12.1)

vs. �6.7 (12.8) Mean

DBP: �3.4 vs. �3.6

NR SBP main effect: �0.2

(p = 0.89); Mean SBP:

�6.8 v�7.5; Mean DBP:

�3.4 v �4.9

6/20 56/12 Good Note:

Results

presented

are for MD-

control

group (i.e. no

physician

intervention)

Diabetes-mellitusSelf-efficacy

Brown, 2002 [60]

N = 256

Study specific health belief

scale (control)

3.63 (1.33) vs. 3.41

(1.40)

NR 3.65 (1.26) vs. 3.43

(1.26)

NR 12/27 NR/NR Poor

Adolfsson, 2007 [58]

N = 101

Study specific questionnaire NR NR Median difference

(1st and 3rd quartile)

9.8 (0.6, 17.2) vs. 4.0

(�4.2, 14.8), p = 0.272

NR 7/5 53/13 Fair

Khunti, 2012 [66],

Davies, 2008 [67]

N = 824

IPQ-R (illness coherence, 5–25) NR Median (inter-quartile

range) 19 (16 to 20) vs.

18 (14 to 20)

Median (inter-quartile

range) 20 (17 to 20) vs.

19 (15 to 20)

NR 1 day or 2 half-

days/1

NA/11 Good

Lorig, 2009 [69]

N = 345

PAM (0–100, "= better) NR 4.52 (15.8) vs. 1.75

(15.3), p = 0.17

NR NR 1.5/6 NA/15 Fair

Diabetes Self-Efficacy scale (1–

10, "= better)

NR 0.495 (1.52) vs. �0.122

(1.709), p = 0.001

NR NR

Lujan, 2007 [70]

N = 150

DHBM (25 item, 100 total point

possible)

53.7 (11.2) vs. 55.7

(11.5)

54.6 (8.4) vs. 50.8 (13.6) NR NR 2/8 NR/6 Fair

Rosal, 2011 [76]

N = 252

Study specific scale (diet &

physical activity change)

NR Mean difference

(range) 0.316 (0.194 to

0.439), p<0.001

Mean difference

(range) 0.235 (0.093 to

0.376), p = 0.001

NR 11/19 57/16 Fair

A.R

. Q

uin

on

es et

al.

/ P

atien

t E

du

catio

n a

nd

Co

un

seling

95

(20

14

) 3

–2

9

21


Study

Sample size

Outcome (range




GV duration

(months);

# sessions

% Participationb

/% Loss to

follow-upc

Study

quality

0–3 mo 4–6 mo 7–12 mo 13+ mo

Rygg, 2012 [77]

N = 146

PAM NR 1.8 (�2.4 to 6.1) 0.6 (�3.4 to 4.6) NR 1.25/3 91/9 Fair

Schillinger, 2009 [80]

N = 339

Diabetes Quality Improvement

Program, self-efficacy measure

(0–100)

NR NR 5.5 (1.4 to 9.6) NR 9/9 73/10 Fair

Steed, 2005 [83]

N = 127

MDS (total) 72.78 (15.74) vs. 69.81

(16.24), p = ns

NR NR NR 1.25 + booster @ 3/6 51/16 Poor

Toobert, 2011 [85]

N = 280

Confidence in Overcoming

Challenges to Self Care

NR 3.5 (0.7) vs. 3.3 (0.8) 3.4 (0.7) vs. 3.3 (3.8) NR 12/37 61/22 Fair


Adolfsson, 2007 [58]

N = 101

Adapted WHO QOL NR NR Median difference (1st

and 3rd quartile) 2.5

(�3.9, 7.5) vs. 0.0 (�6.0,

7.5), p = 0.588

NR 7/5 53/13 Fair

Khunti, 2012 [66],

Davies, 2008 [67]

N = 824

HADS (0–21) NR Median (inter-quartile

range) 2 (1–4) vs. 2 (1–

5)

Median (inter-quartile

range) 2 (1–5) vs. 2 (1–

6)

NR 1 day or 2 half-

days/1

NA/11 Good

Lorig, 2009 [69]

N = 345

PHQ-9 (depression 0–27,

#= better)

NR �1.51 (4.33) vs. 0.541

(4.49), p<0.001


Rygg, 2012 [77]

N = 146

SF-36 (physical) NR 0.3 (�1.9 to 2.5) 0.6 (�1.6 to 2.8) NR 1.25/3 91/9 Fair

SF-36 (mental) NR �1.7 (�4.0 to 0.7) �2.1 (�4.5 to 0.4) NR

EQ-VAS NR 1.0 (�2.7 to 5.9) 2.4 (�2.5 to 7.2) NR

Schillinger, 2009 [80]

N = 339

SF-12 (physical) NR NR �0.1 (�6.9 to 6.7) NR 9/9 73/10 Fair

SF-12 (mental) NR NR �2.9 (�8.6 to 2.9) NR

Steed, 2005 [83]

N = 127

ADDQOL Authors report

significant differences

between groups at

follow up F(1,

97) = 9.33, p<0.01

NR NR NR 1.25+ booster @ 3/6 51/16 Poor

SF-36 No significant

differences between

groups, numeric values

not reported

NR NR NR

Toobert, 2011 [85]

N = 280

CDC Healthy Days (physical, #

of days not good)

NR 7.3 (9.8) vs. 7.5 (9.7) 6.1 (8.4) vs. 8.1 (9.2) NR 12/37 61/22 Fair

CDC Healthy Days (mental, # of

days not good)

NR 6.7 (9.5) vs. 7.4 (10.0) 7.3 (8.7) vs. 7.3 (9.5) NR


Dejesus, 2009 [64]

N = 54

SBP NR 0.35 (15.04) vs. �5.17

(11.23), p = 0.41

NR NR 1 day/1 13/55 Poor

DBP NR 0.53 (8.94) vs. �1.5

(10.25), p = 0.46

Utilization

Dejesus, 2009 [64]

N = 54

RN and MD visits NR No significant

differences between


not reported

NR NR 1 day/1 13/55 Poor

A.R

. Q

uin

on

es et

al.

/ P

atien

t E

du

catio

n a

nd

Co

un

seling

95

(20

14

) 3

–2

92

2

Lorig, 2009 [69]

N = 345

MD visits NR 0.219 (3.88) vs. �0.106

(3.37), p = 0.462


ED visits NR �0.012 (0.742) vs.

0.113 (0.885), p = 0.273

NR NR

Days hospitalized NR 0.00 (4.31) vs. �0.143

(1.75), p = 0.152

NR NR

Rygg, 2012 [77]

N = 146

Clinician visits NR Difference n (%) �9

(�12%), p = 0.106

Difference n (%) 5 (7%),

p = 0.354

NR 1.25/3 91/9 Fair

2

Multiple chronic conditionsSelf-efficacy

Lorig 1999 [93]

N = 952

Cognitive symptom

management (0–5, "= better)

NR 0.38 (0.77) vs. 0.07

(0.73), p = 0.0001


Lorig, 2003 [92]

N = 551

4-item self-efficacy scale (1–10,

"= better)

NR 1.157 (3.08) vs. 0.719

(3.09), p = 0.0006


Fu, 2003 [91]

N = 954

4-item self-efficacy scale.

Managing disease (1–10

increase = improvement)

NR 0.22 (2.52) vs. �0.41

(2.87) p = 0.001


Cognitive symptom

management scale (0–5

increase = improvement)

NR 0.37 (0.95) vs. 0.04

(0.76), p = 0.005

NR NR

Elzen, 2007 [90]

N = 136

GSES-16 (Dutch; 5-point Likert

scale)

No significant

differences between


not reported t = 1.37,

p = 0.09

No significant

differences between



p = 0.06

NR NR 1.5/6 26/10 Poor

Cognitive symptom scale

(Dutch; 6-point Likert scale)

No significant

differences between


not reported t =�1.42,

p = 0.08

No significant

differences between



p = 0.14

NR NR

Quality of life

Lorig, 1999 [93]

N = 952

Self-rated health, (1–5,

#= better)

NR �0.09 (0.72) vs. 0.02

(0.69), p = 0.02


Disability (HAQ), (0–3,

#= better)

NR �0.02 (0.32) vs. 0.03

(0.36), p = 0.002

NR NR

Elzen, 2007 [90]

N = 136

RAND-36 Physical components

(0–100)

No significant

differences between



p = 0.29

No significant

differences between



p = 0.45

NR NR 1.5/6 26/10 Poor

RAND-36 Mental components

(0–100)

No significant

differences between



p = 0.35

No significant

differences between



p = 0.46

NR NR

Utilization

Lorig, 1999 [93]

N = 952

Physician and ER visits in past 6

months

NR �0.77 (5.6) vs. �0.54

(6.3), p = 0.11


Hospital stays in past 6 months NR �0.07 (0.69) vs. �0.05

(1.1), p = 0.047

NR NR

Lorig, 2003 [92]

N = 551

Physician visits in past 4

months

NR �0.475 (2.78) vs.

�0.034 (2.44), p = 0.057


ER visits in past 4 months NR �0.083 (0.622) vs.

0.101 (0.722), p = 0.005

NR NR

Hospital days in past 4 months NR �0.011 (0.836) vs.

�0.079 (3.54), p = 0.481

NR NR

A.R

. Q

uin

on

es et

al.

/ P

atien

t E

du

catio

n a

nd

Co

un

seling

95

(20

14

) 3

–2

9

23


Study

Sample size

Outcome (range




GV duration

(months);

# sessions

% Participationb

/% Loss to

follow-upc

Study

quality

0–3 mo 4–6 mo 7–12 mo 13+ mo

Fu, 2003 [91]

N = 954

Physician visits in past 6

months

NR �1.01 (9.41) vs. �0.84

(7.76), p = 0.72


ER visits in past 6 months NR �0.04 (0.96) vs. �0.03

(0.72), p = 0.44

NR NR

Hospital stays in past 6 months NR �0.06 (0.46) vs. 0.06

(0.92), p = 0.04

NR NR

Chronic painSelf-efficacy

Ersek, 2003 [96]

N = 45

Survey of Pain Attitudes,

control (range 0 to 4)

2.3 (0.6) vs. 2.3 (0.8),

p = ns

NR NR NR 2/7 NA/13 Fair

Vlaeyen, 1996 [95]

GV1 vs. UC N = 131

Pain coping 12.2 (2.7) vs.�1.0 (2.7),

p = 0.001

NR NR NR 1.5/12 NR/20 Fair

Pain control 1.6 (1.4) vs. 0.4 (1.4),

p = 0.003

NR NR NR

Quality of life

Ersek, 2003 [96]

N = 45

SF-36 (role physical) 28.4 (34.8) vs. 28.4

(35.6), p = ns

NR NR NR 2/7 NA/13 Fair

SF-36 (physical functioning) 38.4 (19.0) vs. 43.0

(23.7), p = ns

NR NR NR

Graded chronic pain scale–

activity interference

3.2 (2.9) vs. 3.6 (3.0),

p = ns

NR NR NR

Haugli, 2000 & Haugli,

2003 [94,98]

N = 174

VAS (pain) 45.9 (23.4) vs. 52.7

(21.5)

NR 44.0 (22.9) vs. 54.8

(24.3)

NR 9/12 NA/33 Poor

a Unless otherwise noted, results represent the difference in mean change from baseline (Treatment–Control, 95% CI), or the mean (SD) change from baseline comparing Treatment vs. Control.b Defined as percent eligible for enrollment among those invited to participate.c Defined as percent lost to follow up among those randomized.

A.R

. Q

uin

on

es et

al.

/ P

atien

t E

du

catio

n a

nd

Co

un

seling

95

(20

14

) 3

–2

92

4

Table 4Findings from studies comparing head-to-head interventions.

Study

(sample size)

Arm 1 Arm 2 % Participationa

/% Loss

to follow-upb

Study quality

(Good/Fair/

Poor)

Comment

Arthritis

Hewlett,

2011 [25]

GV1 (7 SME sessions) GV2 (1 DE session) 15/24 Good Beneficial effect of cognitive behavior therapy relative

to didactic-only single session GV assessed at 4.5

months

Ettinger, 1997

[22, p. 2258]

GV1 (3 DE sessions) GV2 aerobic exercise

(36 classes)

53/17 Fair Beneficial effect of either exercise group vs. education

group on pain, disability, and functional performance.

Dose response for patients who completed more

sessions of either exercise program.

GV1 (3 DE sessions) GV3 resistance exercise

(36 classes)

Freeman,

2002 [23]

GV1 (4 SME sessions) GV2 (4 DE sessions) 94/23 Fair Cognitive-behavioral education program did not

significantly improve pain or self-efficacy for patients

newly diagnosed with RA.

Hammond,

2008 [99]

GV1 (9 SME, EE

sessions)

GV2 (5 DE, EE sessions) 46/37 Fair GV1 was effective in improving short-term pain,

functional disability, self-efficacy, and reducing

physician visits compared to GV2. Longer-term benefits

for GV1 for pain, and maintained functional ability

compared to declines in GV2.

Kaplan, 1981 [26] GV1 (13 DE, group

counseling sessions)

GV2 (1 DE session) NR/35 Poor Combination of education and short-term group

counseling led to improved knowledge and self-esteem.

Lorig, 2004 [29] GV (6 SME sessions) Mailed individual

program

84/32 Good Both programs show moderate improvements in self-

efficacy, pain, and disability outcomes. Earlier

advantages of mailed program narrowed after 3 yrs. GV

program had decreased physician visits compared with

mailed program.

Riemsma,

2003 [30]

GV1 (8 SME, EE

sessions)

Patients only

GV2 (8 SME, EE

sessions)

Patients and significant

others

26/17 Fair Participation of significant others led to decreases in

self-efficacy for coping with other symptoms compared

to improvements in patients participating without their

partners.

History of falls

Arnold, 2010 [33] GV1 (22 DE, EE, aquatic

exercise classes)

GV2 (22 EE, aquatic

exercise classes)

55/23 Fair Combination of aquatic exercise and education resulted

in improvements in functional performance vs. aquatic

exercise alone.

Ryan, 1996 [35] GV (1 DE) Individual (1 DE) NR/NR Poor Small study. Control group experienced the most falls in

the post period.

Asthma

Wilson,

1993 [37]

GV (3 SME sessions) IV (3–5 weekly SME

sessions)

56/14 Fair No significant differences between GV and IV. GV and IV

were equally effective compared with UC. Reduced

bother and improved MDI technique observed with

both small group and individual education.

COPD

Bestall

2003 [43]

GV (16 DE + 26 EE

sessions: exercise)

GV (16 DE sessions) NR/16 Fair Compared with DE alone, pts in exercise group had

improved exercise capacity (shuttle walking distance)

that lasted 6 months. For QoL (CRQ, SGRQ) there were

mixed results at 6 months, and no differences between

groups at 1 year.

Effing

2011 [44]

GV (4 DE + up to 120 EE

sessions)

GV (4 DE sessions) 41/11 Fair COPE-active group experienced an improvement in

maximal exercise capacity compared to the steady

decline in the control group.

Kunik,

2008 [46]

GV (8 DE sessions) GV (8 SME sessions:

CBT)

19/55 Good CBT and COPD education groups were comparable and

significantly improved QoL, anxiety, depression, and

6MWD, with no significant differences between groups,

and improvement was maintained till the end of the

study (52 weeks).

Hypertension

Figar,

2006 [54]

N = 60

GV (4 SME sessions) GV (4 DE sessions) NR/17 Good More SME patients had controlled BP (defined as the

proportion of patients with mean 24-h BP <140/90 mm

Hg) compared with DE: 70% vs. 44%, p = 0.04. No

significant reductions in SBP or DBP in either group.

Diabetes mellitus

Deakin,

2006 [63]

GV (6 SME) Individual (3 DE) 20/32 Fair Significant improvements with group compared with

individual visits in glycemic control, total cholesterol

level, body weight, BMI and waist circumference,

reduced requirement for diabetes medication,

increased consumption of fruit and vegetables,

enjoyment of food, knowledge of diabetes, self-

empowerment, self-management skills and treatment

satisfaction.

De Greef,

2011 [62]

GV (3 SME) Individual (3 SME) 78/5 Good No improvement in biophysical health outcomes for

patients in the GV arm compared to individual visit

arm. Individual visit participants showed significant

improvements in waist circumference, FBG, HbA1c, and

total cholesterol compared to control arm.



Study

(sample size)

Arm 1 Arm 2 % Participationa

/% Loss

to follow-upb

Study quality

(Good/Fair/

Poor)

Comment

Kulzer,

2007 [68]

GV1 (4 DE) GV2 (12 SME) GV3 (6

Group + 6 individual

SME)

50/6 Fair GV2 (SME) had significantly lower HbA1c at 15 months

compared with both GV1 (DE) and GV3

(group + individual SME). GV2 (SME) also had

significant improvements in BMI, anxiety, and exercise

relative to GV1 (DE).

Melkus,

2010 [71]

GV1 (11 culturally

relevant SME)

GV2 (10 culturally

neutral DE)

NA/11 Fair Both arms had significant, similar reductions in HbA1c

at 24 months. The culturally relevant SME group had

significantly lower levels of diabetes-related emotional

distress at 24 months compared with the culturally

neutral DE group.

Miller,

2002 [72]

GV1 (10 DE, EE) GV2 (6 DE, or mailed

materials)

NA/6 Fair Intense nutrition education GV improved glycemic

control

Rickheim,

2002 [75]

GV (4 DE) Individual (4 DE) NR/46 Poor Individual and group education resulted in similar

improvements at 6 months in HbA1c, weight, BMI,

health-related QOL, attitudes, and medication regimen.

Schillinger,

2009 [80]

GV (9 SME) ATSM: automated

telephone self-

management (39 SME

calls)

73/10 Fair No statistical differences between GV and ATSM arms in

self-efficacy, or quality of life (physical). Improvement

for ATSM relative to GV in quality of life (mental).

Sperl-Hillen,

2011 [82]

GV (4 DE) Individual (3 DE); Usual

care

82/2 Fair HbA1c deceased significantly more with individual DE

compared with group DE and usual care. Individual DE

significantly reduced distress (PAID) and increased self-

efficacy compared with group DE.

Surwit,

2002 [84]

GV1 (5 DE) GV2 (5 DE + EE) NA/24 Poor At 1-year follow-up, patients who received training in

stress management in addition to DE had a 0.5%

reduction in HbA1c relative to DE alone. No differences

between groups in anxiety (STAI) or psychological

distress (GHQ; PSS) measures.

Weinger,

2011 [87]

GV (5 SME) Individual DE

(unlimited access to

DM nurse and dietitian

visits)

89/3 Fair GV (SME) had significantly greater reduction in HbA1c

levels over 1 year compared with individual DE. No

differences in QOL, and self-efficacy measures.

Zapotoczky,

2001 [88]

GV1 (4 wk + 12 mo DE) GV2 (4 wk DE) 100/0 Poor All subjects received 4-wk group education. GV2

received usual care with no further group education.

Significant reductions in HbA1c and body weight over 1

year in GV1 (12-month continuation DE) compared

with GV2.

Chronic pain

Ersek,

2003 [96]

N = 45

GV (7 SME sessions) Educational booklet on

pain

NA/13 Fair There was a significant improvement in physical role

functioning and in pain intensity directly following

treatment, but not 3 months after treatment, though no

significant effect was noted for other primary outcome

variables including physical functioning, activity

interference, and depression.

Vlaeyen,

1996 [95]

N = 131*

specify

GV (12 SME, EE, DE

sessions)

GV (12 DE, EE sessions) NR/20 Fair Significant improvement of knowledge, pain coping,

pain control, and relaxation for both GV groups

compared to control at immediate follow-up; non-

significant differences between GV groups at 6 and 12-

month follow-up on all primary outcomes.

Gustavsson,

2010 [97]

N = 156

GV (8 session, SME, EE) Individual physical

therapy

84/20 Good GV was positively associated with most assessed

outcomes including pain coping, pain control,

catastrophizing, pain scores, and anxiety, though

effects on depression were non-significant.

Note: Only total scores are shown in the table. The Comments include additional summary statements that are based on subscale findings, but not numerically represented

because there were only subscale scores reported.a Defined as percent eligible for enrollment among those invited to participate.b Defined as percent lost to follow-up among those randomized.


Acknowledgement

Contributors: A. Quinones planned the study, abstracted thedata, interpreted the findings, and drafted the manuscript. J.Richardson, M. Freeman, M. O’Neil, and D. Kansagara assisted withthe planning of the study, abstracting the data, and writing andreviewing the manuscript. R. Fu conducted the meta-analyses, andreviewed the manuscript. M. Motu’apuaka assisted with abstract-ing data, and revising the manuscript. All authors declare noconflicts of interest involved in the research presented in themanuscript. The corresponding author (AQ) declares she had fullaccess to all the data in the study and takes responsibility for the

integrity of the data and the accuracy of the data analysis. Wethank Rose Relevo for library assistance.

Funders: The research reported here was supported by theDepartment of Veterans Affairs, Veterans Health Administration,Health Services Research and Development Service Evidence-based Synthesis Program (ESP) project #05-225. Dr. Quinones is aHealth Science Specialist at the Portland VA Medical Center. Dr.Quinones is also supported by American Diabetes AssociationGrant 7-13-CD-08. The views expressed in this article are those ofthe authors and do not necessarily represent the views of theDepartment of Veterans Affairs nor the American DiabetesAssociation.


Prior presentations: The full-length report prepared for VAnominating stakeholders is available on the VA website: http://vaww.hsrd.research.va.gov/publications/esp/group-visits.cfm. Aposter presentation to the Northwest regional meeting of theSociety for General Internal Medicine was made in February, 2013in Portland, Oregon. A Cyber Seminar presentation for VA HealthServices Research & Development Service on April 1, 2013 isarchived online: http://www.hsrd.research.va.gov/for_research-ers/cyber_seminars/archives/video_archive.cfm?SessionID=676#.UYlNRLWG0qQ.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.pec.2013.12.021.

References

[1] Nutting PA, Miller WL, Crabtree BF, Jaen CR, Stewart EE, Stange KC. Initiallessons from the first national demonstration project on practice transfor-mation to a patient-centered medical home. Ann Fam Med 2009;7:254–60.

[2] Rittenhouse DR, Shortell SM. The patient-centered medical home: will itstand the test of health reform? J Amer Med Assoc 2009;301:2038–40.

[3] Abrams M, Nuzum R, Mika S, Lawlor G. Realizing health reform’s potential—how the affordable care act will stregthen primary care and benefit patients,providers, and payers. Commonw Fund 2011;1466:1–27.

[4] Han A, Judd M, Welch V, Wu T, Tugwell P, Wells GA. Tai chi for treatingrheumatoid arthritis. CDS Rev 2010;CD004849.

[5] Ashworth NL, Chad KE, Harrison EL, Reeder BA, Marshall SC. Home versuscenter based physical activity programs in older adults. CDS Rev2009;CD004017.

[6] Gillespie LD, Robertson MC, Gillespie WJ, Sherrington C, Gates S, Clemson LM,et al. Interventions for preventing falls in older people living in the commu-nity. CDS Rev 2012;9:CD007146.

[7] Anonymous. Effect of intensive blood-glucose control with metformin oncomplications in overweight patients with type 2 diabetes (UKPDS 34). UKProspective Diabetes Study (UKPDS) Group. Lancet 1998;352:854–65.

[8] Lorig K, Stewart A, Ritter P, Gonzalez V, Laurent D, Lynch J. Outcome measuresfor health education and other health care interventions. Thousand Oaks, CA:Sage Publications; 1996. p. 24–5, 41–5.

[9] Hibbard JH, Stockard J, Mahoney ER, Tusler M. Development of the patientactivation measure (PAM): conceptualizing and measuring activation inpatients and consumers. Health Serv Res 2004;39(August (4 Part I)):1005–26.

[10] Greene J, Hibbard JH. Why does patient activation matter? An examination ofthe relationships between patient activation and health-related outcomes. JGen Intern Med 2012;27:520–6.

[11] Harris RP, Helfand M, Woolf SH, Lohr KN, Mulrow CD, Teutsch SM, et al.Current methods of the US preventive services task force: a review of theprocess. Am J Prev Med 2001;30(3 Suppl.):21–35.

[12] Guyatt G, Oxman AD, Akl EA, Kunz R, Vist G, Brozek J, et al. GRADE guidelines:1. Introduction-GRADE evidence profiles and summary of findings tables. JClin Epidemiol 2011;64:383–94.

[13] DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials1986;7:177–88.

[14] Higgins JPT, Altman DG. Chapter 8: assessing risk of bias in included studies.In: Higgins JPT, Green S, editors. Cochrane handbook for systematic reviewsof interventions 2008; Version 5.0.1. Chichester, West Sussex, England: TheCochrane Collaboration and John Wiley & Sons Ltd; 2008 . Available from:www.cochrane-handbook.org.

[15] Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysisdetected by a simple, graphical test. Brit Med J 1997;315:629–34.

[16] Hammond A, Lincoln N, Sutcliffe L. A crossover trial evaluating an education-al-behavioural joint protection programme for people with rheumatoidarthritis. Patient Educ Couns 1999;37:19–32.

[17] Ackerman IN, Buchbinder R, Osborne RH. Challenges in evaluating an arthritisself-management program for people with hip and knee osteoarthritis inreal-world clinical settings. J Rheumatol 2012;39:1–9.

[18] Barlow JH, Turner AP, Wright CC. A randomized controlled study of thearthritis self-management programme in the UK. Health Educ Res2000;15:665–80.

[19] Breedland I, van Scheppingen C, Leijsma M, Verheij-Jansen NP, van Weert E.Effects of a group-based exercise and educational program on physicalperformance and disease self-management in rheumatoid arthritis: a ran-domized controlled study. Phys Ther 2011;91:879–93.

[20] Buszewicz M, Rait G, Griffin M, Nazareth I, Patel A, Atkinson A, et al. Selfmanagement of arthritis in primary care: randomised controlled trial. BritMed J 2006;333:879 [Clinical research ed.].

[21] Patel A, Buszewicz M, Beecham J, Griffin M, Rait G, Nazareth I, et al. Economicevaluation of arthritis self management in primary care. Brit Med J2009;339:b3532.

[22] Ettinger Jr WH, Burns R, Messier SP, Applegate W, Rejeski WJ, Morgan T, et al.A randomized trial comparing aerobic exercise and resistance exercise with ahealth education program in older adults with knee osteoarthritis. TheFitness Arthritis and Seniors Trial (FAST). J Amer Med Assoc 1997;277:25–31.

[23] Freeman K, Hammond A, Lincoln NB. Use of cognitive-behavioural arthritiseducation programmes in newly diagnosed rheumatoid arthritis. Clin Rehabil2002;16:828–36.

[24] Giraudet-Le Quintrec J, Mayoux-Benhamou A, Ravaud P, Champion K, DernisE, Zerkak D, et al. Effect of a collective educational program for patients withrheumatoid arthritis: a prospective 12-month randomized controlled trial. JRheumatol 2007;34:1684–91.

[25] Hewlett S, Ambler N, Almeida C, Cliss A, Hammond A, Kitchen K, et al. Self-management of fatigue in rheumatoid arthritis: a randomised controlled trialof group cognitive-behavioural therapy. Ann Rheum Dis 2011;70:1060–7.

[26] Kaplan S, Kozin F. A controlled study of group counseling in rheumatoidarthritis. J Rheumatol 1981;8:91–9.

[27] Lorig K, Lubeck D, Kraines RG, Seleznick M, Holman HR. Outcomes of self-helpeducation for patients with arthritis. Arthritis Rheum 1985;28:680–5.

[28] Lorig K, Gonzalez VM, Ritter P. Community-based Spanish language arthritiseducation program: a randomized trial. Med Care 1999;37:957–63.

[29] Lorig KR, Ritter PL, Laurent DD, Fries JF. Long-term randomized controlledtrials of tailored-print and small-group arthritis self-management interven-tions. Med Care 2004;42:346–54.

[30] Riemsma RP, Taal E, Rasker JJ. Group education for patients with rheumatoidarthritis and their partners. Arthritis Rheum 2003;49:556–66.

[31] Sevick MA, Miller GD, Loeser RF, Williamson JD, Messier SP. Cost-effective-ness of exercise and diet in overweight and obese adults with knee osteoar-thritis. Med Sci Sports Exerc 2009;41:1167–74.

[32] Taal E, Riemsma RP, Brus HL, Seydel ER, Rasker JJ, Wiegman O. Groupeducation for patients with rheumatoid arthritis. Patient Educ Couns1993;20:177–87.

[33] Arnold CM, Faulkner RA. The effect of aquatic exercise and education onlowering fall risk in older adults with hip osteoarthritis. J Aging Phys Act2010;18:245–60.

[34] Clemson L, Cumming RG, Kendig H, Swann M, Heard R, Taylor K. Theeffectiveness of a community-based program for reducing the incidence offalls in the elderly: a randomized trial. J Am Geriatr Soc 2004;52:1487–94.

[35] Ryan JW, Spellbring AM. Implementing strategies to decrease risk of falls inolder women. J Gerontol Nurs 1996;22:25–31.

[36] Shumway-Cook A, Silver HF, LeMier M, York S, Cummings P, Koepsell TD.Effectiveness of a community-based multifactorial intervention on falls andfall risk factors in community-living older adults: a randomized, controlledtrial. J Gerontol A Biol Sci Med Sci 2007;62:1420–7.

[37] Wilson SR, Scamagas P, German DF, Hughes GW, Lulla S, Coss S, et al. Acontrolled trial of two forms of self-management education for adults withasthma. Am J Med 1993;94:564–76.

[38] Abdulwadud O, Abramson M, Forbes A, James A, Walters EH. Evaluation of arandomised controlled trial of adult asthma education in a hospital setting.Thorax 1999;54:493–500.

[39] Allen RM, Jones MP, Oldenburg B. Randomised trial of an asthma self-management programme for adults. Thorax 1995;50:731–8.

[40] Bolton MB, Tilley BC, Kuder J, Reeves T, Schultz LR. The cost and effectivenessof an education program for adults who have asthma. J Gen Intern Med1991;6:401–7.

[41] Snyder SE, Winder JA, Creer TJ. Development and evaluation of an adultasthma self-management program: Wheezers Anonymous. J Asthma1987;24:153–8.

[42] Wilson JS, Fitzsimons D, Bradbury I, Stuart Elborn J. Does additional supportby nurses enhance the effect of a brief smoking cessation intervention inpeople with moderate to severe chronic obstructive pulmonary disease? Arandomised controlled trial. Int J Nurs Stud 2008;45:508–17.

[43] Bestall JC, Paul EA, Garrod R, Garnham R, Jones RW, Wedzicha AJ. Longitudinaltrends in exercise capacity and health status after pulmonary rehabilitationin patients with COPD. Respir Med 2003;97:173–80.

[44] Effing T, Zielhuis G, Kerstjens H, van der Valk P, van der Palen J. Communitybased physiotherapeutic exercise in COPD self-management: a randomisedcontrolled trial. Respir Med 2011;105:418–26.

[45] Ninot G, Moullec G, Picot MC, Jaussent A, Hayot M, Desplan M, et al. Cost-saving effect of supervised exercise associated to COPD self-managementeducation program. Respir Med 2011;105:377–85.

[46] Kunik ME, Veazey C, Cully JA, Souchek J, Graham DP, Hopko D, et al. COPDeducation and cognitive behavioral therapy group treatment for clinicallysignificant symptoms of depression and anxiety in COPD patients: a random-ized controlled trial. Psychol Med 2008;38:385–96.

[47] Moore SM, Charvat JM, Gordon NH, Pashkow F, Ribisl P, Roberts BL, et al.Effects of a change intervention to increase exercise maintenance followingcardiac events. Ann Behav Med 2006;31:53–62.

[48] Chang BH, Hendricks A, Zhao Y, Rothendler JA, LoCastro JS, Slawsky MT. Arelaxation response randomized trial on patients with chronic heart failure. JCardpulm Rehabil 2005;25:149–57.

[49] Smeulders ES, van Haastregt JC, Ambergen T, Uszko-Lencer NH, Janssen-Boyne JJ, Gorgels AP, et al. Nurse-led self-management group programme for

http://vaww.hsrd.research.va.gov/publications/esp/group-visits.cfm

http://vaww.hsrd.research.va.gov/publications/esp/group-visits.cfm

http://www.hsrd.research.va.gov/for_researchers/cyber_seminars/archives/video_archive.cfm?SessionID=676#.UYlNRLWG0qQ




http://refhub.elsevier.com/S0738-3991(14)00007-X/sbref0005






































http://www.cochrane-handbook.org/










































































































patients with congestive heart failure: randomized controlled trial. J AdvNurs 2010;66:1487–99.

[50] Smeulders ES, van Haastregt JC, Ambergen T, Stoffers HE, Janssen-Boyne JJ,Uszko-Lencer NH, et al. Heart failure patients with a lower educational leveland better cognitive status benefit most from a self-management groupprogramme. Patient Educ Couns 2010;81:214–21.

[51] Nessman DG, Carnahan JE, Nugent CA. Increasing compliance.Patient-operated hypertension groups. Arch Intern Med 1980;140:1427–30.

[52] Rujiwatthanakorn D, Panpakdee O, Malathum P, Tanomsup S. Effectiveness ofa self-management program for Thais with essential hypertension. PacificRim Int J Nurs Res 2011;15:97–109.

[53] Balcazar HG, Byrd TL, Ortiz M, Tondapu SR, Chavez M. A randomizedcommunity intervention to improve hypertension control among MexicanAmericans: using the promotoras de salud community outreach model. JHealth Care Poor Underserved 2009;20:1079–94.

[54] Figar S, Galarza C, Petrlik E, Hornstein L, Rodriguez Loria G, Waisman G, et al.Effect of education on blood pressure control in elderly persons: a random-ized controlled trial. Am J Hypertens 2006;19:737–43.

[55] Scala D, D’Avino M, Cozzolino S, Mancini A, Andria B, Caruso G, et al.Promotion of behavioural change in people with hypertension: an interven-tion study. Pharm World Sci 2008;30:834–9.

[56] Svetkey LP, Pollak KI, Yancy Jr WS, Dolor RJ, Batch BC, Samsa G, et al.Hypertension improvement project: randomized trial of quality improve-ment for physicians and lifestyle modification for patients. Hypertension2009;54:1226–33.

[57] Baghianimoghadam MH, Rahaee Z, Morowatisharifabad MA, Sharifirad G,Andishmand A, Azadbakht L. Effects of education on self-monitoring of bloodpressure based on BASNEF model in hypertensive patients. J Res Med Sci2010;15:70–7.

[58] Adolfsson ET, Walker-Engstrom ML, Smide B, Wikblad K. Patient education intype 2 diabetes: a randomized controlled 1-year follow-up study. DiabetesRes Clin Pract 2007;76:341–50.

[59] Anderson RM, Funnell MM, Nwankwo R, Gillard ML, Oh M, Fitzgerald JT.Evaluating a problem-based empowerment program for African Americanswith diabetes: results of a randomized controlled trial. Ethn Dis 2005;15:671–8.

[60] Brown SA, Garcia AA, Kouzekanani K, Hanis CL. Culturally competent diabetesself-management education for Mexican Americans: the Starr County borderhealth initiative. Diabetes Care 2002;25:259–68.

[61] Brown SA, Blozis SA, Kouzekanani K, Garcia AA, Winchell M, Hanis CL.Dosage effects of diabetes self-management education for Mexican Amer-icans: the Starr County Border Health Initiative. Diabetes Care 2005;28:527–32.

[62] De Greef K, Deforche B, Tudor-Locke C, De Bourdeaudhuij I. Increasingphysical activity in Belgian type 2 diabetes patients: a three-arm randomizedcontrolled trial. Int J Behav Med 2011;18:188–98.

[63] Deakin TA, Cade JE, Williams R, Greenwood DC. Structured patient education:the diabetes X-PERT Programme makes a difference. Diabet Med2006;23:944–54.

[64] Dejesus RS, Chaudhry R, Leutink DJ, Hinton MA, Cha SS, Stroebel RJ. Effects ofefforts to intensify management on blood pressure control among patientswith type 2 diabetes mellitus and hypertension: a pilot study. Vasc HealthRisk Manag 2009;5:705–11.

[65] Hornsten A, Stenlund H, Lundman B, Sandstrom H. Improvements in HbA1cremain after 5 years – a follow up of an educational intervention focusing onpatients’ personal understandings of type 2 diabetes. Diabetes Res Clin Pract2008;81:50–5.

[66] Khunti K, Gray LJ, Skinner T, Carey ME, Realf KHD, et al. Effectiveness of adiabetes education and self management programme (DESMOND) for peoplewith newly diagnosed type 2 diabetes mellitus: three year follow-up of acluster randomised controlled trial in primary care. Brit Med J2012;344:e2333.

[67] Davies MJ, Heller S, Skinner TC, Campbell MJ, Carey ME, Cradock S, et al.Effectiveness of the diabetes education and self management for ongoing andnewly diagnosed (DESMOND) programme for people with newly diagnosedtype 2 diabetes: cluster randomised controlled trial. Brit Med J2008;336:491–5.

[68] Kulzer B, Hermanns N, Reinecker H, Haak T. Effects of self-managementtraining in Type 2 diabetes: a randomized, prospective trial. Diabet Med2007;24:415–23.

[69] Lorig K, Ritter PL, Villa FJ, Armas J. Community-based peer-led diabetesself-management: a randomized trial. Diabetes Educ 2009;35:641–51.

[70] Lujan J, Ostwald SK, Ortiz M. Promotora diabetes intervention for MexicanAmericans. Diabetes Educ 2007;33:660–70.

[71] D’Eramo Melkus G, Chyun D, Vorderstrasse A, Newlin K, Jefferson V,Langerman S. The effect of a diabetes education, coping skills training,and care intervention on physiological and psychosocial outcomes in blackwomen with type 2 diabetes. Biol Res Nurs 2010;12:7–19.

[72] Miller CK, Edwards L, Kissling G, Sanville L. Nutrition education improvesmetabolic outcomes among older adults with diabetes mellitus: results froma randomized controlled trial. Prev Med 2002;34:252–9.

[73] Philis-Tsimikas A, Fortmann A, Lleva-Ocana L, Walker C, Gallo LC. Peer-leddiabetes education programs in high-risk Mexican Americans improve gly-cemic control compared with standard approaches: a Project Dulce promo-tora randomized trial. Diabetes Care 2011;34:1926–31.

[74] Raji A, Gomes H, Beard JO, MacDonald P, Conlin PR. A randomized trialcomparing intensive and passive education in patients with diabetes melli-tus. Arch Intern Med 2002;162:1301–4.

[75] Rickheim PL, Weaver TW, Flader JL, Kendall DM. Assessment of group versusindividual diabetes education: a randomized study. Diabetes Care2002;25:269–74.

[76] Rosal MC, Ockene IS, Restrepo A, White MJ, Borg A, Olendzki B, et al.Randomized trial of a literacy-sensitive, culturally tailored diabetes self-management intervention for low-income latinos: latinos en control. Diabe-tes Care 2011;34:838–44.

[77] Rygg LO, Rise MB, Gronning K, Steinsbekk A. Efficacy of ongoing group baseddiabetes self-management education for patients with type 2 diabetes mel-litus. A randomised controlled trial. Patient Educ Couns 2012;86:98–105.

[78] Sarkadi A, Rosenqvist U. Experience-based group education in Type 2 diabe-tes: a randomised controlled trial. Patient Educ Couns 2004;53:291–8.

[79] Scain SF, Friedman R, Gross JL. A structured educational program improvesmetabolic control in patients with type 2 diabetes: a randomized controlledtrial. Diabetes Educ 2009;35:603–11.

[80] Schillinger D, Handley M, Wang F, Hammer H. Effects of self-managementsupport on structure, process, and outcomes among vulnerable patients withdiabetes: a three-arm practical clinical trial. Diabetes Care 2009;32:559–66.

[81] Sharifirad G, Najimi A, Hassanzadeh A, Azadbakht L. Does nutrition educationimprove the risk factors of cardiovascular diseases among elderly with type 2diabetes? A randomized controlled trial based on an educational model. JDiabetes 2013;5:157–62.

[82] Sperl-Hillen J, Beaton S, Fernandes O, Von Worley A, Vazquez-Benitez G,Parker E, et al. Comparative effectiveness of patient education methods fortype 2 diabetes: a randomized controlled trial. Arch Intern Med2011;171:2001–10.

[83] Steed L, Lankester J, Barnard M, Earle K, Hurel S, Newman S. Evaluation of theUCL diabetes self-management programme (UCL-DSMP): a randomized con-trolled trial. J Health Psychol 2005;10:261–76.

[84] Surwit RS, van Tilburg MA, Zucker N, McCaskill CC, Parekh P, Feinglos MN,et al. Stress management improves long-term glycemic control in type 2diabetes. Diabetes Care 2002;25:30–4.

[85] Toobert DJ, Strycker LA, Barrera Jr M, Osuna D, King DK, Glasgow RE. Out-comes from a multiple risk factor diabetes self-management trial for Latinas:

!

Viva Bien! Ann Behav Med 2011;41:310–23.[86] Toobert DJ, Strycker LA, King DK, Barrera Jr M, Osuna D, Glasgow RE. Long-

term outcomes from a multiple-risk-factor diabetes trial for Latinas:

!

VivaBien! Transl Behav Med 2011;1:416–26.

[87] Weinger K, Beverly EA, Lee Y, Sitnokov L, Ganda OP, Caballero AE. The effect ofa structured behavioral intervention on poorly controlled diabetes: a ran-domized controlled trial. Arch Intern Med 2011;171:1990–9.

[88] Zapotoczky H, Semlitsch B, Herzog G, Bahadori B, Siebenhofer A, Pieber TR,et al. A controlled study of weight reduction in type 2 diabetics treated by tworeinforcers. Int J Behav Med 2001;8:42–9.

[89] Stanford School of Medicine. Chronic Disease Self-Management Program;2012, http://patienteducation.stanford.edu/programs/cdsmp.html [accessed20.09.12].

[90] Elzen H, Slaets JP, Snijders TA, Steverink N. Evaluation of the chronic diseaseself-management program (CDSMP) among chronically ill older people in theNetherlands. Soc Sci Med 2007;64:1832–41.

[91] Fu D, Fu H, McGowan P, Shen YE, Zhu L, Yang H, et al. Implementation andquantitative evaluation of chronic disease self-management programme inShanghai, China: randomized controlled trial. Bull World Health Organ2003;81:174–82.

[92] Lorig KR, Ritter PL, Gonzalez VM. Hispanic chronic disease self-management:a randomized community-based outcome trial. Nurs Res 2003;52:361–9.

[93] Lorig KR, Sobel DS, Stewart AL, Brown Jr BW, Bandura AP, Ritter PP, et al.Evidence suggesting that a chronic disease self-management program canimprove health status while reducing hospitalization: a randomized trial.Med Care 1999;37:5–14.

[94] Haugli L, Steen E, Laerum E, Nygard R, Finset A. Psychological distress andemployment status. Effects of a group learning programme for patients withchronic musculoskeletal pain. Psychol Health Med 2003;8:135–48.

[95] Vlaeyen JW, Teeken-Gruben NJ, Goossens ME, Rutten-van Molken MP,Pelt RA, van Eek H, et al. Cognitive-educational treatment of fibromyalgia:a randomized clinical trial. I. Clinical effects. J Rheumatol 1996;23:1237–45.

[96] Ersek M, Turner JA, McCurry SM, Gibbons L, Kraybill BM. Efficacy of a self-management group intervention for elderly persons with chronic pain. Clin JPain 2003;19:156–67.

[97] Gustavsson C, Denison E, von Koch L. Self-management of persistent neckpain: a randomized controlled trial of a multi-component group interventionin primary health care. Eur J Pain 2010;14:e1–1.

[98] Haugli L, Steen E, Laerum E, Finset A, Nygaard R. Agency orientation andchronic musculoskeletal pain: effects of a group learning program based onthe personal construct theory. Clin J Pain 2000;16:281–9.

[99] Hammond A, Bryan J, Hardy A. Effects of a modular behavioural arthritiseducation programme: a pragmatic parallel-group randomized controlledtrial. Rheumatology 2008;47:1712–8.

[100] Griffin SJ, Borch-Johnsen K, Davies MJ, Khunti K, Rutten GE, Sandbaek A, et al.Effect of early intensive multifactorial therapy on 5-year cardiovascularoutcomes in individuals with type 2 diabetes detected by screening (ADDI-TION-Europe): a cluster-randomised trial. Lancet 2011;378:156–67.









































































































































http://patienteducation.stanford.edu/programs/cdsmp.html







































[101] Kritikos V, Armour CL, Bosnic-Anticevich SZ. Interactive small-group asthmaeducation in the community pharmacy setting: a pilot study. J Asthma2007;44:57–64.

[102] Donesky-Cuenco D, Nguyen HQ, Paul S, Carrieri-Kohlman V. Yoga therapydecreases dyspnea-related distress and improves functional performance inpeople with chronic obstructive pulmonary disease: a pilot study. J AlternComplement Med 2009;15:225–34.

[103] Smeulders ES, van Haastregt JC, Ambergen T, Janssen-Boyne JJ, van Eijk JT,Kempen GI. The impact of a self-management group programme on healthbehaviour and healthcare utilization among congestive heart failure patients.Eur J Heart Fail 2009;11:609–16.

[104] Andryukhin A, Frolova E, Vaes B, Degryse J. The impact of a nurse-led careprogramme on events and physical and psychosocial parameters in patients

with heart failure with preserved ejection fraction: a randomized clinicaltrial in primary care in Russia. Eur J Gen Pract 2010;16:205–14.

[105] Burke V, Mansour J, Beilin LJ, Mori TA. Long-term follow-up of participants ina health promotion program for treated hypertensives (ADAPT). Nutr MetabCardiovasc Dis 2008;18:198–206.

[106] Pierce JP, Watson DS, Knights S, Gliddon T, Williams S, Watson R. A controlledtrial of health education in the physician’s office. Prev Med 1984;13:185–94.

[107] Arnold CM. Fall risk in older adults with hip osteoarthritis: decreasing riskthrough education and aquatic exercise. Canada: University of Saskatche-wan; 2008.

























educational group visits for the management of chronic health conditions: a systematic review

Documents