Article Text

Download PDFPDF

OP26 19-year trajectories of cardio-metabolic factors among patients with type 2 diabetes by dementia status in England
  1. Heidi Lai1,
  2. Kiara Chang1,
  3. Mansour Sharabiani2,
  4. Jonathan Valabhji3,4,5,
  5. Lefkos Middleton6,7,
  6. Azeem Majeed2,
  7. Christopher Millett1,
  8. Alex Bottle2,
  9. Eszter Vamos1
  1. 1Public Health Policy Evaluation Unit, Imperial College London, London, UK
  2. 2Department of Primary Care and Public Health, Imperial College London, London, UK
  3. 3NHS England, London, UK
  4. 4Imperial College Healthcare NHS Trust, London, UK
  5. 5Imperial College London, London, UK
  6. 6Ageing Epidemiology (AGE) Research Unit, School of Public Health, Imperial College London, London, UK
  7. 7Public Health Directorate, Imperial College Healthcare NHS Trust, London, UK


Background Type 2 diabetes (T2D) and dementia are both costly and rapidly growing global health crises, and major contributors to comorbidities and mortality. Recent improvements in treatments for T2D have improved survival, but people with T2D are subjected to a two-fold increase in risk of dementia. Associations of cardio-metabolic factors with dementia among patients with type 2 diabetes (T2D) are uncertain, and possibly driven by reverse causation, as dementia develops progressively long before diagnosis. Thus, assessing longitudinal trajectories over longer-follow up may help elucidate this relationship.

Methods We identified 227,580 patients with T2D aged >42 years between 1st January 1999 and 31st December 2018. Annual mean levels of eight routinely measured cardiometabolic factors were extracted from the Clinical Practice Research Datalink (England). Multivariable-adjusted multilevel piecewise and non-piecewise growth curve models assessed retrospective trajectories of cardio-metabolic factors among patients with T2D by dementia status from up to 19 years before dementia diagnosis (dementia) or last contact with healthcare (no dementia) at baseline.

Results 23,319 patients developed dementia; mean (SD) follow-up was 10.0 (5.8) years. Patients with dementia at baseline had a lower systolic blood pressure (SBP), body mass index (BMI), but higher fasting plasma glucose (FPG), HbA1c, and cholesterol compared to patients without dementia. In the dementia group, mean SBP increased 16–19 years before dementia diagnosis compared with patients without dementia, but declined more steeply from 16 years before diagnosis (-0.16 (95% CI -0.18, -0.14) mmHg per year). Diastolic blood pressure generally declined at similar rates, but levels at baseline by dementia status was not significantly different (0.11 mmHg, p=0.063). Mean FPG and HbA1c were generally higher among patients with dementia throughout follow-up, ranging between 0.09 to 0.21 mmol/L (p<0.03) and 0.66 to 5.07 mmol/mol (p<0.005) respectively. Mean BMI followed a steeper non-linear decline from 11 years before diagnosis in the dementia group, where group differences ranged between -0.22 to -1.26 kg/m2 (p<0.001). Mean blood lipid levels (total cholesterol, low-density lipoprotein, high-density lipoprotein) were generally higher in the dementia group compared with those without dementia and followed similar patterns of change.

Conclusion Marked changes in levels of cardiometabolic factors are apparent up to two decades prior to diagnosis of dementia, and could be potential targets for early intervention to prevent and delay the onset of dementia among older adults.

  • Diabetes
  • Dementia
  • Epidemiology

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.