AJCP / Original Article
Impact of an Educational Intervention on the Frequency of Daily Blood Test Orders for Hospitalized Patients Rajiv N. Thakkar, MD,1 Daniel Kim, MD,1 Amy M. Knight, MD,1 Stefan Riedel, MD,2 Dhananjay Vaidya, MBBS, PhD, MPH,1 and Scott M. Wright, MD1 From the Departments of ¹Medicine and ²Pathology, Johns Hopkins University School of Medicine, Baltimore, MD. Key Words: Phlebotomy; Hospital-acquired anemia; Excessive blood testing; Blood draw; Daily blood test; Laboratory test
CME/SAM
Am J Clin Pathol March 2015;143:393-397 DOI: 10.1309/AJCPJS4EEM7UAUBV
ABSTRACT Objectives: During hospitalizations, blood is drawn for diagnostic laboratory tests to help guide patient care. Often, blood tests continue to be ordered even in the face of clinical and laboratory stability. Blood draws are painful and costly, and they may be associated with anemia. We hypothesized that provider education could reduce the frequency of daily blood tests ordered for hospitalized patients. Methods: During a 2-month intervention period, internal medicine providers were educated through flyers displayed in providers’ offices and periodic email communications reminding them to order daily blood tests only if the results would change patient care. Two-month preintervention data from 982 patients and 2-month postintervention data from 988 patients were analyzed. The primary outcome measured was the number of daily blood tests ordered per patient per day. Results: Mean orders of CBC decreased from 1.46 to 1.37 tests per patient per day (P < .05) after the intervention. Basic metabolic panel orders were reduced from 0.91 to 0.83 tests per patient per day (P < .05). Cost analyses showed a reduction of $6.33 per patient day based on the decrease in the number of daily laboratory tests ordered. Conclusions: Provider education and reminders can reduce the frequency of daily blood tests ordered by providers for hospitalized patients. This can decrease health care costs and may reduce the risk of complications such as anemia.
© American Society for Clinical Pathology
Upon completion of this activity you will be able to: • discuss the risks and cost associated with excessive and or daily phlebotomy. • list factors and risks associated with hospital-acquired anemia. • apply a simple intervention based on available guidelines and evidence to reduce daily blood draws in hospitalized patients. The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 Credit ™ per article. Physicians should claim only the credit commensurate with the extent of their participation in the activity. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module. The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose. Questions appear on p 460. Exam is located at www.ascp.org/ajcpcme.
Patients undergo blood draw (phlebotomy) for laboratory testing when they are hospitalized for acute illnesses.1,2 Both the increase in acuity of hospitalized patients and the availability of new laboratory tests to diagnose or monitor disease have led to increasing numbers of laboratory and other diagnostic test orders. Such trends are evident from recent statistics published by the Centers for Medicare & Medicaid Services (CMS) showing that Medicare’s spending on physician services per beneficiary from 2000 to 2009 rose at a rate of 5.4% a year.3 Beyond higher costs, anemia is commonly encountered in hospitalized patients.4,5 Excessive phlebotomy lowers patients’ hemoglobin levels and can result in hospital-acquired anemia (HAA).6,7 HAA has been associated with increased length of stay, need for blood transfusions, and increased mortality.8-10 HAA is more common in patients with specific admitting diagnoses, and it develops in nearly 50% of patients with acute myocardial infarction and is associated with higher mortality.11,12 Despite this evidence, health care providers routinely order
Am J Clin Pathol 2015;143:393-397 393 DOI: 10.1309/AJCPJS4EEM7UAUBV
Thakkar et al / Impact of Education on Daily Blood Test Orders
Baltimore, Maryland. On the general medical unit, care is delivered and orders are typically placed by (1) members of the hospitalist division (physicians, nurse practitioners, and physician assistants) and (2) inpatient teaching teams (attending physicians, residents, and interns). The Johns Hopkins Medicine Institutional Review Board approved the study. Intervention During March and April 2012, we carried out a 2-month active educational intervention on the general medicine units. The components of our intervention included (1) educational sessions with providers and nurses in the form of interactive didactic presentations as well as discussions at division meetings and noon conferences, (2) educational flyers posted in provider and nurse work areas ❚Image 1❚, and (3) weekly email communications to all providers (faculty members, house staff, nurse practitioners, and physician assistants) and nurses. Consistent across all elements of the intervention were three messages or pieces of information: providers should (1) question the utility of every blood test and order the tests only if the result will affect patient care, (2) think about the sizable impact that costs of blood tests have on health care expenditures, and (3) consider “adding on” tests to blood samples that have already been collected whenever possible. ❚Image 1❚ Educational flyer.
daily blood tests, most commonly CBC, basic metabolic panel (BMP), complete metabolic panel (CMP), and coagulation panels (prothrombin time [PT], partial thromboplastin time [PTT], and international normalized ratio [INR]) in hospitalized patients even in the face of clinical and laboratory stability. Phlebotomy is invasive and painful, and it contributes to soaring health care costs. One of the Society of Hospital Medicine’s and American Society for Clinical Pathology’s recommendations to the American Board of Internal Medicine’s “Choosing Wisely” initiative is for providers and patients to question the need for daily blood tests for stable hospitalized patients and for preoperative blood testing in low-risk surgery patients.13 We conducted this study to determine the effect of an educational intervention for providers and nurses on the frequency of daily blood tests ordered in hospitalized adult patients.
Materials and Methods Setting and Design This prospective pre-postinterventional educational study was conducted at a 400-bed academic hospital in
394 Am J Clin Pathol 2015;143:393-397 DOI: 10.1309/AJCPJS4EEM7UAUBV
Data Collection Data were collected from our computerized provider order entry system (Meditech, Westwood, MA). We used the same two calendar months for the preintervention period (May and June 2011) and the postintervention period (May and June 2012) to minimize the likelihood of seasonal variation in admitting diagnoses and to ensure that the orders placed by trainees (interns and residents) were put in by individuals with equal amounts of experience. No interventions or changes that relate to laboratory ordering or utilization either at the hospital or larger institutional level were made to the order entry system between January 2011 and February 2012. Records from patients who were 18 years and older, were admitted through the emergency department to the general medicine service, and had had at least one daily blood test (CBC, BMP, CMP, PT/INR, PTT) during their hospital stay were included in this study. Patients with gastrointestinal bleeding, acute renal insufficiency, and congestive heart failure are appropriately monitored with frequent laboratory testing and therefore were excluded from the study. Patients transferred to or from a higher level of care (eg, intensive care unit) or to a different clinical service (eg, neurology) were also excluded. Patient demographic and clinical data collected included age, sex, warfarin use, length of stay, and in-hospital
© American Society for Clinical Pathology
AJCP / Original Article
mortality. Patient laboratory data collected included the number of CBC, BMP, CMP, and coagulation panels ordered during the hospitalization, with the data reported as number ordered per patient per day. Data Analysis We tabulated and compared the demographic and clinical characteristics of patients in the preintervention and postintervention periods. Categorical variables were compared by using c2 tests. We used nonparametric rank sum tests for length of stay comparison. Continuous variables (eg, age) and data such as whether the number of daily blood tests ordered per patient per day differed between the two periods were compared by using t tests. We further estimated the total laboratory cost (in dollars) of all the tests of interest performed in each of the two periods and costs per patient day based on the current charges of each test as defined by the CMS. These charges were used in the analyses because they represent a common standard and costs at our institution are fairly close to the CMS rates. Data were analyzed by using STATA software (version 13; StataCorp LP, College Station, TX). Tests were considered significant when P < .05. For the comparative purpose of assessing the impact of the intervention on test ordering across the two time periods, we analyzed laboratory test ordering in other clinical services in the Department of Medicine, the medical and cardiac intensive care units, and the progressive care units. No differences were seen in these units (all P > .05; specific data not shown).
Results Records and laboratory data from 1,970 patients in the preintervention and postintervention groups were compared and are summarized in ❚Table 1❚. There were a higher number of male patients in the postintervention group (50.4% vs 43.5%, P = .002). The pre- and postintervention groups were otherwise similar. There were statistically significant reductions in the number of most targeted tests ordered after the intervention ❚Table 2❚. The mean number of CBCs per day per patient decreased from 1.46 to 1.37 tests (P < .05). There was a reduction in BMP orders from 0.91 to 0.83 tests per patient per day (P < .05). PT orders decreased from 0.59 to 0.50 (P < .05), and PTT reduced from 0.53 to 0.43 per patient per day (P < .05). There was a statistically significant reduction in the number of overall test orders (4.15 vs 3.79 per patient per day, P < .05). ❚Table 3❚ summarizes cost comparisons of the laboratory tests ordered between the two periods. The total number
© American Society for Clinical Pathology
❚Table 1❚ Comparison of Patient Data Across the Two Time Periods Preintervention Postintervention (n = 982) (n = 988) P Value
Patient Data Age, mean (SD), y Male sex, No. (%) Length of stay, median (IQR), d In-hospital mortality, No. (%) Warfarin use, No. (%)
60.5 (17.8) 427 (43.5) 3 (2-5)
60.3 (18.3) 498 (50.4) 3 (2-5)
.867 .002 .683
6 (0.6)
5 (0.5)
.755
122 (12.4)
118 (11.9)
.745
IQR, interquartile range.
❚Table 2❚ Mean (SD) Number of Daily Blood Tests per Patient per Day Across the Two Time Periods Test
Preintervention Postintervention (n = 982) (n = 988) P Value
CBC BMP CMP PT PTT Any test
1.46 (0.67) 0.91 (0.65) 0.65 (0.47) 0.59 (0.56) 0.53 (0.53) 4.15 (2.02)
1.37 (0.69) 0.83 (0.62) 0.64 (0.47) 0.50 (0.61) 0.43 (0.57) 3.79 (2.09)
.003 .008 .611 .001 .001 .001
BMP, basic metabolic panel; CMP, complete metabolic panel; PT, prothrombin time; PTT, partial thromboplastin time.
of tests ordered decreased from 13,742 in the preintervention period to 13,528 in the postintervention period. The total cost per patient day decreased from $48.73 to $42.40; this results in a net saving of $6.33 per patient day.
Discussion This study shows that an educational intervention with providers and nurses focused mainly on heightening awareness related to discomfort, cost, and risk of unnecessary blood draws was associated with about 9% reductions in mean daily blood test orders. This decrease in testing may very well have translated into less phlebotomyrelated discomfort, reduced workload for phlebotomists, less data for providers to follow up on and interpret, and lower costs. Anemia is highly prevalent among unselected internal medicine inpatients, and unfortunately, it is associated with prolonged hospital stays, higher readmission rates, and increased mortality.14 Literature reviews show a strong correlation between phlebotomy and change in hematocrit levels.2,15 Factors known to contribute to HAA include the number of laboratory tests ordered, phlebotomy volumes collected, and tube sizes. A small decrease in phlebotomy volumes has been associated with a reduced need for blood transfusion among intensive care unit patients.9 Similarly,
Am J Clin Pathol 2015;143:393-397 395 DOI: 10.1309/AJCPJS4EEM7UAUBV
Thakkar et al / Impact of Education on Daily Blood Test Orders
❚Table 3❚ Cost Comparisons for Daily Blood Tests Across the Two Time Periods Preintervention: 3,936 Patient Days
Postintervention: 4,480 Patient Days
Test (National Average Charges)
Total No. of Tests
Total Cost, $
Cost, $/Patient Day
Total No. of Tests
Total Cost, $
Cost, $/Patient Day
Cost Difference, $/Patient Day
CBC ($14.34) BMP ($15.59) CMP ($19.48) PT ($7.25) PTT ($11.07) Total
4,820 3,165 2,040 1,973 1,744 13,742
69,118.80 49,342.35 39,739.20 14,304.25 19,306.08 191,810.68
17.56 12.54 10.10 3.63 4.91 48.73
4,836 3,176 2,035 1,857 1,624 13,528
69,348.24 49,513.84 39,641.80 13,463.25 17,977.68 189,944.81
15.48 11.05 8.85 3.01 4.01 42.40
–2.08 –1.49 –1.25 –0.62 –0.90 –6.33
BMP, basic metabolic panel; CMP, complete metabolic panel; PT, prothrombin time; PTT, partial thromboplastin time.
blood conservation devices have also been associated with a reduction in blood transfusion requirements and smaller decreases in hemoglobin among hospitalized patients.16,17 Use of pediatric-sized test tubes reduces the blood loss from diagnostic phlebotomy and may improve clinical outcomes.18-20 Wisser et al1 showed that reporting cumulative patient blood loss to providers influenced behaviors and translated into decreased laboratory testing. Likewise, making health care providers aware of the cost of laboratory tests has been shown to decrease the frequency of blood test orders and can result in significant savings for the hospital.21,22 The primary focus of our study was the frequency of blood draws. Our low-cost intervention—short talks, flyers, emails, and heightened awareness—encouraged providers to order blood tests based on the clinical need. Our intervention did not involve any elements that would compromise provider work flow or efficiency. It is as if ordering daily blood tests represents a habit and that providers have grown accustomed to this culture of collecting blood, which may not be based on patients’ need. The fact that an educational intervention and reminders were able to significantly decrease the numbers of all five daily laboratory test orders shows that the clinical indications for daily blood draws are often missing. Given the current state of health care financing and expenditures in this country, everyone is looking for cost reduction strategies. One could argue that decreasing spending on routine laboratory testing from about $49 to $42 does not represent savings that are adequately substantial to deserve much interest or attention. However, because these savings can be applied to every hospitalized patient every day, the cumulative effect can be significant. An extrapolated reduction in spending on our medicine ward service alone is greater than $150,000 annually. Also to be factored into the equation is the cost of the intervention, which actually was very low. Several limitations of this study should be considered. First, because the study used a pre-post design and was conducted at a single center, generalizability of these 396 Am J Clin Pathol 2015;143:393-397 DOI: 10.1309/AJCPJS4EEM7UAUBV
findings may be limited. It should be noted that there were no other similar institutional interventions locally or nationally addressing this topic during the data collection phases. Second, postintervention data collected were immediately following the intervention; the long-term effects are not known, and it may be possible that fatigue or diminished attention to reminders will make providers less responsive over time. Third, we did not assess clinical outcomes such as anemia. However, 30-day all-cause readmission rates at our facility for the second quarter of fiscal year 2012 (April-June 2012, which coincides with our postintervention period) did not increase compared with the preintervention time frame. Fourth, although the number of blood tests per patient per day reduced, we did not collect data on the number of sticks for blood draw per patient, which may or may not have changed. Finally, as in all nonrandomized pre-post intervention studies, the resulting data suggest associations and cannot prove causality. An educational intervention with guidance about daily laboratory test utilization can reduce phlebotomy frequency and lower laboratory cost for hospitalized patients. Address reprint requests to Dr Thakkar: Division of Hospital Medicine, Johns Hopkins Bayview Medical Center, 5200 Eastern Ave, Mason F. Lord West Tower, 6th Floor, Baltimore, MD 21224; [email protected]. Research analysis was supported by NIH grant UL1TR001079 from the Johns Hopkins Institute for Clinical and Translational Research.
References 1. Wisser D, van Ackern K, Knoll E, et al. Blood loss from laboratory tests. Clin Chem. 2003;49:1651-1655. 2. Thavendiranathan P, Bagai A, Ebidia A, et al. Do blood tests cause anemia in hospitalized patients? the effect of diagnostic phlebotomy on hemoglobin and hematocrit levels. J Gen Intern Med. 2005;20:520-524. 3. Fees, volume and spending at Medicare. New York Times. http://economix.blogs.nytimes.com/2010/12/24/fees-volumeand-spending-at-medicare/. Accessed April 9, 2014.
© American Society for Clinical Pathology
AJCP / Original Article
4. Gianserra CV, Aguero AP, Chapelet AG, et al. Hospitalacquired anemia and decrease of hemoglobin levels in hospitalized patients [in Spanish]. Medicina (B Aires). 2011;71:201-206. 5. Lin RJ, Evans AT, Chused AE, et al. Anemia in general medical inpatients prolongs length of stay and increases 30-day unplanned readmission rate. South Med J. 2013;106:316-320. 6. Tosiri P, Kanitsap N, Kanitsap A. Approximate iatrogenic blood loss in medical intensive care patients and the causes of anemia. J Med Assoc Thai. 2010;93(suppl 7):S271-S276. 7. Languasco A, Cazap N, Marciano S, et al. Hemoglobin concentration variations over time in general medical inpatients. J Hosp Med. 2010;5:283-288. 8. Merono O, Cladellas M, Recasens L, et al. In-hospital acquired anemia in acute coronary syndrome: predictors, in-hospital prognosis and one-year mortality. Rev Esp Cardiol (Engl Ed). 2012;65:742-748. 9. Chant C, Wilson G, Friedrich JO. Anemia, transfusion, and phlebotomy practices in critically ill patients with prolonged ICU length of stay: a cohort study. Crit Care. 2006;10:R140. 10. Koch CG, Li L, Sun Z, et al. Hospital-acquired anemia: prevalence, outcomes, and healthcare implications. J Hosp Med. 2013;8:506-512. 11. Salisbury AC, Alexander KP, Reid KJ, et al. Incidence, correlates, and outcomes of acute, hospital-acquired anemia in patients with acute myocardial infarction. Circ Cardiovasc Qual Outcomes. 2010;3:337-346. 12. Salisbury AC, Amin AP, Reid KJ, et al. Hospital-acquired anemia and in-hospital mortality in patients with acute myocardial infarction. Am Heart J. 2011;162:300-309.e3. 13. Choosing Wisely. An initiative of ABIM Foundation. http:// www.choosingwisely.org/. Accessed May 29, 2014.
© American Society for Clinical Pathology
14. Nathavitharana RL, Murray JA, D’Sousa N, et al. Anaemia is highly prevalent among unselected internal medicine inpatients and is associated with increased mortality, earlier readmission and more prolonged hospital stay: an observational retrospective cohort study. Intern Med J. 2012;42:683-691. 15. Salisbury AC, Reid KJ, Alexander KP, et al. Diagnostic blood loss from phlebotomy and hospital-acquired anemia during acute myocardial infarction. Arch Intern Med. 2011;171:1646-1653. 16. Mukhopadhyay A, Yip HS, Prabhuswamy D, et al. The use of a blood conservation device to reduce red blood cell transfusion requirements: a before and after study. Crit Care. 2010;14:R7. 17. Page C, Retter A, Wyncoll D. Blood conservation devices in critical care: a narrative review. Ann Intensive Care. 2013;3:14. 18. Sanchez-Giron F, Alvarez-Mora F. Reduction of blood loss from laboratory testing in hospitalized adult patients using small-volume (pediatric) tubes. Arch Pathol Lab Med. 2008;132:1916-1919. 19. Valentine SL, Bateman ST. Identifying factors to minimize phlebotomy-induced blood loss in the pediatric intensive care unit. Pediatr Crit Care Med. 2012;13:22-27. 20. McPherson RA. Blood sample volumes: emerging trends in clinical practice and laboratory medicine. Clin Leadersh Manag Rev. 2001;15:3-10. 21. Stuebing EA, Miner TJ. Surgical vampires and rising health care expenditure: reducing the cost of daily phlebotomy. Arch Surg. 2011;146:524-527. 22. Feldman LS, Shihab HM, Thiemann D, et al. Impact of providing fee data on laboratory test ordering: a controlled clinical trial. JAMA Intern Med. 2013;173:903-908.
Am J Clin Pathol 2015;143:393-397 397 DOI: 10.1309/AJCPJS4EEM7UAUBV
Impact of an Educational Intervention on the Frequency of Daily Blood Test Orders for Hospitalized Patients Rajiv N. Thakkar, MD,1 Daniel Kim, MD,1 Amy M. Knight, MD,1 Stefan Riedel, MD,2 Dhananjay Vaidya, MBBS, PhD, MPH,1 and Scott M. Wright, MD1 From the Departments of ¹Medicine and ²Pathology, Johns Hopkins University School of Medicine, Baltimore, MD. Key Words: Phlebotomy; Hospital-acquired anemia; Excessive blood testing; Blood draw; Daily blood test; Laboratory test
CME/SAM
Am J Clin Pathol March 2015;143:393-397 DOI: 10.1309/AJCPJS4EEM7UAUBV
ABSTRACT Objectives: During hospitalizations, blood is drawn for diagnostic laboratory tests to help guide patient care. Often, blood tests continue to be ordered even in the face of clinical and laboratory stability. Blood draws are painful and costly, and they may be associated with anemia. We hypothesized that provider education could reduce the frequency of daily blood tests ordered for hospitalized patients. Methods: During a 2-month intervention period, internal medicine providers were educated through flyers displayed in providers’ offices and periodic email communications reminding them to order daily blood tests only if the results would change patient care. Two-month preintervention data from 982 patients and 2-month postintervention data from 988 patients were analyzed. The primary outcome measured was the number of daily blood tests ordered per patient per day. Results: Mean orders of CBC decreased from 1.46 to 1.37 tests per patient per day (P < .05) after the intervention. Basic metabolic panel orders were reduced from 0.91 to 0.83 tests per patient per day (P < .05). Cost analyses showed a reduction of $6.33 per patient day based on the decrease in the number of daily laboratory tests ordered. Conclusions: Provider education and reminders can reduce the frequency of daily blood tests ordered by providers for hospitalized patients. This can decrease health care costs and may reduce the risk of complications such as anemia.
© American Society for Clinical Pathology
Upon completion of this activity you will be able to: • discuss the risks and cost associated with excessive and or daily phlebotomy. • list factors and risks associated with hospital-acquired anemia. • apply a simple intervention based on available guidelines and evidence to reduce daily blood draws in hospitalized patients. The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 Credit ™ per article. Physicians should claim only the credit commensurate with the extent of their participation in the activity. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module. The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose. Questions appear on p 460. Exam is located at www.ascp.org/ajcpcme.
Patients undergo blood draw (phlebotomy) for laboratory testing when they are hospitalized for acute illnesses.1,2 Both the increase in acuity of hospitalized patients and the availability of new laboratory tests to diagnose or monitor disease have led to increasing numbers of laboratory and other diagnostic test orders. Such trends are evident from recent statistics published by the Centers for Medicare & Medicaid Services (CMS) showing that Medicare’s spending on physician services per beneficiary from 2000 to 2009 rose at a rate of 5.4% a year.3 Beyond higher costs, anemia is commonly encountered in hospitalized patients.4,5 Excessive phlebotomy lowers patients’ hemoglobin levels and can result in hospital-acquired anemia (HAA).6,7 HAA has been associated with increased length of stay, need for blood transfusions, and increased mortality.8-10 HAA is more common in patients with specific admitting diagnoses, and it develops in nearly 50% of patients with acute myocardial infarction and is associated with higher mortality.11,12 Despite this evidence, health care providers routinely order
Am J Clin Pathol 2015;143:393-397 393 DOI: 10.1309/AJCPJS4EEM7UAUBV
Thakkar et al / Impact of Education on Daily Blood Test Orders
Baltimore, Maryland. On the general medical unit, care is delivered and orders are typically placed by (1) members of the hospitalist division (physicians, nurse practitioners, and physician assistants) and (2) inpatient teaching teams (attending physicians, residents, and interns). The Johns Hopkins Medicine Institutional Review Board approved the study. Intervention During March and April 2012, we carried out a 2-month active educational intervention on the general medicine units. The components of our intervention included (1) educational sessions with providers and nurses in the form of interactive didactic presentations as well as discussions at division meetings and noon conferences, (2) educational flyers posted in provider and nurse work areas ❚Image 1❚, and (3) weekly email communications to all providers (faculty members, house staff, nurse practitioners, and physician assistants) and nurses. Consistent across all elements of the intervention were three messages or pieces of information: providers should (1) question the utility of every blood test and order the tests only if the result will affect patient care, (2) think about the sizable impact that costs of blood tests have on health care expenditures, and (3) consider “adding on” tests to blood samples that have already been collected whenever possible. ❚Image 1❚ Educational flyer.
daily blood tests, most commonly CBC, basic metabolic panel (BMP), complete metabolic panel (CMP), and coagulation panels (prothrombin time [PT], partial thromboplastin time [PTT], and international normalized ratio [INR]) in hospitalized patients even in the face of clinical and laboratory stability. Phlebotomy is invasive and painful, and it contributes to soaring health care costs. One of the Society of Hospital Medicine’s and American Society for Clinical Pathology’s recommendations to the American Board of Internal Medicine’s “Choosing Wisely” initiative is for providers and patients to question the need for daily blood tests for stable hospitalized patients and for preoperative blood testing in low-risk surgery patients.13 We conducted this study to determine the effect of an educational intervention for providers and nurses on the frequency of daily blood tests ordered in hospitalized adult patients.
Materials and Methods Setting and Design This prospective pre-postinterventional educational study was conducted at a 400-bed academic hospital in
394 Am J Clin Pathol 2015;143:393-397 DOI: 10.1309/AJCPJS4EEM7UAUBV
Data Collection Data were collected from our computerized provider order entry system (Meditech, Westwood, MA). We used the same two calendar months for the preintervention period (May and June 2011) and the postintervention period (May and June 2012) to minimize the likelihood of seasonal variation in admitting diagnoses and to ensure that the orders placed by trainees (interns and residents) were put in by individuals with equal amounts of experience. No interventions or changes that relate to laboratory ordering or utilization either at the hospital or larger institutional level were made to the order entry system between January 2011 and February 2012. Records from patients who were 18 years and older, were admitted through the emergency department to the general medicine service, and had had at least one daily blood test (CBC, BMP, CMP, PT/INR, PTT) during their hospital stay were included in this study. Patients with gastrointestinal bleeding, acute renal insufficiency, and congestive heart failure are appropriately monitored with frequent laboratory testing and therefore were excluded from the study. Patients transferred to or from a higher level of care (eg, intensive care unit) or to a different clinical service (eg, neurology) were also excluded. Patient demographic and clinical data collected included age, sex, warfarin use, length of stay, and in-hospital
© American Society for Clinical Pathology
AJCP / Original Article
mortality. Patient laboratory data collected included the number of CBC, BMP, CMP, and coagulation panels ordered during the hospitalization, with the data reported as number ordered per patient per day. Data Analysis We tabulated and compared the demographic and clinical characteristics of patients in the preintervention and postintervention periods. Categorical variables were compared by using c2 tests. We used nonparametric rank sum tests for length of stay comparison. Continuous variables (eg, age) and data such as whether the number of daily blood tests ordered per patient per day differed between the two periods were compared by using t tests. We further estimated the total laboratory cost (in dollars) of all the tests of interest performed in each of the two periods and costs per patient day based on the current charges of each test as defined by the CMS. These charges were used in the analyses because they represent a common standard and costs at our institution are fairly close to the CMS rates. Data were analyzed by using STATA software (version 13; StataCorp LP, College Station, TX). Tests were considered significant when P < .05. For the comparative purpose of assessing the impact of the intervention on test ordering across the two time periods, we analyzed laboratory test ordering in other clinical services in the Department of Medicine, the medical and cardiac intensive care units, and the progressive care units. No differences were seen in these units (all P > .05; specific data not shown).
Results Records and laboratory data from 1,970 patients in the preintervention and postintervention groups were compared and are summarized in ❚Table 1❚. There were a higher number of male patients in the postintervention group (50.4% vs 43.5%, P = .002). The pre- and postintervention groups were otherwise similar. There were statistically significant reductions in the number of most targeted tests ordered after the intervention ❚Table 2❚. The mean number of CBCs per day per patient decreased from 1.46 to 1.37 tests (P < .05). There was a reduction in BMP orders from 0.91 to 0.83 tests per patient per day (P < .05). PT orders decreased from 0.59 to 0.50 (P < .05), and PTT reduced from 0.53 to 0.43 per patient per day (P < .05). There was a statistically significant reduction in the number of overall test orders (4.15 vs 3.79 per patient per day, P < .05). ❚Table 3❚ summarizes cost comparisons of the laboratory tests ordered between the two periods. The total number
© American Society for Clinical Pathology
❚Table 1❚ Comparison of Patient Data Across the Two Time Periods Preintervention Postintervention (n = 982) (n = 988) P Value
Patient Data Age, mean (SD), y Male sex, No. (%) Length of stay, median (IQR), d In-hospital mortality, No. (%) Warfarin use, No. (%)
60.5 (17.8) 427 (43.5) 3 (2-5)
60.3 (18.3) 498 (50.4) 3 (2-5)
.867 .002 .683
6 (0.6)
5 (0.5)
.755
122 (12.4)
118 (11.9)
.745
IQR, interquartile range.
❚Table 2❚ Mean (SD) Number of Daily Blood Tests per Patient per Day Across the Two Time Periods Test
Preintervention Postintervention (n = 982) (n = 988) P Value
CBC BMP CMP PT PTT Any test
1.46 (0.67) 0.91 (0.65) 0.65 (0.47) 0.59 (0.56) 0.53 (0.53) 4.15 (2.02)
1.37 (0.69) 0.83 (0.62) 0.64 (0.47) 0.50 (0.61) 0.43 (0.57) 3.79 (2.09)
.003 .008 .611 .001 .001 .001
BMP, basic metabolic panel; CMP, complete metabolic panel; PT, prothrombin time; PTT, partial thromboplastin time.
of tests ordered decreased from 13,742 in the preintervention period to 13,528 in the postintervention period. The total cost per patient day decreased from $48.73 to $42.40; this results in a net saving of $6.33 per patient day.
Discussion This study shows that an educational intervention with providers and nurses focused mainly on heightening awareness related to discomfort, cost, and risk of unnecessary blood draws was associated with about 9% reductions in mean daily blood test orders. This decrease in testing may very well have translated into less phlebotomyrelated discomfort, reduced workload for phlebotomists, less data for providers to follow up on and interpret, and lower costs. Anemia is highly prevalent among unselected internal medicine inpatients, and unfortunately, it is associated with prolonged hospital stays, higher readmission rates, and increased mortality.14 Literature reviews show a strong correlation between phlebotomy and change in hematocrit levels.2,15 Factors known to contribute to HAA include the number of laboratory tests ordered, phlebotomy volumes collected, and tube sizes. A small decrease in phlebotomy volumes has been associated with a reduced need for blood transfusion among intensive care unit patients.9 Similarly,
Am J Clin Pathol 2015;143:393-397 395 DOI: 10.1309/AJCPJS4EEM7UAUBV
Thakkar et al / Impact of Education on Daily Blood Test Orders
❚Table 3❚ Cost Comparisons for Daily Blood Tests Across the Two Time Periods Preintervention: 3,936 Patient Days
Postintervention: 4,480 Patient Days
Test (National Average Charges)
Total No. of Tests
Total Cost, $
Cost, $/Patient Day
Total No. of Tests
Total Cost, $
Cost, $/Patient Day
Cost Difference, $/Patient Day
CBC ($14.34) BMP ($15.59) CMP ($19.48) PT ($7.25) PTT ($11.07) Total
4,820 3,165 2,040 1,973 1,744 13,742
69,118.80 49,342.35 39,739.20 14,304.25 19,306.08 191,810.68
17.56 12.54 10.10 3.63 4.91 48.73
4,836 3,176 2,035 1,857 1,624 13,528
69,348.24 49,513.84 39,641.80 13,463.25 17,977.68 189,944.81
15.48 11.05 8.85 3.01 4.01 42.40
–2.08 –1.49 –1.25 –0.62 –0.90 –6.33
BMP, basic metabolic panel; CMP, complete metabolic panel; PT, prothrombin time; PTT, partial thromboplastin time.
blood conservation devices have also been associated with a reduction in blood transfusion requirements and smaller decreases in hemoglobin among hospitalized patients.16,17 Use of pediatric-sized test tubes reduces the blood loss from diagnostic phlebotomy and may improve clinical outcomes.18-20 Wisser et al1 showed that reporting cumulative patient blood loss to providers influenced behaviors and translated into decreased laboratory testing. Likewise, making health care providers aware of the cost of laboratory tests has been shown to decrease the frequency of blood test orders and can result in significant savings for the hospital.21,22 The primary focus of our study was the frequency of blood draws. Our low-cost intervention—short talks, flyers, emails, and heightened awareness—encouraged providers to order blood tests based on the clinical need. Our intervention did not involve any elements that would compromise provider work flow or efficiency. It is as if ordering daily blood tests represents a habit and that providers have grown accustomed to this culture of collecting blood, which may not be based on patients’ need. The fact that an educational intervention and reminders were able to significantly decrease the numbers of all five daily laboratory test orders shows that the clinical indications for daily blood draws are often missing. Given the current state of health care financing and expenditures in this country, everyone is looking for cost reduction strategies. One could argue that decreasing spending on routine laboratory testing from about $49 to $42 does not represent savings that are adequately substantial to deserve much interest or attention. However, because these savings can be applied to every hospitalized patient every day, the cumulative effect can be significant. An extrapolated reduction in spending on our medicine ward service alone is greater than $150,000 annually. Also to be factored into the equation is the cost of the intervention, which actually was very low. Several limitations of this study should be considered. First, because the study used a pre-post design and was conducted at a single center, generalizability of these 396 Am J Clin Pathol 2015;143:393-397 DOI: 10.1309/AJCPJS4EEM7UAUBV
findings may be limited. It should be noted that there were no other similar institutional interventions locally or nationally addressing this topic during the data collection phases. Second, postintervention data collected were immediately following the intervention; the long-term effects are not known, and it may be possible that fatigue or diminished attention to reminders will make providers less responsive over time. Third, we did not assess clinical outcomes such as anemia. However, 30-day all-cause readmission rates at our facility for the second quarter of fiscal year 2012 (April-June 2012, which coincides with our postintervention period) did not increase compared with the preintervention time frame. Fourth, although the number of blood tests per patient per day reduced, we did not collect data on the number of sticks for blood draw per patient, which may or may not have changed. Finally, as in all nonrandomized pre-post intervention studies, the resulting data suggest associations and cannot prove causality. An educational intervention with guidance about daily laboratory test utilization can reduce phlebotomy frequency and lower laboratory cost for hospitalized patients. Address reprint requests to Dr Thakkar: Division of Hospital Medicine, Johns Hopkins Bayview Medical Center, 5200 Eastern Ave, Mason F. Lord West Tower, 6th Floor, Baltimore, MD 21224; [email protected]. Research analysis was supported by NIH grant UL1TR001079 from the Johns Hopkins Institute for Clinical and Translational Research.
References 1. Wisser D, van Ackern K, Knoll E, et al. Blood loss from laboratory tests. Clin Chem. 2003;49:1651-1655. 2. Thavendiranathan P, Bagai A, Ebidia A, et al. Do blood tests cause anemia in hospitalized patients? the effect of diagnostic phlebotomy on hemoglobin and hematocrit levels. J Gen Intern Med. 2005;20:520-524. 3. Fees, volume and spending at Medicare. New York Times. http://economix.blogs.nytimes.com/2010/12/24/fees-volumeand-spending-at-medicare/. Accessed April 9, 2014.
© American Society for Clinical Pathology
AJCP / Original Article
4. Gianserra CV, Aguero AP, Chapelet AG, et al. Hospitalacquired anemia and decrease of hemoglobin levels in hospitalized patients [in Spanish]. Medicina (B Aires). 2011;71:201-206. 5. Lin RJ, Evans AT, Chused AE, et al. Anemia in general medical inpatients prolongs length of stay and increases 30-day unplanned readmission rate. South Med J. 2013;106:316-320. 6. Tosiri P, Kanitsap N, Kanitsap A. Approximate iatrogenic blood loss in medical intensive care patients and the causes of anemia. J Med Assoc Thai. 2010;93(suppl 7):S271-S276. 7. Languasco A, Cazap N, Marciano S, et al. Hemoglobin concentration variations over time in general medical inpatients. J Hosp Med. 2010;5:283-288. 8. Merono O, Cladellas M, Recasens L, et al. In-hospital acquired anemia in acute coronary syndrome: predictors, in-hospital prognosis and one-year mortality. Rev Esp Cardiol (Engl Ed). 2012;65:742-748. 9. Chant C, Wilson G, Friedrich JO. Anemia, transfusion, and phlebotomy practices in critically ill patients with prolonged ICU length of stay: a cohort study. Crit Care. 2006;10:R140. 10. Koch CG, Li L, Sun Z, et al. Hospital-acquired anemia: prevalence, outcomes, and healthcare implications. J Hosp Med. 2013;8:506-512. 11. Salisbury AC, Alexander KP, Reid KJ, et al. Incidence, correlates, and outcomes of acute, hospital-acquired anemia in patients with acute myocardial infarction. Circ Cardiovasc Qual Outcomes. 2010;3:337-346. 12. Salisbury AC, Amin AP, Reid KJ, et al. Hospital-acquired anemia and in-hospital mortality in patients with acute myocardial infarction. Am Heart J. 2011;162:300-309.e3. 13. Choosing Wisely. An initiative of ABIM Foundation. http:// www.choosingwisely.org/. Accessed May 29, 2014.
© American Society for Clinical Pathology
14. Nathavitharana RL, Murray JA, D’Sousa N, et al. Anaemia is highly prevalent among unselected internal medicine inpatients and is associated with increased mortality, earlier readmission and more prolonged hospital stay: an observational retrospective cohort study. Intern Med J. 2012;42:683-691. 15. Salisbury AC, Reid KJ, Alexander KP, et al. Diagnostic blood loss from phlebotomy and hospital-acquired anemia during acute myocardial infarction. Arch Intern Med. 2011;171:1646-1653. 16. Mukhopadhyay A, Yip HS, Prabhuswamy D, et al. The use of a blood conservation device to reduce red blood cell transfusion requirements: a before and after study. Crit Care. 2010;14:R7. 17. Page C, Retter A, Wyncoll D. Blood conservation devices in critical care: a narrative review. Ann Intensive Care. 2013;3:14. 18. Sanchez-Giron F, Alvarez-Mora F. Reduction of blood loss from laboratory testing in hospitalized adult patients using small-volume (pediatric) tubes. Arch Pathol Lab Med. 2008;132:1916-1919. 19. Valentine SL, Bateman ST. Identifying factors to minimize phlebotomy-induced blood loss in the pediatric intensive care unit. Pediatr Crit Care Med. 2012;13:22-27. 20. McPherson RA. Blood sample volumes: emerging trends in clinical practice and laboratory medicine. Clin Leadersh Manag Rev. 2001;15:3-10. 21. Stuebing EA, Miner TJ. Surgical vampires and rising health care expenditure: reducing the cost of daily phlebotomy. Arch Surg. 2011;146:524-527. 22. Feldman LS, Shihab HM, Thiemann D, et al. Impact of providing fee data on laboratory test ordering: a controlled clinical trial. JAMA Intern Med. 2013;173:903-908.
Am J Clin Pathol 2015;143:393-397 397 DOI: 10.1309/AJCPJS4EEM7UAUBV
