Assessment monitoring of laboratory critical values: a College of American Pathologists Q-Tracks study of 180 institutions.
Objective.--To measure critical laboratory value reporting in multiple institutions over time and to examine the practice patterns and demographic factors associated with sustained improvement in critical value reporting.
Design.--A longitudinal cohort study of 180 clinical laboratories that provided quarterly critical values reporting data for 2 to 16 quarters was conducted using a uniform definition of successful caregiver notification. Mixed linear model analysis of the 2001 through 2004 dataset was performed.
Results.--A decrease in total and inpatient rates of undocumented critical values per 1000 results was associated with (1) the American Association of Blood Banks inspection within the past 2 years (P = .01, for both total and inpatient rates); (2) unit secretary/clerical staff not authorized to accept inpatient critical value notification (P = .004 [total] and .001 [inpatient]); and (3) the mandatory practice of requiring notification of health care providers when handling inpatients known to have results repeatedly in the critical range (P = .01, for both total and inpatient rates). Continued participation in the Q-Tracks monitoring program was associated with significant and progressive improvement in total, inpatient, and outpatient critical value reporting (P = .02, .01, and .003, respectively).
Conclusions.--Critical value reporting improved as the duration of participation in the Q-Tracks monitoring program increased. Improved total and inpatient critical value reporting was associated with factors that may be markers for institutions with priorities of quality management and enhanced communication with responsible caregivers.
Originally defined by Lundberg, (1) a critical value represents a pathophysiologic state at such variance with normal as to be life-threatening unless something is done promptly and for which some corrective action can be taken. As part of Lundberg's procedure, he identified the term critical (panic) values for results with life-threatening abnormalities. The American Society for Clinical Pathology developed a practice parameter addressing critical value reporting in 1997. (2) Other investigators have similarly focused on practices in health care networks or standardized recommendations from other countries. (3,4) The Q-Probes studies, offered by the College of American Pathologists, has been used to examine practices and performance at multiple institutions. (5,6)
Critical values, despite their importance in clinical laboratory operations, are infrequently cited in the literature. In spite of a long history of reporting panic values and the incorporation of critical values into multiple regulations, there is little standardization of procedures. Performance benchmarks and definitions are variable or nonexistent. However, interest in critical values has intensified as various regulatory agencies have incorporated their use into standard operating recommendations. Critical values are described in federal regulations as part of the Clinical Laboratory Improvement Amendments Act of 1988. (7) A recent example of regulatory adoption is the development by the Joint Commission on Accreditation of Healthcare Organizations of a standard requiring the receiving physician or health care professional to "read-back" the critical value results with patient identifiers and to assess the critical value turnaround times. (8)
A recently described summation of Q-Probes data from 623 institutions indicates that critical value lists vary widely for routine chemistry and hematology analytes among participants. (9) Laboratory practices also varied for outpatient clinic versus inpatient reporting, with most outpatient results received by physicians' office staff. A change in therapy resulted from 65% of reported critical values, with 94.9% of physicians indicating that critical results are valuable for patient care.
To develop meaningful data and establish a performance benchmark, a critical values study with cumulative data collections was initiated as a Q-Tracks monitoring program by the College of American Pathologists in 2001. Voluntary participants in a given institution were instructed to longitudinally monitor the total number of inpatient and outpatient laboratory results that qualified as critical values. Participants also documented whether result notification occurred as well as the number of components on the critical value list. Four years of Q-Tracks data provide what we believe to be the most comprehensive continuously recorded data record of critical values. These results demonstrate performance improvement of participants when using multiple-site results as a benchmark. Additionally, innovative techniques used by the best-performing laboratories are described to assist the clinical laboratory community in refining critical value procedures for specific institutional needs.
MATERIALS AND METHODS
Study Format and Data Collection
The study was conducted according to the Q-Tracks monitor format: annual subscribers received mailed data-collection instructions on a quarterly basis. The objective of the monitor was to determine the level of successful critical value reporting in the general laboratory for inpatients and outpatients. The performance indicator was the percentage of critical values with documentation of successful caregiver notification and was calculated as:
Number of documented critical value notifications/ Total number of reportable critical value results x 100
For the purpose of this longitudinal study, the rate of undocumented critical values per 1000 results was defined as:
(1 - proportion of documented critical value notifications) x 1000
Each institution also identified (1) specific institutional practices of the personnel responsible for reporting critical values, (2) the number of analytes on the critical value list, (3) the policy regarding repeat critical values, and (4) whether confirmation was required for faxed/electronically transmitted results. These principles were adhered to throughout the study. The tracking of institution-specific subgroup information was optional and intended for the internal use of the subscribing institution only.
Because they are more consistently listed as critical values in clinical laboratories, (3) results included in this study were for general hematology, coagulation, and chemistry analytes. Clinical microbiology critical values, cardiac markers, drugs of abuse and therapeutic drug levels, urinalysis, and blood gases as well as point-of-care tests and tests performed at reference laboratories were excluded from the study because of the likely variability in the critical value lists at participating institutions resulting from variable patient profiles. Critical values from discharged patients were also excluded.
Data collection was performed quarterly, and the numbers of reportable critical values and documented critical value notifications were collected retrospectively. Participating clinical laboratories were requested to evaluate 120 inpatient and 120 outpatient reportable critical values per month. To ensure statistical validity, institutions with more than 20 inpatient and 20 outpatient reportable values per day collected data on 6 specified randomized dates per quarter. For laboratories with less than 20 inpatient and 20 outpatient reportable values per day, data were collected every day starting with the first day of the month until 120 total values were obtained. Similar methods were applied for laboratories that had fewer than 20 inpatient reportable critical values per day matched with more than 20 outpatient reportable critical values and vice versa. If a patient had multiple critical values called as a single notification, they were counted as a single reportable critical value.
A questionnaire was also provided regarding the subscribing institution's policies and practices related to notification of critical values. This questionnaire was submitted to the College of American Pathologists with the institution's first data submission. This information was updated annually, unless a change occurred before that time. Demographics were collected about the institution type, location, occupied bed size of inpatient facilities, training activities, outpatient services, administrative services such as courier services, level of service, hospital complexity, and whether the institution was a teaching hospital (Table 1). Performance summaries were provided to all annual subscribers from 2001 to 2004. In 2003 and 2004, 12 learning assessment questions were provided to the subscribers. Three questions were collected for each quarter. Best-performers data were collected after the third quarter of each year to survey responses to questions regarding procedures that may have contributed to the reporting of improved critical values (Table 2).
To ensure comparability of participant responses across institutions, several definitions were established (Table 3).
Calculations used to derive the study variables have been described. (10) Quarterly distributions of cumulative quarters for the rates of undocumented critical values per 1000 results and quarterly distributions of overall rates of undocumented critical values per 1000 results were reported for all 4 years of analysis. An aggregate rate of overall compliance with critical value reporting was reported during the 16 quarters since the start of the study. Summaries of general questionnaire responses were compiled as well as a distribution of analytes on participating institutions' critical values list. The best-performing 25% of institutions were identified as achieving a rate of undocumented critical values per 1000 results, at most, 15 (2001), 13 (2002), 6 (2003), and 5 (2004).
As part of this quality assurance monitor, 180 Q-Tracks participants monitored quarterly critical value reporting compliance rates. The number of institutions that participated in each quarter varied from 72 to 99. A total of 669 815 inpatient and 274 299 outpatient critical values were included in the study. To investigate whether there was a tendency to attain lower rates of undocumented critical values per 1000 results as the length of time spent in the program increased, and if there were any practice variables collected over time associated with lower rates of undocumented critical values per 1000 results, data were collected on a quarterly basis for a maximum of 16 quarters and participants were grouped into 3 categories based on how long they had participated in the continuous monitoring program. Participants with only a single observation were excluded from this analysis.
The hypothesis that longer participation results in a decrease in the rates of undocumented critical values per 1000 results was formally tested using the PROC MIXED procedure in SAS (SAS, Cary, NC). A mixed linear model was used to account for repeated measures taken from the same experimental units. It is intuitive that a current rate may be highly correlated with previous and subsequent time-adjacent rates from the same institution. Another underlying assumption for use of this technique is that the data are normally distributed (Gaussian); the original rates were skewed. To minimize the skewness, data from participants who reported 100% critical values reporting rates for all 16 quarters were excluded. A natural log transformation was applied to the rates to make the distribution approximately Gaussian. There were 180 participants included in the final analysis.
Three mixed models were tested using the natural log transformation of the dependent variables, total rate of undocumented critical values per 1000 results, inpatient rate of undocumented critical values per 1000 results, and outpatient rate of undocumented critical values per 1000 results. The independent variables in the 3 models are listed in Table 4.
Mixed Linear Model Analysis
Total Rate of Undocumented Critical Values per 1000 Results.--Lower total rates of undocumented critical values per 1000 results were significantly associated with longer participation in the program (P = .02). The Figure illustrates the improvement in the total rate of undocumented critical values per 1000 results over time.
Variables found to be significantly associated with lower total rates of undocumented critical values per 1000 results are (1) American Association of Blood Banks (AABB) inspection within the past 2 years (P = .01); (2) unit secretary/ clerical staff not authorized to accept inpatient critical values notification (P = .004); and (3) a policy to always notify health care providers with inpatient critical values, irrespective of whether the patient is known to have results repeatedly in the critical value range (P = .01).
Table 5 provides the median total rate of undocumented critical values per 1000 results for the practice variables.
Inpatient Rate of Undocumented Critical Values per 1000 Results.--Lower inpatient rates of undocumented critical values per 1000 results were significantly associated with longer participation in the Q-Tracks monitoring program (P = .01). There was clear improvement in inpatient rates of undocumented critical values per 1000 results over time (Figure). Because the inpatient rates of undocumented critical values per 1000 results are highly correlated with the total rates of undocumented critical values per 1000 results (r = 0.94), the significant associations are similar for the 2 rates.
Variables found to be significantly associated with lower inpatient rates of undocumented critical values per 1000 results are (1) AABB inspection within the past 2 years (P = .01); (2) unit secretary/clerical staff not authorized to accept inpatient critical values notification (P = .001); and (3) a policy to always notify health care providers with inpatient critical values, regardless of whether the patient is known to have results repeatedly in the critical value range (P = .01). Table 5 provides the median inpatient rate of undocumented critical values per 1000 results for the practice variables.
Outpatient Rate of Undocumented Critical Values per 1000 Results.--Lower outpatient rates of undocumented critical values per 1000 results were significantly associated with longer participation in the Q-Tracks monitoring program (P = .003). The Figure illustrates the improvement in outpatient rates of undocumented critical values per 1000 results over time. No other variables were found to be significantly associated with outpatient rate of undocumented critical values per 1000 results.
Distribution of Cumulative Undocumented Critical Value Rates per 1000.--To generate useful comparative data for clinical laboratories, a distribution of cumulative undocumented critical value rates per 1000 results was generated for all 180 laboratories (Table 6). This analysis provides valuable information for benchmarking performance related to total critical values in a given setting. At the 25th percentile (25th percentile being better performance than the 75th percentile), 8.8 undocumented critical values were observed per 1000 events versus 66.4 undocumented critical values observed per 1000 events at the 75th percentile.
Critical value reporting is increasingly scrutinized by regulatory agencies and quality-management organizations as an important marker for excellence in patient centric care. (8) In 2003, the Joint Commission on Accreditation of Healthcare Organizations listed the read-back of critical values by receiving health care professionals as a key patient safety standard. Two years later, the Commission mandated an additional requirement for monitoring the turnaround time for critical values reporting. This new emphasis has caused hospital and health care organization administrators to focus with renewed vigor on critical values and their role in safe and effective patient care.
There is little consensus or benchmark data about critical values in the clinical laboratory. Procedures are not standardized and the definition of a critical value remains uncertain in the minds of many clinicians and laboratory professionals. A recently described analysis of Q-Probes data from 623 institutions highlights this variability. (9) In the present study, the choice of analytes and high and low critical limits varied considerably. Calling of critical values also varied, depending on patient status, with a mean time of 6.1 minutes for inpatient calls versus 13.7 minutes for outpatient calls.
This Q-Tracks monitoring program was designed to determine the rate of undocumented critical values per 1000 results in a group of voluntary participants, monitored on a quarterly basis. Three variables were found to be significantly associated with lower total and inpatient rates of undocumented critical values per 1000 results: (1) unit secretary/clerical staff not authorized to accept inpatient critical values notification; (2) procedures for handling inpatients who are known to have critical values that repeatedly include always notifying the health care provider, regardless of the previous results; and (3) AABB inspection within the past 2 years. The procedure requirement that clerical staff are not authorized to receive critical value results implies that such results are reported instead to licensed or professionally trained health care personnel. An institution making this distinction may be paying close attention to the provision of critical information to personnel capable of interpreting and translating the significance of the results. The presence of such a procedure is, perhaps, an indirect indicator of the direct involvement of the laboratory medical director, laboratory technologists, nurses, and physicians in the critical values reporting process. The role of leaders is to define and communicate the purpose of a procedure. When leadership is involved in a process, more significant progress is attained in achieving patient safety goals. (11)
Similarly, the finding that reporting of all critical values, even when they are repeat values on the same patients, may indicate a higher degree of vigilance in the critical value reporting system of a given institution, and may imply that physicians, nurses, and laboratory professionals acknowledge and respect the critical value reporting mechanism. Such vigilance is essential for sustaining patient safety efforts. This result additionally emphasizes the importance of a patient-centered approach and integration of clinical services. The critical values reporting system as initially described by Lundberg (1) may, in fact, be one of the earliest examples of an attempt to coordinate care of patients across different clinical service settings.
The significant association of AABB inspections with lower total rates of undocumented critical values per 1000 results is an interesting finding. Laboratory services that include transfusion medicine are integrated into a higher level of regulatory oversight. If, in addition to Clinical Laboratory Improvement Amendments and/or College of American Pathologists inspections, a clinical laboratory chooses to seek AABB accreditation, it may be demonstrating a knowledge base and confidence in its quality processes that are additionally reflected in its ability to improve total critical values reporting rates. An AABB accreditation indicates compliance with 10 groups of standards that provide the structure for a well-maintained quality management system. (12) Such systems often become models for other areas of a clinical laboratory.
When the distribution of cumulative undocumented critical value rate per 1000 results for 180 laboratories was analyzed for the first year of participation, 8.8 events/1000 results occurred in better-performing laboratories (25th percentile) versus 66.4 events/1000 results at the 75th percentile. This represents a 7.5-fold difference in performance. Clinical laboratories may use these data to benchmark their own performance and develop mechanisms by which they can create improvements in documentation of critical value reporting.
There was no significant association with other demographics or with responses to questions related to reporting mechanisms for outpatient critical values handled outside normal working hours (physician paged, answering service notified, results faxed or electronically sent, direct patient notification), the primary personnel responsible for reporting critical values (laboratory testing personnel, laboratory clerical staff), and, if critical values are faxed, whether there is a mechanism for confirmation notification.
An encouraging finding of this investigation is that continued participation in Q-Tracks monitoring of critical values reporting was associated with a progressive decrease in the rate of undocumented critical values per 1000 results for total, inpatient, and outpatient critical values. The institutions that had participated for longer periods of time in the Q-Tracks program had progressively decreased rates of undocumented critical values per 1000 results.
In the 1930s, early studies of quality in the manufacturing arena noted that subjects under observation perform better than unobserved subjects. (13) It is not possible to determine explicitly whether participation in the Q-Tracks program brought a higher level of participant awareness to critical value reporting. However, data from other clinical disciplines demonstrate increased quality awareness over time. For example, a study of cardiothoracic surgeons using continuous quality improvement to improve their practices reduced combined mortality rates by 24%. (14) Awareness is a key component of quality improvement. This Q-Tracks monitoring program provides consciousness of the significance of quality-enhancing practices in critical value reporting.
The mixed linear model used to analyze the Q-Tracks longitudinal dataset provides strong support for the factors that may be associated with improved total and inpatient critical values reporting. Association with explicit reporting to health care professionals, repeat reporting of critical values for the same patients, and AABB accreditation are markers for clinical laboratories that emphasize quality management and patient safety. Most encouraging is the continued decreased rate of undocumented critical values per 1000 results for institutions participating in this Q-Tracks monitoring program for 2 to 4 years, which clearly indicates that clinical laboratories can improve with time. These conclusions provide moral support to laboratory managers and directors as they develop quality-improvement and patient-safety projects in their clinical laboratories.
We thank Bushra Yasin, PhD, for her independent assessment and review of this article.
Accepted for publication July 10, 2006.
(1.) Lundberg GD. When to panic over abnormal values. MLO Med Lab Obs. 1972;4:47-54.
(2.) Emancipator K. Critical values: ASCP practice parameter. Am J Clin Pathol. 1997;108:247-253.
(3.) Lum G. Critical limits (alert values) for physician notification: universal or medical center specific limits? Ann Clin Lab Sci. 1998;28:261-271.
(4.) Tillman J, Barth JH. A survey of laboratory "critical (alert) limits" in the UK. Ann Clin Biochem. 2003;40:181-184.
(5.) Novis DA. Detecting and preventing the occurrence of errors in the practices of laboratory medicine and anatomic pathology: 15 years' experience with the College of American Pathologists' Q-Probes and Q-Tracks programs. Clin Lab Med. 2004;24:965-978.
(6.) Howanitz PJ. Errors in laboratory medicine: practical lessons to improve patient safety. Arch Pathol Lab Med. 2005;129:1252-1261.
(7.) Centers for Medicare and Medicaid Services. Medicare, Medicaid and CLIA programs: regulations implementing the Clinical Laboratory Improvement Amendments of 1988 (CLIA)--HCFA: final rule with comment period. Fed Regist. 1992;57:7002-7186.
(8.) Joint Commission on Accreditation of Healthcare Organizations. 2006 National patient safety goals. Available at: http://www.jcipatientsafety.org. Accessed June 8, 2006.
(9.) Howanitz PJ, Steindel SJ, Heard NV. Laboratory critical values policies and procedures: a College of American Pathologists Q-Probes study in 623 institutions. Arch Pathol Lab Med. 2002;126:663-669.
(10.) Wilkinson DS. Critical Values Reporting: Q-Tracks Annual Summary Report. Northfield, Ill: College of American Pathologists; 2004.
(11.) Committee on Quality of Health Care in America, Institute of Medicine. Building organizational supports for change. In: Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: National Academy Press; 2001:111-144.
(12.) Blood Bank/Transfusion Service Standards Program Unit. Standards for Blood Banks and Transfusion Services. 23rd ed. Bethesda, Md: American Association of Blood Banks; 2004.
(13.) Franke RH, Kaul JD. The Hawthorne experiments: first statistical interpretation. Am Sociol Rev. 1978;43:623-643.
(14.) O'Connor GT, Plume SK, Olmstead EM, et al. The Northern New England Cardiovascular Disease Study Group: a regional intervention to improve the hospital mortality associated with coronary artery bypass graft surgery. JAMA. 1996; 275:841-846.
Elizabeth A. Wagar, MD; Ana K. Stankovic, MD, PhD; David S. Wilkinson, MD, PhD; Molly Walsh, PhD; Rhona J. Souers, MS
From the Department of Pathology and Laboratory Medicine, David Geffen School of Medicine, University of California, Los Angeles (Dr Wagar); Becton Dickinson Diagnostics, Preanalytical Systems, Franklin Lakes, NJ (Dr Stankovic); Virginia Commonwealth University Health System, Department of Pathology, Richmond (Dr Wilkinson); and the College of American Pathologists, Department of Biostatistics, Northfield, Ill (Dr Walsh and Ms Souers).
The authors have no relevant financial interest in the products or companies described in this article.
Reprints: Elizabeth A. Wagar, MD, Professor and Vice Chair, Laboratory Medical Director, UCLA Clinical Laboratories, Department of Pathology and Laboratory Medicine, David Geffen School of Medicine at UCLA, Box 951732, AL-206 CHS, 10833 Le Conte Ave, Los Angeles, CA 90095-1732 (e-mail: email@example.com).
Table 1. Demographics of Participating Institutions Institution Data No. (%) of Institutions Type Voluntary, nonprofit 97 (58.1) Proprietary hospital 20 (12.0) State, county, city hospital 14 (8.4) University hospital 7 (4.2) Governmental, federal 7 (4.2) Private, independent laboratory 6 (3.6) System, integrated delivery network 5 (3.0) Children's hospital 3 (1.8) Other, governmental, nonfederal 4 (2.4) Other, nongovernmental 4 (2.4) Location City 85 (50.6) Suburban 47 (28.0) Rural 34 (20.2) Federal installation laboratory 2 (1.2) No. of occupied beds 1-150 67 (44.7) 151-300 45 (30.0) 301-450 27 (18.0) 451-600 8 (5.3) >600 3 (2.0) Table 2. Techniques or Processes Used by Best Performers That May Account for Higher Than Average Critical Values Reporting Rates Written report is issued for failure to document critical value communication Three times per day a computer report is generated to verify critical value notification; an additional report is generated the following morning to verify notification of all critical values from the previous day For panic values, the laboratory computer system retrieves the physician's name and telephone number and requires documentation of the notification Frequent training is provided on the importance of critical values and the correct way to document notification, and follow- up on any failures Table 3. List of Definitions Used in This Study Term Definition Inpatient Any patient who is admitted to the hospital; laboratories reporting values on emergency department patients should include them in this category Outpatient Patients who are registered at the hospital, but who are not admitted, as well as patients who are neither registered nor admitted, such as patients drawn in physicians' offices, outside nursing homes, and outreach programs Successful Situations in which there is documentation notifications that the critical value was communicated according to a given laboratory's policy Institution-specific Tracking institution-specific subgroup subgroups information is optional and intended for internal use only Table 4. Independent Variables in the Three Models Variable Description * Number of quarters since first participation Quarter of participation (ie, 01A, 01B) Institution is inspected/accredited by CAP (Y/N) Institution is inspected/accredited by JCAHO (Y/N) Institution is inspected/accredited by AABB (Y/N) Institution is inspected/accredited by FDA (Y/N) Institution is inspected/accredited by state (Y/N) Teaching hospital (Y/N) Nongovernmental institution (Y/N) Governmental, nonfederal institution (Y/N) Governmental, federal institution (Y/N) Institution trains pathology residents (Y/N) Number of licensed beds LIS used in Anatomic Pathology section (Y/N) LIS used for Chemistry/Hematology section (Y/N) LIS used for Microbiology section (Y/N) LIS used for HIS/LIS interface (Y/N) Who is authorized to accept notification of critical values for inpatients (4 levels) ([dagger]) Who is authorized to accept notification of critical values for outpatients (4 levels) ([dagger]) How are outpatient critical values handled outside of normal working hours (5 levels) ([dagger]) Who is primarily responsible for reporting critical values for inpatients (3 levels) ([dagger]) Who is primarily responsible for reporting critical values for outpatients (3 levels) ([dagger]) Procedure used for handling inpatients who are known to repeatedly have results in the critical range (4 levels) ([dagger]) Procedure used for handling outpatients who are known to repeatedly have results in the critical range (4 levels) ([dagger]) For inpatient critical values that are faxed or sent in an electronic format, there is a procedure in place to confirm that the recipient received the notification (Y/N) ([dagger]) For outpatient critical values that are faxed or sent in an electronic format, there is a procedure in place to confirm that the recipient received the notification (Y/N) ([dagger]) Number of analytes on the critical values list included in this study for inpatients ([dagger]) Number of analytes on the critical values list included in this study for outpatients ([dagger]) * CAP indicates College of American Pathologists; JCAHO, Joint Commission on Accreditation of Healthcare Organizations; AABB, American Association of Blood Banks; FDA, US Food and Drug Administration; LIS, laboratory information system; and HIS, hospital information system. ([dagger]) These variables were included in 2 of the 3 models. They were included in model testing the total rates of undocumented critical values per 1000 results and in the model for either the inpatient or the outpatient rates of undocumented critical values per 1000 results. Table 5. Median Total and Inpatient Rates of Undocumented Critical Values per 1000 Results for the Practice Variables Median Rate of Undocumented Critical Values per 1000 Results Practice Variable * Total Inpatient AABB inspection within past 2 years Yes 20.8 (n = 86) 15.5 (n = 86) No 25.3 (n = 68) 19.8 (n = 64) Unit secretary/clerical staff authorized to accept notification of inpatient critical values Yes 30.6 (n = 51) 25.0 (n = 51) No 20.8 (n = 98) 16.8 (n = 94) Procedure for handling inpatients who are known to have results repeatedly in the critical range Health care providers always notified, re- gardless of the previous results 17.7 (n = 122) 19.7 (n = 121) Other policies 19.3 (n = 27) 20.6 (n = 24) * AABB indicates American Association of Blood Banks. Table 6. Distribution of Cumulative Undocumented Critical Value Rates per 1000 Results for 180 Laboratories During the First Year of Participation Percentile Rate of Undocumented Critical Values per No. of 1000 Results Institutions 25th 50th 75th Total 180 8.8 22.6 66.4 Inpatient 168 6.9 20.2 55.5 Outpatent 167 8.1 27.7 68.9 Mean decrease in rate of undocumented critical values per 1,000 results Improvement in the rate of undocumented critical values per 1000 results over time. Total, inpatient, and outpatient rates of undocumented critical values per 1000 results were significantly lower participation in the program. Years 2 and 3 of participation are shown as combined data. Years of participation 4 years 2-3 years 1 year Outpatients 11.6 10.8 8.2 Inpatients 22.2 12.5 5.5 Total 22.8 13.9 7.8 Note: Table made from bar graph.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||CAP Laboratory Improvement Programs|
|Author:||Wagar, Elizabeth A.; Stankovic, Ana K.; Wilkinson, David S.; Walsh, Molly; Souers, Rhona J.|
|Publication:||Archives of Pathology & Laboratory Medicine|
|Date:||Jan 1, 2007|
|Previous Article:||American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing...|
|Next Article:||Flow cytometric detection of ZAP-70 in chronic lymphocytic leukemia: correlation with immunocytochemistry and Western blot analysis.|