Cabell Huntington Hospital

BACKGROUND: Currently, physicians are limited in their ability to provide an accurate prognosis for COVID-19 positive patients. Existing scoring systems have been ineffective for identifying patient decompensation. Machine learning (ML) may offer an alternative strategy. A prospectively validated method to predict the need for ventilation in COVID-19 patients is essential to help triage patients, allocate resources, and prevent emergency intubations and their associated risks. METHODS: In a multicenter clinical trial, we evaluated the performance of a machine learning algorithm for prediction of invasive mechanical ventilation of COVID-19 patients within 24 h of an initial encounter. We enrolled patients with a COVID-19 diagnosis who were admitted to five United States health systems between March 24 and May 4, 2020. RESULTS: 197 patients were enrolled in the REspirAtory Decompensation and model for the triage of covid-19 patients: a prospective studY (READY) clinical trial. The algorithm had a higher diagnostic odds ratio (DOR, 12.58) for predicting ventilation than a comparator early warning system, the Modified Early Warning Score (MEWS). The algorithm also achieved significantly higher sensitivity (0.90) than MEWS, which achieved a sensitivity of 0.78, while maintaining a higher specificity (p < 0.05). CONCLUSIONS: In the first clinical trial of a machine learning algorithm for ventilation needs among COVID-19 patients, the algorithm demonstrated accurate prediction of the need for mechanical ventilation within 24 h. This algorithm may help care teams effectively triage patients and allocate resources. Further, the algorithm is capable of accurately identifying 16% more patients than a widely used scoring system while minimizing false positive results.

OBJECTIVE: To test the efficacy of single family room (SFR) neonatal intensive care unit (NICU) designs, questions regarding patient medical progress and relative patient safety were explored. Addressing these questions would be of value to hospital staff, administrators and designers alike. STUDY DESIGN: This prospective study documented, by means of Institution Review Board-approved protocols, the progress of patients in two contrasting NICU designs. Noise levels, illumination and air quality measurements were included to define the two NICU physical environments. RESULT: Infants in the SFR unit had fewer apneic events, reduced nosocomial sepsis and mortality, as well as earlier transitions to enteral nutrition. More mothers sustained stage III lactation, and more infants were discharged breastfeeding in the SFR. CONCLUSION: This study showed the SFR to be more conducive to family-centered care, and to enhance infant medical progress and breastfeeding success over that of an open ward.

BACKGROUND: Sepsis remains a common condition with high mortality when multiple organ failure develops. The evidence for therapeutic plasma exchange (TPE) in this setting is promising but inconclusive. Our study aims to evaluate the efficacy of adjunct TPE for septic shock with multiple organ failure compared to standard therapy alone. METHODS: A retrospective, observational chart review was performed, evaluating outcomes of patients with catecholamine-resistant septic shock and multiple organ failure in intensive care units at a tertiary care hospital in Winston-Salem, NC, from August 2015 to March 2019. Adult patients with catecholamine-resistant septic shock (≥ 2 vasopressors) and evidence of multiple organ failure were included. Patients who received adjunct TPE were identified and compared to patients who received standard care alone. A propensity score using age, gender, chronic co-morbidities (HTN, DM, CKD, COPD), APACHE II score, SOFA score, lactate level, and number of vasopressors was used to match patients, resulting in 40 patients in each arm. RESULTS: The mean baseline APACHE II and SOFA scores were 32.5 and 14.3 in TPE patients versus 32.7 and 13.8 in control patients, respectively. The 28-day mortality rate was 40% in the TPE group versus 65% in the standard care group (p = 0.043). Improvements in baseline SOFA scores at 48 h were greater in the TPE group compared to standard care alone (p = 0.001), and patients receiving adjunct TPE had a more favorable fluid balance at 48 h (p = 0.01). Patients receiving adjunct TPE had longer ICU and hospital lengths of stay (p = 0.003 and p = 0.006, respectively). CONCLUSIONS: Our retrospective, observational study in adult patients with septic shock and multiple organ failure demonstrated improved 28-day survival with adjunct TPE compared to standard care alone. Hemodynamics, organ dysfunction, and fluid balance all improved with adjunct TPE, while lengths of stay were increased in survivors. The study design does not allow for a generalized statement of support for TPE in all cases of sepsis with multiple organ failure but offers valuable information for a prospective, randomized clinical trial.

OBJECTIVE: To evaluate angiogenesis in endometrial hyperplasia and stage I endometrial carcinoma, and to investigate the relationship between angiogenesis and tumor grade and depth of invasion. METHODS: Three groups of patients were analyzed: control patients who underwent hysterectomy for benign conditions (n = 19), patients with endometrial hyperplasia (n = 24), and patients with stage I endometrial carcinoma (n = 34). All hysterectomy specimens were stained immunohistochemically for factor VIII-related antigen as a sensitive and specific marker for vascular endothelium. Areas close to the deepest myometrial invasion or those with the highest grade of endometrial hyperplasia and the highest angiogenic intensity were selected. Three fields (x 400) were selected for each slide, and the mean microvessel count was calculated. Statistical analysis included Mann-Whitney U test or Kruskal-Wallis analysis of variance and Dunn post hoc procedure. P < .05 was considered significant. RESULTS: A significant difference was found between the microvessel count of controls versus the group with complex endometrial hyperplasia (median 21, range 16-80, versus median 38, range 20-130; P < .05). Microvessel counts of complex endometrial hyperplasia were significantly higher than those of simple hyperplasia (median 25, range 16-42; P < .05) and significantly lower than counts of endometrial carcinoma (median 77.5, range 19-189; P < .05). Microvessel counts in complex hyperplasia were not significantly different than those of noninvasive stage I endometrial carcinoma (median 38, range 20-130, versus median 44, range 19-119; P = .5). In cases of stage I endometrial carcinoma, a higher number of microvessels was noted in specimens with myometrial invasion than in those without myometrial invasion (median 44, range 19-119, versus median 83, range 19-189; P < .01). A higher number of microvessels was noted in cases with grade 2 than in those with grade 1, stage I endometrial carcinoma (median 44, range 19-98, versus median 96, range 63-189; P < .001). CONCLUSION: Complex endometrial hyperplasia and endometrial carcinoma are angiogenic. Furthermore, in stage I endometrial carcinoma, greater depth of invasion and higher tumor grade are directly correlated with angiogenic intensity.

OBJECTIVE: With neonatal intensive care units (NICUs) evolving from multipatient wards toward family-friendly, single-family room units, the study objective was to compare satisfaction levels of families and health-care staff across these differing NICU facility designs. STUDY DESIGN: This prospective study documented, by means of institutional review board-approved questionnaire survey protocols, the perceptions of parents and staff from two contrasting NICU environments. RESULT: Findings showed that demographic subgroups of parents and staff perceived the advantages and disadvantages of the two facility designs differently. Staff perceptions varied with previous experience, acclimation time and employment position, whereas parental perceptions revealed a naiveté bias through surveys of transitional parents with experience in both NICU facilities. CONCLUSION: Use of transitional parent surveys showed a subject naiveté bias inherent in perceptions of inexperienced parents. Grouping all survey participants demographically provided more informative interpretations of data, and revealed staff perceptions to vary with position, previous training and hospital experience.

BACKGROUND: Severe sepsis and septic shock are among the leading causes of death in the USA. While early prediction of severe sepsis can reduce adverse patient outcomes, sepsis remains one of the most expensive conditions to diagnose and treat. OBJECTIVE: The purpose of this study was to evaluate the effect of a machine learning algorithm for severe sepsis prediction on in-hospital mortality, hospital length of stay and 30-day readmission. DESIGN: Prospective clinical outcomes evaluation. SETTING: Evaluation was performed on a multiyear, multicentre clinical data set of real-world data containing 75 147 patient encounters from nine hospitals across the continental USA, ranging from community hospitals to large academic medical centres. PARTICIPANTS: Analyses were performed for 17 758 adult patients who met two or more systemic inflammatory response syndrome criteria at any point during their stay ('sepsis-related' patients). INTERVENTIONS: Machine learning algorithm for severe sepsis prediction. OUTCOME MEASURES: In-hospital mortality, length of stay and 30-day readmission rates. RESULTS: Hospitals saw an average 39.5% reduction of in-hospital mortality, a 32.3% reduction in hospital length of stay and a 22.7% reduction in 30-day readmission rate for sepsis-related patient stays when using the machine learning algorithm in clinical outcomes analysis. CONCLUSIONS: Reductions of in-hospital mortality, hospital length of stay and 30-day readmissions were observed in real-world clinical use of the machine learning-based algorithm. The predictive algorithm may be successfully used to improve sepsis-related outcomes in live clinical settings. TRIAL REGISTRATION NUMBER: NCT03960203.

Background: Machine learning methods have been developed to predict the likelihood of a given event or classify patients into two or more diagnostic categories. Digital twin models, which forecast entire trajectories of patient health data, have potential applications in clinical trials and patient management. Methods: In this study, we apply a digital twin model based on a variational autoencoder to a population of patients who went on to experience an ischemic stroke. The digital twin’s ability to model patient clinical features was assessed with regard to its ability to forecast clinical measurement trajectories leading up to the onset of the acute medical event and beyond using International Classification of Diseases (ICD) codes for ischemic stroke and lab values as inputs. Results: The simulated patient trajectories were virtually indistinguishable from real patient data, with similar feature means, standard deviations, inter-feature correlations, and covariance structures on a withheld test set. A logistic regression adversary model was unable to distinguish between the real and simulated data area under the receiver operating characteristic (ROC) curve (AUCadversary = 0.51). Conclusion: Through accurate projection of patient trajectories, this model may help inform clinical decision making or provide virtual control arms for efficient clinical trials.

Cancer progression is a complex multistep process comprising of angiogenesis of the primary tumor, its invasion into the surrounding stroma and its migration to distant organs to produce metastases. Nutritional compounds of the "capsaicinoid" family regulate angiogenesis, invasion and metastasis of tumors. Capsaicinoids display robust anti-angiogenic activity in both cell culture and mice models. However, conflicting reports exist about the effect of capsaicinoids on invasion of metastasis of cancers. While some published reports have described an anti-invasive and anti-metastatic role for capsaicinoids, others have argued that capsaicinoids stimulate invasion and metastasis of cancers. The present review article summarizes these findings involving the bioactivity of capsaicin in angiogenesis, invasion and metastasis of cancer. A survey of literature indicate that they are several articles summarizing the growth-inhibitory activity of capsaicinoids but few describe its effects on angiogenesis, invasion and metastasis in detail. Our review article fills this gap of knowledge. The discovery of a second generation of natural and synthetic capsaicin analogs (with anti-tumor activity) will pave the way to improved strategies for the treatment of several human cancers.

BACKGROUND: Racial disparities in health care are well documented in the United States. As machine learning methods become more common in health care settings, it is important to ensure that these methods do not contribute to racial disparities through biased predictions or differential accuracy across racial groups. OBJECTIVE: The goal of the research was to assess a machine learning algorithm intentionally developed to minimize bias in in-hospital mortality predictions between white and nonwhite patient groups. METHODS: Bias was minimized through preprocessing of algorithm training data. We performed a retrospective analysis of electronic health record data from patients admitted to the intensive care unit (ICU) at a large academic health center between 2001 and 2012, drawing data from the Medical Information Mart for Intensive Care-III database. Patients were included if they had at least 10 hours of available measurements after ICU admission, had at least one of every measurement used for model prediction, and had recorded race/ethnicity data. Bias was assessed through the equal opportunity difference. Model performance in terms of bias and accuracy was compared with the Modified Early Warning Score (MEWS), the Simplified Acute Physiology Score II (SAPS II), and the Acute Physiologic Assessment and Chronic Health Evaluation (APACHE). RESULTS: The machine learning algorithm was found to be more accurate than all comparators, with a higher sensitivity, specificity, and area under the receiver operating characteristic. The machine learning algorithm was found to be unbiased (equal opportunity difference 0.016, P=.20). APACHE was also found to be unbiased (equal opportunity difference 0.019, P=.11), while SAPS II and MEWS were found to have significant bias (equal opportunity difference 0.038, P=.006 and equal opportunity difference 0.074, P<.001, respectively). CONCLUSIONS: This study indicates there may be significant racial bias in commonly used severity scoring systems and that machine learning algorithms may reduce bias while improving on the accuracy of these methods.

A recent human immunodeficiency virus (HIV) outbreak among people who inject drugs in Cabell County, West Virginia, underscores the importance of HIV prevention, diagnosis, and care in communities with high rates of substance use disorder.

BACKGROUND: Severe sepsis and septic shock are among the leading causes of death in the United States and sepsis remains one of the most expensive conditions to diagnose and treat. Accurate early diagnosis and treatment can reduce the risk of adverse patient outcomes, but the efficacy of traditional rule-based screening methods is limited. The purpose of this study was to develop and validate a machine learning algorithm (MLA) for severe sepsis prediction up to 48 h before onset using a diverse patient dataset. METHODS: Retrospective analysis was performed on datasets composed of de-identified electronic health records collected between 2001 and 2017, including 510,497 inpatient and emergency encounters from 461 health centers collected between 2001 and 2015, and 20,647 inpatient and emergency encounters collected in 2017 from a community hospital. MLA performance was compared to commonly used disease severity scoring systems and was evaluated at 0, 4, 6, 12, 24, and 48 h prior to severe sepsis onset. RESULTS: 270,438 patients were included in analysis. At time of onset, the MLA demonstrated an AUROC of 0.931 (95% CI 0.914, 0.948) and a diagnostic odds ratio (DOR) of 53.105 on a testing dataset, exceeding MEWS (0.725, P < .001; DOR 4.358), SOFA (0.716; P < .001; DOR 3.720), and SIRS (0.655; P < .001; DOR 3.290). For prediction 48 h prior to onset, the MLA achieved an AUROC of 0.827 (95% CI 0.806, 0.848) on a testing dataset. On an external validation dataset, the MLA achieved an AUROC of 0.948 (95% CI 0.942, 0.954) at the time of onset, and 0.752 at 48 h prior to onset. CONCLUSIONS: The MLA accurately predicts severe sepsis onset up to 48 h in advance using only readily available vital signs extracted from the existing patient electronic health records. Relevant implications for clinical practice include improved patient outcomes from early severe sepsis detection and treatment.

OBJECTIVES: Pancreatic cancer is a lethal disease mostly affecting elderly people. Clinical trials on treatment contain disproportionately fewer elderly patients compared with everyday practice. This retrospective study evaluates differences in the rates of chemotherapy delivered and associated survival in different age groups. METHODS: Data were collected from the Cancer Information Resource Files on patients diagnosed with pancreatic cancer from 1993 to 2008. Patients were divided into 3 age groups, namely, A with younger than 50 years, B with 50 to 70 years old, and C with older than 70 years. RESULTS: Complete data were available on 16,694 patients. Forty-four percent were in group C.Chemotherapy was given to 38% of patients in group C versus 69% of patients in group A. A multivariate analysis revealed a similar chemotherapy benefit in all groups, as follows: group C (hazard ratio [HR], 0.51), group A (HR, 0.74), and group B (HR, 0.55). CONCLUSIONS: We found that elderly patients with pancreatic cancer receive treatment less frequently compared with younger patients. However, elderly patients receiving chemotherapy derive similar benefits. Randomized clinical trials are needed to evaluate pancreatic cancer treatment in the elderly patients, particularly given the increasing occurrence of pancreatic cancer in later life.

The Appalachian region of the United States is experiencing a large increase in hepatitis C virus (HCV) infections related to injection drug use (IDU) (1). Syringe services programs (SSPs) providing sufficient access to safe injection equipment can reduce hepatitis C transmission by 56%; combined SSPs and medication-assisted treatment can reduce transmission by 74% (2). However, access to SSPs has been limited in the United States, especially in rural areas and southern and midwestern states (3). This report describes the expansion of SSPs in Kentucky, North Carolina, and West Virginia during 2013-August 1, 2017. State-level data on the number of SSPs, client visits, and services offered were collected by each state through surveys of SSPs and aggregated in a standard format for this report. In 2013, one SSP operated in a free clinic in West Virginia, and SSPs were illegal in Kentucky and North Carolina; by August 2017, SSPs had been legalized in Kentucky and North Carolina, and 53 SSPs operated in the three states. In many cases, SSPs provide integrated services to address hepatitis and human immunodeficiency virus (HIV) infection, overdose, addiction, unintended pregnancy, neonatal abstinence syndrome, and other complications of IDU. Prioritizing development of SSPs with sufficient capacity, particularly in states with counties vulnerable to epidemics of hepatitis and HIV infection related to IDU, can expand access to care for populations at risk.

PURPOSE: The effect of computerized prescriber order entry (CPOE) on the efficiency of medication-order-processing time was evaluated. METHODS: This study was conducted at a 761-bed, tertiary care hospital. A total of 2988 medication orders were collected and analyzed before (n = 1488) and after CPOE implementation (n = 1500). Data analyzed included the time the prescriber ordered the medication, the time the pharmacy received the order, and the time the order was completed by a pharmacist. RESULTS: The mean order-processing time before CPOE implementation was 115 minutes from prescriber composition to pharmacist verification. After CPOE implementation, the mean order-processing time was reduced to 3 minutes (p < 0.0001). The time that an order was received by the pharmacy to the time it was verified by a pharmacist was reduced from 31 minutes before CPOE implementation to 3 minutes after CPOE implementation (p < 0.0001). The implementation of CPOE reduced the order-processing time (from order composition to verification) by 97%. Additionally, pharmacy-specific order-processing time (from order receipt in the pharmacy to pharmacist verification) was reduced by 90%. This reduction in order-processing time improves patient care by shortening the interval between physician prescribing and medication availability and may allow pharmacists to explore opportunities for enhanced clinical activities that will further positively impact patient care. CONCLUSION: CPOE implementation reduced the mean pharmacy order-processing time from composition to verification by 97%. After CPOE implementation, a new medication order was verified as appropriate by a pharmacist in three minutes, on average.

BACKGROUND AND AIMS: Syringe-sharing significantly increases the risk of HIV and viral hepatitis acquisition among people who inject drugs (PWID). Clearer understanding of the correlates of receptive syringe-sharing (RSS) is a critical step in preventing bloodborne infectious disease transmission among PWID in rural communities throughout the United States. This study aimed to measure the prevalence and correlates of RSS among PWID in a rural county in Appalachia. DESIGN: Observational, cross-sectional sample from a capture-recapture parent study. SETTING: Cabell County, West Virginia (WV), USA, June-July 2018. PARTICIPANTS: The sample was restricted to people who reported injecting drugs in the past 6 months (n = 420). A total of 180 participants (43%) reported recent (past 6 months) RSS. Participants reported high levels of homelessness (56.0%), food insecurity (64.8%) and unemployment (66.0%). MEASUREMENTS: The main outcome was recent re-use of syringes that participants knew someone else had used before them. Key explanatory variables of interest, selected from the risk environment framework, included: unemployment, arrest and receipt of sterile syringes from a syringe services program (SSP). Logistic regression was used to determine correlates of recent RSS. FINDINGS: PWID reporting recent RSS also reported higher prevalence of homelessness, food insecurity and unemployment than their non-RSS-engaging counterparts. In adjusted analyses, correlates of RSS included: engagement in transactional sex work [adjusted odds ratio (aOR) = 2.27, 95% confidence interval (CI) = 1.26-4.09], unemployment (aOR = 1.67, 95% CI = 1.03-1.72), number of drug types injected (aOR = 1.33, 95% CI = 1.15-1.53) and injection in a public location (aOR = 2.59, 95% CI = 1.64-4.08). Having accessed sterile syringes at an SSP was protective against RSS (aOR = 0.57, 95% CI = 0.35-0.92). CONCLUSION: The prevalence of receptive syringe-sharing among people who inject drugs in a rural US county appears to be high and comparable to urban-based populations. Receptive syringe-sharing among people who inject drugs in a rural setting appears to be associated with several structural and substance use factors, including unemployment and engaging in public injection drug use. Having recently acquired sterile syringes at a syringe services program appears to be protective against receptive syringe sharing.

Approximately 17,000 new spinal cord injuries occur each year in the United States, with motor vehicle collisions and falls being the most common causes. Even though 94% of patients survive the initial injury and corresponding hospitalization, their life expectancy is reduced secondary to the long-term complications that stem from their injury. Every patient should be approached in the same manner (i.e., as if they have a spinal cord injury) until proven otherwise to prevent additional injury. Important considerations that should take place in the emergency department include the patient's airway, the presence of shock, and the prevention of complications secondary to the primary injury. This article briefly summarizes the epidemiology and pathophysiology of spinal cord injuries and the therapies that may be recommended and initiated following a thorough assessment of the patient.

AIM: Safe consumption spaces (SCS) are indoor environments in which people can use drugs with trained personnel on site to provide overdose reversal and risk reduction services. SCS have been shown to reduce fatal overdoses, decrease public syringe disposal, and reduce public drug consumption. Existing SCS research in the USA has explored acceptability for the hypothetical use of SCS, but primarily among urban populations of people who inject drugs (PWID). Given the disproportionate impact of the opioid crisis in rural communities, this research examines hypothetical SCS acceptability among a rural sample of PWID in West Virginia. METHODS: Data were drawn from a 2018 cross-sectional survey of PWID (n = 373) who reported injection drug use in the previous 6 months and residence in Cabell County, West Virginia. Participants were asked about their hypothetical use of a SCS with responses dichotomized into two groups, likely and unlikely SCS users. Chi-square and t tests were conducted to identify differences between likely and unlikely SCS users across demographic, substance use, and health measures. RESULTS: Survey participants were 59.5% male, 83.4% non-Hispanic White, and 79.1% reported likely hypothetical SCS use. Hypothetical SCS users were significantly (p < .05) more likely to have recently (past 6 months) injected cocaine (38.3% vs. 25.7%), speedball (41.0% vs. 24.3%), and to report preferring drugs containing fentanyl (32.5% vs. 20.3%). Additionally, likely SCS users were significantly more likely to have recently experienced an overdose (46.8% vs. 32.4%), witnessed an overdose (78.3% vs. 60.8%), and received naloxone (51.2% vs. 37.8%). Likely SCS users were less likely to have borrowed a syringe from a friend (34.6% vs. 48.7%). CONCLUSIONS: Rural PWID engaging in high-risk behaviors perceive SCS as an acceptable harm reduction strategy. SCS may be a viable option to reduce overdose fatalities in rural communities.

Cryptococcosis is a fungal disease which has been characterized by its identified risk groups. There are many risk factors identified. We present a surprising four cases of disseminated cryptococcosis in intravenous drug abuse (IVDA) patients in a short period of time and in one geographical area, this observation suggest that there may be a new association with IVDA and cryptococosis.

From January 1, 2018, through October 9, 2019, 82 HIV diagnoses occurred among people who inject drugs (PWID) in Cabell County, West Virginia. Increasing the use of HIV preexposure prophylaxis (PrEP) among PWID was one of the goals of a joint federal, state, and local response to this HIV outbreak. Through partnerships with the local health department, a federally qualified health center, and an academic medical system, we integrated PrEP into medication-assisted treatment, syringe services program, and primary health care settings. During the initial PrEP implementation period (April 18-May 17, 2019), 110 health care providers and administrators received PrEP training, the number of clinics offering PrEP increased from 2 to 15, and PrEP referrals were integrated with partner services, outreach, and testing activities. The number of people on PrEP increased from 15 in the 6 months before PrEP expansion to 127 in the 6 months after PrEP implementation. Lessons learned included the importance of implementing PrEP within existing health care services, integrating PrEP with other HIV prevention response activities, adapting training and material to fit the local context, and customizing care to meet the needs of PWID. The delivery of PrEP to PWID is challenging but complements other HIV prevention interventions. The expansion of PrEP in response to this HIV outbreak in Cabell County provides a framework for expanding PrEP in other outbreak and non-outbreak settings.

Background: Overdose fatality rates in rural areas surpass those in urban areas with the state of West Virginia (WV) reporting the highest drug overdose death rate in 2017. There is a gap in understanding fentanyl preference among rural people who inject drugs (PWID). The aim of this study is to investigate factors associated with fentanyl preference among rural PWID in WV. Methods: This analysis uses data from a PWID population estimation study conducted in Cabell County, WV in June-July 2018. Factors associated with fentanyl preference were assessed using multivariable Poisson regression with a robust variance estimate. Results: Among PWID who reported having ever used fentanyl (n = 311), 43.4% reported preferring drugs containing fentanyl. Participants reported high levels of socioeconomic vulnerability, including homelessness (57.9%) and food insecurity (66.9%). Recent increases in drug use and injecting more than one drug in the past 6 months were reported by 27.0% and 84.2% of participants, respectively. In adjusted analyses, fentanyl preference was associated with being younger (PrR:0.98, 95% CI: 0.97-1.00), being female (PrR:1.45, 95% CI:1.14-1.83), being a Cabell county resident (PrR:0.60, 95% CI: 0.45-0.81), increased drug use in the past 6 months (PrR:1.28, 95% CI: 1.01-1.63), and injecting fentanyl in the past 6 months (PrR:1.89, 95% CI: 1.29-2.75). Conclusion: Fentanyl preference is highly prevalent among rural PWID in WV and associated with factors that may exacerbate overdose risks. There is an urgent need for increased access to tailored harm reduction services that address risks associated with fentanyl preference.