Awareness of Prediabetes Status and Subsequent Health Behavior, Body Weight, and Blood Glucose Levels

Ibiye Owei, MD, MPH; Nkiru Umekwe, MBBS; Fatoumatta Ceesay, BS; Samuel Dagogo-Jack, MD


J Am Board Fam Med. 2019;32(1):20-27. 

In This Article


In the present report, we observed that POP-ABC participants who were notified that they had developed incident prediabetes showed decreases in FPG and 2hrPG levels, weight, and waist circumference when reassessed 18 months later. These improvements were in comparison with a control group of age-, sex-, and ethnicity-matched POP-ABC participants who had maintained normoglycemia during the same follow-up period. The POP-ABC study was a longitudinal natural history study of the transition from normal glucose regulation to prediabetes among initially normoglycemic AA and EA adults who have 1 or both biological parents with diagnosed T2DM.[16,17] By design, no therapeutic intervention was offered to the participants who progressed to prediabetes. However, a written report of the confirmatory OGTT results indicating the development of IFG and/or IGT was provided to participants reaching those endpoints. Thus, our present findings suggest that the mere notification of incident prediabetes status may trigger beneficial changes in glycemia and body size in AAs and EAs with parental history of T2DM.

The mechanism(s) for the observed benefits in glycemia, glucose tolerance, and adiposity following notification of incident prediabetes status are unclear but could have involved self-directed lifestyle modification. To explore such a mechanism, we analyzed changes in self-reported dietary and physical activity behaviors in the 2 comparison groups of participants across the 18-month interval between conclusion of the POP-ABC study and initiation of the PROP-ABC study. We found that FHQ scores decreased in both the prediabetes and control groups (indicating healthier eating patterns), but the improvement was significantly greater in the prediabetes group. Similarly, participants in the prediabetes group reported a significant 6 MET-hr/wk greater physical activity than did those in the control group. These behavioral alterations, if sustained, could explain the glycemic and weight benefits observed in the prediabetes group. In addition to self-directed behavioral changes, it is possible that some participants may have discussed their prediabetes status with their primary care providers and may have received lifestyle counseling outside our study.

Did the Hawthorne effect or regression to the mean play a role in our findings? Subjects enrolled in research studies may alter their behavior and produce unintended results, from the increased attention and encounters associated with research participation (Hawthorne effect).[21] Usually, the Hawthorne effect dissipates within approximately 4 months or once study procedures cease.[22,23] Notably, our POP-ABC study had a follow-up period of 5.5 years (mean, 2.62 years) and the interval between end of study and retesting of participants was 18 months. Thus, it is unlikely that our present findings could be explained by the Hawthorne effect. Moreover, there is no reason why any Hawthorne effect would be restricted to the prediabetes group. The statistical phenomenon of regression to the mean describes the tendency of individuals with outlier values on a given measure to have spontaneously lower values on retesting, without any intervention. The use of contemporaneous measurements in a matched control group, as was done in the present study, usually obviates the risk of misinterpretation due to regression to the mean.[24,25]

Two previous national surveys had reported discordant findings on the impact of awareness of prediabetes status on health behavior.[14,15] Data from the 2006 National Health Interview Survey showed that of the ~ 4% of US adults who had been told that they had prediabetes, 68% reported active attempts to lose or control weight, 55% reported increased physical activity, 60% reported less fat consumption, and 42% reported trying all 3 approaches.[14] In a different report, investigators analyzed data on 24-hour dietary recall, self-reported diabetes and prediabetes awareness status, and FPG and HbA1c values from the 2005 to 2010 National Health and Nutrition Examination Survey.[15] Persons unaware of diabetes and prediabetes were identified by FPG <126 mg/dL or HbA1c <6.5% and FPG 100 mg/dL to 125 mg/dL or HbA1c of 5.7–6.4%, respectively.[15] People with diagnosed diabetes reported consumption of less carbohydrates and more protein compared with those with undiagnosed diabetes.[15] However, the authors observed no significant differences in macronutrient intake by awareness of prediabetes status.[15] The conclusion from the 2005 to 2010 National Health and Nutrition Examination Survey data was that knowledge of glycemic status induced healthier dietary patterns for people with diabetes but not those with prediabetes. Unlike the cross-sectional surveys based on self-reported prediabetes status, our prospective study used a rigorous ascertainment of incident prediabetes status, using OGTT and independent adjudication. Thus, the mechanism of becoming aware of incident prediabetes in the POP-ABC study was the documentation that a participant's FPG and/or 2hrPG values had drifted upward from normal glucose regulation to prediabetes (IFG and/or IGT). Our findings support a beneficial impact of prediabetes awareness on health behavior and metabolic endpoints. We argue that awareness of incident prediabetes status through direct communication of measured glucose values, as was done in the POP-ABC study, was impactful in triggering behavioral change.

Remarkably, the vast majority of people with prediabetes in the general primary care population remain undiagnosed and unaware of their condition.[2,14,15,26] In the 2006 National Health Interview Survey, only an estimated 4% of US adults had been told they had prediabetes.[15] Although ~34% of US adults had prediabetes in 2015, only 11.6% self-reported that they had been diagnosed with prediabetes by a health care worker.[2] Awareness of and response to prediabetes among health care providers is even much lower. Analyzing HbA1c data from the 2012 National Ambulatory Medical Care Survey that targeted adults aged >45 years without diagnosed diabetes who had an HbA1c test within 90 days (N = 11,167,004 weighted visits), the prevalence of prediabetes was found to be 33.6%.[26] However, <1% of patients with HbA1c results consistent with prediabetes had formal diagnosis and documentation of prediabetes in the medical records.[26] Clearly, there is need for a greater awareness and early action regarding the diagnosis and management of prediabetes. Once a diagnosis is obtained, using FPG, OGTT, or HbA1c values, the clinical significance of prediabetes needs to be communicated clearly to all affected patients (6 to 8), as the present report has shown that notification of prediabetes status (even without active intervention) may trigger beneficial self-directed lifestyle modifications. In addition, current guidelines by the ADA recommend referring people with prediabetes for intensive lifestyle modification.[4]