The dramatic rise in the prevalence of obesity and type 2 diabetes is clearly a strain on health care systems globally (1). Recently published guidelines on obesity management indicate that a sustained weight loss of 3–5% is likely to result in clinically meaningful reductions in triglycerides, blood glucose, and A1C (2). However, the 2013 Obesity Guidelines noted achieving more weight loss (∼10%) will also reduce blood pressure, improve LDL cholesterol and HDL cholesterol, and reduce the need for medications required to control cardiometabolic risk (2). The Look AHEAD (Actions for Health in Diabetes) trial provides evidence, specific to diabetes, that an intensive lifestyle intervention can achieve a 5–10% weight loss and can improve sleep apnea quality-of-life indices, achieve cardiometabolic biomarker improvements, and reduce need for medications (3–6). These benefits are emphasized in the 2013 Obesity Guidelines (2). Challenges to providing an intensive intervention in primary care practice setting, as required to achieve the suggested outcomes, include the lack of staffing and resources required to adhere to the schedule for the intensive follow-up. In this regard, telephone contact, along with other technologies that have the potential to provide frequent contact with a lower staff and patient burden, is worth exploring. The specific question, however, is whether the required weight loss of 3–5% (necessary to result in clinically meaningful benefits) can be achieved by telephone intervention. In this issue, a randomized study by Eakin et al. (7), Living Well With Diabetes (LWWD), is reported and attempts to address that question.
The LWWD study was a pragmatic randomized controlled trial that evaluated the effectiveness of a telephone-delivered behavioral weight loss and physical activity intervention in primary care clinics. Patients with type 2 diabetes and a BMI of ≥25 kg/m2 were randomized to telephone counseling (n = 151) or usual care (n = 151) (7,8). Relative to the usual-care group, telephone counseling participants achieved slightly greater improvements in weight loss (1.4 vs. 0.3% of baseline body weight), 42% greater increase in moderate-to-vigorous intensity physical activity, and 2.7-fold greater improvement in diet quality (7). Despite a greater reduction in waist circumference, the A1C and other cardiometabolic markers were not significantly different between the treatment arms at either 18 or 24 months. Although the outcomes did not show a significant change/deterioration over the maintenance period, only the intervention effect for increased moderate-to-vigorous intensity physical activity remained statistically significant at 24 months. Eakin et al. (7) concluded that their LWWD telephone intervention achieved modest improvements in weight loss and behavior change over usual care, but did not achieve changes in cardiometabolic markers. These findings raise questions about the clinical utility, scalability, and sustainability of telephone interventions.
The strengths of the LWWD telephone intervention study included the low intervention burden for health systems and patients, the rigorous collection of outcomes and process measure, and the positive benefits with regard to achieving a modest weight loss and lifestyle improvements. Weaknesses of the study include not achieving the recommended level of weight loss to improve long-term metabolic improvement and problems with providing the full treatment dose (planned number of consultations) in the telephone intervention. However, when examining weight-loss studies, the RE-AIM (Reach, Effectiveness, Adoption, Implementation and Maintenance) evaluation framework can provide insights for translating research findings into clinical practice (9).
As shown in Table 1, the study “reached” a high proportion of eligible patients in an efficient manner, as the primary care practices “adopted” recruitment procedures that used the electronic medical records (EMR) and provider review. While the telephone intervention achieved a 1.4% weight loss (at 18 months), the investigators had a weight-loss goal of 5–10%, which is the level recommended to substantially improve A1C and other risk markers. The lack of metabolic effect may be due to the patients being close to the recommended targets at baseline as well as the modest change in weight. The intervention team was unable to “implement” the intervention as planned to half of the patients, who received less than 75% of the 27 intended telephone counseling sessions. The lifestyle and weight changes reported at 18 months were “maintained” at 24 months if evaluated from the perspective that the mean values for selected variables did not differ between the 18- and 24-month time points. However, the 95% CI was quite wide and the sample size may be too small to interpret that finding.
Study findings: applying RE-AIM evaluation framework
Reach |
The recruitment was integrated into the practice using EMR to indentify 1,407 potentially eligible patients. Physicians screened out (n = 409) patients with potential contraindications and the practice sent letters inviting patients (n = 908) to join the study. About a third (n = 302) joined the study. The patients in the study reflected the underlying population with diabetes. However, the vast majority were Caucasian, reflecting the demographic of the local community but not the global population with diabetes. |
Effectiveness |
Telephone counseling participants achieved slightly greater improvements in weight loss (1.4 vs. 0.3% of baseline body weight), 42% greater increase in moderate-to-vigorous intensity physical activity, and 2.7-fold greater improvement in diet quality at the end of the 18-month intervention. The intervention did not significantly improve A1C or other cardiometabolic biomarkers. However, metabolic control was close to the target at baseline, e.g., mean A1C level of 7.1% (54 mmol/mol). The supplementary tables and figures provide a more in-depth examination related to the proportion of participants achieving interventions goals. |
Adoption |
The selected primary care (n = 9) practices “adopted” recruitment procedures that used the EMR to identify patients who would be eligible on the basis of their BMI and A1C levels. Primary care providers reviewed their respective lists of EMR-identified patients for contraindications to participating in the study intervention. |
Implementation |
Intervention delivery was monitored via patient feedback regarding content of randomly selected telephone counseling session calls. Call attempts, completions, and duration were tracked in the trial database. However, only about half of the patients completed at least 21 or more of the planned 27 telephone counseling sessions. |
Maintenance |
The small effect size and wide variability (SD of change scores) makes assessment of maintenance challenging. Nonetheless, the modest changes achieved at the end of the 18-month intervention may have been maintained as there was no significant decline between 18 and 24 months. However, only the increase in physical activity was significantly improved from the baseline level. |
Reach |
The recruitment was integrated into the practice using EMR to indentify 1,407 potentially eligible patients. Physicians screened out (n = 409) patients with potential contraindications and the practice sent letters inviting patients (n = 908) to join the study. About a third (n = 302) joined the study. The patients in the study reflected the underlying population with diabetes. However, the vast majority were Caucasian, reflecting the demographic of the local community but not the global population with diabetes. |
Effectiveness |
Telephone counseling participants achieved slightly greater improvements in weight loss (1.4 vs. 0.3% of baseline body weight), 42% greater increase in moderate-to-vigorous intensity physical activity, and 2.7-fold greater improvement in diet quality at the end of the 18-month intervention. The intervention did not significantly improve A1C or other cardiometabolic biomarkers. However, metabolic control was close to the target at baseline, e.g., mean A1C level of 7.1% (54 mmol/mol). The supplementary tables and figures provide a more in-depth examination related to the proportion of participants achieving interventions goals. |
Adoption |
The selected primary care (n = 9) practices “adopted” recruitment procedures that used the EMR to identify patients who would be eligible on the basis of their BMI and A1C levels. Primary care providers reviewed their respective lists of EMR-identified patients for contraindications to participating in the study intervention. |
Implementation |
Intervention delivery was monitored via patient feedback regarding content of randomly selected telephone counseling session calls. Call attempts, completions, and duration were tracked in the trial database. However, only about half of the patients completed at least 21 or more of the planned 27 telephone counseling sessions. |
Maintenance |
The small effect size and wide variability (SD of change scores) makes assessment of maintenance challenging. Nonetheless, the modest changes achieved at the end of the 18-month intervention may have been maintained as there was no significant decline between 18 and 24 months. However, only the increase in physical activity was significantly improved from the baseline level. |
What are the “lessons learned” from this trial? Using EMR preliminary screening combined with primary provider review can efficiently identify patients for a diabetes weight-loss intervention. Delivering a weight-loss intervention by telephone is subject to “session attendance” challenges even though the participant time commitment would be reduced by not having to travel. The mean weight changes do not provide the full picture needed to address translation to practice. The proportion of patients achieving the goal of ≥5% weight loss at 24 months was threefold higher in the telephone intervention than in the usual-care group (15.9 vs. 5.1%). Evaluating the predictors of achieving the ≥5% weight-loss goal was beyond the scope of the study by Eakin et al. (7), and the study did not address cost-effectiveness.
A systematic review of telephone-delivered physical activity and dietary behavior interventions concluded that telephone interventions achieve behavior change, but dissemination research indicates that completing all of the telephone sessions and retention are challenging (10,11). Other studies have concluded that telephone intervention outcomes can be improved by increasing the dose (completing more telephone sessions) (12), using group conference calls (13), combining telephone with intervention modalities (14), text messaging for support and to facilitate self-monitoring (15–18), and evaluating the intervention process via applying the RE-AIM framework (9). Future research needs to address potential capacity of telephone-based interventions to enhance the effectiveness of weight-loss interventions as well as to expand the availability of weight-loss interventions.
See accompanying article, p. 2177.
Article Information
Duality of Interest. No potential conflicts of interest relevant to this article were reported.