Clinical Biochemistry 47 (2014) 1163–1168
Contents lists available at ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
Laboratory critical values: Automated notification supports effective clinical decision making Elisa Piva, Michela Pelloso, Laura Penello, Mario Plebani ⁎ Department of Laboratory Medicine, Padua University School of Medicine, Padua, Italy
a r t i c l e
i n f o
Article history: Received 18 April 2014 Received in revised form 15 May 2014 Accepted 19 May 2014 Available online 2 June 2014 Keywords: Critical value reporting Patient safety Effectiveness of critical value notification Clinical outcomes Clinical decision-making
a b s t r a c t Objective: Failure to adequately communicate a laboratory critical value (CV) is a potential cause of adverse events. The harmonization of CV reporting is increasingly recognized as a key issue in ensuring patient care and minimizing harm. With ongoing improvements in CV reporting, the patient's outcome should be audited to assess the effectiveness of CV notification. Design and methods: We report the data audited throughout a six month-period during which an analysis was made of CVs, and we describe the approach of clinicians and general practitioners (GPs), and their decision making following CV reporting. Results: CV notification led to a change of treatment in 98.0% of patients admitted to surgical and in 90.6% of those admitted to medical wards. Clinicians made a further evaluation of new complications in patients in 70.0% and 60.4% of cases, in surgical and medical wards respectively. In more than 40.0% of cases, CVs were unexpected findings. In the primary care setting, critical hyperkalemia was managed by GPs in 55% of patients, thus sparing patient's hospitalization. For all outpatients with critical INR (international normalized ratio), the GPs changed or stopped warfarin dosage. Twenty-four percent of patients were checked for an additional INR, whereas a further medical examination by a consultant in the hospital setting was requested for 5% of patients. Conclusions: The laboratory plays a key role in ensuring patient safety, especially in CV reporting. An evaluation should be made of the patient's outcome and clinical decision making in order to assess the effectiveness of the CV process. © 2014 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Introduction The original definition of “laboratory critical value” (formerly panic value), introduced by George D. Lundberg more than 40 years ago, is a laboratory result that is life-threatening unless some corrective action is undertaken, for which interventions are possible [1]. In the post-analytical phase of laboratory testing, the issue of critical values continues to be of crucial importance in safeguarding patient safety. Since Lundberg's pioneering work, many laboratories worldwide have implemented systems for notifying critical values, whereas the International Standard for laboratory accreditation, the ISO 15189:2012, has included written procedures for reporting critical values as a requirement, and related lists of quality indicators are also needed to verify the process [2,3]. In 2005, with the aim of improving upon patient safety, the Joint Commission released a laboratory accreditation program, in which the reporting of laboratory critical values became a
⁎ Corresponding author at: Dipartimento Medicina di Laboratorio, Azienda OspedalieraUniversità di Padova, Via Nicolò Giustiniani 2, 35128 Padova, Italy. Fax: +39 049 663240. E-mail address: [email protected] (M. Plebani).
National Patient Safety Goal [4]. However, the different terminologies used, and variations in practices and policies that have been described in literature, indicate the need for a more harmonized and systematic approach in notifying critical results [5]. Harmonization initiatives should take into consideration the following: a) reliable value limits should be chosen for true “life-threatening” analytes, according to the definition; b) critical values should be formulated while considering patients' characteristics (i.e. age, gender and ethnic origin); c) notification should be made with the most efficient reporting and communication tools; d) policies should be laid down to identify the person who should be in charge of notifying critical values, and the caregiver who should be responsible for receiving these values; and e) technological tools should allow the acknowledgment of the critical values, facilitate feedback and data recording, and ensure that the indicators control and monitor the critical value process. The efficacy of laboratory critical values, namely their impact on the clinician's decision-making process, including treatments and related patient outcomes, has been poorly investigated and represents the missing link in the cycle of critical result communication. The aim of the present study is therefore to assess the effectiveness of automated critical value notification on clinical decision making, related patient management issues, and outcomes.
http://dx.doi.org/10.1016/j.clinbiochem.2014.05.056 0009-9120/© 2014 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
1164
E. Piva et al. / Clinical Biochemistry 47 (2014) 1163–1168
Material and methods Study setting The University-Hospital of Padua in north-east Italy, a healthcare institution with 1600 beds and a research center of national and international relevance, provides specialty care in surgical and medical areas, including transplantation, cardiology, hepatology, gastroenterology, immunology, hematology, pediatrics, gynecology, oncology, traumatology, and orthopedics. For the year 2012 a 16.3% increase is estimated in highly complex cases, defined according to the Diagnosis Related Group system, and the ICD-9-CM International Classification of Diseases, 9th revision (available on http://www.sanita.padova.it/all/LibroBianco2013.pdf). In a primary care setting, the health organization provides care for more than 500,000 people in Padua city and the surrounding area, with a network of 200 general practitioners. Outpatients account for the vast majority, with about 16,000 people/month referring to 20 phlebotomy centers. The Department of Laboratory Medicine of the above hospital includes clinical chemistry, hematology, coagulation, autoimmune serology and immunology, emergency testing and specialties such as molecular biology, clinical proteomics, and newborn screening. Policy for communicating critical values Briefly, in the policy of the Department of Laboratory Medicine in Padua, a clear definition is given of critical values and critical tests are specified; provider responsibilities are clearly outlined, and electronic reporting is defined as the quality procedure for critical value notification and critical test reporting. The acceptable time interval between the identification and notification of critical values is considered no more than 40 min. Notification is monitored as follows: critical values are made available on the laboratory information system (LIS) (LM*X, version 22.01, TDLims, Grenoble, France); as soon as values have been validated by the clinical pathologist in charge, they are fed into to the hospital clinical information system (HIS) (e-Health Solutions Medical Software, developed by GMD, version 3.8, Noematica, Bologna, Italy). HIS generates automated notification by means of a specific text message sent to the cell phone of the referring physician (the clinician on duty, supporting for 24 h the ordering clinician) and an alert sent to the ordering clinician, resident physician or nurse in charge of patient care. The alert message, flashed on the monitor until the clinician, resident physician or a nurse in charge of notification acknowledges its receipt, is automatically arrested after 40 min. The status of successful notification is indicated on the HIS by a coding green color, whereas failure to acknowledge within the timeliness is visualized in red coding, and the word “expired” appears. In this case, the notification is communicated by telephone. In primary care, outpatients are recorded in the LIS in association with their own family doctor, the responsible physician who is available until 10 p.m. The notification is made using a text message and a cell-phone call. The traceability of any action and time registration is available on the HIS [6,7]. Patients Over a six-month time period, we investigated a total of 200 consecutive laboratory critical values, observed in 102 patients admitted to different wards: 73 to General Medicine (Female, F = 40; Male, M = 33); 8 to Geriatrics (F = 5; M = 3); 4 to Gastroenterology (F = 4), 2 to the Pulmonary Unit (F = 2), 5 to Liver transplantation (M = 5), 4 to General Surgery (F = 2; M = 2), and 6 to Orthopedics (F = 5; M = 1). The overall number of critical values in the above wards was 2.3. Patient groups were evaluated for age and sex. In the same period, in the primary care setting, we audited 105 general practitioners (GPs) for two groups of outpatients referred to the Department of Laboratory Medicine for investigations.
78 outpatients (F = 50, M = 28) reported a total of 90 critical INR (N4.5), while 27 (F = 9, M = 18) were affected by critically high potassium levels (N 6 mmol/L). In order to calculate the number of critical values, recurrent critical values were considered as separate events when they occurred 24 h apart, while critical values of different analytes were considered as more than one when they occurred simultaneously. Study design A survey was conducted to evaluate the effectiveness of critical value notification in relation to clinical decision-making and clinicians' actions. Medical doctors were asked to provide information on any medical action undertaken following the notification of critical values. The authors interviewed clinicians and residents responsible for inpatient care, and GPs for outpatients, using standard questions in order to gather standardized information, although interviewees were allowed to make individual comments. The response rate was 100%, all the questions being answered. We also interviewed the clinicians to assess how they were notified of critical values, and agreement with the specific value defining the critical limit. The list of questions is provided in the Supplementary material. The following were also investigated: disease and clinical status of patients, the event triggering the critical value, the rate of expectation for the critical value in relation to the patient's clinical status, and the need for admission of outpatients to the Emergency Room. An evaluation was also made of the length of stay (LOS) and the mortality rate in the week following the notification of the critical value. Laboratory tests were performed using the conventional methods described in the Supplementary material. Critical values were notified following the standard operating procedure [7]. The list of critical values is provided in the Supplementary material. Data collection and statistical analysis The critical values reported in the present study were obtained from reports generated by the Laboratory Information System (LIS) and fed into to a Microsoft Excel program (Microsoft® Office Excel 2003 11.5612.5606). Statistical analyses were made using Mann–Whitney Analysis of Variance (ANOVA) and Frequency table and Chi-squared tests were conducted using MedCalc software, version 11.0.1.0 (MedCalc Software, Mariakerke, Belgium). Results Inpatients The number and the relative frequency of critical values evaluated for inpatients are shown in Table 1, and the more important associated diseases are summarized in Fig. 1. Medical actions undertaken following the notification of critical values are shown in Fig. 2. As reported by physicians, the principal action taken was a change of therapy, due to the clinical need to modify dosage and/or to treat patient with new drugs. The notification of critical values led to a change of treatment in 98.0% of patients admitted to surgical wards and in 90.6% of those admitted to medical wards. Clinicians made an additional evaluation of new complications and conditions in 70.0% and 60.4% of patients, in surgical and medical wards respectively, and they took further steps in inpatient care following the critical value notification. A closer monitoring of patient's clinical condition was made in 26.0% and in 25.5% of cases in surgical and medical wards, respectively. Clinicians reported that critical values were unexpected findings in 42.3% of patients admitted to surgical and 43.0% of those admitted to medical wards. The pre-existing serious condition of patients provided a rationale for obtaining critical values in 40.3% of patients in the
E. Piva et al. / Clinical Biochemistry 47 (2014) 1163–1168
1165
Table 1 Total number of critical values, number of inpatients and relative frequency of critical values. Test
Critical values (n)
Patients (n)
Frequency (%)
Value (mean ± SD)
Potassium b2.8 mmol/L Potassium N6 mmol/L Magnesium b0.5 mmol/L Sodium N160 mmol/L Prothrombin Time, INR N4.5 Platelets b10 · 109/L Sodium b120 mmol/L Calcium N3.53 mmol/L Calcium b1.63 mmol/L Calcium, free b0.8 mmol/L Calcium, free N1.54 mmol/L Digoxin N2.8 nmol/L Hemoglobin b60 g/L Total
98 28 17 12 11 9 7 6 3 3 3 2 1 200
42 19 10 5 6 4 5 3 2 2 2 1 1 102
49 14 8.5 6 5.5 4.5 3.5 3 1.5 1.5 1.5 1 0.5 100
2.61 ± 0.2 6.7 ± 0.6 0.43 ± 0.1 170.16 ± 6.9 5.78 ± 2.9 8.44 ± 0.3 118.6 ± 1.8 3.76 ± 0.2 1.44 ± 0.2 0.77 ± 0.01 1.18 ± 0.2 4.15 ± 0.3 ND
INR, international normalized ratio.
surgical wards and in 43.0% of those in the medical wards. Critical values revealed the cause of disease in 17.3% of patients admitted to surgical and in 14.7% of those admitted to medical wards. No significant difference was found between the two inpatients' groups on comparing the frequencies of the different classifications of clinical responses. An assessment was also made of the way in which physicians were notified of critical values. More than 90.0% of the surgeons indicated that they were usually notified of critical values by HIS alert (IT notification), whereas 75.0% of clinicians received IT notification, but were also alerted on checking the patient's clinical records, by text message, doctor on call or call from the laboratory (Fig. 3). The agreement with values defining critical limits was 100%. For inpatients, a review was made of laboratory critical values and computerized records. The most frequently found critical value was extremely low potassium levels, recorded in 44 inpatients (49.0%). In these patients, the three main diseases linked to critically low potassium levels were intestinal disorders with poor absorption and diarrhea (n = 10), acute and sub-acute intestinal obstruction (n = 7), and cirrhosis (n = 6). One of the patients with critically low potassium
levels died during the study period. For inpatients presenting critical values, the LOS was longer than the average mean LOS in that year (N10 against 8.1 days). In the survey, no significant difference was found between the surgical and medical groups for the total number of critical values, the mean for critical values being 4.12 for the surgical and 3.83 for the medical group (P = 0.86). The age of inpatients with critical values ranged from 35 to 95 years (mean, 76.44 years; 95% CI, 73.21 to 79.67; median 81.34 years; 25th percentile = 71.21, and 75th percentile = 87.65). No statistically significant correlation was found between age and the number of critical values (P = 0.157). The rate of critical values differed according to gender, the average of critical values in females being 9.71, and those in males, 5.28 (P = 0.04). However, it was not possible to calculate the gender-related incidence. On repeatedly evaluating the presence of more than one critical value in inpatients, the frequencies were: 25 patients (24.51%), 2 critical values; 13 patients (12.75%), 3 critical values; 7 patients (6.86%), 4 critical values; 3 patients (2.94%), 5 critical values; 1 patient (0.98%), 6 critical values and 1 (0.98%), 7 critical values. The percentage of patients with two simultaneous different
Fig. 1. For inpatients presenting critical values the most important associated diseases are shown.
1166
E. Piva et al. / Clinical Biochemistry 47 (2014) 1163–1168
Fig. 2. Medical actions undertaken by clinicians following the notification of critical values for admitted patients are shown. Inpatients were subdivided into two admission groups.
types of critical value (for example, some patients with abnormal platelet and potassium, or digoxin and potassium or sodium and potassium values) was 11.76%. None of the patients had three or more different types of critical value. Outpatients For outpatients presenting critical INR, the range was from 4.5 to N 7 (mean value, 5.8; 95% CI, 5.6–6.0; median value, 5.6; 25th percentile 4.9; 75th percentile N7). The age ranged from 39 to 94 years (mean, 80.6 years; 95% CI, 77.8-83.5; median value, 84 years; 25th percentile, 76; 75th percentile, 90). As reported by GPs, consequently to critical INR for all patients the change or stop in warfarin dosage was the main clinical action. Further, additional decisions were also undertaken for some of them: 19 (24%) were checked for an additional INR, which in all cases confirmed the previous critical result, while a medical examination was requested by a consultant for 4 patients (5%) in the hospital setting. No patients had bleeding and none were referred for hospitalization by the GPs. In the second group of outpatients the mean value for critically high plasma potassium was 6.72 mmol/L (95% CI from 6.55 to 6.90), and the median value, 6.80 (25th percentile 6.30; 75th percentile 7.02). The age of outpatients with critical potassium levels ranged from 52 to 92 years (mean, 77.4 years; 95% CI 74.8–86.0; median 79 years; 25th percentile,
71.0; 75th percentile, 86.0). Information collected at phone interviews with GPs showed that for 10 outpatients (50%) with essential hypertension and heart failure, the event triggering critical hyperkalemia was related to the use of drugs interfering with potassium homeostasis, such as angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs), and aldosterone receptor antagonists. In 5 outpatients with chronic kidney disease, hyperkalemia was related to poor compliance, and dialysis was undertaken. For the other 5, the hyperkalemic event was related to cancer and other comorbidities. In 65% of the outpatients (n = 13), critically high potassium levels were considered unexpected finding by the GPs; all patients were treated within 4 h of notification, and 9 patients (45%) were admitted to the hospital for additional treatment. Discussion Quality in laboratory testing includes all aspects of the so-called “Brain-to-brain loop”, from the “pre-pre-analytical” phase (“Right test choice at the Right time on the Right patient”) through analytical steps (“Right results in the Right forms”) to the “post-post-analytical” phase (“Right interpretation, at the Right time with the Right advice as to what to do next with the result”) [8,9]. A body of evidence demonstrates that the initial and final steps of the total testing process (TTP) are much more vulnerable to errors
Fig. 3. Notifications of the critical values related to inpatients in surgical and medical wards.
E. Piva et al. / Clinical Biochemistry 47 (2014) 1163–1168
than the analytical phase, many diagnostic errors being associated with this type of problem, and a small percentage of testing-related errors can translate into significant harm [10–13]. Therefore clinical laboratories bear an increasing responsibility in controlling pre-pre- and postpost-analytical processes to improve upon patient safety [14]. Despite the importance of critical values in patient care worldwide, poor harmonization and great variability of policies and procedures persist. The main items in critical value harmonization should be the definition and management of procedures and practices related to critical values that play a key role in assuring quality and safety in laboratory testing [15]. Although the concept of “critical values” was defined 40 years ago, it has remained unchanged. Patient safety is still at risk because of variations in terminology used, doubts concerning analytes to be included in critical value lists, cut-off values, and disparity in the practices among different laboratories and countries, particularly in pediatric setting, as stated by Gong and Adeli [16]. In harmonizing the critical value list and cut-off values, a careful consideration should be made of the following: a) cut off values must reflect a true life-threatening situation, according to Lundberg's definition; b) values should be taken into specific consideration for age, gender and ethnicity; c) test results should have a substantial effect on patient safety. The best possible way to determine appropriateness should also be established. The variability is particularly alarming in the field of pediatric care [17]. Don-Wauchope and colleagues used a survey to investigate the clinicians' agreement with laboratory-defined critical values for some important analytes, and reported that physicians had a poor understanding of the concept of critical value [18]. In two studies conducted to estimate inpatients' outcomes and mortality rates, an evaluation was made of the appropriateness of sodium and calcium cut-off values [19,20]. Without such studies, any awareness of the importance of critical values will be compromised, and efforts to harmonize practices will fail. Comparisons should be made between existing policies worldwide in the attempt to promote improvement. Key aspects to be considered are patient outcomes, follow-up and related clinical decisions (i.e. whether or not physicians respond to the notification by changing treatment) [21]. The results of the present study may add new insights in the field. The majority of critical values provided by our own laboratory have led to a change in patient-management. The principal action undertaken by clinicians has been to change therapy and consequently to treat patients with alternative drugs. But additional actions have been undertaken to improve patient care, and further evaluations with additional laboratory investigations or ECG have been required. In more than 20% of patients in surgical and medical wards, clinicians decided to undertake closer monitoring, usually checking blood pressure, breathing rate, urine output and the patient's O2 saturation using finger pulse oximetry. While acute renal failure, cancer and intestinal diseases were the three main serious conditions responsible for critical values, in only 17.3% (surgical patients) and 14.7% (medical patients) of cases did clinicians believe that the critical values depend on the patient's disease. In addition, the LOS was at least ten days, longer than the average LOS for the year, indicating more severe conditions in patients than those without critical values. In discussing these findings, it is important to bear in mind that the loop of the entire cycle of test ordering and performance will be closed only if clinicians take action. If critical values are used in this fashion, technicians and doctors working in the laboratories become responsible and are driven to improving upon the notification of critical values. Concerning this issue, the behavior of clinical staff (i.e. the clinical decision making based on critical value notification) depends on the timeliness and accuracy of reporting critical value information, which is more effective if IT information is used [7]. As reported in Fig. 3, IT notification is the most commonly used tool in our Hospital setting, above all in the surgery ward. In the clinical wards, probably due to differences in the organization of the clinicians' working hours, notification is received also by the doctor on call or the person entrusted with checking the EMR. Another
1167
essential point concerns the harmonization of critical value lists and cutoffs, which should be considered a key issue in harmonization projects. As medical actions are based upon the results of critical values for a particular analyte, the critical threshold could be deemed appropriate worldwide, and established by following robust guidelines. For example, the European Resuscitation Council Guidelines for Resuscitation recommend that emergency treatment should be undertaken where serum K+ is ≥ 6.5 mmol/L or plasma K+ is ≥6.0 mmol/L, despite ECG modifications, and these should be considered appropriate critical values [22]. Our data show that, following notifications for hyperkalemia, 55% of the outpatients were managed safely in their homes and were thus spared the unfavorable outcomes that can occur in hospitalized patients. Data indicate that hospitalization is often unnecessary, and the adverse event of drug-induced hyperkalemia can be successfully managed in a primary care setting, thus obviating hospitalization costs [23]. Regarding critical hyperkalemia, on the basis of outpatient outcomes, the appropriateness of our cut-off value seems to be optimal in safeguarding patients. In the outpatients, INR showed the highest test volume (data not shown). This may be due to the fact that outpatients taking oral anticoagulants need to be monitored. Of outpatients taking warfarin (Coumadin®) for whom INR had been proven critical, 12.5% presented more than two critical INR values, showing poor therapeutic control. The fact that no major bleeding was observed, and no hospitalization was indicated for the patients, once again indicates the appropriateness of critical INR N4.5 in sparing patients from unfavorable outcomes. Therefore, harmonization projects should be encouraged, and the cut off for all potentially life-threatening critical values identified. Among clinicians, critical values were “unexpected” findings for 43% of the patients as they were for 65% of patients among the GPs. These data indicate a duty of care for laboratories, in both hospitals and primary care settings. To this end, the communication of critical values is still an open issue [24–26]. First, a body of evidence indicates that the physician can effectively change patient management and second, the person notifying a critical value should have enough clinical judgment to understand whether or not a true medical emergency exists. From this viewpoint, healthcare professionals should be held responsible for communicating critical values, instead of the call centers created in order to centralize critical value reporting, particularly for call-back. While there are national differences regarding the policies, it should be stressed that critical value communication needs to become a part of interpretive laboratory medicine [27]. In some clinical cases, especially where outpatients are concerned, for GPs that need to interpret the critical value correctly, any additional information or suggestions should be provided as interpretive advice. In TTP, the notification of the critical values falls directly under the responsibility of laboratories that must be aware that they are providing essential information to clinicians. In fact, in critical value reporting results must be interpreted correctly, as is done in interpretive laboratory medicine. The auditing of patients' outcomes has demonstrated that critical values reporting should be considered intrinsic to healthcare excellence, as it leads to changes in the clinical decision-making and therapeutic process. The impact of critical values on medical decisions points to the need for harmonization projects, whereas the IT tools available to hospitals should optimize CV reporting. Therefore notification, follow-up and documentation of critical values are still essential quality indicators in laboratory medicine for assessing patient care and safety. In conclusion, on considering the entire process of caring for patients, “life-threatening” critical values can be harmful unless medical action is taken. IT notification is an efficient way of communicating with a responsible caregiver, who can initiate treatment promptly, thus closing the loop on CV reporting. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.clinbiochem.2014.05.056.
1168
E. Piva et al. / Clinical Biochemistry 47 (2014) 1163–1168
References [1] Lundberg GD. When to panic over abnormal values. MLO Med Lab Obs 1972;4:47–54. [2] ISO 15189. Medical laboratories — requirements for quality and competence; 2012. [3] Plebani M, Chiozza ML, Sciacovelli L. Towards harmonization of quality indicators in laboratory medicine. Clin Chem Lab Med 2013;51:187–95. [4] The Joint Commission. National Patient Safety Goals. Last access on February 1st, 2014. Available at: http://www.jointcommision.org/assets/1/6/LAB_NPSG_2014. pdf; 2014. [5] Piva E, Plebani M. From “panic” to “critical” values: which path toward harmonization? Clin Chem Lab Med 2013;51:2069–71. [6] Hawkins RC. Laboratory turnaround time. Clin Biochem Rev 2007;28:179–94. [7] Piva E, Sciacovelli L, Zaninotto, Laposata M, Plebani M. Evaluation of effectiveness of a computerized notification system for reporting critical values. Am J Clin Pathol 2009;131:432–41. [8] Plebani M, Laposata M, Lundberg GD. The brain-to-brain loop concept for laboratory testing 40 years after its introduction. Am J Clin Pathol 2011;136:829–33. [9] Tate JR, Johnson R, Barth J, Panteghini M. Harmonization of laboratory testing — current achievements and future strategies. Clin Chim Acta 2013;31 [Epub ahead of print]. [10] Plebani M. Errors in clinical laboratories or errors in laboratory medicine? Clin Chem Lab Med 2006;44:750–9. [11] Plebani M. Exploring the iceberg of errors in laboratory medicine. Clin Chim Acta 2009;404:16–23. [12] Carraro P, Plebani M. Errors in a stat laboratory: types and frequencies 10 years later. Clin Chem 2007;53:1338–42. [13] Epner PL, Gans JE, Graber ML. When diagnostic testing leads to harm: a new outcomes-based approach for laboratory medicine. BMJ Qual Saf 2013;22(Suppl. 2): ii6–ii10. [14] Plebani M. The detection and prevention of errors in laboratory medicine. Ann Clin Biochem 2010;47:101–10. [15] Campbell CA, Horvath AR. Harmonization of critical result management in laboratory medicine. Clin Chim Acta 2013;16 [Epub ahead of print].
[16] Gong Y, Adeli K, CSCC Pediatric Focus Group. A national survey on pediatric critical values used in clinical laboratories across Canada. Clin Biochem 2009;42:1610–5. [17] Geaghan SM. Critical values for the maternal-fetal unit, fetus, infant, child and adolescent: bilirubin reporting practice in North American Children's Hospitals as a paradigm for critical value reporting assessment. Clin Biochem 2011;44:483–4. [18] Don-Wauchope AC, Chetty VT. Laboratory defined critical value limits: how do hospital physicians perceive laboratory based critical values? Clin Biochem 2009;42:766–70. [19] Howanitz JH, Howanitz PJ. Evaluation of total serum calcium critical values. Arch Pathol Lab Med 2006;130:828–30. [20] Howanitz JH, Howanitz PJ. Evaluation of serum and whole blood sodium critical values. Am J Clin Pathol 2007;127:56–9. [21] Piva E, Sciacovelli L, Laposata M, Plebani M. Assessment of critical values policies in Italian institutions: comparison with the US situation. Clin Chem Lab Med 2010;48:461–8. [22] Soar J, Perkins GD, Abbas G, Alfonzo A, Barelli A, Bierens JJ. European Resuscitation Council Guidelines for Resuscitation 2010 Section 8. Cardiac arrest in special circumstances: electrolyte abnormalities, poisoning, drowning, accidental hypothermia, hyperthermia, asthma, anaphylaxis, cardiac surgery, trauma, pregnancy, electrocution. Resuscitation 2010;81:1400–33. [23] Khanna A, White WB. The management of hyperkalemia in patients with cardiovascular disease. Am J Med 2009;122:215–21. [24] Liebow EB, Derzon JH, Fontanesi J, Favoretto AM, Baetz RA, Shaw C, et al. Effectiveness of automated notification and customer service call centers for timely and accurate reporting of critical values: a laboratory medicine best practices systematic review and meta-analysis. Clin Biochem 2012;45:979–87. [25] Piva E, Plebani M. Critical results communication: still an open issue. Clin Biochem 2013;46:184. [26] Derzon JH, Liebow EB. Critical result communication response to Piva and Plebani: best practices for communicating critical values. Clin Biochem 2013;46:185–6. [27] [MLO editor]Bersch C, Hale KN. Critical values launch interpretive lab medicine and clinical pathology. MLO Med Lab Obs 2009;41:18–9.
Contents lists available at ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
Laboratory critical values: Automated notification supports effective clinical decision making Elisa Piva, Michela Pelloso, Laura Penello, Mario Plebani ⁎ Department of Laboratory Medicine, Padua University School of Medicine, Padua, Italy
a r t i c l e
i n f o
Article history: Received 18 April 2014 Received in revised form 15 May 2014 Accepted 19 May 2014 Available online 2 June 2014 Keywords: Critical value reporting Patient safety Effectiveness of critical value notification Clinical outcomes Clinical decision-making
a b s t r a c t Objective: Failure to adequately communicate a laboratory critical value (CV) is a potential cause of adverse events. The harmonization of CV reporting is increasingly recognized as a key issue in ensuring patient care and minimizing harm. With ongoing improvements in CV reporting, the patient's outcome should be audited to assess the effectiveness of CV notification. Design and methods: We report the data audited throughout a six month-period during which an analysis was made of CVs, and we describe the approach of clinicians and general practitioners (GPs), and their decision making following CV reporting. Results: CV notification led to a change of treatment in 98.0% of patients admitted to surgical and in 90.6% of those admitted to medical wards. Clinicians made a further evaluation of new complications in patients in 70.0% and 60.4% of cases, in surgical and medical wards respectively. In more than 40.0% of cases, CVs were unexpected findings. In the primary care setting, critical hyperkalemia was managed by GPs in 55% of patients, thus sparing patient's hospitalization. For all outpatients with critical INR (international normalized ratio), the GPs changed or stopped warfarin dosage. Twenty-four percent of patients were checked for an additional INR, whereas a further medical examination by a consultant in the hospital setting was requested for 5% of patients. Conclusions: The laboratory plays a key role in ensuring patient safety, especially in CV reporting. An evaluation should be made of the patient's outcome and clinical decision making in order to assess the effectiveness of the CV process. © 2014 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Introduction The original definition of “laboratory critical value” (formerly panic value), introduced by George D. Lundberg more than 40 years ago, is a laboratory result that is life-threatening unless some corrective action is undertaken, for which interventions are possible [1]. In the post-analytical phase of laboratory testing, the issue of critical values continues to be of crucial importance in safeguarding patient safety. Since Lundberg's pioneering work, many laboratories worldwide have implemented systems for notifying critical values, whereas the International Standard for laboratory accreditation, the ISO 15189:2012, has included written procedures for reporting critical values as a requirement, and related lists of quality indicators are also needed to verify the process [2,3]. In 2005, with the aim of improving upon patient safety, the Joint Commission released a laboratory accreditation program, in which the reporting of laboratory critical values became a
⁎ Corresponding author at: Dipartimento Medicina di Laboratorio, Azienda OspedalieraUniversità di Padova, Via Nicolò Giustiniani 2, 35128 Padova, Italy. Fax: +39 049 663240. E-mail address: [email protected] (M. Plebani).
National Patient Safety Goal [4]. However, the different terminologies used, and variations in practices and policies that have been described in literature, indicate the need for a more harmonized and systematic approach in notifying critical results [5]. Harmonization initiatives should take into consideration the following: a) reliable value limits should be chosen for true “life-threatening” analytes, according to the definition; b) critical values should be formulated while considering patients' characteristics (i.e. age, gender and ethnic origin); c) notification should be made with the most efficient reporting and communication tools; d) policies should be laid down to identify the person who should be in charge of notifying critical values, and the caregiver who should be responsible for receiving these values; and e) technological tools should allow the acknowledgment of the critical values, facilitate feedback and data recording, and ensure that the indicators control and monitor the critical value process. The efficacy of laboratory critical values, namely their impact on the clinician's decision-making process, including treatments and related patient outcomes, has been poorly investigated and represents the missing link in the cycle of critical result communication. The aim of the present study is therefore to assess the effectiveness of automated critical value notification on clinical decision making, related patient management issues, and outcomes.
http://dx.doi.org/10.1016/j.clinbiochem.2014.05.056 0009-9120/© 2014 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
1164
E. Piva et al. / Clinical Biochemistry 47 (2014) 1163–1168
Material and methods Study setting The University-Hospital of Padua in north-east Italy, a healthcare institution with 1600 beds and a research center of national and international relevance, provides specialty care in surgical and medical areas, including transplantation, cardiology, hepatology, gastroenterology, immunology, hematology, pediatrics, gynecology, oncology, traumatology, and orthopedics. For the year 2012 a 16.3% increase is estimated in highly complex cases, defined according to the Diagnosis Related Group system, and the ICD-9-CM International Classification of Diseases, 9th revision (available on http://www.sanita.padova.it/all/LibroBianco2013.pdf). In a primary care setting, the health organization provides care for more than 500,000 people in Padua city and the surrounding area, with a network of 200 general practitioners. Outpatients account for the vast majority, with about 16,000 people/month referring to 20 phlebotomy centers. The Department of Laboratory Medicine of the above hospital includes clinical chemistry, hematology, coagulation, autoimmune serology and immunology, emergency testing and specialties such as molecular biology, clinical proteomics, and newborn screening. Policy for communicating critical values Briefly, in the policy of the Department of Laboratory Medicine in Padua, a clear definition is given of critical values and critical tests are specified; provider responsibilities are clearly outlined, and electronic reporting is defined as the quality procedure for critical value notification and critical test reporting. The acceptable time interval between the identification and notification of critical values is considered no more than 40 min. Notification is monitored as follows: critical values are made available on the laboratory information system (LIS) (LM*X, version 22.01, TDLims, Grenoble, France); as soon as values have been validated by the clinical pathologist in charge, they are fed into to the hospital clinical information system (HIS) (e-Health Solutions Medical Software, developed by GMD, version 3.8, Noematica, Bologna, Italy). HIS generates automated notification by means of a specific text message sent to the cell phone of the referring physician (the clinician on duty, supporting for 24 h the ordering clinician) and an alert sent to the ordering clinician, resident physician or nurse in charge of patient care. The alert message, flashed on the monitor until the clinician, resident physician or a nurse in charge of notification acknowledges its receipt, is automatically arrested after 40 min. The status of successful notification is indicated on the HIS by a coding green color, whereas failure to acknowledge within the timeliness is visualized in red coding, and the word “expired” appears. In this case, the notification is communicated by telephone. In primary care, outpatients are recorded in the LIS in association with their own family doctor, the responsible physician who is available until 10 p.m. The notification is made using a text message and a cell-phone call. The traceability of any action and time registration is available on the HIS [6,7]. Patients Over a six-month time period, we investigated a total of 200 consecutive laboratory critical values, observed in 102 patients admitted to different wards: 73 to General Medicine (Female, F = 40; Male, M = 33); 8 to Geriatrics (F = 5; M = 3); 4 to Gastroenterology (F = 4), 2 to the Pulmonary Unit (F = 2), 5 to Liver transplantation (M = 5), 4 to General Surgery (F = 2; M = 2), and 6 to Orthopedics (F = 5; M = 1). The overall number of critical values in the above wards was 2.3. Patient groups were evaluated for age and sex. In the same period, in the primary care setting, we audited 105 general practitioners (GPs) for two groups of outpatients referred to the Department of Laboratory Medicine for investigations.
78 outpatients (F = 50, M = 28) reported a total of 90 critical INR (N4.5), while 27 (F = 9, M = 18) were affected by critically high potassium levels (N 6 mmol/L). In order to calculate the number of critical values, recurrent critical values were considered as separate events when they occurred 24 h apart, while critical values of different analytes were considered as more than one when they occurred simultaneously. Study design A survey was conducted to evaluate the effectiveness of critical value notification in relation to clinical decision-making and clinicians' actions. Medical doctors were asked to provide information on any medical action undertaken following the notification of critical values. The authors interviewed clinicians and residents responsible for inpatient care, and GPs for outpatients, using standard questions in order to gather standardized information, although interviewees were allowed to make individual comments. The response rate was 100%, all the questions being answered. We also interviewed the clinicians to assess how they were notified of critical values, and agreement with the specific value defining the critical limit. The list of questions is provided in the Supplementary material. The following were also investigated: disease and clinical status of patients, the event triggering the critical value, the rate of expectation for the critical value in relation to the patient's clinical status, and the need for admission of outpatients to the Emergency Room. An evaluation was also made of the length of stay (LOS) and the mortality rate in the week following the notification of the critical value. Laboratory tests were performed using the conventional methods described in the Supplementary material. Critical values were notified following the standard operating procedure [7]. The list of critical values is provided in the Supplementary material. Data collection and statistical analysis The critical values reported in the present study were obtained from reports generated by the Laboratory Information System (LIS) and fed into to a Microsoft Excel program (Microsoft® Office Excel 2003 11.5612.5606). Statistical analyses were made using Mann–Whitney Analysis of Variance (ANOVA) and Frequency table and Chi-squared tests were conducted using MedCalc software, version 11.0.1.0 (MedCalc Software, Mariakerke, Belgium). Results Inpatients The number and the relative frequency of critical values evaluated for inpatients are shown in Table 1, and the more important associated diseases are summarized in Fig. 1. Medical actions undertaken following the notification of critical values are shown in Fig. 2. As reported by physicians, the principal action taken was a change of therapy, due to the clinical need to modify dosage and/or to treat patient with new drugs. The notification of critical values led to a change of treatment in 98.0% of patients admitted to surgical wards and in 90.6% of those admitted to medical wards. Clinicians made an additional evaluation of new complications and conditions in 70.0% and 60.4% of patients, in surgical and medical wards respectively, and they took further steps in inpatient care following the critical value notification. A closer monitoring of patient's clinical condition was made in 26.0% and in 25.5% of cases in surgical and medical wards, respectively. Clinicians reported that critical values were unexpected findings in 42.3% of patients admitted to surgical and 43.0% of those admitted to medical wards. The pre-existing serious condition of patients provided a rationale for obtaining critical values in 40.3% of patients in the
E. Piva et al. / Clinical Biochemistry 47 (2014) 1163–1168
1165
Table 1 Total number of critical values, number of inpatients and relative frequency of critical values. Test
Critical values (n)
Patients (n)
Frequency (%)
Value (mean ± SD)
Potassium b2.8 mmol/L Potassium N6 mmol/L Magnesium b0.5 mmol/L Sodium N160 mmol/L Prothrombin Time, INR N4.5 Platelets b10 · 109/L Sodium b120 mmol/L Calcium N3.53 mmol/L Calcium b1.63 mmol/L Calcium, free b0.8 mmol/L Calcium, free N1.54 mmol/L Digoxin N2.8 nmol/L Hemoglobin b60 g/L Total
98 28 17 12 11 9 7 6 3 3 3 2 1 200
42 19 10 5 6 4 5 3 2 2 2 1 1 102
49 14 8.5 6 5.5 4.5 3.5 3 1.5 1.5 1.5 1 0.5 100
2.61 ± 0.2 6.7 ± 0.6 0.43 ± 0.1 170.16 ± 6.9 5.78 ± 2.9 8.44 ± 0.3 118.6 ± 1.8 3.76 ± 0.2 1.44 ± 0.2 0.77 ± 0.01 1.18 ± 0.2 4.15 ± 0.3 ND
INR, international normalized ratio.
surgical wards and in 43.0% of those in the medical wards. Critical values revealed the cause of disease in 17.3% of patients admitted to surgical and in 14.7% of those admitted to medical wards. No significant difference was found between the two inpatients' groups on comparing the frequencies of the different classifications of clinical responses. An assessment was also made of the way in which physicians were notified of critical values. More than 90.0% of the surgeons indicated that they were usually notified of critical values by HIS alert (IT notification), whereas 75.0% of clinicians received IT notification, but were also alerted on checking the patient's clinical records, by text message, doctor on call or call from the laboratory (Fig. 3). The agreement with values defining critical limits was 100%. For inpatients, a review was made of laboratory critical values and computerized records. The most frequently found critical value was extremely low potassium levels, recorded in 44 inpatients (49.0%). In these patients, the three main diseases linked to critically low potassium levels were intestinal disorders with poor absorption and diarrhea (n = 10), acute and sub-acute intestinal obstruction (n = 7), and cirrhosis (n = 6). One of the patients with critically low potassium
levels died during the study period. For inpatients presenting critical values, the LOS was longer than the average mean LOS in that year (N10 against 8.1 days). In the survey, no significant difference was found between the surgical and medical groups for the total number of critical values, the mean for critical values being 4.12 for the surgical and 3.83 for the medical group (P = 0.86). The age of inpatients with critical values ranged from 35 to 95 years (mean, 76.44 years; 95% CI, 73.21 to 79.67; median 81.34 years; 25th percentile = 71.21, and 75th percentile = 87.65). No statistically significant correlation was found between age and the number of critical values (P = 0.157). The rate of critical values differed according to gender, the average of critical values in females being 9.71, and those in males, 5.28 (P = 0.04). However, it was not possible to calculate the gender-related incidence. On repeatedly evaluating the presence of more than one critical value in inpatients, the frequencies were: 25 patients (24.51%), 2 critical values; 13 patients (12.75%), 3 critical values; 7 patients (6.86%), 4 critical values; 3 patients (2.94%), 5 critical values; 1 patient (0.98%), 6 critical values and 1 (0.98%), 7 critical values. The percentage of patients with two simultaneous different
Fig. 1. For inpatients presenting critical values the most important associated diseases are shown.
1166
E. Piva et al. / Clinical Biochemistry 47 (2014) 1163–1168
Fig. 2. Medical actions undertaken by clinicians following the notification of critical values for admitted patients are shown. Inpatients were subdivided into two admission groups.
types of critical value (for example, some patients with abnormal platelet and potassium, or digoxin and potassium or sodium and potassium values) was 11.76%. None of the patients had three or more different types of critical value. Outpatients For outpatients presenting critical INR, the range was from 4.5 to N 7 (mean value, 5.8; 95% CI, 5.6–6.0; median value, 5.6; 25th percentile 4.9; 75th percentile N7). The age ranged from 39 to 94 years (mean, 80.6 years; 95% CI, 77.8-83.5; median value, 84 years; 25th percentile, 76; 75th percentile, 90). As reported by GPs, consequently to critical INR for all patients the change or stop in warfarin dosage was the main clinical action. Further, additional decisions were also undertaken for some of them: 19 (24%) were checked for an additional INR, which in all cases confirmed the previous critical result, while a medical examination was requested by a consultant for 4 patients (5%) in the hospital setting. No patients had bleeding and none were referred for hospitalization by the GPs. In the second group of outpatients the mean value for critically high plasma potassium was 6.72 mmol/L (95% CI from 6.55 to 6.90), and the median value, 6.80 (25th percentile 6.30; 75th percentile 7.02). The age of outpatients with critical potassium levels ranged from 52 to 92 years (mean, 77.4 years; 95% CI 74.8–86.0; median 79 years; 25th percentile,
71.0; 75th percentile, 86.0). Information collected at phone interviews with GPs showed that for 10 outpatients (50%) with essential hypertension and heart failure, the event triggering critical hyperkalemia was related to the use of drugs interfering with potassium homeostasis, such as angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs), and aldosterone receptor antagonists. In 5 outpatients with chronic kidney disease, hyperkalemia was related to poor compliance, and dialysis was undertaken. For the other 5, the hyperkalemic event was related to cancer and other comorbidities. In 65% of the outpatients (n = 13), critically high potassium levels were considered unexpected finding by the GPs; all patients were treated within 4 h of notification, and 9 patients (45%) were admitted to the hospital for additional treatment. Discussion Quality in laboratory testing includes all aspects of the so-called “Brain-to-brain loop”, from the “pre-pre-analytical” phase (“Right test choice at the Right time on the Right patient”) through analytical steps (“Right results in the Right forms”) to the “post-post-analytical” phase (“Right interpretation, at the Right time with the Right advice as to what to do next with the result”) [8,9]. A body of evidence demonstrates that the initial and final steps of the total testing process (TTP) are much more vulnerable to errors
Fig. 3. Notifications of the critical values related to inpatients in surgical and medical wards.
E. Piva et al. / Clinical Biochemistry 47 (2014) 1163–1168
than the analytical phase, many diagnostic errors being associated with this type of problem, and a small percentage of testing-related errors can translate into significant harm [10–13]. Therefore clinical laboratories bear an increasing responsibility in controlling pre-pre- and postpost-analytical processes to improve upon patient safety [14]. Despite the importance of critical values in patient care worldwide, poor harmonization and great variability of policies and procedures persist. The main items in critical value harmonization should be the definition and management of procedures and practices related to critical values that play a key role in assuring quality and safety in laboratory testing [15]. Although the concept of “critical values” was defined 40 years ago, it has remained unchanged. Patient safety is still at risk because of variations in terminology used, doubts concerning analytes to be included in critical value lists, cut-off values, and disparity in the practices among different laboratories and countries, particularly in pediatric setting, as stated by Gong and Adeli [16]. In harmonizing the critical value list and cut-off values, a careful consideration should be made of the following: a) cut off values must reflect a true life-threatening situation, according to Lundberg's definition; b) values should be taken into specific consideration for age, gender and ethnicity; c) test results should have a substantial effect on patient safety. The best possible way to determine appropriateness should also be established. The variability is particularly alarming in the field of pediatric care [17]. Don-Wauchope and colleagues used a survey to investigate the clinicians' agreement with laboratory-defined critical values for some important analytes, and reported that physicians had a poor understanding of the concept of critical value [18]. In two studies conducted to estimate inpatients' outcomes and mortality rates, an evaluation was made of the appropriateness of sodium and calcium cut-off values [19,20]. Without such studies, any awareness of the importance of critical values will be compromised, and efforts to harmonize practices will fail. Comparisons should be made between existing policies worldwide in the attempt to promote improvement. Key aspects to be considered are patient outcomes, follow-up and related clinical decisions (i.e. whether or not physicians respond to the notification by changing treatment) [21]. The results of the present study may add new insights in the field. The majority of critical values provided by our own laboratory have led to a change in patient-management. The principal action undertaken by clinicians has been to change therapy and consequently to treat patients with alternative drugs. But additional actions have been undertaken to improve patient care, and further evaluations with additional laboratory investigations or ECG have been required. In more than 20% of patients in surgical and medical wards, clinicians decided to undertake closer monitoring, usually checking blood pressure, breathing rate, urine output and the patient's O2 saturation using finger pulse oximetry. While acute renal failure, cancer and intestinal diseases were the three main serious conditions responsible for critical values, in only 17.3% (surgical patients) and 14.7% (medical patients) of cases did clinicians believe that the critical values depend on the patient's disease. In addition, the LOS was at least ten days, longer than the average LOS for the year, indicating more severe conditions in patients than those without critical values. In discussing these findings, it is important to bear in mind that the loop of the entire cycle of test ordering and performance will be closed only if clinicians take action. If critical values are used in this fashion, technicians and doctors working in the laboratories become responsible and are driven to improving upon the notification of critical values. Concerning this issue, the behavior of clinical staff (i.e. the clinical decision making based on critical value notification) depends on the timeliness and accuracy of reporting critical value information, which is more effective if IT information is used [7]. As reported in Fig. 3, IT notification is the most commonly used tool in our Hospital setting, above all in the surgery ward. In the clinical wards, probably due to differences in the organization of the clinicians' working hours, notification is received also by the doctor on call or the person entrusted with checking the EMR. Another
1167
essential point concerns the harmonization of critical value lists and cutoffs, which should be considered a key issue in harmonization projects. As medical actions are based upon the results of critical values for a particular analyte, the critical threshold could be deemed appropriate worldwide, and established by following robust guidelines. For example, the European Resuscitation Council Guidelines for Resuscitation recommend that emergency treatment should be undertaken where serum K+ is ≥ 6.5 mmol/L or plasma K+ is ≥6.0 mmol/L, despite ECG modifications, and these should be considered appropriate critical values [22]. Our data show that, following notifications for hyperkalemia, 55% of the outpatients were managed safely in their homes and were thus spared the unfavorable outcomes that can occur in hospitalized patients. Data indicate that hospitalization is often unnecessary, and the adverse event of drug-induced hyperkalemia can be successfully managed in a primary care setting, thus obviating hospitalization costs [23]. Regarding critical hyperkalemia, on the basis of outpatient outcomes, the appropriateness of our cut-off value seems to be optimal in safeguarding patients. In the outpatients, INR showed the highest test volume (data not shown). This may be due to the fact that outpatients taking oral anticoagulants need to be monitored. Of outpatients taking warfarin (Coumadin®) for whom INR had been proven critical, 12.5% presented more than two critical INR values, showing poor therapeutic control. The fact that no major bleeding was observed, and no hospitalization was indicated for the patients, once again indicates the appropriateness of critical INR N4.5 in sparing patients from unfavorable outcomes. Therefore, harmonization projects should be encouraged, and the cut off for all potentially life-threatening critical values identified. Among clinicians, critical values were “unexpected” findings for 43% of the patients as they were for 65% of patients among the GPs. These data indicate a duty of care for laboratories, in both hospitals and primary care settings. To this end, the communication of critical values is still an open issue [24–26]. First, a body of evidence indicates that the physician can effectively change patient management and second, the person notifying a critical value should have enough clinical judgment to understand whether or not a true medical emergency exists. From this viewpoint, healthcare professionals should be held responsible for communicating critical values, instead of the call centers created in order to centralize critical value reporting, particularly for call-back. While there are national differences regarding the policies, it should be stressed that critical value communication needs to become a part of interpretive laboratory medicine [27]. In some clinical cases, especially where outpatients are concerned, for GPs that need to interpret the critical value correctly, any additional information or suggestions should be provided as interpretive advice. In TTP, the notification of the critical values falls directly under the responsibility of laboratories that must be aware that they are providing essential information to clinicians. In fact, in critical value reporting results must be interpreted correctly, as is done in interpretive laboratory medicine. The auditing of patients' outcomes has demonstrated that critical values reporting should be considered intrinsic to healthcare excellence, as it leads to changes in the clinical decision-making and therapeutic process. The impact of critical values on medical decisions points to the need for harmonization projects, whereas the IT tools available to hospitals should optimize CV reporting. Therefore notification, follow-up and documentation of critical values are still essential quality indicators in laboratory medicine for assessing patient care and safety. In conclusion, on considering the entire process of caring for patients, “life-threatening” critical values can be harmful unless medical action is taken. IT notification is an efficient way of communicating with a responsible caregiver, who can initiate treatment promptly, thus closing the loop on CV reporting. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.clinbiochem.2014.05.056.
1168
E. Piva et al. / Clinical Biochemistry 47 (2014) 1163–1168
References [1] Lundberg GD. When to panic over abnormal values. MLO Med Lab Obs 1972;4:47–54. [2] ISO 15189. Medical laboratories — requirements for quality and competence; 2012. [3] Plebani M, Chiozza ML, Sciacovelli L. Towards harmonization of quality indicators in laboratory medicine. Clin Chem Lab Med 2013;51:187–95. [4] The Joint Commission. National Patient Safety Goals. Last access on February 1st, 2014. Available at: http://www.jointcommision.org/assets/1/6/LAB_NPSG_2014. pdf; 2014. [5] Piva E, Plebani M. From “panic” to “critical” values: which path toward harmonization? Clin Chem Lab Med 2013;51:2069–71. [6] Hawkins RC. Laboratory turnaround time. Clin Biochem Rev 2007;28:179–94. [7] Piva E, Sciacovelli L, Zaninotto, Laposata M, Plebani M. Evaluation of effectiveness of a computerized notification system for reporting critical values. Am J Clin Pathol 2009;131:432–41. [8] Plebani M, Laposata M, Lundberg GD. The brain-to-brain loop concept for laboratory testing 40 years after its introduction. Am J Clin Pathol 2011;136:829–33. [9] Tate JR, Johnson R, Barth J, Panteghini M. Harmonization of laboratory testing — current achievements and future strategies. Clin Chim Acta 2013;31 [Epub ahead of print]. [10] Plebani M. Errors in clinical laboratories or errors in laboratory medicine? Clin Chem Lab Med 2006;44:750–9. [11] Plebani M. Exploring the iceberg of errors in laboratory medicine. Clin Chim Acta 2009;404:16–23. [12] Carraro P, Plebani M. Errors in a stat laboratory: types and frequencies 10 years later. Clin Chem 2007;53:1338–42. [13] Epner PL, Gans JE, Graber ML. When diagnostic testing leads to harm: a new outcomes-based approach for laboratory medicine. BMJ Qual Saf 2013;22(Suppl. 2): ii6–ii10. [14] Plebani M. The detection and prevention of errors in laboratory medicine. Ann Clin Biochem 2010;47:101–10. [15] Campbell CA, Horvath AR. Harmonization of critical result management in laboratory medicine. Clin Chim Acta 2013;16 [Epub ahead of print].
[16] Gong Y, Adeli K, CSCC Pediatric Focus Group. A national survey on pediatric critical values used in clinical laboratories across Canada. Clin Biochem 2009;42:1610–5. [17] Geaghan SM. Critical values for the maternal-fetal unit, fetus, infant, child and adolescent: bilirubin reporting practice in North American Children's Hospitals as a paradigm for critical value reporting assessment. Clin Biochem 2011;44:483–4. [18] Don-Wauchope AC, Chetty VT. Laboratory defined critical value limits: how do hospital physicians perceive laboratory based critical values? Clin Biochem 2009;42:766–70. [19] Howanitz JH, Howanitz PJ. Evaluation of total serum calcium critical values. Arch Pathol Lab Med 2006;130:828–30. [20] Howanitz JH, Howanitz PJ. Evaluation of serum and whole blood sodium critical values. Am J Clin Pathol 2007;127:56–9. [21] Piva E, Sciacovelli L, Laposata M, Plebani M. Assessment of critical values policies in Italian institutions: comparison with the US situation. Clin Chem Lab Med 2010;48:461–8. [22] Soar J, Perkins GD, Abbas G, Alfonzo A, Barelli A, Bierens JJ. European Resuscitation Council Guidelines for Resuscitation 2010 Section 8. Cardiac arrest in special circumstances: electrolyte abnormalities, poisoning, drowning, accidental hypothermia, hyperthermia, asthma, anaphylaxis, cardiac surgery, trauma, pregnancy, electrocution. Resuscitation 2010;81:1400–33. [23] Khanna A, White WB. The management of hyperkalemia in patients with cardiovascular disease. Am J Med 2009;122:215–21. [24] Liebow EB, Derzon JH, Fontanesi J, Favoretto AM, Baetz RA, Shaw C, et al. Effectiveness of automated notification and customer service call centers for timely and accurate reporting of critical values: a laboratory medicine best practices systematic review and meta-analysis. Clin Biochem 2012;45:979–87. [25] Piva E, Plebani M. Critical results communication: still an open issue. Clin Biochem 2013;46:184. [26] Derzon JH, Liebow EB. Critical result communication response to Piva and Plebani: best practices for communicating critical values. Clin Biochem 2013;46:185–6. [27] [MLO editor]Bersch C, Hale KN. Critical values launch interpretive lab medicine and clinical pathology. MLO Med Lab Obs 2009;41:18–9.
![[Clinical decision making algorithms].](/assets/img/hold.png)
