To accurately record chiropractic encounters, the creation of a large number of new terms was required. This study has shown that by adding chiropractic specific terms to the ICPC-2 PLUS terminology, it is possible to code a large number of chiropractic encounters to enable classification and reporting of chiropractic encounters to the desired level. However this is a work in progress and further data collection will require the addition of new terms.
Although existing ICPC-2 PLUS terms mostly covered the reason for the encounters and processes of care, the PLUS terms were not as successful in representing the diagnoses/problems recorded by chiropractors. Just under half of the total diagnoses/problems recorded in COAST were coded using newly created chiropractic specific terms.
The strength of this study came from using the well-established ICPC-2 PLUS terminology as a base and then adding to this to meet chiropractic specific needs. A large number of chiropractor specific terms were added to record chiropractic encounters. General practice and chiropractic are different in their scope so this had been expected. Using the ICPC-2 PLUS process allowed the straightforward creation of these new terms and then enabled these to be grouped together for ease of reporting.
The new terms generated in this study are a reflection of terms used by chiropractors in practice to represent what occurred in their patient encounters. Having a term assigned does not mean the diagnosis/problem can be substantiated by evidence, it simply means that one or more chiropractors used the term to record their patient encounters. More research is needed into the diagnosis/problem descriptions used by chiropractors and the level of evidence that supports the existence of the condition the chiropractors labeled. This issue has been extensively examined in the general medical practice setting, including that a definitive diagnosis is not apparent in about half of general practitioners’ consultations, that many patients present to general practice without a serious physical disorder, and that there is wide variance in the way general practitioners describe the diagnosis/problem under management .
This study highlighted the wide range of terms used in documentation of chiropractic encounters. This resulted in separate terms being created for what essentially could be considered the same diagnosis/problem. All new terms were mapped to ICPC-2 rubrics and chapters, so the inter-clinician variance in terms used in clinical practice is reduced when reported at these levels, where like terms are classified to the one rubric.
While a consultation process took place among the members of the research team to determine if a new term should be created, 169 new terms were still required. We assume that any further documentation of chiropractic encounters will require the generation of additional terms, and possible merging of the terms already generated, particularly the terms that were not used by the chiropractors in COAST. Future research in this area should include investigation into the terms used in chiropractic to distinguish synonyms from separate terms. A more extensive consultation process with members of the chiropractic profession would potentially allow synonyms to be identified and linked to one term rather than to have several separate terms. For example, Restriction/Fixation;pelvis may be linked to the PLUS term ‘Dysfunction;pelvis’ rather than be a separate term.
Two examples of new chiropractic terms generated in COAST highlight the different meanings of the same term used in the general medical practice profession and the chiropractic profession. First, the term ‘subluxation’ is already present in ICPC-2 PLUS in the L80 chapter ‘Dislocation/Subluxation’. However, this term is listed under the accepted medical definition of subluxation, that is, a partial dislocation. Some chiropractors use this term in a different context hence the series of terms related to ‘chiropractic subluxation’ were generated (see Additional file 1) . Second, in BEACH the sacroiliac joint is not considered a moving joint, like a wrist joint, and sacroiliac recordings are classified in the rubric ‘L03: Low Back Symptom/Complaint’, because the sacroiliac joint is regarded as part of the back. However, some chiropractors consider that the sacroiliac joint is a distinct moving joint hence a series of new codes were generated in COAST to represent this (see Additional file 1) .
The allocation of ICPC-2 rubrics to the new chiropractic specific terms generated in COAST was done primarily for reporting purposes. Investigators had anticipated that any terms created during COAST may be difficult to assign to a chapter, as there can be disagreement as to which body system the diagnosis/problem belongs to. In most cases rubric allocation (including chapter) was straight forward, such as allocating ‘Dysfunction; joint; sacroiliac’ to the Musculoskeletal chapter using ‘Dysfunction; joint’ as a reference guide. However, in some cases rubric selection was more subjective and investigators acknowledge that other researchers may allocate different ICPC-2 rubrics to the J99 codes.
Using the COAST grouping process made it possible to report both the distribution of individual diagnoses/problems relevant to a chiropractic audience and also to the wider health community by using broader groups. The existing groups used by BEACH are general medical practitioner focused; for example Hypertension, Neoplasm and Abdominal Pain. Although the existing groups did include musculoskeletal groups such as Osteoarthritis and Sprains/Strains, in some cases the ICPC-2 PLUS group did not include terms a chiropractor would use. For example, the ICPC-2 PLUS group ‘Back Complaints (all)’ did not include the rubric L20 ‘Joint Symptom/complaint Not Otherwise Specified’ which was considered essential by the research team to include for chiropractic reporting.
Special consideration was required when assigning rubrics to COAST reporting groups, particularly as the majority of the groups were derived from newly created ‘J-codes’. Great care had been taken when classifying new chiropractic terms to ICPC-2; however, with each allocation of an ICPC-2 rubric to a COAST group, a ‘double check’ of the rubric was made. This ensured that the ‘J-code’ had been classified to the most appropriate rubric and that the rubric was assigned to the most appropriate group according to the COAST data. In this way the research team produced the groups they felt were most relevant to chiropractic. For example, to better report chiropractic care, the reporting group ‘Health Maintenance/Preventative Care’ combined any ICPC-2 PLUS term that included ‘Health Maintenance’ and ‘Check Up’ with the J99 code of ‘Wellbeing’.
It should be noted that the term ‘Problem’ has been used to name the COAST groups rather than ‘Symptom’ or ‘Complaint’. Within chiropractic clinical encounters, there is often no symptom or complaint as the reason for encounter, as shown by our large number of encounters being recorded as wellbeing and health maintenance visits. The COAST research team considered that the profession would be more accepting of the alternate title for reporting of results.
Researchers who wish to use the new ICPC-2 PLUS (Chiro) need to be aware of its limitations. The chiropractic version of ICPC-2 PLUS only contains terms recorded by the 52 participants in the COAST study who recorded 4,464 chiropractor-patient encounters, including recording details of 6,225 reasons for encounter and 6,491 diagnoses. Expansion of this study to a wider group of participants would be expected to result in additional terms added to the classification system.
Further, COAST specific reporting groups are not transferable to other studies, because they only include the ICPC-2 PLUS terms used in this study, plus the newly generated chiropractic terms. This is especially true because the COAST groups were created at the rubric level rather than at the term level. For example the COAST group ‘CG103-Back syndrome with radiating pain’ included all terms allocated to the rubric N99 ‘Neurological disease, other’. In the COAST data, only the terms ‘Neuralgia’ (N99 014) and ‘Radiculopathy’ (N99 038) were present from the whole N99 rubric. In the PLUS terminology, there are currently 34 terms allocated to the N99 rubric, including terms such as ‘Narcolepsy’ (N99 013) and ‘Encephalopathy’ (N99 042) which are not relevant in the ‘CG103-Back syndrome with radiating pain’ group.
A comprehensive chiropractic grouping tool would require each of the ICPC-2 PLUS terms to be considered for each of the COAST groups. In some cases, this would result in individual terms within a rubric assigned to different groups. For example, neuralgia might be grouped to ‘CG103-Back syndrome with radiating pain’, while Narcolepsy might not be assigned to a chiropractic group. More work is needed before this grouping can be used by other research teams.
When previous studies have used ICPC in their research, the ICPC classification was used as required for each study’s particular needs. The focus of Meier and Rogers’ (2006) study of Traditional Chinese Medicine encounters was to develop data management and reporting guidelines . While that study demonstrated the use of ICPC outside a general medical practice setting, it did not add to ICPC by producing new classes specific to Traditional Chinese Medicine. Similarly, Van Mil et al. generated a pharmacy specific classification system and provided a subset of pharmacy specific ICPC codes rather than develop a terminology that was then classified to ICPC .
The production of ICPC-2 PLUS (Chiro) for COAST differed in two main ways from these previous studies. COAST used ICPC-2 PLUS to develop the system rather than ICPC-2; this provided coders with a large number of chiropractic relevant terms already present within the terminology. In addition, COAST was able to create new terms specific to chiropractic rather than only using those available. By using the ICPC-2 PLUS system, researchers had a wider range of keywords to search when assigning terms to reasons for encounter, diagnoses and procedures. Although using terms specifically relevant to general practice, the ICPC-2 PLUS keyword list was suited to coding information documented at chiropractic encounters. This was shown by the low percentage of new terms that were created to accurately describe reasons for encounter.
When terms relevant only to chiropractic were not present on the ICPC-2 PLUS term list, researchers were able to add new terms. This enabled a significant number of problems identified by chiropractors to be recorded that would have otherwise been placed under a non-specific term if forced to fit into the existing system. The research team had anticipated the need for new chiropractic specific terms due to the differing styles of practice and the wide range of terminology used in the profession.