Cost display and cost charge induce different test ordering behavior depending on the healthcare setting.
Objectives: To study the effects on the number of laboratory tests ordered after introduction of cost display (showing the cost in the computerized test ordering system at test ordering and test result delivery) and cost charge (requiring all primary healthcare centers to pay full laboratory costs of the ordered tests).
Study Design: The study included cost display for secondary healthcare centers (inpatient hospitals, emergency departments, and outpatient specialist providers) as well as publicly and privately operated primary healthcare centers (sites of nonemergency, nonspecialist healthcare). After 3 months, cost charge was introduced by management for all primary healthcare centers.
Methods: Information on laboratory test cost was appended to the laboratory test name in the test ordering system, resulting in cost display both at the moment of test ordering and at the presentation of the test result. Numbers of laboratory tests were obtained from the laboratory information system and calculated as tests per physician visit. Cost charge was managed through the established laboratory invoicing system.
Results: In the publicly operated primary healthcare centers, neither of the interventions had any effect on laboratory test volume, nor did cost display have an effect in the privately operated primary healthcare centers. However, introduction of cost charge significantly decreased laboratory test ordering in the privately operated primary healthcare centers. In contrast, secondary healthcare centers lowered test volumes when cost display was introduced.
Conclusions: The results support cost awareness and cost charge as means of reducing laboratory utilization. However, the outcome varies with the setting.
Am J Manag Care. 2018;24(5):e164-e169Takeaway Points
Cost display (showing the cost in the computerized test ordering system at test ordering and test result delivery) and cost charge (requiring all primary healthcare centers to pay full laboratory costs of the ordered tests) can reduce laboratory test ordering, although the effect is dependent on the healthcare setting.
The utilization of laboratory services is increasing both worldwide and in Sweden,1-3 which adds to the strain on the limited resources of the healthcare sector. Laboratory services provide the main data source that supports physicians in most medical decisions,4 and a substantial portion of the costs is generated by clinical chemistry tests.5 However, ordering patterns to the chemical laboratory vary based on clinical practice factors,6 such as experience,7 time pressure,8 and uncertainty.9,10 In addition, geographical variation11 in ordering patterns is reported. Statistics indicate that our county in Sweden, Kronoberg, had the highest per capita utilization of chemical laboratory tests of the compared counties.12,13 Therefore, we hypothesized that tests might be ordered that are not clinically relevant according to standards, guidelines, and experience. These tests have limited or no benefit to the patient, and thus it would be possible to decrease the number of laboratory tests without affecting quality. In fact, excess tests are reported to represent up to 40% of test volume,14 and there is no evidence of a correlation between decreased number of selected tests and impaired clinical outcome.15-17 A variety of interventions have been tested in attempts to control escalating costs and excessive resource utilization,18 including peer management,15 data reports,19 education, audits, reviews,20 and multidimensional techniques.21 Most of these interventions are labor intensive; in a setting with limited resources to address these issues, an alternative approach was requested.
Physicians are poorly informed of laboratory costs22 and they have a tendency to underestimate them23; however, it has been reported that they consider availability of price lists to have an impact on cost generation.2,24 In the county of Kronoberg, cost availability had been requested by resident physicians, because price lists were not published and limited information on cost was available.
Charge display and price lists have successfully been introduced to decrease the number of ordered tests17,25,26; we therefore presented cost information on all available tests at our laboratories through the computerized test ordering system. By doing so, laboratory costs were presented to all staff in the county of Kronoberg with access to the computerized patient record, both in primary and secondary healthcare centers. The cost of each laboratory analysis (cost display) could thus be seen at the moment of test request as well as on the result report.
This is, to our knowledge, the first large-scale intervention on cost awareness using cost display at order entry and on the result report that included both inpatients and outpatients. We also describe the effect of introduction of cost charge (ie, all primary healthcare centers were obliged to pay full laboratory costs) on laboratory testing in different healthcare settings.
Study Design and Setting
Laboratory tests in Kronoberg are increasing and show a marked seasonal variation due to multiple factors, such as epidemiologic trends and holidays. This was a longitudinal study assessing the effect on clinical chemistry laboratory test volumes of introducing cost display for all primary and secondary healthcare centers. Cost was defined as the price for each laboratory test, including the costs for equipment, reagents, labor, service contract, and overhead. There are no volume discounts in Kronoberg County, nor are there any insurance company reimbursement policies in Sweden. Test ordering was solely based on the physician’s individual medical assessment for each patient, which can be supported by national or local medical guidelines to assist in decision making.
The cost display intervention started in September 2013 and was followed by introduction of full cost charge for primary healthcare centers, requiring them to pay full laboratory costs, in January 2014. The full cost charge intervention was not initiated by this study, but by a county policy decision. Kronoberg County had a population of 187,156 inhabitants as of December 31, 2013.13 They were served by 22 primary healthcare centers and 2 secondary healthcare centers operated by the County Council, as well as 11 privately operated primary healthcare centers. A primary healthcare center is an open healthcare unit exclusively for outpatients that serves as a first line of healthcare, primarily for medical conditions that are not defined as acute or in need of emergency care. The staff is predominantly nurses and general practitioners. The secondary healthcare centers are hospitals that provide care for patients primarily referred from the primary healthcare centers. The hospitals have emergency departments and provide specialist care within a wide range of medical specialties for both inpatients and outpatients.
There is no difference between privately and publicly operated primary healthcare centers except that the privately operated primary healthcare centers are allowed to make a profit, whereas the publicly operated centers are strictly nonprofit. All primary and secondary healthcare centers are publicly financed by taxes. Prior to the cost charge intervention, primary healthcare centers paid a fixed subscription fee, in addition to 30% of the cost of every test ordered. Secondary healthcare centers paid a fixed annual fee, regardless of the number of ordered laboratory tests.
The laboratory tests were performed by 2 central laboratories operated by the County Council, one at each secondary healthcare center. Point-of-care testing (POCT) was not included in our study. POCT was highly regulated by the central laboratories that restricted both the equipment and the analysis supply; thus, it constituted a very small proportion of the total laboratory tests in the county of Kronoberg. All publicly financed healthcare providers in Kronoberg County were obliged to send all of their test requests to the 2 central laboratories, regardless of private or public operation.
Description of Intervention
Information on laboratory test cost in Swedish currency (crowns) as an integer surrounded by brackets was appended to the laboratory test name in the clinical chemistry laboratory test definition file for the computerized patient record and test ordering system, Cambio Cosmic (Cambio Healthcare Systems Ab; Stockholm, Sweden). This resulted in cost display both at the moment of test ordering (eAppendix [available at ajmc.com]) and at presentation of the test result. The use of Cambio Cosmic was mandatory for all publicly financed healthcare providers in Kronoberg County.
Two weeks before the intervention began in September 2013, information on the introduction of cost display was presented online and an information newsletter was sent to healthcare employees subscribing to the Kronoberg County Council newsletter. The information was repeated the day after the introduction with an email to all physicians and department managers containing the same information as the newsletter. Before this intervention, no price list in the Kronoberg County Council had been published.
Due to a policy change in the county, all primary healthcare centers were obliged to pay full laboratory costs (ie, cost charge), starting January 2014.
The total intervention period for cost display was 13 months, from September 2013 through September 2014, during which 2,519,130 laboratory tests were analyzed. The total intervention period for cost charge was 9 months, from January 2014 through September 2014.
Data Source and Patient Selection
All requests were made in Cambio Cosmic and were exported to the laboratory information system (LIS), FlexLab/Kemi v.3.6.1 (Tieto Sweden Ab; Stockholm, Sweden). Results were registered and saved in the LIS database, then exported back to Cambio Cosmic, where the results were displayed. Information from the LIS database was extracted using ProClarity Analytics Platform 5 software (ProClarity Corp; Boise, Idaho). All test results, except results comprising calculations, ordered by the publicly financed healthcare centers and analyzed by the 2 central laboratories during the intervention period were included. The total blood count and the blood differential count were each counted as a single test. Information about the number of physician visits was retrieved from the Department of Analytical Support, Kronoberg County Council.
The main outcome was the number of tests ordered by the healthcare providers after the interventions.
Statistical analysis was performed using IBM SPSS Statistics for Windows, version 23.0 (IBM Corp; Armonk, New York).
To determine the mean difference in tests ordered before and after the interventions, analysis of the relative number of laboratory test requests per physician visit was performed. Analysis was performed separately for secondary healthcare centers, publicly operated primary healthcare centers, privately operated primary healthcare centers, and all of the healthcare centers combined. Due to a negative trend in the number of tests ordered from January 2010 to September 2011, we compared intervention data with data starting from September 2011 for both publicly operated and privately operated healthcare centers.
We compared the intervention outcomes using analysis of variance with calendar month as fixed factor to account for seasonal variation and time as covariate to assess the overall trend. We considered P values <.05 to be statistically significant.
The project was approved by the Research Ethics Advisory Committee, Kronoberg County Council.
The test volume for the total group (primary and secondary healthcare centers combined) showed a significant decrease of 0.35 tests per physician visit (P <.001) after the introduction of cost display. After cost charge introduction, no additional decrease in test volume was seen (Table).
Secondary Healthcare Centers
In the secondary healthcare center group, the test volume decreased by 0.34 tests per physician visit (P = .001) after the introduction of cost display (Table; Figure 1).
Following the introduction of cost display, the seasonal variation had the same pattern as before the intervention. In addition, the increase in number of tests over time was similar to that observed in the preintervention period. For the secondary healthcare centers, no cost charge was introduced and hence no analysis of change in test volumes was performed for this intervention.
Primary Healthcare Centers
In the publicly operated primary healthcare centers, neither the introduction of cost display (P = .13) nor that of full cost charge (P = .64) had any significant impact on the test volume (Table; Figure 2).
For the privately operated primary healthcare centers, the introduction of cost display did not change the test volume significantly. The introduction of cost charge significantly decreased the test volume by 0.48 tests per physician visit (P <.001). However, these data show high variability and hence lower reliability (Table; Figure 3).
Laboratory utilization is escalating, and laboratory resources are limited. Furthermore, there are increasing demands to lower costs while maintaining quality in patient care. Several studies on noncomputerized cost display strategies have reported various results.22,27,28 Thus, we wanted to investigate if a large-scale computerized intervention introducing cost display in a population of physicians with little previous knowledge of laboratory costs would decrease the number of laboratory tests ordered in the Kronoberg County Council. The setting included secondary healthcare centers and publicly and privately operated primary healthcare centers. The test requesters had 2 major interfaces with the laboratory: the computerized ordering sheet and the result display. We introduced cost display on both, making information available even if a paper ordering sheet was used.
We found an overall significant decrease in the number of tests ordered. However, subgroup analysis revealed differences among the 3 clinical settings, and the only significant effect was seen in secondary healthcare centers.
Further studies are needed to explain the discrepancy between the secondary and primary healthcare centers; therefore, one can only speculate that the differing patient populations and types of diagnoses have an impact on test ordering. Also, the proportion of substitute physicians is higher in the primary healthcare centers than in the secondary healthcare centers, indicating a higher turnover of staff, which has been associated with higher test utilization.29
Our results partly differ from the findings of an earlier American study by Bates et al using computerized cost display for outpatients,30 where the numbers of tests were nonsignificantly reduced. The intervention period in that study was similar to ours, but our study group was larger and the postintervention period was longer. We also included both inpatients and outpatients. However, in the study by Bates et al, cost display was randomized by patient. Therefore, the effect of cost display for the randomized patients might have introduced a spillover effect, influencing physician ordering behavior for nonrandomized patients. Tierney et al showed that the decrease in test ordering was not sustainable over time, which is in agreement with our data.17 The intervention generated a downshift in test ordering, but the increase over time, before and after the intervention, is similar (Figure 1).
A complementary intervention was implemented by management in which 100% cost charge was introduced for all primary healthcare centers 3 months after the introduction of cost display. Because of this, we were able to study the effect of real economic impact on the number of test requests. This ruled out the possibility of studying the longitudinal effects of cost display in the primary healthcare settings. However, it opened the possibility of future assessment of cost charge impact. Cost charge promoted a significant decrease in the number of tests ordered, but only in the privately operated primary healthcare centers, which might possibly be due to a more pronounced financial stake in this setting. However, cost display alone did not have any significant impact. In the publicly operated primary healthcare centers, neither cost display nor cost charge had any effect on ordering behavior.
A number of attempts to explain variability in test ordering have been made.6,29,34,35 The age2,35 and the experience of the physician have been shown to influence test ordering behavior,36 as has geographical location.34 Other plausible causes are demography, number of substitute physicians hired through staffing agencies, and physicians per capita.11
In addition to cost display, a multitude of strategies have been described to reduce laboratory tests or to increase appropriateness of the test requests.37 Feedback strategies38 and unbundling of test panels have been used to decrease test volume.33,39 However, bundling of tests has also been described to decrease the number of tests.37 Automated algorithms using reflex orders have been suggested to decrease tests without compromising medical information.40
No single intervention seems to have resulted in a sustained decrease in laboratory test use. A long-term decrease is favored by combining interventions, and repeating them over time, in a supportive environment.41 It has been reported that key components include committed senior staff, long-term strategies, and providing diverse approaches for different groups of physicians.42
Laboratory costs have decreased over the decades. An American report from the 1970s found that 25% of hospital charges were due to laboratory costs.27 For Europe, laboratory expenditures were described to be 0.8% of total healthcare expenditures in 2014 by the European Diagnostic Manufacturers Association43; however, that figure does not include laboratory labor costs. In Kronoberg, the total cost of the tests analyzed at the Department of Clinical Chemistry and Transfusion Medicine was 1.5% of the total healthcare expenditures of the Kronoberg County Council in 2014. It could be argued that decreasing laboratory tests would have a limited direct economic impact on the healthcare economy. However, appropriate use of laboratory services will not only contribute to containment of costs but also improve medical care.44,45 The main benefits would probably be indirect. For example, laboratory test result reference values are generally 95% CIs, such that every 20th sample will be outside the interval, possibly creating the need for further laboratory testing and investigations, resulting in increasing expenses. Reduced laboratory tests might also increase patient safety, as information overload may obscure information crucial to the physician.31,46
Although this study had several strengths, it also had limitations. The study did not address the appropriateness of the laboratory test requests. Inappropriate testing may include overuse, underuse, and misutilization of laboratory tests.31 It has been reported that, for a selection of tests, the overuse rate was 26% to 98%, depending on setting.31 It has also been suggested that a high number of tests could be justified if it saves inpatient time and costs.32 In contrast, it has been shown that the number of laboratory tests can be selectively reduced without compromising the outcome of medical care.31,33 Another limitation of the study was that the results were not compared with a county that did not implement cost display. The counties in Sweden are politically managed and each county has its own elected political board. This gives each county unique conditions, including the financial setting, that make comparisons difficult. An additional limitation is that data on physician characteristics were not available, except for the proportion of substitute physicians in publicly operated primary healthcare centers, nor was information on individual physician-level test ordering patterns available. Although no major organizational changes or other factors that might have changed ordering behavior were known, the occurrence of such confounders cannot completely be excluded.
Depending on the setting, cost display and cost charge could be used to reduce the number of laboratory tests ordered. The sustainability of decreased test volumes, the underlying mechanisms, and the impact on health outcomes need to be investigated further.
The expert technical assistance of Kent Danielsson and Kristina Rönn at the Department of Clinical Chemistry and Transfusion Medicine, Region Kronoberg, is greatly appreciated. The authors thank Anna Lindgren at the Center for Mathematical Sciences, Lund University, for excellent statistical advice.Author Affiliations: Department of Clinical Chemistry and Transfusion Medicine, Växjö Central Hospital (KE, AP), Växjö, Sweden; Department of Medical Biosciences, Clinical Chemistry, Umeå University (KE), Umeå, Sweden.
Source of Funding: The study was funded by grants from the Department of Research and Development, Region Kronoberg, grant number 3590.
Author Disclosures: Dr Ekblom and Dr Petersson are employees at the study setting and have received grants from the employer.
Authorship Information: Concept and design (KE, AP); acquisition of data (KE, AP); analysis and interpretation of data (KE, AP); drafting of the manuscript (KE, AP); critical revision of the manuscript for important intellectual content (KE, AP); statistical analysis (KE, AP); provision of patients or study materials (KE, AP); and obtaining funding (KE, AP).
Address Correspondence to: Kim Ekblom, MD, PhD, Department of Clinical Chemistry and Transfusion Medicine, Växjö Central Hospital, SE-35185 Växjö, Sweden. Email: firstname.lastname@example.org.REFERENCES
1. Mindemark M, Larsson A. Longitudinal trends in laboratory test utilization at a large tertiary care university hospital in Sweden. Ups J Med Sci. 2011;116(1):34-38. doi: 10.3109/03009734.2010.528071.
2. Salloum S, Franssen E. Laboratory investigations in general practice. Can Fam Physician. 1993;39:1055-1061.
3. Winkens R, Dinant GJ. Evidence base of clinical diagnosis: Rational, cost effective use of investigations in clinical practice. BMJ. 2002;324(7340):783-785. doi: 10.1136/bmj.324.7340.783.
4. Forsman RW. Why is the laboratory an afterthought for managed care organizations? Clin Chem. 1996;42(5):813-816.
5. Larsson A, Palmer M, Hultén G, Tryding N. Large differences in laboratory utilisation between hospitals in Sweden. Clin Chem Lab Med. 2000;38(5):383-389. doi: 10.1515/CCLM.2000.056.
6. Grytten J, Sørensen R. Practice variation and physician-specific effects. J Health Econ. 2003;22(3):403-418. doi: 10.1016/S0167-6296(02)00105-4.
7. Bugter-Maessen AM, Winkens RA, Grol RP, et al. Factors predicting differences among general practitioners in test ordering behaviour and in the response to feedback on test requests. Fam Pract. 1996;13(3):254-258.
8. van der Weijden T, van Bokhoven MA, Dinant GJ, van Hasselt CM, Grol RP. Understanding laboratory testing in diagnostic uncertainty: a qualitative study in general practice. Br J Gen Pract. 2002;52(485):974-980.
9. Sonnenberg A. A medical uncertainty principle. Am J Gastroenterol. 2001;96(12):3247-3250. doi: 10.1111/j.1572-0241.2001.05321.x.
10. West AF, West RR. Clinical decision-making: coping with uncertainty. Postgrad Med J. 2002;78(920):319-321. doi: 10.1136/pmj.78.920.319.
11. Mindemark M, Wernroth L, Larsson A. Costly regional variations in primary health care test utilization in Sweden. Scand J Clin Lab Invest. 2010;70(3):164-170. doi: 10.3109/00365511003642519.
12. Helseplan Nysam AB. Nyckeltal 2012 Rapport - Klinisk kemi. Stockholm, Sweden: Helseplan Nysam AB; 2013.
13. Population by region, marital status, age and sex. Year 1968 - 2017. Statistiska centralbyrån website. http://www.statistikdatabasen.scb.se/pxweb/en/ssd/START__BE__BE0101__BE0101A/BefolkningNy/table/tableViewLayout1/?rxid=51e58cdf-d493-4a1f-9696-90c98ff6e77c. Accessed April 6, 2018.
14. Rao GG, Crook M, Tillyer ML. Pathology tests: is the time for demand management ripe at last? J Clin Pathol. 2003;56(4):243-248. doi: 10.1136/jcp.56.4.243.
15. Neilson EG, Johnson KB, Rosenbloom ST, et al; Resource Utilization Committee. The impact of peer management on test-ordering behavior. Ann Intern Med. 2004;141(3):196-204. doi: 10.7326/0003-4819-141-3-200408030-00008.
16. Kroenke K, Hanley JF, Copley JB, et al. Improving house staff ordering of three common laboratory tests. Reductions in test ordering need not result in underutilization. Med Care. 1987;25(10):928-935.
17. Tierney WM, Miller ME, McDonald CJ. The effect on test ordering of informing physicians of the charges for outpatient diagnostic tests. N Engl J Med. 1990;322(21):1499-1504. doi: 10.1056/NEJM199005243222105.
18. Wu AH. Improving the utilization of clinical laboratory tests. J Eval Clin Pract. 1998;4(3):171-181. doi: 10.1046/j.1365-2753.1998.00001.x.
19. Grivell AR, Forgie HJ, Fraser CG, Berry MN. League tables of biochemical laboratory costs. An attempt to modify requesting patterns. Med J Aust. 1982;2(7):326-328.
20. Everett GD, deBlois CS, Chang PF, Holets T. Effect of cost education, cost audits, and faculty chart review on the use of laboratory services. Arch Intern Med. 1983;143(5):942-944. doi: 10.1001/archinte.1983.00350050100019.
21. Solomon DH, Hashimoto H, Daltroy L, Liang MH. Techniques to improve physicians’ use of diagnostic tests: a new conceptual framework. JAMA. 1998;280(23):2020-2027. doi: 10.1001/jama.280.23.2020.
22. Schilling UM. Cost awareness among Swedish physicians working at the emergency department. Eur J Emerg Med. 2009;16(3):131-134. doi: 10.1097/MEJ.0b013e32831cf605.
23. Long MJ, Cummings KM, Frisof KB. The role of perceived price in physicians’ demand for diagnostic tests. Med Care. 1983;21(2):243-250.
24. Schilling UM. The acceptance of price lists at the emergency department: how do doctors think about it? Scand J Trauma Resusc Emerg Med. 2010;18(suppl 1):P32. doi: 10.1186/1757-7241-18-S1-P32.
25. Schilling UM. Cutting costs: the impact of price lists on the cost development at the emergency department. Eur J Emerg Med. 2010;17(6):337-339. doi: 10.1097/MEJ.0b013e32833651f0.
26. Seguin P, Bleichner JP, Grolier J, Guillou YM, Mallédant Y. Effects of price information on test ordering in an intensive care unit. Intensive Care Med. 2002;28(3):332-335. doi: 10.1007/s00134-002-1213-x.
27. Cummings KM, Frisof KB, Long MJ, Hrynkiewich G. The effects of price information on physicians’ test-ordering behavior. Ordering of diagnostic tests. Med Care. 1982;20(3):293-301.
28. Hampers LC, Cha S, Gutglass DJ, Krug SE, Binns HJ. The effect of price information on test-ordering behavior and patient outcomes in a pediatric emergency department. Pediatrics. 1999;103(4 pt 2):877-882.
29. Kristiansen IS, Hjortdahl P. The general practitioner and laboratory utilization: why does it vary? Fam Pract. 1992;9(1):22-27.
30. Bates DW, Kuperman GJ, Jha A, et al. Does the computerized display of charges affect inpatient ancillary test utilization? Arch Intern Med. 1997;157(21):2501-2508. doi: 10.1001/archinte.1997.00440420135015.
31. Peterson SE, Rodin AE. Prudent laboratory usage, cost containment, and high quality medical care: are they compatible? Hum Pathol. 1987;18(2):105-108. doi: 10.1016/S0046-8177(87)80328-3.
32. Spencely M, Miller DL. Inflation control in the laboratory: does demand equal need? J Infect. 1980;2(1):3-12. doi: 10.1016/S0163-4453(80)91681-3.
33. Attali M, Barel Y, Somin M, et al. A cost-effective method for reducing the volume of laboratory tests in a university-associated teaching hospital. Mt Sinai J Med. 2006;73(5):787-794.
34. Salinas M, López-Garrigós M, Flores E, Uris J, Leiva-Salinas C. Request of acute phase markers in primary care in Spain. Am J Manag Care. 2015;21(10):e591-e596.
35. Peterson S, Eriksson M, Tibblin G. Practice variation in Swedish primary care. Scand J Prim Health Care. 1997;15(2):68-75.
36. Morgan S, Henderson KM, Tapley A, et al. Pathology test-ordering behaviour of Australian general practice trainees: a cross-sectional analysis. Int J Qual Health Care. 2015;27(6):528-535. doi: 10.1093/intqhc/mzv086.
37. Wong ET. Cost-effective use of laboratory tests: a joint responsibility of clinicians and laboratorians. Clin Lab Med. 1985;5(4):665-672.
38. Studnicki J, Bradham DD, Marshburn J, Foulis PR, Straumfjord JV. A feedback system for reducing excessive laboratory tests. Arch Pathol Lab Med. 1993;117(1):35-39.
39. Golden WE, Pappas AA, Lavender RC. Financial unbundling reduces outpatient laboratory use. Arch Intern Med. 1987;147(6):1045-1048. doi: 10.1001/archinte.1987.00370060041008.
40. Schedvin G, Jones I, Hultdin J, Nilsson TK. A laboratory algorithm with homocysteine as the primary parameter reduces the cost of investigation of folate and cobalamin deficiency. Clin Chem Lab Med. 2005;43(10):1065-1068. doi: 10.1515/CCLM.2005.186.
41. Martin AR, Wolf MA, Thibodeau LA, Dzau V, Braunwald E. A trial of two strategies to modify the test-ordering behavior of medical residents. N Engl J Med. 1980;303(23):1330-1336.
42. Young DW. Improving laboratory usage: a review. Postgrad Med J. 1988;64(750):283-289. doi: 10.1136/pgmj.64.750.283.
43. European IVD Market Statistics Report 2014. MedTech Europe website. www.medtecheurope.org/sites/default/files/resource_items/files/European%20IVD%20Market%20Report%202014_1.pdf. Accessed March 22, 2016.
44. Benson ES. The responsible use of the clinical laboratory. Clin Biochem. 1986;19(5):262-270. doi: 10.1016/S0009-9120(86)80038-8.
45. Naugler C, Ma I. More than half of abnormal results from laboratory tests ordered by family physicians could be false-positive. Can Fam Physician. 2018;64(3):202-203.
46. Beasley JW, Wetterneck TB, Temte J, et al. Information chaos in primary care: implications for physician performance and patient safety. J Am Board Fam Med. 2011;24(6):745-751. doi: 10.3122/jabfm.2011.06.100255.