Global diabetes prevalence is increasing, producing a parallel rise in diabetic ketoacidosis (DKA) cases that in many settings outpaces the availability of intensive care unit (ICU) and intermediate care beds. Historically, the complexity of DKA management and safety concerns associated with continuous intravenous insulin infusions have necessitated ICU admission due to the need for intensive monitoring and nursing ratios. These constraints motivated the search for alternative, safe treatment pathways that could reduce critical care demand without compromising patient outcomes.
The primary purpose described was to develop and implement an emergency department (ED) protocol for treating patients with mild to moderate DKA using subcutaneous (SQ) insulin rather than an IV insulin infusion. By avoiding an infusion, the protocol aimed to reduce the intensity of physiologic and nursing monitoring required and thereby provide an alternative disposition pathway that could safely keep eligible patients out of the ICU.
The SQuID (Subcutaneous Insulin in Diabetic Ketoacidosis) protocol was created by a multidisciplinary team and intended for ED use in appropriately selected patients with mild to moderate DKA. The protocol uses subcutaneous insulin dosing as the primary insulin delivery method instead of an intravenous insulin infusion. Details about specific dosing schedules, inclusion/exclusion criteria, laboratory targets, or glucose monitoring frequency were not reported in the abstract.
A multidisciplinary implementation effort introduced SQuID in the ED and subsequently assessed options for admission outside the ICU. After the initial implementation, a second study expanded the acceptable inpatient locations for patients managed on the protocol. Over a four-year period, the protocol was actively incorporated into clinical practice and has begun to be adopted in some community affiliate hospitals.
Education for staff, ongoing monitoring processes, and iterative targeted modifications were part of the implementation strategy. Those efforts focused on safe adoption of the protocol and on operational measures intended to reduce the risk of adverse events, including hypoglycemia.
In its fourth year of use the SQuID protocol demonstrated substantial real-world uptake and operational impact. More than 75% of patients with mild to moderate DKA who were eligible were treated using SQuID. Adoption of the protocol was associated with a greater than 33% reduction in ICU utilization for this patient population. The abstract does not report additional numeric clinical outcomes (for example lengths of stay, rates of hypoglycemia, readmissions, or mortality) or detailed statistical analyses.
Implementation included comprehensive education, ongoing monitoring, and targeted protocol modifications explicitly intended to reduce hypoglycemia risk. The authors report that these combined safety interventions helped advance the institution's efforts to provide a safe alternative to conventional DKA management. Specific educational content, monitoring intervals, or the nature of the protocol modifications were not detailed in the abstract.
Following initial ED implementation, a follow-up study expanded the locations to which patients treated on SQuID could be admitted, enabling care in non-ICU settings when appropriate. The authors also state they are currently expanding the protocol into some community affiliate hospitals, suggesting the protocol is transferable beyond the originating academic center.
The SQuID protocol describes a pragmatic, multidisciplinary approach to treating selected patients with mild to moderate DKA using subcutaneous insulin in the ED and non-ICU inpatient settings. Over four years the protocol achieved high uptake among eligible patients and was associated with a significant reduction in ICU utilization for this population. Key elements cited for success included comprehensive staff education, ongoing outcome monitoring, and targeted safety-focused modifications—particularly those addressing hypoglycemia risk. The abstract does not provide granular clinical outcome data or protocol specifics; those details would need to be obtained from the full text.