time of onset was coded as 4:00 AM; for “afternoon,” time was coded as 12:00 PM; and for “evening,” time was coded as 8:00 PM. If no time of the day was described, then time of onset was coded as 12:00 PM. Only five cases had to be coded with these conventions. Date and time that doses of antipsychotic medications were given The electronic medical record contained the exact times that doses of antipsychotic medications were given as en- tered by nursing staff. Wewere also able to see if medications were scheduled to be given but were held. Date and time that delirium resolved Delirium was considered resolved when the consult- ing psychiatrist’s assessment reflected this in the progress notes. Specific times of resolution were present in the progress notes for 38 patients. For the other eight patients, the progress notes did not contain a specific time when delirium symptoms resolved, so the standard convention for estimation of time that was described earlier was used. If patients were discharged with delirium symptoms, their discharge date and time were used to determine their dura- tion of delirium. Duration of delirium symptoms after medication was started The time it took for delirium symptoms to resolve after medication treatment was initiated was estimated as the number of hours from the first dose of antipsychotic medica- tion until documentation that delirium symptoms resolved. Timeliness of starting medication If patients received a dose of antipsychotic medication within the first 24 hours after deliriumwas first noted, then they were coded as the Timely group. If they did not receive a dose within the first 24 hours, then they were coded as the Delayed group. Type of delirium Mental status exam findings from progress notes were used to classify the types of delirium at the time of diagnosis as hypoactive, hyperactive, and mixed, following methods used in previous studies. 12 Besides the presence of fluc- tuating level of consciousness, criteria for being coded as “hypoactive” delirium included lethargy, sedation, and/ or psychomotor retardation. To be coded as “hyperactive,” symptoms included agitation, aggression, and/or psychotic symptoms. If the patient had symptoms from both, they were coded as “mixed.” Data Analysis All of the statistical analysis was performed using SAS version 9.3 (Cary, NC). Assignment to Scheduled versus PRN groups was determined empirically (see Results).
(scheduled dosing of haloperidol at 4:00 AM, 11:00 AM, 4:00 PM, and 10:00 PM) when compared with the standard ap- proach (which mainly included sporadic and PRN dosing) led to shorter durations of delirium. 4 The first goal of this naturalistic chart review is to ex- amine the duration of delirium after initiation of scheduled doses of antipsychotic medication compared to PRN (as needed) dosing. The existing literature has devoted little attention to the issue of scheduled versus PRN dosing. A second goal is to examine whether duration of delirium re- solves more rapidly when medications (whether scheduled or PRN) are initiated sooner after diagnosis as opposed to later. If delirium resolves more rapidly with prompt initia- tion of treatment (i.e., before the delirium can worsen or have secondary impacts on other physiological systems), this would have direct and immediate implications for clinical practice. PATIENTS AND METHODS Patients The setting was an academic-affiliated hospital in a medium-size city in the southeastern United States. The hospital is situated in the downtown area and provides both general inpatient medical services as well as mul- tiple specialized and tertiary referral services typical of an academically-based institution. Investigators reviewed the paper and electronic charts of all 86 patients admitted to the hospital from January 2007 to January 2013 who received diagnosis of delirium from the consult-liaison psychiatrist. Inclusion criteria were a clinical diagnosis of delirium by the psychiatric consultant, 18 years of age and older, and a recommendation from the consultant to use an antipsychotic for treatment. Subjects were excluded if they had a history of a primary psychotic disorder, diagnosis of delirium tre- mens or substance-induced psychosis (if an illicit substance). Eighty-six patient charts were reviewed, and 42 patients met criteria for inclusion in analyses. Procedure The study protocol was submitted and accepted through the Tulane University Institutional Review Board. Data abstractors were not blinded to hypotheses. The data abstractors were the first author and a research assistant. In this hospital setting, there was not an established protocol for treatment of delirium on any of the hospital services. Data were systematically collected and entered onto a stan- dardized Case Report Form (CRF) developed for this study. Measures Date and time for onset of initial recognition of delirium symptoms Progress notes indicated a specific time that the onset of delirium was noted. If the progress notes did not con- tain a specific time, then a standard convention was used for estimation. If the progress note noted “morning,” then
J La State Med Soc VOL 166 November/December 2014 243
Made with FlippingBook - Online catalogs