Background
The use of lumbar radiography within chiropractic clinical practice is well established as a necessary tool to aid diagnosis and direct appropriate treatment of the low back. The use of radiography is associated with risks to the patient, including risks from ionising radiation; unnecessary diagnosis leading to poorer patient outcomes or potentially unnecessary investigation or treatment; and higher costs [
1]. Guidelines directing the appropriate use of lumbar radiography are necessary to increase the likelihood of clinically relevant information being obtained from the radiographs and minimise the associated risk.
Radiographic guidelines for low back pain (LBP) have been published by both the medical [
2-
4] and chiropractic professions [
5]. Despite potential differences in LBP treatment methods used by the two professions, the published guidelines are consistent in their recommendations. Radiographs of the lumbar spine are recommended in cases of suspected serious pathology (ie. cancer, infection, inflammatory arthridity etc.) or trauma with suspected fracture or dislocation. Radiographs are not initially recommended in cases of nonspecific LBP with or without neurological symptoms. Radiographs may be indicated if the patient exhibits 4 to 6 weeks of non-response to treatment [
2,
3,
5].
Despite the recommendations made in the published guidelines many chiropractors still believe that radiographs of the lumbar spine are useful in clinical scenarios outside the current guidelines. Surveys conducted by Ammendolia et. al. [
6] and Walker et. al. [
7] found that 63 and 68 % of respondents respectively would order lumbar radiographs outside the current clinical guidelines [
6,
7]. Reasons for noncompliance with the guidelines were assessed through focus groups by Ammendolia et. al. [
6] and Bussieres et. al. [
8] and found that chiropractors believed lumbar radiographs were useful to detect spinal misalignment; detect and monitor degenerative change; screen for contraindications; educate patients; and for medicolegal reasons [
6,
8], all of which are not consistent with current guidelines. Radiographic instruction by accredited chiropractic schools may also lack adherence to current guidelines, with 34 % of radiology instructors instructing students to consider using radiography to screen for pathology or contraindications and 25 % to provide patient reassurance [
9]. In a study performed by Bussieres et. al. [
10], it was found that the chiropractic school attended was one of the most influential predictors of future radiograph utilisation [
10] and as such the radiographic instruction received may be important in directing radiographic guideline adherence.
There is some evidence that educating chiropractors regarding current radiographic guidelines is associated with increased guideline compliance [
11,
12] or a reduction in radiographic referrals [
13]. However, to establish whether this would be a useful intervention strategy amongst a specific chiropractic population, the current awareness of radiographic guidelines amongst those chiropractors should be identified.
Therefore, the aims of this study were to quantitatively assess whether Australian chiropractors reported awareness of current radiographic guidelines for LBP and whether they demonstrated adherence to the key messages in the guidelines. Finally, demographic, chiropractic practice and radiographic usage characteristics were investigated for association with poorer guideline adherence.
Discussion
The primary finding of this study was that 49.6 % (95 % CI: 44.9, 54.4) of Australian chiropractors’ report definite awareness of current radiographic guidelines for the low back and only 14.3 % (95 % CI: 11.4, 17.7) were aware of published guidelines for the chiropractic profession. This low awareness of current guidelines is consistent with the lack of adherence found in both chiropractors’ reported likelihood of referring for radiographs and their agreement with scenarios where radiographs could be useful. Poorer reported guideline adherence was associated with chiropractors who: refer to in-house radiographic facilities; practice techniques other than diversified; or are unsure or unaware of current radiographic guidelines.
The results of this study indicate that adherence to radiographic guidelines for LBP by Australian chiropractors needs to be improved. Current guidelines recommend that radiographs are not useful for the majority of acute LBP cases, including those associated with neurological symptoms; to screen for contraindications; or to analyse for biomechanical change. Radiographs are indicated by the suspicion of trauma or pathology as a cause of LBP or after a period of four to 6 weeks non-response to treatment [
5]. Although reported adherence to guidelines was high in clinical scenarios consistent with current guidelines, many chiropractors also reported that radiographs would be useful in cases of acute LBP, LBP with neurological symptoms, to screen for contraindications or to diagnose or analyse biomechanical changes, all of which show poor guideline adherence. It is important that the clinical benefit of taking a radiograph outweighs the inherent risks involved. Indications for referring for radiographs of the lumbar spine in clinical guidelines are based on best current evidence to maximise the clinical benefit to risk ratio. Poor adherence to clinical guidelines results in an increased risk to the patient as there is a decreased likelihood of clinical benefit but the same inherent risks associated with radiography. Furthermore, a lack of adherence to evidence- based clinical guidelines reflects poorly across the chiropractic profession as primary contact healthcare practitioners. If determined significant enough, this lack of adherence may result in legislative changes limiting radiographic referral rights, to improve evidence-based care and reduce unnecessary radiographs.
Adherence to radiographic guidelines may be limited by a lack of awareness of the current guidelines. Low awareness of the current guidelines in Australian chiropractors was seen in this study, particularly of those guidelines published for the chiropractic profession specifically [
5]. Although an awareness of guidelines does not necessarily mean guideline messages will be adhered to, chiropractors need to be aware of the current standards they should be meeting. In addition, the odds of being adherent to the guidelines were between 1.72 to 2.35 times greater if the chiropractor reported guideline awareness in this study. Guideline education, such as online dissemination or reminders, didactic workshops and media campaigns have shown some previous evidence of effectiveness in improving guideline adherence or reducing inappropriate imaging of the lumbar spine within chiropractic populations [
11-
13] and may be cost-effective interventions to increase guideline awareness. Clinical scenarios with lower reported adherence to guidelines should be targeted during these educational strategies, with a focus on explaining the limited clinical benefit of referring for radiographs in these scenarios. In this study, the chiropractic institute of graduation was not significantly associated with awareness of current guidelines and the time since graduation was not significantly associated with guideline adherence. This indicates that similar proportions of chiropractors, graduating from different institutes and at different time points, have similar guideline awareness and adherence respectively. Therefore, educational strategies should be targeted both at practicing chiropractors and within chiropractic teaching institutions.
Poorer guideline adherence was associated with chiropractors referring to in-house radiographic facilities and those practicing techniques other than diversified as their primary manual technique style. Therefore, it may be important to target educational strategies towards these groups of chiropractors to improve adherence to guideline recommendations. Referral to in-house radiographic facilities showed the strongest association with poorer guideline adherence. In chiropractors referring to their own radiographic facilities the odds of being likely to take radiographs of new patients were increased 7.23 times and the odds of agreeing that radiography is useful routinely before spinal manipulative therapy were increased 6.26 times. These findings are of concern since self-referral to practitioner-owned radiographic facilities has the potential to be misused for financial incentives. Strong consideration needs to be given by these practitioners regarding the clinical justification of referring for radiographs, particularly when not in adherence with clinical guidelines. Given their responsibility in owning and operating radiographic facilities and their need to comply with national regulations for ionising radiation, strategies to increase guideline adherence within this group should be prioritised. Diversified technique is the most common spinal manipulative technique practiced among chiropractors [
15-
17], however, there are many other technique systems that may be used in chiropractic practice [
18]. Historically, chiropractors analysed radiographs to assess for spinal alignment and guide appropriate management [
19]. Although this approach is no longer recommended by radiographic guidelines [
5] and is not a component of diversified technique, some chiropractic technique systems still include the use of radiographic analysis as a diagnostic tool to detect biomechanical changes [
18,
20,
21]. Two of these technique systems, Gonstead technique and Chiropractic Biophysics, were the primary technique system used by 10.2 % of respondents, representing more than one quarter of respondents who did not practice diversified technique. The inclusion of techniques emphasising radiography as a diagnostic tool may explain the poorer adherence to guidelines in chiropractors practicing techniques other than diversified technique. Consideration needs to be given by chiropractors practicing these techniques to determine whether radiographic analysis gives sufficient clinical benefit to justify the associated risks.
The awareness of radiographic guidelines for low back pain within Australian chiropractors has not been previously assessed in the literature. Similarly, no studies assessing awareness of guidelines in other chiropractic populations were found. This information is important as this may indicate the need for education of the key guideline messages to the chiropractic population. Three previous studies have quantitatively assessed the adherence of registered chiropractors to radiographic guidelines for the management of LBP. Two surveys performed in Canada with 26 and 32 responses respectively found that 63 and 59 % would use radiography for acute LBP without indicators of potential pathology and 68 and 66 % thought that radiography was useful in the evaluation of acute LBP [
6,
12]. Despite the small sample sizes the results are consistent with those of this study. Similarly, Walker et. al. [
7] surveyed 274 Australian chiropractors and found that 68 % of respondents would order radiographs in clinical situations where it was not indicated [
7]. Previous qualitative research has assessed reasons for poor adherence with guidelines and found that reasons for taking radiographs for LBP include to perform biomechanical assessment, to screen for contraindications, to confirm a diagnosis or direct treatment, to educate patients, for medicolegal reasons or for financial incentives [
6,
8]. The results of this study provide quantitative evidence supporting these findings, with respondents agreeing that radiographs of the low back are useful to screen for contraindications (55.8 %), to confirm a diagnosis or direct treatment (61.3 %) and for biomechanical analysis (30.7 %). In the current study, institute of graduation was not significantly associated with awareness of current guidelines for LBP radiography. In contrast, an American study found that the chiropractic school attended was strongly associated with rates of imaging utilisation [
10]. There are only three currently operating institutes in Australia with graduate chiropractors compared to 21 American institutes in the study by Bussieres et. al. [
10], giving a wider range for variation in American compared to Australian institutes. This variation, the different geographic locations, or the different outcomes measures may account for the differences in study findings.
Key strengths of this study include the primary outcome question and the analysis for potential associations with poorer reported guideline adherence. Previous studies have shown chiropractors’ to demonstrate a lack of adherence to radiographic guidelines for LBP [
6-
8], however, they have not assessed whether chiropractors are aware of current radiographic guidelines. This may impact potential education strategies to increase guideline adherence. Knowledge of potential associations with poorer guideline adherence may also help direct future strategies to improve guideline adherence. Performing multivariate logistic regression analysis enabled assessment for these associations while accounting for potential confounders.
The main limitation of this study is the inability to calculate an accurate response rate. The response rate was calculated as 9.9 % based upon the number of completed surveys compared to the total number of registered chiropractors in Australia. However, this is likely to be an underestimate as the number of people who actually received the survey link could not be accurately determined. The survey was distributed via email distribution to association members and a link in the quarterly newsletter from the Chiropractors’ Board of Australia. Although all registered Australian chiropractors should have received the survey information, it is unknown how many actually received and opened the emails or newsletter and how many followed the link to obtain information about the survey. Calculation of an accurate response rate is necessary to assess non-response bias. As this could not be achieved, assessment of representativeness of the respondent sample was addressed by comparing demographic characteristics from survey respondents to demographic data from the Chiropractic Board of Australia and previous research. Data from the Chiropractors’ Board of Australia [
14] showed a similar distribution of chiropractors by Australian state or territory except for a higher number of survey respondents from New South Wales (41 % compared to 33 %) and a lower number from Victoria (18 % compared to 26 %). Diversified was listed as the primary manual technique style for 61 % of survey respondents, which is consistent with previous research [
16,
17]. Likewise, less than a quarter of respondents referred to in-house radiographic facilities which is also consistent with previously published research [
17,
19]. Although there are some similarities between demographics of survey respondents and previously published data we cannot eliminate the possiblity of non-response bias. The results of this survey should therefore be interpreted with caution as they may not be reflective of the Australian chiropractic population as a whole. Validated questions could not be found for the secondary outcome questions in this study. The questions used were adapted from those previously used in the literature [
9,
12] and were informally piloted prior to use. The secondary outcome questions were asked in two different formats: assessing the respondents reported likelihood of referring for radiographs, and their reported agreement with the usefulness of radiographs, in certain clinical scenarios. Responses and the results of the multivariate logistic regression analysis were similar despite the different question formats used, suggesting that they measured a similar construct.
It is important that Australian chiropractors are aware of current guidelines and demonstrate increased adherence to those guidelines. Implementation of guideline education strategies, both within chiropractic teaching institutes and to chiropractors within the profession, should be undertaken and may be a low-cost strategy to increase awareness and adherence to current guidelines. Further research should be carried out to measure the effectiveness of these educational strategies and whether further intervention is necessary. Targeted interventions may be needed in specific groups within the chiropractic profession, including those with in-house radiographic facilities and those practicing techniques other than diversified technique. Further research may be needed to identify the barriers specific to these populations and design suitable implementation interventions accordingly.