Hemp Growing

A recent study by Arabian et al. revealed the presence of inaccuracy and variability between hospitals, specifically in the areas of data coding and injury severity scoring. Additionally, the type of registry software a hospital utilizes can report injury severity scores differently . This too, renders data subject to informational bias. Information bias is due to inaccurate or incorrect recording of individual data points. When continuous variables are involved, it is called measurement error; when categorical variables are involved, it is called misclassification . In this study, the potential for information bias is mostly due to 1) incomplete data documented in the medical record, or 2) inaccurate entry into the hospital trauma database by hospital staff. Missing data will be analyzed in terms of potential effect for both the independent and dependent variable . While the database captures marijuana exposure through the first recorded positive drug screen within the first 24 hours after first hospital encounter, it is recognized that at times patients will not be screened, even if they have been exposed to marijuana. Marijuana presence can be detected in the urine up to 3-5 days from exposure in infrequent users; marijuana can be detected up to 30 days for chronic users . Therefore, patients could potentially have a positive marijuana toxicology screen even though they may not have ingested marijuana the day of the event. A positive marijuana urine toxicology screen indicates the probability of prior use, not immediate use. This is an important limitation to note. In clinical practice,drying room the determination for a toxicology screen is often symptomology, so it is reasonable to assume that patients who have ingested marijuana a week prior to the event date may not exhibit the expected symptomology.

Unlike other observational cohort studies, the potential of recall bias is minimal due to the availability of an objective marker to measure the independent variable, namely, the presence of marijuana. The presence of marijuana is captured from the hospital lab urinalysis results and is recorded as present within 24 hours after the first hospital encounter. Similarly, the data entered to measure the GCS score is also captured objectively through a numeric recorded score found in the medical record. See analysis section for how this type of bias will be addressed. The final sample size for this study involved 7,875 total unique cases. Those cases represent individuals who sustained a moderate or severe TBI in the NTDB database. Of the 997,970 total cases for 2017, there was a total of 32,896 cases that were identified as having sustained some form of traumatic brain injury, ranging from a concussion to severe injury, using the ICD 10 Diagnosis codes listed below . Of the 32,896 cases, 25,021 were identified as having a concussion diagnosis, and were ultimately excluded from the final sample size. This was because mild concussion diagnosis was found to suffer from large underestimates in documented incidence . A World Health Organization systematic review of mild TBI found that up to 90% of overall TBIs was mild in nature. The WHO has also estimated a yearly incidence of mild TBI anywhere from 100-600 per 100,000 cases, 0.1 to 0.6 respectively . Furthermore, up to 40% of individuals who sustain a mild TBI, or concussion do not seek the attention of a physician . Another study found that 57% of veterans who had returned from Iraq and/or Afghanistan, and had sustained a possible TBI, were not evaluated or seen by a physician . According to the WHO and CDC reports, these numbers may still not represent the actual incidence of TBI worldwide. Furthermore, the data suggests that individuals with a mild TBI for the most part do not go and seek medical attention, and this study focuses on individuals who sustain a moderate or severe TBI as those individuals suffer life-long devastatingly debilitating effects and are the targets of public health initiatives and injury prevention measures.

The Trauma Quality Programs research database housed in the NTDB for the year 2107 is the time frame for this study. Though initially the researcher intended to include data from 2013-2017, data from years other than 2017 had to be excluded. In effort to standardize the type of data collected by local, regional, and state trauma registries, the NTDB designs a National Trauma Data Standard Data Dictionary that is designed to establish a national standard for the collection of trauma registry data while also providing the operational definitions for the NTDB. In summary, the NTDS provides the exact standards for trauma registry data submitted to the NTDB. Prior to the 2017 data dictionary, trauma registry programs had limited selections regarding data related to drug use. The options provided by the NTDB registry only included whether drug use was present and whether it was confirmed by a test or by prescription. It did not allow the trauma data abstractor to specifically identify the type of drug found. In 2017, the data dictionary was revised to include a drug screening category that aimed at recording the first positive drug screen result within 24 hours after the first hospital encounter. Typically, in trauma hospitals reporting to NTDB and within the context of trauma, acquisition of a urine and blood drug and alcohol screen is standard expectation of practice. It then provided a list of 15 options for the abstractor to choose from. Because it was impossible to isolate cannabinoid use in earlier data sets, the researcher was only able to use the 2017 NTDB data set, which at the beginning of the study was the latest available data set by the NTDB. As of February 13th, 2021 the 2018 NTDB data set was not available. All the trauma data used in this study are organized by an element INC_KEY, which is a designated unique identifier for each record. The designated unique identifier INC_KEY expresses a unique clinical visit/episode by an individual at a participating trauma center. It is important to consider that an individual could have been included/counted more than once in the registry because of more than one traumatic event within the year. The Participant Use File Trauma data set contained all the demographic, environmental, and clinical data information. However, it did not identify or delineate TBI cases as such.

Therefore, a separate data set that contained ICD 10 Diagnosis Codes had to be utilized to identify TBI cases which then could be used to create a merged data set that is complete. The 2017 PUF Trauma data set was uploaded to SPSS version 25 on September 10th, 2020. The PUF Trauma data set included a total of 997,970 unique identifier cases. A frequency analysis was performed to ensure no duplicate cases were found . The PUF Trauma data set included 328 unique variables. Next,how to trim cannabis the PUF ICD-10 Diagnosis data set was uploaded and examined. The PUF ICD Diagnosis data set is organized via the same INC_KEY identifiers. The PUF ICD Diagnosis data set included 3 variables: ICD CM diagnosis code, ICD CM diagnoses code Blank Inappropriate Values and ICD Clinical Modification version. This data set was used to distinguish TBI cases from cases related to other traumas such as pneumothorax, liver laceration or femur fractures. The way this was done was first the researcher identified TBI related ICD 10 CM diagnosis codes by visiting the ICD 10 Data website at www.icd10data.com and searching for all head injury related codes. Additionally, the selection of TBI related ICD 10 codes was corroborated by examining a list of codes found in existing studies on TBI which validated the inclusion of the specifically identified TBI codes in this study. Though these other studies included ICD 10 Diagnosis codes related to concussion injuries , these codes were excluded from this study as the researcher was only interested in identifying cases with either a moderate or severe TBI and concussions are designated as mild TBI. The following codes were ultimately selected: S02.0xx ; S02.1 ; S06.1 ; S02.19XD ;S06.2 ; S06.30 ; S06.31 ; S06.32 ; S06.33 ; S09.X . Next, PUF ICD 10 Diagnosis codes were regrouped into the following categories via numerical representation. ICD 10 Diagnosis code S02.0xx was grouped into group 3683-3687; S02.1 into group 3688; S02.19XD into group 3738; S06.1 into group 4008-4025; S06.2 into groups 4026-4045; S06.3, S06.31, S06.32, and S06.33 into groups 4046-4095; S09.X into groups 4310-4311. A missing value analysis for the ICD 10 Diagnosis code variable revealed no missing values. A new variable titled ‘TBI” was created in the PUF ICD-10 Diagnosis data set where if a TBI related ICD 10 code was assigned, the value ‘1’ was given. If not, it was assigned a value of ‘0’. A frequency analysis on the ‘TBI’ variable was then done to determine the number of TBI codes which were found to be 131,518.The new data set contained 324 total variables.

The variables present were identified as subsets of the following categories: work-related injury, patients occupational industry, patient’s occupation, ICD 10 primary external cause, ICD 10 place of injury code, ICD 10 additional External cause code, protective devices, child specific restraint, airbag deployment variables, report of physical abuse, investigation of physical abuse, caregiver at discharge, transport modes, initial emergency service system vital signs , time to EMS response, time from dispatch to ED/hospital, interfacility transfer, pre-hospital cardiac arrest, trauma center criteria for admission, vehicular/pedestrian or other risk, mechanism of injury , total time between ED/hospital arrive and ED discharge, systolic blood pressure, pulse rate, temperature, respiratory rate and assistance, pulse oximetry, supplemental oxygen, height, weight, primary method of payment, signs of life, emergency room disposition, hospital discharge disposition, comorbid conditions , total intensive care unit length of stay, total ventilator days, length of stay , hospital complications, procedural interventions, medications administered, blood transfusions, withdrawal of life support, facility level, year of discharge, ISS, and AIS derived ISS. Variables that would not be included in the final analysis were removed. Example of variables removed were ventilator days, length of stay and blood transfusions. Some of the variables that incorporated more than one value, such as race, ethnicity, alcohol screen result and drugs, were concatenated to form new variables. A description of how each variable was dealt with is delineated below. This was done to facilitate the analysis of more than one categorical variable to be treated as one. A separate-variance t Test table is displayed by SPSS as part of the missing value analysis. This table can help identify variables whose pattern of missing values may be influencing the quantitative variables. When age is missing, the mean alcohol screen result is .0031 compared to .0652 when age is present. This large difference in mean alcohol screen result scores when age is present indicates that the data missing is not missing at random. When age was missing, mean total GCS was 14.77 compared to 13.21. This is not a large difference, indicating that data may be indeed missing at random. When alcohol screen result is missing, the mean age is 28.86 compared to 42.64 when alcohol screen result is present. This indicate that the data may not be missing completely at random it is important to consider that in the alcohol screen result variable, there is a large percentage of missing values. Additionally, since this data set includes patients ages 16 years and younger, it may be that clinicians are not drawing alcohol levels. This can lead to the fact that the values that are missing when these two variables are cross-tabulated, may not be missing at random. Finally, it is important to note that unlike in questionnaires or surveys, these trauma patients are not asked for an alcohol screen result, rather they are tested by the retrieval of a blood sample. Therefore, it is not the patient themselves that chooses to respond or not, rather, it is the hospital system that contributes to whether the data is missing. Data for alcohol screen result may be missing due to lack of time to retrieve the blood sample as can be found when patients present to the ER in traumatic full arrest. Alternatively, the sample may have been drawn but not sent to lab, or sent to lab but not reported by lab, or reported by lab not recorded by the nurse.

Alternatively, being male, having elevated blood alcohol levels and having other drugs present on admission were all found to have a significant influence on GCS scores and TBI severity, with GCS scores being lower for all three variables, implying a more serious TBI. Similarly, having a diagnosis of cancer, mental or personality disorder and alcohol use disorder were found to have an influence on GCS scores. Participants with a diagnosis of cancer or mental/personality disorder were found to have lower GCS scores, again, implying a more serious TBI. Conversely, participants with a diagnosis or history of alcohol use disorder had higher GCS scores, indicating a less serious TBI. While the presence of THC initially did show a hypothesized relationship to GCS score , the relationship became insignificant when adjusted for all the other covariates variables. Because of the large percentage of missing data, the validity of findings, such as THC prevalence rate in this TBI population, should be cautiously interpreted for all the included hypothesized explanatory variables. Further research with datasets that are larger and more complete are needed to fully understand and examine the relationship between marijuana and TBI severity. This study importantly underscores the need for better data to enable better research regarding the relationship between marijuana and TBI severity. Traumatic brain injury is a significant public health concern as it is a leading cause of mortality, morbidity and disability in the United States . According to the World Health Organization, TBI is expected to become the third leading cause of death and disability in the world by 2020. In the United States TBI contributes to a third of all injury related deaths . A traumatic brain injury,grow trays as defined by the Centers for Disease Control and Prevention , is a disturbance of the brain’s normal function that occurs when an individual sustains a blow, jolt, or bump to the head, or sustains a penetrating head injury .

Traumatic brain injuries can lead to a variety of secondary conditions that could result in cognitive, behavioral, motor, and somatic impairments that cause long-term disability and poor quality of life . The leading causes of injuries resulting in TBI prevalence are traffic related, such as motor vehicle crashes, or non-traffic related, such as falls. Falls are the leading cause of TBI with almost 81% of emergency department room visits in adults over the age of 65 attributed to falls . Motor vehicle collisions are the leading cause of TBI related deaths, with rates being highest for adults between the ages of 15-24, 25-35 and older adults greater than 75 . Notably, up to 51% of all TBI patients have substance use exposure at the time of injury . Substance use includes alcohol and drugs such as marijuana. Current existing research suggest that in general, substance-exposed patients may have worse TBI outcomes, including greater rates of mortality and severity of injury. Research has also shown that these patients suffer worse functional outcomes, which can result in socioeconomic burden to patients and the nation at large. This healthcare burden has been calculated to be approximately $76.5 billion in 2010 alone . There is a substantial body of research elucidating the influence of alcohol on TBI prevalence and outcomes . Alcohol use results in impairments such as diminished motor control, blurred vision, and poor decision making, which in turn has been shown to increase the risk for TBI . This research has been used to create public health policies, public education efforts, and prevention programs that have made a significant health impact, such as reducing the number of alcohol-impaired drivers . While it is known that there is significant alcohol use related to TBI, little is known about the influence of marijuana on the prevalence, severity and outcomes related to TBI . Marijuana is an drug that despite being federally and legally regulated, remains the most widely used drug in the U.S. . Marijuana use has been shown to result in similar cognitive impairments as alcohol use, such as lack of coordination, alterations in reaction time, inability to pay attention, and decision-making abilities, suggesting marijuana users are similarly at increased risk for TBI .

There is some indirect evidence of this, in that it has been shown that marijuana users in general are about 25% more likely to be involved in a motor vehicle collision and that the older adult marijuana users have a greater risk for falls . Both short and long-term marijuana exposure has been shown to impair driving ability; marijuana is the drug most often reported in association with impaired motor vehicle collisions, including fatal ones . It has also been shown that the overall risk of being involved in a motor vehicle collision increases by a factor of 2 soon after an individual has used marijuana . Motor vehicle collisions make up almost two thirds of U.S. trauma center admissions and are the leading cause of TBI related deaths . Approximately 60% of MVC patients tested positive for drugs and alcohol . Despite the increase in marijuana use and exposure, concrete data linking marijuana exposure at time of injury and TBI prevalence and severity is scarce . Adding to the concern, national surveys on drug use and health have documented an increase in individual daily marijuana use over the last 5 years. In summary, there is no body of research documenting the relationship between marijuana exposure and TBI prevalence and severity. As the number of states legalizing marijuana for both medical and recreational use increases, it is imperative to resolve the ambiguity within the research available regarding the influence of marijuana exposure on TBI. This study will fill important gaps in knowledge about this emerging public health concern by documenting the prevalence of marijuana exposure in a national sample of TBI patients, and determine the relationship between marijuana exposure, mechanism of injury, and TBI severity. Study aims are to: 1) assess the prevalence of marijuana exposure in patients with moderate or severe TBI at time of injury; 2) examine correlates associated with marijuana exposure at the time of injury; and 3) examine the relationship between marijuana exposure, mechanism of injury and TBI severity. Results will provide the first quantifiable national-level evidence of the impact of marijuana exposure on TBI.

Results will also serve as the basis for research that can inform policy and public safety standards and metrics regarding marijuana exposure and its effect on TBI.A search strategy was implemented by searching the PUBMED electronic bibliographic database between January 17-19 in 2019. No restrictions were applied on publication status and publication date. This systematic review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement. The search strategy included the terms traumatic brain injury, severity, substance, substance abuse, marijuana, THC, cannabis, and drug use. Only publications in English were sought. Reference lists of review papers were searched to ensure all relevant literature was included. An example of the search strategy for this review is shown in Figure 1. To be included in this systematic review, studies must have been peer-reviewed, published in English, involve human subjects only, and must have investigated the use of marijuana in adult patients reported to have sustained a moderate to severe TBI. We did not consider participants below the age of 16 because pediatric trauma patients present differently than do adults,drying marijuana and are treated with different intervention protocols than in adults. A preliminary search identified the fact that articles subsumed marijuana exposure under the broader umbrella term of substance use/abuse. Therefore, substance and substance abuse terms were included to ensure a wide sensitivity to studies involving drugs such as marijuana. Exclusion criteria Patients with a diagnosis of mild TBI were excluded because up to 40% of mild TBI patients do not seek medical attention, and therefore, findings would not be representative . Similarly, the following studies were excluded from this review: studies that did not assess for marijuana exposure at time of injury, marijuana post-TBI, cellular based studies, clinical review papers, editorials, case reports, pediatric studies and studies using nonhuman subjects. Selection Process Study selection was conducted in a two-stage process. First, studies were screened by titles and abstracts for potential inclusion. Next, studies identified as relevant for potential inclusion underwent a full-text evaluation. Studies that included any information about marijuana exposure at the time of injury were included, including studies where marijuana was bundled with other substances as either a variable or via analysis, because it was assumed there would still be relevant information embedded within the study. The studies were reviewed a second time to ensure all inclusion criteria were met and included if they did.Data was extracted from studies that met selection criteria. Data from the studies were used to achieve the primary aims of this systematic review: to examine marijuana exposure and use in TBI prevalence, severity and outcomes.

The following data were abstracted to summarize specific study features and address the review’s aims: 1) study characteristics, including authors names, publication year, country, design, sample size, and methods utilized, 2) participant characteristics such as mean age and type of TBI, 3) information about whether other substances besides marijuana, such as alcohol, methamphetamines, cocaine, opiates, benzodiazepines, narcotics, stimulants, speed, hallucinogens and heroin were documented and/or analyzed 4) results, including the prevalence of marijuana, TBI outcomes, and if a relationship between marijuana and TBI was present.Search results, including abstracts and full-text articles, were exported to an Excel file for data management. The decision for inclusion or exclusion in the review process was recorded in the Excel file, as well as a rationale for exclusion of studies. Reference management was done through the Papers©, a reference management software used to manage bibliographies and references. A reference library of PDF documents was maintained through the software and allows a variety of features such as collecting, curating, merging of studies as well as the insertion of citations in-text. The Levels of Evidence were assessed using the National Heart, Lung and Blood Institute categories. The NHLBI Levels of Evidence framework rates evidence on four major levels, placing the highest rating on evidence that is acquired from Randomized Controlled Trials with an extensive body of data; RCTs are assigned a level “A” according to the NHLBI. Level B studies are RCTs with a limited body of data, usually involving a smaller sample size, include a subgroup analysis of RCTs, and may include study results that are inconsistent. Level C studies are those that employ a non-randomized study design, such as observational studies. Finally, Level D studies include studies that utilized mechanism-based reasoning that involve anecdotal findings based on expert opinion. Risk of bias of included articles was assessed using the National Heart, Lung and Blood Institute quality assessment tool for observational cohort and cross-sectional studies. The NHLBI offers six various study quality assessment tools, three of which apply to observational cohort studies, cross-sectional studies, and case series studies. The quality assessment of observational cohort and cross-sectional studies tool was utilized. The NHLBI quality assessment tool is comprised of 14 criteria/questions that address study objectives, study population, sample size, exposures and outcome measures, and key potential confounding variables. An example of NHLBI quality assessment tool for observational cohort and cross-sectional studies is presented in Table 2. Potential sources of bias were rated as either “yes”, “no”, “cannot determine”, “not applicable”, or “not reported”. Each study was given an overall bias rating of good, fair, or poor. Table 2 delineates responses to each of the 14 questions in the NHLBI quality assessment tool, while Table 3 addresses the types of biases encountered, the presence or lack thereof of confounding variables, and other information that aid in the assessment of biases.Results from the included studies were reviewed for the outcome of interest and were reported under seven themes: presence of marijuana exposure; time frame in which marijuana exposure was measured; method used to measure marijuana exposure; information on other substances if they were bundled with marijuana exposure; and the presence of a specific link between marijuana exposure and TBI severity. Due to the range and diversity of study results and designs, a meta-analysis was not possible. Additionally, given the differences in the conceptualization and definition of marijuana exposure across the studies included, and the heterogeneity in methods, sample data, collection and findings, a narrative interpretation and descriptive analysis of the findings was necessary.

One limitation of this report is the study population consisted of HIV-negative MSM in Hanoi enrolled into a cohort study and is not necessarily representative of the community prevalence among MSM in Hanoi or other cities in Vietnam. Our finding of high rates of extragenital infections is also consistent with regional reports from Thailand. Among a cohort of MSM in Thailand, including those living with HIV, rectal infections accounted for nearly 70% of C. trachomatis infections and approximately 60% of N. gonorrhoeae infections, while 40% of N. gonorrhoeae infections involved the oropharynx. In another report of HIV-negative MSM in Bangkok, approximately 65% of C. trachomatis and 80% of N. gonorrhoeae infections were extragenital. Elsewhere in the region, prevalence among MSM, including those living with HIV, was lower: in Guangzhou, the overall prevalence of rectal C. trachomatis was 11.2% and rectal N. gonorrhoeae was 6.1%, and in Kunming, the anatomic site with the highest prevalence of C. trachomatis was the rectum at 15.5% and the prevalence of N. gonorrhoeae was highest in the oropharynx at 8.1%. While direct comparisons between other studies and ours are limited due to different study populations, recruitment methods, and testing strategies, the relatively higher rate of N. gonorrhoeae infections involving the oropharynx observed here is particularly concerning given this is a potential reservoir for antimicrobial resistance. The overall high prevalence of chlamydia and gonorrhea observed in our study might reflect an increase in rates of STIs within MSM communities in Vietnam. Testing at three anatomic sites also likely contributed to the high prevalence observed in our study. However,curing cannabis increasing rates of STIs among MSM is a trend observed globally and one that is often driven by stigma, discrimination, and limited access to healthcare.

Few of the study participants with CT or NG infections reported a prior diagnosis of an STI, likely reflecting limited access and engagement with appropriate sexual health services, including HIV services, observed among MSM in Vietnam. High levels of stigmatisation and low levels of STI knowledge are structural and individual barriers that can lead to alienation of MSM from sexual health services in Vietnam. A clinic-based, sexual health promotion intervention using health educators among male sex workers, a subgroup of MSM, in Vietnam was effective at increasing testing and treatment for STIs, increasing their knowledge of HIV/STI transmission risk, and health seeking intention. More efforts are needed to expand such measures among MSM in Vietnam, as well as other settings seeking to promote engagement with sexual health services. We found that half of all infections were asymptomatic, although it should be noted that oropharyngeal symptoms were not assessed by the study’s survey instruments, which might lead to an under-estimation of symptoms reported here. Nevertheless, the data shown here support routine triple-site testing for N. gonorrhoeae and C. trachomatis among MSM in Hanoi. Yet in Vietnam, as well as many other low- and middle-income countries, the cost of NAATs for C. trachomatis and N. gonorrhoeae is prohibitive and is a primary barrier limiting the widespread availability of these tests. Many other barriers to diagnosing STIs in low resource settings also exist, including availability of laboratory equipment and infrastructure, as well as limited availability of trained personnel. Expanding access to testing can not only improve the diagnosis and treatment of STIs but also help to identify those at risk for HIV who can benefit from PrEP. While PrEP was not yet available in Vietnam at the time of this study, a free-of-charge PrEP program has since been implemented in Vietnam and has engaged more than 32,000 people by the end of 2021, mostly MSM. There is considerable need for PrEP among MSM in Vietnam, and increasing funding and access to STI diagnostics and treatment are important components of scaling up PrEP recruitment and routine PrEP care.

Further research is needed to optimize STI screening among MSM in low-resource settings, including assessments of diagnostic testing strategies . Frequently observed risk factors for STIs, such as younger age, condomless anal intercourse, and having two or more recent sex partners were independently associated with gonorrhea or chlamydia in this cohort at baseline. In addition, meeting sex partners via mobile apps or the internet was associated with N. gonorrhoeae or C. trachomatis. The use of the internet or mobile apps to meet sex partners has been associated with behaviors that can increase risk for STIs, including HIV, as users tend to have greater frequency of condomless anal intercourse and more sexual partners. However, mobile app use was associated with infections independent of those factors, suggesting an additional mechanism; one plausible explanation could be related to the sexual networks of mobile app users. Determining causality of mobile app use is difficult, as it is not clear if meeting partners online or via mobile apps increases the risk of STIs or is a behavior of an individual who is already at higher risk for STIs. Nevertheless, the internet or mobile apps are becoming increasingly common ways to meet sexual partners worldwide, including in Vietnam, where a recent survey of MSM found that over three-quarters reported meeting their partners online. Given their widespread use, internet and mobile apps should be leveraged to deliver targeted sexual health interventions aimed at improving diagnosis, treatment, and prevention of STIs, among MSM in Vietnam. One limitation of our report was that all behaviors and symptoms were self-reported and might be subject to recall or social desirability biases, although the use of ACASI for the questionnaire would be expected to limit the latter. In summary, our report comprehensively documents the prevalence of N. gonorrhoeae and C. trachomatis infections at urethral, rectal, and oropharyngeal sites among a cohort of HIV-negative MSM living in Hanoi and adds to the body of recent evidence demonstrating the high burden of STIs within MSM populations globally.

Most of these infections are extragenital and asymptomatic, supporting routine screening at multiple anatomic sites. However, multilevel barriers exist that limit access to sexual health services and diagnostic testing for CT and NG in Vietnam, which include costs and availability of tests, stigma, education, and other individual and systemic barriers. Efforts are urgently needed to address these barriers in order to increase access to STI testing and treatment for MSM in Hanoi. Over 35 million people worldwide live with human immunodeficiency virus , and 1.2 million of these people live in the United States. Since the development of combination antiretroviral therapy , HIV-associated mortality has decreased in the United States, such that the lifespan of people living with HIV with reliable access to cART is comparable to those without HIV . Despite these advances in the medical management of HIV disease, the central nervous system remains vulnerable. In fact, HIV targets the CNS within days after infection leading to neurological, behavioral, and cognitive complications . Even in the current cART era, mild neurocognitive deficits are observed in about 45% of PLWH, particularly in the domains of executive function, learning, and memory . Neuroimaging studies suggest that functional and structural abnormalities in subcortical regions underlie these cognitive deficits . Neurocognitive impairment among PLWH is clinically meaningful because it is known to adversely affect daily functioning, conferring an increased risk of poor medication management , impaired driving ability , problems in employment , and early mortality . As the HIV+ population ages,how to dry cannabis understanding and addressing HIV-associated comorbidities that impact cognitive performance and everyday functioning is critical to overall healthcare for PLWH. Multiple adverse experiences such as childhood trauma, sexual abuse, physical violence, unemployment, and poverty are highly prevalent among PLWH and have known CNS consequences. For example, estimates of sexual and/or physical abuse in PLWH range from 30% to over 50% Whereas the physiological response to acute stress is typically adaptive, chronically-elevated stress exposure can disturb brain development and function, and increase risk of psychiatric disease . Chronic exposure to stress and stress hormones, glucocorticoids, can hinder immune mechanisms and amplify inflammation in the CNS and, furthermore, exacerbate injury-induced neuronal death . Chronic stress in healthy adults is linked to structural and functional alterations in the hippocampus and prefrontal cortex , and poorer memory recall ability . Due to the overlap in the inflammatory and immune mechanisms shown to be affected by stress and HIV, traumatic and stressful experiences may contribute to or compound the likelihood of CNS injury via this pathway in PLWH . Thus, PLWH with a history of trauma and adversity may be at increased risk for neurocognitive impairment and decreased functional capacity.

Among men living with HIV, a previous study found that stressful life events were related to worse executive functioning, attention, and processing speed . In women living with HIV, high levels of self-reported stress were associated with verbal memory deficits, as well as prefrontal cortex structural and functional deficits . Conversely, high stress was not associated with verbal memory performance in women without HIV, suggesting that stress may be particularly deleterious to cognitive function in the context of HIV. Another recent study found that PLWH with higher levels of social adversity showed reduced volumes of sub-cortical structures and worse learning/memory performance, and these findings did not extend to the HIV- group . Stress, emotional reactivity, and avoidant coping behaviors are related to important daily functioning behaviors such as medication non-adherence among PLWH . Although multiple studies have examined the effects of stress on cognitive function within cohorts of PLWH or individuals without HIV, few have directly compared the effects between serostatus groups while examining the combined effects of multiple traumatic and stressful experiences, or included standardized measures of daily functional abilities. In the present study, we investigated whether a composite measure of multiple adverse experiences including trauma, economic hardship, and stress exerts a negative impact on cognitive and everyday function in a cohort of adults living with and without HIV. We hypothesized that PLWH would experience more trauma, economic hardship, and stress than their HIV- counterparts. Furthermore, we hypothesized that elevated TES would relate to worse cognitive function and everyday function in both serostatus groups, but the magnitude of the association would be greater for PLWH compared to their HIV- counterparts, after controlling for established predictors of cognitive and functional status.Prior to conducting primary analyses, independent samples t-tests and Chi-square tests were used to compare HIV status groups on demographic, psychiatric, substance use, and clinical variables. Any variables that differed between the HIV+ and HIVgroups at p < .1 were added as covariates when analyzing the relationship between TES and cognitive/functional outcomes. Thus, we included gender, ethnicity, years of education, lifetime MDD, lifetime substance use disorder , lifetime alcohol use , and lifetime cannabis use in the models for cognition. We did not include current MDD as a covariate due to its low prevalence. For functional outcomes, we additionally included global neurocognitive impairment as a covariate. For PLWH-only models, any HIV disease characteristics that related to global cognition or ADL declines at p <.1 in univariable analyses were added as covariates. For our models in which a cognitive domain in PLWH wasthe outcome variable, current CD4 count was included as an additional covariate, given that it was associated with global cognition at p <.1 in univariable analyses. For our model in which a functional outcome in PLWH was the outcome variable, estimated duration of HIV infection was included as an additional covariate, given that it was associated with ADL declines at p <.1 in univariable analyses. We used multi-variable linear regression analyses to examine the independent and interactive effects of the TES composite and HIV status on cognitive function and declines in activities of daily living. Separate univariable models were run for each of the seven cognitive domains and global cognition, and alpha was set at 0.006 . We pursued multi-variable analyses only for those cognitive domains that showed a significant relationship with TES in univariable models, and in multi-variable analyses, alpha was set at 0.017 , based on the number cognitive domains tested. Post-hoc analyses examined how the components of our TES composite score correlated with each other in PLWH with Spearman’s rho correlations for continuous variables and Cohen’s d for dichotomous variables .

Unions were concerned that charter schools would be difficult to unionize, leading to losses of membership and union jobs. As far as school boards were concerned, chartering would diminish their power by allowing newly forming schools to apply directly to “authorizers” under state government jurisdiction. Though teachers unions generally had significant influence, particularly within the Democratic party, members of both parties wanted to stake out strong positions responding to perceived failures in the American education system . In addition, the prospect of vouchers loomed over the charter campaigns, in some cases smoothing the way for chartering. As one advocate in Connecticut said: “charters were advocated by groups whose embrace was intended to head off vouchers and other groups who saw them as a step toward vouchers or as the half-way to seek if they couldn’t get vouchers” . Similarly, in California, which adopted its charter law in 1992, advocates used the threat of a potential ballot initiative establishing a school voucher program to win votes. Though some Republicans would have preferred more radical policy shifts towards vouchers, many saw the charter laws as a step in this direction. Efforts to diffuse charter laws benefited also from a network of policy entrepreneurs sharing information and, in some cases, traveling state-to-state to draft bills and shepherd votes . There was also a growing network of supportive governors led by Lamar Alexander and Bill Clinton and operating through the NGA. Put together, this was a successful formula for the establishment of charter laws: by 2003, a full 40 states had passed charter school legislation.The simple tally of the adoption of charter laws presented in Table 1, though, obscures the significant variation in the specifics of these laws. In many states, charter opponents recognized that even if they could not prevent charter laws from being passed,greenhouse bench top they could use their political sway to dilute the laws and make it difficult for new charter schools to get a foothold. States that simply used “copy-paste” legislation from others were more likely to have weaker laws . By contrast, charter laws crafted to respond to a state’s specific needs generally were better for charter schools.

The favorability of a state’s charter law affected how much charter growth it would see. Growth depended heavily on states that had more generous funding formulas, higher caps on the number of schools that could be newly authorized, and less stringent authorization procedures . By 2000, around two-thirds of the 1497 charter schools that had opened were located in the 8 states whose charter policies were rated as an “A” by the Center for Education Reform .And the majority of charter schools outside of those 8 states were located in the 10 states rated as a “B”. Charter growth has remained concentrated in states with favorable charter laws.In 2017, the top 10 states in terms of charter penetration had an average charter law score from the Center for Education Reform of 151.7 , compared to just 109.4 in other states.The adoption of charter laws in the states has also changed the politics—as the literature on “policy feedback” effects would suggest it might . In the realm of social welfare, program beneficiaries tend to organize to protect and expand the policies that help them . Similarly, policies that benefit organized economic interests like firms and unions “feed back” into the politics by enhancing the resources these interests have at their disposal to make demands in the political sphere . The degree to which policies generate these supportive feedback effects factors critically into their durability and expandability . Reforms that reshape the politics by empowering supporters and disempowering opponents tend to prevail and potentially even expand. Reforms that fail to do so are often retrenched or eroded over time . In the case of charter politics, the obvious candidates to defend and expand charter laws, and thus drive feedback effects, are charter schools themselves. Charter schools owe their existence to the state charter laws adopted mostly in the 90’s and early 2000’s allowing charter schools to establish. They also receive critical funding from federal government programs. Their ability to sustain themselves and grow depends on the preservation and expansion of pro-charter policies—they are vested interests . Thus, a policy feedback perspective suggests that charter schools would leverage their growth in the K-12 education sector toward a greater political presence—both to defend against threats and take opportunities to expand the policies that benefit them.

This has occurred, to an extent. State charter school associations have emerged as important actors in education politics, particularly in states with significant charter school sectors. The California Charter Schools Association, for instance, has become a powerful interest group in California education politics—it spent roughly 18 million on lobbying and campaign contributions in 2015 and 2016.It has also leveraged the ability to mobilize charter parents, alumni, students, and school staff for political advocacy .In addition, leaders of large charter school networks have engaged in several high-profile political battles. For instance, Eva Moskowitz, the head of Success Academy Charter Schools, took center stage in mid-2010’s battles with New York City Mayor Bill de Blasio over co-location of charter schools with traditional public schools in city-owned school buildings . Yet, the political influence of charter schools themselves is also constrained. For one, they tend to have limited budgets. Even with federal funding, charter schools generally receive less money per student than traditional public schools . These funds are needed for educational purposes, generally leaving little left over for lobbying. And though charter schools can sometimes politically mobilize the parents of their students, the families served by charter schools are more likely to be low-income and people of color —groups that generally have less political sway. Nonetheless, as the role of charter schools in K-12 education has grown over the first two decades of the 21st century, so too has a powerful pro-charter advocacy coalition. Unlike the relatively shallow network of policy entrepreneurs driving initial charter school laws, elements of this coalition have maintained a steady presence in state capitols, local politics, and in Washington . Elements have also, in several cases, gone toe to-toe with powerful teachers unions . Many of these groups are associated with national networks—primarily Policy Innovators in Education Network and the newer 50-State Campaign for Achievement Now —that are active on charter policy and other education policy issues. Some accounts have suggested that certain wealthy foundations have played a significant role in bolstering this coalition through strategic grant-making.

Yet, the literature lacks systematic evaluation of the role of philanthropists in growing the pro-charter political coalition, and how this relationship has developed over time. One investigation that can help to illuminate the degree to which the charter school coalition is dependent on foundation funding is examining revenue sources for charter school associations. These associations, as mentioned above, have been critical to defending and expanding charter laws. We might suspect that the role of foundations in supporting charter school associations would shift over time. In the early 2000’s, with fewer charter schools to draw member revenue from, it is reasonable to think they’d be highly dependent on philanthropic funding—but, in the 2010’s, with the charter school sector better-established, they might have greater revenue from member schools, with foundation funding thus playing a relatively smaller role. I collect and analyze original non-profit revenue data for charter school associations to assess how revenue sources have changed over time. These data come from IRS 990 forms,cannabis dry rack which non-profits claiming federal tax-exempt status are required to file on a yearly basis. I examine the 32 organizations listed as partners of the National Alliance for Public Charter Schools ,and the National Alliance itself. The IRS 990 forms provide organizations’ main revenue sources, and the total. The two largest line items, generally, in the revenue streams are 1) contributions and grants, and 2) program service revenue.While contributions and grants also include grants from the government, stakeholder interviews suggest the great majority comes from foundations—whereas the majority of program service revenue comes from member organization dues. The data demonstrate a massive growth in charter school association revenue between 2003 and 2018. The 13 charter school associations operating in 2003 took in just under $6 million in total revenue in that year. But by 2018 the number of charter school associations had grown to 33, and the total revenue had grown over 17-fold to over $100 million. Indeed, charter school association revenue has grown much faster than charter schools themselves. As expected, charter school associations relied heavily on contributions and grants in the sector’s early years. In 2003, for instance, only 21 percent of charter school association revenue came from program services, while 70 percent of revenue came from contributions and grants . What is striking is that this breakdown has remained roughly constant, even as the number of charter schools and size of charter networks has grown considerably. The amount of total charter school association revenue from program services has hovered between 17 and 22 percent of the total—while the amount from contributions and grants has hovered at closer to 80 percent.

Even as the charter school sector grew, the political advocacy arm remained dependent on philanthropic funding for critical financial resources. How did foundations become such important backers of the pro-charter coalition that has emerged over the past two decades? Here, I present a mix of qualitative and quantitative evidence suggesting that a policy feedback lens is critical to understanding the role of foundations in charter school politics. The broad point is that the adoption of charter laws in the states allowed foundations to make investments in new schools—which they later supported with investments in pro-charter political advocacy. Foundations played a limited role in charter school politics during the initial diffusion of charter laws in the 90’s. As discussed above, this was a period when a network of policy entrepreneurs was able to exploit a “window of opportunity” to push through chartering legislation—despite opposition from incumbent K-12 education interests. This changed when charter schools started to establish in the late 90’s in early-adopting states. At the time, newer foundations with living donors like the Gates Foundation were interested in expanding their education portfolios, and had a higher risk tolerance and desire to experiment than more longstanding organizations like the Ford Foundation and Carnegie Foundation . In addition to supporting existing organizations, these newer foundations also sought to promote organizations that would compete with the traditional educational system structure— what Mehta and Teles call “jurisdictional challengers.” Newer foundations were also seeking to invest in organizations in the education policy space that would eventually develop their own revenue streams, and no longer need philanthropic support. The passage of charter laws in the states and emergence of charter schools presented the opportunity for these foundations to do just that by funding charter schools. In addition to providing newer foundations the opportunity to fund jurisdictional challengers, the establishment of charter schools in leading states provided a crucial “proof of concept” for the legitimacy of chartering. Chartering’s legitimacy as an education policy reform was also bolstered by early assessments demonstrating charters’ positive effects on certain metrics of student achievement . The flood of new charter schools also produced significant demand for foundation dollars from the charter school sector. Charter schools generally only begin to receive significant funding from state and local governments once students enroll. While the federal Charter School Program provided some funding for school creation, the sector still had a huge need for seed funding. In the late 90’s and early 2000’s, philanthropists like the Gates Foundation, Walton Family Foundation, and Broad Foundation became a crucial source of funding. In addition to providing funds directly to schools, foundations also supported the initiation and growth of organizations like the NewSchools Venture Fund that would provide its own seed funding to new charter schools. Beyond providing seed funding, foundations offered many charter schools continuing grants that supplemented relatively scarce public monies once the schools were operating. While existing research has documented the important role of foundations in charter school politics , we have limited evidence tracing out empirically how foundation support for the charter school sector has progressed over time. In this section, I draw on IRS 990 data to bring empirical analysis to bear on this question.

More specifically, policy reforms in the states they represent can increase pressures members face from organized groups and individuals in their constituencies to promote aligned federal policies. Empirically, I examine the effects of state marijuana legalization. The inferential design leverages differences across the states in statewide citizen initiative institutions, which provides exogenous variation in legalization. Instrumental variables analysis indicates legalization influenced pro-marijuana bill sponsorship and roll calls in the 116th Congress. The evidence points to growing influence of industry in legalizing states— including the ability to mobilize employees and customers—as the key mechanism, thus underscoring the importance of a political economy perspective for studying interdependencies in American federalism.During his tenure in the Senate , Cory Gardner became a central figure in federal marijuana policy. In 2018, Gardner vowed to block judicial nominees in the Senate until he received a commitment that the federal government would not prosecute marijuana industry . In the 116th Congress, Gardner sponsored core marijuana-related legislation including the SAFE Banking Act and the STATES Act. Gardner was not always so pro-marijuana. He opposed Colorado’s landmark 2012 ballot initiative legalizing marijuana for adult-use , and there is little in his record prior to 2012 that would indicate he would become an important marijuana proponent. At a basic level, Gardner’s pro-marijuana turn appears to be driven by a policy shift in the state he represented. The adoption of adult-use legalization in 2012 led to rapid marijuana industry growth in Colorado, which, as of 2018, took in the most industry revenue per capita of any state. The industry, according to journalist accounts, has gained leverage in Colorado politics,vertical grow racks compelling even conservative politicians like Gardner to support industry demands .

That a policy shift in the state of Colorado might affect the future politics—in this case by shaping the behavior of a member of Congress—accords broadly with the notion of policy feedback, whereby “policy, once enacted, restructures subsequent political processes” . Yet, these dynamics do not fit cleanly within existing policy feedback frameworks for two reasons. First, the policy feedback studies that investigate lawmaking as an outcome tend to rely on broad historical institutional analysis of qualitative data —not micro-level, quantifiable examinations of lawmaker behavior. While, more recently, the policy feedback literature has taken a micro-level turn, research in this vein has focused on the effects of policies on individual-level behavioral outcomes like turnout and attitudes , not the behavior of lawmakers. As a result, we have accumulated much quantitative evidence on how policies affect voters, and to a lesser extent, interest groups, but little on how it matters for lawmaking and public policy decisions. The second reason has to do with how policy feedback mechanisms operate within the federal system of American government. Classic studies of policy feedback examine the political implications of national policies, with scholars only more recently turning their attention to the sub-national level. Most of this sub-national-level work examines the effects of state policies on the politics in the states they were adopted. In this paper, I argue that state policy decisions can also affect how states and districts are represented at the national level. Broadly speaking, this is because members of Congress represent geographically demarcated units that are embedded in state policy landscapes, and these policy landscapes affect the political pressures that they face. First, state policies structure state economies, and in so doing can affect the ability of organized economic interests to engage in politics and make demands on their representatives. Second, state policies can affect the mobilization and preferences of individual voters, and thereby condition the pressures faced by re-election seeking members.

Finally, beyond potentially shaping preferences, state policy enactment might send a signal of constituent preferences that can be difficult for members of Congress to ignore. Put together, these mechanisms suggest the adoption of a policy at the state level can increase the pressure on members of Congress to promote aligned federal policies. Empirically, I examine marijuana policy reform, a case that provides critical analytical leverage for testing the argument. The wave of state-level legalization over the past two decades has produced great variation in policy landscapes across the states. Moreover, the importance of the statewide citizen initiative—only available in 24 of the states—for passing legalization provides exogenous variation that allows for causal estimation of the effect of state policy shifts on representation in Congress. Does marijuana legalization in the states they represent affect members’ behavior in Congress? Studying the 116th Congress, I find evidence that it does. Using whether states permit citizen initiatives as an instrument, I find that members of Congress representing legalizing states were more likely to sponsor or co-sponsor key pro-marijuana pieces of legislation, and also more likely to cast certain pro-marijuana roll-call votes. Bringing quantitative evidence and elite interviews together to investigate mechanisms, I find the most support for the role of growing industry influence in legalizing states, but also find some support for the role of signaling constituent preferences. I find little support for the notion that effects were driven by positive shifts to public favorability wrought by legalization. Though inability to precisely decompose mechanisms is a limitation, such a decomposition is not critical for the paper’s core contributions, which are two-fold. First, this paper provides novel theory and evidence on the ways that the policy terrain affects lawmaking in Congress. Establishing causation using quantitative designs in policy feedback research is notoriously difficult . This study is, to my knowledge, the first to leverage a quantitative causal inference design to estimate the effect of prior policy decisions not just on voter behavior or interest group mobilization, but also on the actions of lawmakers in Congress. In doing so, it has the potential to serve as a bridge between work in policy feedback and scholarship on Congress. Second, this study contributes to a growing body of literature that is fruitfully applying ideas about policy feedback to the study of policy interdependence and diffusion in American federalism.

While recent work has illuminated how state policies can “feed into” the interest group politics in other states and at the national level , this study explores how state policy decisions shape the politics in Congress. In doing so, it demonstrates the importance of federalism as an institution that structures policy and political change over time in American politics. The paper unfolds as follows. First, I develop the core theoretical framework linking state policy decisions to representation in Congress. I proceed to introduce the case—the politics of marijuana—and the design for estimating the causal effect of state legalization on the behavior of members of Congress. I next present the main empirical results, discuss the evidence on the contributions of different mechanisms, and conclude.What determines how members of Congress represent their states and districts? While analyses have highlighted diverse drivers of congressional behavior including ideology and partisanship , a consistent finding in the literature is that members represent the preferences of the citizens and organized groups that make up their constituencies. Individual- and group-level inputs are often taken, in this literature, as exogenous. But these factors, in addition to influencing policy, are also shaped by previously established policy through dynamics of policy feedback. Scholars have identified several mechanisms. Broadly speaking, considering individual behavior, policies both condition the resources that individuals can devote to politics and the way they interpret the role of government in their lives . Considering organized interests, policies can incentivize beneficiaries to form citizens groups to advocate for the preservation or expansion of policies . Public policies can also,indoor grow light shelves by changing the rules governing the economy, increase the political capacities of organized economic interests like firms and unions . If policies can affect the behavior of individuals and the landscape of organized interests, and members of Congress are responsive to the individuals and organized interests that make up their constituencies, we might then expect policy feedback dynamics to ultimately influence the lawmaking process. Indeed, classic historical institutional accounts have traced the full policy feedback cycle, showing how previously adopted policies reshaped the politics, and in doing so, affected the decisions of lawmakers decades later . The strength of this historical institutional scholarship is the wide lens and attention to macro-level change. However, this also means that it is more limited for generating expectations about how shifts to policy landscapes might affect the decisions of individual lawmakers. And because more micro-level policy feedback work generally focuses on intermediate outcomes at the voter and interest group levels, we have accumulated little quantitative evidence that captures the full policy feedback cycle. One important exception is Campbell’s analysis of the relationship between Social Security and senior political participation. While Campbell’s seminal account is mainly concerned with the individual and group-level feedback effects of Social Security, one chapter investigates outcomes in Congress, thereby completing the “participation-policy cycle” . Campbell shows that, while Democratic members across the board tended to oppose cuts to programs that principally benefit seniors, Republican members’ willingness to vote for program cuts depended in part on the number of seniors in their districts. This analysis provides useful evidence but does not provide a clear-cut test of the core hypothesis that the prior adoption of pro-senior policies affected lawmaker behavior in a future political era.

The reason is that we would expect members representing districts with more seniors to vote for more pro-senior policies even in the absence of prior policy adoption driving seniors’ political mobilization.24 Ideally, we would be able to compare the behavior of members representing districts in locales featuring pro-senior policies to the behavior of members representing locales without pro-senior policies. But, because programs like Social Security have national scope, there is limited variation to leverage. This paper relies on state policy variation for empirical leverage. Scholars of policy feedback are increasingly interested in state policy, and broadly speaking, studies have shown that state policies can produce the same sorts of feedback effects on voters and interest groups as national ones. The approach I take here is somewhat different. Instead of examining policies’ political implications in the states where they are passed, I examine the national-level political implications. In particular, I focus on how state policy decisions in the places they represent shape lawmakers’ behavior in Congress. There are several reasons why state policy decisions might affect representation in Congress. The first set of explanations I put forward focus on how state policies structure states’ political economies. State policy decisions can influence what sorts of economic activities are profitable, and as a result, which types of firms establish and grow—as well as which fail. And while this paper focuses on firms, the same is true of another set of powerful organized economic interests: labor unions. The ability of unions to develop and maintain organizational strength is heavily influenced by state policies like collective bargaining rules and “right-to-work” laws . This matters also for politics. These interests will therefore be in a stronger position to influence national politics. Moreover, the groups that benefit from, and are strengthened by, state-level policy decisions are likely to also benefit from the adoption of aligned policies at the national level—so might leverage their newfound strength towards that end. One core potential avenue for doing so is putting pressure on members of Congress representing geographic areas where they have a significant presence. It is well-established that members of Congress generally are more responsive to the interests of industries that employ constituents and provide state and local tax revenue in the districts they represent. What is novel here is the understanding that which interests grow and develop a strong presence can be a function of prior state-level policy decisions. In considering the mechanisms linking state policy to congressional representation, members’ re-election motive is a good starting point . Members generally care about their re-election, and firms and unions have demonstrated an ability to transform their economic presence into political power by engaging in elections. Firms mobilizing employees to support their political interests is widespread in contemporary American politics . Similarly, mobilizing members in elections is a key source of union political strength .

Leveraging state policy data and administrative data on charter schools, I first demonstrate the importance of favorable charter laws for promoting the growth of new charter schools. Though the great majority of the states permit charter schools, charter laws vary markedly in aspects like funding, caps on new schools, and authorization procedures. In 2017, charter schools enrolled over 10 percent of total K-12 public school students in states like Colorado and Louisiana, but less than 1 percent in Washington state and Mississippi. The emergence and growth of charter schools in states with favorable charter laws came at an opportune moment for foundations aiming to expand their education portfolios. Several newer foundations with living donors, dissatisfied with the state of education philanthropy, sought to provide more support to “jurisdictional challengers” —organizations that they hoped would provide fresh ideas and energy by competing with elements of the traditional K-12 education system. Charter schools presented these foundations with an opportunity to do so. Thus, in the early 2000’s, foundations provided a critical source of funds directly seeding new charter schools and developing operational infrastructure for establishing new schools. As charter schools grew and began to present a meaningful challenge to incumbent education interests, the politics became more contested. Charter school boosters increasingly recognized the importance of maintaining and expanding pro-charter policy to sustain growth. Drawing on tax data, I document a shift in the mid-2000’s,hydroponic shelves with foundations greatly increasing their charter advocacy grant-making over time. They expanded from primarily funding charter school operations to also providing grants to pro-charter education reform advocacy groups— thus engaging politically to defend prior investments made in the charter school movement.

In each of the policy areas I study, reforms achieved at the state-level did not just produce learning and competitive pressures for other states—they also fundamentally restructured the broader interest group politics. State-level reforms created and empowered coalitions of organized interests that deployed newfound strength to expand and diffuse preferred policies across multiple states and up to the federal level. In terms of theoretical contribution, these papers thus demonstrate the gains to examining mechanisms of policy interdependence outside of learning and competition. In particular, they suggest that policies that affect the resources of organized interests are likely to have political and policy implications across the federal system. The dissertation also compels scholars of federalism interested in exploring policy interdependencies to study other outcomes outside of the simple spread of policies across jurisdictions. When state policy reforms shift the resources or engagement of organized interests, the effects can go beyond the diffusion of those particular policy reforms. Take the marijuana case. Elements of the marijuana industry newly empowered by state legalization have not only lobbied for the diffusion of legalization to the federal level—they have also lobbied on issues of federal enforcement, banking regulations, tax policy, and others. Similarly, the pro-charter interest group coalition that formed following the passage of charter laws across a range of states in the 90’s advocated for different policies from the federal government than the states. State policies can, in addition to potentially driving a policy diffusion process, shift the broader organized group landscape in a particular policy area —thus opening avenues for additional reforms that were previously unavailable due to unfavorable interest group politics. What this means for reformers seeking to change policy over the long term is that attention to multiple sites and levels of government is essential. There is growing interest in considering how policies might be sequenced over time, leveraging positive feed backs to deconstruct blockages and ratchet up reforms .

These ideas are particularly compelling for considering the politics of decarbonization . This dissertation shows that space, in addition to time, is a relevant factor in considering how to ratchet up policy reforms. Reforms achieved at the state level can play a critical role in building an interest group coalition capable of not only expanding those state-level policies, but also propagating those and aligned policies to new locales.Once considered a backwater, state politics has become a critical arena of American politics. In the face of congressional gridlock, national-level political actors have turned to the states as venues for achieving their policy goals . Policy variation across the states is growing and is increasingly associated with whether Democrats or Republicans control state office . Perhaps as a result, candidates for state office have amassed huge sums of campaign contributions from outside of their states in recent years . Renewed interest in state politics is in part driven by an understanding that state-level decisions have implications for politics and policy across the country. This understanding is reflected in a rich tradition of scholarship in American federalism examining the ways that policies adopted in one federal unit can affect politics and policy making elsewhere . At the core of this literature is the concept of states as “laboratories of democracy” —the idea that state lawmakers learn from policy experiments carried out elsewhere. Building on this concept, existing literature studying the diffusion of policies across the federal system has focused on mechanisms of learning and competition . This paper2 argues that traditional policy diffusion mechanisms do not account for an important source of policy interdependence—namely, the effects of sub-national policies on the capacities of interest groups to influence policy in the broader federal system. My theoretical argument builds on the classic finding in the policy feedback literature that policies shape the landscape of organized interests represented in the political system . I identify and explore mechanisms by which these dynamics can manifest across units and levels of government—bringing a focus to the intergovernmental effects of policy on interest group politics.

Broadly speaking, sub-national reforms that benefit particular organized interests also tend to strengthen them politically. These interests, in turn, might have an economic incentive to apply newfound strength to seek to propagate the reforms that benefit them. In this way, mechanisms of policy feedback can—like learning and competition—drive policy diffusion . Empirically, I examine cross-state policy feedback—the effects of state policy on the politics in other states—in distributed, or rooftop, solar policy.Unlike traditional utility-owned centralized power sources, rooftop solar arrays are connected to distribution systems and are generally owned or leased by customers. State-level policy played a key role in promoting the strong growth of rooftop solar over the last decade. I show that, by empowering new business interests that subsequently engaged politically across federal system,vertical growing racks state solar policies affected the politics in other states. The clear role of state policy in the growth of a new industry makes rooftop solar an instructive case for examining cross-state feedback effects. But this is also a hard case to observe effects due to the power and opposition of incumbent electric utilities.The empirical analysis proceeds in three general steps. First, I bring together data measuring state rooftop solar policies with administrative data on solar installations to investigate the relationship between policy and solar growth. Results from two-way fixed effects regression models indicate faster rooftop solar growth in states with pro-rooftop solar policies. Though findings are consistent with advocates’ and industries’ understanding , this finding—by providing empirical evidence for the substantive importance of state policy—lays the groundwork for the subsequent analyses. In the second step, I bring together firm-level lobbying disclosure data with firm-level system installation data to examine the feedback effects of state rooftop solar policies both in the states where they are adopted and in other states. I specifically examine the political engagement of large installer firms that have been central to efforts to expand and defend pro-distributed solar policies. Results from multilevel modeling indicate that a rooftop solar installer’s lobbying in a particular state depends on its economic presence in that state—but also its economic presence across the states. In addition, by tracking the economic expansion and political activity of the two largest rooftop solar installers, I find evidence of firms seeking to influence policy in markets where they did not yet have an economic presence in preparation for potential expansion. These findings, combined with the results indicating the importance of state policy to industry growth, suggest that state policy decisions affected political contestation in other states. Installers relied on growth in early adopters of favorable rooftop solar policies to accumulate resources, and then deployed those resources to propagate favorable policies more broadly. Third, I present analysis suggesting these cross-state feedback effects had policy consequences. Two-way fixed effects models indicate that installer lobbying is associated with more favorable state policies, with larger effects in states with lower levels of distributed solar penetration.

Qualitative analysis of the case of South Carolina affirms the plausibility of installers influencing policy even in states where they did not have an economic presence. By partnering with local groups and hiring well-connected lobbyists, Sunrun was able to drive policy shifts that led to the construction of a new market it could then expand into. This paper contributes to a growing body of literature at the intersection of federalism, policy feedback, and interest groups . Recent advances have documented how federated unions rely heavily on resources from affiliates in states with favorable labor laws , and how renewable energy interests leveraged states with favorable policies as “beach-heads” for their expansion across the country . Empirically, I build on this developing literature by using a rich array of evidence to trace out a causal chain from state policy to shifts in interest group engagement to policy decisions in other states. Conceptually, while existing work in this area, especially that on labor unions, has focused on organizational maintenance, I bring a focus to the role of state policy in driving lobbying and policy expansion—thereby bridging this literature with scholarship on policy diffusion and policy interdependence. By integrating policy feedback and policy diffusion concepts, this paper provides a framework for understanding and examining a myriad of inter dependencies in our federal system that are difficult to study with existing theoretical frameworks. Sub-national policies do not just motivate learning and competition: they also fundamentally affect the resources of organized interests that, in many cases, engage politically across the federal system. This can serve as a mechanism of policy diffusion, as groups that benefit from, and are strengthened by, particular sub-national reforms deploy their newfound strength to propagate the reforms that benefit them. However, the cross-unit political engagement of organized interests empowered by sub-national reforms will, in most cases, go beyond simply seeking to propagate those reforms. As a result, the perspective put forward here suggests that, in addition to potentially initiating a process of diffusion, sub-national reforms can also more durably shape interest group competition in the broader federal system over long time horizons. In addition to theoretical contributions, this paper also has practical implications for climate advocates. Well-designed climate policies not only drive shifts from fossil fuel energy infrastructure to renewables infrastructure, but also replace fossil fuel political interests with clean energy interests . As I show, this positive feedback can manifest across state lines, lending weight to an institutionally and geographically pluralistic advocacy approach to climate politics. Prominent theoretical perspectives for studying policy interdependence in American federalism have focused on three key mechanisms: political learning, competition, and firm preferences for unified standards. The general concept of political learning goes at least as far back as Supreme Court justice Louis Brandeis’s famous characterization of the states as “laboratories of democracy” . The basic logic of political learning is straightforward. Re-election motivated government officials generally prefer policies that benefit their constituents. If officials observe that a policy is performing well in another federal unit, this indicates that the policy has a greater likelihood of succeeding in their own locale, so they are more likely to adopt it. As a result, well-performing policies, the theory suggests, will diffuse across units and levels of government . Of course, in practice, it is not so simple. As Gilardi points out, officials are concerned broadly with the political effects of adopting a particular policy, not just whether that policy worked well elsewhere. In Gilardi’s analysis, whether learning leads to policy diffusion depends on officials’ prior beliefs and ideologies, and experimental work supports the notion that ideology moderates political learning .4 Broadly speaking, this finding suggests that political polarization can weaken learning as a diffusion mechanism.

As pharmacological antagonism of CB1 receptors caused BD rats to seize, we also evaluated the relationship between changes in anandamide levels and seizure phenomena. Our results suggest that anandamide acts as both an endogenous antidyskinetic and anticonvulsive compound, in part via interactions with the opioid system. Russia’s overall disease burden attributable to substance use is one of the highest in the world . It is among the top three countries with the largest estimated populations of persons who inject drugs . Injection drug use is the main driver of HIV transmission in Russia , a country where HIV incidence is steadily increasing. This same trend is occurring throughout parts of Eastern Europe where substance use is widespread and stands in contrast to the decline or stabilization of HIV incidence currently seen in other parts of the world , reflecting a strong relationship between injection drug use and HIV transmission. Estimates from 2017 suggest 30.4% of all PWID in Russia were living with HIV . Despite the country’s fast growing HIV epidemic, access to prevention and harm reduction services, such as needle and syringe exchange programs and opioid agonist treatment are limited and non-existent, respectively . Further, available statistics suggest the majority of Russian adults living with HIV are not on antiretroviral treatment . Numerous barriers prevent access to treatment, including the multiple steps required to enter into HIV care, discrimination towards people living with HIV overall, and conservative legislation placing restrictions on same-sex and other non-traditional relationships, drug use and sex work . Effective,vertical grow system evidence-informed prevention strategies are urgently needed in Russia to prevent both HIV acquisition and transmission.

Understanding the country’s HIV dynamics and patterns of transmission and acquisition are essential for designing approaches to reach key populations, including PWID and their sexual partners. Globally, there are more male than female PWID but available data suggest that among PWID, women have a higher HIV prevalence in many settings , including Eastern Europe. A 2012 meta-analysis including data collected from 128,745 PWID, drawn from 117 studies in 14 countries found a modest but significantly higher HIV prevalence among female PWID, relative to male PWID . Many possible explanations exist for elevated risk for HIV infection among women relative to men, irrespective of drug use behaviors. Clear biological mechanisms underlie the differential outcomes of HIV infection in women and men. For example, male-to-female HIV sexual transmission is more efficient than female-to male transmission because HIV-1 infected women have lower infectious potential . Additionally, sex hormones in women contribute to enhanced susceptibility by affecting the vaginal mucosa. It is hypothesized that women have lower viral reservoirs . It is also widely recognized that sociocultural factors, particularly as they relate to attitudes and practices surrounding sexual behaviors contribute to disparities in HIV infection and transmission between men and women, in the context of heterosexual partnerships. Further, gender-based inequalities prevent many women from protecting themselves and/or their partner against HIV infection . Less is known about the elevated risk for HIV infection and transmission among women who inject drugs. It has been found that unsafe injection practices and condomless sex heighten HIV infection risk among all PWID, but evidence suggests women who inject drugs are disproportionately more likely to engage in these behaviors, compared to men . Despite these findings, female PWID are often underrepresented in research with drug users and in studies on access to HIV care .

Evidence-based prevention measures aimed at PWID are urgently needed and have been specifically called for in Russia, where PWID account for the largest proportion of new HIV diagnoses, relative to any other risk group in the country and where evidence is limited on gender differences in HIV risk behaviors among PLHIV who have ever injected drugs. A prior study with women who inject drugs in St. Petersburg found 64% were HIVpositive and, in the past year, over 50% had two or more sexual partners, 40% transacted sex, 40% had condomless sex and 40% shared injecting needles. Transactional sex and sexual violence were both associated with increased injection drug equipment sharing and violence was associated with increased condomless sex. A second St. Petersburg study , conducted with male and female PWID living with HIV found internalized stigma surrounding HIV and drug use was correlated with poorer health outcomes and lower likelihood of service utilization. Although this study did not examine gender differences in these relationships, other research suggests that, compared to men, women who use drugs experience more stigma related to gendered cultural norms which contributes to increased risk for negative HIV outcomes . “Women-specific” research and prevention approaches have been called for to better understand the true context in which drug-using women experience health risks, and to design programs that account for the social, micro, and macro levels of women’s lives . Further, it is imperative that efforts be placed on developing gender-specific strategies for conducting research and programs to understand and reduce female PWID’s risk for both HIV acquisition from and transmission to sexual and injection drug use partners. Although a developing body of research informs our ability to design gender-tailored programs to prevent HIV infection among HIV-negative female PWID , less is known about how to effectively prevent HIV transmission by HIV-positive PWID to HIV-negative sexual partners or injection drug use partners.

We aimed to assess the association between female gender and drug risk behaviors and sex risk behaviors among a population of HIV positive men and women who had ever injected drugs in St. Petersburg, Russia. Recognizing injection drug use as a chronic condition, and that most PWID go through repeated periods of injection cessation and relapses during their injection careers , we included participants who reported past month injection drug use and/or injection drug use prior to their HIV-positive diagnosis. Alcohol use prior to sharing injecting equipment and surrounding sex were secondary outcomes in our analysis because alcohol consumption overall including by PLHIV has been associated with significantly higher drug and sex risk behaviors that heighten vulnerability for HIV acquisition, as well as transmission to others . Additionally, some of the most risky patterns of drinking  have been observed in Russia and Ukraine. Based on what has been found in other studies in Russia and other settings, we hypothesized that among Russian PLHIV who had ever injected drugs, women would have higher odds of engaging in high risk drug use, sexual behaviors, and use of alcohol prior to sex or injecting drugs, relative to men. This study involved secondary analysis of data from the Russia ARCH cohort, which is part of the three site Uganda, Russia, Boston Alcohol Network for Alcohol Research Collaboration on HIV/AIDS Consortium. Russia ARCH is an observational prospective cohort study conducted to assess the longitudinal association between alcohol consumption and biomarkers of microbial translocation and inflammation/ altered coagulation, which also encompasses a nested randomized controlled trial aimed at assessing the efficacy of zinc supplementation on markers of inflammation. A sample of 351 Russia ARCH participants were recruited into the study between November 2012 and June 2015 from clinical HIV and addiction care sites, non-clinical sites,indoor vertical garden systems and via snowball recruitment in St. Petersburg. Eligibility criteria for inclusion in the cohort included the following: 18-70 years old; documented HIV-infection; documented ART-naïve status; the ability to provide contact information for two contacts to assist with follow-up; stable address within St. Petersburg or districts within 100 kilometers of St. Petersburg; possession of a home or mobile phone. The current analysis was restricted to people who had ever injected drugs, defined as individuals who reported a history of injection drug use prior to HIV diagnosis, or past 30-day injection drug use at study visit. Participants were excluded from the cohort if they were not fluent in Russian or had a cognitive impairment resulting in inability to provide written informed consent. Eligibility was verified and informed consent was obtained. Participants provided a blood sample and were administered an interview assessment. Institutional Review Boards of Boston University Medical Campus and First St. Petersburg Pavlov State Medical University approved this study. The main independent variable for this study was female gender. Gender was self-reported as male or female. We did not assess other gender categories. The two primary dependent variables of interest were sharing of injecting equipment in the past 30 days and condomless sex in the past 90 days.

Condomless sex was defined as vaginal or anal sex with any sexual partner without the use of a condom or other protective barrier. Three secondary outcomes of interest were also examined, including alcohol use prior to sharing injecting equipment in past 30 days, alcohol use before or during sex in past 90 days, and reporting of both of the primary dependent variables . The following were selected as covariates for inclusion in the adjusted models, due to their potential confounding effects: age, education , income , partnered status, and recent ART use at follow-up. We also controlled for recent ART use at follow-up using data from participants’ responses to the following question about ART use at the 12 month and 24 month follow-up, “in the last 6 months, have you taken anti-retroviral medications for treating HIV?” Covariates were selected based on clinical knowledge and the literature. Partnered status was a 3 level covariate with the following categories: not partnered, partnered HIV discordant partner , and partnered HIV concordant partner . There were 21 observations over the course of the study whereby a participant reported they did not know the HIV status of their partner. Data from these observations were excluded from the analyses. Partner denoted being married, in a domestic partnership/living with a partner, or in a long-term relationship . Also measured at baseline were median income in Russian Rubles with interquartile range , mean CD4 count , heroin or other opioid use in past 30 days, and cannabis use in past 30 days. Since measures for past month heroin or opioid use did not distinguish the mode by which the drug was taken, we included data from a question assessing any injection drug use in the past 30 days . Those who indicated past 30 day injection drug use were asked to specify the type of drug injected . Heavy alcohol use in past 30 days was measured via the 30-day Timeline Follow back Method and defined as heavy if meeting NIAAA at-risk drinking amounts . Because involvement in transactional sex was associated with injection risk among women in a recent St. Petersburg study , we measured whether both male and female participants reported having given sex to a partner, received sex from a partner or both given and received sex to/from a partner in exchange for money, alcohol, drugs, or other things in the past 12 months.We assessed baseline frequencies of demographic characteristics, covariates, CD4 count, heroin or other opioid use, any injection drug use , cannabis use, heavy alcohol use and transactional sex, overall and by gender. For descriptive purposes, we assessed differences between male and female participants at baseline using chi-square and Fisher’s exact tests for categorical variables, and t-tests and Wilcoxon rank-sum tests for continuous variables. We assessed baseline, 12 month and 24 month follow-up frequencies for the primary and secondary outcomes and 12 month and 24 month follow-up frequencies for ART use in the past 6 months. To account for the correlation from using repeated observations from the same study participants, separate generalized estimating equations logistic regression models were used to evaluate the association between gender and each of the binary outcomes controlling for potential confounders. An independence working correlation was used and robust standard errors from the GEE approach are reported. Odds ratios and 95% confidence intervals are presented from the logistic regression models.We then fit a partially adjusted model controlling only for the demographic covariates specified above . Lastly, we fit the final, fully adjusted model controlling for demographic covariates and marital/partner status and respondents’ reports of ART use in the past 6 months, as recorded at the 12 and 24 month follow-up interviews. A posthoc sensitivity analysis was performed excluding observations from participants who used ARTs. All analysis was done using the statistical package SAS 9.3At baseline the sample included 291 HIV-positive participants from the Russia ARCH cohort who had ever injected drugs.

Although a significant minority of patients remained ARVnaı¨ve throughout the study follow-up, we wanted to estimate adherence to combination highly active antiretroviral therapy stratified by psychiatric diagnosis and SU diagnosis status for study participants who did receive HAART. Adherence was measured using electronic pharmacy dispensing refill records; the “days supply of HAART medication was divided by the “total time elapsed between first day of HAART initiation and last day of HAART medication supply’’ over the first 12 and 24 months of study follow-up. All data analyses were conducted using SAS software, version 9.1 .The distributions of demographic and HIV-related clinical and behavioral characteristics by psychiatric diagnosis status are presented in Table 1. The results of w2 tests indicate significant differences in most characteristics between those patients with and without a psychiatric diagnosis. However it can be seen that the categories of these characteristics were still very similar in distribution in both groups. Finding significant results for very small differences in distributions is likely the consequence of having a very large sample size in this study. The majority of patients were white, male, 30–49 years of age at baseline, and belonged to the men who have sex with men HIV transmission risk group. CD4 Tlymphocyte cell counts measured at or near time of study entry were comparable in both patients with and without a psychiatric diagnosis. Similar results were observed for HIV RNA levels. Of the 2472 patients with a psychiatric diagnosis, 83.9% had one or more psychiatry department visits. The proportion of patients with any ARV therapy experience at baseline was similar across psychiatric disorder status,plant grow trays with on average 35% of all patients having no ARV experience. Throughout study follow-up, approximately 25% of all patients remained ARV naı¨ve.

Among those who were receiving HAART during study follow-up, mean adherence was estimated as 82.4% among patients with a psychiatric diagnosis at 12 months after initiation of HAART and 83.7% among patients with no psychiatric diagnoses; similar mean adherence was observed at 24 months. Patients diagnosed with SU problems showed mean adherence of 81.1% at 12 months after initiating HAART in comparison to 83.5% among patient without a SU problem diagnosis. Because adherence rates were similar across diagnostic status, we did not conduct a sub-analysis of ARV-experienced patients only, where adherence would have been included as a covariate in the regression model. The distribution of cause of death cross-tabulated by psychiatric diagnosis is presented in Table 2. The majority of deaths among patients with or without a psychiatric diagnosis were attributed to HIV/AIDS. The remaining causes of death had proportionately the same distribution across categories of psychiatric diagnosis status, with the possible exception of suicide which was twice as common among patients with a psychiatric diagnosis in comparison to patients with no diagnosis. Examining all-cause mortality for the entire study follow-up, we found an age-adjusted mortality rate of 28.6 deaths per 1000 person–years for patients with a psychiatric diagnosis versus 17.5 deaths for those without a psychiatric diagnosis. To examine the joint effects of psychiatric diagnosis, psychiatric treatment visits, SU diagnosis, and SU treatment on mortality, relative hazards were estimated using Cox proportional hazards regression. As mentioned in Statistical methods, the effects of psychiatric diagnosis/treatment and SU diagnosis/treatment were not additive, with statistically significant interactions between these covariates. RHs and 95% Confidence Intervals estimated from unadjusted and adjusted models are presented in Table 3. Categories of diagnosis and treatment are ordered from lowest to highest RH in the unadjusted model 1.

In comparison to patients with neither a psychiatric diagnosis nor a SU diagnosis , the highest risk of dying was found among patients with dual diagnoses but who had no psychiatric treatment visits and no SU treatment . This effect was somewhat attenuated after adjustment for potential confounders but remained statistically significant . Similar results were observed for patients who had a psychiatric diagnosis but no psychiatric services and no SU diagnosis that were very similar to those parameter estimates in model 2.During 12 years of follow-up , we observed a higher mortality risk for HIV-infected patients diagnosed with both psychiatric and SU disorders in comparison to patients with neither diagnosis. However, we observed that psychiatric and SU treatment, in general, reduced mortality risk in single and dual diagnosed patients, and remained statistically significant even after adjustment for age, race, immune status, HIV viral load, anti-retroviral therapy use, and other potential confounders. Accessing psychiatric treatment reduced mortality risk among dual diagnosed patients who were treated or not treated for SU disorder. Previous studies of individuals with HIV infection have found that those with psychiatric disorders are at elevated risk for poor medication adherence and clinical outcomes.There is substantial evidence that depression, stressful life events and trauma affect HIV disease progression and mortality.This effect has been found even controlling for medication adherence, in a study that showed that HAART adherent patients with depressive symptoms were 5.90 times more likely to die than adherent patients with no depressive symptoms.Depressive symptoms independently predicted mortality among women with HIV,and also in a separate study of men.Similarly, in multivariate analyses controlling for clinical characteristics and treatment, women with chronic depressive symptoms were 2 times more likely to die than women with limited or no depressive symptoms .Among women with CD4 cell counts of less than 200 10/ L, HIV-related mortality rates were 54% for those with chronic depressive symptoms and 48% for those with intermittent depressive symptoms compared with 21% for those with limited or no depressive symptoms.

Chronic depressive symptoms were also associated with significantly greater decline in CD4 cell counts after controlling for other variables.These mechanisms could help to explain the greater risk of mortality observed in our sample. Our findings strongly highlight the importance of access to psychiatric and SU disorder treatment for this population. It was estimated that during a 6-month period, 61.4% of 231,400 adults in the United States receiving treatment for HIV/AIDS used psychiatric or SU disorder treatment services.A significant number of HIV-infected patients report accessing psychiatric services.Such visits are associated with decreased risk of discontinuing HAART.Burnam et al.found that those with less severe HIV-related illness were less likely to access psychiatric or SU disorder treatment. One study found that engagement in SU disorder treatment was not associated with a decrease in hospital use by HIV-infected individuals with a history of alcohol problems.Improvement in depression was associated with increase in HAART adherence among injection drug users.26The translation of preclinical theories of alcoholism etiology to clinical samples is fundamental to understanding alcohol use disorders and developing efficacious treatments. Human subjects research is fundamentally limited in neurobiological precision and experimental control, whereas preclinical models permit fine grained measurement of biological function. However, the concordance between preclinical models and human psychopathology is often evidenced by face validity alone. The aim of this study, therefore, is to test the degree to which one prominent preclinical model of alcoholism etiology, the Allostatic Model, predicts the behavior and affective responses of human subjects in an experimental pharmacology design. The Allostatic Model was selected for translational investigation due to its focus on reward and reinforcement mechanisms in early vs. late stages of addiction. In this study, we advance a novel translational human laboratory approach to assessing the relationship between alcohol-induced reward and motivated alcohol consumption. A key prediction of the Allostatic Model is that chronic alcohol consumptions results in a cascade of neuroadaptations,custom grow room which ultimately blunt drinking-relate hedonic reward and positive reinforcement, while simultaneously leading to the emergence of persistent elevations in negative affect, termed allostasis. Consequently, the model predicts that drinking in late-stage dependence should be motivated by the relief of withdrawal related negative affect, and hence, by negative reinforcement mechanisms. In other words, the Allostatic Model suggests a transition from reward to relief craving in drug dependence.

The Allostatic Model is supported by studies utilizing ethanol vapor paradigms in rodents that can lead to severe withdrawal symptoms, escalated ethanol self-administration, high motivation to consume the drug as revealed by progressive ratio breakpoints, enhanced reinstatement, and reduced sensitivity to punishment. Diminished positive reinforcement in this model is inferred through examination of reward thresholds in an intracranial self-stimulation protocol. Critically, these allostatic neuroadaptations are hypothesized to persist beyond acute withdrawal, producing state changes in negative emotionality in protracted abstinence. Supporting this hypothesis, exposure to chronic ethanol vapor produces substantial increases in ethanol consumption during both acute and protracted abstinence periods. . Despite strong preclinical support, the Allostatic Model has not been validated in human populations with AUD.Decades of human alcohol challenge research has demonstrated that individuals differences in subjective responses to alcohol predict alcoholism risk. The Low Level of Response Model suggests that globally decreased sensitivity to alcohol predicts AUD. Critically however, research has demonstrated that SR is multi-dimensional. The Differentiator Model as refined by King et al suggests that stimulatory and sedative dimensions of SR differentially predict alcoholism risk and binge drinking behavior. Specifically, an enhanced stimulatory and rewarding SR, particularly at peak BrAC is associated with heavier drinking and more severe AUD prospectively. The Differentiator Model also suggests blunted sedative SR is an AUD risk factor, however, effect sizes for sedation are generally smaller. Both the LR and Differentiator models have garnered considerable empirical support in alcohol challenge research ; however, both models share some limitations. Human subjects research has not adequately tested whether SR represent a dynamic construct across the development of alcohol dependence and whether the motivational structure of alcohol consumption is altered in dependence vs. early non-dependent drinking. Recently, King et al. reported that the elevated stimulating and rewarding SR in heavy drinkers remained elevated over a 5-year period. Furthermore, this outcome was particularly strong among the ~10% of heavy drinking participants who showed high levels of AUD progression. In two previous alcohol challenge studies, we showed that stimulation/hedonia and craving are highly correlated among non-dependent heavy drinkers, whereas no stimulation-craving association was evidenced among alcohol dependent participants. These results were interpreted as being consistent with the Allostatic Model, insofar as the function of stimulation/ hedonia in promoting craving appeared diminished in alcohol dependence. Of note, however, neither study observed the hypothesized relationship between negative affect and craving among dependent participants. A primary limitation of these previous studies was the utilization of craving as a proxy end point for alcohol motivation and reinforcement. A recent study of young heavy drinkers found that both stimulation and sedation predicted free-access self-administration via craving. However, since this study did not include moderate–severe AUD participants, it is unclear whether the association between stimulation and self-administration is blunted in later-stage dependence. This study was designed to test whether SR predicts motivated alcohol self-administration and whether this relationship is moderated by alcohol use severity, thus providing much needed insight about the function of SR in alcohol reinforcement and advancing an experimental framework for translational science. Heavy drinkers ranging in their severity of alcohol use and problems completed a novel intravenous alcohol administration session consisting of a standardized alcohol challenge followed by progressive-ratio alcohol reinforcement. On the basis of the Allostatic Model, we predicted a strong relationship between stimulation and self-administration at low alcohol use severity, whereas no such association would be observed at greater alcohol use severity. Conversely, it was hypothesized that negative affect would be a stronger predictor of alcohol self administration among more severe participants. These two hypotheses would thus capture dependence-related blunting of positive reinforcement and enhancement of negative reinforcement.This study was approved by the Institutional Review Board at UCLA. Non-treatment seeking drinkers were recruited between April 2015 and August 2016 from the Los Angeles community through fliers and online advertisements . Initial eligibility screening was conducted via online and telephone surveys followed by an in-person screening session. After providing written informed consent, participants were breathalyzed, provided urine for toxicology screening, and completed a battery of self-report questionnaires and interviews. All participants were required to have a BrAC of 0 mg% and to test negative on a urine drug screen . Female participants were required to test negative on a urine pregnancy test. Inclusion criteria were as follows: age between 21 and 45, caucasian ethnicity , fluency in English, current heavy alcohol use of 14+ drinks per week for men or 7+ for women, if female, not pregnant or lactating, and using a reliable method of birth control , and body weight of less than 265lbs to reduce the likelihood of exhausting the alcohol supply during the infusion.

The first is whether endogenously produced anandamide and 2-AG participate in the modulation of specific disease states. Drugs that block endocannabinoid inactivation should magnify this adaptive function in the same way as serotonin reuptake or monoaminooxidase inhibitors heighten the mood-regulating actions of endogenous biogenic amines. The second question is whether inhibiting endocannabinoid clearance provides a therapeutic advantage over direct activation of cannabinoid receptors with agonist drugs. The latter approach has been generally favored thus far, and several classes of subtype-selective cannabinoid agonists are already available for preclinical use . Thus, demonstrating that inhibitors of endocannabinoid inactivation possess a unique pharmacological profile is essential to justify the substantial efforts associated with the development of a new class of drugs. In the following sections, we illustrate with some examples the endocannabinoids’ rolein pathology and discuss the potential therapeutic value of drugs that target endocannabinoid inactivation. Pain. For example, in vivo microdialysis experiments have shown that peripheral injections of the chemical irritant formalin are accompanied by increases in anandamide outflow within the PAG, a brain region intimately involved in pain processing . Since activation of CB1 receptors in the PAG causes profound analgesia, it has been argued that inhibitors of anandamide inactivation “may form the basis of a modern pharmacotherapy of pain, particularly in instances where opiates are ineffective” . The fact that the endocannabinoid transport inhibitor AM404 has no antinociceptive effect in models of acute pain seems to contradict this possibility . It should be noted, however,4×4 grow table that neither AM404 nor any other inhibitor of anandamide clearance has yet been tested in animal models that are directly relevant to pathological pain states in humans.

In models that mimic such states , the CB1 receptor antagonist SR141716A exacerbates pain when administered alone, suggesting that inflammation and nerve injury may be associated with compensatory increases in cannabinergic activity . If this hypothesis is correct, one would expect endocannabinoid inactivation inhibitors to alleviate inflammatory or neuropathic pain. This possibility has not yet been tested, however. Hypotensive Shock. During hemorrhagic and septic shock, anandamide and 2-AG may be released from macrophages and platelets, activate CB1-type receptors on the surface of vascular smooth muscle cells, and produce vasodilatation . The physiological significance of this response is still unclear. Nevertheless, the fact that a CB1 antagonist reduces survival time in “shocked” rats suggests that activation of the endocannabinoid system may have beneficial effects, possibly by redistributing cardiac output to or improving microcirculation in vital organs such as the kidneys . If this is true, inhibitors of endocannabinoid inactivation that do not appear to exert direct vasoactive effects could be used to prolong life expectancy in hemorrhagic and septic shock. Disorders of Dopamine Transmission. Functional interactions between dopamine and endocannabinoids are well documented. CB1 receptors are highly expressed in CNS regions that are innervated by dopamine-releasing neurons . In one of these regions, the striatum, anandamide release is stimulated by activation of dopamine D2-family receptors . Furthermore, the CB1 antagonist SR141716A, which has no effect on motor activity when administered alone, enhances the motor hyperactivity elicited by D2-family agonists . These findings suggest that one role of the endocannabinoid system in the CNS may be to act as an inhibitory feedback mechanism countering dopamine-induced facilitation of psychomotor activity . A corollary of this idea is that drugs that prevent endocannabinoid clearance should antagonize dopamine-mediated responses.

As a test of this hypothesis, the endocannabinoid transport inhibitor AM404 was injected into the cerebral ventricles of rats that were then systemically treated with the mixed D1/D2 dopamine agonist apomorphine or the selective D2-family agonist quinpirole. AM404 blocked the yawning evoked by apomorphine and reduced the motor stimulation elicited by quinpirole. By contrast, when administered alone, AM404 produced only a mild hypokinesia, not other cannabinoid actions such as catalepsy . The effects of AM404 were also studied in juvenile spontaneously hypertensive rats . Juvenile SHR are not yet hypertensive but are hyperactive and show a number of attention deficits, which have been linked to alterations in mesocorticolimbic dopamine transmission and dopamine receptor expression . Systemic administration of AM404 normalizes the behavior of juvenile SHR without affecting that of control rats . These findings suggest that inhibitors of endocannabinoid inactivation may be used to alleviate certain symptoms of dopamine dysfunction. Clinical data showing that 9 -tetrahydrocannabinol ameliorates tics in Tourette’s syndrome patients lend further support to this possibility . Future Challenges. In conclusion, three major challenges lie before the pharmacologist interested in the mechanisms of endocannabinoid inactivation from the perspective of drug discovery. The first is the need for a deeper molecular understanding of these mechanisms. Considerable insight has been gained in the last few years on the structure and catalytic properties of AAH, but many questions remain unanswered, including the identity of the putative endocannabinoid transporter and the existence of additional hydrolytic enzymes for anandamide and 2-AG. The second challenge lies in the development of potent and selective inhibitors of endocannabinoid inactivation. Future AAH inhibitors should combine the potency of those currently available with greater pharmacological selectivity and biological availability. A second generation of endocannabinoid transport blockers that overcome the limitations of AM404 and its congeners is also needed.

The third challenge is the validation of endocannabinoid mechanisms as targets for therapeutic drugs. This task is intertwined, of course, with that of understanding the endocannabinoids’ roles in normal physiology, one on which much research is currently focused.Alcohol abuse is a global problem, constituting the seventh leading risk factor for death and disability . Worldwide, over 100 million people had an alcohol use disorder in 2016. Statistics from the National Survey on Drug Use and Health show that >85% of adults in the United States report ever having consumed alcohol, with >25% reporting binge drinking in the past month . The proportion of adults in the United States with an AUD is estimated to be 6.2% . Alcohol use behaviors are complex, and how and why people drink is partially influenced by genetic factors. However, identifying the genetic factors that increase the risk for harmful drinking has been challenging, partially because patterns of alcohol use are dynamic across the lifespan. The terms used to describe alcohol use and abuse are as diverse as the behaviors themselves. Hazardous drinking describes heavy drinking that places an individual at risk for future harm. Harmful drinking and alcohol abuse are defined as drinking that causes mental or physical damage to the individual. These descriptive terms were devised to identify individuals who would benefit from brief interventions and are assessed using screening questionnaires such as the Alcohol Use Disorders Identification Test . Alcohol dependence was, until recently, defined according to the DSM-IV and required the presence of 3 or more of 7 criteria in a 12-month period. The DSM-IV made a distinction between alcohol abuse and dependence that was removed under DSM-V and replaced with ‘mild’ to ‘severe’ definitions of AUD. Genetic studies encompass the wide range of alcohol use phenotypes; in this review we mirror the language used in the original studies. AUD can be viewed as the end point of a series of transitions ,cannabis drying system which begin with the initiation of use, continue with the escalation to hazardous drinking and culminate in compulsive harmful use that persists despite negative consequences. Genome-wide association studies have been instrumental in discovering novel genetic loci associated with multiple psychiatric conditions. In the field of AUD genetics, studies have mostly focused on either levels of consumption or AUD diagnosis. Recent GWAS have now begun to identify hundreds of genome-wide significant variants, and provide evidence that the components of alcohol use behavior have a distinct genetic architecture.In this review, we provide an overview of recent molecular genetic findings of alcohol use behaviors from the largest GWAS performed to date. Other reviews have elegantly summarized findings from twin and family studies of heritability, linkage, candidate gene and GWAS [e.g. ], and we extend on recent reviews of the molecular genetics of AUD by including additional GWAS of alcohol use behaviors that identify genome-wide significant hits . In addition, we discuss the application of polygenic methods, which provide mounting evidence that alcohol use and misuse are partially distinct. Finally, we delineate future directions to investigate the different etiologic sources that underlie the life course of alcohol use behaviors. For decades, candidate gene studies were used to determine the contribution of specific genes that increase risk for AUD.

Candidate gene studies tended to focus on genes that influenced pharmacokinetic and pharmacodynamic factors. Larger genetic studies have generally not replicated the findings from candidate gene studies . One exception to this are the genes encoding ethanol metabolizing enzymes, particularly alcohol dehydrogenase and aldehyde dehydrogenase , which have repeatedly been shown to have the largest impact on alcohol consumption and risk for AUD . As study designs have evolved to incorporate GWAS, researchers have been able to scan the whole genome without any hypotheses about the underlying biology of alcohol use behaviors. Initial efforts focused on collecting clinically defined cases of AUD, but these ascertainment strategies could not amass the large sample sizes required for GWAS . Accordingly, multi-ethnic and clinically defined samples have been combined through the Psychiatric Genomic Consortium of Substance Use Disorders working group. The efforts of the PGC-SUD have led to a trans-ancestral meta-analysis consisting of almost 15,000 AD cases and almost 38,000 controls from 28 independent cohorts , identifying a single locus , which was robustly associated with AD. More recently, using information from electronic health records to infer AUD status, a GWAS of 274,424 multi-ethnic individuals from the Million Veterans Program cohort identified 10 loci associated with AUD . Kranzler et al showed that alcohol consumption and AUD were genetically correlated but distinct, thus allowing them to adjust for consumption in the AUD GWAS and for AUD in the GWAS of consumption. In parallel with these efforts, which have focused on clinical diagnoses, other GWAS have incorporated continuous measures of alcohol use. These include self reported weekly alcohol intake or the scores from screening questionnaires such as the AUDIT . The AUDIT can be decomposed to provide a measure of alcohol use from the first 3 questions and misuse from questions 4-10 . These quantitative measures are available in large population-based cohorts such as the UK Biobank , MVP and 23andMe. The GWAS meta-analysis of AUDIT identified 10 associated risk loci . Large consortia were also formed to collate quantitative measures of alcohol use, including AlcGen and the GWAS & Sequencing Consortium of Alcohol and Nicotine Use . GSCAN have recently identified nearly 100 loci associated with alcohol consumption . The MVP study also examined alcohol consumption, allowing for an explicit comparison between AUD and consumption in a single population; of the 18 loci detected in that study, 5 were common to both AUD diagnosis and alcohol consumption. As the prior two paragraphs make clear, population based cohorts have provided larger sample sizes, which are critical for obtaining adequate power for GWAS. Their use can come at the cost of missing more severe alcohol use phenotypes. For example, the frequency of AUD in the UKB is lower than the population average [7% ], indicating that certain population studies may be under powered to detect genetic effects specific to dependence . The frequency of AUD in the MVP, on the contrary, was much higher [20%, ]. Despite these limitations, population based cohorts provide a cost-effective strategy for obtaining very large samples, compared to traditional study designs that require obtaining a diagnosis from clinically trained staff. Table 1 summarizes the most recent GWAS of alcohol use behaviors ; Figure 2 provides an overview of the chronology of these studies. Figure 3 shows that the list of genes identified by these studies is highly heterogeneous. These data suggest incomplete genetic overlap between measures of alcohol use behaviors , though ascertainment bias and limited power are likely to be additional contributing factors. The 4q23 region, which contains the genes for several alcohol metabolizing enzymes, has been associated with multiple alcohol use behaviors. This association is one of the most consistently replicated findings in the field of psychiatric genetics, although the effects are clearly ancestry-specific . There appear to be multiple signals in this region, including ADH1C , ADH4 , ADH5 and the METAP1/EIF4E region .

The national prevalence rate was estimated to be 0.37 %. The southern region accounted for almost 50 % of total cases, and had the highest number of cases compared to the other three regions of Vietnam: northern, central and highland. Vietnam is still facing an HIV epidemic that has occurred primarily in PWIDs and female sex workers . Recently, the epidemic has been rising significantly among MSM . The team visited these hotspots to estimate the numbers of MSM in each. With the assistance of MSM and hotspot owners, additional hotspots were identified, yielding a total of 745. Local health staff, with the help of MSM peers, accessed these venues and conducted rapid interviews of hotspot owners and several MSM to get information for estimating the size of the MSM population and how to approach MSM in each hotspot. MSM were invited to participate in this survey if they were at least 16 years old and self reported having had oral and/or anal sex with another male in the past 12 months. Those with any history of poor blood clotting were excluded due to the risk of prolonged bleeding after drawing of blood, and those with hearing disorders were excluded due to the difficulty for them to clearly hear and understand the questions being asked and responding to them correctly.There were differences between provinces in sample sizes because of variations in prevalence estimates and/or limited funding. The HIV prevalence among MSM per site was estimated using proxy data of nearby provinces . We also had personal communications with peer educators and staff of provincial AIDS centers from the study provinces to gain insights into the probable HIV prevalence and risk behaviors among MSM to estimate the HIV prevalence for selecting suitable sample sizes. The prevalence of HIV in MSM in southern Vietnam was estimated to be approximately 4 %,horticulture solutions and the desired precision was set at 2 %, indicating that a sample size of 369 was needed; allowing 10 % for incomplete data and specimen damage, the sample size was rounded to 400.

However, since funding was insufficient, the sample size was lower for four provinces, where the estimated prevalence was approximately 3 %, and the desired precision was set at 2 %. The sample size needed was 279, rounded to 300. For Vinh Long, a sample size of 338 was obtained, since more individuals were willing to participate. The surveys were conducted in the listed hotspots in each province , in which the number of MSM was estimated. The sample size in each province was stratified based on the estimated size of MSM population in each district, then in each hotspot. All interviewers, medical technicians, and physicians attended a three-day training course specific for conducting the study. Informed consent was obtained prior to face-to-face interviews to collect data on sociodemographic characteristics, sexual identity, sexual behaviors, knowledge related to HIV and sexually transmitted infection , history of STIs, alcohol and recreational drug use, and access to HIV/STI intervention programs. After the interview, four ml of blood and 50 ml of urine were collected. Interviews were conducted by health staff or staff with a background in social sciences who were trained to administer the questionnaire. Biological samples were taken by trained phlebotomists according to national protocols. HIV testing was performed using ELISA and a rapid test . All specimens were tested at provincial AIDS centers. Syphilis was screened using RPR at the AIDS centers. Positive specimens were transported to the Pasteur Institute in Hochiminh City for further confirmation by the Treponema pallidum haemagglutination assay . If positive for both tests, the specimen was considered positive for syphilis. Due to limited funding, syphilis testing was only performed in seven provinces . Neisseria gonorrhoeae and Chlamydia trachomatis were tested by PCR at the PIHCM for only six provinces . The test results were returned to the participants through local voluntary HIV counseling and testing clinics. Men infected with syphilis, NG, and/or CT were referred to local STI centers for free treatment according to national STI treatment syndrome guidelines .

HIV-positive individuals were referred to local outpatient clinics. All interview answer sheets were checked by the interviewers for any missing information, then sent to the supervisors for futher checking before being sent to PIHCM. Interview answer sheets were stored in locked cabinets in the Provincial AIDS Centers and sent to PIHCM. Data were entered using Epi-Data version 3.1 , and all statistical analyses were carried out using Stata version 13.0 . Frequency distributions and percentages were used to describe the HIV infection rate and several qualitative variables. Mean, median and variance were estimated for quantitative continuous variables. These parameters were also used to clean data before further analysis. To partially reduce the effect of temporal relationships between HIV and risk behaviors, those who had been tested for HIV previously and knew they were HIV-positive were removed from the univariate and multivariate analyses, because they might have altered their risk behaviors, and this could possibly cause an inverse association if binary logistic regression analysis was used. Potential covariates were first identified in the existing literature or by subjective prior knowledge plus those variables with p values of ≤0.25 in univariate analysis, and were entered in the full model. Backward elimination was used. Any variable which had a p value over 0.05 was removed from the model. A log likelihood ratio test was performed to compare the “bigger” and “reduced” models. If the log likelihood ratio test gave a p value of ≤0.05, the corresponding variable was retained in the model. The procedure was repeated until no other variables in the model yielded p values of >0.05. The final estimates were also adjusted for cluster effects . The median number of male oral sex partners in the past 3 months was two, while more than one-third of participants reported having 2–4 male anal sex partners in the past three months. The majority of participants were unmarried, and 89.8 % engaged in sex with male partners, but 30.7 % also had sex with females/girl-friends. Few had engaged in sex with a foreigner in the past 12 months.

We found that 49.2 % of those who had ever engaged in sex with a foreigner had ever had transactional sex with male or female clients. Additionally, 24.9 % of those who never engaged in sex with a foreigner ever had transactional sex with male or female clients . One-fourth had had sex with male clients, and 10.4 % had had sex with a male sex worker in the past 12 months. Only 43.5 % had consistently used condoms with any anal sex partners, and 22.7 % never used condoms. Unprotected anal intercourse was slightly higher among unmarried MSM than ever-married MSM . Participants also engaged in sex with their wives/cohabiting partners or female sex workers in the past 12 months, and female clients in the past three months. The rate of consistent condom use with female sex workers was 68.4 %. Lubricant was also used by almost 40 % for anal sex with either males or females .In univariate analysis, HIV infection was more prevalent among older MSM,grow benches those residing in the southeastern provinces , small businessmen/vendors or freelance singers/barbers, those reporting having a religion, ever having sex with a foreigner, consuming alcohol on a daily basis, ever using recreational drugs , and those who thought that they were likely or very likely to be infected with HIV. HIV was less prevalent among those who had higher education levels, and/or never or only sometimes consumed alcohol immediately before having sex. In multivariate analysis, 10 factors were associated with HIV in the final model, including having ever married, having a religion, exclusively/frequently receptive, engaging in sex with a foreigner in past 12 months, consuming alcohol before anal sex in the past 3 months, using condoms during anal sex in the past three months, ever using recreational drugs, using amphetamine-type stimulants /heroin, perceiving oneself to be likely/very likely to be infected, and testing positive for syphilis. When age was increased by one year , the risk of HIV infection increased by 13 % . HIV infection was higher among MSM who had a religion , ever engaged in anal sex with a foreigner , and/or were syphilis-seropositive . Compared with those who had never used recreational drugs, those who reported previously but no longer using , currently inhaling/swallowing drugs , or currently injecting drugs were at significantly increased risk of HIV. When the drug use route was replaced by types of drug in the final model, compared with those who had never used recreational drugs, those who reported using ATS or heroin were at a higher risk of HIV infection. Moreover, MSM who thought that they were likely or very likely to be infected with HIV were at a higher risk of HIV infection.

MSM who had ever married , were exclusively or frequently receptive , sometimes consumed alcohol immediately before having sex , and/or frequently used condoms during anal sex in the past three months were less likely to be infected with HIV. The observed prevalence of HIV among MSM in the eight provinces was low compared with other provinces in Vietnam [>5 % in Hanoi, Hochiminh City, Can Tho and An Giang ], except for Dong Nai. The prevalence of HIV in the southwestern provinces was lower than that observed in southeastern provinces, including Dong Nai . Dong Nai borders with Hochiminh City, which has amongst the highest prevalence of HIV in Vietnam in all high-risk groups, including those who inject drugs, MSM, and female sex workers. Previous studies among MSM in Vietnam were carried out in urban populations, whereas our study was conducted in rural or small urban areas, except for Dong Nai which is an industrial province where HIV prevalence may be lower. The prevalence of HIV in the current study, 2.6 %, was lower than in other countries, including 13.6 % in Brazil, 12.9 % in northern Thailand, and 4.8 % in Beijing, China. Several correlates of HIV infection were identified in this study. Increasing age was found to be correlated with a higher likelihood of HIV infection, perhaps due to cumulative exposure, as was observed in studies in Malawi, Namibia, and Botswana and China . Ever being married was associated with a lower likelihood of HIV, similar to that observed in China; unmarried and homosexual MSM who did not have female sex partners were six-fold more likely to be infected with HIV compared to married or non-homosexual MSM with a female partner. Both that study and ours found that unprotected anal intercourse among married MSM was lower than among those who had never married. The association between having a religion and HIV infection found in this study might be due to infected individuals seeking consolation with religion. However, it is possible that people may believe that their destinies are decided by God and therefore take fewer precautions. It has been shown that personal sexual behaviors and cultures are sometimes related to religion. Hence, education about HIV transmission and prevention should be discussed with religious leaders so they can deliver appropriate messages to MSM and their partners or families. Recreational drug use, especially injecting, was shown to be highly associated with HIV, consistent with a number of other studies. Drug injection was associated with a higher risk of HIV than inhalation, smoking, or swallowing drugs. The fact that those who had previously but no longer used drugs had higher rates of HIV infection suggests either under-reporting current drug use or quitting drug use when learning they were HIV-positive. The risk of HIV infection was different according to drug used: cannabis , ATS , and heroin . Receptive anal intercourse was found to be an important risk factor for sexual HIV transmission in several studies. However, in our study, receptive anal intercourse was associated with a lower likelihood of HIV infection than for those who were exclusively or frequently insertive. This could be partly explained by a higher rate of recreational drug use in the “insertive” group than the “receptive” group in our study. Although a low proportion of MSM engaged in sex with foreigners, this was significantly associated with a higher risk of HIV infection.