Skip to main content

Table 1 Characteristics of the included studies

From: Laboratory test ordering in inpatient hospitals: a systematic review on the effects and features of clinical decision support systems

References Study design Study duration Setting Population Sample size Intervention description Main finding regarding the proportion of laboratory tests Conclusion
Bates et al. [28] RCT 4 months A tertiary care hospital Inpatients at the hospital CG: 5886 patients IG:5700 patients CPOE reminder: In the intervention group, if a test had previously been ordered within its test-specific interval, the physician received a reminder that the test had been performed recently or was pending; the result was showed if available. For the control group, duplication was determined in exactly the same way, but there was no reminder (1) In the IG 69% cancelled the order after the reminder
(2) In the CG, 51% of ordered redundant tests were performed, whereas in the IG only 27% of ordered redundant tests were performed (P < 0.001)
(3) During the preceding period, 20.5% of target tests were performed earlier than specific intervals, whereas during the study period, this rate was significantly lower in the IG (18.5%, P = 0.004) but not in the control group (19.6%, P = 0.19)
(4) In the 4-month period preceding the intervention, there were 4.84 target tests per admission, compared with 4.24 during the study period in the intervention group and 4.28 in the control group (both P < 0.0001)
Delivering reminders about orders for apparently redundant laboratory tests were effective. However, since many tests were conducted without corresponding computer orders and many orders were not screened for duplication overall effect was limited
Boon-Falleur et al. [31] Before-after 6 months A pediatric liver disease unit Patients with liver transplant Before: 42 patients After: 175 patients A rule-based expert system allows static and dynamic requesting rules to be defined for different clinical classifications of patients. The static rules allow the definition of "baseline" proposals within a precise time schedule. Dynamic rules allow the system to react to results of previously ordered tests. The attending physician may accept or amend the system's proposals by adding or removing requests to the proposed schedule (1) An increase of the total number of tests requested per patient was observed
(2) An overall reduction in laboratory resources consumption for transplanted patients (27%)
(3) A decrease in the percentage of "STAT" requested tests (− 44%)
(4) The percentage of tests ordered in agreement with the protocols for those patients increased from 33% before the introduction of the expert system to 45% when the system was used
The clinicians’ perspective was that the system would increase the total benefits in clinical resources use, improve the management of laboratory data, and save time for doing laboratory ancillary tasks
Bridges et al. [34] Before-after 6 months A tertiary care hospital Patient admitted to the department of medicine Before: 674 patients After: 692 patients The intervention consisted of displaying a computerized alert informing that the clinician is ordering a recently ordered test (1) In the pre-intervention period, 53 (7.9%) were duplicated and post-intervention 18 (2.6%) were duplicated (P < 0.001)
(2) The alert significantly reduce associated costs of duplicated acute hepatitis profile tests (P ≤ 0.001)
Computerized alerts may be effective in reducing redundant laboratory tests and enhancing efficiency of healthcare system
Dalal et al. [35] Before-after 6 months A teaching hospital All TSH, T3, and T4 ordered in Department of Medicine Before: 2611 tests After: 2454 tests A clinical algorithm for CDS and Hard Stops were incorporated into the EMR to decline ordering freeT3 or freeT4 without an abnormal TSH, also certain exceptions were predefined. In addition, if the TSH was abnormal a reflex rule was triggered and could automatically order freeT3 and freeT4 (1) The fT3 to TSH ordering ratio similarly decreased by 55.2%, from 6.2 to 2.9% (P < 0.0001)
(2) Post-intervention there was a decrease in the ratio of fT4 to TSH orders (fT4/TSH) of 35.2%, from 44.6% to 28.9% (P < 0.0001)
(3) TFT/TSH pre-intervention ratio was 52.2%, which decreased by 39.1%, to 31.8% post-intervention (P < 0.0001)
By a clinical decision support about when to order TFTs, they observed a decrease in the number of unnecessary tests ordered
Eaton et al. [36] Time-series 30 months Hospital Inpatient population admitted to general medicine service Before: 14,193 patients After: 13,751 patients Educational guide, nonintrusive ordering message, and noon conference. Appropriate indications for selected tests were incorporated into text accompanying the laboratory orders in hospital’s HER. Physicians could ignore the text and proceed with the order (1) The rate of folate tests ordered per monthly admissions showed no significant level change at the time of the intervention with only a slight decrease in rate of 0.0109 (P = 0.07)
(2) There was a 43% decrease in the rate of hepatitis C virus tests per monthly admissions immediately AI with a decrease of 0.0135 tests per monthly admissions (P = 0.02)
Nonintrusive CDSS do not have significant effect on utilization of laboratory test
Gottheil et al. [30] Time series 12 months A tertiary care hospital Erythrocyte sedimentation rate orders Not mentioned Educational content and CDSS: a series of appropriateness criteria for Erythrocyte Sedimentation Rate was incorporated into CDSS After CDS, ESR orders per week decreased from 386 to 151. When unlimited access was provided to select subspecialties, there was an increase in ESR orders per week to 241. This represents a decrease of almost 40% from baseline Their quality improvement initiative could reduce inappropriate Erythrocyte Sedimentation Rate testing by computerized CDS
Klatte et al. [37] Time series 12 months A tertiary hospital, a 53-bed satellite facility Specimens from children ≤ 12 months 485 specimens Educational intervention, an evidence-based algorithm for appropriate clostridium difficile ordering, and CPOE requiring clinicians to mandatory complete 2 extra fields. Non diarrheal stool were automatically declined by laboratory, unless in cases with severe ileus or toxic megacolon After the intervention, the average percentage of specimens tested dropped to 53.8% Their CDSS intervention resulted in a sustained drop in the number of specimens tested, which saved laboratory and patient cost significantly. They observed no sustained change in clinicians’ ordering practices in spite of multiple educational efforts
Levick et al. [38] Time series 6 months Three not-for-profit hospitals Patients with B-type natriuric peptide test 41,306 patients CPOE with embedded CDS: The CDS intervention is an expert rule that searches the system for a B-Type natriuric peptide lab value for the patient. An advisory alert was indicated to the ordering clinician if there was a value for the test and it was within the current hospital stay (1) The CDS intervention reduced B-Type natriuric peptide orders by 21% relative to the mean Using CDSS alerts has the potential for improving care, but should be used judiciously and in the appropriate environment
Lippi et al. [32] Before-after 6 months A teaching hospital A variety of tests requests including C reactive protein, TSH, ferritin, brain natriuretic peptide, etc 3539 test requests CDSS: an electronic alert is automatically triggered by a potentially inappropriate test request. The alert contains a detailed explanation of the specific rule for appropriateness of the test The total number of test requests violating the preset criteria of inappropriateness constantly decreased over time (26% in the first three months of implementation versus 17% in the following period; P < 0.001) A CDSS alert may be effective to decrease the inappropriateness of laboratory test orders, generate significant cost saving and educate physicians to use laboratory resources more efficiently
Nicholson et al. [39] Before-after non-equivalent control group 26 months A tertiary-care pediatric hospital Children < 36 months of age Before: 141 patients After: 55 patients An alert advising against ordering C. difficile tests in infants and young children based on the American Academy of Pediatrics recommendations. Physicians could override it optionally (1) The average monthly testing rate significantly decreased for children 0–11 months old ( P < 0.001) and 12–35 months old (P < 0.001), but not for those children ≥ 36 months old (P = 0.3) The average monthly testing rate for C. difficile for children < 35 months old decreased without complication after the use of a CPOE alert in those who tested positive for C. difficile
Niès et al. [33] Time series 36 months A university teaching hospital Patients with hepatitis B antigen test Before: 2888 patients After: 1572 patients CDSS: The alert is triggered when one of the targeted serological tests for hepatitis B virus is selected to be ordered. The Serology-CDSS stores a record of its execution each time a physician selects a viral serology test order. An alert is displayed if the most recent result of the targeted laboratory test for the patient is less than 90 days old (1) In pre-intervention period 15.5% of viral serology tests were unnecessarily repeated. During the intervention period, 15.8% were repeated. Before the intervention, the mean proportion of unnecessarily repeated HBs antigen tests increased by 0.4% per month (P < 0.001). After the intervention, a significant trend change occurred, with a monthly difference estimated at -0.4% (P = 0.02) resulting in a stable proportion of unnecessarily repeated HBs antigen tests After CDSS implementation an immediately decrease was observed in the proportion of unnecessarily duplicate tests. CDSS alerts could also improve compliance rate
Quan et al. [40] Before-after 24 months An academic hospital Patients with C. difficile infection test Before: 284 tests After: 268 tests Clinicians were required to verify the determined criteria for appropriate ordering of C. difficile infection test. A warning email was sent to the physicians ordering the test without appropriate approval Baseline CDI testing rate declined from 284/10,000 to 268/10,000 patient-days post-intervention (P = 0.02). The intervention decreased inappropriate testing by 64% The protocol increased appropriate testing as well as decreasing hospital-onset standardized infection ratio of C. difficile infection
Procop et al. [41] Time series 24 months The Cleveland Clinic More than 1000 tests of all patients Not mentioned CDSS: This tool informs the provider that the test being ordered is a duplicate. It also block unnecessary duplicate test orders during the computerized physician order entry (1) The proportions of reductions in the number of stool ova/parasite examinations was 54.1% (P < 0.0001)
(2) The proportions of reductions in the number of Giardia/Cryptosporidium enzyme immunoassay tests was 22.58% (P = 0.2807)
(3) The proportions of reductions in the number of stool culture tests was 49.1% (P < 0.0001)
Real-time interaction between the laboratory and the physician through CDS tools could decrease duplicate orders. It saves healthcare costs and should also increase patient satisfaction and well-being
Rosenbloom et al. [42] Time series 5 years An academic inpatient tertiary care facility Clinicians at a university hospital 194,192 patients The CDSS exhorted users to discontinue unnecessary tests recurring more than 72 h into the future
(2) Education regarding appropriate indications for testing. (3) CDS and CPOE systems targeted only magnesium ordering, displayed recent results, limited testing to one instance per order, summarized indications for testing, and required users to select an indication
At baseline, there were 539 magnesium tests ordered per week. This decreased to 380 (P = 0.001) per week after the first intervention, increased to 491 per week (P, 0.001) after the second, and decreased to 276 per week (P, 0.001) after the third A clinical decision support intervention intended to regulate testing increased test order rates as an unintended result of decision support
Rudolf et al. [43] Time series 36 months A tertiary care teaching hospital Laboratory tests 61,644 laboratory test ordes Alert in the CPOE system: the alert appeared in the CPOE each time an order with frequency greater than one occurrence was selected. The justification for the order was also captured by the CPOE, as providers were required to select one of three approved indications for the daily laboratory test or manually enter another indication (1) 6,463 orders for recurrent daily laboratory tests were placed for a mean daily rate of 71.8 orders per day
(2) AI 44,900 orders for recurrent daily laboratory tests were placed for a mean daily rate of 44.8 orders per day, representing a highly significant decrease in daily laboratory test ordering
(3) Total inpatient test volumes were not affected
Our experience suggests auditing and continued feedback are additional crucial components to changing ordering behavior. Curtailing daily orders alone may not be a sufficient strategy to reduce in-laboratory costs
Samuelson et al. [44] Before-after 16 months Two academic medical hospitals Patients evaluated for heparin-induced thrombocytopenia Before: 265 patients After: 146 patients CDSS: a decision-support tool required providers to calculate the 4Ts (heparin-induced thrombocytopenia risk) score prior to ordering laboratory-based tests for anti-PF4/heparin antibody enzyme-linked immunosorbent assay testing (1) We observed a significant decrease from 43 tests/month before to 22 tests/month ( P  < 0.001) after the intervention
(2) We observed a trend toward decrease in the proportion of tested patients with low 4Ts scores (66% vs 56%, P = 0.069),
Our study demonstrates that a clinical decision support tool embedded within the electronic ordering process can decrease unnecessary testing for heparin-induced thrombocytopenia
  1. CDSS clinical decision support system, CG control group, CPOE computerized physician order entry, ED emergency department, IG intervention group, RCT randomized control trial, TFT thyroid function test, TSH thyroid stimulation hormone
\