The ICD-10 coding system in chiropractic practice and the factors influencing compliancy

Abstract

Background: The International Classification of Diseases (ICD) provides codes to classify diseases in such a manner, that every health condition is assigned to a unique category. Some of the most common diagnoses made by chiropractors are not included in the ICD-10 coding system, as it is mainly medically orientated and does not accommodate these diagnoses. This can potentially lead to reimbursement problems for chiropractors in future and create confusion for medical aid schemes as to what conditions chiropractors actually diagnose and treat. Aim: To determine the level of compliancy of chiropractors, in South Africa, to the ICD-10 coding procedure and the factors that may influence the use of correct ICD-10 codes. As well as to determine whether the ICD-10 diagnoses chiropractors commonly submit to the medical aid schemes, reflect the actual diagnoses made in practice. Method: The study was a retrospective survey of a quantitative nature. A self-administered questionnaire was e-mailed and posted to 380 chiropractors, practicing in South Africa. The electronic questionnaires were sent out four times at two week intervals for the duration of eight weeks; and the postal questionnaires sent once. A response rate of 16.5% (n = 63) was achieved. Raw data was received from the divisional manager of the coding unit of Discovery Health (Pty) Ltd. in the form of an excel spreadsheet containing the most common ICD-10 diagnoses made by chiropractors in South Africa, for the period June 2006 to July 2007, who had submitted claims to the Medical Scheme. The spreadsheet also contained depersonalised compliance statistics of chiropractors to the ICD-10 system from July 2006 to October 2008. SPSS version 15 was used for descriptive statistical data analysis (SPSS Inc., Chicago, Ill, USA). Results: The age range of the 63 participants who responded to the questionnaire was 26 to 79 years, with an average of 41 years. The majority of the participants were male (74.6%, n = 47). KwaZulu-Natal had 25 participants (39.6%), Gauteng 17 (26.9%), Western Cape 12 (19%), Eastern Cape four (6.3%), Free State and Mpumalanga two (3.1%) each and North West one (1.5%). The mean knowledge score for ICD-10 coding was 43.5%, suggesting a relatively low level of knowledge. The total percentage of mistakes for electronic claims was higher for both the primary and unlisted claims (3.93% and 2.18%), than for manual claims iv (1.57% and 1.59%). The total percentage of mistakes was low but increased marginally each year for both primary claims (1.43% in 2006; 1.99% in 2007; 2.33% in 2008) and unlisted claims (0% in 2006; 2.61% in 2007; 3.07% in 2008). CASA members were more likely to be aware of assistance offered, in terms of ICD-10 coding through the medical schemes and the association (p = 0.131), than non-members. There was a non-significant trend towards participants who had been on an ICD-10 coding course (47.6%; n = 30), having a greater knowledge of the ICD-10 coding procedures (p = 0.147). Their knowledge was almost 10% higher than those who had not been on a course (52.4%; n = 33). Most participants (38.1%; n = 24) did not use additional cause codes when treating cases of musculoskeletal trauma, nor did they use multiple codes (38.7%; n = 24) when treating more than one condition in the same patient. Nearly 70% of participants (n = 44) used the M99 code in order to code for vertebral subluxation and the majority (79.4%; n = 50) believed the definition of subluxation used in ICD-10 coding to be the same as that which chiropractors use to define subluxation. According to the medical aid data, the top five diagnoses made by chiropractors from 2006 to 2007 were: Low back pain, lumbar region, M54.56 (8996 claims); Cervicalgia, M54.22 (6390 claims); Subluxation complex, cervical region, M99.11 (2895 claims); Other dorsalgia, multiple sites in spine, M54.80 (1524 claims) and Subluxation complex, sacral region, M99.14 (1293 claims). According to the questionnaire data, the top five diagnoses (Table 4.24) were: Lumbar facet syndrome, M54.56 (25%); Lumbar facet syndrome, M99.13 (23.3%); Cervical facet syndrome, M99.11 (21.7%); Cervicogenic headache, G44.2 (20%) and Cervicalgia, M54.22 (20%). Conclusion: The sample of South African chiropractors were fairly compliant to the ICD-10 coding system. Although the two sets of data (i.e. from the medical aid scheme and the questionnaire) regarding the diagnoses that chiropractors make on a daily basis correlate well with each other, there is no consensus in the profession as to which codes to use for chiropractic specific diagnoses. These chiropractic specific diagnoses (e.g. facet syndrome) are however, the most common diagnoses made by chiropractors in private practice. Many respondents indicated that because of this they sometimes use codes that they know will not be rejected, even if it is the incorrect code. For more complicated codes, the majority of respondents indicated that they did not know how to or were not interested in submitting the correct codes to comply with the level of specificity required by the medical aid schemes. The challenge is to make practitioners aware of the advantages of correct coding for the profession.