Hollman, John

Abstract Objective The objective of this study was to compare the lexical sophistication of personal statements submitted by professional physical therapist education program applicants with those generated by OpenAI’s Chat Generative Pretrained Transformer (ChatGPT). Methods Personal statements from 152 applicants and 20 generated by ChatGPT were collected, all in response to a standardized prompt. These statements were coded numerically, then analyzed with recurrence quantification analyses (RQAs). RQA indices including recurrence, determinism, max line, mean line, and entropy were compared with t-tests. A receiver operating characteristic curve analysis was used to examine discriminative validity of RQA indices to distinguish between ChatGPT and human-generated personal statements. Results ChatGPT-generated personal statements exhibited higher recurrence, determinism, mean line, and entropy values than did human-generated personal statements. The strongest discriminator was a 13.04% determinism rate, which differentiated ChatGPT from human-generated writing samples with 70% sensitivity and 91.4% specificity (positive likelihood ratio = 8.14). Personal statements with determinism rates exceeding 13% were 8 times more likely to have been ChatGPT than human generated. Conclusion Although RQA can distinguish artificial intelligence (AI)-generated text from human-generated text, it is not absolute. Thus, AI introduces additional challenges to the authenticity and utility of personal statements. Admissions committees along with organizations providing guidelines in professional physical therapist education program admissions should reevaluate the role of personal statements in applications. Impact As AI-driven chatbots like ChatGPT complicate the evaluation of personal statements, RQA emerges as a potential tool for admissions committees to detect AI-generated statements.

PurposeIncreasing stride‐rate by 5%‐10% has been used for injury prevention and rehabilitation but evidence is limited if an independent program can alter stride‐rate in runners. The objective of this study was to determine whether the preferred stride‐rate of recreational runners can be increased by 5%‐10% following a 6‐week independent training program.MethodsThirty‐eight runners running a minimum of 15 miles/wk with a preferred stride‐rate ≤85 strides/min determined during treadmill testing were randomized into two groups: (a) Experimental group was instructed to increase stride‐rate by 10% using a watch and foot pod for stride‐rate feedback; and (b) Control group was instructed to continue normal running. Compliance was tracked with a training log. One stride being defined as two sequential steps. Preferred running stride‐rate was retested at 6 weeks using the same testing protocol.ResultsThere was no significant difference in baseline preferred running stride‐rates between the experimental or control groups. A significant (P < .001) increase in stride‐rate from 79.9 ± 4.8 to 86.8 ± 5.7 strides/min (8.6% increase) was found in the experimental group. The control group did not significantly change their stride‐rate (Baseline: 80.4 ± 4.2, 6 weeks: 81.3 ± 3.3 strides/min).ConclusionA 6‐week home training program using a watch and foot pod can increase the preferred running stride‐rate by 5%‐10% in runners with a stride‐rate of 85 or less. Clinicians may be able to apply this type of gait retraining in recreational runners when designing injury prevention and rehabilitation programs.

Background: Cervical muscle strength, proposed as a modifiable risk factor in concussions, can be assessed using various methods. The purpose of this study was to compare the reliability and force outputs of 3 methods that use handheld dynamometry (HHD) for assessing cervical muscle strength. Hypothesis: All 3 testing methods are reliable, and force outputs are significantly different between methods. Study Design: Repeated-measures reliability. Level of Evidence: Level 5. Methods: The study used a convenience sample of 30 participants. HHD “make tests” for cervical extension, flexion, and right and left side bending were performed using lying push tests, sitting push tests, and sitting pull tests. A sole examiner performed all tests. Two testing sessions were conducted 1 week apart. Analysis included intraclass correlation coefficients (ICCs), repeated-measures analyses of variance (α = 0.05) with post hoc Bonferroni tests, and minimal detectable change (MDC) calculations. Results: All testing methods were reliable; the lying push test had the greatest point estimate values (ICC, 0.89-0.95). Significant differences in force were found between the 3 testing methods. The MDC was most sensitive for the lying push method. Conclusion: Of the 3 cervical muscle testing methods investigated, the lying position with a push test had the largest ICC according to the point estimate and the most sensitive MDC. Force values between the 3 methods were significantly different, which suggests that consistent testing methods should be used. Clinical Relevance: Results from this study support the clinical use of an HHD “make test” in a lying position for assessing cervical muscle strength. The test is reliable and more sensitive to change compared with tests in a seated position.

Background and Purpose/Objective: Few studies have examined the efficacy of fall risk assessment tools for patients undergoing hematopoietic cell transplant (HCT). The purpose of this study was to determine whether the Hendrich II Fall Risk Model (HFRM II) can distinguish patients who fell from those who did not fall in the inpatient HCT population. Methods: Observational study. Design: Retrospective chart review. Setting: Academic medical center. Patients: Patients undergoing HCT who fell during the first 100 days posttransplant. Intervention: None. Measurements: Independent-sample t tests were performed to determine differences in HFRM II score, age, pulse, systolic and diastolic blood pressure, platelet count, hemoglobin concentration, and hematocrit for fallers compared with nonfallers. Chi-square tests of independence determined differences regarding sex, assistive device used on the day of fall, transplant type, and physical therapy. Logistic regression analysis detected odds of falling dependent on transplant type, pulse rate, and HFRM II score. Results: Forty-seven patients who fell within 100 days after HCT were control matched by transplant date with 2 nonfallers each. Significant differences included HFRM II mean score (5.36 for fallers vs 3.41 for nonfallers; t = 4.510, P < .001) and mean pulse rates (100.09 vs 88.70 bpm; t = 3.398, P = .001). Patients with allogeneic HCT were 3.46 times more likely to have fallen than those with autologous HCT (P = .007). Limitations: Potential selection bias of retrospective review. Conclusions: HCT patients with higher HFRM II scores, greater pulse rate, or allogeneic transplant were more likely to fall.

Background: External rotation (ER) strengthening of the shoulder is an integral component of rehabilitative and preventative programs for overhead athletes. A variety of shoulder ER strengthening exercises are reported, including those intended to integrate the core musculature. The purpose of this study was to examine ER torque and electromyographic (EMG) activation of shoulder and trunk muscles while performing resisted isometric shoulder ER in 3 positions (standing, side lying, and side plank). Hypothesis: Significantly greater force and shoulder muscle activation would be generated while side lying given the inherent stability of the position, and greater trunk muscle activation would be generated in the less stable plank position. Study Design: Quasi-experimental repeated-measures study. Level of Evidence: Level 5. Methods: A convenience sample of 25 healthy overhead recreational athletes (9 men, 16 women) participated in this study. EMG electrodes were placed on the infraspinatus, posterior deltoid, middle trapezius, multifidi, internal obliques, and external obliques. EMG signals were normalized to a maximal isometric contraction. Participants performed resisted isometric ER in standing, side-lying, and side plank positions. Results were analyzed using a repeated-measures analysis of variance with post hoc Bonferroni corrections (α = 0.05). Results: There was no significant difference in ER torque between positions (α = 0.05). A significant difference in EMG activity of shoulder and trunk musculature between positions was found in 7 of the 8 muscles monitored. Significantly greater EMG activity in the infraspinatus, middle trapezius, and the nondominant external and internal obliques was found in the side plank position as compared with standing and side lying. Conclusion: While there was no difference in ER torque between the 3 exercise positions, EMG activity of the shoulder and trunk muscles was dependent on body position. Clinical Relevance: If a clinician is seeking to integrate trunk muscle activation while performing shoulder ER strengthening, the side plank position is preferred as compared with standing or side lying.