No abstract

Online interventions for older adults should be tailored to their unique needs to increase the efficacy of and adherence to the intervention. The agile development cycle is a dynamic model to solicit and incorporate feedback from older adults during the design process. We combined this approach with the framework of Harvard University’s clinical and translational phases that provide a clear structure for evaluating new health programs before they are offered in the community. We based our online memory program on the empirically validated in-person Memory and Aging Program. The aim of the present study was to combine the agile development cycle with the clinical and translational phases framework to develop and pilot an online memory program tailored to the unique needs of older adults. Study 1 involved piloting individual program modules on site and integrating participant feedback into the program’s design to optimize usability. Study 2 involved two sequential pilots of the program accessed remotely to evaluate preliminary clinical outcomes and obtain feedback for iterative modifications. Plans for further validation and limitations are discussed. The successful application of the agile development cycle implemented in this series of studies can be adapted by others seeking to offer online content for targeted end users.

Previous studies have suggested that cardiovascular autonomic neuropathy (CAN) may predict rapid kidney function decline among persons with diabetes. We analyzed the association between baseline CAN and subsequent glomerular filtration rate (GFR) decline among individuals with type 1 diabetes (T1D) from the Preventing Early Renal Loss in Diabetes (PERL) study (N=469) and with type 2 diabetes (T2D) from the Action to Control Cardiovascular Risk in Diabetes (ACCORD) study (N=7,973). Baseline CAN was ascertained using ECG-derived heart rate variability indices. Its association with GFR slopes, rapid kidney function decline (GFR loss ≥-5 ml/min/1.73 m2/year), and ≥40% GFR loss was evaluated by linear mixed effect, logistic, and Cox regression, respectively. Participants with CAN experienced more rapid GFR decline, by an excess 1.15 (95%CI [-1.93, -0.37], P= 4.0x10-3) ml/min/1.73m2/year in PERL and 0.34 (95%CI [-0.49, -0.19], P= 6.3x10-6) ml/min/1.73m2/year in ACCORD. This translated in 2.11 (95% CI [1.23-3.63], P=6.9x10-3) and 1.39 (95% CI [1.20-1.61], P=1.1x10-5) odds ratios of rapid kidney function decline in PERL and ACCORD, respectively. Baseline CAN was also associated with a greater risk of ≥40% GFR loss events during follow-up (HR=2.60, 95%CI [1.15-5.45], p=0.02 in PERL and HR=1.54, 95%CI [1.28-1.84], P=3.8×10-6 in ACCORD). These associations remained significant after adjustment for potential confounders, including baseline GFR and albuminuria. Our findings indicate that CAN is a strong, independent predictor of rapid kidney function decline in both T1D and T2D. Further studies of the link between these two complications may help develop new therapies to prevent kidney function decline in patients with diabetes.

No abstract