Mental health, physical health, and social support in health profession students: a proof-of-concept study Authors: Edie L. Sperling, Jennifer M. Hulett Published: 2026 ABSTRACT BackgroundResearch is limited as to how physical and mental symptoms may relate to emotional social support in health profession students, who often experience social isolation. The ORBIT Model Phase Ia was used to guide this proof-of-concept study. MethodsSurveys were distributed at two timepoints to preclinical medical, dental, veterinary, and physical therapy students at a health sciences university. Surveys included the Perceived Stress Scale, the Generalized Anxiety Disorder scale, the Patient Health Questionnaire scales for depression and somatization, and the NIH Toolbox: Emotional Social Support subscale. ResultsN = 38 separate individuals completed the timepoints. The outcome measures appeared tolerable and acceptable to the participants at two time-points. Emotional social support was below recommended levels, particularly among Hispanic/Latino students. Social support was significantly negatively correlated with perceived stress (rho = -0.43, p < 0.01), anxiety (rho = -0.34, p < 0.01), and depression (rho = -0.40, p < 0.01) at T1, and depression (rho = -0.43, p < 0.01) and physical symptoms (rho = -0.46, p < 0.01) at T2. ConclusionsResults support proof-of-concept for an association between these variables in health profession students. Further investigation into emotional support among health profession students in a larger sample is warranted. Background Health profession students have elevated stress, anxiety, depression, and physical symptoms while in school [1, 3, 4, 34]. These students commonly report social isolation due to multiple factors including academic workload, relocating for school, and competition with classmates [29]. Social support has been shown to be protective against psychosocial symptoms among healthcare students [14], predicting well-being in veterinary students [12, 31]; lower stress, anxiety, and depression in medical students [5, 14, 35]; and lower depression and better quality of life in dental students [10, 18]. Seeking social support is one of the most common coping methods among health profession students [30], but research is limited on comparisons of social support across disciplines or years in training, as well as its potential association with physical symptoms in this population [34]. However, social support has a well-established association to physical health outcomes through both biophysiological and psychological mediators in non-health profession students [6, 36]. Undergraduate college students, for example, have demonstrated greater severity of physical symptoms associated with lower social support [13]. Emotional social support is a sub-category of social support which encompasses having someone with whom one can share their feelings and struggles and whom will listen and demonstrate caring, understanding, and empathy [8]. Emotional support has been identified as the most important type of social support for students [22]. Students who come from marginalized backgrounds or identities, such as those with genders, sexual orientations, races, or ethnicities which differ from the cultural majority, may be at heightened risk for social isolation and poor mental health [25, 32]. We collected demographic information to assess whether emotional support interacted with mental or physical health in different ways within different demographic groups. Due to the early stage of research on emotional support, mental health, and physical symptoms in health profession students, we designed a Phase Ia study based on the Obesity-Related Behavioral Intervention Trials (ORBIT) Model [9]. The ORBIT Model lays out a progressive approach to obtaining evidence for behavioral trials, with Phase Ia as the first phase [9]. In Phase Ia researchers are guided to identify a hypothesized pathway by which variables interact, for example, how emotional support can influence mental and physical health; assess the tolerability and acceptability of the study design in the population; and determine if there is a significant relationship between variables which can provide an evidentiary basis for larger studies [9]. With our known population of interest health profession students, we enlisted institutional stakeholders in discussion regarding which population subset would be ideal for conducting a pilot study. Students who are in clinical rotations have highly variable schedules and experiences and often work up to 12 h per day along with studying, whereas preclinical students are typically more accessible and have more leisure time. This paper reports findings from our recent ORBIT Model phase Ia study [9] which aimed to establish proof-of-concept regarding a need for studies examining mental and physical health correlates in health professions students, including (1) assessment of the tolerability and acceptability of our selected outcome measures and collecting data from a multi-disciplinary health science students at two timepoints, and (2) exploring potential relationships between social support and mental or physical symptoms in health profession students. Methods Sample The Institutional Review Board (approval #2165953) approved this study. The study was conducted in accordance with the Declaration of Helsinki. All participants provided informed consent to participate and for their anonymous responses to be used in research. Potentially eligible participants at a single academic health science university were recruited through their university email addresses. Participants were eligible if they were enrolled as a first or second-year student in a health professions program, including medical, dental, veterinary, or physical therapy. Because of our Phase 1a approach, a power calculation was not performed [9]. Procedures Inclusion and exclusion criteria First- and second-year students in osteopathic medicine, dental medicine, veterinary medicine, and physical therapy at a single health sciences university with multiple colleges were identified as the target population. Students on a leave of absence, repeating a year in school, or from another college were excluded, as stress levels may have been different. Approximately 1,100 student were eligible; we aimed to recruit n = 10 from each of the four colleges for a total of n = 40. Data collection Students were contacted through university emails, available to the PI who is employed at the sister campus. Data were collected via Qualtrics prior to beginning classes in August (T1) and at the end of the Fall semester in late November/early December (T2). Retention Reminder emails with the survey link were delivered twice more weekly. Participants were entered in a drawing to receive one of four $100 gift cards (one per program). Measures Demographics Demographic data included students’ gender, age, race, ethnicity, and sexual orientation, as well as program of study and enrollment year. Perceived stress scale-14 (PSS-14) The Perceived Stress Scale-14 (PSS-14) scores each item on a 0–4 Likert scale from 0 = never to 4 = very often, with a higher number indicating higher perceived stress on the 0–56 scale [7]. For example, item two asks, “In the last month, how often have you felt unable to control the important things in your life?” [7]. There are two sub-scales, self-efficacy and helplessness [2]. Validity and reliability have been sufficient in previous studies, with Cronbach’s α > 0.70 [19, 23]. The generalized anxiety disorder scale-2 (GAD-2) assesses risk of anxiety, with items anchored from 0 = not at all to 3 = nearly every day [17]. For example, item one asks, “Over the last two weeks, have often have you been bothered by feeling nervous, anxious or on edge?” [17]. The final score is between 0 and 6 with higher scores indicating higher risk of anxiety [17]. A score of 3 or higher is recommended to trigger further screening of anxiety disorders (sensitivity 86%, specificity 83%; [17]). The patient health questionnaire-2 (PHQ-2) screens for depression, with items anchored from 0 = not at all to 3 = nearly every day [16]. For example, item one asks, “Over the last two weeks, how often have you been bothered by little interest or pleasure in doing things?” [16]. The final score ranges from 0 to 6 with higher scores indicative of higher risk of depression [16]. A score of 3 or higher is recommended to trigger further screening of depressive disorders (sensitivity of 82.9%, specificity of 90%; [16]). The patient health questionnaire-15 (PHQ-15) assesses 15 physical symptom such as headaches, trouble sleeping, and intestinal issues, asking “During the past four weeks, how much have you been bothered by any of the following problems?” [16]. Items at anchored from 0 = not bothered at all to 2 = bothered a lot [16]. The final score ranges between 0 and 30, with higher numbers indicating more physical symptoms [16]. Severity cut-offs have been suggested to be 5–9, 10–14, and 15 + for mild, moderate, and high physical symptom severity, respectively [16]. Studies have shown good reliability (Cronbach’s α > 0.80) and criterion validity (sensitivity and specificity > 0.70; [16]). The NIH toolbox: emotional support subscale was used to measure emotional social support [8]. The subscale has eight items, for example, item one asks, “In the past month, please rate how often I have someone who understands my problems” [8]. Respondents score each item between 0 = never to 4 = always [8]. The score is converted to an age-adjusted score and standardized to a mean of 50 and standard deviation of 10 [8]. One standard deviation either above or below the mean of 50 indicates relatively high or low levels of social support respectively, with a score of 40 or below indicating a higher risk of social isolation [33]. The instrument has displayed good reliability (Cronbach’s α > 0.9) and concurrent validity (> 0.70) [8]. Data analysis Aim 1 Proof of concept evaluated with tolerability via recruitment rates and completeness of data at two time points, and acceptability via stakeholder responses and university support. Aim 2 Proof-of-concept evaluated using non-parametric statistics to explore potential relationships between social support and mental and physical symptoms in health profession students. SPSS version 29.0.2 and R Studio were used to analyze data (R Core Team, [28]). Nonparametric tests were used due to small sample size and unbalanced subgroups, including Wilcoxon signed-rank test for related groups to compare means between time-points and Spearman’s correlations for variable associations. Results Participant characteristics N = 38 students completed both time points. The mean age of students was 24.8 (SD = 2.5). See Table 1 for demographics of the matched-data participants. Native American/American Indian and Native Hawaiian/Pacific Islanders were combined for analysis due to small sample size. Table 1Participant demographics Program n % DO2052.6 DVM1128.9 DDM615.8 DPT12.7 Year in school 1st year2976.3 2nd year923.7 Race White1744.7 Asian1744.7 Native Amer*410.6 Ethnicity Hispanic615.8 Non-Hispanic3284.2 Gender Female2668.4 Male1231.6 Sexual Orientation LGBTQIA+1026.3 Heterosexual2873.7*Native Americans/American Indians and Native Hawaiians/Pacific Islanders combined Table 1Participant demographics Program n % DO2052.6 DVM1128.9 DDM615.8 DPT12.7 Year in school 1st year2976.3 2nd year923.7 Race White1744.7 Asian1744.7 Native Amer*410.6 Ethnicity Hispanic615.8 Non-Hispanic3284.2 Gender Female2668.4 Male1231.6 Sexual Orientation LGBTQIA+1026.3 Heterosexual2873.7*Native Americans/American Indians and Native Hawaiians/Pacific Islanders combined Participant demographics *Native Americans/American Indians and Native Hawaiians/Pacific Islanders combined Aim 1: Proof-of-concept: tolerability and acceptability Tolerability appeared high, with survey completion rate of 99%, and n = 38 out of 40 targeted participated filling out both surveys, although groups were unbalanced and more medical students were able to be recruited than other disciplines, particularly physical therapy. Acceptability was satisfactory; stakeholder feedback from the College Deans and Offices of Student Affairs was positive. University support was positive overall, including support with Qualtrics and survey distribution. Aim 2: Proof-of-concept: potential relationships between social support and mental or physical symptoms Means of outcomes at each time-point are presented in Table 2. Emotional support did not significantly change over time, but mean scores were < 40.0 at both timepoints, indicating high risk of social isolation and associated health problems [8]. Perceived stress, anxiety, depression, and physical symptoms all significantly increased between the first and second time-point. Table 2Outcome measure means and changes over three monthsOutcomeT1 Mean (SD)T2 Mean (SD)Wilcoxon signed-rank statistic (W)Score InterpretationEmotional social support37.8 (6.5)38.1 (6.3)-0.53≤ 40.0 at riskPerceived stress23.1 (6.9)29.5 (6.7)4.34***No cut-offAnxiety1.9 (1.6)2.8 (2.0)2.90**≥ 3 more screeningDepression0.7 (1.0)1.6 (1.7)3.17**≥ 3 more screeningPhysical symptoms6.86 (4.44)8.58 (3.98)2.82**Mild symptoms***p ≤ 0.001 **p ≤ 0.01 *p ≤ 0.5 Emotional support did not significantly differ by discipline, year in school, race, gender, or sexual orientation. However, it did differ by ethnicity. Over time, the gap significantly widened between Hispanics, with lower emotional support, and non-Hispanics, with higher emotional support (F(1,32) = 7.07, p = 0.01). Table 2Outcome measure means and changes over three monthsOutcomeT1 Mean (SD)T2 Mean (SD)Wilcoxon signed-rank statistic (W)Score InterpretationEmotional social support37.8 (6.5)38.1 (6.3)-0.53≤ 40.0 at riskPerceived stress23.1 (6.9)29.5 (6.7)4.34***No cut-offAnxiety1.9 (1.6)2.8 (2.0)2.90**≥ 3 more screeningDepression0.7 (1.0)1.6 (1.7)3.17**≥ 3 more screeningPhysical symptoms6.86 (4.44)8.58 (3.98)2.82**Mild symptoms***p ≤ 0.001 **p ≤ 0.01 *p ≤ 0.5 Outcome measure means and changes over three months ***p ≤ 0.001 **p ≤ 0.01 *p ≤ 0.5 Spearman’s correlations are presented in Table 3. Emotional support was significantly negatively correlated with perceived stress, anxiety, and depression at T1, and depression and physical symptoms at T2. Table 3Spearman’s rho correlations of outcomes in August and November (n = 38)(T1) AugustPercd stresseiveAnxietyDepressionPhysical SymptomsEmotional support Perceived Stress0.65**0.54**0.45**-0.43** Anxiety0.61**0.47*-0.34* Depression-0.40** Phys symptoms-0.19(T2) November Perceived Stress0.76**0.59**0.21-0.18 Anxiety0.66**0.28-0.23 Depression0.36*-0.43** Phys symptoms-0.46*****p ≤ 0.001 **p ≤ 0.01 *p ≤ 0.5 Table 3Spearman’s rho correlations of outcomes in August and November (n = 38)(T1) AugustPercd stresseiveAnxietyDepressionPhysical SymptomsEmotional support Perceived Stress0.65**0.54**0.45**-0.43** Anxiety0.61**0.47*-0.34* Depression-0.40** Phys symptoms-0.19(T2) November Perceived Stress0.76**0.59**0.21-0.18 Anxiety0.66**0.28-0.23 Depression0.36*-0.43** Phys symptoms-0.46*****p ≤ 0.001 **p ≤ 0.01 *p ≤ 0.5 Spearman’s rho correlations of outcomes in August and November (n = 38) ***p ≤ 0.001 **p ≤ 0.01 *p ≤ 0.5 Discussion This study established proof-of-concept regarding a need for studies examining mental and physical health correlates in health professions students, including (1) assessment of the tolerability and acceptability of our selected outcome measures and collecting data from a multi-disciplinary health science students at two timepoints, and (2) exploring potential relationships between social support and mental or physical symptoms in health profession students. We established the tolerability and acceptability of recruiting health profession students into a study about emotional support, mental and physical symptoms across two time-points. We established a successful plan for how to recruit students, how to distribute the survey, which outcome measures to use, and how to retain participants. While scaling may require minor modifications to the study plan such as providing more gift cards for retention, the logistics are feasible. We found significantly increased perceived stress, depression, anxiety, and physical symptoms in three months between T1 and T2, and significant correlations between emotional support, perceived stress, anxiety, depression, and physical symptoms at different time points, which supports further research on these variables among health profession students. Several considerations for tolerability and acceptability exist. The PI is employed at a satellite campus of the medical school (not included in the study), which may have positively impacted recruitment of medical students, who had the most participants out of the sample. Higher recruitment of medical students may also simply be due to the medical school being the largest of the four colleges. Physical therapy students were most difficult to recruit, and it is not known why other than the physical therapy cohort was the smallest. Although efforts were made to identify and avoid exam dates, surveying multiple colleges at the same time means there may have been other stressful events happening for one college’s population and not another’s which influenced recruitment. Feedback from the separate colleges and the university was positive, including at the Dean level and Office of Student Affairs. This may not be the case at every university, particularly when mental or physical health symptoms are being queried. The PI’s employee-employer relationship with the university may have positively impacted acceptability of the study. Proof-of-concept was demonstrated regarding a need for studies exploring potential relationships between social support and mental or physical symptoms in health profession students. Participants as a whole reported significantly lower emotional support than the U.S. adult population, and were below the score cut-off of 40.0, indicating high risk of social isolation and subsequent mental and physical health issues [8]. Emotional support did not significantly change over the course of the first semester, despite our postulation that students might seek social support from classmates as the semester went on. Second-year students also did not report significantly different emotional support compared to first-year students, as might be expected with students who have been in a cohort together longer. This may suggest that emotional support from classmates plays a relatively small role in health profession students’ emotional support-seeking, or that there are barriers to seeking emotional support among classmates. Emotional social support was significantly lower among Hispanic students than non-Hispanics at both time-points, although the sample size was very small (n = 6 Hispanic vs. n = 32 non-Hispanic students). Hispanic students may face inequities in accessing graduate health profession schooling and systemic racism within medical institutions which make emotional support even more important [11]. Additionally, with a smaller overall population of Hispanic students in the school (7.4%) compared to White (31.9%) and Asian/Asian American students (53.2%), students who seek emotional support amongst peers from their same ethnic background may be at a disadvantage. [26] Further investigation in a larger sample, and eventually inquiry into targeted emotional support for Hispanic students, if warranted, may be beneficial to their wellbeing and long-term mental and physical health. Students’ increased levels perceived stress, anxiety, and depression from the start of school to the end of the first semester is consistent with the literature and reinforces what is known about mental health struggles in health profession schooling [1, 4, 15, 20, 21, 24, 27]. At T1, emotional support showed significant negative correlations with perceived stress, anxiety, and depression, and depression remained significantly correlated at T2. With these correlations, examining strategies to increase emotional social support may be worth investigating, particularly for preventing or reducing depression. Physical symptoms among health profession students are under-researched, although globally studies have shown significant correlations between mental health and physical symptoms in health profession students [34]. As a significant correlation between social support and physical symptoms existed at T2 but not T1, this may suggest a temporal relationship in which increasing physical symptoms resulted in a more reliance on protective factors, like emotional support, over time. As this provides preliminary evidence of an association between physical symptoms and emotional support in health profession students, further investigation may be warranted. This Phase Ia ORBIT study has several limitations. The small sample size and unbalanced groups preclude the ability to conduct sub-group analyses. Representation of under-resourced populations was limited. Because duration of the study was three months, the full range of what might take place regarding mental and physical health during health profession schooling is not captured. As all outcomes were self-reported, response bias may exist. This study was done at a single health science university, which prevents generalization to other healthcare students or institutions. Future research is indicated on emotional support’s relationship with mental and physical symptoms in health profession students. Seeking support from classmates appeared to be limited in this sample, so future studies with powered sample sizes would be useful to determine if this is maintained across other health sciences programs. Investigation of sources from which students receive emotional support and their relative importance, research into interventional methods to increase emotional support early in the academic year, and evaluating how classmates provide or block emotional support, may all be reasonable areas for future research. A potential temporal relationship between mental health issues preceding physical health issues in health profession students has not been examined and is another avenue which might reveal valuable information for preventing and reducing mental and physical health issues in health profession students. Conclusions The results of this Phase Ia ORBIT study support proof-of-concept with further research warranted into the relationships between emotional support, mental and physical symptoms in health profession students [9]. This study revealed low emotional support and increasing stress, anxiety, depression, and physical symptoms among medical, dental, veterinary, and physical therapy students over the course of one semester, particularly among Hispanic students. Future investigation into where and how health students seek and receive emotional support is indicated, as emotional support did not increase with time spent in a cohort. Including diverse students, finding a clinical signal, and identifying independent and interactive treatment components for future interventional trials may also be beneficial.