Insulin use in long term care settings for patients with type 2 diabetes mellitus: a systematic review of the literature.

Objective: To summarize currently available data about insulin therapy in patients with diabetes mellitus (DM), focusing on patients with type 2 DM (T2DM), in long term care (LTC) settings.

Data Sources: Ovid Medline, EMBASE, Cochrane Library databases, and United Kingdom National Health Service (NHS) Economic Evaluation Database, last accessed on November 12, 2012.

Study Eligibility Criteria: We included studies that reported insulin use in patients with T2DM, and studies with combined samples of patients with type 1 DM or T2DM, that were conducted in LTC settings. Excluded were review articles and studies published before 2000.

Results: We identified 11 articles that met all inclusion and exclusion criteria. Insulin use in patients with DM in LTC settings varied widely, from 2.7% to 58.0%. It is difficult to draw conclusions from these proportions, as many studies did not define whether their populations were exclusively patients with T2DM. Despite recommendations against its use by the American Diabetes Association, the American Geriatrics Society, and the American Medical Directors Association, treatment with sliding-scale insulin (insulin injections adjusted to current blood glucose levels) was prevalent in the LTC setting. Although the recommended target hemoglobin A1c (A1C) for this patient population varies from ≤6.5% to ≤8.0%, higher A1C values (8.0%-8.9%) were associated with better patient outcomes in a study examining insulin treatment in community-dwelling elderly patients enrolled in an outpatient LTC setting. Insulin pen-devices seemed associated with a high incidence of needle-stick injuries in workers in LTC settings but, compared with insulin vials, showed cost advantages for use in very short-term (≤30 days) patients with DM in LTC settings.

Limitations: Paucity of available data; only published studies for which full-text articles could be retrieved and which were identified by our search strategy were included; insufficient detail about patient samples were available in many included studies; and potential biases across studies might be introduced by funding sources or study designs.

Conclusions: Available data about insulin therapy in patients with DM in LTC settings are very scarce and great treatment variability of this patient population seems to prevail in the current clinical practice. Additional, randomized, prospective clinical trials are needed to expand our knowledge and allow clinicians to make informed treatment decisions for patients with DM in LTC settings.