About

Jim Harris is a Senior Lecturer in the Department of History at Ohio State University, where he received his Ph.D. in 2017. His research focuses on modern European history, with a particular emphasis on Great Britain, as well as the history of science, medicine, and the environment. Harris has authored the book 'Children’s Health and Urban Ecology in England, 1885–1919,' which explores the localized history of urban public health campaigns in major English cities such as Birmingham, Liverpool, and Manchester. He has also published articles on the history of the 1918 flu pandemic in Great Britain, British anthropologist Sir Arthur Keith, and pedagogy related to teaching the history of public health. Harris co-designed the interdisciplinary course 'Pandemic Pedagogies: An Interdisciplinary Model for Teaching about the Human Impacts of Pandemics,' funded by the 2020 Global Arts and Humanities COVID-19 Special Grants Initiative. At Ohio State, he regularly offers courses on modern European history, global environmental history, and the history of public health, disease, and medicine. Additionally, he teaches interdisciplinary courses on vaccines and the climate crisis, collaborating with colleagues from pharmacy, earth science, and ecology departments.

Research topics

• Medicine
• Surgery
• Physical therapy
• Orthodontics
• Anatomy

Selected publications

• College Generation Status and Financial Socialization: Implications for Financial Knowledge, Self-Efficacy, and Behavior of First-Generation College Students

  Journal of Family and Economic Issues · 2026-01-24

  articleSenior author
  PublisherDOI

• Deep Gluteal Space Anatomy, Pathology, Clinical Presentation, and Management

  Clinics in Sports Medicine · 2026-02-01

  article1st authorCorresponding
  PublisherDOI

• Review GIDE -- Restaurant Review Gastrointestinal Illness Detection and Extraction with Large Language Models

  ArXiv.org · 2025-03-12

  preprintOpen access

  Foodborne gastrointestinal (GI) illness is a common cause of ill health in the UK. However, many cases do not interact with the healthcare system, posing significant challenges for traditional surveillance methods. The growth of publicly available online restaurant reviews and advancements in large language models (LLMs) present potential opportunities to extend disease surveillance by identifying public reports of GI illness. In this study, we introduce a novel annotation schema, developed with experts in GI illness, applied to the Yelp Open Dataset of reviews. Our annotations extend beyond binary disease detection, to include detailed extraction of information on symptoms and foods. We evaluate the performance of open-weight LLMs across these three tasks: GI illness detection, symptom extraction, and food extraction. We compare this performance to RoBERTa-based classification models fine-tuned specifically for these tasks. Our results show that using prompt-based approaches, LLMs achieve micro-F1 scores of over 90% for all three of our tasks. Using prompting alone, we achieve micro-F1 scores that exceed those of smaller fine-tuned models. We further demonstrate the robustness of LLMs in GI illness detection across three bias-focused experiments. Our results suggest that publicly available review text and LLMs offer substantial potential for public health surveillance of GI illness by enabling highly effective extraction of key information. While LLMs appear to exhibit minimal bias in processing, the inherent limitations of restaurant review data highlight the need for cautious interpretation of results.

  PublisherOA PDFDOI

• The Trend to Use the Word “Trend” to Describe Nonsignificant Results in Orthopaedic Literature

  JBJS Open Access · 2025-04-01 · 1 citations

  reviewOpen access

  Introduction: Use of the word “trend” for statistical judgment with no statistical significance has been recognized in multiple fields of medical literature. “Trend” is sometimes used to describe “nearly significant” findings when they are not statistically significant. The purpose of this study was to characterize the use of the word “trend” to describe nonsignificant findings in orthopaedic publications. Methods: Orthopaedic journals indexed in MEDLINE were identified using the terms: orthopedics OR orthopaedics OR sports medicine OR orthopedic OR orthopaedic. Abstracts from these journals that contained the word “trend” from 1985 to 2022 were identified. Journals containing literature on nonhuman orthopaedics were excluded. “Trend” used for statistical judgment without a p-value or with a nonsignificant p-value was labeled “NS Trend”. Use of “trend” not in the context of statistical tests was labeled as “dictionary usage” of the word. Results: A total of 59,839 article abstracts were identified, with “trend” being used in 1,029 abstracts (1.7%). “NS Trend” was used in 617 abstracts (60%) to describe significant results when the p-value was not given or not actually significant. There was a strong correlation over time with increasing use of the word “trend” in orthopaedic abstracts (r = 0.956). An exponential rise in “NS Trend” usage was found (r = 0.924; R 2 = 0.853; p < 0.001). An exponential fit was confirmed using a Kolmogorov-Smirnov test (p = 0.283). Conclusion: The word “trend” is increasingly being used in orthopaedic literature, including suggesting statistical significance when absent. Misinterpretation of statistical analyses can lead to research conclusions that push readers toward inappropriate treatments. It is imperative that we clearly communicate our statistical findings and not use ambiguous language. The word “trend” should only be used to describe an actual trend analysis and not to imply statistical significance when none is present.

  PublisherDOI

• OP2.1 MRI has a low sensitivity and specificity for accurate diagnosis of labral tear for patients undergoing HA+PAO

  Journal of Hip Preservation Surgery · 2025-03-01

  articleOpen accessSenior author

  Abstract Introduction: MRI interpretation has been utilized by surgeons and payors to determine the necessity for labral repair. Some PAO surgeons have stated that hip arthroscopy (HA) is unnecessary for PAO patients and others have reported that approximately 10% of patients undergoing PAO would require HA for labral fixation based on the MRI. Purpose was to determine the sensitivity and specificity of MRI for labral tear in patients undergoing HA+PAO. Hypothesis was that MRI would show low sensitivity and specificity for labral tear decision-making. Methods: IRB approval was obtained. Inclusion all patients undergoing HA+PAO with MRI radiologist interpretation undergoing primary HA + PAO. Exclusion any revision patients. MRIs were from a referral base encompassing a four-state area and read by MSK radiologist. Time from MRI to surgery in months was recorded. Presence of a tear was defined arthroscopically, and treatment was based on surgeons’ discretion at time of HA. HA + PAO technique was post-less and previously published. Power analysis performed 90 patients were needed to detect a significance P<0.05 for tear. Welch’s t-test was used for non-categorical and Fisher’s exact for categorical data. Linear regression model utilized for factors associated with identifying labral tear. Results: 167 patients met inclusion, 91% female. LCEA 19± 3, all Tonnis 0-1. Mean age 22.8 yrs, and time to surgery 8.3 months from MRI. 148 had definitive MRI reads of tear 98 (56%) or no tear 50 (28.6%), and 19 (10.8%) of patients were read as indeterminate. At time of HA of the positive tear MRI 68 (69.4 %) had a tear and 43 (63.3%) were repaired. In the negative MRI group for labral tear, 33 (66%) had a tear and 19 (58%) were repaired. The sensitivity of MRI for labral tear was 67.3% and specificity 36.2%. A subgroup of 39 patients had definitive MRA reads the SN of 70 and SP of 75. MRA nor plain MRI were predictive of labral tear. Regression analysis showed no association with MRI/MRA/radiologist/time between scan and surgeon to identify a labral tear. Conclusion: MRI and MRA has low sensitivity and specificity for labral tear for patients with DDH undergoing HA+PAO. There is no predictive was to determine presence of tear on imaging. MRI alone should not be considered reliable for sole decision making for the presence of a tear nor operative decisions to repair or not repair the labrum at the time of PAO.

  PublisherDOI

• EP5.11 Utility of the Stryker hip Map does not change the clinical decision for detecting cam

  Journal of Hip Preservation Surgery · 2025-03-01

  articleOpen accessSenior author

  Abstract Introduction: The optimal radiographic view in detection of cam deformity has been shown to be the 45-degree Dunn. Recently mapping software has been utilized to produce 3D models and report calculated measures including alpha angle. There is no study comparing the alpha angle measurements of the Stryker Hip Map 3D modeling tools to alpha angle measurements obtained with simple radiographs. Purpose to compare the alpha angle on x-ray to the Stryker Hip Map software. Hypothesis Stryker hip map would be more sensitive for cam lesion. Study Design: IRB approval. Retrospective review of registry data. Inclusion patients who underwent primary hip arthroscopy due to FAI syndrome with Stryker Hip Map CT and XR scan with modified Dunn. Exclusion imaging of poor quality, revisions, PAO. 2 reviewers, and alpha angles calculated by standard method using 45-degree modified Dunn view. r-values calculated for intra/inter-observer reliability. Measurements compared to values from the Stryker Hip Map \at 12, 1, 3, and 3 o’clock. Alpha defined on x-ray as 55 deg or >. t-test utilized for continuous variables and McNemar’s Chi-Square for paired samples. Linear regression for prediction of alpha. Significance p < 0.50. Minimum power of 0.80 sample size of 35 patient data. Results: 72 met inclusion. Mean patient age 27.8 5. Procedures labral repair, femoroplasty and capsular closure. Intra/inter observer reliability 0.89/0.9 respectively on X-ray. R-values correlating x-ray to CT were 0.56 and 0.54 for 12 and 1 o’clock. Mean alpha in the Dunn group was 60.5, and 49.6, 59, 50, and 55 at the 12, 1, 2, and 3 o’clock positions in the CT group. Alpha measures were significantly different between hip map and x-ray at all measured CT clock faces, p<0.01. Cam sensitivity and specificity were 100 and 89 at all clock faces for CT and 100 and 100 for x-ray alone. Cam lesion was detected by both models when present. Conclusions: X-ray and CT show high specificity and sensitivity for detecting cam. Dunn view showed a moderate correlation for the 12 and 1 o’clock positions as measured on CT. Plain radiographs adequately report alpha and under report cam lesion at the 2 and 3 o’clock positions as they are not in plane for this view. Stryker hip map does not appear change the clinical decision making for identifying cam lesion to perform femoral neck osteochondroplasty. Hip map appears to more readily identify the extent of the cam vs plain xray.

  PublisherDOI

• Healthy LLMs? Benchmarking LLM Knowledge of UK Government Public Health Information

  ArXiv.org · 2025-05-09

  preprintOpen access1st authorCorresponding

  As Large Language Models (LLMs) become widely accessible, a detailed understanding of their knowledge within specific domains becomes necessary for successful real world use. This is particularly critical in the domains of medicine and public health, where failure to retrieve relevant, accurate, and current information could significantly impact UK residents. However, while there are a number of LLM benchmarks in the medical domain, currently little is known about LLM knowledge within the field of public health. To address this issue, this paper introduces a new benchmark, PubHealthBench, with over 8000 questions for evaluating LLMs' Multiple Choice Question Answering (MCQA) and free form responses to public health queries. To create PubHealthBench we extract free text from 687 current UK government guidance documents and implement an automated pipeline for generating MCQA samples. Assessing 24 LLMs on PubHealthBench we find the latest proprietary LLMs (GPT-4.5, GPT-4.1 and o1) have a high degree of knowledge, achieving >90% accuracy in the MCQA setup, and outperform humans with cursory search engine use. However, in the free form setup we see lower performance with no model scoring >75%. Therefore, while there are promising signs that state of the art (SOTA) LLMs are an increasingly accurate source of public health information, additional safeguards or tools may still be needed when providing free form responses.

  PublisherOA PDFDOI

• <b>Podium Presentation Title:</b> Functional Outcomes and Return to Sport for Borderline Dysplasia Patients: Total Hip Arthroscopy vs. Hip Arthroscopy

  Arthroscopy The Journal of Arthroscopic and Related Surgery · 2025-01-21

  articleOpen access

  Borderline dysplasia studies have shown good to excellent treatment results with HA in a young adult population. The optimal treatment for this pathology has not been established for those older with a non-arthritic hip, and some hip surgeons may use age as a reason to perform THA over HA. The purpose of this study was to compare a non-arthritic BD population of HA vs THA. Hypothesis is that THA patients would have less pain interference and HA patients would return to sport at a higher rate at 1 and 2 years.

  PublisherOA PDFDOI

• EP40 Self-Rated Pain Tolerance in Hip Arthroscopy Patients – If Someone Tells You They Have a High Pain Tolerance, They Probably Don’t

  Journal of Hip Preservation Surgery · 2025-12-01

  articleOpen accessSenior author

  Abstract Introduction The effective management of postoperative pain is an ongoing focus among orthopedic surgeons. Previous studies have attempted to determine if self-rated pain tolerance correlates with true pain tolerance. However, limited data exists regarding subjective pain tolerance levels in hip arthroscopy patients. Anecdotally, many patients report that they “have a high pain tolerance”. Though objectives of pain tolerance should be distributed normally across the population, we hypothesize that the Dunning-Kruger effect will cause patients to overestimate their own pain tolerance. Purpose The purpose of this study is to quantify the self-rated pain tolerance level among hip arthroscopy patients. The secondary objective of this study is to determine if differences exist in self-rated pain tolerance between different sexes or age groups. Methods A retrospective case series of all patients undergoing primary hip arthroscopy by a single surgeon beginning April 1, 2016 through to April 9, 2024. Data was collected prospectively and analyzed retrospectively for each patient, including demographic information, type of femoroacetabular impingement morphology, self-rated pain tolerance, and prior opioid use. The percentage of patients that reported themselves as having a “high” pain tolerance was calculated. The data was then further stratified by; age group, sex, prior opioid use, and allergies. Linear regression analysis was performed to determine if significant differences existed between groups. Results A total of 1,388 patients underwent hip arthroscopy during the eligibility period. Two-hundred and five subjects were included (139 female, 66 male). Most patients (87.8%) self-reported a high pain tolerance. A greater percentage of men (94%) reported high pain tolerance, compared to 85% of women. The greatest percentage of high self-rated pain tolerance among age groups was within the 30-39 age group (96%). Conversely, the lowest percentage was in patients &amp;lt;20 years (75%). 89% of patients with prior opioid use endorsed high pain tolerance, and 87.4% of patients with allergies endorsed high pain tolerance. None of the differences between any of the above groups were statistically significant. Conclusion Most (88%) hip arthroscopy patients self-report a high pain tolerance, which is a statistical improbability. Therefore, self-rated pain assessments may not be reliable as a true indicator of pain tolerance.

  PublisherOA PDFDOI

• EP169 The Impact of Pain Tolerance and Grit on Patients with FAI Syndrome – A Prospective Study

  Journal of Hip Preservation Surgery · 2025-12-01

  articleOpen access

  Abstract Purpose To prospectively determine the relationship between preoperative measures of grit and pain tolerance and hip-specific symptoms in patients with Femoroacetabular Impingement Syndrome (FAIS). Methods This is a prospective comparative investigation of individuals with FAIS undergoing primary hip arthroscopy (n=59; 37.2 ± 9.9 years; 56% female; labral preservation, cam/pincer/sub-spine correction, capsular closure) with a single surgeon fellowship-trained in hip preservation and eligible adult healthy volunteers (HV; control n=48). Subjects underwent a cold pressor test (CPT; hands and wrists submerged in 1.1° Celsius water) to objectively determine pain tolerance and completed self-reported surveys to determine grit, pain perception, catastrophizing, kinesiophobia, and other psychosocial behaviors and attitudes before surgery. Subjects were dichotomized into two groups: Group 1 (CPT 3 minutes), Group 2 (CPT &amp;lt;3 minutes). Descriptive and comparative statistics were performed. Results Forty-six surgical patients self-reported high pain tolerance (46/59=78%) versus 30 HV (30/48=63%; p=0.079). Surgical patients demonstrated significantly higher preoperative kinesiophobia, fear avoidance, and catastrophizing than HV but similar demographics, self-reported pain tolerance, pain during CPT, and CPT time. Twenty-four (50%) HV completed the 3-minute CPT. Only 27 patients (46%; 12 females) completed the 3-minute CPT. The other 32 patients (54%, 21 females) had a mean CPT time of 49.3 ± 32.7 seconds. Twenty-two patients (22/59=37%) had CPT less than 60 seconds; 17 of those 22 (77%) had self-reported high pain tolerance; 13 of those 17 (76%) were female. Nine patients (9/59=15%) had CPT less than 30 seconds; seven of those nine (78%) had self-reported high pain tolerance; six of those seven (86%) were female. Conclusions Nearly four out of five patients (78%) who underwent arthroscopic hip preservation surgery for FAIS self-reported high pain tolerance, despite this being statistically improbable (Lake Wobegon and Dunning-Kruger effects). However, 54% failed to successfully complete three minutes on the CPT. Interestingly, in patients with CPT less than 60 and 30 seconds (objectively low pain tolerance), most of them self-reported a high pain tolerance (77% and 78%, respectively). Clearly, pain perception is highly subjective, prone to overestimation in hip preservation patients, and requires careful interpretation by surgeons.

  PublisherOA PDFDOI

Frequent coauthors

• Patrick C. McCulloch

  Methodist Sports Medicine

  350 shared

• Kyle R. Sochacki

  Stanford University

  210 shared

• Shane J. Nho

  Rush University Medical Center

  144 shared

• David Dong

  141 shared

• L. E. Peterson

  Rice University

  119 shared

• Robert A. Jack

  112 shared

• David M. Lintner

  Methodist Sports Medicine

  103 shared

• Geoffrey D. Abrams

  Stanford University

  91 shared

Education

• Fellowship, Orthopedic Surgery, Sports Medicine

  Rush University Medical Center

  2013

• Residency, Orthopedic Surgery

  The Ohio State University Wexner Medical Center

  2012