Abstract:
Background: Good glycemic control reduces the risk of diabetic complications. Despite this, achieving good glycemic
control remains a challenge in diabetic patients. The objective of this study is to identify determinants of glycemic control
among insulin treated diabetic patients at Jimma University Hospital, Southwest Ethiopia.
Methods: Hospital-based cross-sectional study was conducted on systematically sampled 284 insulin-treated diabetic
patients with a regular follow up. Data was collected by interviewing patients during hospital visits and reviewing respective
databases of September 2010 to December 2011. Data collection took place from February 20 to May 20, 2012. Poor
glycemic control was defined as fasting blood sugar (FBS) $126 mg/dL. Binary logistic regression analysis was conducted to
identify predictors of poor glycemic control.
Results: Patients had a mean age of 41.37 (615.08) years, 58.5% were males, the mean duration of insulin treatment was 4.9
(65.1) years, 18.3% achieved good glycemic control (FBS#126 mg/dL), 95% self-reported repeated use of disposable insulin
syringe-needle and 48% correctly rotating insulin injection sites. Most (83.1%) of study participants had one or more
complications. On multivariable logistic regression analyses, body weight of .70 Kg (AOR = 0.21; P,0.001), total daily dose
of insulin #35 IU/day (AOR = 0.26; P,0.001), total daily dose variation without checking glycemic level (AOR = 3.39;
P = 0.020), knowledge deficit about signs and symptoms of hyperglycemia (AOR = 3.60; P = 0.004), and non-adherence to
dietary management (AOR = 0.35; P = 0.005) were independent predictors of poor glycemic control.
Conclusions: The proportion of patients with poor glycemic control was high, which resulted in the development of one or
more complications regardless of duration on insulin treatment. Hence, appropriate management of patients focusing on
the relevant associated factors and independent predictors of poor glycemic control would be of great benefit in glycemic
control.