Evaluation of chiropractic students’ knowledge and attitudes following pain interventions: A randomized educational trial at 2 institutions

Chronic back pain is 1 of the largest contributors to disability worldwide.¹  Globally, chronic low back pain (LBP) prevalence is calculated to affect 9442.5 per 100,000 individuals.¹  The United States spends over $100 billion per year on management of LBP.²  Approximately 70%–80% of the Western population will develop LBP at least once in their lifetime. Chronic pain is associated with an increase in morbidity and mortality, with 20%–50% of these patients reporting comorbid depression.³  This epidemic of back pain burdens not only individuals, but healthcare systems, economies, and societies.⁴ 

Healthcare providers’ education curricula often focus heavily on the structural or biomechanical factors, rather than broadening the lens to the multifactorial experience for patients.⁵  Chronic pain is a multidimensional neurological disorder underpinned by altered connectivity with maladaptive structural and functional brain neuroplasticity.⁶  The advances in understanding the multidimensional nature of chronic pain have been impacted by the recent development of resting-state functional magnetic resonance imaging (fMRI). It reveals the complex dynamics of functional connectivity in large-scale brain networks altered by the chronicity of the pain experience.⁶  Treatment for chronic pain includes nonpharmacologic options, including manual therapy.⁷  Brain network responses to spine mobilization in patients with chronic LBP have been identified by human brain mapping (fMRI), resulting in intrinsic brain functional connectivity (salience resting-state network) modulation.⁸  Biomedical pain models present tissue damage as the source of nociceptive input, but this is insufficient in the explanation of chronic pain. The complexity of chronic pain hinders the ability of musculoskeletal providers to yield effective treatments.⁹  A provider’s biomedical beliefs and attitudes towards musculoskeletal conditions can negatively affect a patient’s belief of their pain. This can result in a cascade of activity avoidance, unnecessary procedures to treat structural deficits, and negative health outcomes.⁶ 

To provide effective care for patients with chronic pain, healthcare providers need an understanding of pain neuroscience, chronic pain, and patient-centered care to effectively apply the biopsychosocial principles.³  Pain neuroscience education (PNE) offers healthcare providers the opportunity to gain a broader understanding of the pain experience, which can allow them to approach patients with a biopsychosocial lens and offer individualized pain management.^8,10  PNE focuses on educating clinicians about the neuroscience of pain and analyzing psychosocial factors contributing to chronicity. It creates a framework and metaphors to explain the cognitive interpretation of pain, understand why hypersensitivity occurs in the central nervous system, and identify factors which contribute to this process.¹¹  Training, which includes multidisciplinary clinical cases, has been shown to improve knowledge, understanding, and biopsychosocial assessment of the pain narrative.¹² 

In the United States, medical students on average are exposed to 11.13 + 8.23 hours of PNE throughout their training. Other countries, such as Canada, incorporate more than twice this amount of PNE in their medical school curricula (27.56 ± 22.02).¹³  In a 2017 study, Stilwell and Harman addressed the need for chiropractors to engage in PNE to stay informed with current pain research, examine their own beliefs of pain, and provide accurate information and effective care to patients.¹⁴ 

In healthcare training, there has been a call for an immediate incorporation of chronic pain education (CPE) in the curricula. The current knowledge and clinical practices have not been sufficient to obtain reliable and optimal patient outcomes.^12,15,16  A scoping review of 14 studies of healthcare students across multiple disciplines highlighted inadequate knowledge of psychosocial factors contributing to the development and persistence of low back pain.¹⁷  Additional studies suggested that CPE interventions can positively affect students’ attitudes and beliefs regarding patient pain and disability.^15,16  According to the International Association for the Study of Pain (IASP), these interventions should address both cognitive and affective dimensions of pain and include active participation by students.¹²  Mills et al recommended emphasizing a comprehensive biopsychosocial assessment that should include an identification of the pain type.³  Furthermore, the Scottish Intercollegiate Guidelines Network’s (SIGN) guidelines for chronic pain care recommend self-management of pain, including self-help resources, consideration of pain management referral, exercise and advice to stay active, and complementary therapies.^3,18 

Patient-centered care (PCC) is an approach that requires providers to be responsive and respectful of patient preferences, values, and needs.¹⁹  Elements essential to this approach include the patient and the doctor’s individuality, biopsychosocial perspective, shared power and responsibility, and therapeutic alliance.¹⁹  Domains of PCC have been suggested to include cultivating communication, providing respectful and compassionate care, engaging patients in managing their health, and integrating care.¹⁹  Studies suggest PCC is associated with higher patient satisfaction, improved patient outcomes, and lower healthcare costs.^19–21  Yet, in a systematic review, healthcare students from various disciplines had low attitudes towards PCC when measured by the Patient Practitioner Orientation Scale (PPOS).²⁰  It has been strongly recommended for curricula to include practice, role play, and discussion on how to incorporate PCC elements into clinical care.²⁰ 

There is a deficiency in the number of studies regarding these integral concepts of clinical training of chiropractic physicians and students. This study aimed to evaluate the impact of PNE, CPE, and PCC educational interventions on randomized groups of 2 universities’ Year 3 chiropractic students’ knowledge, attitudes, and beliefs on these topics. A secondary aim was to compare measures of these skills between cohorts at different time points throughout their training.

This randomized controlled study included students at Parker University (university A) and Logan University (university B). The study and its materials were approved by the institutional review board of Logan University (IRB#RD0801220589). Informed consent was obtained by all student participants in this study.

This education intervention trial had Year 3 chiropractic students from 2 universities randomized to attend 1 of 3 different 50-minute lectures on the topics of PNE, CPE, or PCC. Each lecture group served as a control for their concurrent educational intervention groups (Table 1). For comparison purposes, outcome measures were collected at university B for Year 1 and Year 2 students who were not exposed to the interventions.

Stratified by location, randomization allocated 281Year 3 chiropractic students into the 3 intervention groups using REDCap’s randomization module.^22,23  Year 3 students at university A were preclinical students, and those from university B were entry-level clinical students. Both chiropractic education programs were 3.5 years in length. Additionally, 160 Year 1 students and 118 Year 2 students also participated.

All Year 3 students at university A were invited by email to attend a lecture for this research study outside of their assigned courses. Interested students were asked to complete a demographic form, after which they were randomized to 1 of the 3 intervention lectures. As a participation incentive, university A’s students were rewarded with a meal and extra credit toward an assignment in 1 of their courses. Year 3 students at university B were invited to participate during a required course in which they had the opportunity to observe an extra lecture. Interested students were then randomly assigned 1 of the 3 lecture rooms and invited by email. The intervention topics and contents were not disclosed to participants prior to the intervention. At both universities, students were asked to fill out the consent form and demographic information through REDCap prior to the start of the lecture.

A 12-week follow-up survey was incentivized by offering a $5 e-gift card to all Year 3 participants. Year 3 students at university B were given time during the course to complete the survey, whereas students at university A completed it outside of their courses. For the comparison groups, Year 1 and 2 students from university B were invited to participate in the survey with the 3 questionnaires during a required course.

All 3 educational interventions were delivered at both chiropractic universities in person at different times. The content of these educational interventions is outlined in Table 2. The PNE and CPE lectures were prepared at university B and presented by faculty from each university who were familiar with the lecture content at each university. The PCC lecture was prepared and delivered at both locations by the same presenter: a faculty clinician at university B.

All participants were asked to complete all 3 questionnaires regardless of which lecture they attended. This was done to compare concurrent attitudes and knowledge regarding PN, CPE, and PCC principles. The 3 questionnaires used in this project have been studied in both clinical^24–26  and educational^15,27,28  settings. The Revised Neurophysiology of Pain Questionnaire (RNPQ) offers a valid and reliable way of evaluating the knowledge of nociceptive pathways and chronic pain.^11,29,30  Catley et al²⁹  found the test-retest reliability of the NPQ to be 0.97 (95% confidence interval [CI]: 0.93–0.99), and internal consistency was Cronbach’s α = 0.84 among chronic spinal pain patients.^29,31  The RNPQ has no established minimal clinically important difference (MCID); however, based on data, the MCID knowledge was set at 0.9 points of 7.3%.^10,11  The study used a RNPQ questionnaire with 12 questions featuring the options of “yes,” “no,” “unknown,” or “choose not to answer.”³¹  Each correct answer was awarded 1 point and the score was translated into a percentage. Higher percentages represented a more accurate understanding of underlying neurophysiological mechanisms influencing pain.

Studies have found that scores from the Health Care Providers’ Pain and Impairment Relationship Scale (HC-PAIRS) are reliable³²  and valid³³  measures of providers’ attitudes and beliefs about the relationship between patients’ pain and disability or impairment.^34,35  This 15-item questionnaire evaluates healthcare providers’ perspectives regarding functional expectations of patients with chronic LBP.³²  The tool uses a 1–7 point scale (1 = completely disagree; 7 = completely agree), where the higher the total score means the stronger the belief that CP validates disability and decreases function.³² 

Data gathered from the Patient-Practitioner Orientation Scale (PPOS) have been shown to be valid and reliable to assess attitudes towards patient-centered care.¹⁹  The self-administered questionnaire consists of 18 items, with each item scored on a scale from 0–6 (“strongly disagree” to “strongly agree”). Higher scores are associated with clinicians exhibiting patient-centered behaviors.³⁶  The questionnaire consists of 2 subscales related to 2 qualities: sharing and caring. The first 9 items make up the sharing subscale, and the last 9 items make up the caring sub-scale. Few studies reported the psychometric properties of factorial validation. A study showed the internal consistency of a 10-item PPOS was acceptable (0.77) but low for individual subscales (0.70–0.55).³⁶ 

Demographic and outcome measures were all calculated using descriptive statistics. Using a significant level of 0.05, independent sample t tests evaluated changes within the groups at each time point. Comparisons between groups were evaluated at each time point using 1-way analysis of variance with a Sidak correction. Analysis was conducted with STATA v14.2 (StataCorp LP 2012, Stata Statistical Software: Release 14.2).

As shown in Table 3, the mean age of Year 3 participants was 26.4 years (SD = 4.20). The mean self-reported GPA was 3.38/4.0 (SD = 0.34), and 52% of the participants were males. Of the participants, 4% had earned a master’s degree and 3% held a doctorate degree. The majority of participants (77%) reported an undergraduate major in a health science-related program or exercise sport science.

Of the 281 Year 3 students invited, 162 (58%) participated in the assigned lectures. A total of 47 students (29%) were randomly assigned to the PNE lecture, 56 (35%) to the CPE lecture, and 59 (36%) to the lecture on PCC. All participating students completed the pre- and post-surveys. The 12-week follow-up survey had 108 (67%) participants. For the comparison groups, 117 of the 160 invited Year 1 students (73%) and 77 of the 118 invited Year 2 (65%) students participated.

As shown in Figures 1–3, all lecture groups had an immediate improvement (pre- to post-intervention) in their specific outcome measure, but only the PNE group did not have a statistically significant improvement (PNE: 3.99/100, p = .18 [95% CI: 10.10 to −1.77]; CPE: 0.42/7, p = .02 [95% CI: 0.72 to 0.07]; PCC: 0.25/6, p = .05 [95% CI: 0.12 to 0.51]). However, the 12-week follow up questionnaire yielded no significance compared to pre-intervention scores for all groups (PNE: −6.25/100, p = .15 [95% CI: 14.93 to −2.42], 7.95; CPE: −0.08/7, p = .19 [95% CI: 0.66 to 0.01]; PCC: −0.14/6, p = .30 [95% CI: 0.41 to −0.13]).

Figures of the cross-sectional data can be found in the online Supplement that accompanies this article. Specifically, these data found that, between the comparison groups, RNPQ scores for Year 1 (44.1, SD:17.54) were significantly different than Year 2 (61.0, SD: 20.30) and Year 3 (66.1, SD: 15.15), but no significant difference was found between Year 2 and Year 3 (p < .09, f = 2.37; Year 1–Year 2, p = .82; Year 1–Year 3, p = .09; Year 2–Year 3, p = .64).

The HC-PAIRS scores also had significant differences between Year 1 (4.5, SD: 0.57) and Year 2 (4.0, SD: 0.73), and then Year 3 (4.1, SD: 0.66) had a significant difference with Year 1, but not Year 2 (p < .05, f = 14.63; Year 1–Year 2, p < .05; Year 1–Year 3, p < .05, Year 2–Year 3, p = .24). The PPOS scores among Year 1 (4.0, SD: 0.45), Year 2 (4.0, SD: .47), and Year 3 (4.0, SD: .60) students were similar with no significant differences (p = .37, f = .69; Year 1–Year 2: p = .99, Year 1–Year 3: p = .92, Year 2–Year 3: p = .81).

This study found that a 50-minute Pain Neuroscience Education (PNE) intervention improved RNPQ scores, but the change was not statistically significant (pre = 66, post = 71). Scores had declined significantly at the 12-week follow-up (60). Studies have found promising outcomes of PNE interventions of different durations (up to 3 hours) with healthcare students in osteopathy, physical therapy, and nursing programs.^10,11,30  This includes 1 study from the Journal of Allied Health that conducted a short 45-minute PNE intervention and found physical therapists had a significant increase in RNPQ scores.¹¹  Another promising study of a 70-minute intervention with physical therapy students yielded an increase in pain knowledge immediately after intervention and persisted 6 months later.³⁰  Although there is evidence of some promising outcomes with 40–75 minute PNE interventions, another study of a 50-minute PNE intervention did not demonstrate improved RNPQ scores.³⁵  Interestingly, our study’s PNE intervention did find an immediate increase in knowledge and beliefs about chronic pain, measured by the HC-PAIRS survey.

The CPE group’s post-intervention HC-PAIRS scores were initially statistically significant but did not sustain significance at the 12-week follow-up. Physiotherapists and physiotherapy students in 2 studies found HC-PAIRS scores improved for those who received PNE versus those who did not.^31,37  These studies suggest the need for additional research in this area, emphasizing the importance of providing CPE.

Interestingly, the CPE group was shown to improve RNPQ scores post-intervention. At the 12-week follow-up, the PNE intervention group’s RNPQ scores had significantly worsened, while the CPE group’s RNPQ scores were maintained. The improvement in CPE comprehension has been noted in other PNE lecture studies.^10,11  This suggests the integration of CPE and PNE may increase RNPQ scores initially and help students to retain long-term information on these topics.

In our study, the patient-centered care (PCC) educational intervention improved chiropractic students' attitudes toward patient-centeredness. This was measured by the PPOS. These findings are consistent with another study that showed improvement in PPOS scores following a course focusing on psychosocial aspects of PCC.³⁸  In our study, these changes persisted, but at the 12-week follow-up, the scores had decreased. The PPOS scores in our study were consistent with the average scores among international chiropractic students (mean = 4.18)²⁷  and among healthcare students in general (mean = 4.16).²⁰  Ironically, PPOS scores decreased among the group of students who received the PNE intervention. This suggests placing a stronger emphasis on basic science concepts, such as pain neurophysiology, may shift the focus away from the patient as a person, and therefore negatively affect attitudes towards PCC.

The majority of cross-sectional data from Year 1 and Year 2 students displayed a linear relationship with increased knowledge and attitudes as students progressed through their training, except for PCC. Year 1 and Year 2 students had similar scores, which may indicate that PCC education is unlikely to improve with progression through the chiropractic program. The limited exposure to patient encounters during these first 2 years of training may contribute to the lower scores on PPOS.

Future studies are needed to evaluate, improve, and standardize the educational content for these clinical topics. While the interventions in our study showed immediate improvement, additional studies should explore the effects of a lecture series, or stronger reinforcement throughout the curriculum, to evaluate whether long-term changes can be sustained or even improved. Studies could evaluate cohorts of students in a longitudinal manner, assessing and monitoring on an interim basis as they progress through the curriculum. More studies could investigate mechanisms underlying the surprising finding of PNE and PCC inverse relationship.

Response rates varied between university A and B due to several factors. While all participants were from a convenience sample and were voluntary, university B invited students who were attending a required course to participate in the study and complete the follow-up surveys during the required class hour, yielding a larger initial participation of students. University A offered the interventions outside of required course times with follow-up surveys sent to students’ personal email addresses for completion on their own time. This resulted in a lower participation rate of 29 (21%) students. Additionally, students at both universities received incentives to participate in the form of $5 e-gift cards at the 12-week follow-up. Students at university A were also offered extra credit towards an assignment in 1 of their courses, while students at university B were not. This incentive was an attempt to increase participation, as the sample size was smaller than anticipated. The discrepancy between the incentives among the 2 universities may have introduced selection bias. Both university A and B generalized their results, and no significant differences were found between the programs; however, it is important to note some differences. First, PNE and CPE lectures were not delivered by the same presenter at both universities A and B. Additionally, the Year 3 students at university A did not have exposure to clinical experience yet, while the university B Year 3 students were starting their clinical experience at the time of the interventions. The difference between the universities’ application and clinical experience pertinent to the development of some competencies could have impacted the study.

The educational interventions were limited to a 1-time, 50-minute lecture. While the most effective length of an intervention remains unknown, it is possible a series of interventions, or stronger reinforcement within the curricula, could result in significant changes in students’ knowledge, attitudes, and beliefs pertaining to these important clinical topics.^10,39,40  Although we encountered a variety of study limitations, we are confident the results of our data may offer clinically useful chiropractic educational principles.

After 50-minute lecture interventions, this study found immediate improvement for the targeted topics, with only PNE not showing statistically significant change. However, at 12 weeks, these changes were not sustained. In comparison to Year 1 and Year 2 students, Year 3 students at baseline were only significantly different from Year 1 and Year 2 in their HC-PAIRS scores. Educational programs have the responsibility of equipping future chiropractors to provide optimal patient care through an evidence-based curriculum that includes training on pain neuroscience, chronic pain, and patient-centered care.^2,5,27  More studies are needed to evaluate the effects of short and long-term educational interventions on chiropractic students’ attitudes, beliefs, and knowledge of PN, CP, and PCC.