Knowledge of and adherence to radiographic guidelines for low back pain: a survey of chiropractors in Newfoundland and Labrador, Canada

Lumbar radiography plays an important role in the management of low back pain (LBP) when pathology (tumor, infection, or inflammatory arthropathy) or trauma (fracture or dislocation) are suspected [1]. However, these LBP causes are extremely rare, comprising approximately 5% of cases [2]. Strong evidence indicates that routine imaging of the spine does not improve patient outcomes, increases exposure to unnecessary harms, and increases costs [3, 4]. Accordingly, guidelines in both the fields of medicine [5] and chiropractic [6] recommend delaying imaging, in the absence of red flags, for 4 to 6 weeks. If there is no response to conservative care after this period, then imaging may be indicated [3, 6]. Despite this, a large proportion of clinicians request lumbar imaging outside of these guideline recommendations [7, 8].

In the field of chiropractic, several survey-based studies have found lumbar spine radiography utilization rates vary widely worldwide from 25 to 93% [9‐14]. In Canada, a 1997 study found utilization rates of 63–68% in an Ontario community with a large proportion for reasons outside the recommendations of clinical guidelines [15], self-reported utilization rates by interviewed chiropractors in Ontario and Quebec were 10–50% in 2010 [16], while rates of lumbar spine radiographs funded by the Ontario Health Insurance Plan and Workplace Safety and Insurance Board in 2000/2001 were estimated at 3.25 and 3.30% respectively [17]. Unsupported reasons for ordering lumbar films include: to screen for congenital abnormalities and contraindications to spinal manipulative therapy [13], biomechanical/postural analysis of the spine and to educate the patient [15], and medicolegal reasons [16].

Current utilization rate of lumbar radiography for LBP by chiropractors in the province of Newfoundland and Labrador (NL) is unknown. From informal data tracked by the Newfoundland and Labrador Chiropractic Board, it is estimated that the rate could be as high as 36% (based on data from the Avalon Peninsula for 2015 assuming 2 new patients/week). The awareness of current radiographic guidelines, the uptake of those guidelines, and beliefs towards plain film radiography for patients with LBP has not been previously studied in this clinician population. Therefore, the purpose of this study was to survey chiropractors to determine the level of knowledge, adherence to, and beliefs about, clinical practice guidelines related to the use of lumbar radiography for LBP in NL, Canada. A secondary objective was to estimate level of adherence to radiographic guidelines using clinical vignette responses (i.e., behaviour simulation), and to compare our findings to published estimates by Walker et al. (2011) [18] in Australia.

Chiropractors in NL have comparable levels of knowledge of radiographic guidelines to a previously published study in Australia [12]; with approximately 50% of the participants being aware of at least one guideline for lumbar spine radiography. Twenty-five percent of participants indicated that they do not use guidelines to inform clinical decisions and 4% indicated that they were relying on information presented to them when in chiropractic college. Together, these results suggest that interventions aimed at improving awareness and uptake of clinical guidelines, and ultimately improve quality of care, may be warranted.

The majority of beliefs held by participants regarding the use of lumbar spine x-rays for LBP suggest that chiropractors in NL are generally well informed about LBP and its management. The results of this section of the survey are consistent with those from the original study conducted by Jenkins et al. (2016) [12], with a few notable exceptions. These included a lower proportion of respondents indicating a role for x-rays for LBP when neurological signs are present, for routine x-rays prior to initiating SMT, in acute LBP, in patients with sciatica, or to confirm diagnosis and to direct treatment. However, a higher proportion of respondents agreed that they would refer LBP patients for x-ray because patients expect them to. These differences may be due to the fact that chiropractors in NL are mandated to complete continuing education in the area of radiology each year in order to maintain licensure. However, given that Jenkins et al. (2016) [12] demonstrated that chiropractors taking their own x-rays, who are practicing techniques other than diversified, or are unaware of current radiographic guidelines were associated with poorer adherence, it may be due to the fact that this population does not have access to in-house radiographic facilities and mainly (87%) practice diversified.

Despite larger proportions demonstrating beliefs that are generally in line with current recommendations, not all practitioners are clear on the role of lumbar spine x-rays for sciatica, chronic LBP, in the general management of LBP, and for the determination of spinal manipulative therapy contraindications. Therefore, these topics do need to be revisited more frequently throughout the mandated continuing education program. For example, the survey identified a few areas for further consideration. First, 21% of participants agreed that there is a role for the use of lumbar spine x-rays in the evaluation of patients with acute LBP (< 1-month duration) in the absence of red flags for serious disease, despite strong evidence recommending against imaging in this case. Secondly, 13% believed that full spine x-rays (other than for patients with scoliosis) have a role to play in chiropractic practice when the diagnostic value from full spine x-rays is known to be poor. Finally, while the majority of practitioners indicated that they were not likely to refer LBP patients for x-ray because patients often expect them to do so (87%, 45/52), 4% of the population indicated that patient expectations do play a role in their decision making. These results suggest there is room for improvement to enhance patient care. Evidence based approaches to address these concerns include the use of shared decision aids [30], aimed at both chiropractors and their patients, and clinician decision supports, such as modified referral forms allowing for guideline-appropriate indications for imaging and targeted reminders such as short educational messages promoting correct imaging practices [31].

While not a perfect representation of actual clinical practice [32], the responses to the vignettes suggest that the adherence to radiographic guidelines is fairly high (75% for not ordering an x-ray when it is not indicated by guidelines and 91% for ordering an x-ray when it is indicated by guidelines) for many cases typical of practice, and that these levels are significantly greater than those reported previously by Walker et al. (2011) [18]. This does suggest that there is room to improve the adherence to guidelines for the reduction of imaging where it is not recommended (low-value care). These results also suggest that there does not appear to be an underuse of imaging when it would be necessary, which also could be a problem, as discussed by Jenkins et al. (2018) [7]. Finally, the increased proportion of respondents indicating they would order imaging in Vignette 3, which included contextual factors of a busy clinic and a frustrated/unhappy patient, suggests that patient expectations may play a role in decision making. This is not unexpected and is consistent with the literature [33], however, it is contradictory to the result of the belief statement in this study (87% indicated patient expectations do not play a role) suggesting that further education and training on how to handle these situations is warranted.

The strong disagreement regarding the belief that full spine x-rays play a role in the management of LBP (excluding scoliosis) was reflected in lower proportions of participants in this study recommending full spine views compared to the results of Walker et al. (2011) [18]. There may be a few reasons for the difference in adherence. First, this study was conducted approximately 10 years after the study by Walker et al. [18], and beliefs about the role of x-rays for the management of LBP may have changed since then. Second, in Newfoundland and Labrador all radiographs are taken in public healthcare facilities (i.e. hospitals) and true “full spine” (14″× 51″ cassette) are not taken. When full spine requests are ordered, separate cervical, thoracic and lumbar series are done instead which may be different enough from the single view that practitioners do not order them as frequently. Finally, we are comparing populations from different countries. However, we would expect that since international guidelines around LBP and imaging are similar enough, we should see similar behaviours in these groups, thus justifying the comparison of these populations.

There was a large degree of uniformity across responses particularly in the beliefs and clinical vignettes sections. This uniformity may be due to the practitioner population being in practice for less than 24 years (12 years or less (40%) and between 13 and 24 years (44%), the high proportions of the population primarily using Diversified Technique (86%), a non-x-ray driven chiropractic technique, and the majority being trained by the same institution (58% graduated from the Canadian Memorial Chiropractic College). Anecdotally, it has been suggested that the practitioners in the province of NL have a strong sense of community and generally have similar practice styles. Whether or not this local culture has had an effect on these results is unknown; however, it may be worth investigating in the future.

Chiropractors in Newfoundland and Labrador appear to be a relatively homogenous group in terms of years of practice, education, and clinical approach, and they appear to have fairly unified beliefs regarding the use of lumbar spine x-ray for LBP. While the majority of practitioners in this province appear to follow the current recommendations for lumbar spine radiography in patients with LBP, areas for improvement have been identified. Future research should aim to determine factors hindering guideline awareness and adherence in this population and evaluate the impact of tailored knowledge translation strategies to reduce unnecessary LBP imaging.