A1C as a Predictor of Mortality in Patients with Diabetes

Can variability in HbA1c lead to a higher risk of mortality in persons with diabetes?

A study published in the Diabetes/Metabolism Research and Review journal indicated an increased risk of mortality in adult patients with variability in HbA1c levels over five years. Previous studies have established a relationship between HbA1C and diabetic complications. Variability in HbA1c levels has been linked to higher rates of adverse outcomes. These adverse outcomes include cognitive decline, amputations, microvascular and macrovascular complications. Data analyzed from about 300,000 adults with diabetes showed that HbA1c trends could be a predictor of mortality. A greater risk of death was observed in all adults with a decreasing HbA1c trend. Patients with diabetes 70 years and older have a higher mortality risk with an increasing HbA1c trend. Patients who maintain a stable HbA1c level are at a lower risk of mortality than patients with varying HbA1c.

This study enrolled 293,314 adults between the ages 35 to 89 years who had diabetes listed at least four times in the National Diabetes Registry from 2012 to 2016. This study was a historical cohort observational study using the machine learning method to determine the best variables. Participants’ HbA1c values were collected for each year to decide which clusters they would be assigned. Mortality data was received from the national population register from January 1, 2017, to December 31, 2019. A K-meaning clustering machine learning algorithm was used to select the correct variables and classify participants into clusters. A multivariable logistic regression model was used to estimate odds ratios. The Cox proportional hazard model was not used because the hazards were not proportionally distributed over time.

In this study, 54.8% of the cohort had diabetes for more than ten years, and 25% of participants used insulin. The median age of participants was 68.0 years. There were 46.3% participants with obesity in the study. The cohort’s last median HbA1c was 7%.

 37.9% of participants had at least one hospital admission during the study, 11.1% had at least three hospital admissions. There was higher HbA1c variability in the younger age participants. Participants who did not have a stable HbA1c level were on insulin for longer and had a longer duration of diabetes than participants in the stable group.

Participants were divided based on age, then divided into subgroups classified as participants with stable HbA1c, increasing HbA1c, and decreasing HbA1c based on their year-to-year HbA1c trends. The stable HbA1c subgroup included 67% of participants aged 35 to 54 years, 75% of adults aged 55 to 69 years and, 79% of those aged 70 to 89 years.

Regardless of age, all adults who had an increasing or decreasing HbA1c had a higher likelihood of mortality than individuals with a stable HbA1c. For all age groups, HbA1c was a predictor of mortality. In the decreasing HbA1c subgroup, adults with HbA1c greater than 7% had a higher risk for mortality in the 35-to-89-year age range. Conversely, in the increasing HbA1c subgroup, only adults in the oldest age group had increased mortality.

This study suggests excess mortality with participants with an HbA1c of greater than 7% and not those who had an HbA1c of 7% or less. This excess in mortality can be linked to early exposure to high glucose levels.  To reduce the risk of mortality in patients with diabetes, it is crucial to obtain glycemic control early in the diagnosis and maintain a stable HbA1c level for as long as possible. Physicians have an essential role in patient care and need to consider the long-standing glycemic burden on their patients.

Practice Pearls:

• Older patients with diabetes who have a varying HbA1c are at a higher risk for mortality than patients with a stable HbA1c.
• The health care provider should emphasize the importance of stable glycemic control and its role in preserving life.
• Only patients with a stable HbA1c were showed to have a lower risk of mortality.

Cahn, Avivit et al. “Machine learning-based study of longitudinal HbA1c trends and their association with all-cause mortality: Analyses from a National Diabetes Registry.” Diabetes/metabolism research and reviews, e3485. July 7. 2021, doi:10.1002/dmrr.3485

Jerrine Fletcher, Fourth Year Doctor of Pharmacy Candidate, Florida A&M University College of Pharmacy