What Proportion of Common Diagnostic Tests Appear Redundant?

doi:10.1016/S0002-9343(98)00063-1

The American Journal of Medicine

Volume 104, Issue 4, April 1998, Pages 361-368

https://doi.org/10.1016/S0002-9343(98)00063-1 Get rights and content

Abstract

Purpose: To identify ancillary tests for which there are criteria defining the earliest interval at which a repeat test might be indicated, to determine how often each test is repeated earlier than these intervals and, if repeated, provides useful information.

Subjects and Methods: We performed a retrospective cohort study of 6,007 adults discharged from a large teaching hospital during a 3-month period in 1991. We measured the proportion of commonly performed diagnostic tests that were redundant, and their associated charges.

Results: Of the 6,007 patients discharged, 5,289 (88%) had at least one of 12 target tests performed. Overall, 78,798 of the target tests were performed during the study period, of which 22,237 (28%) were repeated earlier than test-specific predefined intervals. This percentage varied substantially by test (range, 2% to 62%). To assess how many early repeats were justified, we performed chart reviews in a random sample stratified by test. For two tests, nearly all the initial results in the sample were abnormal, and all repeats were considered justified. Of early repeats following a normal initial result for the remaining 10 tests, chart review found no clinical indication for 92%, and a weighted mean of 40% appeared redundant. Overall, 8.6% of these 10 tests appeared redundant; if these were not performed, the annual charge reductions would be $930,000 at our hospital, although the impact on costs would be much smaller.

Conclusions: For some tests, an important proportion are repeated too early to provide useful clinical information. Most such tests might be eliminated using computerized reminder systems.

Section snippets

Patient Population

The study sample included all adults discharged from Brigham and Women’s Hospital, a 726-bed tertiary care hospital, during a 3-month period in 1991.

Test Selection and Definitions

We identified tests that met at least one of two criteria: they were high volume based on data from our hospital’s financial system, or the labor cost or reagent cost of performing the test was high. We then reviewed the literature to identify published guidelines for performing each test. Tests that we initially included because they were either

Results

During a 3-month period in 1991, 6,007 adults were discharged from Brigham and Women’s Hospital (Table 2). Of these patients, 5,289 (88%) had at least one of the 12 target tests performed during their hospitalization. During the study period, 78,798 of the target laboratory tests were performed, of which 22,237 (28%) appeared to be early repeats based on the financial database (Table 3). This proportion was highest for digoxin levels (62%) and chest radiography (58%), and lower for some drug

Discussion

Tests that are repeated too early to provide useful information represent only a small proportion of those that are unnecessary or of marginal yield. However, they represent a group that is easy to target. In this study, we found that 8.6% of a defined group of commonly performed ancillary tests appeared redundant.

There are many reasons that redundant tests are performed. Sometimes nursing or other hospital personnel send more than one sample for a given order (eg, for a urine or sputum

Acknowledgements

We would like to thank Anthony Komaroff, MD, for comments on an earlier version of the manuscript.

References (37)

WH Theodore
Rational use of antiepileptic drug levels
Ther Drug Monit
(1992)
JW Winkelman
Less utilization of the clinical laboratory produces disproportionately small true cost reductions
Hum Pathol
(1984)
JS Spiegel et al.
Changing physician test ordering in a university hospital. An intervention of physician participation, explicit criteria, and feedback
Arch Intern Med
(1989)
DM Berwick et al.
Feedback reduces test use in a health maintenance organization
JAMA
(1986)
AM Epstein et al.
Physician characteristics and organizational factors influencing use of ambulatory tests
Med Decis Making
(1985)
SA Schroeder et al.
Variation among physicians in use of laboratory tests. Relation to quality of care
Med Care
(1974)
M Daniels et al.
Variation among physicians in use of laboratory tests. II. Relation to clinical productivity and outcomes of care
Med Care
(1977)
PF Griner et al.
Misuse of laboratory tests and diagnostic procedures
NEJM
(1982)
SA Schroeder et al.
Use of laboratory tests and pharmaceuticals. Variation among physicians and effect of cost audit on subsequent use
JAMA
(1973)
JM Eisenberg et al.
Computer-based audit to detect and correct overutilization of laboratory tests
Med Care
(1977)

SV Williams et al.

A controlled trial to decrease the unnecessary use of diagnostic tests

J Gen Intern Med

(1986)

RH Dixon et al.

Ultilization of clinical chemistry services by medical house staff. An analysis

Arch Intern Med

(1974)

P Axt-Adam et al.

Influencing behavior of physicians ordering laboratory testsa literature study

Med Care

(1993)

K Kroenke et al.

Improving house staff ordering of three common laboratory tests. Reductions in test ordering need not result in underutilization

Med Care

(1987)

DL Siegel et al.

Inappropriate testing for diarrheal diseases in the hospital

JAMA

(1990)

K Kroenke et al.

The admission urinalysisimpact on patient care

J Gen Intern Med

(1986)

VA Lawrence et al.

The unproven utility of preoperative urinalysis. Clinical use

Arch Intern Med

(1988)

HI Bussey et al.

A prospective evaluation of therapeutic drug monitoring

Ther Drug Monit

(1983)

Cited by (116)

Evaluation of clinical decision support
2023, Clinical Decision Support and beyond: Progress and Opportunities in Knowledge-Enhanced Health and Healthcare
Clinical decision support (CDS) systems interpret patient data from the electronic health record (EHR) against a central repository of information to provide patient-specific guidance to clinicians through electronic tools such as alerts and reminders. CDS is not limited to tools that make an explicit suggestion to the user (such as alerts and reminder); it also includes tools like condition-specific order sets, focused patient data summaries, note templates, and contextually relevant references or guide-lines (e.g., through info buttons). Since their inception, CDS interventions have demonstrated improved compliance with clinical guidelines among providers and have sometimes been associated with a reduction in adverse event rates. However, though there have been several success stories with clinical decision support, there have also been probably just as many, if not more, failures. There is still a great deal of opportunity for improvement in the way most of the clinical decision support in place today is being implemented.
The Frequency, Reasoning, and Impact of Repeated Radiographs at the Initial Orthopedic Arthroplasty Visit
2021, Journal of Arthroplasty
Patients presenting to an orthopedic clinic with joint pain often seek prior care and imaging before consultation. It is unknown how often orthopedic surgeons must repeat imaging and whether repeat imaging has an impact on diagnosis or management. The purpose of this study was to determine the frequency, reason, and impact of repeating radiographs in outpatient orthopedic clinics.
Patients ≥18 years of age presenting with hip and/or knee pain were prospectively enrolled at five arthroplasty clinics from January 2019 until June 2020. Before the initial visit, surveys were distributed to patients regarding the reason for their visit, prior care, and prior diagnostic imaging. At the conclusion of the visit, surgeons reported if repeat radiographs were obtained, and if so, surgeons documented the views ordered, the reasoning for new films, and if diagnosis or management changed as a result. Patients were grouped based on repeat imaging status, and of those with repeat imaging, subgroup analysis compared patients based on if management changed.
Of 292 patients, 256 (88%) had radiographs before their office visit, and 167 (65%) obtained repeat radiographs. Radiographs were most commonly repeated if they were inaccessible (47%), followed by if they were non–weight-bearing (40%). Repeated radiographs changed the diagnosis in 40% of patients and changed management in 22% of patients.
Most patients underwent repeat radiography at their orthopedic visit. The primary reasons were owing to accessibility or the patient being non–weight-bearing. Repeat radiographs changed management in almost one-quarter of patients.
Utility of postoperative hemoglobin testing following total shoulder arthroplasty
2021, JSES International
Identifying areas of excess cost for shoulder arthroplasty patients can play a role in effective health care spending. The purpose of this study was to assess the utility of postoperative complete blood count (CBC) testing after total shoulder arthroplasty (TSA) and identify which patients benefit from routine CBC testing.
We performed a retrospective review of a cohort of patients who underwent primary TSA from January 2018 through January 2019. All patients in this cohort received tranexamic acid. Patient demographic characteristics and patient-specific risk factors such as American Society of Anesthesiologists score, Elixhauser index, body mass index, smoking status, and coagulopathy history were obtained. Perioperative values including length of surgery, preoperative and postoperative hemoglobin (Hgb) levels, and need for transfusion were also obtained.
This study included 387 TSA patients in the final analysis. Comparison between the cohort requiring transfusion and the cohort undergoing no intervention revealed no statistically significant differences in age, sex, body mass index, American Society of Anesthesiologists score, or Elixhauser index. The group receiving transfusions was found to have significantly lower levels of preoperative Hgb (11.3 g/dL) and postoperative Hgb (8.1 g/dL) (P < .0001). Additionally, the percentages of patients with abnormal preoperative Hgb levels (<12 g/dL) (72.3%) and postoperative day 1 Hgb levels < 9 g/dL (81.8%) were significantly higher in the group receiving transfusions (P < .0001). A multivariate regression model identified an abnormal preoperative Hgb level (<12 g/dL), with an odds ratio of 3.8 (95% confidence interval, 1.5-6.2; P < .001), and postoperative day 1 Hgb level < 9 g/dL, with an odds ratio of 3.3 (95% confidence interval, 0.4-6.1; P < .03), as significant predictors of the risk of transfusion with a sensitivity of 64% and specificity of 96.2% with an area under the curve of 0.87.
Routine CBC testing may not be necessary for patients who receive tranexamic acid and have preoperative Hgb levels > 12 mg/dL and first postoperative Hgb levels > 9 mg/dL. This translates to potential health care cost savings and improves current evidence-based perioperative management in shoulder arthroplasty.
Inappropriate repeat testing of complete blood count (CBC) and electrolyte panels in inpatients from Alberta, Canada
2020, Clinical Biochemistry
Citation Excerpt :
Between 2004 and 2009, Calgary Laboratory Services (CLS) reported that there was a 36% increase in chemistry test volumes in the Calgary region (unpublished data). Specifically related to inpatient repeated tests, this is a problem that is faced by many jurisdictions and involves a number of laboratory tests that are conventionally ordered as pre- and post-operative evaluations, or simply as standing orders for the length of hospital stay [16–22]. The potential harmful impact of rapid repeat testing on the prognosis of the patient is a major dilemma.
The avoidance of repeat chemistry testing such as Complete Blood Count (CBC) and Electrolyte Panel (EP) on clinically stable patients was identified as important utilization goals by Choosing Wisely Canada. The purpose of this study was to assess the volume of overutilization of CBC and EP in an inpatient setting in Alberta, Canada, and provide an estimated cost assessment of unnecessary testing.
The total laboratory testing volumes of two common test panels were collected retrospectively for one-year from January to December 2018. Data was collected on test panels performed in an emergency room (ER) and inpatient setting from three separate Laboratory Information Systems covering the provincial population in Alberta, Canada. Total initial test panel instances, total repeated panels, repeated panels that were previously normal or abnormal, and estimated costs were examined. Cost assessment was completed based on Reference Median Cost (RMC) analysis for each of these two common test panels.
During the study period, 2,020,467 (CBC) and 1,455,983 (EP) initial test panel instances were recorded, of which 67.7% and 73.5% were repeated for the CBC and EP, respectively. There was a higher proportion of EP repeated inappropriately (previously normal; 35.6%) compared to CBCs (5.4%). The cost to the province for inappropriately repeating CBC and EP were estimated to be RMC $0.52 million and RMC $1.90 million CAD, respectively.
Results from this study can assist policy makers in implementing utilization management initiatives and update clinical practice guidelines to reduce costs to healthcare without compromising patient care.
Reducing repeat imaging in hepato-pancreatico-biliary surgical cancer care through shared diagnostic imaging repositories
2019, HPB
Citation Excerpt :
This decreased use of repeat imaging resulted in more efficient care with shorter wait times to definitive surgery within the examined population. Repeat imaging rates have been documented between 9 and 40% in prior single-centre reports mostly focussing on a single imaging modality.8,28–30 Detailed data on use of repeat imaging and DIR is sparse.
With regionalization of care, patients often undergo treatment in institutions other than where the initial investigation is conducted. This study assessed the impact of a shared diagnostic imaging repository (SDIR) on processes of care and outcomes in hepato-pancreatico-biliary (HPB) cancer surgery.
Provincial administrative datasets were linked to study HPB cancer patients operated at a regional cancer centre (2003–2014). SDIR and non-SDIR groups were based on where initial imaging (CT or MRI) was conducted. Outcomes were repeat imaging before surgery and wait times for surgery from initial imaging and surgical consultation.
Of 839 patients, 474 were from SDIR institutions. Fewer SDIR patients underwent any repeat imaging (55.9% vs. 75.3%; p < 0.01) and repeat imaging with same modality and protocol (24.7% vs. 43.0%; p < 0.01). Median wait time to surgery from initial imaging (64 Vs. 79 days; p < 0.01) and surgical consultation (39 Vs. 45 days; p = 0.046) was shorter with SDIR. SDIR patients had lower adjusted odds of any repeat imaging (OR 0.20 [0.14–0.30]), and repeat imaging with same modality and protocol (OR 0.58 [0.41–0.80]).
Radiology sharing with SDIR reduced repeat imaging for HPB cancer surgery, including potentially redundant repeat imaging with same protocol, and shortened wait time to surgical care.
Optimizing Order Entry Automaticity Reduces Inpatient Laboratory Utilization
2018, American Journal of Medicine
Citation Excerpt :
Although CDS tools can reduce clinically unnecessary lab testing, when they are set up in an opt-out manner in which an override function is available, the bypass option is often used, limiting their effectiveness. In accordance with other studies,4,9,19,20 our results support the notion that the most successful interventions to the CPOE interface limit ease of ordering (particularly automatically repeating laboratory studies), thus prompting active clinical reassessment by providers to determine order necessity. This is particularly relevant in inpatient settings, where routine daily lab tests account for a significant portion of wasted care in the form of low-cost, high-volume testing.

View all citing articles on Scopus

^☆: Supported in part by grants from the Culpeper Foundation, and the Agency for Health Care Policy and Research, Rockville, Maryland (Research Grant R01-HS08927).

View full text

Clinical StudiesWhat Proportion of Common Diagnostic Tests Appear Redundant?☆

Abstract

Section snippets

Patient Population

Test Selection and Definitions

Results

Discussion

Acknowledgements

Ther Drug Monit

Hum Pathol

Changing physician test ordering in a university hospital. An intervention of physician participation, explicit criteria, and feedback

Arch Intern Med

Feedback reduces test use in a health maintenance organization

JAMA

Physician characteristics and organizational factors influencing use of ambulatory tests

Med Decis Making

Variation among physicians in use of laboratory tests. Relation to quality of care

Med Care

Variation among physicians in use of laboratory tests. II. Relation to clinical productivity and outcomes of care

Med Care

Misuse of laboratory tests and diagnostic procedures

NEJM

Use of laboratory tests and pharmaceuticals. Variation among physicians and effect of cost audit on subsequent use

JAMA

Computer-based audit to detect and correct overutilization of laboratory tests

Med Care

A controlled trial to decrease the unnecessary use of diagnostic tests

J Gen Intern Med

Ultilization of clinical chemistry services by medical house staff. An analysis

Arch Intern Med

Influencing behavior of physicians ordering laboratory testsa literature study

Med Care

Improving house staff ordering of three common laboratory tests. Reductions in test ordering need not result in underutilization

Med Care

Inappropriate testing for diarrheal diseases in the hospital

JAMA

The admission urinalysisimpact on patient care

J Gen Intern Med

The unproven utility of preoperative urinalysis. Clinical use

Arch Intern Med

A prospective evaluation of therapeutic drug monitoring

Ther Drug Monit

Clinical Studies
What Proportion of Common Diagnostic Tests Appear Redundant?☆