Megan Johnson · University of Southern California · PhdFit
Resume-aware faculty matching
Find professors who actually fit you
Upload your resume. PhdFit's six research agents compare your background with faculty profiles, recent publications, lab focus, and outreach opportunities, then rank professors with evidence you can review.
Resume-aware matchingEvidence from recent researchOutreach drafts included
Advisor Match Brief Resume uploaded
Resume signals5 found
NLPHCIFairnessGraph MLPyTorch
Six-agent analysisRunning
Axel · Candidate AnalystResume parsed
Nova · Professor ResearcherRecent papers
Echo · Match ExplainerEvidence linked
Atlas · Strategy AdvisorFit ranked
Lyra · Outreach WriterDraft queued
Juno · Meeting Prep CompilerNext step ready
SC
Dr. Sarah Chen
Stanford · NLP & interpretability
91
fit
Fairness + NLP projects overlap with her recent work.
Strong methods fit: model evaluation and interpretability.
Clear outreach angle for current trustworthy AI research.
Outreach angle
Ask how her lab is extending interpretability methods into fairness audits for real-world AI systems.
Megan Johnson
· Assistant Professor of Emergency Medicine
University of Southern California · Emergency Medicine
Megan Johnson is a Clinical Assistant Professor of Emergency Medicine (Clinician Educator) at the Keck School of Medicine of USC. She is involved in education and clinical practice within the Department of Emergency Medicine. Her role focuses on training future healthcare professionals and providing clinical care in emergency medicine, contributing to the academic and practical advancement of the field at the Keck School of Medicine.
Abstract Introduction Advances in emergency, critical and surgical care have markedly improved survival after major burn injury (≥50% TBSA), yet survivors experience major physical and psychological sequelae. We aimed to characterize long-term physical and mental health outcomes of people with burn injuries ≥50% TBSA to inform patients, families, providers and payors about expected trajectories, rehabilitation needs, functional abilities and common impairments, when compared to patients with moderate sized burn injuries (20-49.9% TBSA). We hypothesized that larger burn sizes (≥50% TBSA) would be associated with greater loss of physical function but similar mental health outcomes compared to those with more moderate burn sizes across all timepoints. Methods Adults with burns ≥20% TBSA were identified from a multicenter longitudinal cohort study and stratified into 20–49.9%, 50–69.9% and ≥ 70% TBSA groups. Data were collected at discharge (pre-injury recall) and 6, 12, 24 and 60 months post-injury. The Short-Form-12 (SF-12), Veterans RAND-12 (VR-12) or PROMIS Global Health-10 were administered at all timepoints and linked with validated bridges to create Global Physical Health (GPH) and Global Mental Health (GMH) scores. Besides GPH and GMH, the outcomes also included SF-12 and VR-12 subdomains to explore more granular trends. Demographic and injury characteristics were summarized and mixed-effects linear regression models were used to evaluate changes in GPH, GMH and subdomain scores over time. Results A total of 1113 participants were analyzed. GPH declined more with larger burns, with significant net differences relative to the 20–49.9% reference group (p<.0001; Fig. 1). GMH did not differ by burn size (Fig. 2). Subdomain analyses showed more declines with larger burns compared to the 20-49.9% group in physical function, bodily pain and role limitations due to physical problems (p<.05). Conversely, the ≥70% TBSA group demonstrated significant recovery in mental health at 24 months (p<.05). Conclusions This analysis confirms lasting deficits in physical function, role limitations and greater bodily pain for patients with ≥50% TBSA burns. In contrast, mental health outcomes were less influenced by burn size, and people living with ≥70% TBSA burns demonstrated significant recovery at 24 months. These persistent deficits support the classification of burns as a chronic condition. Future studies should examine factors driving recovery, map findings to individualized rehabilitation plans, and visually provide prognostic information for patients, families and providers caring for those with major burn injuries. Applicability of Research to Practice Physical and mental health recovery after 6 – 12 months of challenging rehabilitation and adaptation is the common trajectory for people with massive burn injuries. This information can help guide providers, patients and families in prognostication and formulating recommendations. Funding for the study The contents of this abstract were developed under a grant from the National Institute on Disability, Independent Living, and Rehabilitation Research (NIDILRR grant #90DPBU0005). NIDILRR is a Center within the Administration for Community Living (ACL), Department of Health and Human Services (HHS). The contents of this abstract do not necessarily represent the policy of NIDILRR, ACL, HHS, and do not assume endorsement by the Federal Government.
Patients with burn injuries exhibit one of the most intense hypermetabolic responses among critically ill populations, making them highly susceptible to malnutrition-linked to prolonged hospital stays and delayed wound healing. While cannabis is recognized for its appetite-stimulating properties in acute settings, its association with the nutritional demands of burn injuries remains underexplored. A single-institution retrospective study was conducted on adult patients with burn injuries having < 20% TBSA who tested positive for cannabis on admission urine toxicology between 2015 and 2024. These patients were matched 1:1 with controls who tested negative for cannabis. The primary predictor variable was cannabis use, while outcomes included burn characteristics, prealbumin and albumin levels, overall outcomes, and complications. Significance was set at P < .05. We analyzed 76 cannabis-positive patients with burn injuries and 76 controls. No significant differences were found in demographics or outcomes. When controlling for body mass index, cannabis intoxication was not significantly associated with changes in admission prealbumin (18.8 vs 19.2, P = .804) or admission albumin (3.9 vs 4.0, P = .375) levels. There was also no significant variation in the number of days postadmission required to achieve peak prealbumin (3.8 vs 3.9, P = .876) and albumin level (0.3 vs 1.0, P = .088). Increased age was associated with a reduction in admission albumin (P < .001), and Caucasian patients had increased albumin compared to other races (P = .048). Cannabis intoxication had no significant association with preburn injury nutritional status. Further research with larger sample sizes is necessary to fully understand the complex relationship.
Abstract Introduction Nicotine delivery from non-tobacco products (e.g., electronic cigarettes, patches, pouches, gums, lozenges, and oral sprays) is becoming increasingly prevalent. While cigarettes and tobacco products are known risk factors for impaired wound healing, the isolated effects of non-tobacco nicotine dependence (NTND) on post-burn recovery are unclear. The present study sought to assess the impact of NTND on short and long term complications following burn injury. Methods We conducted a retrospective cohort analysis through the United States Collaborative Network on TriNetX, a federated database. Adult burn patients (18+ years) with NTND were compared to burn patients without any nicotine dependence. Patients with a documented history of cigarette smoking or tobacco product usage were excluded. Propensity score matching (1:1) was performed for demographics, body mass index, comorbidities, concurrent substance use disorders, burn characteristics (region of bodily injury, total body surface area affected, and degree of burn), and treatment modalities (non-steroidal anti-inflammatory medications, autografts, and debridements). Primary outcomes at 90 days were wound complications (infection, disruption, and hematoma formation), sepsis, and healthcare utilization (antimicrobial prescriptions and emergency department [ED] visits). Secondary outcomes were mortality and new diagnosis of opioid use disorder within 1 year. Associations were quantified using risk ratios (RRs), with statistical significance set at p&lt;.05. Results An unmatched total of 594 308 burn patients were identified (21 050 with NTND; 573 258 controls). After 1:1 matching, 20 906 patients remained in each cohort. Within 90 days post-burn, the matched NTND patients had significantly higher risk of wound disruption (RR 1.81, p=.004), wound infection (RR 2.65, p&lt;.001), sepsis (RR 1.50, p=.003), antimicrobial prescriptions (RR 1.35, p&lt;.001), ED visits (RR 1.51, p&lt;.001), and mortality (RR 1.20, p=.033) when compared to matched controls. Risk of hematoma formation was higher in the NTND cohort, though not statistically significant (RR 1.54, p=.193). Within 1 year after burn injury, patients with NTND had increased risk of mortality (RR 1.20, p=.002) and opioid use disorder (RR 4.24, p&lt;.001) compared to those without nicotine dependence. Conclusions Non-tobacco nicotine dependence was significantly associated with elevated risk of complications, healthcare utilization, and long-term mortality as well as onset of opioid use disorder after burn injury. Given these findings, nicotine, independent from tobacco, may impair immune function and tissue healing, which could exacerbate burn prognosis. Applicability of Research to Practice Burn care providers are encouraged to screen patients for regular use of non-tobacco nicotine products upon admission and offer cessation counseling, as addressing this risk factor may reduce complications and improve recovery. Funding for the Study N/A.
Abstract Introduction Three philosophical models of disability that have been robustly defended in the ethics literature are the medical model, the social model, and—most recently—the welfarist model (Table 1). Each model has different implications for policy and practice. No empirical ethics study to date has specifically interrogated the degree to which burn survivors agree with the central “thesis” of each philosophical model. Methods The study protocol was approved by our IRB (HS-23-00757). A survey describing the key thesis of each model of disability was administered in person to adult burn survivors at an outpatient clinic for a large regional burn center between July and September of 2025. Descriptive statistics were calculated, and proportions of agreement and disagreement were compared across models using chi-square analysis (alpha = 0.05). Results Fifty-six burn survivors completed the survey (mean age 38.8 ± 15.4 years; mean %TBSA 21.2 ± 22.8). Twenty-two (39.3%) burn survivors self-identified as having a disability. Among the three models of disability, the central thesis of the welfarist model drew the highest proportion of strong agreement (21.4%) compared with the medical (14.3%) and social (10.7%) models (Fig. 1). The overall proportion of agreement across models was not significantly different (p=.39). However, in response to a forced-choice question, 64.3% (36/56) selected the welfarist model, 25.0% (14/56) the social model, and 10.7% (6/56) the medical model. Nearly all respondents (98.2%, 54/55) characterized disability as instrumentally—rather than intrinsically—undesirable; this corroborates a core feature of the welfarist model of disability. Burn survivors who self-identified as having a disability showed significantly stronger agreement with the welfarist model compared to those who did not self-identify as disabled (p=.011). Conclusion Burn survivors at a large regional burn center were most likely to strongly endorse the central thesis of the welfarist model of disability. This tendency was particularly strong when respondents self-identified as having a disability. Applicability to Practice This is the first time burn survivors have been directly queried regarding the central thesis of each of the three philosophical models of disability. Our findings suggest a need to re-examine conventional accounts of disability used in policy, research, and clinical care.
Abstract Introduction Violent or intentional burn injuries represent a severe form of trauma often stemming from interpersonal conflicts and high-risk social circumstances. There is a paucity of literature examining the socioeconomic factors associated with violent burn injuries and how their clinical outcomes differ from non-violent burns. The objective of this study is to identify the sociodemographic factors associated with such injuries and evaluate their association with morbidity and mortality. Methods A retrospective review was conducted for all patients admitted to a large urban burn center between 2012 and 2025. Admission records were screened to identify burns of violent origin. Patients identified as having self-inflicted burn injuries were excluded. Independent t-tests were used for continuous variables and chi-square tests for categorical variables, and regression analyses were performed to assess the relationship between violent burn etiology and clinical outcomes. Results Of the total 4939 included patients, 213 (4.3%) had intentional burns. Patients in the violent burn cohort were younger (35.4 ± 20.1 vs. 38.9 ± 22.8 years, p=.03), more often Black (31.5% vs. 14.3%, p&lt;.01), homeless (40.4% vs. 13.7%, p&lt;.01), single/unmarried (73.2% vs. 55.6%, p&lt;.01), and insured by Medicaid (75.1% vs. 54.2%, p&lt;.01). Drug use was also more common among intentional burn patients (54.5% vs. 31.2%, p&lt;.01). The violent burn cohort had more severe injury. Rates of inhalation injury (13.6% vs. 8.7%, p=.013) and larger affected total body surface area (TBSA) (8.0 ± 7.7 vs. 5.5 ± 15.3, p&lt;.01) were present in this cohort. Multivariate regression analysis showed that violent burns, after controlling for age, sex, TBSA, and inhalation injury, were significantly associated with longer intensive care unit stay (coeff: 3.18, p=.031), more days requiring mechanical ventilation (coeff: 3.45, p=.026), and longer inpatient length of stay (coeff: 2.60, p=.02). Mortality was not significant (coeff: -0.01, p=.324). Conclusions Patients with violent burns represent a vulnerable subgroup with higher social risk factors, greater injury severity, increased critical care needs, and longer inpatient stays. Providing comprehensive social services, including access to social workers, psychologists, and substance use treatment, may help target underlying socioeconomic and psychosocial vulnerabilities and reduce the risk of violent burn injuries. Applicability of Research to Practice Patients with violent burn should be prioritized for social work, mental health, and substance use support to address underlying risks and improve recovery. Funding for the study N/A.
Abstract Introduction Survivors of massive burn injury frequently face enduring psychological and social sequelae including anxiety, depression, impaired role function and difficulty with community integration. We aimed to characterize these long-term mental and social health outcomes and compare recovery trajectories across burn size strata (20-49.9%, 50-69.9%, ≥70%). We hypothesized that people living with larger burn sizes would initially report poorer outcomes compared to those with moderate sized burns, but that these deficits would attenuate over time. Methods Adults with burns ≥20% TBSA were identified from a multicenter longitudinal cohort and stratified into groups of 20-49.9%, 50-69.9% and ≥ 70% TBSA. Data were collected at discharge (pre-injury recall) and 6, 12, and 24 months post-injury. Mental health was assessed using PROMIS Depression and Anxiety instruments; higher scores indicated worse outcomes. While social health and community integration were evaluated using PROMIS Ability to Participate in Social Roles (APSR) and Community Integration Questionnaire (CIQ); higher scores indicated better outcomes. Demographic and injury characteristics were summarized, and mixed-effects linear regression models were applied to examine longitudinal changes in outcomes. Results A total of 816 participants were analyzed. The majority were white (71%), non-Hispanic (78%), and males (76%). Depression and Anxiety scores did not differ significantly at 12 or 24 months compared with 6 months post-injury across all burn size groups and remained slightly above the national population average. APSR scores improved at 12 and 24 months compared to 6 months post-injury in the 20-49.9% TBSA group (Fig. 1; p&lt;.05). Participants with ≥70% TBSA reported the lowest APSR across all post-injury timepoints, although changes were not statistically significant. CIQ scores were significantly lower at all timepoints post-injury compared to pre-injury in both the 20-49.9% and 50-69.9% TBSA groups (p&lt;.05; Fig. 2). Conclusions This analysis details the improving trajectories for mental health, social role participation, and community integration for people living with massive burn injury. Although scores generally improved over time for most burn size groups, people living with burn injury, especially those with the largest injuries, continue to face substantial and enduring challenges. Applicability of Research to Practice Findings highlight the chronic psychosocial burden of burn injury, and the need for long-term, targeted support for social and community reintegration. Importantly, people living with massive burn injury generally demonstrate improvement over time across most domains, offering patients and providers reason for optimism regardless of burn size. Funding for the study The contents of this abstract were developed under a grant from the National Institute on Disability, Independent Living, and Rehabilitation Research (NIDILRR grant #90DPBU0005). NIDILRR is a Center within the Administration for Community Living (ACL), Department of Health and Human Services (HHS). The contents of this abstract do not necessarily represent the policy of NIDILRR, ACL, HHS, and do not assume endorsement by the Federal Government.
Abstract Introduction In 2013, a summary of the 2012 ABA burn quality consensus conference was published. In it, experts recommended universal PTSD and depression (Dep) screening in burn care given their prevalence and negative, long-term impact on outcomes. Twelve years later, significant advancements have been made, including expansion of ABA verification criteria related to mental health screening and referral (2019) and BCQP data fields for PTSD, acute stress, and Dep screening (2024). However, there is a gap in published implementation models or clinical practice guidelines to support burn centers in meeting these requirements. The current study evaluated the adaptation and implementation of tiers 1-2 (see Fig. 1) of the Multi-tier Approach to Psychological Intervention After Traumatic Injury (MAPIT) Program in a psychology-naive burn center in a large metropolitan area safety net hospital. Methods Burn surgeons worked with hospital administration and burn psychology experts to create a FTE for a health psychologist with clinical and research expertise in traumatic injury and burns and program development and implementation. A collective goal was set to create a comprehensive psychological screen and treat approach for inpatient burns that satisfies ABA requirements and is scalable to other units and outpatient settings. Expert consensus based on current evidence base was used to select the MAPIT Program framework and to screen adult inpatient burn admissions for substance abuse (CAGE-AID: 4 yes/no items; urine and serum toxicology, U/S Tox), and post-injury risk for PTSD and Dep (Injured Trauma Survivor Screen: 9 yes/no items). Hospital and burn unit stakeholders were engaged to create the clinical pathway. Acceptability, feasibility, and applicability of the MAPIT Program were assessed. Results Social work administered screens to patients during routine psychosocial assessment. 197 patients were screened; 97% (n = 191) completed the screens. 33.5% (n = 65) scored high-risk for post-injury PTSD, 30.2% (n = 59) scored high-risk for post-injury Dep, and 18% (n = 35) met high-risk for both. 28.4% (n = 55) screened positive for potentially problematic pre-burn substance use on the CAGE-AID. U/S Tox levels were captured in the dataset for 131 patients, 51.9% (n = 68) resulted positive (excluding likely iatrogenic positives) and 44.1% (n = 30) of those also self-reported substance abuse on the CAGE-AID. Conclusions An interdisciplinary team-based approach that leverages hospital stakeholders and existing resources facilitated implementation of MAPIT. Routine, universal mental health screening of admitted burn patients is feasible, acceptable, and applicable to acute burn care. Applicability of Research to Practice A significant minority of burn patients are at high risk for post-injury PTSD, Dep, and substance abuse. Early identification of patients at high risk is essential to guide targeted intervention and optimization of limited mental health resources. Funding for the study This work was supported by the National Institute of Mental Health (1K23MH141296-01). The contents do not necessarily represent the policy of NIMH and endorsement by the Federal Government should not be assumed.
Abstract Introduction In the burn intensive care unit (ICU), pain control is frequently achieved with intravenous opioids (IV). When bolused, these medications have a faster onset but shorter duration of action, which can result in less consistent pain control and greater addictive potential. Our study evaluates whether the early initiation of oral opioids (measured by days until initiation) in the burn intensive care unit (ICU) impacted total narcotic equivalents delivered and important clinical outcomes in early burn care. Methods A retrospective review of adult patients admitted to a single burn center from 2015 to 2024 was conducted. Inclusion criteria were age over 18 and total body surface area burned greater than 20%. Exclusion criteria included inpatient mortality, polytrauma, ICU stay less than 72 hours, admission for dermatologic disease, and significant pre-existing comorbidity. Opioids from all sources delivered in the first 14 days of admission were converted to IV morphine equivalents (IME) to standardize the data for comparison. Outcomes included total IME (oral and IV), total number of days on parenteral opioids, total ICU days, and total ventilator days. Statistical analysis included multivariate linear regression to assess the association between the number of days until enteral opioid administration and these outcomes. Results A total of 192 patients with a mean age of 43.5 ± 14.3 were included in this study. The cohort had a mean ICU length of stay of 25.1 ± 28.3 days and a mean duration of mechanical ventilation of 13.3 ± 24.2 days. On average, oral opioids were initiated 2.9 ± 3.3 days after admission, with 18.8% of patients not receiving oral opioids at all in the first 14 days. The total IME across the first 14 days was 1150.6 ± 1041.0. Multivariate linear regression analysis controlling for TBSA showed that there was a statistically significant association between later oral opioid initiation and increased total IME use (Coeff 91.692, p&lt;.005), longer ICU stays (Coeff 1.256, p&lt;.005), and more ventilator days (Coeff 0.941, p&lt;.005). Conclusions Patients who were started on oral opioids earlier in their treatment of burn injuries showed improved outcomes, including decreased total opioid use, shorter length of ICU stay, and less time on a mechanical ventilator. Although confounding factors like severity of injury and severity of critical illness could impact this relationship, our findings suggest that early initiation of enteral opioids may benefit critically ill burn patients. Applicability of Research to Practice The early initiation of oral opioids in burn patients is associated with improved clinical outcomes. These findings support consideration of early enteral opioid therapy, even in patients anticipated to require prolonged sedation or aggressive pain regimens. Funding for the study N/A.
Advances in burn care have markedly improved survival after major injuries. However, survivors often experience significant physical and psychosocial sequelae. We aimed to characterize long-term health-related quality of life (HRQoL) among burn survivors to inform expected trajectories, rehabilitation needs, and common impairments. Adult burn survivors from a multicenter, longitudinal cohort study were stratified into 20-49.9%, 50-69.9%, and ≥70% total body surface area burn size groups. Patient-reported physical and mental HRQoL and life satisfaction were assessed using validated outcome measures at discharge (pre-injury recall), 6, 12, 24 months, and 5 years post-injury. Standardized summary scores were derived using validated bridges. Mixed-effects linear regression models evaluated longitudinal changes and between-group differences. A total of 1,113 participants were analyzed. All outcomes declined early after injury but improved progressively thereafter. Notably, by 24 months, mental health and life satisfaction approached pre-injury levels across all burn size groups. Physical health deficits were greater with increasing burn size, with significant net differences relative to the 20-49.9% reference group (p<.0001). Mental health and life satisfaction outcomes showed minimal between-group differences overall, though individuals with the largest burns exhibited significantly better relative mental health at 24 months (p<.05). Although outcomes improve over time following major burn injury, persistent physical deficits support the classification of major burns as a chronic condition. These findings characterize burn size-specific recovery trajectories and demonstrate that, despite persistent physical deficits, mental health and life satisfaction can return to near pre-injury levels even after the most extensive injuries.