Abstract
Objective: To determine the rates of simultaneous antinuclear antibodies (ANA) screening and extractable nuclear antigen (ENA) testing that do not follow recommendations.
Design, Setting, and Participants: Retrospective cohort study of adult patients (≥18 years) with a HEp-2 ANA or ENA ordered in the Marshfield Clinic Health System.
Main Outcome(s) and Measure(s): Counts of patients having simultaneous ANA and ENA laboratory testing or ENA testing without ANA screening. Relevant ENA positivity in ANA negative patients. Secondary measures included relative timing of ANA and ENA ordering, potential cost savings of unnecessary testing, and provider ordering characteristics including specialty and provider type.
Results: Of 58,627 cohort patients, 39,155 (66.8%) were women, and the mean (SD) age at first laboratory testing was 48.7 (19.0) years. The negative ANA with positive ENA rate was 2%. Further stratification identified only 23 diagnosed autoimmune connective tissue diseases (AI-CTDs) in this 2%, with a resulting negative ANA with relevant positive ENA rate of 0.37%. Simultaneous ANA and ENA testing occurred in 8.3% of patients, and an ENA only was ordered in 24.2% of patients. The simultaneous or non-sequential ordering of ANA and ENA testing resulted in significant health care costs of $2,293,251.80 over 20,112 unique patients.
Conclusions and Relevance: A significant percentage of providers do not follow recommendations to sequentially order ANA and ENA testing on patients with suspected AI-CTDs. Significant saving in health care spending without failure to diagnose AI-CTDs can be achieved if ANA testing is performed first, followed by ENA testing when suspecting AI-CTDs in patients.
- ANA
- Autoimmune connective tissue disease
- Choosing Wisely guideline
- Economic Impact
- ENA
- Laboratory Ordering
Antinuclear antibody (ANA) and extractable nuclear antigen (ENA) antibody testing are commonly used for screening and diagnosis of autoimmune connective tissue diseases (AI-CTDs). Guidelines recommend only ordering ANA sub-serologies, such as ENA testing, after a positive ANA and clinical suspicion of immune-mediate disease.1 However, ENA testing is often ordered simultaneously with ANA testing or in the setting of a negative ANA.2 Implementing autoantibody testing algorithms that order sub-serologies only with a positive ANA has resulted in cost savings.3,4 This study seeks to assess relative timing and results of ANA and ENA laboratory testing in association with subsequently diagnosed AI-CTDs. The goal is to determine potential cost savings of ordering ANA alone as an initial screening test without compromising patient care outcomes.
Methods
This retrospective cohort study was approved by Marshfield Clinic Research Institute institutional review board with waiver of informed consent. Patient data included demographic information, laboratory testing timing (ANA and ENA ordered alone, simultaneously, or sequentially), and laboratory test results. Provider data included specialty and provider type (physician vs allied provider). Data were retrieved from Marshfield Clinic Health System (MCHS) electronic medical records. Adult patients (≥18 years) seen at MCHS facilities from February 2002 to November 2020 with a human epithelial cell tumor (HEp-2) ANA or ENA testing during the study period were included. Titers for ANA of <1:80 were considered negative, while ENA subcomponent positivity on enzyme-linked immunosorbent assay (ELISA) was determined using the laboratory’s reference ranges. Presence of AI-CTD was determined using ICD-9 and ICD-10 with a relevant diagnosis defined as two diagnosis codes within 1 year of test date. Cost analysis was determined by MCHS 2020 testing costs.
The composition of patients tested, test results, timing of laboratory orders, and patients diagnosed with an AI-CTD within 1 year of testing were reported as counts and percentages. Test results were cross-tabulated for ANA and ENA tests and stratified by the timing of laboratory orders. Potential cost savings associated with ENA laboratory tests when ordered outside of guidelines were calculated. Patient demographics and AI-CTD diagnoses were described using means and standard deviation (SD) for continuous variables and counts and percentages for dichotomous variables. Patient characteristics by test composition were compared using ANOVA or Fisher’s exact test. Frequency of patients with simultaneous tests ordered and with positive tests were compared across provider specialty and type, with differences compared using chi-squared test.
Results
There were 58,627 unique patients with ANA or ENA lab testing meeting inclusion criteria. The mean (SD) age at first laboratory testing was 48.7 (19.0±) years, and 39,155 (66.8%) were women. There were 44,422 patients who had ANA testing ordered with 17,102 (38.5%) reported positive, and 28,607 who had ENA testing ordered with 801 (2.8%) reported positive. Simultaneous ANA and ENA testing occurred in 4,873 (8.3%) patients, and ENA only was ordered in 14,205 (24.2%) patients.
When ANA was negative, ENA positivity rate was 2.0% (124/6193) regardless of order sequence (Table 1). Of these, 23 (18.5%) were eventually diagnosed with one or more AI-CTD (Table 2). Positive ENA sub-serologies are shown in Table 3. The rate of a negative ANA, positive ENA, and subsequently diagnosed AI-CTD was 0.37% (23/6193).
Distribution of laboratory results for ANA and ENA testing
AI-CTD for 124 patients with negative ANA and positive ENA
Positive ENA subcomponents for 124 patients with negative ANA and positive ENA
Non-indicated ENA testing with negative ANA screening test occurred in 6069 patients, with a potential cost savings of $645,741.60. Non-indicated testing of no ANA test ordered and negative ENA occurred in 13,919 patients, with a potential cost savings of $1,480,981.60. Non-indicated testing with a negative ANA and positive ENA occurred in 124 patients, and only 23 patients had a subsequently diagnosed AI-CTD. The cost savings of those 101 unnecessary positive ENA tests was $166,528.80. Avoiding simultaneous ANA and ENA ordering could have resulted in a potential cost saving of $2,293,251.80 over 20,112 unique patients.
Dermatology, rheumatology, and primary care providers simultaneously ordered ANA and ENA tests 32.9%, 33%, and 1.3% of the time, respectively (odds ratio 1.37 [95% CI: 1.36-1.38]). Of these orders, ANA positivity rate was 40.7%, 47.1%, and 32.8% for dermatology, rheumatology, and primary care, respectively, and ENA positivity rate was 5.4%, 4.3%, and 1.6% for dermatology, rheumatology, and primary care, respectively. The odds ratio for an ANA and ENA test being positive for a subspecialty provider (dermatology and rheumatology) than for primary care was 1.14 (95% CI: 1.12-1.15) and 1.03 (95% CI: 1.02-1.03), respectively (Table 4). Physicians and non-physicians ordered ANA and ENA simultaneously 9.8% and 2% of the time, respectively.
Order characteristics by provider department
Discussion
This study identifies significant practice gaps in following guideline-based ANA and ENA ordering recommendations, especially by dermatologists and rheumatologists. Tests for ANA and ENA were ordered simultaneously 8.3% of the time, and ENA alone was ordered 24.2% of the time. With guidelines to avoid ANA sub-serology testing without a positive ANA and clinical suspicion of AI-CTD, this non-sequential ANA and ENA ordering strategy resulted in significant health care costs of $2,293,251.80 over 20,112 unique patients. This study found a very low rate of relevant ENA positive test results in patients with a negative ANA (0.37%), consistent with other reports demonstrating a relevant anti-ENA positivity rate of 0.64% (3/468) in patients with negative screening HEp-2 ANA tests.7
Ordering providers may not be aware of the Choosing Wisely guideline to order ENA only after ANA screening and significant concern for AI-CTDs. They may desire the most expeditious means of rendering a final diagnosis, they may wish to avoid patient inconvenience of subsequent blood draws, or they may be unaware of the high costs associated with simultaneous ANA and ENA testing rather than sequential testing.
Educating providers, particularly dermatologists and rheumatologists, about guidelines and cost savings associated with sequential ordering of ANA and ENA may help change behavior. Most conditions under investigation with ANA and ENA are not acute, with minimal harm to delaying ENA testing until the ANA result is known. To minimize patient inconvenience and costs of additional blood draws, clinicians can order an extra vial of blood to be placed on hold. If ANA screening results are positive, the laboratory can then run the ENA test on the extra blood. Cost savings can occur by focusing the subcomponent screening to just the subcomponent antigen of clinical interest, rather than requesting the full ENA panel. For rare times when ANA screening is negative but ENA is positive with a subsequent AI-CTD diagnosed, we encourage clinical suspicion of the AI-CTD to guide testing strategy and avoid using the ENA panel as a screening instrument.
This study is limited by its single-center and retrospective design. Because AI-CTD diagnosis was based on MCHS coding data, we cannot verify diagnosis accuracy. Cost analysis did not consider costs of delayed AI-CTD diagnosis if ENA had not been ordered in the very small percentage of ANA negative, ENA positive patients with eventual AI-CTD diagnoses. Finally, patients may have received preceding or subsequent care and laboratory testing outside of this single health system that was not captured in our data.
In conclusion, many dermatologists and rheumatologists do not follow guidelines to sequentially order ANA and ENA testing on patients with suspected AI-CTDs. Significant savings in health care spending without compromise to quality of care can be achieved if ANA testing is performed first, followed by ENA testing when suspecting AI-CTD in patients.
Footnotes
Disclosures: The research was supported by a Resident Research grant from the Marshfield Clinic’s Division of Education. The authors have declared no conflicts of interest related to this work.
Ethics statement: This study and its protocol were approved by the institutional review board at the Marshfield Clinic Health System with waiver of informed consent. The authors attest that the study was performed in accordance with the protocol. There was no commercial support for this study.
- Received June 5, 2024.
- Accepted August 1, 2024.
References
- 1.↵Yazdany J, Schmajuk G, Robbins M, American College of Rheumatology Core Membership Group. Choosing wisely: The American College of Rheumatology’s top 5 list of things physicians and patients should question. Arthritis Care Res (Hoboken). 2013;65(3):329-339. doi:10.1002/acr.21930.
- 2.↵Davis LA, Goldstein B, Tran V, Applying Choosing Wisely: Antinuclear Antibody (ANA) and Sub-Serology Testing in a Safety Net Hospital System. Open Rheumatol J. 2015;9(1):82-87. doi:10.2174/1874312901409010082.
- 3.↵Bonaguri C, Melegari A, Ballabio A, Italian multicentre study for application of a diagnostic algorithm in autoantibody testing for autoimmune rheumatic disease: Conclusive results. Autoimmun Rev. 2011;11(1):1-5. doi:10.1016/j.autrev.2011.06.006.
- 4.↵Man A, Shojania K, Phoon C, An evaluation of autoimmune antibody testing patterns in a Canadian health region and an evaluation of a laboratory algorithm aimed at reducing unnecessary testing. Clin Rheumatol. 2013;32(5):601-608. doi:10.1007/s10067-012-2141-y.
- 5.Maddison PJ, Provost TT, Reichlin M. Serological findings in patients with “ANA-negative” systemic lupus erythematosus. Medicine (Baltimore). 1981;60(2):87-94. doi:10.1097/00005792-198103000-00002.
- 6.Solomon DH, Kavanaugh AJ, Schur PH; American College of Rheumatology Ad Hoc Committee on Immunologic Testing Guidelines. Evidence-based guidelines for the use of immunologic tests: Antinuclear antibody testing. Arthritis Care Res. 2002;47(4):434-444. doi:10.1002/art.10561.
- 7.↵Thomson KF, Murphy A, Goodfield MJ, Misbah SA. Is it useful to test for antibodies to extractable nuclear antigens in the presence of a negative antinuclear antibody on Hep-2 cells? J Clin Pathol. 2001;54(5):413. doi:10.1136/jcp.54.5.413.
