0
Original Investigation |

Comparative Effectiveness of Patient Education Methods for Type 2 Diabetes:  A Randomized Controlled Trial FREE

JoAnn Sperl-Hillen, MD; Sarah Beaton, PhD; Omar Fernandes, MPH; Ann Von Worley, RN, BSHS, CCRP; Gabriela Vazquez-Benitez, PhD, MSc; Emily Parker, MPH, PhD; Ann Hanson, BS; Jodi Lavin-Tompkins, RN, CNP, CDE, BC-ADM; Patricia Glasrud, MS, RD, CDE; Herbert Davis, PhD; Kenneth Adams, PhD; William Parsons, MS; C. Victor Spain, DVM, PhD
[+] Author Affiliations

Author Affiliations: HealthPartners Research Foundation and HealthPartners Medical Group, Minneapolis, Minnesota (Drs Sperl-Hillen, Vazquez-Benitez, Parker, and Adams, Mr Fernandes, and Ms Hanson); LCF Research, Albuquerque, New Mexico (Drs Beaton and Davis, Ms Von Worley, and Mr Parsons); HealthPartners Clinics, Minneapolis (Ms Lavin-Tompkins); ABQ Health Partners, Albuquerque (Ms Glasrud); and Merck and Co Inc, North Wales, Pennsylvania (Dr Spain).


Arch Intern Med. 2011;171(22):2001-2010. doi:10.1001/archinternmed.2011.507.
Text Size: A A A
Published online

Background Group education for patients with suboptimally controlled diabetes has not been rigorously studied.

Methods A total of 623 adults from Minnesota and New Mexico with type 2 diabetes and glycosylated hemoglobin (HbA1c) concentrations of 7% or higher were randomized to (1) group education (using the US Diabetes Conversation Map program), (2) individual education, or (3) usual care (UC; ie, no assigned education). Both education methods covered content as needed to meet national standards for diabetes self-management education and were delivered through accredited programs from 2008 to 2009. General linear mixed-model methods assessed patient-level changes between treatment groups in mean HbA1c levels from baseline to follow-up at 6.8 months. Secondary outcomes included mean change in general health status (Medical Outcomes Study 12-Item Short Form Health Survey [SF-12]), Problem Areas in Diabetes (PAID), Diabetes Self-Efficacy (DES-SF), Recommended Food Score (RFS), and Physical Activity (PA, min/wk).

Results Mean HbA1c concentration decreased in all groups but significantly more with individual (−0.51%) than group education (−0.27%) (P = .01) and UC (−0.24%) (P = .01). The proportion of subjects with follow-up HbA1c concentration lower than 7% was greater for individual education (21.2%) than for group (13.9%) and UC (12.8%) (P = .03). Compared with UC, individual education (but not group) improved SF-12 physical component score (+1.88) (P = .04), PA (+42.95 min/wk) (P = .03), and RFS (+0.63) (P = .05). Compared with group education, individual education reduced PAID (−3.62) (P = .02) and increased self-efficacy (+0.1) (P = .04).

Conclusions Individual education for patients with established suboptimally controlled diabetes resulted in better glucose control outcomes than did group education using Conversation Maps. There was also a trend toward better psychosocial and behavioral outcomes with individual education.

Trial Registration clinicaltrials.gov Identifier: NCT00652509

Figures in this Article

Diabetes self-management education (DSME) is a collaborative process through which people with diabetes gain the knowledge and skills needed to modify behavior and successfully self-manage the disease.1 However, few large well-designed randomized trials have been conducted to compare competing models for delivering this key element of diabetes care,2 and systematic reviews rate the quality of the DSME literature as moderate to poor owing to small participant numbers, inadequate allocation concealment, high dropout rates, and lack of intention-to-treat analysis.35

Despite limited high-quality evidence, DSME has been well accepted and widely endorsed nationally through American Diabetes Association6 (ADA) practice recommendations and public health initiatives. A Healthy People 20107 objective was to increase the proportion of people receiving formal diabetes education from the 1998 baseline of 45% to 60% by 2010. In addition, DSME provided through accredited individuals or entities is reimbursed by the Centers for Medicare and Medicaid Services (CMS), although follow-up training is limited to 2 hours annually per beneficiary. After the first year of diagnosis, diabetes patient education is usually provided on an individual as-needed basis.

By demonstrating that only 16% of patients with type 2 diabetes carry out self-management recommendations, the Diabetes Attitudes Wishes and Needs (DAWN) Study8 highlights the need for new models of patient education. One newer model for diabetes education endorsed by the ADA called the U.S. Diabetes Conversation Map program9 differs from traditional methods by using a highly interactive group approach with peer support to more actively engage patients with the information they learn and to help them make workable plans for achieving their personal health goals. The program is now used in more than 90 countries; however, to date, its effectiveness has not been rigorously evaluated.10 We conducted the present study, called the Journey for Control of Diabetes Interactive Dialogue to Educate and Activate (IDEA) study, to determine if group education (GE) using this approach improves glucose control and psychosocial and behavioral outcomes compared with usual care (UC, no assigned education) and with individual education (IE, the conventional approach) for patients with established type 2 diabetes and suboptimal control (HbA1c concentration, ≥7%).

STUDY SETTING AND PATIENT POPULATION

Adult members with type 2 diabetes at 2 large medical groups, ABQ Health Partners in New Mexico and HealthPartners Medical Group in Minnesota, were targeted for study enrollment. A total of 9971 people meeting the electronic health record (EHR)-assessed eligibility criteria of type 2 diabetes, the most recent HbA1c test result in the last 6 months of 7% concentration or higher, and no codes for GE in the last 2 years or IE in the last year, were mailed letters of invitation to participate in the study.11 A total of 623 patients (337 from Minnesota and 286 from New Mexico) consented at a baseline enrollment visit between June 2008 and May 2009 and were randomly assigned using a computer-generated random allocation sequence to GE, IE, or UC groups at each site. Figure 1 shows the CONSORT (Consolidated Standards of Reporting Trials) flow diagram. Participants were given gift cards worth $50 for completing the baseline visit and $25 for each of 4 follow-up surveys.

Place holder to copy figure label and caption
Graphic Jump Location

Figure 1. CONSORT (Consolidated Standards of Reporting Trials) flow diagram. HbA1c indicates hemoglobin A1c measurement. *Declined participation after learning more about the study; had visual, hearing, or cognitive impairment; was age 85 years or older; or was unable to read English. †Baseline results carried forward to follow-up period if data were missing.

TRIAL DESIGN

The study is a prospective multisite randomized controlled trial with parallel interventions for UC, IE, and GE. A total of 623 subjects provided their consent at 1 enrollment visit and were randomized and scheduled for educational sessions if assigned to IE or GE. This analysis includes results through the first planned follow-up evaluation period 6.8 months after enrollment (4 months after the last scheduled interventions). See Figure 2 for a timeline.

Place holder to copy figure label and caption
Graphic Jump Location

Figure 2. Timeline for preintervention, intervention, and follow-up evaluation periods. *Follow-up surveys were mailed 1, 4, 7, and 10 months after the patient’s last scheduled individual education or group education sessions. A corresponding date for mailing surveys to patients in the usual care group was calculated based on the pooled average of the intervention group mailing dates. The timeline shows the mean number of months after baseline and randomization that the surveys were mailed. This analysis includes results of the first follow-up period.

DATA SOURCES, MEASUREMENT, AND OUTCOME MEASURES

Data sources included measurements collected at enrollment (weight and blood pressure [BP]), passive surveillance of EHR observations, and responses to mailed surveys in the postintervention period. The clinical analysis used the latest value (eg, HbA1c) observed in the EHR in 1-year preenrollment and in the follow-up evaluation period. All HbA1c measurements were ordered in the course of usual patient care and conducted at the accredited laboratory associated with the care system with no changes made in HbA1c assay procedures during the study period.

Psychosocial and behavioral measurements were conducted using standardized surveys administered at the enrollment visit and mailed in follow-up to IE and GE subjects at 1 and 4 months after the last scheduled educational session, commensurate with a mean of 3.8 months and 6.8 months after enrollment. Subjects in the UC group were mailed surveys at 3.8 months and 6.8 months after enrollment. The variables measured in the surveys are described in Table 1.

Table Graphic Jump LocationTable 1. Survey Variable Descriptions
INTERVENTIONS

Interventions were delivered through the ADA-recognized education programs of the participant's care system from June 2008 through July 2009. The curricula used for both GE and IE methods were consistent with the American Association of Diabetes Educators Seven (AADE7) Self-Care Behaviors20 to improve knowledge and skills. The AADE7 content areas were healthy eating, monitoring, taking medications, problem solving, risk reduction, healthy coping, and being active. As needed to meet national standards for diabetes self-management education,21 both IE and GE methods strived to help patients set personalized action-oriented goals.

The IE consisted of the approach used conventionally (identical to the accredited education method for members not enrolled in the study) for follow-up diabetes education. Subjects were scheduled for 3 individual 1-hour sessions with the certified diabetes educators (nurses and dietitians) within approximately 1-month intervals. The GE consisted of four 2-hour sessions with groups scheduled at 1-week intervals. Patients were encouraged to bring a significant other (eg, spouse or support person) to the group, and the target group size was 8 to 10 people. The mean number of subjects who actually attended the GE sessions was 5 (range, 1-10). The sessions covered the recommended content areas using 4 different Conversation Maps, large, laminated 3  5-foot tabletop visuals with colorful images of situations familiar to people with diabetes, to facilitate group interaction.9 The mean time interval to complete the scheduled sessions was 77 days for GE and 90 days for IE.

Educators in both care systems received expert training on how to facilitate GE sessions through the creator of the Conversation Map program, Healthy Interactions Inc (Chicago, Illinois).9 An additional training session was conducted by the same expert about a month later to reinforce GE facilitator techniques. Educators completed self-assessments of each session, resulting in mean scores between 8 and 10 (scale of 1 to10, with 10 being the highest) on aspects of the group session, including overall success, coverage of all learning topics, patient participation, responsiveness to patient concerns, and comfort level in facilitating the group. Further details about the educator experience have been previously published.22

Engagement in concurrent education outside of the scheduled research sessions was not prohibited in any treatment group (including UC). Greater engagement in outside education was considered a positive secondary outcome of the IE and GE interventions.

ANALYTIC METHODS

All statistical analyses were conducted with SAS software, version 9.2 (SAS Institute Inc, Cary, North Carolina). Means and standard deviations for continuous variables and percentages for categorical variables were computed. The HbA1c outcome was analyzed in the logarithm scale and transformed back to its original scale. General mixed modeling was used to account for the nested structure as a result of observations over time, and efficacy was measured by the analysis of variance F test for the group  time interaction term. The model was adjusted for baseline HbA1c level, patient age, study site, and duration of diabetes. Within-intervention effects and pairwise comparisons between groups were performed, and multiple comparisons were controlled using a Bonferroni procedure. Intention-to-treat principles were used to test a priori hypotheses. Baseline values (no change) were assigned to subjects missing values post randomization. General linear mixed models were used for secondary normally distributed continuous variables (systolic and diastolic BP, weight, and scores from the following instruments: the Medical Outcomes Study 12-Item Short Form Health Survey [SF-12],14 Diabetes Care Profile [DCP],15 Problem Areas in Diabetes [PAID],16 Diabetes Empowerment Scale–Short Form [DES-SF],17 Recommended Food Score [RFS],18 and Behavioral Risk Factor Surveillance System physical activity [PA]19). Logistic regression was used for binary variables (proportion of subjects meeting HbA1c, BP, and self-monitored blood glucose [SMBG] goals at follow-up) adjusting for age, study site, and duration of diabetes in addition to the baseline outcome. Corrections for multiple comparisons were not conducted for secondary outcomes. Several post hoc sensitivity analyses were performed to examine the efficacy of interventions among patients who completed the interventions, among those with an HbA1c measurement in the follow-up period, and among those with HbA1c concentration of 8% or higher at baseline. An exploratory analysis was conducted to determine whether DES and PAID at 1-month post intervention mediated the intervention group HbA1c concentration effect at 4 months.

Minimum detectable standardized between-group HbA1c concentration effect sizes were determined for group comparisons on the basis of the sample size in each group, anticipated dropout rates, and the nesting of patients within educators, according to specifications that assumed an enrollment of 621 (UC, 133; IE, 244; GE, 244), an HbA1c testing rate of 85% to 95% at 6 months of follow-up, 15 to 25 patients per educator, and an 85% retention rate. We estimated power for the primary HbA1c outcome (power = 0.80; 2-tailed α = 0.05) to detect a 0.32% difference between IE and GE and a 0.36% difference between IE and UC and between GE and UC. Enrollment numbers met the prespecified sample size criteria.

Table 2 lists the baseline characteristics of IDEA study subjects. Mean age was 61.8 years, and mean duration of diabetes was 11.7 years; 49.4% of the patients were women; 65.2% were non-Hispanic whites; 22.1% were Hispanic; 5.5% were non-Hispanic blacks; and 7.3% were another race or ethnicity. A total of 22.1% had a high school education or less, and 15.6% had an annual income of less than $20 000. Randomization resulted in balanced groups at baseline, other than duration of diabetes (P = .04).

Table Graphic Jump LocationTable 2. Baseline Characteristics of IDEA Study Subjectsa

Table 3 and Table 4 list the baseline, follow-up, and mean change in HbA1c values. Hemoglobin A1c concentration was lower in the follow-up period in UC, IE, and GE by an absolute −0.24%, −0.51%, and −0.27%, respectively, with a significant group  time interaction (P = .01). Concentration of HbA1c fell significantly more with IE than with UC (−0.27%) (P = .01) and with GE (−0.24%) (P = .01). When restricted to those who completed the intervention, the mean change in HbA1c level at follow-up was similar and the group  time interaction remained significant (P = .01). When restricted to those with an HbA1c measurement in the follow-up period, group  time interactions were also similar (P = .02). The mean change in HbA1c level was greater when the analysis was restricted to those with a baseline HbA1c level of 8% or higher (UC, −0.64%; IE, −1.06%; and GE, −0.75%), although the sample size was smaller (n = 257), and results were not significant (P = .13).

Table Graphic Jump LocationTable 3. Glucose Control Outcomes Within Interventionsa
Table Graphic Jump LocationTable 4. Pairwise Group Comparisons of Change in Mean HbA1c at Follow-upa

Table 5 and Table 6 list the results of prespecified secondary outcome measurements. Compared with UC, both IE and GE improved DCP understanding scores by 0.28 (P < .001) and 0.19 (P = .02), respectively. There was no significant effect for other DCP scales such as support, attitudes, care ability, and importance of care. Individual education (but not GE) significantly improved nutrition (RFS +0.63) (P = .05) and PA (+42.95 min/wk) (P = .03) compared with UC. Compared with GE, IE reduced diabetes distress (−3.62) (P = .02) and increased self-efficacy (+0.1) (P = .04). We analyzed whether PAID and DES measured at 1-month post intervention mediated the intervention effect of HbA1c measured at 4 months. Neither variable explained the association between interventions and HbA1c effects. Intervention groups were associated with improvement of PAID and DES scores at 1 month post intervention (IE vs UC intervention effect, 0.16 [P = .01] for DES and −0.45 [P = .01] for PAID). In addition, changes in PAID and DES at 1 month were not associated with HbA1c level improvement at 4 months (β = −0.009, P = .34 for DES, and β = 0.0002, P = .64 for PAID, with HbA1c concentration in the log scale), resulting in nonsignificant indirect effects for DES (IE vs UC, −0.002) (P = .07) and for PAID (−0.0009) (P = .49).

Table Graphic Jump LocationTable 5. Within Intervention Effects of Prespecified Secondary Clinical, Psychosocial, and Behavioral Outcomesa
Table Graphic Jump LocationTable 6. Pairwise Group Comparisons of Change in Secondary Outcomes at Follow-upa

There were no significant effects of either method of education on BP outcomes or weight. Table 7 and Table 8 list the results of additional analyses. The IE group was significantly more likely than UC to have a follow-up HbA1c measurement of 7% or lower (odds ratio [OR], 1.83; 95% CI, 1.06-3.16) (P = .03). The GE group was less likely to have an HbA1c measurement below 7% at follow-up than the IE group (OR, 0.60; 95% CI, 0.39-.94) (P = .03). The likelihood of SMBG testing occurring 2 times a day or more trended higher for IE vs UC (OR, 1.63; 95% CI, 0.90-2.96) (P = .11) than for GE vs UC (OR, 0.67; 95% CI, 0.41-1.10) (P = .11).

Table Graphic Jump LocationTable 7. Mean Percentage of Patients at Goal and Odds Ratios at Follow-up
Table Graphic Jump LocationTable 8. Process Measures in the Follow-up Evaluation Period

Between-group differences were identified in the proportion of patients who received 1 or more educational visits outside of the study-assigned interventions (IE, 40.6%; GE, 25.3%; UC, 14.4%) (P < .001). Intervention nonadherence (the proportion who did not complete any education) was greater for GE (IE, 4.1%; GE, 12.4%) (P < .001), and the proportion who completed all education was greater for IE (IE, 86%; GE, 72%) (P < .001). There were no significant between-group differences in time to follow-up HbA1c measurement.

Patient satisfaction was evaluated by questionnaire after each education session. Evaluations were collected from 87% of GE and 93% of IE subjects. On a 100-point scale, the overall score for satisfaction with educational content was 91.5 for IE and 86.1 for GE (P < .001). Overall scores for satisfaction with the educator were 94.6 for IE and 90.3 for GE (P < .001).

Our results show that patients with type 2 diabetes of relatively long duration and HbA1c levels of 7% or higher had improved short-term HbA1c outcomes and a greater likelihood of achieving an HbA1c level below 7% if they were educated using the IE vs the GE method. Outcomes that improved significantly more with IE compared with UC (but not GE compared with UC) included HbA1c concentration, SF-12 physical health score, nutrition (RFS) score, and PA score. Based on glycemic control outcomes, trends of secondary psychosocial and behavioral outcomes, and patient satisfaction, we believe that these short-term results support the use of individual diabetes education for this patient population.

The lower completion rates for GE compared with IE sessions could be indicative of patient preference for IE and/or logistical difficulties associated with the GE approach (eg, less scheduling flexibility). Similar barriers to GE completion have been described by other educators, especially those in underserved settings.23,24 But even when results were restricted to completers of the intervention, the HbA1c results remained better with IE. Hemoglobin A1c outcomes also remained better for IE subjects when analysis was restricted to those with an HbA1c measurement taken in the follow-up period (82%). Missing HbA1c data were equally distributed across treatment groups and would not likely have changed the direction of results for the pairwise comparisons.

This study has implications for diabetes patient education guidelines, public policy, and CMS reimbursement for people with diabetes. The results support the current CMS reimbursement policy for IE for follow-up education. Total educator contact time was greater for the GE (8 hours) than the IE intervention (3 hours), and the results of this trial do not agree with previous statements that educator contact time, regardless of group or individual approach predicts HbA1c level improvement.3,25

A previous meta-analysis of group diabetes education suggests that the overall group approach has merit, and that group-based education improves diabetes control in both the short term and long term.4 Our results may differ because this study overcomes many of the quality concerns described for previous studies in its larger numbers, blinded allocation, low dropout rates, and intention-to-treat analysis. Other possible explanations for our lower GE effects include the type of group visits (using conversation maps), the long duration of diabetes in the study population, and lack of extensive educator experience facilitating Conversation Map sessions.22 More than 33 000 educators have been trained to facilitate the Conversation Map program internationally,9 and more research is needed to compare this approach with different GE delivery models and to evaluate it among specific demographic subgroups, in those with newly diagnosed diabetes, and in those with a clear preference for GE.

The generalizability of results is strengthened by several pragmatic design features of the study. First, few inclusion/exclusion criteria were applied; and second, interventions were tested in “real world” settings (conducted through the patient's affiliated care system). The study has high internal validity, and the observed similar HbA1c results at the 2 geographically disparate care systems with different ethnic mixes is evidence for potential external validity. However, replication of these results with other patient groups would be desirable.

A noted limitation is that the full HbA1c impact of the interventions may have been diminished because data published as the study was implemented26 led to changes in local and national guidelines toward a more individualized approach, with goal HbA1c below 8% rather than below 7% considered acceptable in many situations,6 potentially resulting in less effect than what was originally anticipated in those with baseline HbA1c concentrations in the 7.0% to 7.9% range (58.8% of the subject population). The effects of education were much greater in the group of subjects with HbA1c level of 8% or higher, and they remained more favorable for IE. Finally, the UC group in this study also showed a modest 0.24% HbA1c level reduction. The reason for the UC improvement is unclear, but it may have been owing to confounding diabetes improvement activities in the care systems,27 more medication use, selection effects of research subjects, or a greater than expected interventional effect of the mailed follow-up surveys.

Previous findings suggest that HbA1c effects observed in the short term with education may not be sustained,3 which would not be surprising if the outcomes are predicated on maintenance of behavior change. Additional long-term analysis is desirable and planned, but the short-term outcomes are nevertheless important to document. The present study also provides a rich data set of psychosocial and behavioral variables over time. We were unable to demonstrate that decreased diabetes distress (PAID) or greater self-efficacy (DES) were significant mediators of the observed improvement in HbA1c outcomes, suggesting other more complex mechanisms. Future long-term analysis (eg, using structural equation modeling) would be helpful to evaluate the pathways by which psychosocial characteristics and behavior change impact HbA1c and quality of life outcomes.

In conclusion, among patients with type 2 diabetes of relatively long duration and HbA1c levels of 7% or higher, short-term glucose control improved more in those receiving individual diabetes education than in those receiving group diabetes education or assigned to no education. Several important psychosocial and behavioral outcomes also trended better for individual education. Future analysis is needed to evaluate the effect of the educational interventions on drug use, adherence, cost-effectiveness, and long-term outcomes.

Correspondence: JoAnn Sperl-Hillen, MD, HealthPartners Research Foundation, 8170 33rd Ave S, MS 21111R, Minneapolis, MN 55440 (joann.m.sperlhillen@healthpartners.com).

Accepted for Publication: June 23, 2011.

Published Online: October 10, 2011. doi:10.1001/archinternmed.2011.507

Author Contributions: Drs Sperl-Hillen and Beaton had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Sperl-Hillen, Beaton, Parker, and Spain. Acquisition of data: Fernandes, Von Worley, Hanson, and Parsons. Analysis and interpretation of data: Sperl-Hillen, Beaton, Fernandes, Von Worley, Vazquez-Benitez, Parker, Hanson, Lavin-Tompkins, Glasrud, Davis, Adams, Parsons, and Spain. Drafting of the manuscript: Sperl-Hillen, Beaton, and Parker. Critical revision of the manuscript for important intellectual content: Sperl-Hillen, Beaton, Fernandes, Von Worley, Vazquez-Benitez, Parker, Hanson, Lavin-Tompkins, Glasrud, Davis, Adams, Parsons, and Spain. Statistical analysis: Sperl-Hillen, Beaton, Vazquez-Benitez, Parker, Hanson, Davis, Adams, and Parsons. Obtained funding: Sperl-Hillen, Beaton, and Spain. Administrative, technical, and material support: Fernandes, Von Worley, Hanson, Lavin-Tompkins, Glasrud, and Spain. Study supervision: Sperl-Hillen and Beaton.

Financial Disclosure: Dr Sperl-Hillen has received research support administered through HealthPartners Research Foundation for clinical trials from Merck and Co Inc, GlaxoSmithKline, Lilly, and Merck-ScheringPlough Pharmaceuticals. Drs Beaton, Davis, and Parsons and Mss Von Worley and Glasrud have received research support administered through a subcontract between LCF Research and HealthPartners Research Foundation (as a result of the primary contract between HealthPartners Research Foundation and Merck and Co Inc). Mr Fernandes, Drs Vazquez-Benitez, Parker, and Adams, and Mss Hanson and Lavin-Tompkins have received research support administered through HealthPartners Research Foundation from Merck and Co Inc. Dr Spain is a full-time employee of and owns stock in Merck and Co Inc.

Funding/Support: This study was funded by Merck and Co Inc, North Wales, Pennsylvania.

Role of the Sponsors: The project has been observed from the beginning by project managers from Merck and Co Inc US Outcomes Research Division. The current project manager is C. Victor Spain, DVM, PhD, an author of this article. Dr Spain has been involved through phone meetings twice a month in the management, analysis, and interpretation of the data; and preparation, review, and approval of the manuscript. He has not been involved in the design and conduct of the study or in data collection.

Additional Contributions: We gratefully acknowledge the assistance and support of the ABQ Health Partners diabetes education staff in New Mexico and HealthPartners Diabetes Program in Minnesota for providing the study interventions to participants as well as participating in the Conversation Map training. We also thank Barbara Eichorst, MS, RD, CDE, for her help in providing training to the diabetes educators at both study sites. We thank Dorothy Baumer, MS, Maureen Busch, Colleen King, Dave Butani, Shelly Eudy, Jaime Sekenski, BS, and Louise Hillen, BA, for their contributions to programming, patient recruitment, and survey completion. In addition, we thank Jeremy Gleeson, MD, Patrick O’Connor, MD, MPH, for their valuable consultative roles and support of the study. Finally, we thank Mary VanBeusekom, BA, for her meticulous editing skills.

Martin C, Daly A, McWhorter LS, Shwide-Slavin C, Kushion W.American Association of Diabetes Educators.  The scope of practice, standards of practice, and standards of professional performance for diabetes educators.  Diabetes Educ. 2005;31(4):487-488
PubMed   |  Link to Article
Rickheim PL, Weaver TW, Flader JL, Kendall DM. Assessment of group versus individual diabetes education: a randomized study.  Diabetes Care. 2002;25(2):269-274
PubMed   |  Link to Article
Norris SL, Lau J, Smith SJ, Schmid CH, Engelgau MM. Self-management education for adults with type 2 diabetes: a meta-analysis of the effect on glycemic control.  Diabetes Care. 2002;25(7):1159-1171
PubMed   |  Link to Article
Deakin T, McShane CE, Cade JE, Williams RD. Group based training for self-management strategies in people with type 2 diabetes mellitus.  Cochrane Database Syst Rev. 2005;(2):CD003417
PubMed  |  Link to Article
Duke S-AS, Colagiuri S, Colagiuri R. Individual patient education for people with type 2 diabetes mellitus.  Cochrane Database Syst Rev. 2009;(1):CD005268
PubMed  |  Link to Article
American Diabetes Association.  Standards of medical care in diabetes—2010.  Diabetes Care. 2010;33:(suppl 1)  S11-S61
PubMed   |  Link to Article
Centers for Disease Control and Prevention.  Healthy People 2010: progress review focus area 5—diabetes. http://www.cdc.gov/nchs/healthy_people/hp2010/focus_areas/fa05_diabetes.htm. Accessed November 16, 2010
Funnell MM. The Diabetes Attitudes, Wishes, and Needs (DAWN) Study.  Clin Diabetes. 2006;24:154-155
Link to Article
Healthy Interactions.  The U.S. Diabetes Conversation Map Program. http://healthyinteractions.com/conversation-map-programs/conversation-map-experience/current-programs/usdiabetes. Accessed November 17, 2010
Belton AB. Conversation Maps in Canada: the first 2 years.  Diabetes Spectrum. 2008;21(1):139-142
Link to Article
Beaton SJ, Sperl-Hillen JM, Worley AV,  et al.  A comparative analysis of recruitment methods used in a randomized trial of diabetes education interventions.  Contemp Clin Trials. 2010;31(6):549-557
PubMed   |  Link to Article
Babor TF, Higgins-Biddle JC, Dauser D, Burleson JA, Zarkin GA, Bray J. Brief interventions for at-risk drinking: patient outcomes and cost-effectiveness in managed care organizations.  Alcohol Alcohol. 2006;41(6):624-631
PubMed
Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure.  J Gen Intern Med. 2001;16(9):606-613
PubMed   |  Link to Article
Ware JE, Kosinski M, Keller SD. SF-12: How to Score the SF-12 Physical and Mental Health Summary Scales. 2nd ed. Boston, MA: Health Institute, New England Medical Center; 1995
Fitzgerald JT, Davis WK, Connell CM, Hess GE, Funnell MM, Hiss RG. Development and validation of the Diabetes Care Profile.  Eval Health Prof. 1996;19(2):208-230
PubMed   |  Link to Article
Welch GW, Jacobson AM, Polonsky WH. The Problem Areas in Diabetes Scale: an evaluation of its clinical utility.  Diabetes Care. 1997;20(5):760-766
PubMed   |  Link to Article
Anderson RM, Fitzgerald JT, Gruppen LD, Funnell MM, Oh MS. The Diabetes Empowerment Scale-Short Form (DES-SF).  Diabetes Care. 2003;26(5):1641-1642
PubMed   |  Link to Article
Kant AK, Schatzkin A, Graubard BI, Schairer C. A prospective study of diet quality and mortality in women.  JAMA. 2000;283(16):2109-2115
PubMed   |  Link to Article
Centers for Disease Control and Prevention, Office of Surveillance, Epidemiology, and Laboratory Services; Behavioral Risk Factor Surveillance System.  2007 Behavioral Risk Factor Surveillance System questionnaire. http://www.cdc.gov/brfss/questionnaires/english.htm. Accessed November 16, 2010
American Association of Diabetes Educators.  AADE7 Self-Care Behaviors. http://www.diabeteseducator.org/ProfessionalResources/AADE7/. Accessed June 16, 2011
Funnell MM, Brown TL, Childs BP,  et al.  National standards for diabetes self-management education.  Diabetes Care. 2008;31(1):(suppl 1)  S97-S104
PubMed   |  Link to Article
Fernandes O, Von Worley A, Sperl-Hillen J,  et al.  Educator experience with the U.S. Diabetes Conversation Map education program in the journey for control of diabetes: The IDEA Study.  Diabetes Spectrum. 2010;23(3):194-198
Link to Article
Kahn LS, Glaser K, Fox CH, Patterson A. Diabetes educators in safety-net practices: a qualitative study.  Diabetes Educ. 2011;37(2):212-219
PubMed   |  Link to Article
Glasgow RE, Eakin EG, Toobert DJ. How generalizable are the results of diabetes self-management research? the impact of participation and attrition.  Diabetes Educ. 1996;22(6):573-585
PubMed   |  Link to Article
Mensing C, Eichorst B. Educating the patient with diabetes. In: Holt RG, Cockram C, Flyvbjerg A, eds, et al. Textbook of Diabetes. 4th ed. Oxford, England: Wiley-Blackwell; 2010:334-345
Gerstein HC, Miller ME, Byington RP,  et al; Action to Control Cardiovascular Risk in Diabetes Study Group.  Effects of intensive glucose lowering in type 2 diabetes.  N Engl J Med. 2008;358(24):2545-2559
PubMed   |  Link to Article
Sperl-Hillen JM, O’Connor PJ. Factors driving diabetes care improvement in a large medical group: ten years of progress.  Am J Manag Care. 2005;11(5):(suppl)  S177-S185
PubMed

Figures

Place holder to copy figure label and caption
Graphic Jump Location

Figure 1. CONSORT (Consolidated Standards of Reporting Trials) flow diagram. HbA1c indicates hemoglobin A1c measurement. *Declined participation after learning more about the study; had visual, hearing, or cognitive impairment; was age 85 years or older; or was unable to read English. †Baseline results carried forward to follow-up period if data were missing.

Place holder to copy figure label and caption
Graphic Jump Location

Figure 2. Timeline for preintervention, intervention, and follow-up evaluation periods. *Follow-up surveys were mailed 1, 4, 7, and 10 months after the patient’s last scheduled individual education or group education sessions. A corresponding date for mailing surveys to patients in the usual care group was calculated based on the pooled average of the intervention group mailing dates. The timeline shows the mean number of months after baseline and randomization that the surveys were mailed. This analysis includes results of the first follow-up period.

Tables

Table Graphic Jump LocationTable 1. Survey Variable Descriptions
Table Graphic Jump LocationTable 2. Baseline Characteristics of IDEA Study Subjectsa
Table Graphic Jump LocationTable 3. Glucose Control Outcomes Within Interventionsa
Table Graphic Jump LocationTable 4. Pairwise Group Comparisons of Change in Mean HbA1c at Follow-upa
Table Graphic Jump LocationTable 5. Within Intervention Effects of Prespecified Secondary Clinical, Psychosocial, and Behavioral Outcomesa
Table Graphic Jump LocationTable 6. Pairwise Group Comparisons of Change in Secondary Outcomes at Follow-upa
Table Graphic Jump LocationTable 7. Mean Percentage of Patients at Goal and Odds Ratios at Follow-up
Table Graphic Jump LocationTable 8. Process Measures in the Follow-up Evaluation Period

References

Martin C, Daly A, McWhorter LS, Shwide-Slavin C, Kushion W.American Association of Diabetes Educators.  The scope of practice, standards of practice, and standards of professional performance for diabetes educators.  Diabetes Educ. 2005;31(4):487-488
PubMed   |  Link to Article
Rickheim PL, Weaver TW, Flader JL, Kendall DM. Assessment of group versus individual diabetes education: a randomized study.  Diabetes Care. 2002;25(2):269-274
PubMed   |  Link to Article
Norris SL, Lau J, Smith SJ, Schmid CH, Engelgau MM. Self-management education for adults with type 2 diabetes: a meta-analysis of the effect on glycemic control.  Diabetes Care. 2002;25(7):1159-1171
PubMed   |  Link to Article
Deakin T, McShane CE, Cade JE, Williams RD. Group based training for self-management strategies in people with type 2 diabetes mellitus.  Cochrane Database Syst Rev. 2005;(2):CD003417
PubMed  |  Link to Article
Duke S-AS, Colagiuri S, Colagiuri R. Individual patient education for people with type 2 diabetes mellitus.  Cochrane Database Syst Rev. 2009;(1):CD005268
PubMed  |  Link to Article
American Diabetes Association.  Standards of medical care in diabetes—2010.  Diabetes Care. 2010;33:(suppl 1)  S11-S61
PubMed   |  Link to Article
Centers for Disease Control and Prevention.  Healthy People 2010: progress review focus area 5—diabetes. http://www.cdc.gov/nchs/healthy_people/hp2010/focus_areas/fa05_diabetes.htm. Accessed November 16, 2010
Funnell MM. The Diabetes Attitudes, Wishes, and Needs (DAWN) Study.  Clin Diabetes. 2006;24:154-155
Link to Article
Healthy Interactions.  The U.S. Diabetes Conversation Map Program. http://healthyinteractions.com/conversation-map-programs/conversation-map-experience/current-programs/usdiabetes. Accessed November 17, 2010
Belton AB. Conversation Maps in Canada: the first 2 years.  Diabetes Spectrum. 2008;21(1):139-142
Link to Article
Beaton SJ, Sperl-Hillen JM, Worley AV,  et al.  A comparative analysis of recruitment methods used in a randomized trial of diabetes education interventions.  Contemp Clin Trials. 2010;31(6):549-557
PubMed   |  Link to Article
Babor TF, Higgins-Biddle JC, Dauser D, Burleson JA, Zarkin GA, Bray J. Brief interventions for at-risk drinking: patient outcomes and cost-effectiveness in managed care organizations.  Alcohol Alcohol. 2006;41(6):624-631
PubMed
Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure.  J Gen Intern Med. 2001;16(9):606-613
PubMed   |  Link to Article
Ware JE, Kosinski M, Keller SD. SF-12: How to Score the SF-12 Physical and Mental Health Summary Scales. 2nd ed. Boston, MA: Health Institute, New England Medical Center; 1995
Fitzgerald JT, Davis WK, Connell CM, Hess GE, Funnell MM, Hiss RG. Development and validation of the Diabetes Care Profile.  Eval Health Prof. 1996;19(2):208-230
PubMed   |  Link to Article
Welch GW, Jacobson AM, Polonsky WH. The Problem Areas in Diabetes Scale: an evaluation of its clinical utility.  Diabetes Care. 1997;20(5):760-766
PubMed   |  Link to Article
Anderson RM, Fitzgerald JT, Gruppen LD, Funnell MM, Oh MS. The Diabetes Empowerment Scale-Short Form (DES-SF).  Diabetes Care. 2003;26(5):1641-1642
PubMed   |  Link to Article
Kant AK, Schatzkin A, Graubard BI, Schairer C. A prospective study of diet quality and mortality in women.  JAMA. 2000;283(16):2109-2115
PubMed   |  Link to Article
Centers for Disease Control and Prevention, Office of Surveillance, Epidemiology, and Laboratory Services; Behavioral Risk Factor Surveillance System.  2007 Behavioral Risk Factor Surveillance System questionnaire. http://www.cdc.gov/brfss/questionnaires/english.htm. Accessed November 16, 2010
American Association of Diabetes Educators.  AADE7 Self-Care Behaviors. http://www.diabeteseducator.org/ProfessionalResources/AADE7/. Accessed June 16, 2011
Funnell MM, Brown TL, Childs BP,  et al.  National standards for diabetes self-management education.  Diabetes Care. 2008;31(1):(suppl 1)  S97-S104
PubMed   |  Link to Article
Fernandes O, Von Worley A, Sperl-Hillen J,  et al.  Educator experience with the U.S. Diabetes Conversation Map education program in the journey for control of diabetes: The IDEA Study.  Diabetes Spectrum. 2010;23(3):194-198
Link to Article
Kahn LS, Glaser K, Fox CH, Patterson A. Diabetes educators in safety-net practices: a qualitative study.  Diabetes Educ. 2011;37(2):212-219
PubMed   |  Link to Article
Glasgow RE, Eakin EG, Toobert DJ. How generalizable are the results of diabetes self-management research? the impact of participation and attrition.  Diabetes Educ. 1996;22(6):573-585
PubMed   |  Link to Article
Mensing C, Eichorst B. Educating the patient with diabetes. In: Holt RG, Cockram C, Flyvbjerg A, eds, et al. Textbook of Diabetes. 4th ed. Oxford, England: Wiley-Blackwell; 2010:334-345
Gerstein HC, Miller ME, Byington RP,  et al; Action to Control Cardiovascular Risk in Diabetes Study Group.  Effects of intensive glucose lowering in type 2 diabetes.  N Engl J Med. 2008;358(24):2545-2559
PubMed   |  Link to Article
Sperl-Hillen JM, O’Connor PJ. Factors driving diabetes care improvement in a large medical group: ten years of progress.  Am J Manag Care. 2005;11(5):(suppl)  S177-S185
PubMed

Correspondence

CME
Meets CME requirements for:
Browse CME for all U.S. States
Accreditation Information
The American Medical Association is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The AMA designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM per course. Physicians should claim only the credit commensurate with the extent of their participation in the activity. Physicians who complete the CME course and score at least 80% correct on the quiz are eligible for AMA PRA Category 1 CreditTM.
Note: You must get at least of the answers correct to pass this quiz.
You have not filled in all the answers to complete this quiz
The following questions were not answered:
Sorry, you have unsuccessfully completed this CME quiz with a score of
The following questions were not answered correctly:
Commitment to Change (optional):
Indicate what change(s) you will implement in your practice, if any, based on this CME course.
Your quiz results:
The filled radio buttons indicate your responses. The preferred responses are highlighted
For CME Course: A Proposed Model for Initial Assessment and Management of Acute Heart Failure Syndromes
Indicate what changes(s) you will implement in your practice, if any, based on this CME course.
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).
Submit a Comment
Physician Practice -- Two Doctors
Posted on October 16, 2011
Richard Weiss, MD, ABIM
PRIVATE PRACTICE
Conflict of Interest: None Declared
We have a busy office practice (20 patients/day). Overhead costs prevent us from employing health coaches, nurse practioners, RNs, and other medical professionals often used for "ideal practices, medical homes." So comparative studies that require more resources are not generalizable to our practice. A crucial deficit in the randomized trial are questions related to how much time it takes for one doctor to intervene for designs of comparative, prospective, and randomized trials.
Why is there no mention of the time it takes physicians with small office practices to perform education, HIT, coordination and integration of consultants, procedures, and also talk with families in the office or phone?
These questions that are salient are never answered in articles or in the letters to editors that I write.

Conflict of Interest: None declared
Submit a Comment

Multimedia

Some tools below are only available to our subscribers or users with an online account.

Web of Science® Times Cited: 11

Related Content

Customize your page view by dragging & repositioning the boxes below.

See Also...
Articles Related By Topic
Related Topics
PubMed Articles
JAMAevidence.com