1. Rebecca Pratt Gregory, MS, RD, CDE and Kerri Cavanaugh, MD, MHS Addressing The Role of Nutrition Education and Health Literacy in Diabetes Care

2. 2 Learning Objectives Review of impact of CDEs and medical nutrition therapy (MNT) on diabetes outcomes Review health literacy & numeracy in diabetes self-care Describe design of new randomized controlled trial to: –Examine the value of CDE in diabetes care –Examine the role of different approaches to MNT Learn about study results, interpretation and applications to clinical practice

3. 3 Role of CDEs in Medical Nutrition Therapy in Diabetes Care Diabetes self-management education (DSME) can improve patient knowledge, behavior, and glycemic control –Medical Nutrition Therapy is an essential part of DSME Addressing carbohydrates as a nutritional strategy is endorsed by the ADA and the AADE –Clinical trials have shown that MNT can improve A1C by 1-2% –No studies have compared carbohydrate counting and plate method Facilitating positive behavior as well as transferring knowledge is a priority, but little research in this area exists

4. 4 Components of Literacy Literacy Cultural and Conceptual Knowledge Speaking Reading NumeracyListening Writing Print Literacy Oral Literacy IOM, Health Literacy, 2004 Defining Health Literacy

5. 5 Health literacy is associated with outcomes Diabetes Schillinger, JAMA, 2002

6. 6 Components of Literacy Literacy Cultural and Conceptual Knowledge Speaking Reading NumeracyListening Writing Print Literacy Oral Literacy IOM, Health Literacy, 2004

7. 7 Definition of Numeracy The ability to use numbers in daily life. [Rothman RL et al. AJPM 2006] Examples of numeracy skills –Calculations –Interpretation of graphs/labels –Time –Probability –Ability to deduce when and what math is needed for a given situation.

8. 8 Numeracy in diabetes care Glucose monitoring Carbohydrate counting Sliding/correction scale Insulin Calculating insulin:carbohydrate ratios Insulin pump adjustment Sick day management

9. 9 Measurement of Diabetes Numeracy Diabetes Numeracy Test (DNT) http://www.mc.vanderbilt.edu/diabetes/drtc/preventionandcontrol/tools.php Experts 43-items Diabetes and Numeracy Domains No time limit Calculators could be used Kuder-Richardson-20 coefficient=0.95 Diabetes Care Domains Nutrition Exercise Blood Glucose Monitoring Oral Medication Use Insulin Use Numeracy Domains Addition Subtraction Multiplication Division Fractions/Decimals Multi-step mathematics Time Numeration/Counting/Hierarchy Huizinga MM, et al. BMC Health Services Research 2008: 8;96

10. 10 Diabetes Numeracy & A1C Characteristic  A1c 95% Confidence Intervalp-value DNT Score (per 10%) -0.09[-0.16 to -0.01] 0.03 Age-0.17[-0.24 to -0.10]<0.001 Sex 0.09[-0.22 to 0.40] 0.59 Race 0.17[-0.17 to 0.52] 0.34 Years of Diabetes 0.04[ 0.02 to 0.06]<0.001 * Also adjusted for income, type of diabetes, and clinic Adjusted GLS regression model Cavanaugh K, et al. Ann Intern Med 2008; 148: 737-746

11. Diabetes Literacy & Numeracy Education Toolkit (DLNET): A RCT

12. 12 Diabetes Literacy & Numeracy Education Toolkit (DLNET)RCT Objective –Evaluate a literacy and numeracy-focused diabetes self-management education intervention on patient self-efficacy, satisfaction and glycemic control Design –Randomized controlled trial Setting –Enhanced diabetes education programs Intervention

13. 13 DLNET Intervention Control (Enhanced Care) Intervention (Enhanced Care Plus Literacy/Numeracy) Nurse practitioner/CDE visits (1-3) Dietitian/CDE visits (1-3) Usual diabetes patient education materials Diabetes Literacy & Numeracy Education Toolkit (DLNET) Clear health communication training Usual care from primary care or endocrine physician

14. 14 DLNET Toolkit Wolff K et al. The Diab Educ 2009 Goals Facilitate diabetes education and self-management Type 1 or Type 2 diabetes mellitus Oral medications or insulin Individual modules to customize for each patient Blood Glucose Monitoring Exercise planning Foot care Nutritional management Carbohydrates Medications Logbooks/worksheets Available at: www.mc.vanderbilt.edu/diabetes/drtc/preventionandcontrol/tools.php

15. 15 DLNET Toolkit Wolff K et al. The Diab Educ 2009 Text at 5 th grade reading level Color coding Pictures for key concepts Step-by-step instructions Simplified medication instructions Practice skills worksheets

16. 16 DLNET Study Results Cavanaugh KL et al. Diabetes Care 2009 Mean [95% bootstrap Confidence Interval] *Adjusting for age, gender, race, type of diabetes, income level, site of intervention and baseline DNT score and Hba1c levels A1c 3-monthsAdjusted p-value [Intervention vs. Control]* 6-monthsAdjusted p- value [Intervention vs. Control]* Intervention-1.63 [ -2.03, -1.23]0.03-1.11 [ -1.54,-0.65]0.437 Control-0.97 [-1.37, -0.53 ]-1.17 [-1.61,-0.71]

17. Diabetes Nutrition Education Study (DINES): A Randomized Controlled Trial

18. 18 Diabetes Nutrition Study (DINES) Objectives –To perform a randomized controlled trial to determine the efficacy of RD CDE medical nutrition therapy compared to usual care in the treatment of type 2 diabetes mellitus –To evaluate differences in glycemic control by MNT strategy (carbohydrate counting vs. modified plate method)

19. 19 Diabetes Nutrition Study (DINES) Methods Design: Randomized controlled trial Setting: –Vanderbilt University Medical Center –Regional primary care clinics (middle TN) Participants: Inclusion criteriaExclusion criteria Adults, Type 2 DMUsing flexible insulin A1c > 7%Poor vision No MNT past yearCognitive impairment English-speakingTerminal illness

20. 20 Diabetes Nutrition Study (DINES) Nutrition Education Intervention

21. Control Group 2-3 patient encounters Covered general non- nutrition topics: –Foot care –Fall prevention –Immunizations –Osteporosis –Diabetic Retinopathy –Oral care 21

22. 22 Diabetes Nutrition Study (DINES) Nutrition Education Intervention materials

23. 23

24. Modified Plate Method Number of carb portions defined Based on glucose response to meals Higher carb foods listed with amounts per carb serving 24

25. 25

26. Methods: Measures 26 Primary outcome –Hemoglobin A1C (%) at 3- and 6-Months Secondary outcomes –Perceived Self-efficacy of Diabetes Self-management Scale (PDSMS) –Summary of Diabetes Self-Care Activities Measure (SDSCA) –Diabetes Treatment Satisfaction Questionnaire (DTSQ) Potential Confounding variables –Patient characteristics/demographics Diabetes-related numeracy (DNT) Health literacy (Rapid Estimate of Adult Literacy in Medicine-REALM) Statistical Analyses –Wilcoxon rank-sum or Kruskal-Wallis test, as appropriate –Adjusted analyses: Linear regression modeling with Huber-White robust covariance matrix estimate for repeated measurements –Pre-specified subgroup analysis: Baseline A1C 7-10%

27. 27 Diabetes Nutrition Study (DINES) Referred 293 Enrolled: 150 Refused: 80 Excluded/Not eligible: 63 Carb Counting: 50 Control: 50 Withdrew/Dropped: 8Withdrew/Dropped: 4 Plate Method: 50 6M: 42 (84%) 6M: 45 (90%) Withdrew/Dropped: 5 6M: 46 (92%)

28. 28 Diabetes Nutrition Study (DINES) Participant Characteristics Characteristic Carb Counting n=50 Plate Method n=50 Control n=50 Age (yrs)54 (47, 68)55 (45, 60)57 (48, 62) % Male38%54%48% % White58%73%66% Education (yrs)14 (13, 16)14 (12, 16)14 (13, 16) % Income <$20k/yr22%26%22% % Smoke14%22%10% Diabetes Duration (yrs)8 (4, 10)7 (3, 10)8 (3, 13) % Insulin28%42%34% % Prior DM education52%54%68% Body mass index (kg/m 2 )34 (30, 37)34 (30, 39) % Literacy <9 th grade8%14%10% DNT Score (0-100%)73 (40, 93)67 (40, 85)67 (47, 80) Median (Interquartile Range)

29. 29 Diabetes Nutrition Study (DINES) Participant Characteristics Baseline Characteristic Carb Counting n=50 Plate Method n=50 Control n=50 Hemoglobin A1C (%)8.4 (7.6, 9.7)8.3 (7.5, 10.4)8.0 (7.5, 9.7) Weight (lbs)218 (190, 252)224 (189, 259)216 (180, 243) Self-efficacy: PDSMS [8-40]24 (18, 29)24 (21, 27)24 (21, 29) Self-management: SDSCA [0-7] General Diet3.5 (1.5, 5)4.0 (2.5, 5)3.5 (2.5, 5.5) Specific Diet3.5 (2, 4.5)3.0 (2, 4.4)3.2 (2, 4.5) Exercise1.5 (0.1, 3.4)2.5 (0.5, 3.9)2.0 (0.5, 3.9) Blood glucose monitoring3.5 (1.1, 5.9)6.0 (2.1, 7)5.0 (0.6, 7) Foot care3.5 (1.5, 5.9)3.5 (1.1, 5.9)3.5 (2, 6.5) Medications7.0 (5.2, 7)7.0 (7, 7) Satisfaction: DTSQ [0-48]24 (17, 30)22 (17, 29)26 (19, 31) Median (Interquartile Range)

30. 30 Diabetes Nutrition Study (DINES) A1C at 3-months & 6-months by study group VariableGroupBaselineBaseline to 3 months Baseline to 6 months A1C (%) ∆ A1C Carb Count 8.4 (7.6, 9.7) n=50 7.3 (6.9, 8.5) n=37 7.8 (6.9, 9.7) n=41 -0.70 (-1.2, -0.1)-0.30 (-1, 0.2) Plate8.3 (7.5, 10.4) n=50 7.5 (6.9, 8.4) n=42 7.5 (6.9, 8.4) n=43 -0.60 (-1.5, -0.3)-0.50 (-1.2, 0.05) Control8.0 (7.5, 9.7) n=50 7.3 (6.9, 7.9) n=41 7.8 (7, 9.2) n=42 -0.60 (-1.4, 0.0)-0.30 (-0.80, 0.4) Median (Interquartile Range)

31. 31 Diabetes Nutrition Study (DINES) A1C at 3-months & 6-months by study group

32. 32 Diabetes Nutrition Study (DINES) Adjusted A1C Mean ΔA1C (%) 6-Months 95% Confidence Intervalp-value Carb Counting vs. Control -4.9[-11.9 – 1.7]0.220 Plate vs. Control -6.6[-13.3 – 0.1]0.051 CDE vs. Control -5.9[-12.5 – 0.4]0.065 Adjusted for age, gender, race, income, years of diabetes, baseline A1C, and time interval

33. 33 Diabetes Nutrition Study (DINES) Adjusted A1C: Subgroup Analysis Mean ΔA1C (%) 6-Month 95% Confidence Intervalp-value Carb Counting vs. Control -9.3[-17.9 - -1.8]0.005 Plate vs. Control -8.2[-16.6 - -0.9]0.005 CDE vs. Control -8.6[-17.3 - -1.6]0.002 Participants with baseline A1C > 7.0% & <10.0% Adjusted for age, gender, race, income, years of diabetes, baseline A1C, and time interval

34. 34 Diabetes Nutrition Study (DINES) Secondary outcomes at 3- & 6-months VariableTreatment Arm BaselineBaseline to 3 months Baseline to 6 months Weight, lbsCarb218 (189, 253)-1 (-5.3, 0.3)-1.8 (-6.5, 1) Plate224 (189, 259)-1 (-3.5, 0)-1.0 (-10, 5) Control216 (180, 243)0 (-2, 2.6)0 (-4, 5.8) Treatment Satisfaction (range 6-36) Carb24 (17, 30)3 (0.3, 8.5)4 (-1, 9) Plate22.5 (17, 29)7 (0.5, 11)7 (1, 13) Control26 (19, 31)2 (-1, 5)3 (-2, 9) Self-Efficacy (range 8-40) Carb23.5 (18, 29)2 (0, 9) Plate24 (21, 27)4 (-1, 10)5 (-2, 8) Control24 (21, 29)3 (-2, 7)2.5 (0, 8) Median (Interquartile Range) Adjusted for age, gender, race, income, years of diabetes, baseline A1C, and time interval

35. 35 Diabetes Nutrition Study (DINES) Secondary outcomes at 3- & 6-months VariableTreatment Arm BaselineBaseline to 3 months Baseline to 6 months Weight, lbsCDE220 (189, 258)-1 (-5, 0)-1.7 (-10, 4.2) Control216 (180, 243)0 (-2, 2.6)0 (-4, 5.8) Treatment Satisfaction (range 6-36) CDE23 (17, 29)6 (0, 11)6 (0, 12) Control26 (19, 31)2 (-1, 5)3 (-2, 9) Self-Efficacy (range 8-40) CDE24 (20, 28)3 (-1, 9)4 (-1, 8.8) Control24 (21, 29)3 (-2, 7)2.5 (0, 8) Median (Interquartile Range) Adjusted for age, gender, race, income, years of diabetes, baseline A1C, and time interval

36. 36 Diabetes Nutrition Study (DINES) Secondary outcomes change: Subgroup Analysis VariableTreatment Arm BaselineBaseline to 3 months Baseline to 6 months Weight, lbsCDE216 (187, 249)-1 (-5, 0)-2 (-10, 2.8) Control200 (180, 234)0 (-2, 4.5)0 (-4, 3.0) Treatment Satisfaction (range 6-36) CDE23 (17, 29)5 (-0.5, 11)7 (0, 12) Control26 (19, 31)0 (-2, 4)2 (-2, 7) Self-Efficacy (range 8-40) CDE24 (20, 28)3 (-1, 8)4 (0, 8) Control24 (21, 29)4 (-2, 8)2 (0, 8) Median (Interquartile Range) Participants with baseline A1C > 7.0% & <10.0% Adjusted for age, gender, race, income, years of diabetes, baseline A1C, and time interval

37. 37 Summary At 3- and 6-months the A1C of intervention and control groups significantly improved from baseline At 6-months there was a trend for greater improvement in A1C for both intervention groups compared to control In subgroups analyses for participants with baseline A1C 7-10%, both plate and carb counting resulted in significant improvement in glycemic control CDE delivered MNT resulted in greater reduction in weight and improvement in patient satisfaction scores compared to control

38. 38 Limitations Minimal prevalence of low health literacy subjects limited ability to evaluate literacy intervention effect on A1C Highly motivated group as shown by significant improvement in glycemic control in control arm Pts self-initiated changes within meal planning group –Some went from plate to carb gram counting; –Some in carb group didn’t count carbs Losses to follow-up/ missing data Short duration of follow-up limits examination of persistence of MNT skills in intervention arms

39. 39 Lessons Learned: Applications to clinical practice Diabetes MNT must be tailored to the individual: - no pre-determined meal planning strategy - no pre-set calorie/carb levels - nutrition intervention needs to be based on assessment

40. Lessons Learned The tool is only as good as the user Carb portioning regardless of method is beneficial 40

41. 41 Lessons Learned Low literacy and picture based materials well-received by all ACP Living with Diabetes GuideDLNET

42. 42 Conclusions CDE delivered MNT is an important component of comprehensive diabetes care and all methods improve glycemic control Tailored education may benefit patients, but larger studies are needed

43. Acknowledgements Vanderbilt Program on Effective Health Communication Russell Rothman, MD, MPP Kerri Cavanaugh MD MPH Dianne Davis RD CDE Becky Gregory RD CDE Kathleen Wolff, Ken Wallston PhD Duff Green BA MDiv Tom Elasy MD MPH Robert Dittus MD MPH Ayumi Shintani PhD Svetlana Eden, MS Matt Kennon Shari Barto Funding –American Association of Diabetes Educators –NIH/NIDDK K23DK080952 (Cavanaugh) K23DK065294 (Rothman) 5P60DK020593 (VUMC DRTC)

44. Extra slides

45. 45 Health literacy is associated with outcomes Design: RCT Setting: Primary Care Intervention ► Diabetes Education ► Evidence-based medication algorithms ► Database to track and manage patient outcomes ► Diabetes Care Coordinator ► Addressed health literacy –Individualized verbal education –Materials –Clear communication –“Teach back” techniques Diabetes intervention study

46. 46 Factors for health communication Baker DW JGIM 2006; 21: 878-83 MNT

47. 47 Health literacy & outcomes Health Outcomes/ Health Services - BMI - General Health Status - Hospitalization - Mortality - Emergency department care - Depression - Diabetes Control - HIV Control - Prostate Cancer Stage - Mammography - Pap smear, STD screening - Immunizations - Cost Behaviors - Breastfeeding - Medication adherence - Smoking, substance abuse Knowledge - Food label & portion size estimation - Birth control - Emergency department instructions - Asthma - Hypertension - Diabetes

48. 48 DNT Example Items Your target blood sugar is between 60 and 120. Circle the values below that are in the target range (circle all that apply): 55 145 118 Correct Response: Circle 118 only Percent Correct: 74% Correct Response: 63 g Correct: 44% If you ate the entire bag of chips, how many total grams of carbohydrate would you eat? Nutrition Facts Serving Size 1oz. (28g/About 10 chips) Servings Per Container 3.5 Amount Per Serving Calories 140 Calories from Fat 60 % Daily Value* Total Fat 6g 10% Saturated Fat 0.5g 4% Cholesterol 0mg 0% Sodium 150 mg 7% Total Carbohydrate 18g 6%

49. Diabetes Nutrition Study (DINES) Secondary outcomes: Subgroup Analysis VariableTreatment Arm BaselineBaseline to 3 months Baseline to 6 months Weight, lbsCarb218 (188, 252)-2 (-6.2, 0)-2.5 (-10, 1) Plate212 (181, 238)-0.7 (-3.5, 0)-1.5 (-10, 4.2) Control200 (180, 234)0 (-2, 4.5)0 (-4, 3.0) Treatment Satisfaction (range 6-36) Carb24 (18, 30)3 (0, 8.5)4 (-1, 10) Plate22 (18, 28)6 (-1, 11)7.5 (1, 13.5) Control26 (19, 31)0 (-2, 4)2 (-2, 7) Self-Efficacy (range 8-40) Carb24 (20, 29)2 (-0.8, 7)2 (0, 8) Plate24 (21, 26)4 (-1, 9)5 (-0.3, 8.5) Control23 (20, 29)4 (-2, 8)2 (0, 8) Median (Interquartile Range) Participants with baseline A1C > 7.0% & <10.0% Adjusted for age, gender, race, income, years of diabetes, baseline A1C, and time interval