Background:
Patients with diabetes and comorbid cardiovascular disease may be at particularly high risk for medication non-adherence and adverse outcomes after hospital discharge. The objective of this study was to design, implement, and evaluate an intensive discharge intervention for inpatients with type 2 diabetes.
Methods:
We randomly assigned inpatients on medicine and cardiology services with type 2 diabetes and comorbid cardiac disease, likely to be discharged home on insulin therapy, to a strategy of an intensive transitional intervention or to usual care. The intervention included: 1) a nurse practitioner to coordinate care and patient education; 2) an inpatient pharmacist to perform intensive medication reconciliation and patient counseling; 3) standardized visiting nurse visits; 4) a post-discharge clinic visit staffed by the NP and an outpatient pharmacist; and 5) telemonitoring of point-of-care glucose levels. The primary end point was adherence to insulin 90 days after discharge based on pharmacy prescription refill information. Secondary outcomes included adherence to all medications, change from preadmission to 90-day post-discharge Hgb A1c levels, frequency of hypoglycemia per monitored patient-day, and 30-day post-discharge health care utilization based on medical records and post-discharge patient phone calls. Unadjusted analyses were conducted using Fisher exact test and Wilcoxon rank sum as appropriate. Adjusted analyses were conducted using weighted propensity scoring methods with general estimating equations to account for clustering by admitting physician.
Results:
The study included 180 patients, including 88 assigned to the intervention and 92 to usual care. The mean medication possession ratio to all insulin types was 84.5% (SD 22.6) among the intervention patients and 76.4% (SD 25.1) among usual care patients (difference 8.1, 95% CI -1.04 to 17.2, p=0.06). A smaller difference was seen for adherence to all medications (86.3% vs. 82.0%). Hgb A1c levels after discharge decreased by 1.09 in the intervention patients and by 0.11 in usual care (difference -0.98 (-2.03 to -0.07), p=0.04). No significant differences were seen in rates of hypoglycemic episodes per monitored patient-day, 30-day readmissions, or 30-day ED visits. In adjusted and clustered analyses, the difference in A1c reduction remained statistically significant (adjusted difference -1.06 (-1.89 to -0.22), p=0.01), while differences in all other outcomes remained non-significant.
Conclusions:
An intensive discharge intervention in patients with diabetes and heart disease was associated with improvements in glycemic control 90 days after discharge, possibly a result of improved adherence to insulin, although the latter result was not significant. No improvements in post-discharge health care utilization were seen, suggesting that different interventions (and larger studies) may be necessary to affect that outcome.