Optimizing outcomes: System and patient factors
medwireNews: An EASD session on optimizing patient outcomes included research into the effect of guideline adherence, predictors of changes in medication adherence, and overtreatment of older people with diabetes.
In the first presentation, Bob Young, from Salford Royal NHS Foundation Trust in the UK, reported that when healthcare providers followed specific guideline-recommended annual care processes, it significantly reduced their diabetes patients’ risk for mortality.
The three processes were measurement of patients’ glycated hemoglobin, blood pressure, and cholesterol. Using the UK’s National Diabetes Audit, Young’s team compared more than 250,000 diabetes patients who had complete (ie, 21) care processes over 7 years (2006–2013) with nearly 100,000 who had 12 or fewer, and found a significant difference in mortality rates over the subsequent 3 years, favoring patients with complete care processes.
Also predictive of mortality was glycated hemoglobin (HbA1c) levels, with consistently high age-adjusted levels over time associated with increased mortality in type 2 diabetes patients, relative to consistently low levels. Young noted that the analysis only considered patients in the top and bottom fifths of HbA1c over the entire 7 years, saying that “for simplicity, we’ve just presented the extremes.”
However, the reverse was true for blood pressure and cholesterol, with low levels associated with increased mortality, which Young described as “counter intuitive” and attributed to the confounding effects of heart failure and statin treatment, respectively.
Next, John Cooper (Stavanger University Hospital, Norway) presented findings from a national registry of adult patients with type 1 diabetes, reporting risk factors for having an HbA1c level above 9%, which was the case for 18.7% of the 7601 patients.
Risk factors, which were significant after accounting for multiple variables, were age of 25 years or younger, smoking, and living alone, whereas greater education was a protective factor. A greater frequency of self-monitored blood glucose tests was also protective, with the largest reduction in risk seen for people who tested nine to 10 times/day versus less than once a week, as was a higher frequency of exercise.
Cooper reported a “striking” rise in the risk of high HbA1c with increasing insulin dose per kg bodyweight, which, of all variables studied, gave the largest risk increase (for the highest vs lowest dose categories).
“Whether this is really something to do with residual beta-cell function or whether it may just be an artifact, because we increase the dose when control is poor, I’m not quite sure,” he said.
Cooper conceded that the data contained “no really sensational results,” but suggests that they still have useful clinical implications, by highlighting where healthcare resources may be best invested, such as in transition clinics for 18 to 25-year-olds.
The final two presentations of the session focused on older people with diabetes. In the first of these, Julie Schmittdiel (Kaiser Permanente, Oakland, California, USA) reported that some modifiable factors are predictive of changes in medication adherence. Specifically, having a greater than 90-day supply of medication on hand and making use of mail order prescription refills significantly reduced the likelihood of becoming nonadherent among 46,406 diabetes patients aged 65 years or older. Around a quarter of these patients became nonadherent during 4 years of follow-up.
Factors increasing the odds for nonadherence were not modifiable: age of 75 years or more; being female or American Indian/Alaska Native or African American; and having multiple comorbidities.
Finally, Huberta Hart (University Medical Center, Utrecht, the Netherlands) discussed whether older diabetes patients (≥70 years) receive appropriate treatment, considering that intensive treatment can increase hypoglycemia risk and reduce quality of life in these patients.
Referring to the current Dutch guidelines, which vary the target HbA1c level according to age and diabetes duration and treatment, Huberta’s team found that 51.7% of a cohort of 319 patients aged 70 years or more had an HbA1c target above the usual 53 mmol/mol threshold.
Of these patients, 32.7% were optimally treated (at age-appropriate HbA1c target), but 28.5% were undertreated (above target) and 38.8% overtreated (met a lower target), meaning that, in all, 67.3% of patients were incorrectly treated.
Surprisingly, nearly a third of the overtreated patients with the highest HbA1c target, of 64 mmol/mol, were using insulin. Half were considered frail, 22% had recorded hypoglycemia and nearly a third had falls. Almost all were using at least five medications.
Among other recommendations, Hart suggested that guidelines should contain age-specific minimum HbA1c limits, as well as upper thresholds.
medwireNews is an independent medical news service provided by Springer Healthcare. © 2017 Springer Healthcare part of the Springer Nature group