Overtreatment of older people with type 2 diabetes-a high impact frequent occurrence in need of a new definition

Prevalence of tight glycemic control based on frailty status and associated factors in community-dwelling older adults

BACKGROUND

Tight control of type 2 diabetes (T2DM) in frail older adults has shown to be associated with adverse outcomes. The objective of this study is to determine the prevalence of tight glycemic control based on underlying frailty status and its association with functional and cognitive measures in community-dwelling older adults.

METHODOLOGY

Ancillary study of the Singapore Population Health Studies on older adults aged ≥65 years with T2DM. Tight glycemic control cut-offs were based on the 2019 Endocrine Society guideline using HbA1c target range based on a patient's overall health status measured by the FRAIL scale. Data on basic demographics, frailty, cognitive, and functional statuses were collected. Multivariable regression was used to assess potential factors associated with tight glycemic control.

RESULTS

Of 172 community-dwelling older adults with diabetes mellitus and HbA1c done, frail (65%) and pre-frail (64.4%) participants were more likely to have tight glycemic control than robust participants (31.6%, P < 0.001). In multi-variate analysis, frailty (OR 6.43, 95% CI 1.08-38.1, P = 0.041), better cognition (OR 1.15, 95% CI 1.02-1.32, P = 0.028), and multi-morbidity (OR 7.36, 95% CI 1.07-50.4, P = 0.042) were found to be significantly associated with increased odds of tight glycemic control.

CONCLUSION

Tight glycemic control was highly prevalent in frail and pre-frail older adults, especially in those with multi-morbidity and better cognition. Future prospective longitudinal studies are required to evaluate effectiveness of frailty screening in making treatment decisions and long-term outcomes. Key messages What is already known on this topic:  There is growing recognition that glycemic targets should be adjusted based on health or frailty status. However, there is no consensus on how health status or frailty should be defined when determining glycemic control targets. What this study adds:  Our study found that tight glycemic control was highly prevalent in frail and pre-frail older adults. Our findings highlight the importance of assessing for tight glycemic control based on frailty status and further work is needed to aid implementation of screening and intervention policies to avoid the attendant harms of tight glycemic control.

Diabetes mellitus in older persons with neurocognitive disorder: overtreatment prevalence and associated structural brain MRI findings

BACKGROUND

Tight diabetes control is often applied in older persons with neurocognitive disorder resulting in increased hypoglycemic episodes but little is known about the pattern of brain injury in these overtreated patients. This study aims to: (a) quantify the prevalence of diabetes overtreatment in cognitively impaired older adults in a clinical population followed in an academic memory clinic (b) identify risk factors contributing to overtreatment; and (c) explore the association between diabetes overtreatment and specific brain region volume changes.

METHODS

Retrospective study of older patients with type 2 diabetes and cognitive impairment who were diagnosed in a memory clinic from 2013 to 2020. Patients were classified into vulnerable and dependent according to their health profile. Overtreatment was defined when glycated hemoglobin was under 7% for vulnerable and 7.6% for dependent patients. Characteristics associated to overtreatment were examined in multivariable analysis. Grey matter volume in defined brain regions was measured from MRI using voxel-based morphometry and compared in patients over- vs. adequately treated.

RESULTS

Among 161 patients included (median age 76.8 years, range 60.8-93.3 years, 32.9% women), 29.8% were considered as adequately treated, 54.0% as overtreated, and 16.2% as undertreated. In multivariable analyses, no association was observed between diabetes overtreatment and age or the severity of cognitive impairment. Among patients with neuroimaging data (N = 71), associations between overtreatment and grey matter loss were observed in several brain regions. Specifically, significant reductions in grey matter were found in the caudate (adj β coeff: -0.217, 95%CI: [-0.416 to -0.018], p = .033), the precentral gyri (adj βcoeff:-0.277, 95%CI: [-0.482 to -0.073], p = .009), the superior frontal gyri (adj βcoeff: -0.244, 95%CI: [-0.458 to -0.030], p = .026), the calcarine cortex (adj βcoeff:-0.193, 95%CI: [-0.386 to -0.001], p = .049), the superior occipital gyri (adj βcoeff: -0.291, 95%CI: [-0.521 to -0.061], p = .014) and the inferior occipital gyri (adj βcoeff: -0.236, 95%CI: [-0.456 to - 0.015], p = .036).

CONCLUSION

A significant proportion of older patients with diabetes and neurocognitive disorder were subjected to excessively intensive treatment. The association identified with volume loss in several specific brain regions highlights the need to further investigate the potential cerebral damages associated with overtreatment and related hypoglycemia in larger sample.

Tertiary centre study highlights low inpatient deintensification and risks associated with adverse outcomes in frail people with diabetes

INTRODUCTION

The community deintensification rates in older people with diabetes are low and hospital admission presents an opportunity for medication review. We audited the inpatient assessment and deintensification rate in people with diabetes and frailty. We also identified factors associated with adverse inpatient outcomes.

METHODS

A retrospective review of electronic charts was conducted in all people with diabetes and clinical frailty score ≥6 who were discharged from the medical unit in 2022. Data on demographics, comorbidities and background glucose-lowering medications were collected.

RESULTS

Six-hundred-and-sixty-five people with diabetes and moderate/severe frailty were included in our analysis. For people with no HbA1c in the last six months preceding admission, only 9.0% had it assessed during inpatient. Deintensification rates were 19.1%. Factors that were associated with adverse inpatient outcomes included inpatient hypoglycaemia, non-White ethnicity, and being overtreated (HbA1c <7.0% [53 mmol/mol] with any glucose-lowering medication).

CONCLUSION

The assessment and deintensification rate in secondary care for people with diabetes and frailty is low. Inpatient hypoglycaemia, non-White ethnicity, and overtreatment are important factors in determining inpatient outcomes highlighting the importance of deintensification and the need for an evidence-based risk stratification tool.

Therapeutic Inertia in the Management of Type 2 Diabetes: A Narrative Review

Adequate glycemic control is key to prevent morbi-mortality from type 2 diabetes (T2D). Despite the increasing availability of novel, effective, and safe medications for the treatment of T2D, and periodically updated guidelines on its management, the overall rate of glycemic goal attainment remains low (around 50%) and has not improved in the past decade. Therapeutic inertia (TI), defined as the failure to advance or de-intensify medical therapy when appropriate to do so, has been identified as a central contributor to the lack of progress in the rates of HbA1c goal attainment. The time to treatment intensification in patients not meeting glycemic goals has been estimated to be between 1 and 7 years from the time HbA1c exceeded 7%, and often, even when an intervention is carried out, it proves insufficient to achieve glycemic goals, which led to the concept of intensification inertia. Therefore, finding strategies to overcome all forms of TI in the management of T2D is a fundamental initiative, likely to have an enormous impact in health outcomes for people with T2D. There are several factors that have been described in the literature leading to TI, including clinician-related, patient-related, and healthcare system-related factors, which are discussed in this review. Likewise, several interventions addressing TI had been tested, most of them proving limited efficacy. Within the most effective interventions, there appear to be two common factors. First, they involve a team-based effort, including nurses, pharmacists, and diabetes educators. Second, they were built upon a framework based on results of qualitative studies conducted in the same context where they were later implemented, as will be discussed in this article. Given the complex nature of TI, it is crucial to use a research method that allows for an in-depth understanding of the phenomenon. Most of the literature on TI is focused on quantitatively describing its consequences; unfortunately, however, not many study groups have undertaken qualitative studies to deeply investigate the drivers of TI in their diverse contexts. This is particularly true in the United States, where there is an abundance of publications exploring the effects of different strategies to overcome TI in type 2 diabetes, but a severe shortage of qualitative studies aiming to truly understand the phenomenon.

Impact of deintensifying hypoglycaemic drugs in older adults with type 2 diabetes: protocol for an emulation of a target trial

INTRODUCTION

In older adults with type 2 diabetes (T2D), overtreatment with hypoglycaemic drugs (HDs: sulfonylureas, glinides and/or insulins) is frequent and associated with increased 1-year mortality. Deintensification of HD is thus a key issue, for which evidence is though limited. The primary objective of this study will be to estimate the effect of deintensifying HD on clinical outcomes (hospital admission or death) within 3 months in older adults (≥75 years) with T2D.

METHODS

We will emulate with real-world data a target trial, within The Health Improvement Network cohort, a large-scale database of data collected from electronic medical records of 2000 general practitioners in France. From 1 January 2010 to 28 February 2019, we will include eligible patients ≥75 years who will have T2D, a stable dose of HDs, glycated haemoglobin A1c (HbA1c) value <75 mmol/mol (9.0%) and no deintensification in the past year. The target trial will be sequentially emulated (ie, eligibility assessed) every month in the database. Patients will be classified at baseline of each sequential trial in the intervention arm (deintensification of HDs: decrease of ≥50% in the total dose of HDs, including complete cessation) or control arm (no deintensification of HDs). The pooled dataset for all sequential emulated trials will be analysed. The primary outcome will be time to first occurrence of hospital admission or death, within 3 months. Secondary outcomes will be hospitalisation, death, appropriateness of glycaemic control and occurrence of HbA1c >75 mmol/mol within 1 year. Participants will be followed from baseline to 12 months after randomisation, administrative censoring, or death, whichever occurs first. A pooled logistic regression will be used to estimate the treatment effect on the incidence of the outcomes.

DISSEMINATION AND ETHICS

No ethical approval is needed for using retrospectively this fully anonymised database. The results will be disseminated during conferences and through publications in scientific journals.