• Center on Health Equity & Access
  • Clinical
  • Health Care Cost
  • Health Care Delivery
  • Insurance
  • Policy
  • Technology
  • Value-Based Care

Management of Procalcitonin Test Overuse in an Emergency Department Through a Computer Algorithm

Publication
Article
The American Journal of Managed CareSpecial Issue: Health IT
Volume 30
Issue SP 6
Pages: SP464-SP467

Procalcitonin test demand from the emergency department is growing, necessitating the implementation of strategies to address overuse. Successful interventions must be based on information technology.

ABSTRACT

Objectives: To show the procalcitonin (PCT) test demand from an emergency department (ED) over several years, to decrease PCT measurement via a computerized algorithm based on C-reactive protein (CRP) value, and to evaluate the subsequent economic savings.

Study Design: A cross-sectional study was performed from January 1, 2018, to May 31, 2019, to evaluate an intervention to avoid PCT measurement in the ED of Hospital Universitario San Juan in Alicante in Spain, when CRP values are low.

Methods: A PCT result of at least 1.5 ng/mL was agreed upon with ED providers in our study as the value for clinical decision-making, with values less than 1.5 ng/mL considered negative. We retrospectively reviewed all PCT and CRP values for ED patients and calculated the diagnostic indicators for PCT at 4 different CRP cutoffs using the PCT quantification as the gold standard. From July 1, 2019, to April 30, 2021, the agreed-upon strategy was implemented, and we counted the PCT tests avoided and calculated the savings.

Results: PCT was not measured when CRP values were less than the selected CRP cutoff of 0.8 mg/dL, at which false-negative results were 1% and the 99th percentile of PCT was 1.5 ng/mL. In the postintervention period, 1091 PCT values were not measured and $11,553.69 was saved.

Conclusions: An intervention to decrease PCT measurement in the ED designed by the clinical laboratory staff in consensus with requesting clinicians and based on CRP values decreased PCT testing and generated significant economic savings.

Am J Manag Care. 2024;30(Spec Issue No. 6):SP464-SP467. https://doi.org/10.37765/ajmc.2024.89554

_____

Takeaway Points

  • Sepsis is a major health problem. Early detection of sepsis and timely administration of appropriate antibiotics are important for improving the outcomes of affected patients.
  • Procalcitonin (PCT) test demand in our emergency department has increased significantly in recent years.
  • Interventions to address PCT test overuse should be based on information technology.
  • An intervention to decrease PCT measurement in the emergency department based on C-reactive protein values decreased PCT testing and generated significant economic savings.

_____

Sepsis is a major health problem estimated to affect more than 30 million individuals worldwide each year and cause 6 million deaths annually.1 In addition to short-term mortality, patients with sepsis can experience numerous long-term complications that impact quality of life. Early detection of sepsis and timely administration of appropriate antibiotics are important for improving the outcomes of these patients. However, the initial signs and symptoms of sepsis are often nonspecific, generally leading to a delay in diagnosis. This scenario is common in patients with septic shock in the emergency department (ED), who frequently have vague, nonspecific symptoms of infection, and a delay in the diagnosis of sepsis could delay the administration of antibiotics.2

Diagnosis is challenging because there is not yet a sufficiently sensitive and specific diagnostic test for sepsis. In the early 1990s, with the discovery that procalcitonin (PCT) concentrations were correlated with the severity of microbial infection,3 it was hypothesized that the PCT level could become a valid marker for identifying patients with sepsis. As time passed, PCT began to be considered a suboptimal diagnostic/prognostic marker for sepsis,4 such that unrestricted PCT testing is acceptable only in intensive care units, when aiding in the decision to continue or discontinue antibiotics, and in pediatric wards. In other settings, the use of PCT testing must be protocol guided. In the undifferentiated ED population, the benefit of PCT test use is unclear.5

A pilot study in Spain that evaluated the appropriateness of PCT test use in an ED setting through comparison of geographical areas identified a high and variable PCT test request rate ranging from 0.8 to 305.7 tests per 1000 ED admissions,6 resulting in significant financial expenditure. The PCT measurement rate was lower in laboratories with restricted availability of PCT testing,6 showing that it is possible to affect demand through management interventions.

The superiority of PCT compared with C-reactive protein (CRP) for the diagnosis of sepsis is clear,5 but our hypothesis was that CRP values could be used to decide whether to measure PCT, which is a more expensive test. Our objectives for this study were to evaluate the evolution of PCT test requests from the ED over several years, to investigate how potential overuse can be corrected by means of a computerized algorithm based on CRP values, and to quantify potential subsequent economic savings.

METHODS

Setting and Patients

The clinical laboratory is located in a 370-bed suburban university community hospital, Hospital Universitario de San Juan, that serves the population of the area served by the Departamento de Salud de Alicante - Sant Joan d’Alacant in Spain (234,551 individuals). The stat laboratory is independent but located in the main laboratory, and it receives samples from inpatients and ED patients. Laboratory requests are generated online, and the reports are sent electronically from the laboratory information system (iGestlab) to the patient’s electronic health record. ED care providers can automatically review the report via the intranet.

Study Design

Preintervention study. A cross-sectional study was performed from January 1, 2018, to May 31,2019, to design an intervention to decrease PCT measurement in ED patients based on their CRP values.

In a meeting with ED physicians, a PCT level of at least 1.5 ng/mL was agreed upon as the clinical decision-making value. We retrospectively obtained all the PCT and CRP values in ED patients with laboratory requests for both markers from the laboratory information system. For diagnostic indicators, we calculated the sensitivity and specificity, positive predictive value and negative predictive value, positive likelihood ratio and negative likelihood ratio, and 99th percentile (p99) using PCT as the gold standard and considered PCT values less than 1.5 ng/mL negative. The calculations were done for 4 different CRP cutoffs (0.5, 0.8, 1.0, and 2.0 mg/dL). Once we selected the CRP cutoff, we studied the potential number of PCT tests that could have been avoided and the consequent savings in reagents.

Strategy design and agreement. A second meeting was held between laboratory staff and ED practitioners to design an intervention to manage inappropriate requests for PCT testing based on the results of the preintervention study.7 A flowchart or algorithm to be used in daily practice to decrease PCT measurement based on preintervention study results was proposed and agreed upon (Figure 1).

Intervention study. From July 1, 2019, to April 30, 2021, we applied the agreed-upon algorithm based on CRP as a frontline screening tool to decide whether to measure PCT in ED patients. We counted the number of PCT tests not measured and calculated the reagent cost savings, taking into account the nonmeasured PCT, the nonrequested but measured CRP, and the price of each test ($10.59 for PCT and $0.87 for CRP).

Laboratory Methods

CRP was measured using Dimension RxL Max (Siemens Healthineers). The within-lab coefficients of variation for CRP were 3.8% at 0.42 mg/dL and 2.3% at 5.7 mg/dL. PCT was measured using AQT90 FLEX (Radiometer Medical ApS), and the within-lab coefficients of variation for PCT were 2.0% at 0.30 ng/mL and 1.3% at 2 ng/mL. CRP and PCT in both analyzers presented acceptable internal quality control and external quality assurance.

Statistical Analysis

Data were statistically processed and analyzed using SPSS Statistics 22.0 (IBM) and Excel (Microsoft). Continuous data were tested for normal distribution using the Kolmogorov-Smirnov test. Descriptive statistics were presented as mean (SD) and percentages for continuous data and categorical data, respectively.

RESULTS

The annual increase in the ED demand for PCT testing from 2012 to 2018 is shown in Figure 2. During the preintervention study, 4384 ED patients had both tests—CRP and PCT—requested. The demographic characteristics and percentage of patients with PCT levels less than 1.5 ng/mL are shown in Table 1.

Table 2 shows the diagnostic indicators and hypothetical savings for the 4 different CRP cutoffs. As observed, more PCT tests could have been avoided, resulting in less reagent expenditure, by using 1 mg/dL or even 2 mg/dL as CRP cutoffs (Table 2, columns 3 and 4), but at the expense of having more false-negative results.

A CRP cutoff value of 0.8 mg/dL (Table 2, column 2) was selected, which resulted in 1% false-negative results, and PCT p99 was 1.5 ng/mL, which was the agreed-upon value chosen by ED practitioners as the appropriate cutoff for clinical decision-making. A total of 1045 PCT tests (23.8%) could have been avoided during this period, with hypothetical reagent savings of $11,066.55.

Figure 1 shows our proposed and agreed-upon flowchart or algorithm to be used in daily practice to decrease PCT measurement. When PCT testing is requested, we first measure CRP. If that CRP level is less than 0.8 mg/dL, PCT is not measured and—based on p99 and false-negative results at this CRP cutoff value—the following comment is automatically generated in the laboratory report: “Based on CRP measurement, with a probability of 99%, PCT result is below 1.5 ng/mL.”

During the 22-month intervention period, 4557 PCT tests were requested but 1091 (23.9%) were not measured because the patient’s CRP level was less than 0.8 mg/dL. Ninety-eight CRP tests were automatically registered and measured because they were not requested. The savings in reagent expenses were $11,553.69.

DISCUSSION

In recent years, there has been a growing demand for PCT tests from the ED, leading us to develop and execute a management intervention in collaboration with clinicians and a laboratory pathologist. This intervention, grounded in performance validation, aims to restrict the overutilization of PCT tests, employing a CRP cutoff value of 0.8 mg/dL.

It is known that PCT and CRP tests do not have the same clinical utility and that a CRP cutoff of 0.8 mg/dL performed poorly for the prediction of positive bacterial culture,8 but our data prove this cutoff’s utility as a CRP threshold for measuring PCT in the ED setting, where PCT testing can be overrequested. The agreed-upon PCT level of 1.5 ng/mL remained a very conservative cutoff because a PCT level of at least 2 ng/mL has been reported as a highly accurate biomarker for differentiating sepsis from other noninfectious triggers when used in conjunction with clinical data.9 This cutoff is considered efficient for categorizing patients in the ED with severe sepsis, septic shock, or bacteremia.10 However, our results show that values of 0.8 mg/dL and 1.5 ng/mL for CRP and PCT, respectively, are appropriate cutoffs for the proposed algorithm to limit inappropriate PCT measurement in the ED.

In a previous publication, we showed progressively increasing use of CRP testing in the ED11; we now show that PCT test use in the ED has also been continuously increasing. PCT testing is far more expensive than CRP testing, its usefulness in the ED is still unproven,4 and its use is controversial12 because the utility of this test is limited by the clinical setting and patient population.13 CRP test demand doubled in 3 years, and an educational intervention based on a request protocol designed and agreed upon by emergency, laboratory, rheumatology, pediatric, and infectious disease clinicians did not work.11 The decline in CRP test requests lasted just 5 months, despite the monthly delivery of intervention key performance indicator results to every clinician.11

The success of our PCT testing intervention has shown lasting efficacy more than a year after establishment, demonstrating the superiority of interventions based on information technology compared with simple educational interventions.

Limitations

The study had several limitations. First, despite our interventions, PCT testing could still be overrequested. Second, the chosen PCT and CRP cutoffs for the algorithm design are the ones calculated based on the findings of the retrospective study with our ED patients’ results. However, it would be necessary to replicate the diagnostic performance study in every setting to determine its potential use and the appropriate PCT and CRP cutoff values. Third, results from the past 2 years are not reported because a large proportion of the ED visits were related to COVID-19, which could distort our results. Fourth, we did not show information on additional clinical outcomes or a clinical end point, such as mortality, duration of antibiotic therapy, or blood culture results. Fifth, the calculated economic reagent savings may not apply in other settings because the authors’ laboratory belongs to a public health network, where reagent prices are relatively low. Finally, other costs were not included.

CONCLUSIONS

Designed by the clinical laboratory in consensus with requesting clinicians and based on CRP values, the intervention to decrease PCT measurement in the ED decreased PCT testing and generated significant economic savings.

Author Affiliations: Clinical Laboratory (MS, ML-G, RT, EF) and Emergency Department (ED), Hospital Universitario San Juan de Alicante, Alicante, Spain; Department of Biochemistry and Molecular Pathology (ML-G) and Department of Clinical Medicine (EF), Universitas Miguel Hernández, Elche, Spain; CIBER de Epidemiología y Salud Pública (ML-G), Madrid, Spain; Department of Radiology, University of Missouri School of Medicine (CL-S), Columbia, MO.

Source of Funding: None.

Author Disclosures: The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (MS, ML-G, RT, EF, ED, CL-S); acquisition of data (ML-G, RT, EF); analysis and interpretation of data (ML-G, RT, EF); drafting of the manuscript (MS, ML-G, RT, EF, ED, CL-S); critical revision of the manuscript for important intellectual content (MS, ML-G, RT, EF, ED, CL-S); statistical analysis (ML-G, EF); provision of patients or study materials (ED); and supervision (MS).

Address Correspondence to: Maria Salinas, PhD, Hospital Universitario San Juan de Alicante, Carretera Alicante-Valencia, s/n 03550 San Juan de Alicante, Alicante, Spain. Email: salinas_mar@gva.es.

REFERENCES

1. Fleischmann C, Scherag A, Adhikari NK, et al; International Forum of Acute Care Trialists. Assessment of global incidence and mortality of hospital-treated sepsis: current estimates and limitations. Am J Respir Crit Care Med. 2016;193(3):259-272. doi:10.1164/rccm.201504-0781OC

2. Filbin MR, Lynch J, Gillingham TD, et al. Presenting symptoms independently predict mortality in septic shock: importance of a previously unmeasured confounder. Crit Care Med. 2018;46(10):1592-1599. doi:10.1097/CCM.0000000000003260

3. Assicot M, Gendrel D, Carsin H, Raymond J, Guilbaud J, Bohuon C. High serum procalcitonin concentrations in patients with sepsis and infection. Lancet. 1993;341(8844):515-518. doi:10.1016/0140-6736(93)90277-n

4. Aloisio E, Dolci A, Panteghini M. Procalcitonin: between evidence and critical issues. Clin Chim Acta. 2019;496:7-12. doi:10.1016/j.cca.2019.06.010

5. van der Does Y, Limper M, Schuit SCE, et al. Higher diagnostic accuracy and cost-effectiveness using procalcitonin in the treatment of emergency medicine patients with fever (The HiTEMP study): a multicenter randomized study. BMC Emerg Med. 2016;16:17. doi:10.1186/s12873-016-0081-6

6. Salinas M, López-Garrigós M, Flores E, Uris J, Leiva-Salinas C; Pilot Group of the Appropriate Utilization of Laboratory Tests (REDCONLAB) working group. Procalcitonin in the emergency department: a potential expensive over-request that can be modulated through institutional protocols. Am J Emerg Med. 2018;36(1):158-160. doi:10.1016/j.ajem.2017.07.033

7. Salinas M, López-Garrigós M, Flores E, et al. Managing inappropriate requests of laboratory tests: from detection to monitoring. Am J Manag Care. 2016;22(9):e311-e316.

8. Lin CT, Lu JJ, Chen YC, Kok VC, Horng JT. Diagnostic value of serum procalcitonin, lactate, and high-sensitivity C-reactive protein for predicting bacteremia in adult patients in the emergency department. PeerJ. 2017;5:e4094. doi:10.7717/peerj.4094

9. Sinha M, Desai S, Mantri S, Kulkarni A. Procalcitonin as an adjunctive biomarker in sepsis. Indian J Anaesth. 2011;55(3):266-270. doi:10.4103/0019-5049.82676

10. Wiwatcharagoses K, Kingnakom A. Procalcitonin under investigation as a means of detecting severe sepsis, septic shock and bacteremia at emergency department, Rajavithi Hospital. J Med Assoc Thai. 2016;99(suppl 2):S63-S68.

11. Salinas M, López-Garrigós M, Gutiérrez M, Lugo J, Sirvent JV, Uris J. Achieving continuous improvement in laboratory organization through performance measurements: a seven-year experience. Clin Chem Lab Med. 2010;48(1):57-61. doi:10.1515/CCLM.2010.003

12. Yurttutan Uyar N, Sayar AK, Kocagöz AS, et al. Sepsis biomarkers for early diagnosis of bacteremia in emergency department. J Infect Dev Ctries. 2023;17(6):832-839. doi:10.3855/jidc.17221

13. Cleland DA, Eranki AP. Procalcitonin. In: StatPearls (Internet). StatPearls Publishing; 2024. Updated April 23, 2023. Accessed April 18, 2024. https://www.ncbi.nlm.nih.gov/books/NBK539794/

Related Videos
Milind Desai, MD
Masanori Aikawa, MD
1 KOL is featured in this series.
1 KOL is featured in this series.
Justin Oldham, MD, MS, an expert on IPF
Mei Wei, MD, an oncologist specializing in breast cancer at Huntsman Cancer Institute at the University of Utah.
Dr Bonnie Qin
Screenshot of an interview with Ruben Mesa, MD
Justin Oldham, MD, MS, an expert on IPF
Ruben Mesa, MD
Related Content
© 2024 MJH Life Sciences
AJMC®
All rights reserved.