Severity is important because it provides a level of specificity about the injury which determines management of care. Additionally, TBI severity can yield valuable insight about proximal and distal outcomes. It seems reasonable that it would be an important measure to include when examining the relationship between TBI and all included variables. Additional tools, such as the AIS scores and imaging studies, may be necessary in accurately capturing TBI severity in study participants; these studies, in addition to GCS, should be considered an essential variable that must be accounted for. All of the studies measured presence of marijuana, yet the methods by which marijuana was measured varied. For example, urine was the most common way to measure marijuana concentration in patients in reviewed studies, but urine tests results are not specific to time of injury: The detectable level of marijuana can be present in urine for approximately 4.6 to 15.4 days after last use for infrequent and chronic users respectively. The presence of marijuana on a urine toxicology screen may not accurately reflect or correlate marijuana levels in an individual’s system at time of injury, rather, it reflects recent use. Therefore, when considered as a variable, a marijuana level should be considered as reflective of recent use at time of injury, not directly at time of injury. Finally, this review and other systemic reviews consistently identify blood alcohol concentration as an important potential confounder in TBI studies. Much has been studied about the relationship between alcohol and TBI. As a prominent pre-disposing factor in TBI,vertical farming supplies the implications alcohol intoxication has on TBI is important and must be accounted for when examining the effects of marijuana on TBI.
The current systematic literature review has several limitations, the first of which was the inability to perform a meta-analysis with the studies acquired. There was heterogeneity across the studies addressing marijuana exposure and TBI; from different criteria used to classify TBI, to diverse populations of interest, to varied outcomes of measures, the studies varied widely preventing a meta-analysis of the 8 included studies. Additionally, the studies differed in the type of data they collected, especially individual level data, which do not provide the necessary statistical measures that would make a meta-analysis meaningful.As a trauma clinician, my theoretical orientation focuses on injury prevention and favorable outcomes in the context of severe trauma, especially TBI. Epidemiological observation shows that the landscape of TBI prevalence and outcome is changing, but there is not a good understanding of how, which would make visible actionable areas for beneficial intermediation. Therefore, this study orients to the phenomenon of TBI epidemiologically. The definition of epidemiology is the study of the distribution and determinants of diseases and injuries in human populations . Epidemiological data includes data gathered via interviews, archival research, and record review as well as via direct observation. The unit of analysis in epidemiology is the individual, yet focuses on identifying factors that are deleterious to the public . For the phenomenon of marijuana use, there are many reasons why individuals choose to use marijuana, either for recreational purposes or medicinal purposes. While the social aspects and context for marijuana exposure and use are important in and of themselves, potentially adverse clinical outcomes are equally as important and valid. Findings of this epidemiological study may uncover important demographic characteristics and health effects that warrant further study to help better understand the positive and negative effects of marijuana exposure in the context of traumatic brain injury.
In summary, while the individual and social characteristics of marijuana use may be diverse and in need of study, findings from the literature review in Chapter 2 identified that these individuals as a collective group are potentially at more at risk for incurring an injury that could eventually lead to a TBI. Determining this risk and identifying characteristics of the population that are potentially at greater risk than others is an appropriate focus for study.A correlate is defined in this study as a characteristic of the marijuana-positive patient. Correlates were identified through the literature review , other observational research, and clinical practice expertise. Identified correlates include age, gender, race, ethnicity, other substances, alcohol, alcohol use disorder, chemotherapy for cancer, disseminated cancer and mental/personality disorders. Age will be included because research has shown that certain age groups comprise of a larger percentage of current marijuana users . While no research has associated gender and ethnicity with marijuana use, they are both variables commonly studied in most observational research, therefore, it will be included in this study. Alcohol and other drug use at time of injury will be included in this study because research has shown that 1 in 8 individuals had both alcohol and an drug use disorder in the past year . Cancer dissemination and cancer treatment are included as correlates because studies show that smoked marijuana may be helpful in the treatment of nausea and vomiting because of chemotherapy . Other studies have found that smoked marijuana may be beneficial in the treatment of neuropathic pain related to chemotherapy treatment . Finally, mental illness will be included as a variable as Cannabis Use Disorder is much higher in individuals with schizophrenia, personality disorders, post-traumatic stress disorder, mood and anxiety disorders, and other types of mental illnesses when compared to the general population . Identified correlates were both identified and confirmed as extractable in the NTDB. There were no identified correlates that were NOT present in the NTDB. See Chapter 4 Table 6 for details about how each variable will be operationalized. Confounding is a type of bias where a variable is associated with both the exposure and a given outcome resulting in a misrepresentation of the true relationship .
Confounding variables may conceal a true association, or they may falsely demonstrate an existent association between an intervention or exposure and an outcome when no association actually exists . For this study, a confounder is defined as a variable that is associated with both the independent and dependent variable. Confounders were identified through the literature review , other observational research, and clinical practice expertise. Identified confounders include age, gender, alcohol exposure at time of TBI, and alcohol use disorder, which is defined by the DSM-V as medical diagnosis indicating that the problem of drinking has become severe and chronic for the patient . Physicians typically diagnosis this disorder through their history and physical assessment, which is documented in the patient’s medical record and extracted into the NTDB by trained trauma program registrars. Participants’ age as well as gender may be potential confounders, with males being at higher risk of sustaining a TBI . Another potential confounder in this study is alcohol, or alcohol use disorder. There have been extensive studies conducted on the relationship between alcohol and TBI related outcomes, with alcohol identified in 35-50% of individuals who sustain a TBI . Another confounding variable that will be examined is the use of other drugs. Evidence suggests that there is an increase in the presence of other drugs, aside from alcohol,vertical farms companies in injured and fatally injured drivers . Furthermore, findings from the literature review showed that the presence of other drugs in combination with marijuana was a common occurrence.The first phase of the data cleaning process will be data screening. When screening data, four types of abnormalities will be assessed: missing data, in consistences and outliers, odd patterns of distributions, and unexpected analysis results, inferences or abstractions . Descriptive tools, such as Statistical Package for the Social Sciences will be utilized to facilitate the screening process and ensure the process is objective and systematic. A potential source of problem in this study that may be encountered during data collection is missing data, outliers, and inconsistencies due to the use of a database that includes existing data that was not specifically collected for the purposes of this study. Errors such as blank fields, unintentional deletions or duplications during data entry, blank data fields, or values incorrectly entered must be accounted for . Screening methods involving graphical exploration of distributions and statistical outlier detection will be utilized. The second phase in the data cleaning process is the diagnostic phase. In this phase, a diagnosis regarding the nature of concerning data points or patterns will be attempted. Potential diagnoses for each data point include the following: erroneous, true normal, true extreme, or idiopathic . The correct value or data point for certain fields can be obvious and easily noticed . For such erroneous or missing data points, processes regarding dealing with missing data will be implemented and corrected prior to analysis.
The treatment phase of identified erroneous data involves correcting, deleting or leaving the error unchanged . For purposes of this study, if impossible or missing values are observed, they will have to be deleted, as there would be no way of correcting that value related to the retrospective and secondary nature of the data. For data points that are true extremes, further examination on the influence of these data points, individually and collectively, on analysis will be made prior to determining whether or not that data point will be deleted or left unchanged . It is important to deal with missing data because missing data can create bias. First, an exploratory analysis will be performed to look at frequencies or percentages of missing data, and to help identify how much data is missing. Next, an analysis of the mechanisms, or types, of missingness will be performed to identify whether the missing data is missing completely at random , missing at random , or not missing at random using statistical tests, such as Little’s test for MCAR. Following this, an analysis for patterns of missingness will be performed using a missing pattern value chart. There are two patterns that may be potentially observed: 1) a monotone pattern where data is missing systematically, or 2) an arbitrary pattern where data are missing at random . While the analyses are not definitive, they can bring attention to blatant anomalies in the missingness of data and help to make decisions on the missing data handling procedures. There are a variety of methods that can be utilized to deal with missing data. The type of method utilized will depend on the percentage of missing data present and cannot be specified beforehand. Simple methods, such as list wise or pairwise deletion are helpful when the percentage of missing data is less than 5%. List wise deletion, also known as complete-case analysis, removes all data for a case with one or more missing values. In other words, that case is omitted completely. A disadvantage when using list wise deletion is that it can reduce the sample size. On the other hand, pairwise deletion, also known as available-case analysis, aims at minimizing the loss of other potential data incurred with list wise deletion. Pairwise deletion still uses that case when analyzing other variables with non-missing values; it just excludes that one value with a missing data. An advantage to pairwise deletion over list wise is that it can help increase statistical power. However, pairwise deletion does have its disadvantages in that most software packages use the average sample size across analyses which can create over or underestimation. If the percentage of missing data is greater than 5%, then more advanced methods of dealing with missing data can be utilized, such as imputation. Imputation methods will depend on the pattern of missingness identified and the type of variable requiring imputation . In patterns where missing data is systematic or monotone, methods such as regression, predicted mean matching or propensity scoring are helpful. In patterns where missing data is arbitrary or at random, methods such as multiple imputation using maximum likelihood regression methods to predict missing values based on observed values and sensitivity analyses that simulate the results based on a range of plausible values can be used. For Aim 2, the objective is to determine the correlates associated with the presence of marijuana exposure at the time of injury. The correlates included in Aim 2 will be also collected for the sample of participants without marijuana exposure at time of injury. Measures of central tendency, including range, means, proportions and standard deviations will be calculated.