Comparison of Two Methods Using Intranasal Lidocaine to Alleviate Discomfort Associated With Administration of Intranasal Midazolam in Children.

We will enroll 50 children to determine whether the PREMED method is non-inferior to the PREMIX method. We based our sample size on the outcome of pain and distress associated with the administration of IN midazolam, which will be measured using our primary outcome measure of the Observational Scale of Behavioral Distress-Revised (OSBD-R). The OSBD-R is an observational measure of distress that has been well validated in the pediatric population for evaluating painful and distressing procedures, and has been used in children as young as 1 year of age [1,2]. The sample size of 50 patients was determined based on a prestated margin of non-inferiority (delta) of 1.8 (SD 2.25). This value was based on the minimum clinically significant differences used in prior studies of painful procedures in children in the emergency department [3,4,5]. To determine noninferiority using a delta of 1.80 (SD 2.25), with a 1-tailed alpha of 0.025 and power of 80%, we would require 25 patients in each arm, for a total of 50 patients. OSBD-R scores will be determined independently by two blinded trained assessors who will review the videotapes of the study procedures. Interrater reliability of the OSBD-R between the two assessors will evaluated by determining the intraclass correlation coefficient. The period of administration of the midazolam alone in the PREMED group and the period of administration of the midazolam/lidocaine mixture in the PREMIX group are the two phases which will be compared to each other to determine our primary outcome.

Secondary outcome measures of pain and distress associated with IN midazolam administration will include the Children's Hospital of Eastern Ontario Pain Scale (CHEOPS); the Faces-LegsActivity-Crying-Consolability (FLACC) scale; and cry duration. These are all continuous measures that will be analyzed using the independent samples t-test.We will also evaluate parental and provider satisfaction across various domains using a 5-point Likert scale (see attached document for questions to be asked). Responses will be dichotomized into "agree" (i.e. if respondent answers "agree" or "strongly agree") or "disagree" (i.e. if respondent responds "undecided", "disagree", or "strongly disagree") and analyzed using the chi square test.