Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Aug 1.
Published in final edited form as: Arch Phys Med Rehabil. 2015 Aug;96(8 0):S256–S273.e14. doi: 10.1016/j.apmr.2015.01.025

Psychotropic Medication Use during Inpatient Rehabilitation for Traumatic Brain Injury

Flora M Hammond 1,2, Ryan S Barrett 3, Timothy Shea 4, Ronald T Seel 5, Thomas W McAlister 2, Darryl Kaelin 6, David Ryser 7, John D Corrigan 4, Nora Cullen 8, Susan D Horn 3
PMCID: PMC4516906  NIHMSID: NIHMS685876  PMID: 26212402

Abstract

Objective

To describe psychotropic medication administration patterns during inpatient rehabilitation for traumatic brain injury (TBI) and their relationship to patient pre-injury and injury characteristics.

Design

Prospective observational cohort.

Setting

multiple acute inpatient rehabilitation units or hospitals.

Participants

2,130 individuals with TBI (complicated mild, moderate, or severe) admitted for inpatient rehabilitation.

Interventions

NA

Main Outcome Measure(s)

NA

Results

Most frequently administered was narcotic analgesics (72% of sample) followed by antidepressants (67%), anticonvulsants (47%), antianxiolytics (33%), hypnotics (30%), stimulants (28%), antipsychotics (25%), antiparkinson agents (25%), and miscellaneous psychotropics (18%). The psychotropic agents studied were administered to 95% of the sample with 8.5% receiving only 1 and 31.8% receiving 6 or more. Degree of psychotropic medication administration varied widely between sites. Univariate analyses indicated younger patients were more likely to receive anxiolytics, antidepressants, antiparkinson agents, stimulants, antipsychotics, and narcotic analgesics, while those older were more likely to receive anticonvulsants and miscellaneous psychotropics. Men were more likely to receive antipsychotics. All medication classes were less likely administered to Asians, and more likely to those with more severe functional impairment. Use of anticonvulsants was associated with having seizures at some point during acute care or rehabilitation stays. Narcotic analgesics were more likely for those with history of drug abuse, history of anxiety and depression (premorbid or during acute care), and severe pain during rehabilitation. Psychotropic medication administration increased rather than decreased during the course of inpatient rehabilitation in each of the medication categories except for narcotics. This observation was also true for medication administration within admission functional levels (defined by cognitive Functional Independence Measure (FIM) scores), except for those with higher admission cognitive FIM scores.

Conclusion(s)

Many psychotropic medications are used during inpatient rehabilitation. In general, lower admission FIM Cognitive groups were administered more of the medications under investigation, compared to those with higher cognitive function at admission. Considerable site variation existed regarding medications administered. The current investigation provides baseline data for future studies of effectiveness.

Keywords: Brain injuries, Rehabilitation, Antipsychotic agents, Antidepressive agents, Central nervous system stimulants, Amantadine, Polypharmacy, Physician's practice patterns, Drug therapy, Medication therapy management, Patient care


Individuals with traumatic brain injury (TBI) frequently present to acute inpatient rehabilitation facilities with pain, hypoarousal, sleep dysregulation, behavioral dysregulation, spasticity, confusion, slowed cognitive processing, impaired memory, and affective disorders prompting prescription of multiple psychotropic medications.1 Some of these medications are aimed at controlling behaviors to prevent harm and allow safer and more effective management of the patient (e.g. use of stimulants, benzodiazepine and antipsychotic agents to control agitation). Other medication uses are aimed at preventing comorbidities (such as seizures), and some are aimed at enhancing function (such as sleep medications, stimulants, and antiparkinson agents).2

Upon admission and throughout the rehabilitation stay, the rehabilitation physician typically reviews prescribed medications to continually reassess the patient's needs. This includes discontinuing medications that no longer appear necessary or may cause an adverse response, while adding other agents as deemed necessary. There is sparse literature to guide such clinical decision-making, and there are no medications that are currently approved by the United States Food and Drug Administration for the treatment of TBI. Additionally, the small body of published research is commonly limited by scientific rigor, such as lack of controlled trials, non-blinded prescribers, lack of information regarding injury, limited information on relevant data (such as severity of injury and time of injury to treatment), mixed brain injury samples, and small sample size. Evidence of medication benefit and safety is usually extrapolated from therapeutic trials targeting common post-TBI conditions that also occur in other patient populations. An example would be the use of antipsychotic agents studied in patient populations other than brain injury and settings other than acute inpatient rehabilitation. There is a small but growing literature regarding which pharmacologic agents may be helpful in the acute rehabilitation setting for persons who sustain TBI. For example, a randomized, placebo-controlled trial of 184 TBI rehabilitation patients in vegetative state or minimally conscious state showed that amantadine was more effective than placebo in accelerating the rate of functional recovery.3

Various agents commonly used to manage the effects of TBI may cause adverse effects on health, function, and treatment efficiency.4-10 For example, a retrospective review of 182 consecutive patients with moderate-to-severe TBI revealed commonly prescribed neuroleptics were associated with 7 days longer of post-traumatic amnesia (PTA).1 In a study of individuals with TBI undergoing residential treatment, polypharmacy and use of anticholinergic medications were associated with an increased risk of falls.11

The degree to which psychotropic medications are used early after TBI during the course of inpatient rehabilitation is unknown. Use of psychotropic medications late after TBI was evaluated in a retrospective cohort study of 306 moderate-to-severe TBI survivors who had all been discharged from a TBI rehabilitation unit and were tracked up to 24 years post-injury. This study found that at follow-up, 58.9% were currently prescribed at least 1 medication. On average, persons with TBI were prescribed 2.64 (SD = 2.14) medications with a range of 1-12. The most prescribed medication types were anticonvulsants (25.8%), followed by antidepressants (8.2%), analgesics (8.2%) and anxiolytics (5.9%). 12

Due to a lack of evidence on medication effects in TBI patients, medication management during acute rehabilitation is driven largely by a patient's clinical presentation and physician subjective experience or preferences. Consequently, highly variable prescribing practices exist.2,13 There is significant need to study physicians' medication administration patterns during acute TBI rehabilitation. Medication pattern data could then be used as the basis for future research. Specifically, such data could help identify commonly used types of medicine that would benefit from effectiveness analyses, inform research design (including sample size determination), and identify the degree to which sociodemographics, injury severity, and other potential confounds (such as time from injury to rehabilitation, medical co-morbidities, function, insomnia, agitation) would need to be addressed.

The TBI Practice-Based Evidence (TBI-PBE) project provides a unique opportunity to describe patterns of psychotropic medication administration at specialized inpatient brain injury rehabilitation units in the United States and Canada including: 1) the medication agents prescribed; 2) if they were prescribed as needed (PRN) or scheduled; and 3) timing of medication initiation and discontinuation across the course of rehabilitation. The TBI-PBE data also allows evaluation of the relationship between medication prescription and patient demographic, injury, medical, and function.

Methods

Study design, study sites, and participants

The TBI-PBE Project is a 5-year, multi-center investigation of the TBI inpatient rehabilitation process.14 2,130 patients who received acute inpatient rehabilitation were enrolled in the project and used for the current study. The project sites included10 inpatient rehabilitation facilities: 9 in the United States and 1 in Canada. The study was approved by the local institutional review board at each study site. Inclusion criteria included: participant age of at least 14 years, informed consent from participant or their parent/guardian, and admission to the facility's brain injury unit for initial rehabilitation following TBI.

Variables and Data Collection

Collection and classification of Medications

Medication data were collected either through manual chart abstraction or electronic data download, depending on the site and availability and dependability of electronic data. Only those medications actually administered were recorded. Medications ordered but not given for any reason were not recorded. As customary during inpatient rehabilitation, medications were administered and recorded by nursing staff. Also per routine practice, a rehabilitation physician wrote the admission medication orders within minutes to hours of the patient's arrival to the inpatient rehabilitation unit and performed history and physical examination within 24 hours.

Common drug classification schemes vary, based on factors such as the chemical type of the active ingredient (e.g. “benzodiazepines”), presumed mechanisms of action (e.g. “serotonin re-uptake inhibitor”), or clinical indications for use (e.g. “antidepressant”). Medications were grouped primarily by common clinical usage/purpose and then by general mechanism of action. We also were aware that many drugs could be classified into more than 1 class (e.g. divalproex sodium as an anticonvulsant and as a mood stabilizer). For the purpose of this study, medications were classified in only 1 category. The classification scheme is outlined in Table 1. Patients may have been administered medications from multiple classes or more than 1 agent within a class, simultaneously or successively.

Table 1. Classification for the Psychoactive Medications Administered.
Major Drug Class & General Mechanism Pharmacologic Agents Received # (%)* Total # Patients Receiving Agent
Anxiolytic
Gaba-aminobutyric acid-A (GABA-A) agonist lorazepam (478; 68%), clonazepam (85; 12%), alprazolam (67; 10%), diazepam (66; 9%), chlordiazepoxide (5; <1%) 701
H-1 receptor antagonist hydroxyzine (21; 100%) 21
Other buspirone (151; 100%) 151
Anticonvulsant
Calcium channel antagonist levetiracetam (440; 61%), gabapentin (219; 30%), pregabalin (65; 9%) 724
GABA-A agonist tiagabine (4; 100%) 4
Sodium channel antagonist valproic acid (239; 39%), phenytoin (229; 37%), carbamazepine (56; 9%), topiramate (38; 6%), lamotrigine (23; 4%), oxycarbamazepine (13; 2%), fosphenytoin (12; 2%) primidone (3; <1%), zonisamide (2; <1%) 612
Other lacosamide (3; 100%) 3
Antidepressant
Norepinephrine-Dopamine reuptake inhibitor (NDRI) bupropion (30; 100%) 30
(NaSSA) mirtazapine (70; 100%) 70
Serotonin antagonist and reuptake inhibitor (SARI) trazodone (1124; 100%) 1124
Serotonin and norepinephrine reuptake inhibitor (SNRI) duloxetine (54; 52%), venlafaxine (45; 44%), milnacipran (4; 4%) 103
Selective serotonin reuptake inhibitor (SSRI) paroxetine (44; 8%), fluoxetine (37; 6%) 81
Tricyclic Antidepressant (TCA) - secondary amine nortriptyline (34; 92%), desipramine (3; 8%) 37
TCA - tertiary amine amitriptyline (62; 95%), doxepin (3; 5%) 65
Antiparkinson
Catechol-O -methyltransferase (COMT) inhibitor entacapone (1; 100%) 1
Dopamine agonist bromocriptine (190; 95%), pramipexole (7; 3%), ropinirole (4; 2%) 201
Monoamine oxidase (MAO) inhibitor benzatropine (15; 79%), rasagiline (2; 11%), selegiline (2; 11%) 19
N-Methyl-D-aspartate (NMDA) antagonist amantadine (361; 100%) 361
Other carbidopa + levodopa (28; 88%), levodopa (4; 13%) 32
Stimulant
Norepinephrine agonist atomoxetine (56; 100%) 56
Norepinephrine -Dopamine-5HT agonist sulfate + dextroamphetamine saccharate + dextroamphetamine sulfate (24; 5%), amphetamine + dextroamphetamine (6; 1%), dextroamphetamine (3; <1%) 490
Other modafinil (117; 96%), armodafinil (6; 4%) 123
Antipsychotic
First generation / Typical 11%) 55
Second generation / Atypical quetiapine (307; 48%), risperidone (119; 19%), olanzapine (93; 15%), ziprasidone (92; 14%), aripiprazole (25; 4%), paliperidone (1; <1%) 637
Hypnotic
Benzodiazepine GABA-A agonist temazepam (63; 62%), midazolam (38; 38%) 101
Non-benzodiazepine GABA-A agonist zolpidem (482; 88%), eszopiclone (62; 11%), zaleplon (3; <1%) 547
Melatonin agonist ramelton (13; 100%) 13
Other chloral hydrate (36; 57%), propofol (26; 41%), phenobarbital (1; 2%) 63
Narcotic Analgesic
Narcotic oxycodone (864; 37%), acetaminophen (APAP) + hydrocodone (688; 30%), morphine (205; 9%), fentanyl (145; 6%), tramadol (142; 6%), hydromorphone (85; 4%), propoxyphene N + APAP (84; 4%), codeine (48; 2%), methadone (44; 2%), APAP + codeine (14; <1%), meperidine (4; <1%), buprenorphine (4; <1%), propoxyphene N (4; <1%) 2234
Miscellaneous Psychotropic
Acetylcholinesterase inhibitor (AChE-I) donepezil (178; 95%), rivastigmine (6; 3%), physostigmine salicylate (3; 2%) 187
NMDA antagonist memantine (29; 100%) 29
Other nicotine (204; 98%), interferon beta 1a (2;<1%), glatiramer acetate (1; <1%), varenicline (1; <1) 208
*

#patients who received agent among sample of 2130 with medication data; % of patients who received the agent among the other agents in that mechanism within that classification

Patients may receive more than one agent within a mechanism

The medications studied included: anxiolytic agents, anticonvulsants, antidepressants, antiparkinson agents, stimulants, antipsychotics, hypnotics, miscellaneous psychotropics, and narcotic analgesics. These agents were selected among the many medications due to the need to focus the study, commonality of use in acute brain injury care, and the agent's use specifically for their central-acting property. Other psychotropic agents exist that were not studied, such as some centrally-acting antihypertensives, gastrointestinal agents, and others.

Descriptive variables

The variables for this study were chosen by the study investigators and clinicians at the onset of the project based on their clinical impressions and literature review of factors relevant to brain injury care and outcome. These data were obtained through medical record abstraction and interview with the study participants and their close others (proxy). Variables were chosen to represent patient characteristics prior to injury, post-injury before admission to rehabilitation, and during inpatient rehabilitation.

Premorbid variables studied for association with medication use included age (both continuous and categorical), gender, race, history of psychosis/schizophrenia/bipolar disorder, and history of alcohol or drug abuse.

Patient injury and medical data were abstracted from patient medical records by trained data collectors. Several variables were used to describe injury severity, including post-resuscitation Glasgow Coma Scale score in the Emergency Department, duration of PTA, and time from injury to rehabilitation admission. Any mention of presence of depression or anxiety in the medical record during acute care or at rehabilitation admission was recorded representing problems in this area premorbidly or during acute care. The extent and severity of medical illness during the rehabilitation stay was captured using the maximum Comprehensive Severity Index (CSI®) score. The CSI is derived by scoring the extent of deviation from normal physiological status for each medical complication and comorbidity present, with a higher CSI score denoting greater medical severity.15 A brain injury CSI subscore was used to establish the severity of central nervous system illness, while a non-brain injury CSI subscore established severity of illness of all other injuries, existing chronic disorders, complications, and comorbidities. The CSI score used for this study represented the maximum CSI score for the entire course of rehabilitation.14 Functional status and need for assistance were measured at rehabilitation admission by the Functional Independence Measure (FIM®). The FIM Cognitive and Motor scale scores were Rasch-transformed to a ratio scale using 0-100 scores.14,16

Rehabilitation variables included: presence of seizures at any point up to rehabilitation discharge (premorbid, during acute care, or during rehabilitation), percent of rehabilitation days with fewer than 5 hours of sleep between the hours of 9 PM and 6 AM, percent of rehabilitation stay agitated (defined as 6 shifts with Agitated Behavior Scale scores >21 out of twelve 4-hour shifts),17 and average level of effort over the stay for physical therapy, occupational therapy, and speech therapy, combined.18 Severity of pain was operationalized as percent of the rehabilitation stay with a patient-reported pain score of 7 or higher (out of a possible score of 10, which was the worst pain).19

Data processing and analysis

Description of medication administration during course of rehabilitation

Percentages were used to portray the frequency of psychotropic medication administration for each pharmaceutical class during rehabilitation.

Comparison by cognitive function at rehabilitation admission

Five relatively homogenous subgroups were created based on admission FIM Cognitive scores to stratify the impact of patients' cognitive impairments on outcomes and facilitate between group comparisons of medications administered.14 The admission FIM Cognitive categories used were: <6, 7-10, 11-15, 16-20, >21.

Factors related to medications administered

Data were analyzed to determine patient characteristics that may differentiate whether medications in each pharmaceutical class were either administered or not administered. Medication administration patterns were also compared across treatment sites (details in the Results section under subheading of medication administration across sites). Categorical variables with more than 2 categories (e.g. site, age, race/ethnicity) were evaluated using the chi-square test; categorical variables with 2 categories (e.g. gender) were evaluated with Fisher's exact test. Continuous variables (e.g. brain injury and non-brain injury CSI) were evaluated using the independent-samples t-test. To minimize Type I error, only differences reaching an alpha level of p<.001 were considered significant. Correction for multiple comparisons was not performed due to the exploratory nature of this descriptive paper.

Calculation of rehabilitation weeks

In order to study the timing of medication initiation and discontinuation across the course of rehabilitation we depicted medication administration by week of stay in rehabilitation. All patients with a rehabilitation length of stay (RLOS) of 8 days or less were considered to have only 1 admission week. All others have an admission (week 1) and a discharge week, at a minimum. Patients with RLOS of 9-15 days have 2 weeks; RLOS of 16 and 17 have 3 weeks (with the admission week comprised of only 6 days). All patients with an RLOS ≥18 days have the following: 18-22 have 3 weeks; 23-29 have 4 weeks; 30-36 have 5 weeks; etc. There are no weeks shorter than 4 days, and none longer than 8. For RLOS with remainders of 1 when divided by 7 (e.g. 22, 29, etc.) the extra day is added to the discharge week to create an 8-day week.

Results

Study Sample

Our sample of 2,130 patients with TBI was 73% male, 74% white, 37% married, and 51% employed at the time of injury. Average age of the sample was 45 years. Cause of injury was most commonly vehicular accidents (56%), followed by falls or flying objects (32%), violence (7%), and sports (2%). Mean RLOS was 27 days (SD = 20). The mean Rasch-transformed FIM Motor score at admission was 33 (SD = 19) and mean Rasch-transformed FIM Cognitive score was 37 (SD = 20). The mean time from injury to rehabilitation admission was 29 days (SD = 34). The first article in this series14 further summarizes the demographic and injury characteristics for the sample.

Patterns of Medication Administration

Medication use by admission FIM cognitive categories

Medication use is summarized by admission FIM Cognitive subgroup in tables 2 and 3, based on time-variant factors. For all medication classes except anticonvulsants, use was less frequent among those in the highest FIM Cognitive subgroup than in the lower groups. Conversely, medication use was greater for those with worse cognitive function at the time of rehabilitation admission. Use was higher in the 2 lower FIM Cognitive groups than middle and higher functioning subgroups for antiparkinson agents, stimulants, and anxiolytics, while antipsychotic and miscellaneous psychotropics had the opposite pattern. In general, as admission FIM Cognitive groups increased in function, antidepressant use was less. For example, antiparkinson agents were used for 35% and 26% of the patients in the 2 lowest FIM Cognitive subgroups, with frequency decreasing with higher admission cognitive function. Anticonvulsant use was higher for the 2 highest FIM cognitive groups, although use did not substantially vary across the 5 subgroups.

Table 2. Summary information on psychoactive medications administered during rehabilitation, by pharmaceutical categoryand level of functional cognition.
Pharmaceutical category FIM Cognitive score at admission N* % ever rec'd % rec'd first 2 days % rec'd last 2 days % rec'd first & last 2 days % rec'd 5+ days mean # days (for those rec'd) median # days (for those rec'd) mean % stay (for those rec'd) median % stay (for those rec'd) % days PRN % days Sched % days Unknown PRN/Sched
Anxiolytic Overall 2130 33 19 19 11 23 17 12 56 58 37 53 10
Adm cog <=6 339 48 24 26 14 31 21 18 50 50 40 45 14
Adm cog 7-10 374 44 23 26 12 32 19 14 53 52 37 55 7
Adm cog 11-15 495 31 18 18 11 22 17 13 60 67 36 57 7
Adm cog 16-20 408 28 20 17 13 20 13 9 62 73 33 56 11
Adm cog >=21 504 20 13 12 7 14 12 10 60 65 39 50 11
Anticonvulsant Overall 2130 47 35 39 28 43 23 17 81 100 6 76 19
Adm cog <=6 339 50 32 40 25 48 35 28 76 93 8 71 21
Adm cog 7-10 374 52 34 42 26 48 27 23 77 94 3 85 12
Adm cog 11-15 495 46 34 39 28 44 22 18 83 100 4 80 16
Adm cog 16-20 408 46 38 41 33 43 19 15 87 100 4 73 22
Adm cog >=21 504 41 37 32 28 36 13 12 83 100 10 67 24
Antidepressant Overall 2130 67 44 55 37 61 23 18 78 93 27 60 13
Adm cog <=6 339 77 47 64 40 73 34 29 79 94 23 62 15
Adm cog 7-10 374 76 48 63 39 72 26 22 78 90 25 65 10
Adm cog 11-15 495 66 46 54 38 60 21 19 80 95 30 57 13
Adm cog 16-20 408 69 49 58 42 62 17 14 78 95 28 57 15
Adm cog >=21 504 53 35 41 28 42 15 12 74 90 30 57 13
Antiparkinson Overall 2130 25 11 20 8 23 25 21 73 83 3 83 14
Adm cog <=6 339 53 24 40 17 49 30 26 70 79 4 84 12
Adm cog 7-10 374 40 18 31 13 36 25 22 71 85 1 87 12
Adm cog 11-15 495 21 8 19 7 20 24 19 78 86 2 82 15
Adm cog 16-20 408 15 4 12 4 13 16 14 71 75 5 72 23
Adm cog >=21 504 6 4 6 3 5 14 14 80 100 5 73 23
Antipsychotic Overall 2130 25 16 15 10 21 20 15 65 75 23 62 15
Adm cog <=6 339 38 18 23 10 34 27 20 57 55 29 49 22
Adm cog 7-10 374 34 21 23 13 30 21 17 66 78 24 61 15
Adm cog 11-15 495 28 21 18 13 23 18 15 71 89 18 74 8
Adm cog 16-20 408 22 16 11 7 18 15 13 63 75 17 71 12
Adm cog >=21 504 10 8 6 5 7 12 9 65 70 33 50 16
Hypnotic Overall 2130 30 14 20 10 23 18 13 60 67 48 42 10
Adm cog <=6 339 36 14 23 9 29 24 21 57 58 34 51 16
Adm cog 7-10 374 37 17 26 12 31 22 18 62 70 41 50 9
Adm cog 11-15 495 31 15 20 10 25 18 14 61 68 47 46 7
Adm cog 16-20 408 25 13 16 8 16 11 9 57 64 55 35 9
Adm cog >=21 504 24 13 16 9 17 11 8 62 70 68 23 10
Narcotic analgesic Overall 2130 72 55 45 36 59 16 13 65 77 63 26 11
Adm cog <=6 339 71 50 35 26 59 21 17 56 51 63 25 12
Adm cog 7-10 374 74 50 40 29 60 18 14 57 59 61 30 9
Adm cog 11-15 495 73 56 42 35 59 16 13 62 70 64 26 10
Adm cog 16-20 408 75 60 51 43 60 14 12 69 90 57 32 12
Adm cog >=21 504 69 59 52 46 58 14 11 78 100 69 19 12
Miscellaneous Psychotropic Overall 2130 18 8 15 6 16 19 15 69 75 9 78 13
Adm cog <=6 339 24 4 22 3 23 26 21 58 59 16 76 8
Adm cog 7-10 374 19 7 14 5 17 21 17 65 68 10 77 13
Adm cog 11-15 495 21 10 17 8 20 17 16 73 81 4 84 12
Adm cog 16-20 408 19 12 16 9 17 16 14 75 90 5 78 17
Adm cog >=21 504 10 7 7 6 8 13 9 72 98 14 69 18
Stimulant Overall 2130 28 7 22 6 26 23 18 66 72 5 83 12
Adm cog <=6 339 57 16 41 12 54 29 27 67 75 4 79 17
Adm cog 7-10 374 44 10 36 9 43 25 22 68 78 3 89 8
Adm cog 11-15 495 25 5 20 4 23 19 16 63 64 5 84 11
Adm cog 16-20 408 15 5 12 4 13 13 11 65 67 9 78 14
Adm cog >=21 504 8 2 7 2 6 11 9 64 64 10 79 10
*

10 patients were excluded due to missing Admission FIM cognitive score

r'cd = received

sched = scheduled

Table 3. Sample size by week and percent with psychoactive medication administration by pharmaceutical class, week of rehabilitation, and level of cognitive function at admission.
Pharmaceutical class FIM Cognitive score at admission % rec'd week 1 % rec'd week 2 % rec'd week 3 % rec'd week 4 % rec'd week 5 % rec'd week 6 % rec'd week 7 % rec'd week 8 % rec'd week 9
Sample size by week* Overall 2130 2008 1551 1065 707 482 339 215 153
Adm cog <=6 339 333 323 288 223 160 107 79 59
Adm cog 7-10 374 371 337 266 185 121 90 56 34
Adm cog 11-15 495 482 387 236 127 73 48 30 27
Adm cog 16-20 408 381 253 126 74 50 42 24 17
Adm cog >=21 504 432 242 140 91 71 47 23 16
Anxiolytic Overall 24 22 23 23 26 28 30 32 31
Adm cog <=6 31 30 29 30 30 29 30 33 27
Adm cog 7-10 30 27 27 26 30 35 40 41 38
Adm cog 11-15 24 23 21 20 24 29 29 30 37
Adm cog 16-20 23 19 19 19 19 20 24 21 29
Adm cog >=21 16 14 14 13 14 15 15 22 19
Anticonvulsant Overall 39 39 39 41 42 43 42 46 48
Adm cog <=6 37 38 38 42 41 45 43 49 59
Adm cog 7-10 39 39 41 45 52 52 51 52 53
Adm cog 11-15 38 40 41 43 46 48 46 50 52
Adm cog 16-20 41 41 40 41 49 44 40 29 24
Adm cog >=21 39 34 31 23 12 13 13 22 19
Antidepressant Overall 56 59 62 65 66 67 66 67 69
Adm cog <=6 59 65 65 68 71 72 71 63 73
Adm cog 7-10 61 65 69 68 71 72 74 70 62
Adm cog 11-15 58 59 62 67 61 63 65 73 78
Adm cog 16-20 60 63 64 64 68 60 52 63 59
Adm cog >=21 44 47 47 48 51 54 53 65 63
Antiparkinson Overall 16 21 25 29 32 32 32 31 27
Adm cog <=6 35 41 41 44 44 45 45 42 36
Adm cog 7-10 26 29 32 35 38 42 43 39 32
Adm cog 11-15 15 19 22 25 28 30 25 27 30
Adm cog 16-20 9 14 15 18 18 12 14 13 12
Adm cog >=21 4 6 7 6 4 3 0 0 0
Antipsychotic Overall 20 19 19 20 21 22 23 24 29
Adm cog <=6 25 24 24 27 27 30 33 35 42
Adm cog 7-10 26 25 23 23 23 25 27 27 29
Adm cog 11-15 24 21 20 21 19 15 17 17 26
Adm cog 16-20 19 19 19 15 16 18 17 8 6
Adm cog >=21 9 7 8 7 7 7 6 4 6
Hypnotic Overall 21 22 24 26 26 26 25 25 27
Adm cog <=6 21 24 23 26 26 25 22 25 25
Adm cog 7-10 25 26 29 32 30 34 33 34 38
Adm cog 11-15 23 23 26 31 32 36 35 33 33
Adm cog 16-20 18 18 19 18 18 12 10 0 6
Adm cog >=21 20 18 18 16 14 14 15 13 19
Narcotic analgesic Overall 65 60 55 49 49 45 40 42 40
Adm cog <=6 59 56 52 47 46 45 41 42 41
Adm cog 7-10 63 59 52 47 50 48 44 48 38
Adm cog 11-15 67 60 55 51 50 52 46 47 48
Adm cog 16-20 69 63 58 52 54 36 31 25 29
Adm cog >=21 66 62 59 48 42 37 30 35 38
Miscellaneous Psychotropic Overall 11 13 15 14 16 18 19 20 18
Adm cog <=6 6 10 15 19 20 23 27 28 25
Adm cog 7-10 11 13 13 13 16 18 14 20 15
Adm cog 11-15 14 16 18 16 16 15 23 23 19
Adm cog 16-20 15 18 18 14 14 14 12 13 12
Adm cog >=21 8 8 7 6 7 10 9 4 0
Stimulant Overall 15 21 27 33 38 39 38 35 34
Adm cog <=6 30 42 46 51 56 57 57 52 49
Adm cog 7-10 25 33 38 43 44 51 49 45 38
Adm cog 11-15 13 18 21 28 33 34 35 30 37
Adm cog 16-20 9 13 13 11 14 6 5 0 0
Adm cog >=21 4 6 7 7 5 4 4 0 0
*

10 patients were excluded due to missing Admission FIM cognitive score

r'cd = received

The most commonly prescribed agents were narcotic analgesics (72% of the sample), followed in decreasing frequency by antidepressants (67%), anticonvulsants (47%), antianxiety agents (33%), hypnotics (30%), stimulants (28%), antiparkinson agents (25%), antipsychotics (25%), and miscellaneous psychotropics (18%). Expanded detail on the frequency of specific medications at the level of general mechanism within each pharmaceutical class by admission FIM Cognitive category, is available in tables A and B in the supplemental digital content.

Anxiolytic agents

The percentage of patients administered anxiolytic medication remained roughly the same from admission to discharge for the overall sample and for all FIM Cognitive subgroups. Only 19% received an anxiolytic during the first 2 days and 19% during the last 2 days, with 33% receiving this class at some point during the stay. The primary anxiolytics prescribed were benzodiazepines, with 29% of patients receiving them at some point during the rehabilitation stay––approximately half of individuals receiving it on a regular basis, and half on an “as needed” basis. Lorazepam was the most common benzodiazepine prescribed, accounting for 68% of the benzodiazepine-based anxiolytics administered, followed by clonazepam (12%), and alprazolam (10%). H1 receptor antagonists (i.e., hydroxyzine) were rarely utilized, and were prescribed PRN more often than scheduled. Of the entire sample, 7% of patients received buspirone, which was predominately prescribed on a scheduled basis, with usage increasing over the RLOS.

Anticonvulsant agents

Nearly half (47%) of patients received an anticonvulsant at some point during their rehabilitation stay, with 35% receiving 1 during the first 2 days, 39% the last 2 days, and 28% during both intervals. The most commonly used anticonvulsants were the calcium channel and sodium channel antagonists. The most common calcium channel antagonist used was levetiracetam (61% of agents in this class administered to 21% of the sample); the most common sodium channel antagonists used were valproic acid (39% of agents in this class), phenytoin (37%), and carbamazepine (9%).

Antidepressant agents

Two thirds of the patients (67%) received an antidepressant at some point during their rehabilitation stay, 44% during the first 2 days, 55% the last 2 days, and 37% during both intervals. The most commonly used antidepressants were serotonin antagonist and reuptake inhibitor (SARI; i.e., trazodone) and selective serotonin reuptake inhibitors (SSRIs; i.e., citalopram, escitalopram, fluoxetine, paroxetine, sertraline), while only a minority of patients received tricyclic antidepressants (TCAs; i.e., desipramine, nortriptyline, amitriptyline, and doxepin), norepinephrine-dopamine reuptake inhibitor (NDRI; i.e., bupropion), noradrenergic and specific serotonergic antidepressant (NaSSA; i.e., mirtazapine), and serotonin and norepinephrine reuptake inhibitors (SNRIs; i.e., duloxetine, venlafaxine, and milnacipran). Antidepressants were generally prescribed as scheduled with only occasional PRN use, with the exception of SARI (i.e. trazodone), which was used in both manners, consistent the common practice of prescribing this agent for insomnia.

Antiparkinson agents

Antiparkinson agents were administered to only 25% of the patients at some point during rehabilitation, with use substantially increasing over the stay, from 11% receiving this class of medication during the first 2 days to 20% the last 2 days. The most commonly used antiparkinson was an NMDA antagonist (i.e., amantadine) administered to 17% of the sample, followed by dopamine agonist (i.e., bromocriptine, pramipexole, ropinirole). These agents were generally administered on a scheduled basis with rare PRN use. Bromocriptine accounted for 95% of the dopamine agonists administered.

Stimulant agents

Stimulants were administered to only 28% of the sample. Similar to the antiparkinson and miscellaneous therapeutic agents, stimulants were predominately started after admission. Use of stimulants increased over the course of the stay, and these agents were commonly used among those with long RLOS. Patterns of administration appear consistent within the various agents contained in the stimulant class. The most commonly used were the norepinephrine-dopamine-5HT agonists (i.e., agents containing amphetamine, dextroamphetamine, or methylphenidate) that were used by 23% of the sample. Less commonly used were armodafinil/modafinil and the norepinephrine agonist atomoxetine. The stimulant agents were generally used on a scheduled basis.

Antipsychotic agents

Antipsychotic agents were received by a quarter of the sample at some point during their stay. The overall percentage of use did not increase during the stay, with 16% receiving it the first 2 days and 15% the last 2 days, and 10% receiving during both intervals. Second generation antipsychotics were administered more frequently (24% of the sample) than first generation ones (3%). Second generation medications were most commonly received as scheduled but were also used PRN. On the other hand, first generation antipsychotics were more often administered PRN than scheduled. Those with longer RLOS had slightly higher usage. Of the second generation antipsychotics administered, quetiapine accounted for 48%, followed by risperidone (19%), olanzapine (15%), and ziprasidone (14%).

Hypnotic agents

Hypnotic agents were administered to 30% of the sample. Use increased slightly from admission to later in the stay and was particularly common for those with longer RLOS. Most commonly prescribed in this class were non-benzodiazepine GABA-A agonists [i.e., zolpidem (88% of the class), eszopiclone (11%), zaleplon (<1%)], followed by occasional (5%) use of benzodiazepine GABA-A agonists [i.e., temazepam (62% of the class), midazolam (38%)] and 3% use of other hypnotics (i.e., chloral hydrate, propofol, phenobarbital). Melatonin agonists were rarely used. In general, hypnotics were slightly more likely to be used PRN than scheduled.

Miscellaneous Psychotropic agents

Miscellaneous psychotropics were used relatively less often than other agents, with 18% of patients receiving 1 of these agents at some point during their rehabilitation stay. They were most commonly initiated later in the stay and more frequently administered to those with longer RLOS. The most commonly prescribed in this class were acetylcholinesterase inhibitors (AChE-I; i.e., donepezil, physostigmine, rivastigmine) at 9% and “other” (i.e., glatiramer acetate, interferon beta 1a, nicotine, varenicline) at 9%. The AChE-I were generally prescribed after rehabilitation admission and later in the stay. Use was greatest in the later weeks of the rehabilitation stay and for those with longer RLOS. On the other hand, administration of the “other” psychotherapeutics was greatest during the first 2 days of rehabilitation with decreased use over the remaining stay. For those with longer RLOS, these agents were used less over time. These findings are largely accounted for by the prescription of nicotine or nicotine patch, which accounted for 98% of the use in the “other” category. This class of medications was most commonly administered as scheduled with occasional PRN use. The AChE-I were used PRN for 17% of the patients receiving this agent.

Narcotic Analgesics

The majority of patients received narcotics during their rehabilitation stay (72% overall) with a high use across FIM categories, even among those with lower levels of function. Most of the use occurred at admission (55% of sample during the first 2 days of rehabilitation) with decreased use occurring over the rehabilitation stay––45% of the sample received narcotic analgesics during last 2 days of rehabilitation. Narcotics were consumed for an average of 16 days, accounting for a mean 65% of the RLOS administered. Narcotics were received as both scheduled and PRN. PRN administration was used as commonly in the lower functioning group who are expected to have impaired communication as in the higher functioning groups. Scheduled use occurred across functional groups with less scheduled narcotic administration in the highest functioning group.

Relation of Patient Factors and Medication Administration

Table 4 shows the relationship between receiving a medication from a psychotropic pharmaceutical class at any time during rehabilitation and pre-injury characteristics and injury related variables. Age was highly associated with receiving most medications, the exception being hypnotics. In general, younger patients were more likely to receive anxiolytics, antidepressants, antiparkinson, stimulants, antipsychotics, and narcotic analgesics. In contrast, older patients were more likely to receive anticonvulsants and miscellaneous psychotropics. Males were more likely to receive antipsychotics. History of psychosis, bipolar disorder, or schizophrenia was also associated with being more likely to receive an antipsychotic, but was unrelated to receiving other classes of medications. Anxiolytics, antidepressants, and hypnotics were less likely to be used in minority populations. Anxiolytics, anticonvulsants, antidepressants, and narcotic analgesics were more likely to be used when there was a history of depression or anxiety (premorbid history or during acute care). Anti-depressants, anti-psychotics, and psychotropics were more likely to be used when a patient had a prior history of substance abuse.

Table 4. Bivariate Associations of Patient Pre-injury and Injury Characteristics with Ever Receiving Medication During Rehabilitation.

Variable Anxiolytic Anti-Convulsant Anti-Depressant Anti-Parkinson Anti-Psychotic Hypnotic Narcotic-
Analgesic
Miscellaneous
Psychotropic
Stimulant
% or
Mean(SD)*
p-
value
% or
Mean(SD)*
p- value % or
Mean(SD)*
p-
value
% or
Mean(SD)*
p-
value
% or
Mean(SD)*
p- value % or
Mean(SD)*
p-
value
% or
Mean(SD)*
p-
value
% or
Mean(SD)*
p-
value
% or
Mean(SD)*
p-
value
Age (continuous) 42(19); -3 <.001 47(22); 5 <.001 44(20); -3 0.008 42(20); -3 0.004 41(19); -4 <.001 44(20); -1 0.526 42(20); -8 <.001 47(20); 4 0.001 40(20); -6 <.001
Age (category) <.001 <.001 <.001 0.009 <.001 0.018 <.001 <.001 <.001
<30 years 32% . 41% . 67% . 26% . 27% . 28% . 76% . 13% . 33% .
≥30-<45 years 40% . 44% . 75% . 30% . 32% . 35% . 82% . 22% . 30% .
≥ 45-<65 years 37% . 50% . 69% . 25% . 26% . 31% . 74% . 21% . 25% .
≥ 65-<75 years 26% . 57% . 59% . 21% . 17% . 31% . 61% . 16% . 20% .
≥75-<85 years 19% . 58% . 61% . 16% . 20% . 28% . 53% . 23% . 19% .
≥85 years 15% . 44% . 46% . 18% . 6% . 17% . 39% . 17% . 14% .
Gender 0.959 0.408 0.235 0.082 <.001 0.314 0.516 0.077 0.481
Female 33% . 48% . 65% . 22% . 20% . 28% . 73% . 16% . 26% .
Male 33% . 46% . 68% . 26% . 28% . 31% . 72% . 19% . 28% .
Race/Ethnicity <.001 0.011 <.001 0.014 0.004 <.001 0.01 0.003 0.015
Asian/Other/Unknown 19% . 30% . 47% . 18% . 19% . 17% . 58% . 9% . 16% .
Black 27% . 49% . 58% . 31% . 23% . 24% . 71% . 14% . 27% .
White Non-hispanic 35% . 47% . 71% . 24% . 27% . 32% . 74% . 20% . 29% .
White Hispanic 24% . 47% . 60% . 19% . 15% . 23% . 70% . 15% . 21% .
History of drug abuse 0.001 0.957 <.001 0.664 <.001 0.178 <.001 <.001 0.51
No 31% . 47% . 65% . 25% . 23% . 29% . 70% . 16% . 28% .
Yes 39% . 47% . 74% . 26% . 35% . 32% . 81% . 27% . 26% .
History of alcohol abuse 0.06 0.526 <.001 0.075 <.001 0.03 0.511 <.001 0.649
No 31% . 47% . 64% . 26% . 21% . 31% . 72% . 14% . 28% .
Yes 35% . 46% . 72% . 23% . 33% . 27% . 73% . 26% . 27% .
History of psychosis/bipolar disorder/schizophrenia 0.049 0.05 0.908 0.9 <.001 0.556 0.185 0.091 <.001
No 32% . 46% . 67% . 25% . 24% . 30% . 72% . 18% . 28% .
Yes 43% . 57% . 66% . 24% . 55% . 33% . 79% . 25% . 12% .
Depression prior to or during acute care <.001 <.001 <.001 0.441 0.002 <.001 <.001 0.385 0.11
No 30% . 44% . 61% . 24% . 24% . 27% . 70% . 17% . 27% .
Yes 39% . 53% . 82% . 26% . 30% . 36% . 79% . 19% . 30% .
Anxiety prior to or during acute care <.001 <.001 <.001 0.448 <.001 0.064 <.001 0.355 0.502
No 29% . 45% . 64% . 25% . 24% . 29% . 69% . 18% . 28% .
Yes 50% . 55% . 80% . 23% . 33% . 34% . 86% . 19% . 26% .
Post-resuscitation Glasgow Coma Scale score <.001 <.001 <.001 <.001 0.036 0.003 <.001 0.238 <.001
Intubated/Sedated 41% . 47% . 78% . 31% . 29% . 39% . 79% . 22% . 36% .
Mild (13-15) 29% . 46% . 65% . 14% . 21% . 26% . 76% . 18% . 15% .
Moderate (9-12) 27% . 33% . 64% . 17% . 24% . 28% . 76% . 13% . 24% .
Severe (3-8) 36% . 40% . 71% . 31% . 29% . 31% . 73% . 17% . 36% .
Missing 29% . 57% . 60% . 23% . 23% . 28% . 67% . 18% . 23% .
*

%=percent of patients with that characteristic who received the specified medication class; mean (SD) = mean (SD) for those patients with that characteristic who received the specified medication class

The difference between those who received the specified medication class and those who did not receive the specified medication class.

A negative value indicates that the characteristic is less common for those who received the medication class as opposed to those who did not recieve the medication class (e.g. -3 for age indicates patients who received anxiolytics were on average 3 years younger than those who did not).

In contrast to table 4, table 5 shows the relationship between having ever been administered a medication and patient characteristics during the rehabilitation stay. Multiple indices of more severe impairment (percent of stay agitated, effort given in therapies, severity of brain impairment, severity of non-brain comorbidities, and length of PTA) were related to increased drug administration in nearly all categories. Other indices indicative of greater difficulties during rehabilitation (i.e., percent of days in pain and percent of days with less than 5 hours sleep) were related to increased medication administration with the exceptions of anti-psychotics and psychotropics. Having seizures during rehabilitation increased the likelihood of administration of anticonvulsants as well as narcotic analgesics.

Table 5. Bivariate Associations of Rehabilitation Characteristics with Ever Receiving Medication During Rehabilitation.

Variable Anxiolytic Anti-Convulsant Anti-Depressant Anti-Parkinson Anti-Psychotic Hypnotic Narcotic-
Analgesic
Miscellaneous
Psychotropic
Stimulant
% or
Mean(SD)*
p-
value
% or
Mean(SD)*
p-
value
% or
Mean(SD)*
p-
value
% or
Mean(SD)*
p-
value
% or
Mean(SD)*
p-
value
% or
Mean(SD)*
p-
value
% or
Mean(SD)*
p-
value
% or
Mean(SD)*
p-
value
% or
Mean(SD)*
p-
value
Seizure any time up to rehabilitation discharge 0.023 <.001 0.153 0.621 0.208 0.463 <.001 0.011 0.785
 No seizure 32% . 41% . 68% . 25% . 26% . 30% . 74% . 17% . 28% .
 Yes, 1 or more seizure 38% . 80% . 63% . 26% . 23% . 28% . 64% . 23% . 28% .
Days from injury to rehabilitation admission 35(42); 8 <.001 28(32); -3 0.036 31(37); 7 <.001 37(40); 10 <.001 29(32); 0 0.998 31(30); 2 0.159 28(33); -6 0.001 29(33); -0 0.937 37(43); 10 <.001
Admission FIM cognitive category <.001 0.031 <.001 <.001 <.001 <.001 0.204 <.001 <.001
 Score <=6 48% . 50% . 77% . 53% . 38% . 36% . 71% . 24% . 57% .
 Score 7-10 44% . 52% . 76% . 40% . 34% . 37% . 74% . 19% . 44% .
 Score 11-15 31% . 46% . 66% . 21% . 28% . 31% . 73% . 21% . 25% .
 Score 16-20 28% . 46% . 69% . 15% . 22% . 25% . 75% . 19% . 15% .
 Score >=21 20% . 41% . 53% . 6% . 10% . 24% . 69% . 10% . 8% .
Admission FIM motor category <.001 0.167 <.001 <.001 <.001 <.001 <.001 0.604 <.001
 Score <22.05 44% . 50% . 77% . 43% . 29% . 37% . 77% . 19% . 45% .
 Score 22.05-28.75 30% . 46% . 58% . 21% . 30% . 28% . 72% . 16% . 27% .
 Score 28.75-40.65 31% . 46% . 70% . 16% . 25% . 27% . 77% . 19% . 18% .
 Score 40.65-44.25 23% . 46% . 67% . 12% . 23% . 26% . 79% . 14% . 14% .
 Score 44.25-53.36 24% . 46% . 66% . 14% . 26% . 27% . 68% . 17% . 16% .
 Score >53.36 21% . 41% . 51% . 9% . 17% . 23% . 59% . 17% . 14% .
Percent of days with <5 hours sleep 36(23); 8 <.001 33(23); 3 <.001 32(22); 2 0.079 34(22); 5 <.001 33(22); 3 0.005 36(23); 8 <.001 33(23); 7 <.001 31(23); 0 0.936 31(20); -0 0.942
Average Therapy Level of Effort 4(1); -0.4 <.001 4(1); -0.2 <.001 4(1); -0.3 <.001 4(1); -0.8 <.001 4(1); -0.4 <.001 4(1); -0.2 <.001 4(1); 0.0 0.618 4(1); -0.3 <.001 4(1); -0.6 <.001
Maximum CSI brain injury component 59(25); 16 <.001 50(25); 4 <.001 52(25); 12 <.001 67(23); 24 <.001 58(22); 13 <.001 55(25); 10 <.001 50(25); 5 <.001 55(24); 8 <.001 66(23); 25 <.001
Maximum CSI non-brain injury component 30(24); 7 <.001 26(22); 3 <.001 27(21); 6 <.001 30(23); 6 <.001 30(23); 7 <.001 30(23); 7 <.001 27(22); 9 <.001 24(19); -1 0.578 28(21); 5 <.001
Posttraumatic amnesia duration (days) 49(54); 18 <.001 40(46); 4 0.045 42(47); 14 <.001 61(54); 32 <.001 45(47); 10 <.001 44(45); 9 <.001 37(42); -2 0.324 43(44); 7 0.006 60(54); 31 <.001
Time of posttraumatic amnesia clearing <.001 0.151 <.001 <.001 <.001 <.001 0.007 <.001 <.001
 Cleared PTA prior to rehab admission 26% . 47% . 64% . 11% . 17% . 28% . 78% . 14% . 12% .
 Cleared PTA on rehab admission day 14% . 40% . 44% . 12% . 9% . 9% . 88% . 14% . 9% .
 Cleared PTA after rehab admission day 39% . 50% . 72% . 36% . 32% . 33% . 73% . 21% . 40% .
Percent of days with pain score 7+ 20(26); 4 <.001 20(27); 5 <.001 19(26); 7 <.001 12(20); -7 <.001 18(24); 1 0.528 19(26); 3 0.007 23(27); 20 <.001 19(27); 2 0.117 11(18); -8 <.001
Percent of rehabilitation stay agitated 16(25); 10 <.001 11(22); 5 <.001 11(21); 5 <.001 13(22); 5 <.001 20(27); 15 <.001 13(24); 6 <.001 9(20); 2 0.04 14(24); 6 <.001 13(23); 6 <.001
*

%=percent of patients with that characteristic who received the specified medication class; mean (SD) = mean (SD) for those patients with that characteristic who received the specified medication class

The difference between those who received the specified medication class and those who did not receive the specified medication class.

A negative value indicates that the characteristic is less common for those who received the medication class as opposed to those who did not recieve the medication class (e.g. 8 for days injury to rehab admission indicates that patient who received anxiolytics had on average 8 days longer days from injury to rehab admission than those who did not).

Psychotropic medication exposure summary and concurrent use

Table 6 depicts the percentage of patients receiving specific quantities of psychotropic medications during rehabilitation, overall, and by admission cognitive category. Only 5.0% of the patients were never administered psychotropic medications during their rehabilitation stay, while 8.5% were prescribed only 1 of the psychotropic medications; 31.8% were prescribed 6 or more of these agents at some point during their stay. These results could occur if all 6 were prescribed simultaneously, or sequentially (1 after the other) while the physician was searching for an effective drug. More likely, some were given at the same time, with some dropping off and others being added. During the first 2 days of rehabilitation 5.5% of patients were on at least 6 psychotropic medications, while 13.5% were on at least 6 of these medications during the last 2 rehabilitation days. In general, those in the lower admission FIM Cognitive categories received a greater number of psychotropic medications studied (3-8 agents) than those in higher FIM Cognitive categories, in which most received 0-5 agents.

Table 6. Percent of patients who received psychoactive medications during rehabilitaiton by number received and admission cogntitive functional level.
Overall Adm FIM Cog ≤6 Adm FIM Cog 7-10 Adm FIM Cog 11-15 Adm FIM Cog 16-20 Adm FIM Cog ≥ 21
#
psycho
active
meds
rec'd
%
ever
rec'd*
%
rec'd*
1st 2
days
%
rec'd*
last 2
days
%
ever
rec'd*
%
rec'd*
1st 2
days
%
rec'd*
last 2
days
%
ever
rec'd*
%
rec'd*
1st 2
days
%
rec'd*
last 2
days
%
ever
rec'd*
%
rec'd*
1st 2
days
%
rec'd*
last 2
days
%
ever
rec'd*
%
rec'd*
1st 2
days
%
rec'd*
last 2
days
%
ever
rec'd*
%
rec'd*
1st 2
days
%
rec'd*
last 2
days
0 5.0 13.0 12.7 2.4 13.9 7.7 2.4 11.8 9.4 4.4 11.5 14.1 4.2 9.6 10.8 10.1 17.3 18.7
1 8.5 21.9 17.3 3.2 15.6 10.3 4.5 20.6 10.7 7.5 23.0 15.2 8.6 22.1 19.9 16.3 25.6 27.4
2 12.6 23.4 17.4 5.9 24.2 15.9 7.0 21.1 14.7 10.7 21.6 16.6 15.7 25.2 19.1 20.4 25.0 19.8
3 14.6 17.5 17.0 9.7 18.0 16.2 11.2 18.2 16.8 17.4 19.4 19.0 16.2 18.6 17.6 16.3 13.9 15.5
4 14.3 12.0 13.9 10.9 14.2 16.2 12.6 11.0 17.4 15.8 14.5 14.1 17.2 10.5 14.2 14.3 10.1 9.1
5 12.9 6.6 8.4 14.7 6.2 10.6 13.9 9.6 10.4 13.5 6.1 8.5 14.2 7.4 8.1 9.5 4.4 5.2
6 10.4 3.5 6.0 11.2 5.0 11.2 15.5 5.3 7.0 13.1 1.8 7.1 8.8 3.9 3.4 4.4 2.4 2.8
7 7.3 1.4 3.8 12.4 1.8 6.8 11.0 1.3 7.2 5.5 1.4 2.2 4.7 1.7 3.7 5.0 1.0 0.8
8 4.9 0.5 1.5 8.8 0.9 2.9 7.2 0.8 2.7 4.4 0.2 0.8 3.9 0.5 1.0 2.0 0.4 0.8
9 3.5 0.1 1.1 6.5 0.3 1.2 4.5 0.3 1.6 2.2 0.0 1.6 3.7 0.2 1.0 1.6 0.0 0.0
10 2.2 0.0 0.3 4.1 0.0 0.0 4.5 0.0 1.1 1.8 0.2 0.4 1.2 0.0 0.2 0.2 0.0 0.0
>10 3.5 0.0 0.4 10.0 0.0 0.9 5.7 0.0 1.0 3.6 0.2 0.4 1.6 0.2 0.9 0.0 0.0 0.0
*

r'cd = received

Medication administration across sites

Medication administration patterns varied greatly across treatment sites as summarized in table 7. Sites with high antipsychotic use had lower use of anxiolytics, and vice versa. Sites with high antiparkinson administration had less antipsychotic use, and vice versa. For anticonvulsant use, most sites were similar except 1 site where 80% of their patients received an anticonvulsant agent during their rehabilitation stay. With a range of 7-31%, miscellaneous psychotropic agents were used relatively infrequently at some sites. Antidepressant use was uncommon at 1 site (27%), with use ranging 46-91% across the others. The site with the highest use of antidepressants had a practice pattern of using the antidepressants SARI and tertiary amine TCAs as their first line treatment of insomnia. Across sites, antiparkinson agent use ranged 1% - 57% and stimulants use 5-50%.

Table 7. Percent that Received Medication Classification by Treatment Site.
Variable Anxiolytic Anti-
Convulsant
Anti-
Depressant
Anti-
Parkinson
Anti-
Psychotic
Hypnotic Narcotic-
Analgesic
Miscellaneous
Psychotropic
Stimulant
% patients
received*
% patients
received*
% patients
received*
% patients
received*
% patients
received*
% patients
received*
% patients
received*
% patients
received*
% patients
received*
 Site 1 39% 42% 52% 10% 14% 23% 77% 11% 12%
 Site 2 50% 42% 91% 57% 18% 49% 87% 23% 50%
 Site 3 17% 44% 50% 4% 62% 13% 79% 9% 40%
 Site 4 20% 46% 64% 16% 13% 18% 78% 7% 5%
 Site 5 21% 60% 64% 25% 24% 21% 58% 37% 40%
 Site 6 20% 51% 46% 11% 27% 13% 48% 8% 28%
 Site 7 33% 52% 89% 25% 35% 29% 90% 31% 25%
 Site 8 27% 80% 27% 46% 23% 31% 49% 21% 6%
 Site 9 38% 47% 76% 33% 28% 44% 77% 15% 35%
 Site 10 22% 22% 48% 1% 15% 15% 33% 12% 5%
Average 29% 49% 61% 23% 26% 26% 68% 17% 25%
*

All p-values were <0.001 for differences across sites.

highest percentage for medication class

lowest percentage for medication class

Discussion

This large sample, multicenter study documents the extent to which psychotropic medications are administered to treat patients with TBI during inpatient rehabilitation. In 9 broad categories of medications, the percent of overall use varied from 18 to 72% with a mean of 42% (table 2: % ever received), and 31.8% were exposed to at least 6 of the psychotropic agents studied during rehabilitation (table 6). These results suggest 1) a strong “culture of intervention”20 with the prevalent use of unproven medications to advance recovery in this group of facilities that specialize in brain injury management; 2) an urgent need to control patient behavior; and/or 3) a strong desire to stimulate recovery. We found considerable variation across sites. Marked variation in clinical practice is likely a reflection of the relative lack of high quality research available in neuropharmacology post TBI. With the absence of solid data, clinicians may base their treatment decisions on information gleaned from accepted treatments for other impairment groups with similar problems to treat issues such as agitation, headache, pain, insomnia, and sleep disorder. In the absence of better evidence, the prescriber is often reliant on their subjective clinical impressions, expert opinion, and a multitude of case studies and open-label case series reinforced by and overlying natural recovery.

In this study, univariate analyses indicated potential differences related to age and race in the percentage of patients prescribed varying classes of medications. The extent to which younger patients may be more likely to be administered anxiolytics, antidepressants, antiparkinson/stimulants, antipsychotics, and narcotic analgesics requires further analysis that controls for injury severity and secondary conditions. Further testing for nonlinear relationships between age and medication administration (i.e., both very young and very old patients being less likely to be prescribed medications) is also warranted.21 Anxiolytics, antidepressants, antipsychotics, hypnotics, and antiparkinson agents were less likely to be used with ethnic minorities, particularly those of Asian and Hispanic descent. Given the relatively small number of Asian and Hispanic patients in this sample, further investigation is warranted to evaluate the extent that injury severity and secondary conditions versus unmeasured factors (such as differential cultural preferences or site differences in ethnicity and prescribing preferences) are related to medication use.

This study did not capture information about the primary symptom(s) that physicians targeted for each medication prescribed. Tables 4 and 5 indirectly provide insight into the potential variability in symptoms associated with the pharmaceutical classes of medication administered. For example, 29% of those who received anxiolytics did not have anxiety mentioned in their medical record (as having been present premorbidly, during acute care, or at the time of rehabilitation admission), suggesting that many may be treated with this class of medication for other reasons such as agitation or insomnia. Similarly, 61% of those who received antidepressants did not have mention of depression present premorbidly or during acute care, suggesting that pain, sleep disorders, and/or behavior are being treated by commonly prescribed medications that were classified as antidepressants. For example, the SARI trazodone is often used in this population for sleep induction. Similar findings were observed for antipsychotics (24% lacked mention of premorbid history of psychosis, bipolar disorder, or schizophrenia). Of those administered anticonvulsants, 41% did not have a seizure during acute care or rehabilitation indicating use for seizure prophylaxis or other reasons (such as behavior control or pain management). The broad range of medication applications highlights the importance of patient education and communication with co-treating physicians regarding the targeted use of medications prescribed at the time of discharge from rehabilitation and after.

Our univariate analyses found statistically significant center effects across pharmaceutical classes. Given the wide variability between centers with regard to age, time from injury to rehabilitation admission, injury etiology and severity, and levels of functional impairment,22 further analyses are required to determine the extent that center effects exist independent of other confounds. With the limited literature on neuropharmacology effectiveness post TBI to guide treatment decisions, practice variation at least between physicians would not be surprising.23,24

Antiparkinson and stimulant administration was low in comparison to the authors' expectations and in comparison to other psychotropic medications (narcotic analgesics, antidepressants, anticonvulsants, antianxiolytics, and hypnotics). Antiparkinson agents were administered to 25% of patients at some point during rehabilitation (most commonly amantadine and bromocriptine). In clinical practice, these medications are often used in the treatment of several rehabilitation relevant issues including: poor arousal, agitation, disinhibition, lack of initiation, akinetic mutism, and cognitive impairment. Similarly, stimulant administration (28% of sample received) was surprisingly low given that symptoms of inattention, lack of initiation, poor arousal, and slow processing speed are cardinal features of moderate and severe TBI. Stimulants were administered predominately to those with lower admission FIM Cognitive scores. The most commonly used stimulants were methylphenidate, modafanil, and atomoxetine. Considering the greater use of other classes (such as antidepressants, anticonvulsants, antianxiolytics, hypnotics), perhaps antiparkinson and stimulant agents could have a greater role in the management of the TBI patient (such as the agitated, confused, difficult to manage, or slow to recover patient) 25-30 than is currently being used by some physicians. In studies of subacute TBI, patients receiving methylphenidates have shown short-term improvements in attention, concentration, motor memory, cognitive processing speed, and overall function. 26,27 Scientific evidence suggests amantadine may help minimize the impact of many deficits commonly following TBI, particularly disordered consciousness, cognitive impairments, and behavioral dysregulation.3, 29-31

Conversely, prescription of narcotics was surprisingly high, despite the risk of their cognitive sedating properties. Narcotic use is very high across all functional cognitive levels, with nearly 75% of all patients receiving these medications at least once during their stay. While narcotics were overwhelmingly prescribed on a PRN basis, the median percent of days that patients were administered these medications suggests that in practice they were fairly regularly used. Applying these findings clinically, the clinician is advised to use caution with administering pain medication and consider incorporation of objective measures of function as well as pain into the assessment and ongoing administration.

Antipsychotic agents were received by 25% of the sample at some point during their stay. It is common for practitioners to use this class of medication to assist with controlling agitation post TBI. This particular use is somewhat controversial as the blocking of dopamine is not always considered to be productive in terms of recovery.1,4 However, second generation antipsychotics have less D2 dopamine receptor effect, and are thought to be preferable over first generation agents; though they still have a considerable side effect profile. Second generation antipsychotics have been proposed by some in the field as preferred treatment for agitation and psychosis due to TBI.32,33 Quetiapine accounted for 48% of the second generation antipsychotics administered, followed by risperdone (19%), olanzapine (15%), and ziprasidone (14%).

Future Research Directions

The use of this multi-center, longitudinal data to evaluate the effectiveness of medication treatments in real-world clinical settings offers both opportunities and challenges. Findings from this initial investigation of medication administration patterns during TBI inpatient rehabilitation provides valuable data that can inform the research design of future medication comparative effectiveness studies. Ninety percent of the patients in our study were administered 2 or more psychotropic medications during their stay, with 60% administered between 3 and 7. Because of the administration of multiple medications at the same time or within the short time frame of rehabilitation, future research requires that study designs carefully evaluate the effects of psychotropic medications alone and in combination on the primary outcomes of interest. Future research will also need to take into account dosing levels and duration of treatment, while controlling for participant-specific effects. Mixed effects quantile stratification propensity adjustment strategies for longitudinal analyses may be suited for such treatment effectiveness analyses.34,35 Based on our findings, participant effects that should be considered for stratified propensity adjustment for each primary outcome include age, timing of administration, history of Axis I mental health disorders, severity of cognitive impairment, and pain. The potential confounding effects of center and race should be further evaluated to determine whether these are true effects or are encapsulated within the covariates already listed for potential stratification adjustment. Evaluation, and where necessary, adjustment of individual covariates for nonlinear relationships and outlier effects is essential given the frequent observance of large SDs.

Study Limitations

The findings of this study represent the patterns of administration at highly specialized brain injury rehabilitation centers and may not represent the patterns of use at all rehabilitation units. In particular, this study may be unique in regards to the medical complexity and neurologic functional level of the patients, training and experience of the clinicians, academic environment, resources of the facilities, and demographics of the study sample (primarily Caucasian). The acute care hospital medical records were not consistently available, thus we did not include medications used during acute care. The study focused on key agents commonly used to improve arousal, behavior, function, and control central nervous system issues associated with TBI. The study was limited to 9 medications categories. There are several psychotropic medications that were not examined here, but were administered, such as alpha agonist and beta-blocking antihypertensive agents, metoclopramide, proton pump inhibitors, and a host of agents with anticholinergic effects.

The targeted goals for medication prescription are not known in this study. Medications designed and approved for 1 use are commonly used for other purposes. For instance, antidepressants may be useful for correction of sleep disorders, pain, and anxiety as well as depression. Anxiolytics may be used for sleep and behavior modification as well as anxiety. Anticonvulsants are commonly used for neuropathic pain and mood stabilization as well as seizure prevention or management. Antipsychotics may be administered for insomnia, anxiety, psychosis, and agitation. The present study reveals the type of psychotropic agents used but not the purpose. Data about severity of injury, duration of PTA, agitation, pain, seizures, sleep, and cognition were assessed for association with administration of these agents, and thereby, provide some information on use. However, caution should be used in presuming the use of the medications in this study.

Conclusion

Many psychotropic medications are used during inpatient rehabilitation. A wide variety of applications are perceived for each class of psychotropic medications and individual agents within classes. Knowledge of prescribing patterns may inform further research such as comparative effectiveness studies. In general, lower admission FIM Cognitive groups were administered more of the medications under investigation, compared to those with higher cognitive function at admission. Considerable site variation existed regarding medications administered.

Supplementary Material

Supplemental Digital Content Table A: Expanded psychoactive medications data administered by pharmaceutical class and level of admission cognitive function

Acknowledgments

We gratefully acknowledge the contributions of clinical and research staff at each of the 10 inpatient rehabilitation facilities represented in the Improving Outcomes in Acute Rehabilitation for TBI Study and Individualized Planning for the First Year Following Acute Rehabilitation, collectively known as the TBI Practice Based Evidence (TBI-PBE) study. The study site directors included: John D. Corrigan, PhD and Jennifer Bogner, PhD (Ohio Regional TBIMS at Ohio State University, Columbus, OH); Nora Cullen, MD (Toronto Rehabilitation Institute, Toronto, ON Canada); Cynthia L. Beaulieu, PhD (Brooks Rehabilitation Hospital, Jacksonville, FL); Flora M. Hammond, MD (Carolinas Rehabilitation, Charlotte, NC [now at Indiana University]); David K. Ryser, MD (Neuro Specialty Rehabilitation Unit, Intermountain Medical Center, Salt Lake City, UT); Murray E. Brandstater, MD (Loma Linda University Medical Center, Loma Linda, CA); Marcel P. Dijkers, PhD (Rehabilitation Medicine, Icahn School of Medicine at Mount Sinai, New York, NY); William Garmoe, PhD (Medstar National Rehabilitation Hospital, Washington, DC); James A. Young, MD (Physical Medicine and Rehabilitation, Rush University Medical Center, Chicago, IL); Ronald T. Seel, PhD (Brain Injury Research, Shepherd Center, Atlanta, GA).

We want to acknowledge members of the staff of the Institute for Clinical Outcomes Research, International Severity Information Systems, Inc, Salt Lake City, UT, who also contributed significantly to the success of this study: Susan D. Horn, PhD (Senior Scientist); Randall J. Smout, MS (Vice President, Analytic Systems); Ryan S. Barrett (Project Manager and Analyst); Michael Watkiss (Study Coordinator); and Patrick B. Brown (Project Manager and Systems Administrator). In addition, we acknowledge the help of Gale G. Whiteneck, PhD (Craig Hospital, Englewood, CO).

Funding for this study came from the National Institutes of Health, National Center for Medical Rehabilitation Research (grant 1R01HD050439-01), the National Institute on Disability and Rehabilitation Research (grant H133A080023), and the Ontario Neurotrauma Foundation (grant 2007-ABI-ISIS-525).

Abbreviations

AChE-I

Acetylcholinesterase inhibitors

APAP

Acetaminophen

COMT

Catechol-O-methyltransferase

CSI

Comprehensive Severity Index

FIM

Functional Independence Measure

GABA-A

Gaba-aminobutyric acid-A

MAO

Monoamine oxidase

NaSSAs

Noradrenergic and specific serotonergic antidepressants

NDRI

Norepinephrine-dopamine reuptake inhibitor

NMDA

N-Methyl-D-aspartate

PRN

Pro re nata (as needed)

PBE

Practice-based evidence

PTA

Posttraumatic amnesia

RLOS

Rehabilitation length of stay

SARI

Serotonin antagonist and reuptake inhibitors

SNRI

Serotonin and norepinephrine reuptake inhibitors

SSRI

Selective serotonin reuptake inhibitor

TBI

Traumatic brain injury

TCA

Tricyclic antidepressant

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Mysiw WJ, Bogner JA, Corrigan JD, Fugate LP, Clinchot DM, Kadyan V. The impact of acute care medications on rehabilitation outcome after traumatic brain injury. Brain Inj. 2006 Aug;20(9):905–11. doi: 10.1080/02699050600743972. [DOI] [PubMed] [Google Scholar]
  • 2.Francisco GE, Walker WC, Zasler ND, Bouffard MH. Pharmacologic management of neurobehavioral sequelae of traumatic brain injury: a secondary survey of current physiatric practice. Brain Inj. 2007;21(10):1007–1014. doi: 10.1080/02699050701559558. [DOI] [PubMed] [Google Scholar]
  • 3.Giacino JT, Whyte J, Bagiella E, Kalmar K, Childs N, Khademi A, Eifert B, Long D, Katz DI, Cho S, Yablon SA, Luther M, Hammond FM, Nordenbo A, Novak P, Mercer W, Maurer-Karattup P, Sherer M. Placebo-controlled trial of amantadine for severe traumatic brain injury. N Engl J Med. 2012 Mar 1;366(9):819–26. doi: 10.1056/NEJMoa1102609. [DOI] [PubMed] [Google Scholar]
  • 4.Madeira G, Montmirail C, Decat M, Gersdorff M. TRT: Results after one year treatment. Laryngol Otol Rhinol. 2007;128(3):145–148. [PubMed] [Google Scholar]
  • 5.Feeney DM, Gonzalez A, Law WA. Amphetamine, haloperidol, and experience interact to affect rate of recovery after motor cortex injury. Science. 1982;217:855–857. doi: 10.1126/science.7100929. [DOI] [PubMed] [Google Scholar]
  • 6.Hoffman AN, Cheng JP, Zafonte RD, Kline AE. Administration of haloperidol and risperdone after neurobehavioral testing hinders the recovery of traumatic brain injury-induced deficits. Life Sciences. 2008;83:602–607. doi: 10.1016/j.lfs.2008.08.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Massagli TL. Neurobehavioral effects of phenytoin, carbamazepine, and valproic acid: implications for use in traumatic brain injury. Arch Phys Med Rehabil. 1991;72:219–226. [PubMed] [Google Scholar]
  • 8.Balon R. SSRI-associated sexual dysfunction. Am J Psychiatry. 2006;163:1504–1509. doi: 10.1176/ajp.2006.163.9.1504. [DOI] [PubMed] [Google Scholar]
  • 9.Edwards JG. Unwanted effects of psychotropic drugs. 1. Effects on human physiological systems, mechanisms and methods of assessment. In: King DJ, editor. Seminars in Clinical Psychopharmacology. 2nd. London: Gaskell; 2004a. pp. 573–600. [Google Scholar]
  • 10.Edwards JG. Unwanted effects of psychotropic drugs. 2. Drug interactions, effects during pregnancy and breast-feeding, pharmaco-vigilance and medico-legal considerations. In: King DJ, editor. Seminars in Clinical Psychopharmacology. 2nd. London: Gaskell; 2004b. pp. 601–660. [Google Scholar]
  • 11.Murphy MP, Carmine H, Kolakowsky-Hayner S. Modifiable and nonmodifiable risk factors for falls after traumatic brain injury: an exploratory investigation with implications for medication use. Rehabil Nurs. 2014 May-Jun;39(3):113–22. doi: 10.1002/rnj.89. [DOI] [PubMed] [Google Scholar]
  • 12.Yasseen B, Colantonio A, Ratcliff G. Prescription medication use in persons many years following traumatic brain injury. Brain Inj. 2008 Sep;22(10):752–7. doi: 10.1080/02699050802320132. [DOI] [PubMed] [Google Scholar]
  • 13.Fugate LP, Spacek LA, Kresty LA, Levy CE, Johnson JC, Mysiw WJ. Measurement and treatment of agitation following traumatic brain injury: II. A survey of the brain injury special interest group of the American academy of physical medicine and rehabilitation. Arch Phys Med Rehabil. 1997;78(9):924–928. doi: 10.1016/s0003-9993(97)90051-4. [DOI] [PubMed] [Google Scholar]
  • 14.Horn SD, Corrigan JD, Bogner J, Hammond FM, Seel RT, Smout RJ, Barrett RS, Watkiss M, Dijkers MP, Whiteneck G. Traumatic Brain Injury Practice-Based Evidence Study: Design and Description of Patient, Treatment, and Outcome Variables. Arch Phys Med Rehabil. Paper A for TBI supplement can be added but we have not finalized the title yet. [Google Scholar]
  • 15.Ryser DK, Egger MJ, Horn SD, Handrahan D, Gandhi P, Bigler ED. Measuring medical complexity during inpatient rehabilitation following traumatic brain injury. Arch Phys Med Rehabil. 2005;86:1108–1117. doi: 10.1016/j.apmr.2004.11.041. [DOI] [PubMed] [Google Scholar]
  • 16.Heinemann AW, Linacre A, Wright BD, Hamilton B, Granger C. Measurement characteristics of the functional independence measure. Topics in stroke rehabilitation. 1994;1:1–15. doi: 10.1080/10749357.1994.11754030. [DOI] [PubMed] [Google Scholar]
  • 17.Bogner J, Corrigan J, Beaulieu C, Hammond FM, Horn S, Seel R, Garmoe W, Randall Smout, Barrett R, Cullen N, Young JA, Waszkiewicz MA, Rosenthal J, Simental L, Reddin C. Changes in Agitation Associated with Sleep-Wake Cycle, Cognitive Functioning, and Pharmacologic Treatment during Rehabilitation for TBI Patients. Arch Phys Med Rehabil. Paper H for this TBI supplement can be added before publication. [Google Scholar]
  • 18.Seel R, Horn SD, Barrett RS, Smout RJ, Bogner J, Deutscher D, Dijkers M, Corrigan J, Garmoe W, Sullivan E. Average Level of Effort per Week for Each Therapy Discipline and Associations with TBI Patient Pre-Injury and Injury Characteristics. Arch Phys Med Rehabil. Paper G for this TBI supplement can be added before publication. [Google Scholar]
  • 19.Forchheimer MB, Richards JS, Chiodo AE, Thomas N, Bryce TN, Dyson-Hudson TA. Cut point determination in the measurement of pain and its relationship to the psychosocial and functional measures after traumatic spinal cord injury: a retrospective model spinal cord injury system analysis. Arch Phys Med Rehabil. 2011;92(3):419–424. doi: 10.1016/j.apmr.2010.08.029. [DOI] [PubMed] [Google Scholar]
  • 20.Whyte J. Pharmacologic treatment of cognitive and behavioral sequelae of traumatic brain injury: practicing in the absence of strong evidence. Eur J Phys Rehabil Med. 2010 Dec;46(4):557–62. [PubMed] [Google Scholar]
  • 21.Corrigan JD, Horn SD, Barrett RS, Smout RJ, Bogner JB, Hammond FM. Effects of Patient Pre-Injury and Injury Characteristics on TBI Outcomes. Arch Phys Med Rehabil. Paper C for this TBI supplement can be added before publication. [Google Scholar]
  • 22.Seel R, Horn SD, Barrett RS, Smout RJ, Hammond FM, Ryser D. TBI-PBE Rehabilitation Facility Characteristics. Arch Phys Med Rehabil. Paper B for this TBI supplement can be added before publication. [Google Scholar]
  • 23.Evidence-based review of moderate to severe acquired brain injury. http://www.abiebr.com/
  • 24.Warden DL, Gordon B, McAllister TW, Silver JM, Barth JT, Bruns J, Drake A, Gentry T, Jagoda A, Katz D, Kraus J, Labbate LA, Ryan LM, Sparling MB, Walters B, Whyte J, Zapata A, Zitnay G. Guidelines for the pharmacologic treatment of neurobehavioral sequelae of traumatic brain injury. J of Neurotrauma. 2006;23(10):1468–1501. doi: 10.1089/neu.2006.23.1468. [DOI] [PubMed] [Google Scholar]
  • 25.Talsky A, Pacione LR, Shaw T, Wasserman L, Lenny A, Verma A, Hurwitz G, Waxman R, Morgan A, Bhalerao S. Pharmacological interventions for traumatic brain injury. BCMJ. 2010;53(1):26–31. [Google Scholar]
  • 26.Whyte J, Hart T, Schuster K, Polansky M, Coslett H. Effects of methylphenidate on attentional function after traumatic brain injury. Am J Phys Med Rehabil. 1997;76:440–450. doi: 10.1097/00002060-199711000-00002. [DOI] [PubMed] [Google Scholar]
  • 27.Plenger P, Dixon E, Castillo R, Frankowski R, Yablon SA, Levin HS. Subacute methylphenidate treatment for moderate to moderately severe traumatic brain injury: A preliminary double-blind placebo-controlled study. Arch Phys Med Rehabil. 1996;77:536–540. doi: 10.1016/s0003-9993(96)90291-9. [DOI] [PubMed] [Google Scholar]
  • 28.Siddall OM. Use of methylphedinate in traumatic brain injury. Ann Pharmacother. 2005;39:1309–1313. doi: 10.1345/aph.1E637. [DOI] [PubMed] [Google Scholar]
  • 29.Schneider W, Drew-Cates J, Wong T, Dombovy ML. Cognitive and behavioural efficacy of amantadine in acute traumatic brain injury: An initial double-blind placebo-controlled study. Brain Injury. 1999;13:863–872. doi: 10.1080/026990599121061. [DOI] [PubMed] [Google Scholar]
  • 30.Zafonte R, Lexell J, Cullen N. Possible applications for dopaminergic agents following traumatic brain injury: Part 1. J Head Trauma Rehabil. 2000;15:1179–1182. doi: 10.1097/00001199-200010000-00014. [DOI] [PubMed] [Google Scholar]
  • 31.Meythaler J, Brunner R, Johnson A, Novack TA. Amantadine to improve neurorecovery in traumatic brain injury-associated diffuse axonal injury: A pilot double-blind randomized trial. J Head Trauma Rehabil. 2002;31:300–313. doi: 10.1097/00001199-200208000-00004. [DOI] [PubMed] [Google Scholar]
  • 32.Elovic EP, Lansang R, Li Y, Ricker JH. The use of atypical antipsychotics in traumatic brain injury. J Head Trauma Rehabil. 2003 Mar-Apr;18(2):177–95. doi: 10.1097/00001199-200303000-00008. [DOI] [PubMed] [Google Scholar]
  • 33.Burnett DM, Kennedy RE, Cifu DX, Levenson J. Using atypical neuroleptic drugs to treat agitation in patients with brain injury: a review. NeuroRehabilitation. 1999;13(3):165–172. [Google Scholar]
  • 34.Leon AC, Hedeker D, Teres JJ. Bias reduction in effectiveness analyses of longitudinal ordinal doses with a mixed-effects propensity adjustment. Stat Med. 2007;26:110–123. doi: 10.1002/sim.2458. [DOI] [PubMed] [Google Scholar]
  • 35.Leon AC, Hedeker D. A comparison of mixed-effects quantile stratification propensity adjustment strategies for longitudinal treatment effectiveness analyses of continuous outcomes. Stat Med. 2007;26:2650–2665. doi: 10.1002/sim.2732. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Digital Content Table A: Expanded psychoactive medications data administered by pharmaceutical class and level of admission cognitive function

RESOURCES