Study Design, Setting, and Sample
After institutional review board approval, a retrospective chart review was conducted, using medical records of youth who presented in the emergency department at two separate hospitals in a Southern U.S. state for NSSI. Over 1,900 patient encounters were initially identified for another study examining youth suicide that fit the following inclusion criteria: (a) patients who were older than 4 and less than 19 years and (b) patients who presented because of self-injury from January 1, 2011, through August 31, 2012. For the present study, the 1,900 patient encounters were coded for suicide attempt, self-harming ideation, or self-harming act to examine the characteristics and management of patients who presented for self-harm ideation or self-harm. Of the original 1,900 patients, 644 were eligible for inclusion in the present study.
The 644 patient encounters were then screened against the following additional inclusion criteria: (a) patient encounter that involved the evaluation of some form of self-harm (i.e., act or ideation), and (b) self-inflicted harm that did not include intent to die. Two investigators independently reviewed all documents associated with the identified patient encounter for inclusion, and when a case was unclear, it was discussed between the investigators, and consensus was reached regarding inclusion or exclusion. The two investigators were able to reach consensus on all patient encounters. For youths who had more than one patient encounter that met the initial inclusion criteria, only the first encounter was included. A total of 135 patient encounters met the final inclusion and exclusion criteria.
Once the final sample was obtained, comprehensive data were extracted from each included medical record by reviewing all documents and laboratory work results associated with the patient encounter. Examples of documents reviewed were triage nursing notes, social worker consultant notes (when available), emergency department nurse and physician notes, and psychiatric consultant notes (when available). The extracted data included detailed demographic information, hospital data (e.g., length of stay, highest level of care, and discharge disposition), characteristics of NSSI (e.g., mechanism of harm, identification of precipitating event, and previous history of NSSI), personal and family mental health history (e.g., diagnoses, treatments, trauma history), and social issues (e.g., living situation, family structure).
Descriptive statistics were used to examine nominal and ordinal study variables. The data regarding the characteristics of the NSSI and circumstances contributing to engagement in NSSI were often qualitative, coming from social work and psychiatric consultations in the emergency department. Content analysis was used to analyze youth self-reports of reasons for engaging in NSSI and additional aggravating circumstances that might have contributed to engagement in NSSI as documented by the providers who cared for the youths in the hospital. Three independent analysts reviewed the text data for key words and additional meaning units. The researchers collectively reviewed the coding of key words and meaning units throughout the data analysis process and worked collaboratively to organize codes into themes and subthemes.
J Pediatr Health Care. 2017;31(3):334-341. © 2017 Mosby, Inc.