Characteristics and effects of nurse dosing over-rides on computer-based intensive insulin therapy protocol performance

Abstract

When over-riding IIT CDSS recommendations, nurses overwhelmingly administered LTR doses, which emphasized prevention of hypoglycemia but interfered with hyperglycemia control, especially when BG was >150 mg/dl. Nurses appeared to consider the amount of a recommended insulin dose, not a patient's trend of insulin resistance, when administering LTR doses overall. Over-rides affected IIT CDSS protocol performance.

Integrating computerized clinical decision support systems into clinical work: A meta-synthesis of qualitative research

“…All but one (Dowding et al [23]), were undertaken in the United States, and in all but one (Feldstein et al [24]), data collection was undertaken between 2006 and 2010. All but two studies (Campion et al [25,26]) were undertaken in outpatient care settings. Inpatient environments were under-represented.…”

“…User control and inefficiency issues involved users' inability to actively manage alerts by delaying or retrieving them at a later more opportune time [30,31], or by turning them off especially when redundant [24]. Data entry associated with alerts was found to be rigid and inflexible often using repetitive data entry modes that were thought to be overengineered or 'mouse-heavy' [25,27,[29][30][31]. Alerts also failed to adequately inform or were unintelligible.…”

“…Six studies included prescribing clinicians (physicians, nurse practitioners) and three studies [23,25,26] included registered nurses. The studies reflect a diverse set of commercial [27][28][29], government [23,30,31], and in-house [24][25][26] developed products. The majority (6) of the CDSS applications are event-based alerts (see Table 1) supporting medication prescribing decisions and processes.…”

“…The findings were divided into three main subthemes. Nuisance issues included alerts that compelled documentation compliance [26]; poor fit between alerts and actual patient need (i.e., poor prioritization [24], inappropriate signal to noise [26]), repetitive and redundant alerts [25][26][27]31], inappropriate alert presentation within a process where alerts became distractions that predispose to missing or forgetting information [30,31]. User control and inefficiency issues involved users' inability to actively manage alerts by delaying or retrieving them at a later more opportune time [30,31], or by turning them off especially when redundant [24].…”

“…The majority (6) of the CDSS applications are event-based alerts (see Table 1) supporting medication prescribing decisions and processes. Other CDSS applications include multi-step protocols involving dosing calculators [23,25,26].…”