This Article Abstract Free Full Text (Free PDF) Free Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to Saved Citations Download to citation manager Request Permissions Request Reprints Add to My Marked Citations Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (1) Citing Articles via Google Scholar Citing Articles via Scopus Google Scholar Articles by Povey, R. C. Articles by Clark-Carter, D. Search for Related Content PubMed PubMed Citation Articles by Povey, R. C. Articles by Clark-Carter, D. Social Bookmarking                         What's this?

Diabetes and Healthy Eating

A Systematic Review of the Literature

From the Center for Health Psychology, Faculty of Sciences, Staffordshire University, Stoke-on-Trent, United Kingdom.

Correspondence to Rachel Povey, PhD, Centre for Health Psychology, Faculty of Sciences, Staffordshire University, Stoke-on-Trent, ST4 2DE UK (r.povey{at}staffs.ac.uk).

	Abstract

Top Abstract Methods Results Appendix List of Articles... References

 
Purpose

The purpose of this study is to review the literature on healthy eating interventions within diabetes care.

Data Sources

The databases PubMed, Cumulative Index to Nursing and Allied Health Literature, and PsycINFO were searched for the terms healthy eating or dietary and intervention and diabetes in the abstracts. In addition, electronic searches of the contents of Diabetes, Clinical Diabetes, Diabetes Care, Diabetes Spectrum, and Diabetic Medicine were conducted.

Study Selection

Articles were included in the review if the participants were diagnosed with type 1 or type 2 diabetes, results from an intervention to promote dietary change were reported, measurements were taken from 2 time points (preintervention and postintervention), at least 1 outcome measured eating behavior, and the articles had been published between 1990 and the present date.

Data Extraction

Data were extracted from the 23 studies found to be eligible. Studies were compared in terms of sample types and sizes, duration, and type and content of intervention.

Data Synthesis

Comparisons were made between the characteristics of those studies that produced a statistically significant result and those that did not, for the different outcome measures.

Conclusions

For studies measuring the outcomes of weight, fat intake, saturated fat intake, and carbohydrates, there was a tendency for successful interventions to include an exercise dimension and group work. Some outcomes had a tendency to show significant changes in studies of longer duration (eg, serum cholesterol), whereas others were more likely to show significant changes in studies of shorter duration (eg, weight, fiber), suggesting that certain outcomes may be more difficult to maintain. Future research would benefit by ensuring that sample sizes are adequate to give sufficient power, and interventions should be designed that focus on the maintenance in addition to the initiation of eating behavior change.

There currently exists a large number of studies that have delivered healthy eating and dietary change interventions in a range of settings. These studies use many different mechanisms to promote dietary change, ranging from those that take a prescriptive approach to those that have taken more of a patient-centered empowerment approach.⁵ In addition, such studies have used a range of outcome measures to examine the effectiveness of interventions, varying from behavioral outcomes such as changes in nutrients (eg, fat and fiber intake) to clinical outcomes such as hemoglobin A1c (HbA1c) and weight. However, until this literature is reviewed systematically and the effectiveness of such interventions documented,⁶ it is impossible to derive any specific conclusions regarding the relative effectiveness of the mechanisms of such interventions, and this is what this article seeks to do.

The central objective of this article is to review systematically published reports of healthy eating interventions in diabetes care since 1990. The ultimate aims are to evaluate the effectiveness of the mechanisms of different healthy eating interventions for people with diabetes; to provide guidance to diabetes educators intending to promote healthy eating, where possible; and to identify new directions for the development of future research.

	Methods

Top Abstract Methods Results Appendix List of Articles... References

 
Searching
The following databases were searched for the terms healthy eating or dietary and intervention and diabetes in the abstracts: PubMed (March 13, 2006), Cumulative Index to Nursing and Allied Health Literature (Cinahl; March 13, 2006), and PsycINFO (February 15, 2006). Further limits were imposed where possible to English language only (Cinahl, PubMed), peer reviewed (Cinahl), and humans only (PubMed). Finally, searches using the same terms were carried out electronically for the specific journals Diabetes, Clinical Diabetes, Diabetes Care, Diabetes Spectrum, and Diabetic Medicine, as these were considered to be journals highly relevant to the subject area.

Selection
Abstracts were included in the review if they were original articles; the participants were diagnosed with type 1 or type 2 diabetes; results from an intervention to promote dietary change were reported; measurements were taken from 2 time points, preintervention and postintervention,⁶ at least 1 outcome measured dietary behavior, and the articles had been published between 1990 and the present date. Articles were excluded if they had not been peer reviewed; they were descriptive articles such as reviews, advice, lectures, or conference proceedings; they were not written in English; they did not use human participants; and the intervention was targeted toward a group that was so specific that the extent to which it could be applied to a general population with diabetes was reduced. Articles were also excluded if they reported cross-sectional studies, as this type of study design did not make it possible to examine the impact of an intervention on outcomes over a period of time. Finally, samples were not restricted to the United States only or to any specific age group.

In the first instance, 2 independent reviewers (the authors) read all the abstracts resulting from the database searches and agreed on whether they would be selected for the review. Full-text copies of the articles were retrieved, unless they were rejected during this initial screening process. They were rejected if there was enough information in the abstract to demonstrate that they did not meet the inclusion criteria or that they fulfilled 1 or more of the exclusion criteria. Once retrieved, each of the articles was scrutinized separately by the 2 reviewers to check whether it met the inclusion and exclusion criteria. At this stage, 1 further exclusion criterion was included to meet the nutrition principles and recommendations of the American Diabetes Association⁷ and also to meet standard 8 of the National Standards for Diabetes Self-management Education.⁸ This criterion therefore stated that any study that did not "include some type of individualized assessment, development of a plan, and periodic reassessment between instructor(s) and participant when directing the selection of appropriate education materials and intervention" was excluded. This also served to remove all of those studies that compared prescribed diets. This therefore enabled us to focus our review on the differences in effectiveness of the mechanisms of the different types of interventions used to promote healthy eating and dietary change rather than on the differences in the effectiveness of different prescribed diets. Any articles that did not meet all the criteria were excluded from the final review.

Validity Assessment
Quality assessment of evidence for each eligible article was determined using the American Diabetes Association evidence grading system.⁹ Evidence was divided into 3 levels: A, B, or C. Articles that were graded A were based on large, well-conducted, generalizable, randomized controlled trials that were adequately powered; articles that were graded B provided supportive evidence from well-conducted cohort studies or well-conducted case-control studies; and articles that were graded C provided supportive evidence from poorly controlled or uncontrolled studies. All articles were assessed for quality by the 2 independent reviewers, and any disagreements in categorization were discussed.

Data Abstraction
Data were extracted from the eligible studies. Each article included in the review was scrutinized thoroughly and methodically by the 2 independent reviewers, and the relevant data were extracted. During this process, 2 extra articles were removed, as under closer scrutiny it was found that they did not meet all the inclusion criteria.^10,11

Study Characteristics
Data extracted from the final eligible studies included descriptive information (ie, diagnosis, age [mean and standard deviation/range where available], gender, setting, description of sample), information about the methodology (ie, description of the intervention and control conditions, recruitment methods, study design, method of allocation to study conditions), and, finally, details of the relevant outcome measures. Outcomes were defined in terms of learning (eg, change in knowledge or skills), behavior changes (eg, change in fat intake or fruit and vegetable intake), clinical improvement (eg, change in glycemic control or weight), or improved health status (eg, quality of life or well-being) in line with the standards proposed by the American Association of Diabetes Educators.^2,6 Notes were also taken regarding possible confounding effects and limitations of the different studies.

Data Synthesis
Results from studies were examined by comparing the characteristics of those studies that produced a statistically significant result (successful studies) with those that did not (unsuccessful studies) for the different outcome measures. To make the task more manageable, this process was limited to those outcomes that had been measured by at least 6 studies. Studies were compared in terms of sample types and sizes, duration, and type and content of the intervention.

	Results

Top Abstract Methods Results Appendix List of Articles... References

 
Trial Flow
The total number of potentially relevant articles identified from all databases and abstracts screened for retrieval was 497 abstracts (totals: PubMed, 357 abstracts; Cinahl, 76 abstracts; PsycINFO, 17 abstracts; specific diabetes journals, 47 abstracts). From these 497 abstracts, 86 articles were retrieved for a more detailed evaluation (33 were cited by more than 1 database). After retrieval, a further 61 were excluded from the systematic review (for not fulfilling the inclusion/exclusion criteria), resulting in a total number of 25 from which data were to be extracted. During the data extraction process, 2 additional articles were removed, as under closer scrutiny it was found that they did not meet all the inclusion criteria.^10,11 Finally, a further 3 studies were found to be reporting results from the same sample as other articles that were already included^12-14; these were therefore removed for the purpose of analysis. For the full list of 86 articles and details of reasons why specific articles were excluded, please see the appendix. A profile summarizing the trial flow can be seen in Figure 1.

View larger version (20K): [in this window] [in a new window]   Figure 1. Profile summarizing trial flow. a This number is smaller than the total number of articles meeting the inclusion criteria (n = 128) because of the overlap of articles across databases.

Table 1 summarizes the different outcome measures taken, the number of studies that used each measure, and the number that found a statistical difference.

In addition to there being a variety of outcomes measured, the interventions differed in a number of other aspects. These included whether they were part of a more general intervention into lifestyle (including looking at exercise); whether they targeted specific aspects such as weight, risks for coronary heart disease, or specific groups based on ethnicity or a medical condition; whether they dealt with aspects of behavior change such as self-efficacy, healthy eating barriers, problem solving, or stages of change; and whether they employed a particular approach to learning such as tackling only a limited number of concepts at a time, concentrating on skills, providing feedback about laboratory results, or using behavior modification. Also, the interventions differed in terms of how they were delivered: whether computers were used to communicate with participants or to deliver the intervention, whether participants could communicate with those running the program by telephone, whether discussion groups or videos were employed, whether peer counselors were involved, whether a dietitian or nutritionist delivered the intervention, and whether specific recipes were provided (and if so, whether they came from the participants or those running the study). Finally, the nature of the samples differed in terms of age, gender, ethnicity, and type of diabetes. Although type 1 and type 2 diabetes are metabolically different with different eating priorities, only 2 studies^21,22 specifically examined people with type 1 diabetes alone. As it would be inappropriate to carry out between-group comparisons on such small numbers, it was decided to combine the studies for the purpose of this review. See Table 2 for descriptions of the specific studies included in the final review.

Data Synthesis
There was 91.9% agreement between the reviewers for the 86 articles reviewed, with consensus reached for the remaining 7 articles through discussion. Agreement on quality criteria on the final 23 articles was moderate, with 75% agreement between the 2 reviewers. Disagreement over the remaining articles was because they did not fit neatly into any of the 3 possible categories, but again, consensus was reached by discussion.

Given the heterogeneity of studies, in an attempt to distinguish studies that did produce a statistically significant result from those that did not, characteristics of each type of study were explored. However, the multidimensional nature of healthy eating behavior resulted in a total of 65 dietary outcomes measured for the different studies (with 16 different measures of nutrient intake; see Table 1). To make the comparison between studies more manageable, results are therefore reported only for those outcomes that have been measured by at least 6 studies.

Weight
There were 3 significant^12,19,23 and 9 nonsignificant studies.^{16,18,21,24-29} Two of the 3 significant studies involved an exercise element,^12,23 while only 4 of the 9 nonsignificant studies involved exercise.^16,21,24,28 One additional study,³⁰ which involved drugs and exercise, found a statistically significant increase in weight with intensive therapy, despite significant reductions in total energy in both groups. It was argued that a possible reason for this discrepancy may be underreporting of food intake within the sample. The sample size was slightly smaller for the significant studies (mean = 118.3, SD = 141.5 vs mean = 125.7, SD = 146.5), and the duration was shorter (mean = 0.55 years, SD = 0.42 vs mean = 0.69 years, SD = 0.46). The focus of the significant studies was very different, varying from a Mediterranean Lifestyle Program,¹² in which participants attended weekly meetings for 6 months, consisting of a combination of physical activity, stress management, tips for adhering to a Mediterranean diet and support groups; to an intervention in which participants received intensive, personalized nutrition counseling to follow either high-protein or high-carbohydrate diets¹⁹; to an intervention specifically designed for overweight African Americans,²³ in which each participant received an individualized weight reduction diet and attended 12 weekly group sessions, 1 individual diet counseling session, and 6 biweekly group sessions. Whereas none of the significant studies included anyone with type 1 diabetes, 2 of the nonsignificant studies did.^21,25 Two^12,23 of the 3 significant studies involved some form of group work, while only 2 of the 9 nonsignificant studies involved an element of group work.^16,24 The study that showed a significant weight gain³⁰ included group work that involved spouses and lasted 4 years.

HbA1c
There is no obvious pattern to distinguish the significant and nonsignificant studies. Two^23,30 of the 7 significant studies^{17,19,23-25,27,30} involved an exercise component, one consisting of advice about exercise activities, positive feedback, and educational exercises³⁰ (this study also involved drugs [polypharmacological therapy]) and the other consisting of 20 minutes of low-impact aerobic activity.²³ Two^16,31 of the 9 nonsignificant studies^{16,18,20,22,26,29,31-33} also involved exercise. An additional study, which also included exercise,²¹ showed inconsistent results in that one group showed no significant change over the intervention whereas a crossover group subsequently given the same intervention did show a significant improvement. In this group, there was little change in HbA1c levels in the first 6 months when the patients were receiving standard care, but the levels decreased significantly (P = .002) over the following 6 months when the group received intensive dietary advice. Sample size could not be seen as an explanation for the difference between significant and nonsignificant studies (significant studies: mean sample size = 139.85, SD = 165.99; nonsignificant studies: mean = 118.33, SD = 79.84; the smallest sample among the significant studies¹⁹ was 12, whereas for the nonsignificant studies, the smallest sample²² was 32). Although the mean duration of treatment was longer for the significant studies, this was because 1 significant study³⁰ lasted 4 years, and so duration could not be seen as an explanation where this could be ascertained (1 significant study failed to give details of duration,¹⁷ and 2 nonsignificant studies provided imprecise information about it^22,32; significant studies: mean duration = 1.04 years, SD = 1.48; nonsignificant studies: mean = 0.79 years, SD = 0.42). Two of the significant studies lasted only 8 weeks. Finally, 4 of the significant studies^17,23,24,30 and 4 of the nonsignificant studies involved groups.^16,20,22,32

Fasting Blood Glucose
For this measure, it is also difficult to draw any conclusions about factors that might make a study successful or unsuccessful, as there were only 2 nonsignificant studies^17,28 and 1 failed to give its exact duration.¹⁷ One of these studies¹⁷ compared 183 participants who took part in an educational program consisting of a mixture of individual and group sessions with 95 participants from a control group. Although the intervention group showed no significant reduction in fasting blood sugar levels, it is interesting that it did, however, demonstrate a significant reduction in HbA1c levels. The 5 significant studies^{18,19,22,24,27} had a mean sample size of 31.80 (SD = 30.20) and mean duration of 0.43 years (SD = 0.40); 1 study gave insufficient details about its duration.²² The nonsignificant studies had sample sizes of 278 and 8. One of the nonsignificant studies involved exercise,²⁸ whereas none of the significant studies did. One of the significant studies had used a selection criterion that the participants should be newly diagnosed with blood glucose levels of 6.7 mmol/L or greater in repeated measurements.¹⁸ Two of the significant studies^22,24 and 1 of the nonsignificant studies¹⁷ involved groups.

Triglycerides
Three of the 8 studies produced significant results.^18,26,30 Results showed serum triglycerides to decrease significantly within an intensified dietary education intervention¹⁸ and a multifactorial intervention consisting of behavior modification and polypharmacological therapy.³⁰ In the third study,²⁶ triglycerides were found to decrease significantly in a weight management group from recruitment to month 18, but no significant differences were found for participants allocated to modified lipid or high-carbohydrate diets. Triglyceride levels were also found to be significantly lower in the weight management group than in the high-carbohydrate group throughout the study. What seems to characterize the significant studies is a longer duration combined with a larger sample size (mean duration = 2.17 years, SD = 1.61 years; mean sample size = 99.00, SD = 44.44). In contrast, the nonsignificant studies^{19,21,23,24,27} had a mean sample size of 33.60 (SD = 26.44) and mean duration of 0.34 years (SD = 0.18). Even though the mean duration for the significant studies was inflated by 1 study that lasted 4 years,³⁰ the shortest duration of any of them was 1 year,¹⁸ whereas the longest duration of the nonsignificant studies was 6 months.^21,23 One³⁰ of the significant studies involved advice about exercise activities (in which participants were urged to start or continue exercise by positive feedback and educational exercises) and polypharmacological therapy. Two of the nonsignificant studies also involved exercise,^21,23 and none involved drugs. Only 1 significant study³⁰ involved some group work, while 2 of the nonsignificant studies did.^23,24

Serum Cholesterol
The 3 significant studies^29,30,33 had good sample sizes (mean = 205.33, SD = 59.18) and durations (mean = 1.75 years, SD = 1.95), and only 1 included exercise³⁰ (this study also included polypharmacological therapy). Two^16,23 of the 9 nonsignificant studies included exercise interventions. The mean sample size for the nonsignificant studies^* was 69.33 (SD = 54.50), and the mean duration was 0.62 years (SD = 0.44). One of the significant studies³⁰ and 5 of the nonsignificant studies involved some group work.^{16,20,23,24,32} This group work varied from an educational approach based on the needs of participants and their spouses in the significant study³⁰ to community-based group sessions,¹⁶ a peer-directed discussion forum,²⁰ sessions focusing on nutrition education and exercise,²³ and peer-professional discussion groups^24,32 in the nonsignificant studies. An additional study, which included exercise,²¹ showed inconsistent results in that one group showed no significant change over the intervention whereas a crossover group subsequently given the same intervention did show a significant improvement.

High-Density Lipoproteins
One²¹ of the 4 significant studies^18,19,21,26 included an element of exercise by providing participants with intensive lifestyle education for 6 months consisting of individualized dietary and exercise programs. Participants went to monthly meetings with the research team, which consisted of reinforcement of advice, feedback on laboratory results, and the use of behavior modification techniques to improve metabolic control. Three^16,23,30 of the 5 nonsignificant studies^{16,23,24,27,30} also involved exercise interventions (1 of which also included a drug intervention³⁰). There was no advantage in terms of sample size or duration for the significant studies (mean sample size = 55.25, SD = 30.59; mean duration = 0.79 years, SD = 0.59) over the nonsignificant studies (mean sample size = 83.00, SD = 73.44; mean duration = 1.21 years, SD = 1.59). The higher mean duration of the nonsignificant studies is inflated by 1 study that lasted 4 years.³⁰ None of the significant studies involved groups, while 4 of the nonsignificant studies did.^16,23,24,30

Low-Density Lipoproteins
Two of the 6 studies^{19,21,23,24,26,27} that measured low-density lipoprotein (LDL) cholesterol were statistically significant.^19,21 One of these studies¹⁹ compared participants on high-protein diets (40% carbohydrate, 30% protein, 30% fat) and high-carbohydrate diets (55% carbohydrate, 15% protein, 30% fat) and found a significant decrease in the level of LDL cholesterol in only the high-protein group. The other study²¹ examined an intensive lifestyle education intervention and showed inconsistent results in that 1 group showed no significant change over the intervention whereas a crossover group subsequently given the same intervention initially showed a significant increase in LDL cholesterol while receiving standard care and later showed a significant decrease in LDL cholesterol. There was wide overlap between the sample sizes of the significant (mean = 36.50, SD = 34.65) and nonsignificant studies (mean = 39.75, SD = 28.00) and between their durations (significant studies: mean = 0.32 years, SD = 0.24; nonsignificant studies: mean = 0.64 years, SD = 0.59). One of the significant studies²¹ and 1 of the nonsignificant studies²³ involved exercise. Neither of the significant studies involved group work, whereas 2 of the nonsignificant studies did.^23,24

Kilocalories
There were 4 significant^17,21,33,34 and 6 nonsignificant^{16,18,19,26,27,30} studies that measured kilocalories as an outcome, with an additional study²³ that found an initially significant result after half the intervention had been completed but a nonsignificant result after the full duration of the intervention. Interventions that found a significant improvement included displays of suitable food and general advice on meal planning¹⁷; specific targets for dietary goals; information on nutrition, timing of meals, eating out, reading of food labels, adapting recipes, and the use of artificial sweeteners plus feedback from laboratory results²¹; touch-screen computer-assisted assessment of likely barriers to healthier eating and targets for a specific aspect of diet (eg, fat intake)³³; and a focus on skills to promote change, with specific targets for dietary fat, saturated fat, cholesterol, and fiber, demonstrations of food preparation and identification of potential problems, help with label reading, and analysis of low-fat meals.³⁴ Among the studies that did not find a significant improvement in kilocalories was a study that specifically concentrated on improving intake of fat, saturated fat, and carbohydrates¹⁶; one that had goals and targets for weight, normoglycemia, correction of dyslipidemia, blood pressure, and intake of fats, cholesterol, carbohydrates, fiber, and sucrose¹⁸; one that placed participants in high-protein or high-carbohydrate conditions and gave calorie-restricted alternatives for meals and advice on serving sizes¹⁹; one that targeted intake of carbohydrates and fats, gave advice on portion size, and introduced new foods³⁰; one that targeted body mass index with specific groups either concentrating on weight management, lipids, or carbohydrates and fiber, in which all participants were shown food displays and given advice on adapting favorite recipes²⁶; and one that concentrated on weight maintenance but had targets for balance of carbohydrates, protein, fat, and amount of fiber.²⁷ There was no strong pattern to distinguish the studies, except for a difference in sample size. There was a big difference in mean sample size, with the mean sample size for the significant studies being 144.25 (SD = 114.33), with durations of 6, 9, and at least 7 months (1 duration was not provided¹⁷). For nonsignificant studies, on the other hand, the mean sample size was almost half: 82.50 (SD = 66.52), with a mean duration of 1.30 years (SD = 1.42). Only 1 of the 4 significant studies included exercise,²¹ and 2 of the nonsignificant studies included an exercise intervention,^16,30 with 1 of these also including a drug intervention.³⁰ Two of the significant studies^17,34 and 3 of the nonsignificant studies^16,23,30 involved groups. One of these significant studies¹⁷ encouraged family members to attend the sessions and found that in the intervention group postintervention, a significantly higher proportion of participants reduced their total calorie intake when they detected that they were overweight compared with before the intervention; there was not a significant reduction in the control group. The other significant study³⁴ analyzed kilocalorie intake and, adjusting for baseline intake, found that after 3 months, the intervention group had a significantly lower intake of kilocalories than the control group did.

Fat
There were 10 significant and 4 nonsignificant^18-20,26 studies measuring fat intake. Among the significant studies, one¹⁷ used booklets on meal planning and general health care and presented food displays of meals suitable for people with diabetes. In the intervention group only, the authors found that a significantly higher proportion of the participants postintervention reported cutting down on oily/fatty food compared with reports preintervention. Another study³⁰ was specifically aimed at fat and carbohydrate intake, giving advice on portion size, new foods, and menus. The authors found that both intensive and control (standard treatment) groups had significantly reduced total fat (as a percentage of energy intake) but that the intervention group had a significantly lower intake than the control group. A further study²³ also included targets for fat and carbohydrate intake. Advice was based on the culture of the participants and on recipes provided by participants. This included advice on meal planning, food shopping, label reading, and food selection at restaurants. The authors found a significant difference between the usual care and intervention groups in change scores between baseline and 3 months, but the difference was not significant at 6 months. Another study²⁷ had specific targets for the proportions of carbohydrate, protein, and fat in the diet. The authors found a significant reduction in the amount of fat intake. However, there was no control group with which to compare this result. A further study²¹ gave out booklets that contained information on nutrition, the timing of meals, eating out, reading food labels, and adapting recipes. There were specific targets for intake of carbohydrates, fat, and fiber; feedback from laboratory results was given; and behavior modification techniques were used. The study used a crossover design whereby 1 group received an intervention for 6 months followed by 6 months of usual care while the other received usual care for 6 months followed by an intervention for 6 months. Both groups had reduced fat intake significantly after 6 months, with the intervention group having a significantly lower intake than the other group. At 12 months, intake was compared with the 6-month levels. The group that had been given the intervention followed by usual care showed a further significant reduction in intake, whereas the other group, which was now in the intervention phase, did not show a significant improvement. The next study³³ used a touch-screen to identify dietary barriers and gave feedback on potential problem areas for diet. This was followed by individual sessions that set goals and gave advice on problem solving. Participants were split into 2 groups depending on their self-efficacy scores, and the groups were shown a video that was adapted for their group. Both videos looked at strategies for dealing with barriers to healthy eating, with those with lower self-efficacy having their video presented in an interactive format. The data for the 2 self-efficacy groups were combined, and postintervention consumption rates were adjusted to allow for baseline rates. The intervention groups had significantly better intake than the usual care group did. Another study²⁹ had 4 conditions, each of which received a touch-screen assessment and tailored feedback at the start of the program and after 3 months. One group received no more than this; 1 group received telephone follow-up; 1 group received community support, including details of community resources and 4 newsletters; and 1 group had a combination of telephone follow-up and community resources. The researchers report a significant improvement in fat intake in the telephone follow-up conditions. The next study³⁵ looked specifically at reducing coronary heart disease risks and recommended a Mediterranean diet (rich in -linolenic acid). According to a fat and fiber questionnaire, the intake of the intervention group, adjusted for baseline level, was significantly better than for the usual care condition. The final significant study³⁴ focused on skills to promote dietary change using Bandura's social learning theory. It had specific targets for intake of various aspects of diet, including fat, and involved demonstrations of food preparation, shopping, and eating practices. Adjusting for baseline kilocalorie intake, the intervention group had a significantly lower fat intake than the control group did. There does not appear to be any pattern of differences between studies here, apart from sample size. The mean sample size for the significant studies was 179.60 (SD = 138.87), with a mean duration of 1.05 years (SD = 1.23 for 8 studies; 1 not given¹⁷ and another given as "at least 7 months"³⁴). The mean sample size for nonsignificant studies, on the other hand, was 73.25 (SD = 50.08), with a mean duration of 0.73 years (SD = 0.64). Significant studies were therefore more likely to have bigger samples, and the interventions were more likely to last longer. However, there is large overlap in the range of durations between the significant and nonsignificant studies. Four of the interventions in the significant studies also included an exercise component^21,23,30,35 (1 of which also included drugs³⁰), whereas none of the nonsignificant interventions included exercise. Five of the significant studies involved groups,^{17,23,30,34,35} while only 1 of the nonsignificant studies involved some group work.²⁰

Saturated Fat
There were 7 significant^{18,21,23,30,33,34,35} and 3 nonsignificant^16,19,26 studies that measured saturated fat intake. All the significant studies were also significant for fat intake, apart from 1 study,¹⁸ which had goals and targets for weight, normoglycemia, correction of dyslipidemia, blood pressure, and intake of fats, cholesterol, carbohydrates, fiber, and sucrose. As well as information on the principles of diets for people with diabetes, it involved attempts to modify behavior via improving motivation and recognizing difficult situations such as parties and traveling. The intervention group was found to have significantly lowered saturated fat intake, while the conventional treatment group did not show a significant improvement. In addition, although the means and standard deviations are not reported in the article to any decimal places, a t test based on the reported figures shows that the intervention group had a significantly lower fat intake than the conventional treatment group did (t₍₈₂₎ = 2.04, P = .04, d = 0.45). There were large amounts of overlap between the significant and nonsignificant groups for both sample size and duration, with differences between the means being accentuated by 1 significant study having a particularly long duration³⁰ (4 years) and 1 nonsignificant study having a particularly short duration (8 weeks) and small sample (12 participants).¹⁹ For significant interventions, the mean sample size was 125.00 (SD = 81.93), whereas for nonsignificant interventions, the mean sample size was 82.33 (SD = 81.07). The duration for significant interventions varied between 6 months²¹ and 4 years³⁰ (mean = 1.29 years for 6 studies, SD = 1.35: the other was at least 7 months³⁴), whereas for nonsignificant studies, the duration varied between 8 weeks¹⁹ and 18 months²⁶ (mean = 0.88 years, SD = 0.68). Finally, 4 of the significant interventions also included an exercise component: 1 study³⁰ (which also included drugs) encouraged participants by positive feedback and by educational exercise, giving the example of demonstrations of the immediate lowering effect on blood glucose of moderate physical activity. Another study²³ had participants take 30 minutes of exercise during the sessions with the researchers and encouraged them to exercise for 2 additional days per week. Another study²¹ individualized the activity and duration of exercise depending on the participant's level of fitness and goals. Finally, 1 study³⁵ started with the goal of 30 minutes of moderate exercise on most days of the week, but once this had been achieved, the goal was extended to 1 hour of moderate aerobic activity per day. Only 1 of the nonsignificant interventions included exercise.¹⁶ Four of the significant studies involved group work,^{23,30,34, 35} while only 1 of the nonsignificant studies involved groups.¹⁶

Carbohydrates
Four significant studies^19,21,30,34 and 3 nonsignificant studies^18,26,27 measured carbohydrate intake. An additional study²³ showed significance after half the duration of the intervention, but this became nonsignificant after the full duration of the study. There was little difference between the studies in mean sample size; for significant interventions, the mean sample size was 65.00 (SD = 59.47), whereas for nonsignificant interventions, the mean sample size was 54.33 (SD = 35.50). The mean duration for significant interventions, however, was almost twice that of nonsignificant interventions (mean duration for significant interventions = 1.55 years, SD = 2.13, with 1 study that lasted "at least 7 months"³⁴; mean duration for nonsignificant interventions = 0.88 years, SD = 0.68). However, it should be noted that 1 of the significant interventions lasted for 4 years,³⁰ therefore inflating the mean. Three of the significant studies included an exercise component,^21,23,30 whereas none of the nonsignificant studies did. Three of the significant studies involved groups,^23,30,34 with 1 study including spouses,³⁰ while none of the nonsignificant studies involved any group work.

Fiber
Three significant studies^17,19,27 and 5 nonsignificant studies^{18,21,23,26,34} measured fiber intake. Among the significant studies, 1 study,¹⁷ which took place in Singapore, asked specifically about the intake of unpolished rice as well as other high-fiber foods and found that there was a significant increase in the percentage intake in the intervention group but not in the control group. One study¹⁹ compared participants on high-protein diets (40% carbohydrate, 30% protein, 30% fat) and high-carbohydrate diets (55% carbohydrate, 15% protein, 30% fat) and found that the high-carbohydrate group had a significantly lower intake of fiber (mean = 15 g/d, SE = 1) than the high-protein group (mean = 18 g/d, SE = 1). The other significant study²⁷ had a recommended target of 35 g of fiber in the diet. After the intervention, participants had significantly increased their fiber intake to a mean of 32.7 g (SE = 1.1; as ascertained by 24-hour dietary recall to evaluate preintervention intake and 3-day dietary record for postintervention intake). There was no control condition for comparison. There was little difference in sample size (significant studies: mean = 101.67, SD = 152.72, which included 1 sample of 278 participants¹⁷ and the remainder with less than 20 participants; nonsignificant studies: mean = 62.2, SD = 16.35). Interestingly, the interventions for nonsignificant studies were more likely to be longer duration (significant studies: mean = 0.15 years, SD = 0.00, with 1 duration not provided; nonsignificant studies: mean = 0.88 years, SD = 0.48, with 1 other of at least 7 months' duration³⁴). None of the significant studies included an exercise component, whereas 2 of the nonsignificant studies did.^21,23 Finally, only 1 significant study involved groups,¹⁷ while 2 of the nonsignificant studies involved groups.^23,34

Protein
There were 4 significant studies^19,21,27,30 and 2 nonsignificant studies^18,23 that measured protein. One study¹⁹ compared participants on high-protein diets (40% carbohydrate, 30% protein, 30% fat) with those on high-carbohydrate diets (55% carbohydrate, 15% protein, 30% fat) and found that the high-carbohydrate group had a significantly lower intake of protein than the high-protein group did. Another significant study³⁰ was specifically aimed at fat and carbohydrate intake, giving advice on portion size, new foods, and menus. The authors found that both the intervention and control (standard treatment) groups had a significantly higher intake of protein (as a percentage of energy intake) at the end of the study period and that the intervention group had significantly higher protein intake than the control group did. Another study²⁷ had specific targets for the proportions of carbohydrate, protein, and fat in the diet. The authors found a significant increase in the intake of protein, although there was no control group with which to compare this result. A further study²¹ gave out booklets that contained information on nutrition, the timing of meals, eating out, reading food labels, and adapting recipes. There were specific targets for intake of carbohydrates, fat, and fiber, together with feedback from laboratory results and behavior modification techniques. The study used a crossover design (described above in the section on fat), and the group that received the intervention first had significantly increased their intake of protein after 6 months. However, neither group had increased their intake of protein significantly between the 6th and 12th month. The mean sample size for the significant studies was 59.25 (SD = 63.90), with 2 studies having only 12 participants¹⁹ and 15 participants,²⁷ respectively, while for the nonsignificant studies, the mean sample size was 74.00 (SD = 14.14). The mean duration of the significant studies was 1.20 years (SD = 1.87), but this was inflated by one study³⁰ that lasted 4 years. The mean duration for the nonsignificant studies was 0.75 years (SD = 0.35). Two significant studies involved exercise,^21,30 as did 1 of the nonsignificant studies.²³ One³⁰ of the significant studies, which also involved drugs, encouraged participants by positive feedback and educational exercise, giving demonstrations of the immediate lowering effect of moderate physical activity on blood glucose. The other significant study²¹ individualized the activity and duration of exercise depending on the participant's level of fitness and goals. One of the nonsignificant studies²³ involved some group work, as did one of the significant studies³⁰; the latter study also included spouses in the intervention.

Discussion
The heterogeneity of the studies considered in this review and the variety of different outcome measures made comparison between them extremely difficult. Rarely did a study examine changes in learning, behavior, clinical improvement, and health status, as recommended by the standards proposed by the American Association of Diabetes Educators.^2,6 Standardization of outcomes in future studies examining food and nutrition changes would make comparisons between different interventions much easier to examine.

Comparisons between successful (significant) and unsuccessful (nonsignificant) studies revealed some interesting differences between them, although there did not appear to be any clear patterns across the different outcomes, making it difficult to identify any clear distinctions. There was, however, a tendency among studies measuring the outcomes of weight, fat intake, saturated fat intake, and carbohydrates for successful interventions to include an exercise dimension and group work. It could be argued from this evidence, therefore, that the inclusion of exercise and an element of group work in interventions designed to reduce weight and fat intake and increase carbohydrate intake would be more successful than individualized programs focusing on food only.

There was no clear pattern for sample size or duration of studies. However, it is interesting that studies of longer duration were more likely to be significant when measuring the outcomes of HbA1c, triglycerides, serum cholesterol, fat, saturated fat, protein, and carbohydrates, whereas those of shorter duration were more likely to be significant when measuring the outcomes of weight, high-density lipoproteins, low-density lipoproteins, kilocalories, and fiber. Although these findings should be treated with caution, as 1 study lasting 4 years³⁰ will have raised the mean values of some of the significant studies, they are interesting, as they may reflect differences in the outcome variables measured. For example, the finding that studies were more likely to find interventions to be successful in reducing serum cholesterol if they were of longer duration may suggest that changes in cholesterol are more likely to be detected over longer periods of time and that any study seeking to do so should consider a longer follow-up period. On the other hand, studies that examined interventions seeking to change weight and fiber intake were more likely to be successful if they were of shorter duration, suggesting that these outcomes may be more difficult to maintain over a longer period. These findings lead us to stress that the importance of future research is not only to develop interventions to promote dietary change but also to focus on the maintenance of healthy eating behaviors.^3,36

Although there was not a straightforward relationship between a study being significant and the sample size, many studies were very underpowered and would have needed beyond a large effect size to have achieved statistical significance. It is becoming increasingly recognized that if statistical significance is to be used as a means to evaluate the outcome of a study, then an appropriate sample size needs to be chosen to give the study sufficient statistical power. Accordingly, there should be a more consistent use of power analysis at the design stage. An important element in choosing an appropriate sample size is the effect size.³⁷ Researchers could decide what would be the minimum effect that would be clinically useful and use that in their power calculations. Additional improvements that could be made in future studies would be for more consistent use of randomized allocation and inclusion of a control condition. Without these, confidence in the efficacy or lack of efficacy of a given intervention is undermined.

Before concluding, it is important to be aware of the particular aims of this review and therefore of the limits on generalization of the findings. One potential limitation is the search terms used in this review. The review was carried out to examine interventions that have been designed to promote healthy eating in people with diabetes; hence, search terms were chosen accordingly. However, it is possible that with alternative search terms, the searches would have revealed different studies. Second, although past reviews, meta-analyses, and practice guidelines were used as important background information for this review, they were not used as sources of potentially eligible studies. It is possible that this may have led to the omission of some relevant studies, and this should therefore be a consideration before drawing any conclusions from the results from this review. Finally, because of the considerable size of the topic of healthy eating and diabetes, the decision was made to focus on the different mechanisms of the dietary interventions (rather than on differences between prescribed diets), and therefore, studies were included in this review only if they contained some type of individualized assessment. A further systematic review of interventions comparing specific diets would be useful in providing important additional knowledge of the self-care behavior of healthy eating for people with diabetes.

In conclusion, this review demonstrates that although numerous studies have been designed to examine the efficacy of dietary interventions in changing healthy eating behaviors, the diversity of intervention types and outcome measures makes them difficult to compare. The quality of future research would be improved by ensuring that sample sizes are adequate to give sufficient power and also by ensuring that outcomes in all 4 areas—learning, behavior, clinical improvement, and health status^2,6—are measured. Quality would also be improved by ensuring studies last over a period of months rather than weeks, to ensure that potential changes in both behavior and clinical outcomes are detected. Finally, the fact that some outcome variables were less likely to show significant changes in studies of longer duration would indicate that there is clearly a need to design interventions that focus on not only the initiation of behavior change but also, importantly, its maintenance.

	Appendix List of Articles Retrieved Showing Those That Were Included in the Final Review and Reasons for Exclusion of Those That Were Not Included

Top Abstract Methods Results Appendix List of Articles... References

 
^aAgurs-Collins TD, Kumanyika SK, Ten Have TR, Adams-Campbell LL. A randomized controlled trial of weight reduction and exercise for diabetes management in older African-American subjects. Diabetes Care. 1997;20:1503-1511.

^ceAlbisser AM, En Chao SC, Parson ID, Sperlich M. Information technology and home glucose clamping. Diabetes Technol Ther. 2001;3:377-386.

^ceAlbisser AM, Harris RI, Sakkal S, Parson ID, Chao SC. Diabetes intervention in the information age [published correction appears in Med Inf (Lond). 1997;22:205]. Med Inform (Lond). 1996;21:297-316.

^cdeiAnderson EJ, Richardson M, Castle G, et al. Nutrition interventions for intensive therapy in the Diabetes Control and Complications Trial. The DCCT Research Group. J Am Diet Assoc. 1993; 93:1104.

^aAnderson-Loftin W, Barnett S, Bunn P, et al. Soul food light: culturally competent diabetes education. Diabetes Educ. 2005;31:555-563.

^aAnderson-Loftin W, Barnett S, Sullivan P, Bunn PS, Tavakoli A. Culturally competent dietary education for southern rural African Americans with diabetes... including commentary by Melkus GD. Diabetes Educ. 2002;28:245-257.

^mAsh S, Reeves MM, Yeo S, Morrison G, Carey D, Capra S. Effect of intensive dietetic interventions on weight and glycaemic control in overweight men with type II diabetes: a randomised trial. Int J Obes Relat Metab Disord. 2003;27:797-802.

^eBarr Mazzuca K, Farris NA, Mendenhall J, Stoupa RA. Demonstrating the added value of community health nursing for clients with insulin-dependent diabetes. J Community Health Nurs. 1997;14:211-224.

^emBlaak EE, Glatz JF, Saris WH. Increase in skeletal muscle fatty acid binding protein (FABPC) content is directly related to weight loss and to changes in fat oxidation following a very low calorie diet. Diabetologia. 2001;44:2013-2017.

^eBourn DM, Mann JI, McSkimming BJ, Waldron MA, Wishart JD. Impaired glucose tolerance and NIDDM: does a lifestyle intervention program have an effect? Diabetes Care. 1994;17:1311-1319.

^bBowerman S, Bellman M, Saltsman P. Implementation of a primary care physician network obesity management program. Obes Res. 2001;9:321S-325S.

^cdhmCarlson A, Rosenqvist U. Diabetes care organization, process, and patient outcomes: effects of a diabetes control program. Diabetes Educ. 1991;17:42-48.

^eClark M, Hampson SE, Avery L, Simpson R. Effects of a brief tailored intervention on the process and predictors of lifestyle behaviour change in patients with type 2 diabetes. Psychol Health Med. 2004;9:440-449.

^eDavies MJ, Metcalfe J, Day JL, Grenfell A, Hales CN, Gray IP. Effect of sulphonylurea therapy on plasma insulin, intact and 32/33 split proinsulin in subjects with type 2 diabetes mellitus. Diabet Med. 1994;11:293-298.

^eDavies MJ, Metcalfe J, Day JL, Grenfell A, Hales CN, Gray IP. Improved beta cell function, with reduction in secretion of intact and 32/33 split proinsulin, after dietary intervention in subjects with type 2 diabetes. Diabet Med. 1994;11:71-78.

^mDullaart RP, Hoogenberg K, Riemens SC, et al. Cholesteryl ester transfer protein gene polymorphism is a determinant of HDL cholesterol and of the lipoprotein response to a lipid-lowering diet in type 1 diabetes. Diabetes. 1997;46:2082-2087.

^eEriksson KF, Lindgarde F. Prevention of type 2 (non-insulin-dependent) diabetes mellitus by diet and physical exercise. The 6-year Malmo feasibility study. Diabetologia. 1991;34:891-898.

^mEsmaillzadeh A, Tahbaz F, Gaieni I, Alavi-Majd H, Azadbakht L. Concentrated pomegranate juice improves lipid profiles in diabetic patients with hyperlipidemia. J Med Food. 2004;7:305-308.

^aGaede P, Beck M, Vedel P, Pedersen O. Limited impact of lifestyle education in patients with type 2 diabetes mellitus and microalbuminuria: results from a randomized intervention study. Diabet Med. 2001;18:104-108

^dhGalasso P, Amend A, Melkus GD, Nelson GT. Barriers to medical nutrition therapy in black women with type 2 diabetes mellitus. Diabetes Educ. 2005;31:719-725.

^emGamsu DS, Sutton MS, Bennett L, Ward JD. The development of a psychoeducational group intervention for overweight women with type 2 diabetes mellitus: a service evaluation. Practical Diabetes International. 2002;19:43-50.

^aGlasgow RE, La Chance PA, Toobert DJ, Brown J, Hampson SE, Riddle MC. Long term effects and costs of brief behavioural dietary intervention for patients with diabetes delivered from the medical office. Patient Educ Couns. 1997;32:175-184.

^aGlasgow RE, Toobert DJ. Brief, computer-assisted diabetes self-management counseling: effects on behaviour, physiologic outcomes, and quality of life. Med Care. 2000;38:1062-1073.

^aoGlasgow RE, Toobert DJ, Barrera M, Strycker LA. The Chronic Illness Resources Survey: cross-validation and sensitivity to intervention. Health Educ Res. 2005;20:402-409.

^aoGlasgow RE, Toobert DJ, Hampson SE. Effects of a brief office-based intervention to facilitate diabetes dietary self-management. Diabetes Care. 1996;19:835-842.

^cdGlasgow RE, Toobert DJ, Hampson SE, Noell JW. A brief office-based intervention to facilitate diabetes dietary self-management. Health Educ Res. 1995;10:467-478.

^aoGlasgow RE, Toobert DJ, Hampson SE, Strycker LA. Implementation, generalization and long-term results of the "choosing well" diabetes self-management intervention. Patient Educ Couns. 2002;48:115-122.

^fGlasgow RE, Toobert DJ, Mitchell DL, Donnelly JE, Calder D. Nutrition education and social learning interventions for type II diabetes. Diabetes Care. 1989;12:150-152.

^cdehmGulliford MC, Mahabir D. A five-year evaluation of intervention in diabetes care in Trinidad and Tobago. Diabet Med. 1999;16:939-945.

^aHalford WK, Goodall TA, Nicholson JM. Diet and diabetes (II): a controlled trial of problem solving to improve dietary self-management in patients with insulin dependent diabetes. Psychol Health. 1997;12:231-238.

^bHampson SE, Glasgow RE, Strycker LA. Beliefs versus feelings: a comparison of personal models and depression for predicting multiple outcomes in diabetes. Br J Health Psychol. 2000;5:27-40.

^cmHanefield M, Fischer S, Schmechel H, et al; Diabetes Intervention Study. Multi-intervention trial in newly diagnosed NIDDM. Diabetes Care. 1991;14:308-317.

^emHarder H, Dinesen B, Astrup A. The effect of a rapid weight loss on lipid profile and glycemic control in obese type 2 diabetic patients. Int J Obes Relat Metab Disord. 2004;28:180-182.

^bHaynes RB, Kris-Etherton P, McCarron DA, et al. Nutritionally complete prepared meal plan to reduce cardiovascular risk factors: a randomized clinical trial. J Am Diet Assoc. 1999;99:1077-1083.

^aJones H, Edwards L, Vallis TM, et al. Changes in diabetes self-care behaviors make a difference in glycemic control: the Diabetes Stages of Change (DiSC) Study. Diabetes Care. 2003;26:732-737.

^fKaplan RM, Wilson DK, Hartwell SL, Merino KL, Wallace JP. Prospective evaluation of HDL cholesterol changes after diet and physical conditioning for programs for patients with type II diabetes mellitus. Diabetes Care. 1985;8:343-348.

^fKarlstrom B, Nydahl M, Vessby B. Dietary habits and effects of dietary advice in patients with type 2 diabetes: results from a one-year intervention study. Eur J Clin Nutr. 1989;43:59-68.

^aKeyserling TC, Samuel-Hodge CD, Ammerman AS, et al. A randomized trial of an intervention to improve self-care behaviors of African-American women with type 2 diabetes: impact on physical activity. Diabetes Care. 2002;25:1576-1583.

^bejKnobler H, Schattner A, Zhornicki T, et al. Fatty liver—an additional and treatable feature of the insulin resistance syndrome. QJM. 1999;92:73-79.

^aLaitinen JH, Ahola IE, Sarkkinen ES, Winberg RL, Harmaakorpi-Iivonen PA, Uusitupa MI. Impact of intensified dietary therapy on energy and nutrient intakes and fatty acid composition of serum lipids in patients with recently diagnosed non-insulin-dependent diabetes mellitus. J Am Diet Assoc. 1993;93:276-283.

^eLomasky SJ, D'Eramo G, Shamoon H, Fleischer N. Relationship of insulin secretion and glycemic response to dietary intervention in non-insulin-dependent diabetes. Arch Intern Med. 1990;150: 169-172.

^mLu ZX, Walker KZ, Muir JG, O'Dea K. Arabinoxylan fibre improves metabolic control in people with type II diabetes. Eur J Clin Nutr. 2004;58:621-628.

^bmMarkovic TP, Campbell LV, Balasubramanian S, et al. Beneficial effect on average lipid levels from energy restriction and fat loss in obese individuals with or without type 2 diabetes. Diabetes Care. 1998;21:695-700.

^bMau MK, Glanz K, Severino R, Grove JS, Johnson B, Curb JD. Mediators of lifestyle behavior change in native Hawaiians: initial findings from the Native Hawaiian Diabetes Intervention Program. Diabetes Care. 2001;24:1770-1775.

^bMcCarron DA, Oparil S, Chait A, et al. Nutritional management of cardiovascular risk factors: a randomized clinical trial. Arch Intern Med. 1997;157:169-177.

^aMcKay HG, Glasgow RE, Feil EG, Boles SM, Barrera M. Internet-based diabetes self-management and support: initial outcomes from the Diabetes Network Project. Rehabil Psychol. 2002;47:31-48.

^eMendez FJ, Belendez M. Effects of a behavioral intervention on treatment adherence and stress management in adolescents with IDDM. Diabetes Care. 1997;20:1370-1375.

^bMetz JA, Kris-Etherton PM, Morris CD, et al. Dietary compliance and cardiovascular risk reduction with a prepared meal plan compared with a self-selected diet. Am J Clin Nutr. 1997;66:373-385.

^aMilne RM, Mann JI, Chisholm AW, Williams SM. Long-term comparison of three dietary prescriptions in the treatment of NIDDM. Diabetes Care. 1994;17:74-80.

^aNeyestani TR, Alipour-Birgani R, Siassi F, Rajayi M, Djalali M, Mohamadi M. Glycemic optimization may reduce lipid peroxidation independent of weight and blood lipid changes in type 2 diabetes mellitus. Diabetes Nutr Metab. 2004;17:275-279.

^mNicholson AS, Sklar M, Barnard MD, Gore S, Sullivan R, Browning S. Toward improved management of NIDDM: a randomized, controlled, pilot intervention using a low fat, vegetarian diet. Prev Med. 1999;29:87-91.

^aPerry TL, Mann JI, Lewis-Barned NJ, Duncan AW, Waldron MA, Thompson C. Lifestyle intervention in people with insulin-dependent diabetes mellitus (IDDM). Eur J Clin Nutr. 1997;51:757-763.

^dPick ME, Hawrysh ZJ, Gee MI, Toth E, Garg ML, Hardin RT. Oat bran concentrate bread products improve long-term control of diabetes: a pilot study. J Am Diet Assoc. 1996;96:1254-1261.

^ePijls LT, de Vries H, van Eijk JT, Donker AJ. Adherence to protein restriction in patients with type 2 diabetes mellitus: a randomized trial. Eur J Clin Nutr. 2000;54:347-352.

^ePijls LT, de Vries H, van Eijk JT, Donker AJ. Protein restriction, glomerular filtration rate and albuminuria in patients with type 2 diabetes mellitus: a randomized trial. Eur J Clin Nutr. 2002;56: 1200-1207.

^cdhmPinelli L, Mormile R, Gonfiantini E, et al. Recommended dietary allowances (RDA) in the dietary management of children and adolescents with IDDM: an unfeasible target or an achievable cornerstone? J Pediatr Endocrinol Metab. 1998;11:335-346.

^mPi-Sunyer FX, Maggio CA, McCarron DA, et al. Multicenter randomized trial of a comprehensive prepared meal program in type 2 diabetes. Diabetes Care. 1999;22:191-197.

^mQvigstad E, Mostad IL, Bjerve KS, Grill VE. Acute lowering of circulating fatty acids improves insulin secretion in a subset of type 2 diabetes subjects. Am J Physiol Endocrinol Metab. 2003; 284:E129-E137.

^bRacette SB, Weiss EP, Obert KA, Kohrt WM, Holloszy JO. Modest lifestyle intervention and glucose tolerance in obese African Americans. Obes Res. 2001;9:348-355.

^bRamachandran A, Snehalatha C, Shobana R, Vidyavathi P, Vijay V. Influence of life style factors in development of diabetes in Indians—scope for primary prevention. J Assoc Physicians India. 1999;47:761-763.

^emReyna NY, Cano C, Bermúdez VJ, et al. Sweeteners and beta-glucans improve metabolic and anthropometrics variables in well controlled type 2 diabetic patients. Am J Ther. 2003; 10:438-443.

^mRosenfalck AM, Almdal T, Viggers S, Madsbad S, Hilsted J. A low-fat diet improves peripheral insulin sensitivity in patients with type 1 diabetes. Diabet Med. 2006;23:384-392.

^ceSamann A, Muhlhauser I, Bender R, Kloos Ch, Muller UA. Glycaemic control and severe hypoglycaemia following training in flexible, intensive insulin therapy to enable dietary freedom in people with type 1 diabetes: a prospective implementation study. Diabetologia. 2005;48:1965-1970.

^cSamaras K, Ashwell S, Mackintosh AM, Fleury AC, Campbell LV, Chisholm DJ. Will older sedentary people with non-insulin-dependent diabetes mellitus start exercising? A health promotion model. Diabetes Res Clin Pract. 1997;37:121-128.

^aSargrad KR, Homko C, Mozzoli M, Boden G. Effect of high protein vs. high carbohydrate intake on insulin sensitivity, body weight, hemoglobin A1c, and blood pressure in patients with type 2 diabetes mellitus. J Am Diet Assoc. 2005;105:573-80.

^emSchlundt DG, Rea M, Hodge M, et al. Assessing and overcoming situational obstacles to dietary adherence in adolescents with IDDM. J Adolesc Health. 1996;19:282-288.

^dhSchlundt DG, Stetson BA, Plant DD. Situation taxonomy and behavioral diagnosis using prospective self-monitoring data: application to dietary adherence in patients with type 1 diabetes. J Psychopathol Behav Assess. 1999;21:19-36.

^bceSchulz AJ, Zenk S, Odoms-Young A, et al. Healthy eating and exercising to reduce diabetes: exploring the potential of social determinants of health frameworks within the context of community-based participatory diabetes prevention. Am J Public Health. 2005;95:645-651.

^eSinha R, Ismail A, Herbert S, Hoyte R, Gill G. Effect of diet and fenofibrate on lipid and glycaemic control in type 2 diabetes. Practical Diabetes International. 2001;18:269-273.

^emSpannagl M, Drummer C, Froschl H. Plasmatic factors of haemostasis remain essentially unchanged except for PAI activity during n-3 fatty acid intake in type I diabetes mellitus. Blood Coagul Fibrinolysis. 1991;2:259-265.

^beStern L, Iqbal N, Seshadri P, et al. The effects of low-carbohydrate versus conventional weight loss diets in severely obese adults: one-year follow-up of a randomized trial. Ann Intern Med. 2004;140:778-785.

^fSulway M, Tupling H, Webb K, Harris G. New techniques for changing compliance in diabetes. Diabetes Care. 1980;3:108-111.

^aTan AS, Yong LS, Wan S, Wong ML. Patient education in the management of diabetes mellitus. Singapore Med J. 1997;38:156-160.

ⁿTapsell LC, Hokman A, Sebastiao A, et al. The impact of usual dietary patterns, selection of significant foods and cuisine choices on changing dietary fat under "free living" conditions. Asia Pac J Clin Nutr. 2004;13:86-91.

^mThomsen C, Rasmussen OW, Ingerslev J, Hermansen K. Plasma levels of von Willebrand factor in non-insulin-dependent diabetes mellitus are influenced by dietary monounsaturated fatty acids. Thromb Res. 1995;77:347-356.

^aoToobert DJ, Strycker LA, Glasgow RE, Barrera M, Angell K. Effects of the Mediterranean Lifestyle Program on multiple risk behaviors and psychosocial outcomes among women at risk for heart disease. Ann Behav Med. 2005;29:128-137.

^cToobert DJ, Strycker LA, Glasgow RE, Barrera M, Bagdade JD. Enhancing support for health behavior change among women at risk for heart disease. The Mediterranean Lifestyle Trial. Health Educ Res. 2002;17:574-585.

^dTsang MW, Mok M, Kam G, et al. Improvement in diabetes control with a monitoring system based on a hand-held, touch-screen electronic diary. J Telemed Telecare. 2001;7:47-50.

^mTwo Feathers J, Kieffer EC, Palmisano G, et al. Racial and Ethnic Approaches to Community Health (REACH) Detroit Partnership: improving diabetes-related outcomes among African American and Latino adults. Am J Public Health. 2005;95:1552-1560.

^eUusitupa MIJ. Early lifestyle intervention in patients with non-insulin-dependent diabetes mellitus and impaired glucose tolerance. Ann Med. 1996;28:445-449.

^eVanninen E, Uusitupa M, Lansimies E, Siitonen O, Laitinen J. Effect of metabolic control on autonomic function in obese patients with newly diagnosed type 2 diabetes. Diabet Med. 1993;10:66-73.

^aVazquez IM, Millen B, Bissett L, Levenson SM, Chipkin SR. Buena alimentacion, buena salud: a preventive nutrition intervention in Caribbean Latinos with type 2 diabetes. Am J Health Promot. 1998;13:116-119.

^fWeisweiler P, Drosner M, Schwandt P. Dietary effects on very low-density lipoproteins in type 2 (non-insulin-dependent) diabetes mellitus. Diabetologia. 1982;23:101-103.

^aWhittemore R, Chase S, Mandle CL, Roy C. The content, integrity, and efficacy of a nurse coaching intervention in type 2 diabetes. Diabetes Educ. 2001;27:887-898.

^aWhittemore R, Melkus GD, Sullivan A, Grey M. A nurse-coaching intervention for women with type 2 diabetes. Diabetes Educ. 2004;30:795-804.

^eYong A, Power E, Gill G. Improving glycaemic control of insulin-treated diabetic patients—a structured audit of specialist nurse intervention. J Clin Nurs. 2002;11:773-776.

	Acknowledgments

 
We would like to express our thanks to Professor Jim Fain and Dr Sue Boren for kindly helping us to retrieve a number of the articles.

	FOOTNOTES

 
^g Has not been peer reviewed.

^k Is not in English.

^l Does not have human participants.

^* References ¹⁶, ¹⁸, ¹⁹, ²⁰, ²³, ²⁴, ²⁶, ²⁷, ³².

References ¹⁷, ²¹, ²³, ²⁵, ²⁷, ²⁹, ³⁰, ³³-³⁵.

^a Included article.

^c Does not report an intervention to promote dietary change.

^e At least 1 outcome does not measure dietary behavior.

^d Does not measure at 2 time points, preintervention and postintervention.

ⁱ Is a review/advice/descriptive article/conference proceeding/lecture.

^m Does not include some type of individualized assessment, development of a plan, and periodic reassessment between instructor(s) and participant when directing the selection of appropriate education materials and intervention.

^b Results not provided for participants diagnosed with type 1 or type 2 diabetes.

^h Is a cross-sectional study.

^o Reports results from the same sample as another article already included in the review and hence was excluded from the final analysis.

^f Article not published between 1990 and present.

^j Describes an intervention that is targeted toward a group that is so specific that the extent to which the intervention can be applied to a general diabetic population is reduced.

ⁿ Article was unobtainable.

	References

Top Abstract Methods Results Appendix List of Articles... References

 

1. American Diabetes Association. Nutrition recommendations and interventions for diabetes: a position statement of the American Diabetes Association. Diabetes Care.2007; 30:S48 -S65.[Free Full Text]
2. Mulcahy K, Maryniuk M, Peeples M, et al. Diabetes self-management education core outcomes measures. Diabetes Educ.2003; 29:768 -803.[Free Full Text]
3. Wing RR, Goldstein MG, Acton KJ, et al. Behavioral science research in diabetes: lifestyle changes related to obesity, eating behavior, and physical activity. Diabetes Care.2001; 24:117 -123.[Abstract/Free Full Text]
4. Marrero DG, Peyrot M, Garfield S. Promoting behavioral science research in diabetes. Diabetes Care.2001; 24:1 -2.[Free Full Text]
5. Anderson RM, Funnell M. The Art of Empowerment: Stories and Strategies for Diabetes Educators. Alexandria, VA: American Diabetes Association; 2000.
6. American Association of Diabetes Educators. Position statement: standards for outcomes measurement of diabetes self-management education.Diabetes Educ . 2003;29:804 -816.[Free Full Text]
7. American Diabetes Association. Nutrition principles and recommendations in diabetes. Diabetes Care.2004; 27:S36 -S46.[CrossRef][Medline] [Order article via Infotrieve]
8. Mensing C, Boucher J, Cypress M, et al. National standards for diabetes self-management education. Diabetes Care.2006; 29:S78 -S85.[Free Full Text]
9. American Diabetes Association. Introduction. Diabetes Care. 2006;29:S1 -S2.[Free Full Text]
10. Hampson SE, Glasgow RE, Strycker LA. Beliefs versus feelings: a comparison of personal models and depression for predicting multiple outcomes in diabetes. Br J Health Psychol.2000; 5:27 -40.[CrossRef][Web of Science]
11. Pijls LT, de Vries H, van Eijk JT, Donker AJ. Adherence to protein restriction in patients with type 2 diabetes mellitus: a randomized trial.Eur J Clin Nutr . 2000;54:347 -352.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
12. Glasgow RE, Toobert DJ, Barrera M, Strycker LA. The Chronic Illness Resources Survey: cross-validation and sensitivity to intervention.Health Educ Res . 2005;20:402 -409.[Abstract/Free Full Text]
13. Glasgow RE, Toobert DJ, Hampson SE. Effects of a brief office-based intervention to facilitate diabetes dietary self-management. Diabetes Care. 1996;19:835 -842.[Abstract]
14. Glasgow RE, Toobert DJ, Hampson SE, Strycker LA. Implementation, generalization and long-term results of the "choosing well" diabetes self-management intervention. Patient Educ Couns.2002; 48:115 -122.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
15. Two Feathers J, Kieffer EC, Palmisano G, et al. Racial and Ethnic Approaches to Community Health (REACH) Detroit Partnership: improving diabetes-related outcomes among African American and Latino adults. Am J Public Health. 2005;95:1552 -1560.[Abstract/Free Full Text]
16. Keyserling TC, Samuel-Hodge CD, Ammerman AS, et al. A randomized trial of an intervention to improve self-care behaviors of African-American women with type 2 diabetes: impact on physical activity. Diabetes Care. 2002;25:1576 -1583.[Abstract/Free Full Text]
17. Tan AS, Yong LS, Wan S, Wong ML. Patient education in the management of diabetes mellitus. Singapore Med J.1997; 38:156 -160.[Medline] [Order article via Infotrieve]
18. Laitinen JH, Ahola IE, Sarkkinen ES, Winberg RL, Harmaakorpi-Iivonen PA, Uusitupa MI. Impact of intensified dietary therapy on energy and nutrient intakes and fatty acid composition of serum lipids in patients with recently diagnosed non-insulin-dependent diabetes mellitus.J Am Diet Assoc . 1993;93:276 -283.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
19. Sargrad KR, Homko C, Mozzoli M, Boden G. Effect of high protein vs. high carbohydrate intake on insulin sensitivity, body weight, hemoglobin A1c, and blood pressure in patients with type 2 diabetes mellitus. J Am Diet Assoc. 2005;105:573 -580.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
20. McKay HG, Glasgow RE, Feil EG, Boles SM, Barrera M. Internet-based diabetes self-management and support: initial outcomes from the Diabetes Network Project. Rehabil Psychol.2002; 47:31 -48.[Medline] [Order article via Infotrieve]
21. Perry TL, Mann JI, Lewis-Barned NJ, Duncan AW, Waldron MA, Thompson C. Lifestyle intervention in people with insulin-dependent diabetes mellitus (IDDM). Eur J Clin Nutr. 1997;51 : 757-763.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
22. Halford WK, Goodall TA, Nicholson JM. Diet and diabetes (II): a controlled trial of problem solving to improve dietary self-management in patients with insulin dependent diabetes. Psychol Health.1997; 12:231 -238.[Web of Science]
23. Agurs-Collins TD, Kumanyika SK, Ten Have TR, Adams-Campbell LL. A randomized controlled trial of weight reduction and exercise for diabetes management in older African-American subjects. Diabetes Care.1997; 20:1503 -1511.[Abstract]
24. Anderson-Loftin W, Barnett S, Sullivan P, Bunn PS, Tavakoli A. Culturally competent dietary education for southern rural African Americans with diabetes... including commentary by Melkus GD. Diabetes Educ. 2002;28:245 -257.[Abstract/Free Full Text]
25. Jones H, Edwards L, Vallis TM, et al. Changes in diabetes self-care behaviors make a difference in glycemic control: the Diabetes Stages of Change (DiSC) Study. Diabetes Care.2003; 26:732 -737.[Abstract/Free Full Text]
26. Milne RM, Mann JI, Chisholm AW, Williams SM. Long-term comparison of three dietary prescriptions in the treatment of NIDDM. Diabetes Care. 1994;17:74 -80.[Abstract]
27. Neyestani TR, Alipour-Birgani R, Siassi F, Rajayi M, Djalali M, Mohamadi M. Glycemic optimization may reduce lipid peroxidation independent of weight and blood lipid changes in type 2 diabetes mellitus. Diabetes Nutr Metab. 2004;17:275 -279.[Web of Science][Medline] [Order article via Infotrieve]
28. Whittemore R, Chase S, Mandle CL, Roy C. The content, integrity, and efficacy of a nurse coaching intervention in type 2 diabetes.Diabetes Educ . 2001;27:887 -898.[Abstract/Free Full Text]
29. Glasgow RE, Toobert DJ. Brief, computer-assisted diabetes self-management counseling: effects on behaviour, physiologic outcomes, and quality of life. Med Care.2000; 38:1062 -1073.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
30. Gaede P, Beck M, Vedel P, Pedersen O. Limited impact of lifestyle education in patients with type 2 diabetes mellitus and microalbuminuria: results from a randomized intervention study. Diabet Med.2001; 18:104 -108.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
31. Whittemore R, Melkus GD, Sullivan A, Grey M. A nurse-coaching intervention for women with type 2 diabetes. Diabetes Educ.2004; 30:795 -804.[Free Full Text]
32. Anderson-Loftin W, Barnett S, Bunn P, et al. Soul food light: culturally competent diabetes education. Diabetes Educ.2005; 31:555 -563.[Abstract/Free Full Text]
33. Glasgow RE, La Chance PA, Toobert DJ, Brown J, Hampson SE, Riddle MC. Long-term effects and costs of brief behavioural dietary intervention for patients with diabetes delivered from the medical office. Patient Educ Couns. 1997;32:175 -184.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
34. Vazquez IM, Millen B, Bissett L, Levenson SM, Chipkin SR. Buena alimentacion, buena salud: a preventive nutrition intervention in Caribbean Latinos with type 2 diabetes. Am J Health Promot.1998; 13:116 -119.[Web of Science][Medline] [Order article via Infotrieve]
35. Toobert DJ, Strycker LA, Glasgow RE, Barrera M, Angell K. Effects of the Mediterranean Lifestyle Program on multiple risk behaviors and psychosocial outcomes among women at risk for heart disease. Ann Behav Med. 2005;29:128 -137.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
36. Kumanyika SK, Van Horn L, Bowen D, et al. Maintenance of dietary behavior change. Health Psychol.2000; 19:42 -56.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
37. Clark-Carter D. Quantitative Psychological Research: A Student's Handbook. Hove, UK: Psychology Press;2004 .

The Diabetes Educator, Vol. 33, No. 6, 931-959 (2007)
DOI: 10.1177/0145721707308408


CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				American Association of Diabetes Educators AADE Guidelines for the Practice of Diabetes Self-Management Education and Training (DSME/T) The Diabetes Educator, November 1, 2009; 35(3_suppl): 85S - 107S. [Full Text] [PDF]

				M. J. Franz Diabetes and Healthy Eating: A Commentary The Diabetes Educator, November 1, 2007; 33(6): 960 - 961. [Full Text] [PDF]

This Article Abstract Free Full Text (Free PDF) Free Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to Saved Citations Download to citation manager Request Permissions Request Reprints Add to My Marked Citations Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (1) Citing Articles via Google Scholar Citing Articles via Scopus Google Scholar Articles by Povey, R. C. Articles by Clark-Carter, D. Search for Related Content PubMed PubMed Citation Articles by Povey, R. C. Articles by Clark-Carter, D. Social Bookmarking                         What's this?