Self-reported recognition of undiagnosed life threatening conditions in chiropractic practice: a random survey

There are several different definitions of what constitutes a primary care physician [1, 2]. Most of these definitions incorporate the concept of “assumption of longitudinal responsibility for the patient regardless of the presence or absence of disease” [3]. Based on this definition a physician, at a minimum, has the moral and legal responsibility to be able to identify and refer, when necessary, conditions which if left unrecognized and untreated may result in serious and even deadly consequences to the patient.

There is an ongoing controversy within the healthcare community regarding role of the chiropractic physician as well as debates on this issue within the chiropractic profession itself. There are three basic positions argued; primary care, neuromusculoskeletal care and subluxation based care. The Council on Chiropractic Education mandates that chiropractic educational institutions “train a competent doctor of chiropractic who will provide quality patient care and serve as a primary care physician” [4]. Duenas et al. found 94% of chiropractic colleges utilize the term “primary care provider” or similar variants when describing the product of their educational programs [1]. In spite of primary care training some argue the proper role of the chiropractor is as a neuromusculoskeletal or musculoskeletal specialist [5, 6]. This position is supported at least in part by the profile of the typical chiropractic patient. According to The National Board of Chiropractic Examiners, 82.9% of presenting complaints in the typical chiropractor’s office consist of back pain, extremity pain, neck pain and headache. Less than 10% of patients present with non-neuromusculoskeletal complaints [7]. A third position argues chiropractic is a separate and distinct health care specialty that should limit itself to detecting and correcting subluxations [8].

The training students receive in chiropractic college is broad based and similar to medical training [9]. A 2005 study tested 4th year chiropractic students and 4^th year medical students on their knowledge of primary care activities. Overall chiropractic students scored slightly lower than medical students in all areas but musculoskeletal diagnosis [10]. Considering the differences on emphasis relating to primary care between the disciplines the outcomes are not surprising, except possibly that the chiropractic student did so well compared to their medical counterpart. Additionally, a 1995 survey of 753 chiropractors reported 78.5% of respondents considered their training as primary care providers as “adequate” [11].

It has been suggested chiropractic training is deficient to medical training for primary care due to the lack of patient contact [12]. The chiropractic student usually encounters patients with musculoskeletal conditions or asymptomatic patients presenting for wellness care during their internship [13]. Although a medical doctor can be granted a license to practice with as little as 1 year in a post graduate residency [14], most have a minimum of 2 to 3 years in a residency program. This additional exposure provides patient contact with a wide variety of conditions. Essentially the chiropractic physician is limited to knowledge gained in the classroom and limited patient contact. Whether this difference produces a physician that is insufficiently trained to recognize undiagnosed life threatening conditions (LTC) has not been determined.

A review of the Job Analysis of Chiropractic reveals that chiropractors have contact with patients with a wide variety of serious conditions and often co-manage these patients with a medical provider [7]. What is not included in the analysis is whether the chiropractor was the first to recognize these conditions in an undiagnosed patient.

There is relatively little information in the scientific literature relating to the chiropractic physician’s ability to recognize LTC. The purpose of this paper is to begin the process to determine whether chiropractic training equips the chiropractic physician to recognize conditions which were previously undiagnosed, may be fatal if left untreated and usually fall outside the scope of chiropractic practice. Thus, the specific purpose was to identify the type and frequency of self-reported cases of undiagnosed life threatening conditions which doctors of chiropractic claim have presented in their offices during their careers. If these conditions were commonly reported, it was also of interest to explore any potential associations between college of graduation, years in practice, type of practice or prior medical training and the rate of recognition.

The study was a self-administered, epidemiological survey involving randomly selected doctors of chiropractic (DC) practicing in the United States. The survey instrument was a self-administered one paged questionnaire that asked DCs to state the number of cases (patients), based on their recall over the course of their careers, which presented in their offices with LTC which had not been previously diagnosed or recognized. Included in the survey was a list of the following conditions: carcinoma, abdominal aneurysm, deep vein thrombosis, stroke, myocardial infarction, subdural hematoma and other. This list was developed from an earlier unpublished study of chiropractors which did not include a list of conditions. The operational definition of “life threatening” provided to the doctor was “an illness or injury that, if it remains unrecognized and untreated, could result in the death on the individual within 24 months.” The initial and follow up survey included a request to complete the survey even if the doctor had never encountered a LTC in their practice. Based on the responses given it is possible some doctors did not read or ignored the definition of “life threatening condition” the authors provided. There was evidence this occurred in some responses. In reviewing the surveys the authors attempted to correct for these errors. The authors recognize their exclusions were not systematic. Since this is a survey and by nature biased, the authors did not feel it necessary to develop a system to rule out specific responses. The authors used their training as physicians to eliminate those responses that were, in their opinion, not life threatening or excessive reporting. Data were reviewed prior to statistical analysis and data points that were determined to be invalid were excluded.

The survey also solicited demographic information such as gender, college of graduation of the doctor and whether or not they have had any prior health care experience. They were also asked to provide information on some practice characteristics such as number of years in practice, type of practice (solely subluxation-based or not), and practice settings (rural, suburban, urban, solo, multi-DC, or multi-discipline practice) as well as whether or not they were employed by the Veteran Administration (VA) or the military. In early November 2010, 2000 doctors were randomly selected by computer program from a mailing list of 57878 chiropractors provided by MPA Media. Two mailings to each doctor were performed over a two month period. The study was approved by both the research committee and the institutional review board of the institution where this research was conducted.

A total of 588 completed surveys were received (a 29.9% response rate) and 35 uncompleted surveys were returned as undeliverable (31), doctor was deceased (2), or retired (2). Of the completed surveys 12 (2%) were eliminatedfrom the analysis due invalid responses. Thirty seven responses were removed from the list of reported LTC which in the opinion of the authors did not qualify as life threatening. Additionally the responses of 12 doctors who reported between 171 to 500 encounters were also excluded as being excessive.

The purpose of this study was not to establish whether chiropractic training prepares the chiropractor to be a primary care physician, neuromusculoskeletal specialist or subluxation based provider. The role of the chiropractor in healthcare has been eloquently debated in several articles [1, 2, 5, 16–18]. Our interest was to provide preliminary data in the process of determining if chiropractic training provides the chiropractor the ability to recognize undiagnosed LTC which is a fundamental requirement of all physicians. The 576 surveys validated for our study represent 11442 years in practice and a reported 3861 patients with an undiagnosed LTC. Based on this data the typical chiropractor could expect to have 1 patient with an undiagnosed LTC to present in their office every 2.5 practice years using Poisson Regression analysis. Respondents identified over 50 different conditions that fit the operational definition of “life threatening”. Considering the total number and wide variety of diagnoses reported the authors feel there is preliminary evidence to support the position that chiropractic training provides an educational base to the chiropractor to be able to recognize undiagnosed LTC when they present in their office.

The concern relating to missed or delayed diagnosis is common throughout the healthcare professions. Between 2007 and 2009 the NCMIC Insurance Company reported 7.5% of claims made against their insured chiropractors were related to missed diagnosis or delayed referral based on personal communications with Keith P. Henaman, Assistant Vice President of Claims. Within the medical community the most common malpractice lawsuits were related to missed diagnosis, failure to diagnose or delayed diagnosis [19] representing 34% of primary care cases according to Phillips et al. [20]. Interestingly the most common missed diagnoses in medical practice are similar to the most common presentations of LTC in chiropractor’s offices. These include carcinoma, myocardial infarct, stroke and abdominal aneurysm [20–22]. This is not to suggest chiropractors are diagnosing conditions missed in the medical doctor’s office but that these type conditions are common, easy to misdiagnose and can have deadly consequences. As one author noted the gold standard of diagnosis, the clinical autopsy, would not have been necessary in 12% of cases if the proper diagnosis had been made and the proper treatment given [23].

A very important component of our study was to identify the most common conditions chiropractors reported encountering. In rank order they include carcinoma, abdominal aneurysm, deep vein thrombosis, stroke, myocardial infarction and subdural hematoma. There was a large group of other conditions commonly reported which included severe hypertension, other cardiovascular disease, infection, diabetes, meningitis, cervical fracture and appendicitis. It is interesting to note that many of the most commonly reported presenting conditions often share one of two characteristics. The first is that many patients present with conditions that may manifest themselves as neuromusculoskeletal in origin thereby causing them to choose a chiropractor as the physician of choice. These conditions could include myocardial infarction manifesting as thoracic tightness, carcinoma resulting in perceived muscle/bone pain, deep vein thrombosis resulting in leg pain and stroke or subdural hematoma resulting in headache or extremity weakness. Secondly many of the conditions reported could appear to be discovered incidental to radiographic examination such as abdominal aneurysm and carcinoma.

This paper had several limitations which are common to survey instruments. Recall bias may have resulted in over reporting or under reporting of LTC.

The presence of LTC was determined by the survey takers themselves and did not require confirmation by any outside source. It is possible that what a responder thought was a LTC may have been misdiagnosed or not truly life threatening and as a result weakened our findings. It should be noted a total of 37 conditions were removed from data analysis because in the opinion of the authors they did not meet the operational definition of life threatening. Two examples of conditions removed included subluxation and mononucleosis. The response rate for this survey was lower than desired. The findings in a study by Russell et al. indicated that the response rates for postal surveys of chiropractors range from 7.0% to 91.4% with an a mean of 52.7% [24]. It should be noted many surveys reported in the Russell et al. study were of specific populations of chiropractors which generally provide higher response rates, while surveys which include all chiropractors generally have lower response rates [25–27].

Lastly this survey did not attempt to determine “missed” diagnoses on the part of the chiropractor as that was not the purpose of the paper. The authors acknowledge that diagnoses may have been missed which would have increased the frequency of these conditions.

The data from this study provides preliminary evidence that chiropractic education prepares the DC to recognize LTC when they present in their offices. Additionally this study provides preliminary identification and frequency of LTC which commonly present in the chiropractor’s office. This information is important to chiropractic colleges, continuing education programs and the practicing DC. Heightened awareness and knowledge of these conditions should improve patient care by reducing the risk of missed or delayed diagnosis.

As stated previously this is a preliminary study. Additional prospective studies are required to develop more reliable and accurate information in this area of investigation.