Skip to main content
Journal of Chiropractic Medicine logoLink to Journal of Chiropractic Medicine
. 2012 Dec;11(4):242–248. doi: 10.1016/j.jcm.2012.10.003

Radiograph utilization and demographics in a chiropractic college teaching clinic

Makani Lew a,, Gregory J Snow b
PMCID: PMC3706708  PMID: 23843755

Abstract

Objective

The purpose of this study is to present radiograph utilization at a chiropractic college teaching clinic, the associated patient demographics, and the utilization rates by body region.

Methods

Data for outpatient services over a 3-year period were extracted from a college clinic administrative software program. Radiographic data were matched with patient demographic information providing the age, sex, and financial class for all patients.

Results

The overall radiograph utilization rate was 8%, with the highest frequency occurring in the spine in the order of lumbar, cervical, and then thoracic regions. Spinal radiographs made up 66% of the total radiographs taken. The utilization rate increased as the age of the patients increased. The average patient age was 46, and 48% were female.

Conclusion

The radiograph utilization rate at this teaching clinic was lower than previous studies. This study provides new information regarding overall and regional radiography rates and associated patient demographics from an American chiropractic college.

Key indexing terms: Chiropractic, Education, Guideline adherence, Radiography

Introduction

Practitioners of musculoskeletal medicine occasionally require radiographic imaging to help establish or clarify a clinical diagnosis. The decision to use radiography should be based on clinical indicators such as trauma (past or present), abnormal neurologic, blood or history and physical examination findings, and failure to respond to therapy.1 Findings from radiographic imaging may alter patient management, care, and/or prognosis. Such findings include significant anomalies (congenital or biomechanical); fractures, dislocations, pathologies, or other conditions that are contraindications to high-velocity/low-amplitude vector forces; age-related conditions; and degenerative processes.1-3

Diagnostic imaging utilization rates may be important for a variety of reasons to many audiences such as policy makers, insurance companies, office managers, practitioners, health care educators, and patients. Utilization rates provide background data to answer the question of over- or underutilization and help provide foundational information that may help shape future guidelines for chiropractic and other health care professions.

In the larger health care arena, routine imaging of patients with low back pain presenting without red flags has been demonstrated to increase harm and cost without improving outcomes.4-8 In addition to the direct costs involved with obtaining radiographs, indirect costs are incurred from incidental findings on the images as the health care provider attempts to further clarify the nature of the finding. Radiographic imaging exposes the patient to low levels of potentially harmful radiation, the risk of which should be balanced against the benefit of the information that might be obtained from the images.

The value of plain film radiography specific to the chiropractic profession has been a point of debate, with little evidence of the actual impact on care.9-11 Opponents of routine imaging argue its limited value, cost, and harm,9,10 whereas proponents suggest that it may reveal underlying issues of concern,1,12 may allow for better care,11 or may provide useful information with less exposure to ionizing radiation compared with a computed tomographic scan13 and significantly less cost than computed tomography or magnetic resonance imaging.14

According to the 2010 National Board of Chiropractic Examiners Chiropractic Practice Analysis,15 chiropractors review more radiographs than they take or order, and take radiographs on a weekly basis to rule out fracture, dislocation, or pathology. Chiropractors also use radiology on a weekly basis to identify spinal listings or chiropractic subluxations, but less commonly than for ruling out fracture, dislocation, or pathology.15 Reported chiropractic radiograph utilization rates in United States and Canada are quite variable, ranging from 6% to 93%, and came from a handful of studies dating back over 3 decades. Much of the data were derived from doctor or patient surveys regarding real or hypothetical patients.16-21 Two studies from the 1990s using data from clinical records from 2 different American states showed utilization rates as 6% to 14%.16,17 From a 3-year study (1986-1990) at a Canadian chiropractic college, the overall utilization rate was 35%.22 A 3-year study (1999-2001) of Ontario, Canada, insurance billing showed the overall radiograph utilization rates as 8% for work-related injuries and 14% for non–work-related injuries.21 From that same Canadian study, there were less work-related radiographs taken on female patients (35%) when compared with the non–work-related radiographs taken (53%).21

The purpose of this article is to describe the radiograph utilization rate and associated patient demographics of a chiropractic college teaching clinic derived from objective data from administrative software.

Methods

A retrospective analysis of radiographic images obtained over a 3-year period (2008-2010) at a chiropractic college teaching clinic was conducted by 2 employees of the college with approval from the college's Institutional Review Board. Data were extracted from the clinic's administrative software (version 9.5.0; Raintree Systems, Inc) and then sorted, filtered, and analyzed with Microsoft Excel (versions 2003 and 2007; Microsoft, Redmond, WA). Data extraction was conducted by one coauthor with full access to clinic information and then deidentified for the other coauthor for analysis. Blinding before analysis was done to protect patient privacy and conform with Health Insurance Portability and Accountability Act regulations.

The data collected were all radiographic series entered by Current Procedural Terminology code into patients’ visit ledgers for each calendar year from January 1, 2008, through December 31, 2010. Chiropractic students and their dependents, outreach clinic patients, and radiographs taken for local community doctors were excluded from the study. Excel “Unique records only” data filter determined annual unique patient lists, as patients may have had multiple series on a single visit, within a year, or from year to year. To calculate the age of patients in the total unique patient list, the middle date of a year (day 183 or July 2) was subtracted from the date of birth. This approach was used because multiple radiographic views may have been taken on the same patient on different dates during any 1 year.

All radiographs were taken at the chiropractic college facility on a Quantum Odyssey High-Frequency Radiograph Generator and through a 200-line 10:1 Bucky onto digital cassettes. They were processed with an AFGA CR35-X digital processor. Radiographs were ordered by the clinician responsible for the patient's case and in accordance with the campus clinic guidelines for absolute and relative radiologic indications. Absolute indications for radiographic examination include the following: significant recent trauma that may have resulted in a fracture, dislocation, or substantial soft tissue injury; suspicion of active or aggressive bone or joint pathology; history of cardiac or pulmonary complaints; or significant or progressive scoliosis. Relative indications for radiographic examination include the following: older than 50 years; acute or progressive neurologic signs; clinical indications of joint instability; history of bone softening conditions or metastasis; unexplained weight loss; lack of physical, historical, or mechanical findings to explain patient symptoms; failure to respond to care as expected; and medicolegal implications. A Certified Radiologic Technologist and assigned interns performed the radiographic studies according to the clinic-established standard views for each series.

Full-spine radiography includes anterior to posterior and lateral projections of the cervical, thoracic, and lumbar regions. To accurately evaluate radiograph utilization by series and body region, all full-spine film series were converted into 3 separate limited series (cervical, thoracic, and lumbar) for data analysis.

The data provided all radiographs taken over 3 years with respective patient ID number, billing code, radiograph series name, date of birth, sex, and financial class. Data were scoured for missing or abnormal data entry errors and corrected where possible. The final data gave the total number of radiograph series taken for each of 3 consecutive years.

A data set was created that categorized radiographs taken into body regions. The 6 categories were cervical, thoracic (includes rib cage), chest, lumbopelvic (includes combined hip-pelvis), upper extremity (clavicle to fingers), and lower extremity (hip to toes).

The data were sorted and filtered to determine the radiograph utilization rates per year, per radiograph type, and per age of patient. Statistical analysis was done using Microsoft Excel functions for mean (average), median, standard of deviation, frequency, Student t test, and χ2 test. Graphs used to analyze data included bar graphs, normal distribution, and a scatter plot.

Results

The results were pooled into 4 basic categories:

  • 1.

    Radiograph utilization rates overall

  • 2.

    Radiograph utilization rates sorted by radiograph series type

  • 3.

    Radiograph utilization rates of unique patients

  • 4.

    Radiograph utilization rates by region of the body

From 2008 to 2010, there were 4010 unique outpatients seen at the clinic (age range, 0-101; average age, 38; and 51% female). There were a total of 466 radiograph series and 44 series types taken. After the full spine to sectional conversion, the total number of radiograph series taken was 614. After filtering the data, the number of unique radiographed patients was 320 (3 patients had no date of birth listed). The unique radiographed patients’ demographics were as follows: age range, 11 to 89; average age, 46; and 48% female (Table 1).

Table 1.

Patient demographics

Characteristics of study patients
Total patients 4010
 Age range (y) 0-101
 Mean (average) age (y) 38
 Median age (y) 34
 Female 51%
Total unique radiographed patients 320
 Age range (y) 11-89
 Mean (average) age (y) 46
 Median age (y) 45
 Female 48%
Total radiograph series taken 466
Total radiograph with full spine converted to sectionals 614
Total types of radiograph series 44

Two sets of numbers were evaluated: total patients and unique patients. The highest percentage of patients seen in the clinic was the 21- to 30-year-old patients (1337 out of 4010). This same group had the greatest number of patients radiographed (77 out of 1337 total patients) (Table 2).

Table 2.

Raw data and utilization rates by age range

Age (years) No DOB 0-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 > 90 Total (n)
Radiographed patients 3 0 19 77 37 55 59 38 22 10 0 320
Total patients 91 71 241 1337 778 578 467 263 135 46 3 4010
Utilization (%) 3% 0% 8% 6% 5% 10% 13% 14% 16% 22% 0% 8%

The χ2 analysis was used to detect the presence or absence of a correlation of radiographs taken to age of patients. The χ2 analysis (using n = 317 to exclude data missing date of birth) was performed in comparison to both flat frequency (equal number of radiographs for all age groups) and Gaussian distribution. Both flat and Gaussian comparisons showed a significant difference between age groups in terms of numbers of radiographs taken (χ2 [n = 317], P < .001) (Fig 1). However, when comparing the number of unique radiographed patients to the total patient numbers, the frequency of radiographs demonstrated a radiograph utilization rate that increased with increasing age (Table 2, Fig 2).

Fig 1.

Fig 1

The χ2 analysis of age groups (n = 317).

Fig 2.

Fig 2

Percentage radiograph utilization by age range (unique radiographed patients, n = 320; total patients, N = 4010). (Color version of figure is available online.)

The overall radiograph utilization rate for unique patients over 3 years was 8% (2008, 10%; 2009, 7%; and 2010, 8%). Radiograph utilization rates by series were based on 614 total radiograph series. Of the 44 radiograph series types taken, the 5 most common (comprising 66% of the total series taken) were as follows: limited lumbar spine (n = 145, 24%), limited cervical spine (n = 101, 16%), limited thoracic spine (n = 101, 16%), complete chest (n = 32, 5%), and posterior to anterior (PA) chest (n = 25, 4%).

When the data were sorted by region of the body, the frequencies were as follows (in descending order): lumbopelvic, 25%; cervical, 23%; thoracic, 18%; lower extremity, 12%; upper extremity, 12%; and chest, 9% (Fig 3).

Fig 3.

Fig 3

Radiograph utilization rates by region of body (2008-2010) (n = 614). (Color version of figure is available online.)

Average patient ages per body region were as follows: lumbopelvic, 49.1 (SD = 18.6); cervical, 47.2 (SD = 17.73); thoracic, 45.3 (SD = 18.09); lower extremity, 43.3 (SD = 18.77); upper extremity, 42 (SD = 17.78); and chest, 59.8 (SD = 15.77). A Student t test on the distribution of age per region radiographed showed no statistical difference except when comparing chest films to the other regions, which suggests a significant age bias.

Scatter plot analysis was used to evaluate the patient age distribution across the 6 body regions radiographed. All 3 types of spinal radiographs showed increased density around 45 to 50 years old. The upper extremity region was clearly bimodal, with clusters around ages 23 and 60 (Fig 4). Closer evaluation of the raw data of the age clusters showed that, for ages 54 to 66, there were several views of the same body region for the same patient (9 unique patient radiographs out of 22 including 13 radiographs on 3 patients). However, for the group aged 17 to 29, there were less repeated views (23 unique patient radiographs out of 25). Because of the lack of clinical data, there was no way to determine if these extremity views were of the same extremity.

Fig 4.

Fig 4

Scatter plot of age distribution of radiographs by region (n = 614).

Discussion

When compared with radiograph utilization rates derived from chiropractic clinic records in the United States (6%-68%)16,19 and Canadian college clinics (12.3%-34%),22,23 our resultant 8% total utilization was low. Underutilization was demonstrated in a previous study of Canadian chiropractic student interns who, when given clinical scenarios, did not opt to take radiographs in spite of red flags.24

Because of the nature of our data source as an administrative database, our study did not attempt to determine the clinical necessity of radiographs or intern/clinician adherence to radiographic guidelines. The low utilization rate may be in part explained by the lower-complexity patient presentations found in chiropractic college clinics.25,26 Another contributing factor may have been patients who fall into the clinical imaging gray area, with interns failing to consider relative indicators for imaging in their haste to initiate care combined with experienced clinicians comfortable proceeding with care without imaging. On the opposite side of the spectrum, interns approaching graduation may advocate taking films on patients with fewer indications in an effort to obtain the necessary radiograph imaging requirements for graduation. If this is so, then it may be that the ordering tendencies of the teaching clinicians and the interns counterbalance each other. This conjecture, of course, warrants further study.

Our female radiograph utilization rate of 48% was lower than those in previous studies16,17,23; however, this may largely be explained by the fact that female patients comprised only 51% of the clinic's total patient base, lower than the characteristic 60% female patients seen in chiropractic practices.15

The spinal regions combined showed a 66% radiograph utilization rate (26% lumbopelvic, 19% cervical, and 12% thoracic), which is to be expected because spinal complaints make up 56% of a chiropractic practice.15 The data in this study found that there was no significant difference between each spinal region, whereas one might have expected a higher proportion of lumbar films.15 There was a trend of increasing radiographs with increasing age, which is to be expected because of increased significant medical history events that come with age.3 The chest film rate of 9% is largely an artifact of the routine radiographic views for the college clinic. For patients presenting for thoracic imaging with thoracic pain, those 40 to 60 years old routinely have a single PA chest view taken, and those older than 60 years routinely have a complete (PA/lateral) chest series taken. There was no attempt to discriminate those series taken as part of a thoracic spine series from those taken as primary chest series.

Although the absolute number of films in the 21- 30-year age group appears large (77 out of 320 views), this was also the largest cohort of clinic patients. Therefore, the relative utilization rate for this age group was 6% and was one of the lowest.

The scatter plot bimodal clustering for the upper extremity around ages 23 and 60 caused us to look at the data again. This reevaluation revealed a higher repeated radiograph rate in patients aged 60 ± 6 years compared with the predominantly unique radiographed patients aged 23 ± 6 years. The active lifestyle of 20-year-old patients and the inclination to degenerative changes for patients in their 60s may partially explain this finding; however, further study including clinical data is necessary to determine the true reason behind this finding.

Limitations

Our data do not entirely apply to the private chiropractic practice and perhaps other chiropractic college clinics. The clinic studied does not take Worker's Compensation, Personal Injury, or insurance cases other than Medicare. Historically, the chiropractic college patient base is not reflective of those in the private practice setting.25,26 In this study, an attempt was made to partly mitigate some of the nonconforming aspects of the chiropractic college clinic by excluding student, student family, and other non–fee-for-service patients from the data.

Limitations of this study exist. Data entry errors of omission or commission may have occurred, but likely not to the extent to significantly impact the outcomes. The clinic that was studied may not be reflective of other chiropractic college clinics relative to patient demographics or radiographic guidelines and/or their impact on imaging decision making. The impact of the inclusion of chest film series for patients older than 40 years with thoracic pain inflates the number of chest series taken and the overall series taken. The impact of intern behaviors and graduation requirements on clinician ordering tendencies does not exist outside the college clinic setting.

Conclusion

The radiograph utilization rate of 8% at this chiropractic college teaching clinic was at the low end of the published range of previous studies, perhaps because of underutilization or an uncomplicated patient base. Spinal radiograph series comprised 66% of the series taken, and utilization rates increased as the patients aged. Further study on this subject across multiple chiropractic college clinics is warranted to give a more accurate and informative picture of their radiographic utilization rates.

Funding sources and potential conflicts of interest

No funding sources or conflicts of interest were reported for this study.

References

  • 1.Yochum T.R., Rowe L.J. Essentials of skeletal radiology. 3rd ed. Lippincott Williams and Wilkins; Baltimore: 2005. Chapter 7: guidelines for obtaining skeletal radiology, chapter 3: congenital anomalies and normal skeletal variants, and chapter 5: natural history of spondylolysis and spondylolisthesis. [Google Scholar]
  • 2.Sherman R. Chiropractic x-ray rationale. J Can Chiropr Assoc. 1986;30(1):33–35. [Google Scholar]
  • 3.Brown J. Prevalence of pathology seen on lumbar x-rays in patients over the age of 50 years. Br J Chiropr. 2001;5(1–2):23–30. [Google Scholar]
  • 4.Srinivas S.V., Deyo R.A., Berger Z.D. Application of “less is more” to low back pain. Arch Intern Med. 2012;172(13):1016–1020. doi: 10.1001/archinternmed.2012.1838. [DOI] [PubMed] [Google Scholar]
  • 5.Chou R., Qaseem A., Owens D.K., Shekelle P., Clinical Guidelines Committee of the American College of Physicians Diagnostic imaging for low back pain: advice for high-value health care from the American College of Physicians. Ann Intern Med. 2011;154(3):181–189. doi: 10.7326/0003-4819-154-3-201102010-00008. [DOI] [PubMed] [Google Scholar]
  • 6.Chou R., Deyo R.A., Jarvik J.G. Appropriate use of lumbar imaging for evaluation of low back pain. Radiol Clin North Am. 2012;50(4):569–585. doi: 10.1016/j.rcl.2012.04.005. [DOI] [PubMed] [Google Scholar]
  • 7.Rubinstein S.M., van Tulder M. A best-evidence review of diagnostic procedures for neck and low-back pain. Best Pract Res Clin Rheumatol. 2008;22(3):471–482. doi: 10.1016/j.berh.2007.12.003. [DOI] [PubMed] [Google Scholar]
  • 8.Chou R., Fu R., Carrino J.A., Deyo R.A. Imaging strategies for low-back pain: systematic review and meta-analysis. Lancet. 2009;373:463–472. doi: 10.1016/S0140-6736(09)60172-0. [DOI] [PubMed] [Google Scholar]
  • 9.Ammendolia C., Hogg-Johnson S., Pennick V., Glazier R., Bombardier C. Implementing evidence-based guidelines for radiography in acute low back pain: a pilot study in a chiropractic community. J Manipulative Physiol Ther. 2004;27(3):170–179. doi: 10.1016/j.jmpt.2003.12.021. [DOI] [PubMed] [Google Scholar]
  • 10.Haas M., Taylor J.A.M., Gillette R.G. The routine use of radiographic spinal displacement analysis: a dissent. J Manipulative Physiol Ther. 1999;22(4):254–259. doi: 10.1016/s0161-4754(99)70053-9. [DOI] [PubMed] [Google Scholar]
  • 11.Harrison D.E., Harrison D.D., Troyanovich S.J. Reliability of spinal displacement analysis of plain x-rays: a review of commonly accepted facts and fallacies with implications for chiropractic education and technique. J Manipulative Physiol Ther. 1998;21(4):252–266. [PubMed] [Google Scholar]
  • 12.Young K.J., Aziz A. An accounting of pathology visible on lumbar spine radiographs. Chiropr J Aust. 2009;39(2):63–69. [Google Scholar]
  • 13.HPS.org [Internet]. Virginia: Health Physics Society Specialists in Radiation Safety: radiation exposure from medical exams and procedures June 2012 fact sheet and background radiation January 2010 fact sheet. Available from: hps.org/hpspublications/radiationfactsheets.html. Accessed September 28, 2012.
  • 14.Chou R. Diagnostic imaging for low back pain: advice for high-value health care from the American College of Physicians. Ann Intern Med. 2011;154:181–189. doi: 10.7326/0003-4819-154-3-201102010-00008. [DOI] [PubMed] [Google Scholar]
  • 15.Christensen M.G., Kollasch M.W., Hyland J.K. Practice analysis of chiropractic 2010: a project report, survey analysis, and summary of chiropractic practice in the United States. National Board of Chiropractic Examiners; Greely, CO: 2010. [Google Scholar]
  • 16.Mootz R.D., Cherkin D.C., Odegard C.E., Eisenberg D.M., Barassi J.P., Deyo R.A. Characteristics of chiropractic practitioners, patients, and encounters in Massachusetts and Arizona. J Manipulative Physiol Ther. 2005;28(9):645–653. doi: 10.1016/j.jmpt.2005.09.019. [DOI] [PubMed] [Google Scholar]
  • 17.Coulter I.D., Shekelle P.G. Chiropractic in North America: a descriptive analysis. J Manipulative Physiol Ther. 2005;28(2):83–89. doi: 10.1016/j.jmpt.2005.01.002. [DOI] [PubMed] [Google Scholar]
  • 18.Kopansky-Giles D., Papadopoulos C. Canadian Chiropractic Resources Databank (CCRD): a profile of Canadian chiropractors. J Can Chiropr Assoc. 1997;41(3):155–191. [Google Scholar]
  • 19.Cherkin D.C., MacCornack F.A., Berg A.O. Managing low back pain—a comparison of the beliefs and behaviors of family physicians and chiropractors. West J Med. 1988;149:475–480. [PMC free article] [PubMed] [Google Scholar]
  • 20.Carey T.S., Garrett J. Patterns of ordering diagnostic tests for patients with acute low back pain. Ann Intern Med. 1996;125(10):807–814. doi: 10.7326/0003-4819-125-10-199611150-00004. [DOI] [PubMed] [Google Scholar]
  • 21.Ammendolia C., Cote P., Hogg-Johnson S., Bombardier C. Utilization and costs of lumbar and full spine radiography by Ontario chiropractors from 1994 to 2001. Spine J. 2009;9(7):556–563. doi: 10.1016/j.spinee.2009.03.009. [DOI] [PubMed] [Google Scholar]
  • 22.Waalen D.P., White T.P., Waalen J.K. Demographic and clinical characteristics of chiropractic patients: a five year study of patients treated at the Canadian Memorial Chiropractic College. J Can Chiropr Assoc. 1994;38(2):75–82. [Google Scholar]
  • 23.Ammendolia C., Cote P., Hogg-Johnson S., Bombardier C. Do chiropractors adhere to guidelines for back radiographs? A study of chiropractic teaching clinics in Canada. Spine. 2007;32(22):2509–2514. doi: 10.1097/BRS.0b013e3181578dee. [DOI] [PubMed] [Google Scholar]
  • 24.Ammendolia C. Adherence to radiography guidelines for low back pain: a survey of chiropractic schools worldwide. J Manipulative Physiol Ther. 2008;31:412–418. doi: 10.1016/j.jmpt.2008.06.010. [DOI] [PubMed] [Google Scholar]
  • 25.Wyatt L.H., Perle S.M., Murphy D.R., Hyde T.E. The necessary future of chiropractic education: a North American perspective. Chiropr Osteopat. 2005;13:10. doi: 10.1186/1746-1340-13-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Nyiendo J.A., Haldeman S. A critical study of the student interns' practice activities in a chiropractic college teaching clinic. J Manipulative Physiol Ther. 1986;3:197–207. [PubMed] [Google Scholar]

Articles from Journal of Chiropractic Medicine are provided here courtesy of National University of Health Sciences

RESOURCES