- Email: [email protected]
Implementation, management and continuous quality improvement of point-of-care testing in an academic health care setting
Clinica Chimica Acta 307 Ž2001. 49–59 www.elsevier.comrlocaterclinchim
Implementation, management and continuous quality improvement of point-of-care testing in an academic health care setting Ellis Jacobs a,b,) , Karen A. Hinson b, Judit Tolnai b, Elkin Simson a,c a
b
Department of Pathology, The Mount Sinai School of Medicine, New York, NY 10029, USA Stat Laboratories and POC Testing, The Mount Sinai Hospital, Mount Sinai-NYU Health, Box 1519, One GustaÕe L. LeÕy Place, New York, NY 10029, USA c Center for Clinical Laboratories, The Mount Sinai Hospital, Mount Sinai-NYU Health, New York, NY 10029, USA
Abstract Point-of-care testing ŽPOCT. has economic and medical benefits in the areas of immediate medical management, resource utilization and time management. Starting with bedside glucose, the Mount Sinai Medical Center has, over the past 11 years, implemented 23 POC tests, spanning complexity from blood gasrelectrolyte testing to occult blood, in compliance with all regulatory and accreditation requirements. QC data are reviewed on a daily and weekly basis and all patient results are in the electronic medical record. A variety of healthcare workers; nurses, physicians, respiratory therapists and technologists, perform testing. Since POCT impacts on a variety of hospital departments, proper implementation and management requires a multi-disciplinary team approach with focus on the financial, regulatory, quality assurance and data integration issues. Established in 1996, the institutional committee, with laboratory leadership, handles the establishment, compliance review and future direction setting of the program. In 1999, over 1300 individuals performed over 440,000 POC tests within the institution. A formalized continuous quality improvement ŽCQI. program for the POCT program was developed in the fall of 1999. All testing sites are reviewed on a monthly basis for various quality indicators that cover QC performance, maintenance performance, proficiency testing, patient identification, and alert value confirmations. q 2001 Elsevier Science B.V. All rights reserved. Keywords: Quality assurance; Laboratory management; Point-of-care; Glucose; Blood gas; Coagulation; Accreditation
1. Introduction Point-of-care testing ŽPOCT. has economic and medical benefits w1–5x in the areas of immediate
) Corresponding author. Stat Laboratories and POC Testing, The Mount Sinai Hospital, Mount Sinai-NYU Health, Box 1519, One Gustave L. Levy Place, New York, NY 10029, USA. Tel.: q1-212-241-3891; fax: q1-212-876-0651. E-mail address: [email protected] ŽE. Jacobs..
medical management, resource utilization and time management. POCT has some unique characteristics relative to other forms of laboratory testing. It is a broad-based process that is potentially unrestricted to location, personnel or test menu. Furthermore, for it to be successful, because of the fact that it crosses the boundaries of many disciplines and departments within a medical setting, POCT must be a collective, multi-disciplinary effort w6–8x. In general, it employs simple-to-use technology, which is often performed in relatively low volume testing situations.
0009-8981r01r$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 9 - 8 9 8 1 Ž 0 1 . 0 0 4 3 2 - 6
50
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
Table 1 Point-of-care testing committee responsibilities Implementation phase
Oversight phase
Define basic system Define what record keeping is required Select methodology Assign responsibility for test performance
Review QCrQA Monitor compliance Utilization review Approve expansionr contraction of programs Assess impactroutcomes
The Mount Sinai Medical Center is part of the Mount Sinai NYU Health system that includes The Mount Sinai Hospital and Medical Center, NYU Hospital Center, Hospital for Joint Diseases, NYU Downtown Hospital and The Mount Sinai Hospital of Queens. The Mount Sinai Hospital is a 1171-bed tertiary care facility that was reorganized several years ago, utilizing a patient care focus model w9,10x, into various care centers, i.e., Cardiac, GIrGU, Maternal and Child Health, Neuroscience and Restorative, Oncology, Peri-Operative Services, Primary Care and Medical Services, and Psychiatry. In 1999, The Mount Sinai Hospital had 405,572 out-patient visits, 78,172 emergency room visits, 44,021 discharges and 4701 births. In 1989, the first formalized point-of-care ŽPOC. test was started at The Mount Sinai Hospital with the introduction of bedside glucose testing. This program was implemented in full compliance with all regulatory requirements ŽCollege of American Pathologists, Joint Commission for Accreditation of Healthcare Organizations, New York State.. By 1996, 11 additional POCT programs were implemented Žblood gas, Na, K, Ca, hematocrit, activated clotting time, fetal scalp pH, O 2 saturation, prothrombin time ŽPT. and activated partial thromboplastin time ŽPTT.. In addition to these tests, there were a number of other POCT tests, e.g., urinalysis, pregnancy, rapid strep A, occult blood, and spun microhematocrit, being performed by various groups within the Medical Center but not under laboratory oversight and not meeting all regulatory requirements. In September 1996, the decision was made that all POCT performed within the Medical Center would be brought into full regulatory compliance under the guidance of the clinical laboratories.
2. Methods and actions A survey of POCT throughout the medical center was immediately performed. An institutional ban on the performance of spun microhematocrits outside the clinical laboratories was initiated, predominately as a safety issue. An institutional POCT committee, with laboratory leadership, was convened in December 1996 with representation from laboratory, physicians, upper administration, care center administration, nursing education, nursing practice committee, material management, infection control and regulatory affairs. It was charged with reviewing current POCT, determining which tests should continue, and establishing programs that would bring all POCT performed within the Medical Center into regulatory compliance. Furthermore, on an on-going basis, the committee was given the responsibility to monitor
Table 2 The Mount Sinai Medical Center point-of-care testing QA indicators General indicators Number of unidentifiable samples QC material in use within expiration date Correct QC lot number and expiration date programmed in memory QC bottle dated when opened Reagents in use within expiration date Reagents dated when opened QC results documented on days when patient testing was performed QC statistics Routine maintenance performed and documented System specific indicator Activated clotting time—Well temperature checked monthly Blood gasrelectrolytes—Number of aborted samples Occult blood—On-board QC zone checked with each sample Specific gravity by refractometry—Zero setting adjustment completed and documented Glucose Monthly audit, minimum 2 patientsrunit, Hct values between 20–60%, results charted on capillary blood glucose flowsheet Daily maintenance entered into meter memory - 30 or ) 500 mgrdl sent for confirmation Control cuvette and analyzer have matching serial number
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
programs, authorize expansionrcontraction of existing programs, and approve new POCT for the Medi-
51
cal Center. The functions of the committee, both during the implementation phase of a POC test, as
Fig. 1. Quality improvement exception request form.
52
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
well as during the oversight phase, are listed in Table 1. It is through such a committee that authorization and control of POCT is maintained within a health care institution. The POCT committee reviewed the results of the survey and immediately determined that hemoglobin testing would replace spun microhematocrits in dialysis and emergency room ŽER. only. Additionally, it was decided that gram stains were not approved for POC testing anywhere within the institution. A full program was developed by the committee. Additional staffing and supplies needed for the existing POCT program to assume the new responsibilities were added to the budget. Total laboratory staffing for the POCT program was increased to five individuals. Joint nursingrlaboratory policies and procedures were written for the various testing systems. A formalized mechanism was established for care areas to petition the committee for authorization to perform POCT. The following are the five questions to be answered in the request for POCT. Ž1. What is the medical andror financial justification for the testing? Ž2. What is the anticipated frequency and volume of testing? Ž3. Why are current laboratory services insufficient? Ž4. Who will perform testing, i.e., job classification, and how many individuals? Ž5. Who will be the key person at the testing site for assuring quality control ŽQC.rquality assurance ŽQA. requirements? A formalized continuous quality improvement ŽCQI. program for the entire POCT program was developed in the fall of 1999. All testing sites are reviewed on a monthly basis for various quality indicators that cover QC performance, maintenance performance, proficiency testing, patient identification, and alert value confirmations ŽTable 2.. When compliance is below set thresholds, a multi-part form ŽFig. 1. is sent to all nurse managers requesting investigation, conclusion and what the follow-up actions will be.
3. Results 3.1. Program growth The program started with one POC test, glucose, in May 1989. By 2000, there were 23 different POC
tests performed at the Mount Sinai Medical Center ŽTable 3. The greatest surge in the performance of regulated testing occurred in 1997, due to the mandate to bring all testing into regulatory compliance. Since 1997, there has been a slow but steady growth of the POCT program at The Mount Sinai Hospital, as the ability to meet the medical needs of the institution via POCT has increased andror financial justification of POC testing has improved. In 1999, over 440,000 POC tests were performed within the institution ŽTable 4. by 1285 nurses, 64 physicians, 21 medical technologists, 20 respiratory therapists, 20 dialysis technicians, 8 perfusionists and 6 physician assistants. QC data are reviewed on a daily and weekly basis and all patient results are entered into the electronic medical record. Currently, only the blood gasrelectrolyterhematocrit POC testing utilizes a data management system that collects data on a real-time basis via the hospital ethernet and transmits the results into the laboratory information system ŽLIS.. All other data are either collected via laptop computers or faxed to the laboratory on a daily basis for review and entry into the LIS. Average daily testing ranged from less than 1 for PTT testing to 741 for glucose testing. These two tests also demonstrated the extremes in the samplerliquid control testing ratios, 0.2 and 5.1 for PTT and glucose testing, respectively. The mean samplerQC test ratio for the entire POCT program was 2.4. Those tests with a daily testing volume ) 10 had a mean ratio of 3.4, whereas those tests performed less than 10 times a day had a mean ratio of 0.8. The distribution of POCT by patient type within the medical center is shown in Fig. 2. Patients were classified either as in-patients, out-patients, ER patients, or non-patients. Non-patients are those individuals seen in the private practice areas of the medical center or are referred to the out-patient specimen drawing and testing facility. Over 61% of all POC testing sites are at in-patient locations vs. 39% in out-patient locations and - 1% in non-patient locations. Of the 23 different POC tests, glucose, occult blood and urinalysis are most commonly performed, at 66%, 55% and 35% of all testing sites, respectively. Glucose, blood gasrelectrolytesr hematocrit, activated clotting time and fetal scalp pH testing are predominately performed on in-patients;
Table 3 The Mount Sinai Medical Center point-of-care testing regulated through the clinical laboratories 1992–1993
1994–1996
1997–1998
1998–1999
2000
Blood glucose
Blood glucose Blood gas Na, K, Ca and hematocrit
Blood glucose Blood gas, Na, K, Ca and hematocrit Activated clotting time Fetal scalp pH Oxygen saturation PTrPTT
Blood glucose Blood gas, Na, K, Ca and hematocrit Activated clotting time Fetal scalp pH Oxygen saturation PTrPTT Fecal occult blood Gastric occult blood Hemoglobin Pregnancy PT—Home health agency Rapid strep A Specific gravity by refractometry Urinalysis
Blood glucose Blood gas, Na, K, Ca and hematocrit Activated clotting time Fetal scalp pH Oxygen saturation PTrPTT Fecal occult blood Gastric occult blood Hemoglobin Pregnancy PT—Home health agency Rapid strep A Specific gravity by refractometry Urinalysis Hemoglobin A1C Urine creatinine and microalbumin
Blood glucose Blood gas, Na, K, Ca and hematocrit Activated clotting time Fetal scalp pH Oxygen saturation PTrPTT Fecal occult blood Gastric occult blood Hemoglobin Pregnancy PT—Home health agency Rapid strep A Specific gravity by refractometry Urinalysis Hemoglobin A1C Urine creatinine and microalbumin Cholesterol and triglyceride
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
1989–1991
53
54
Table 4 1999 point-of-care testing statistics of the Mount Sinai Medical Center Number of sites
Average daily volume
Annual volume
Sampler QC ratio
Imprecision Low %CV Žmean.
High %CV Žmean.
68 5 7 4 31 2 6
741 22a 31 24a 76 2 52
270,492 5595 11,322 5737 27,779 805 18,981
5.1 4.1 4.4 4.5 2.6 0.3 1.8
Electrolytes ŽNa, K, Ca2q .
6
52
18,981
1.8
Activated clotting time Oxygen saturation Rapid strep A Fetal scalp pH PTT Microalbuminrcreatinine
7 3 5 1 1 3
30 7 5a 2 – 12a
11,012 2505 1091 848 65 2859
2.4 1.5 1.1 0.7 0.2 3.6
7.8 Ž33 mgrdl. 5.0 Ž5.2%. NrA 5.0 Ž11.7 s. NrA 1.7 Ž8 grdl. 0.1 Ž7.01., 4.0 Ž18 mm Hg., 3.0 Ž65 mm Hg. 1.3 Ž119., 1.9 Ž2.6., 2.5 Ž0.86 mmolrl. 10.7 Ž158 s. 0.3 Ž48.6 %. NrA 0.1 Ž7.18. 4.2 Ž56 s. 3.5 Ž33.5 mgrl.r 2.6 Ž102.5 mgrdl.
6.1 Ž441 mgrdl. 3.8 Ž13.1%. NrA 3.4 Ž19.1 s. NrA 1.2 Ž15.7 grdl. 0.1 Ž7.61., 3.0 Ž69 mm Hg., 1.0 Ž144 mm Hg. 0.9 Ž155., 1.1 Ž6.2., 1.2 Ž1.45 mmolrl. 5.6 Ž270 s. 0.5 Ž48.0%. NrA 0.1 Ž7.62. 3.7 Ž104 s. 4.0 Ž218.0 mgrl.r 3.5 Ž412.0 mgrdl.
Glucose Hemoglobin A1C Pregnancy PT Urinalysis Hemoglobin Blood gas ŽpH, P CO 2 , P O 2 .
a
Based on 5 daysrweek.
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
Test
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
55
Fig. 2. Distribution of POCT by patient type at the Mount Sinai Medical Center.
whereas pregnancy, hemoglobin, PTrPTT, rapid strep A, hemoglobin A1C, urinary creatinine and microalbumin are predominately performed on outpatients and non-patients. The utilization of POCT by the various care centers is shown in Fig. 3. As would be predicted due to the breadth of medical needs, the Maternal and Child, as well as, General Medicine Care Centers had the largest menu of POC tests available. In several cases, there are POC tests that are uniquely performed only within a specific care center. Maternal and Child Care Center was the only area that performed rapid strep A and fetal scalp pH testing, at 29% and 7% of 14 sites within the care center, respectively. Activated clotting time was performed at 44% of the 16 care sites within the Cardiac Care Center only. PTrPTT, hemoglobin A1C and urinary microalbuminrcreatinine are three other POC tests that were only performed in one care center, General Medicine, at 11%, 5%, and 5% of the 19 sites, respectively. At the Mount Sinai Medical Center, the performance of POC testing has expanded beyond the hospital walls into the community at large. Under the
auspices of the Maternal and Child Care Center, testing services are provided to the youth of the community through the Adolescent Health Center and public school-based clinic programs. In both of these settings, glucose, pregnancy, rapid strep A, and urinalysis are made available. The Mount Sinai Home Health Agency provides fingerstick PT testing. The other off-campus site is the dialysis center, where glucose and hemoglobin testing is performed. 3.2. Quality improÕement In the last quarter of 1999, 346 quality improvement ŽQI. exception requests were sent to 567 different testing sites. The major QI issues quantified were instrument maintenance documentation, QC documentation for manual tests, and documentation of actions taken associated with the correction of failed QC tests. Significant improvement was made in the QC documentation of urine dipstick testing, with the number of sites exceeding the threshold of 5% of QC results not being documented being reduced from 30% to 15%. The plans are that, when 95% of all test sites are in full compliance, the threshold limits
56
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
Fig. 3. Care center distribution of POCT at the Mount Sinai Medical Center. Number in parenthesis represents the total number of testing sites within that care center.
for the various indicators will be reduced, i.e., continuous quality improvement.
Figs. 4 and 5 track the quality performance indicators for glucose and urine dipstick testing for the
Fig. 4. POC glucose testing quality assurance indicators. Number of meters s 80. Symbols: closed symbols meet threshold limits, open symbols do not meet threshold limits; v —daily QC documentation, `—within range QC, \—failed QC documentation, I—daily maintenance documentation, '—weekly maintenance documentation, e—alert value follow-up.
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
57
Fig. 5. POC urine dipstick testing quality assurance indicators. Number of testing sites s 30. Symbols: closed symbols meet threshold limits, open symbols do not meet threshold limits; v —daily QC documentation, `—proper patient ID, I—within range QC, \—failed QC documentation.
first 6 months of the program. The daily QC documentation of 100% and the average weekly maintenance documentation of 85% for glucose testing were within acceptable limits. The 93% of QC data within the acceptable range just missed the acceptability threshold of 95%. The other three glucose indicators of daily maintenance documentation Ž58%., alert value follow-up Ž58%. and documentation of corrective action taken for failed QC Ž18%. all show the need for significant improvement. Of the four quality assurance indicators for urine dipstick testing, only daily QC documentation, at 80%, was within acceptable threshold limits Ž80%.. Both proper patient ID and percentage of QC data within acceptable range, at 86% and 81%, respectively, require some improvement to meet the acceptable threshold limits of 95%. As with glucose testing, there is a significant problem with the documentation of corrective actions for failed urine dipstick QC, 0% compliance. An analysis of all the problem calls received by the POCT program is shown in Fig. 6. From January through April 2000, 105 calls were made and resolved. All of these calls were generated by only 10 of the 23 tests performed Žblood gases s blood gas, Na, K, Ca and hematocrit.. Despite being one of the
lowest volume tests perfomed, fetal scalp pH testing accounted for 15% of the calls received. Most of these were due to instrument-related problems. Glucose testing generated the greatest number of calls Ž72%., as would be expected due to the size of that testing program. The majority of these calls Ž55%. were due to the glucose meter being dirty ŽFig. 7., followed by the inability to use the device due to QC lock-out Ž21%.. Only 25% of the problems with the
Fig. 6. POCT CQI program—Incident analysis, January–April 2000, ns105.
58
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
Fig. 7. POCT CQI program—Glucose incident resolution, January–April 2000, 30 meters involved, ns 77.
glucose testing were due to problems with the system, either device or QC solution. Thus, 75% of the time problems with glucose testing were due to some type of operator error.
4. Discussion Success of a medical center point-of-care testing program depends upon a multi-disciplinary team approach. Over the last 11 years, the regulated POCT program at the Mount Sinai Medical Center has grown from a single-analyte program with limited impact on the in-patient medical system to a 23-analyte program with significant impact on patient care throughout the entire spectrum of medical care, from in-patient to home patients. Significant changes occurred after the creation of the institutional POCT committee. Sixty one percent of all testing sites are at in-patient locations vs. 39% in out-patient locations and - 1% in non-patient locations. Out of the 23 different POC tests, glucose, occult blood and urinalysis are most commonly performed, at 66%, 55% and 35% of all test sites, respectively. Maternal and Child Health and General Medicine Care Centers have the most diverse menu of POC tests. Glucose, blood gasrelectrolytes, activated clotting time and fetal scalp pH tests are predominately performed on in-patients, whereas pregnancy, hemoglobin, PTrPTT, rapid strep A, hemoglobin A1C, urinary microalbumin and creatinine are predominately performed on out-patients and non-patients.
The implementation of a formalized CQI program helped quantify what was already previously known to be a problem area in the POCT program-documentation, i.e., instrument maintenance, manual QC tests and actions taken associated with correction of failed QC tests. More importantly, it raised the consciousness of the end-users to the significance of this documentation and has helped to improve the overall program. Furthermore, along with the analysis of the various problem calls received, the POCT coordinators have been able to adjust the training and recertification programs in order to address the identified problems. Overall, the CQI program has been well received by the entire nursing and medical staff and has effected improved communications. One of the goals of the committee is to determine the true impact of POCT on a patient’s hospital stay andror healthcare, not just the cost of providing a test result. Other future plans include establishing AJust-in-TimeB supply distribution based on recorded workload statistics.
Acknowledgements The authors would like to acknowledge Dr. Maria Vezina, Director of Nursing Education, and Ms. Julie Fisher, Associate Hospital Director, without whose support and understanding, the program would not have been successful. Additionally, a most sincere thank you to the hard work of Ms. Svetlana Babayeva, Mr. Abelardo Gorospe and Ms. Fira Steinberg, point-of-care testing chemists.
References w1x DeCresce RP, Phillips DL, Howanitz PT. Financial justification of alternative site testing. Arch Pathol Lab Med 1995; 119:899–901. w2x Jacobs E. Analysis of economic models used to justify point-of-care testing. In: D’Orazio P, editor. Preparing for critical care analysis in the twenty first century. Proceedings of an International Symposium. Medfield, MA: Electrolyter Blood Gas Div. vol. 16. Washington, DC: AACC 1996: 126–36. w3x Handord CR. POC testing: must quality cost more? Med Lab Obs 1993;9S:28–33. w4x Kost G. Guidelines for point-of-care testing: improving
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59 patient outcomes. Am J Clin Pathol 1995;104ŽSuppl. 4.: S111–27. w5x Suver JD, Nuemann BR, Boles EE. Accounting for the costs of quality. Health Care Financ Manage 1992;9:29–37. w6x Jacobs E. TQM and POC testing. Med Lab Obs 1993; 25Ž9S.:2–6. w7x Jacobs E. The laboratory as a leader of point-of-care programs. Med Lab Obs 1998;30Ž9S.:6–10.
59
w8x NCCLS Approved Guideline AST2-A, Point-of-care in vitro diagnostic ŽIVD. testing. Vilanova, PA: National Committee for Clinical Laboratory Standards, 1999. June. w9x Lathrop JP, Seufert GE, MacDonald RJ, Martin SB. The patient-focused hospital: a patient care concept. J Soc Health Syst 1991;3:33–50. w10x Borzo G. Patient-focused hospitals begin reporting good results. Health Care Strategic Manage 1992;10Ž8.:17–22.
Implementation, management and continuous quality improvement of point-of-care testing in an academic health care setting Ellis Jacobs a,b,) , Karen A. Hinson b, Judit Tolnai b, Elkin Simson a,c a
b
Department of Pathology, The Mount Sinai School of Medicine, New York, NY 10029, USA Stat Laboratories and POC Testing, The Mount Sinai Hospital, Mount Sinai-NYU Health, Box 1519, One GustaÕe L. LeÕy Place, New York, NY 10029, USA c Center for Clinical Laboratories, The Mount Sinai Hospital, Mount Sinai-NYU Health, New York, NY 10029, USA
Abstract Point-of-care testing ŽPOCT. has economic and medical benefits in the areas of immediate medical management, resource utilization and time management. Starting with bedside glucose, the Mount Sinai Medical Center has, over the past 11 years, implemented 23 POC tests, spanning complexity from blood gasrelectrolyte testing to occult blood, in compliance with all regulatory and accreditation requirements. QC data are reviewed on a daily and weekly basis and all patient results are in the electronic medical record. A variety of healthcare workers; nurses, physicians, respiratory therapists and technologists, perform testing. Since POCT impacts on a variety of hospital departments, proper implementation and management requires a multi-disciplinary team approach with focus on the financial, regulatory, quality assurance and data integration issues. Established in 1996, the institutional committee, with laboratory leadership, handles the establishment, compliance review and future direction setting of the program. In 1999, over 1300 individuals performed over 440,000 POC tests within the institution. A formalized continuous quality improvement ŽCQI. program for the POCT program was developed in the fall of 1999. All testing sites are reviewed on a monthly basis for various quality indicators that cover QC performance, maintenance performance, proficiency testing, patient identification, and alert value confirmations. q 2001 Elsevier Science B.V. All rights reserved. Keywords: Quality assurance; Laboratory management; Point-of-care; Glucose; Blood gas; Coagulation; Accreditation
1. Introduction Point-of-care testing ŽPOCT. has economic and medical benefits w1–5x in the areas of immediate
) Corresponding author. Stat Laboratories and POC Testing, The Mount Sinai Hospital, Mount Sinai-NYU Health, Box 1519, One Gustave L. Levy Place, New York, NY 10029, USA. Tel.: q1-212-241-3891; fax: q1-212-876-0651. E-mail address: [email protected] ŽE. Jacobs..
medical management, resource utilization and time management. POCT has some unique characteristics relative to other forms of laboratory testing. It is a broad-based process that is potentially unrestricted to location, personnel or test menu. Furthermore, for it to be successful, because of the fact that it crosses the boundaries of many disciplines and departments within a medical setting, POCT must be a collective, multi-disciplinary effort w6–8x. In general, it employs simple-to-use technology, which is often performed in relatively low volume testing situations.
0009-8981r01r$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 9 - 8 9 8 1 Ž 0 1 . 0 0 4 3 2 - 6
50
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
Table 1 Point-of-care testing committee responsibilities Implementation phase
Oversight phase
Define basic system Define what record keeping is required Select methodology Assign responsibility for test performance
Review QCrQA Monitor compliance Utilization review Approve expansionr contraction of programs Assess impactroutcomes
The Mount Sinai Medical Center is part of the Mount Sinai NYU Health system that includes The Mount Sinai Hospital and Medical Center, NYU Hospital Center, Hospital for Joint Diseases, NYU Downtown Hospital and The Mount Sinai Hospital of Queens. The Mount Sinai Hospital is a 1171-bed tertiary care facility that was reorganized several years ago, utilizing a patient care focus model w9,10x, into various care centers, i.e., Cardiac, GIrGU, Maternal and Child Health, Neuroscience and Restorative, Oncology, Peri-Operative Services, Primary Care and Medical Services, and Psychiatry. In 1999, The Mount Sinai Hospital had 405,572 out-patient visits, 78,172 emergency room visits, 44,021 discharges and 4701 births. In 1989, the first formalized point-of-care ŽPOC. test was started at The Mount Sinai Hospital with the introduction of bedside glucose testing. This program was implemented in full compliance with all regulatory requirements ŽCollege of American Pathologists, Joint Commission for Accreditation of Healthcare Organizations, New York State.. By 1996, 11 additional POCT programs were implemented Žblood gas, Na, K, Ca, hematocrit, activated clotting time, fetal scalp pH, O 2 saturation, prothrombin time ŽPT. and activated partial thromboplastin time ŽPTT.. In addition to these tests, there were a number of other POCT tests, e.g., urinalysis, pregnancy, rapid strep A, occult blood, and spun microhematocrit, being performed by various groups within the Medical Center but not under laboratory oversight and not meeting all regulatory requirements. In September 1996, the decision was made that all POCT performed within the Medical Center would be brought into full regulatory compliance under the guidance of the clinical laboratories.
2. Methods and actions A survey of POCT throughout the medical center was immediately performed. An institutional ban on the performance of spun microhematocrits outside the clinical laboratories was initiated, predominately as a safety issue. An institutional POCT committee, with laboratory leadership, was convened in December 1996 with representation from laboratory, physicians, upper administration, care center administration, nursing education, nursing practice committee, material management, infection control and regulatory affairs. It was charged with reviewing current POCT, determining which tests should continue, and establishing programs that would bring all POCT performed within the Medical Center into regulatory compliance. Furthermore, on an on-going basis, the committee was given the responsibility to monitor
Table 2 The Mount Sinai Medical Center point-of-care testing QA indicators General indicators Number of unidentifiable samples QC material in use within expiration date Correct QC lot number and expiration date programmed in memory QC bottle dated when opened Reagents in use within expiration date Reagents dated when opened QC results documented on days when patient testing was performed QC statistics Routine maintenance performed and documented System specific indicator Activated clotting time—Well temperature checked monthly Blood gasrelectrolytes—Number of aborted samples Occult blood—On-board QC zone checked with each sample Specific gravity by refractometry—Zero setting adjustment completed and documented Glucose Monthly audit, minimum 2 patientsrunit, Hct values between 20–60%, results charted on capillary blood glucose flowsheet Daily maintenance entered into meter memory - 30 or ) 500 mgrdl sent for confirmation Control cuvette and analyzer have matching serial number
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
programs, authorize expansionrcontraction of existing programs, and approve new POCT for the Medi-
51
cal Center. The functions of the committee, both during the implementation phase of a POC test, as
Fig. 1. Quality improvement exception request form.
52
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
well as during the oversight phase, are listed in Table 1. It is through such a committee that authorization and control of POCT is maintained within a health care institution. The POCT committee reviewed the results of the survey and immediately determined that hemoglobin testing would replace spun microhematocrits in dialysis and emergency room ŽER. only. Additionally, it was decided that gram stains were not approved for POC testing anywhere within the institution. A full program was developed by the committee. Additional staffing and supplies needed for the existing POCT program to assume the new responsibilities were added to the budget. Total laboratory staffing for the POCT program was increased to five individuals. Joint nursingrlaboratory policies and procedures were written for the various testing systems. A formalized mechanism was established for care areas to petition the committee for authorization to perform POCT. The following are the five questions to be answered in the request for POCT. Ž1. What is the medical andror financial justification for the testing? Ž2. What is the anticipated frequency and volume of testing? Ž3. Why are current laboratory services insufficient? Ž4. Who will perform testing, i.e., job classification, and how many individuals? Ž5. Who will be the key person at the testing site for assuring quality control ŽQC.rquality assurance ŽQA. requirements? A formalized continuous quality improvement ŽCQI. program for the entire POCT program was developed in the fall of 1999. All testing sites are reviewed on a monthly basis for various quality indicators that cover QC performance, maintenance performance, proficiency testing, patient identification, and alert value confirmations ŽTable 2.. When compliance is below set thresholds, a multi-part form ŽFig. 1. is sent to all nurse managers requesting investigation, conclusion and what the follow-up actions will be.
3. Results 3.1. Program growth The program started with one POC test, glucose, in May 1989. By 2000, there were 23 different POC
tests performed at the Mount Sinai Medical Center ŽTable 3. The greatest surge in the performance of regulated testing occurred in 1997, due to the mandate to bring all testing into regulatory compliance. Since 1997, there has been a slow but steady growth of the POCT program at The Mount Sinai Hospital, as the ability to meet the medical needs of the institution via POCT has increased andror financial justification of POC testing has improved. In 1999, over 440,000 POC tests were performed within the institution ŽTable 4. by 1285 nurses, 64 physicians, 21 medical technologists, 20 respiratory therapists, 20 dialysis technicians, 8 perfusionists and 6 physician assistants. QC data are reviewed on a daily and weekly basis and all patient results are entered into the electronic medical record. Currently, only the blood gasrelectrolyterhematocrit POC testing utilizes a data management system that collects data on a real-time basis via the hospital ethernet and transmits the results into the laboratory information system ŽLIS.. All other data are either collected via laptop computers or faxed to the laboratory on a daily basis for review and entry into the LIS. Average daily testing ranged from less than 1 for PTT testing to 741 for glucose testing. These two tests also demonstrated the extremes in the samplerliquid control testing ratios, 0.2 and 5.1 for PTT and glucose testing, respectively. The mean samplerQC test ratio for the entire POCT program was 2.4. Those tests with a daily testing volume ) 10 had a mean ratio of 3.4, whereas those tests performed less than 10 times a day had a mean ratio of 0.8. The distribution of POCT by patient type within the medical center is shown in Fig. 2. Patients were classified either as in-patients, out-patients, ER patients, or non-patients. Non-patients are those individuals seen in the private practice areas of the medical center or are referred to the out-patient specimen drawing and testing facility. Over 61% of all POC testing sites are at in-patient locations vs. 39% in out-patient locations and - 1% in non-patient locations. Of the 23 different POC tests, glucose, occult blood and urinalysis are most commonly performed, at 66%, 55% and 35% of all testing sites, respectively. Glucose, blood gasrelectrolytesr hematocrit, activated clotting time and fetal scalp pH testing are predominately performed on in-patients;
Table 3 The Mount Sinai Medical Center point-of-care testing regulated through the clinical laboratories 1992–1993
1994–1996
1997–1998
1998–1999
2000
Blood glucose
Blood glucose Blood gas Na, K, Ca and hematocrit
Blood glucose Blood gas, Na, K, Ca and hematocrit Activated clotting time Fetal scalp pH Oxygen saturation PTrPTT
Blood glucose Blood gas, Na, K, Ca and hematocrit Activated clotting time Fetal scalp pH Oxygen saturation PTrPTT Fecal occult blood Gastric occult blood Hemoglobin Pregnancy PT—Home health agency Rapid strep A Specific gravity by refractometry Urinalysis
Blood glucose Blood gas, Na, K, Ca and hematocrit Activated clotting time Fetal scalp pH Oxygen saturation PTrPTT Fecal occult blood Gastric occult blood Hemoglobin Pregnancy PT—Home health agency Rapid strep A Specific gravity by refractometry Urinalysis Hemoglobin A1C Urine creatinine and microalbumin
Blood glucose Blood gas, Na, K, Ca and hematocrit Activated clotting time Fetal scalp pH Oxygen saturation PTrPTT Fecal occult blood Gastric occult blood Hemoglobin Pregnancy PT—Home health agency Rapid strep A Specific gravity by refractometry Urinalysis Hemoglobin A1C Urine creatinine and microalbumin Cholesterol and triglyceride
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
1989–1991
53
54
Table 4 1999 point-of-care testing statistics of the Mount Sinai Medical Center Number of sites
Average daily volume
Annual volume
Sampler QC ratio
Imprecision Low %CV Žmean.
High %CV Žmean.
68 5 7 4 31 2 6
741 22a 31 24a 76 2 52
270,492 5595 11,322 5737 27,779 805 18,981
5.1 4.1 4.4 4.5 2.6 0.3 1.8
Electrolytes ŽNa, K, Ca2q .
6
52
18,981
1.8
Activated clotting time Oxygen saturation Rapid strep A Fetal scalp pH PTT Microalbuminrcreatinine
7 3 5 1 1 3
30 7 5a 2 – 12a
11,012 2505 1091 848 65 2859
2.4 1.5 1.1 0.7 0.2 3.6
7.8 Ž33 mgrdl. 5.0 Ž5.2%. NrA 5.0 Ž11.7 s. NrA 1.7 Ž8 grdl. 0.1 Ž7.01., 4.0 Ž18 mm Hg., 3.0 Ž65 mm Hg. 1.3 Ž119., 1.9 Ž2.6., 2.5 Ž0.86 mmolrl. 10.7 Ž158 s. 0.3 Ž48.6 %. NrA 0.1 Ž7.18. 4.2 Ž56 s. 3.5 Ž33.5 mgrl.r 2.6 Ž102.5 mgrdl.
6.1 Ž441 mgrdl. 3.8 Ž13.1%. NrA 3.4 Ž19.1 s. NrA 1.2 Ž15.7 grdl. 0.1 Ž7.61., 3.0 Ž69 mm Hg., 1.0 Ž144 mm Hg. 0.9 Ž155., 1.1 Ž6.2., 1.2 Ž1.45 mmolrl. 5.6 Ž270 s. 0.5 Ž48.0%. NrA 0.1 Ž7.62. 3.7 Ž104 s. 4.0 Ž218.0 mgrl.r 3.5 Ž412.0 mgrdl.
Glucose Hemoglobin A1C Pregnancy PT Urinalysis Hemoglobin Blood gas ŽpH, P CO 2 , P O 2 .
a
Based on 5 daysrweek.
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
Test
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
55
Fig. 2. Distribution of POCT by patient type at the Mount Sinai Medical Center.
whereas pregnancy, hemoglobin, PTrPTT, rapid strep A, hemoglobin A1C, urinary creatinine and microalbumin are predominately performed on outpatients and non-patients. The utilization of POCT by the various care centers is shown in Fig. 3. As would be predicted due to the breadth of medical needs, the Maternal and Child, as well as, General Medicine Care Centers had the largest menu of POC tests available. In several cases, there are POC tests that are uniquely performed only within a specific care center. Maternal and Child Care Center was the only area that performed rapid strep A and fetal scalp pH testing, at 29% and 7% of 14 sites within the care center, respectively. Activated clotting time was performed at 44% of the 16 care sites within the Cardiac Care Center only. PTrPTT, hemoglobin A1C and urinary microalbuminrcreatinine are three other POC tests that were only performed in one care center, General Medicine, at 11%, 5%, and 5% of the 19 sites, respectively. At the Mount Sinai Medical Center, the performance of POC testing has expanded beyond the hospital walls into the community at large. Under the
auspices of the Maternal and Child Care Center, testing services are provided to the youth of the community through the Adolescent Health Center and public school-based clinic programs. In both of these settings, glucose, pregnancy, rapid strep A, and urinalysis are made available. The Mount Sinai Home Health Agency provides fingerstick PT testing. The other off-campus site is the dialysis center, where glucose and hemoglobin testing is performed. 3.2. Quality improÕement In the last quarter of 1999, 346 quality improvement ŽQI. exception requests were sent to 567 different testing sites. The major QI issues quantified were instrument maintenance documentation, QC documentation for manual tests, and documentation of actions taken associated with the correction of failed QC tests. Significant improvement was made in the QC documentation of urine dipstick testing, with the number of sites exceeding the threshold of 5% of QC results not being documented being reduced from 30% to 15%. The plans are that, when 95% of all test sites are in full compliance, the threshold limits
56
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
Fig. 3. Care center distribution of POCT at the Mount Sinai Medical Center. Number in parenthesis represents the total number of testing sites within that care center.
for the various indicators will be reduced, i.e., continuous quality improvement.
Figs. 4 and 5 track the quality performance indicators for glucose and urine dipstick testing for the
Fig. 4. POC glucose testing quality assurance indicators. Number of meters s 80. Symbols: closed symbols meet threshold limits, open symbols do not meet threshold limits; v —daily QC documentation, `—within range QC, \—failed QC documentation, I—daily maintenance documentation, '—weekly maintenance documentation, e—alert value follow-up.
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
57
Fig. 5. POC urine dipstick testing quality assurance indicators. Number of testing sites s 30. Symbols: closed symbols meet threshold limits, open symbols do not meet threshold limits; v —daily QC documentation, `—proper patient ID, I—within range QC, \—failed QC documentation.
first 6 months of the program. The daily QC documentation of 100% and the average weekly maintenance documentation of 85% for glucose testing were within acceptable limits. The 93% of QC data within the acceptable range just missed the acceptability threshold of 95%. The other three glucose indicators of daily maintenance documentation Ž58%., alert value follow-up Ž58%. and documentation of corrective action taken for failed QC Ž18%. all show the need for significant improvement. Of the four quality assurance indicators for urine dipstick testing, only daily QC documentation, at 80%, was within acceptable threshold limits Ž80%.. Both proper patient ID and percentage of QC data within acceptable range, at 86% and 81%, respectively, require some improvement to meet the acceptable threshold limits of 95%. As with glucose testing, there is a significant problem with the documentation of corrective actions for failed urine dipstick QC, 0% compliance. An analysis of all the problem calls received by the POCT program is shown in Fig. 6. From January through April 2000, 105 calls were made and resolved. All of these calls were generated by only 10 of the 23 tests performed Žblood gases s blood gas, Na, K, Ca and hematocrit.. Despite being one of the
lowest volume tests perfomed, fetal scalp pH testing accounted for 15% of the calls received. Most of these were due to instrument-related problems. Glucose testing generated the greatest number of calls Ž72%., as would be expected due to the size of that testing program. The majority of these calls Ž55%. were due to the glucose meter being dirty ŽFig. 7., followed by the inability to use the device due to QC lock-out Ž21%.. Only 25% of the problems with the
Fig. 6. POCT CQI program—Incident analysis, January–April 2000, ns105.
58
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59
Fig. 7. POCT CQI program—Glucose incident resolution, January–April 2000, 30 meters involved, ns 77.
glucose testing were due to problems with the system, either device or QC solution. Thus, 75% of the time problems with glucose testing were due to some type of operator error.
4. Discussion Success of a medical center point-of-care testing program depends upon a multi-disciplinary team approach. Over the last 11 years, the regulated POCT program at the Mount Sinai Medical Center has grown from a single-analyte program with limited impact on the in-patient medical system to a 23-analyte program with significant impact on patient care throughout the entire spectrum of medical care, from in-patient to home patients. Significant changes occurred after the creation of the institutional POCT committee. Sixty one percent of all testing sites are at in-patient locations vs. 39% in out-patient locations and - 1% in non-patient locations. Out of the 23 different POC tests, glucose, occult blood and urinalysis are most commonly performed, at 66%, 55% and 35% of all test sites, respectively. Maternal and Child Health and General Medicine Care Centers have the most diverse menu of POC tests. Glucose, blood gasrelectrolytes, activated clotting time and fetal scalp pH tests are predominately performed on in-patients, whereas pregnancy, hemoglobin, PTrPTT, rapid strep A, hemoglobin A1C, urinary microalbumin and creatinine are predominately performed on out-patients and non-patients.
The implementation of a formalized CQI program helped quantify what was already previously known to be a problem area in the POCT program-documentation, i.e., instrument maintenance, manual QC tests and actions taken associated with correction of failed QC tests. More importantly, it raised the consciousness of the end-users to the significance of this documentation and has helped to improve the overall program. Furthermore, along with the analysis of the various problem calls received, the POCT coordinators have been able to adjust the training and recertification programs in order to address the identified problems. Overall, the CQI program has been well received by the entire nursing and medical staff and has effected improved communications. One of the goals of the committee is to determine the true impact of POCT on a patient’s hospital stay andror healthcare, not just the cost of providing a test result. Other future plans include establishing AJust-in-TimeB supply distribution based on recorded workload statistics.
Acknowledgements The authors would like to acknowledge Dr. Maria Vezina, Director of Nursing Education, and Ms. Julie Fisher, Associate Hospital Director, without whose support and understanding, the program would not have been successful. Additionally, a most sincere thank you to the hard work of Ms. Svetlana Babayeva, Mr. Abelardo Gorospe and Ms. Fira Steinberg, point-of-care testing chemists.
References w1x DeCresce RP, Phillips DL, Howanitz PT. Financial justification of alternative site testing. Arch Pathol Lab Med 1995; 119:899–901. w2x Jacobs E. Analysis of economic models used to justify point-of-care testing. In: D’Orazio P, editor. Preparing for critical care analysis in the twenty first century. Proceedings of an International Symposium. Medfield, MA: Electrolyter Blood Gas Div. vol. 16. Washington, DC: AACC 1996: 126–36. w3x Handord CR. POC testing: must quality cost more? Med Lab Obs 1993;9S:28–33. w4x Kost G. Guidelines for point-of-care testing: improving
E. Jacobs et al.r Clinica Chimica Acta 307 (2001) 49–59 patient outcomes. Am J Clin Pathol 1995;104ŽSuppl. 4.: S111–27. w5x Suver JD, Nuemann BR, Boles EE. Accounting for the costs of quality. Health Care Financ Manage 1992;9:29–37. w6x Jacobs E. TQM and POC testing. Med Lab Obs 1993; 25Ž9S.:2–6. w7x Jacobs E. The laboratory as a leader of point-of-care programs. Med Lab Obs 1998;30Ž9S.:6–10.
59
w8x NCCLS Approved Guideline AST2-A, Point-of-care in vitro diagnostic ŽIVD. testing. Vilanova, PA: National Committee for Clinical Laboratory Standards, 1999. June. w9x Lathrop JP, Seufert GE, MacDonald RJ, Martin SB. The patient-focused hospital: a patient care concept. J Soc Health Syst 1991;3:33–50. w10x Borzo G. Patient-focused hospitals begin reporting good results. Health Care Strategic Manage 1992;10Ž8.:17–22.