Interventions to reduce low-value imaging – a systematic review of interventions and outcomes

Kjelle, Elin; Andersen, Eivind Richter; Soril, Lesley J. J.; van Bodegom-Vos, Leti; Hofmann, Bjørn Morten

doi:10.1186/s12913-021-07004-z

BMC Health Services Research

Table 3 Characteristics of the included studies, outcome of the intervention and quality assessment result

From: Interventions to reduce low-value imaging – a systematic review of interventions and outcomes

Author (year)	Country	Methods	Population	Clinical setting	Intervention	Low-value practice	Outcome of intervention(s)	Quality assessment
Arora et al. (2020) [126]	USA	Evaluation of a quality improvement project	1535 children	Emergency department	Education, decision support, and performance feedback	Head CT	4.6%-points reduction in use	*
Ashykian et al. (2019) [37]	USA	Retrospective review	1000 reports	Orthopedic and family practice departments	Education	Repeat radiographs for routine follow-up of osteoarthrosis	50% reduction in use	*
Bailey et al. (2013) [38]	USA	Longitudinal data analysis	800 patients	Emergency department	Health information exchange	Repeated lumbar or thoracic imaging	64% lower odds of repeat	*
Bailey, Wan et al. (2013) [39]	USA	Longitudinal data analysis	1252 patients	Emergency department	Health information exchange	Diagnostic neuroimaging	62% lower odds of repeat	*
Bairstow et al. (2010) [40]	Australia	Pre/post audit	1061 patients	Emergency department	Education, request forms	Chest CTA, abdominal XR, imaging with a provisional diagnosis of renal colic and XR despite negative Ottawa Ankle Rule	13–28% reduction in use	*
Baker et al. (2020) [41]	USA	Retrospective registry review	445 patients	Emergency department	Education and guideline	C-spine CT	> 30% reduction in use	*
Ballard et al. (2019) [127]	USA	Nonrandomized clinical trial - secondary analysis	3859 children	Emergency department	Electronic clinical decision support	Head CT	2.6% unsignificant reduction in use	*
Bhatia et al. (2013) [44]	USA	Prospective, time series analysis	1711 patients	Academic medical center	Education	Transthoracic echocardiography	26% reduction in use
Bhatia et al. (2014) [42]	USA	Randomized Control Trial	88 internal medicine residents and 24 cardiology fellows	Academic medical center	Education	Transthoracic echocardiography	62% reduction in use	*
Bhatia et al. (2017) [43]	Canada/USA	Multicenter, randomized controlled trial	196 physicians	8 hospitals	Education and feedback	Transthoracic echocardiogram	1.3% significant reduction in use	*
Blackmore et al. (2011) [45]	USA	retrospective cohort study	6141 patients	Medical Center	Decision support system	Lumbar MRI, Brain MRI in headache, and sinus CT	23.2–26.8% significant lower use
Bookman et al. (2017) [46]	USA	Longitudinal, before/after study	235,858 patient visits	5 emergency departments	Clinical decision support system	Head/c-spine CT, Chest CTA	> 6% significant decrease in head and c-spine CT 2% non-significant reduction chest CTA	*
Boutis et al. (2013) [47]	Canada	Interrupted time series with pair matched control design.	2151 children	6 emergency departments	Education, reminders, and computerized decision support system	Ankle XR	22% reduction in use	*
Breakell et al. (2018) [48]	UK	Retrospective audit	101 children	District General Hospital	Education and guideline	Chest XR for Bronchiolitis	16% reduction in use
Buntine et al. (2018) [49]	Australia	Before and after study	2931 scans	3 hospitals	Flowchart	Chest CTA and NM ventilation perfusion	6 per 1000 scans reduction in use	*
Carnevale et al. (2015) [128]	USA	Before-and-after study	29,395 encounters	Emergency department	Decision support and education	Several	5–10% reduction in use	*
Carpenter et al. (2020) [50]	USA	Retrospective chart review	1010 children	Hospital	Choosing Wisely	CT/MR or US in cryptorchidism	No significant difference	*
Chandra et al. (2019) [51]	Canada	Evaluation of a quality improvement project	37 physicians	Community teaching hospital	Didactic seminar, Survey	Imaging for low back pain	4%-points significant increase	*
Chang et al. (2018) [52]	USA	Before-and-after study	202 family physicians and 8 general internists	Clinics	Feedback reports	CT, MRI, and PET	14.5% reduction in use	*
Char et al. (2014) [53]	USA	Retrospective chart review	510 patients	Emergency department	Increased D-dimer threshold value and clinical probability assessment	Chest CTA	7%-points Increase in diagnostic yield
Chen et al. (2020) [54]	USA	Before-and-after study	Physicians	Hospital	Alerts	Imaging for lower back pain	9.6% reduction in total imaging rate and MRI (14.9%), No significant difference in use of CT/XR	*
Chien et al. (2017) [55]	USA	Block randomized controlled trial	1205 clinicians	Multidisciplinary medical group	Financial incentives	CT or MRI for single headache or lower back pain, acute, uncomplicated rhinosinusitis, or DEXA for low risk for osteoporosis	No significant difference	*
Depinet et al. (2016) [56]	USA	Interrupted time series trial	1886 children	Urban tertiary care hospital	Clinical decision support system and pathway	Abdominopelvic CT/US	2%-points increased use of US and 5%-points reduction in use of CT	*
Doyle et al. (2019) [57]	USA	Randomized study	3524 practitioners	15 hospitals and 150 clinics	Best practice alerts	Several	6% reduction in use
Drescher et al. (2011) [58]	USA	Before and after study	404 cases	Emergency department	Algorithm	Chest CTA	4.4% increase in use	*
Dudzinski et al. (2016) [59]	USA	Before and after study	65 cardiologists	Ambulatory cardiology practices in hospital	Education and feedback	Transthoracic echocardiography	6%-points reduction in use	*
Dunne et al. (2015) [60]	UK	Before and after study	5892 examinations	Hospital	Clinical decision support system	Chest CTA	12.3% reduction in use	*
Durand et al. (2013) [61]	USA	Randomized controlled trail	10 imaging tests	Tertiary teaching hospital	Cost display	Several	No significant difference	*
Ehrlichman et al. (2017) [129]	USA	Before-and-after study	104,454 patients	Emergency department	Feedback	Several	2.3% reduction	*
Fallon et al. (2016) [62]	USA	Prospective, longitudinal study	Children	Level I pediatric trauma center	Development of Trauma Protocol	Abdominal CT for abdominal trauma	18%-points increase in diagnostic yield	*
Ferguson et al. (2017) [63]	USA	Interventional improvement project	1 Emergency department	Hospital	Education and diagram	Abdominal XR constipation	38%-points reduction	*
Flamm et al. (2013) [64]	Austria	Non-randomized controlled trial with a historical control group	1363 patients	Hospital	PReOPerative evaluation’ (PROP)	Chest XR	21.7%-points reduction
French et al. (2013) [65]	Australia	Cluster Randomized trail	112 general practitioners	Practices	Guideline/ facilitated interactive workshops	Lumbar CT or XR	No significant difference
Gertz et al. (2016) [66]	USA	Before-and-after study	941 patients	Hospital	Computerized order entry tool	Cardiac stress tests with imaging	No significant difference
Goldberg et al. (2011) [67]	USA	Retrospective cohort chart review	742 patients	Hospital	Guideline	Head CT	16% reduction in use
Graves et al. (2018) [68]	USA	Interrupted time series	76,119 compensation claims	Regional	Policy implementation	Imaging for lower back pain	5,6%-points significant decrease in use of MRI. 2.46% increase in the use of XR No change in use of CT
Hardin et al. (2017) [69]	USA	Pre−/post-test design	339 patients	Hospital	Complex Care Map	CT scans	62% reduction in use
Hess et al. (2018) [70]	USA	Cluster Randomized Trial	172 clinicians	Emergency department	Shared decision-making	Head CT (children)	No significant difference
Hong et al. (2017) [71]	USA	Retrospective chart review	1,547,870 patients	Several	Choosing wisely	Imaging for lower back pain	4% reduction in use
Hoo et al. (2011) [72]	USA	Retrospective chart review	457 examinations	Hospital	Mandatory clinical decision rule and selective d-dimer use	Chest CTA	13%-points Increase in diagnostic yield	*
Hui et al. (2014) [73]	USA	Prospective cohort study and retrospective review	762 patients	Hospital	Education and guideline	Pelvic US	58% reduction in use
Hurley et al. (2017) [74]	USA	Before-and-after study	10,554 patients	Hospital	MUSIC imitative Collaborative	Bone scan and CT for prostate cancer	4.5–7%-points reduction in use	*
Ip et al. (2013) [78]	USA	Before-and-after study	1.8 million patient-months	Hospital	Computerized order entry tool with clinical decision support systems and accountability tools	Multiple	12% reduction in use	*
Ip et al. (2014) [75]	USA	Retrospective cohort study	21,445 LBP-related primary care visits	Academic quaternary care hospital	Clinical decision support system	Imaging for low back pain	30.8% reduction in use	*
Ip et al. (2015) [77]	USA	Before-and-after study	Emergency department patients	Hospital	Clinical decision support system	Head CT	13.4% reduction in use	*
Ip et al. (2017) [76]	USA	Before-and-after study	98,894 radiologyorders	Four institutions	Clinical decision support system	Several	1%-point reduction in use
Jennings et al. (2017) [79]	USA	Evaluation of a quality improvement project	1346 Children	Community emergency department	Protocol, education, and individual feedback.	CT head of children, minor head injury	12% reduction in use
Judkins et al. (2013) [80]	Australia	Retrospective chart review	659 children	Tertiary children’s hospital	NICE guidelines	Ultrasound urinary system, MUCG and dimercaptocsuccinic acid scintigraphy	50% reduction in the use
Kandiah et al. (2020) [81]	Canada	Evaluation of a quality improvement project	4480 patients	Hospital	Education and information packages	MRI and CT of joints without red flags	CT 43% reduction in use MRI 0.6% reduction of use	*
Kanaan et al. (2013) [82]	USA	Retrospective chart review	200 patients	Tertiary emergency department	Education	Chest CTA	No significant difference	*
Keveson et al. (2017) [83]	USA	Evaluation of a quality improvement project	All invasive mechanical ventilator patients	Tertiary hospital	Education and change in routines/referral system	Daily CXR of ventilated patients	64% reduction in use	*
Kobes et al. (2020) [130]	Canada	Retrospective chart review	28 medical imaging sites	Mobile radiography	Guideline	Chest XR	3.2% reduction in use	*
Lacson et al. (2017) [84]	USA	Retrospective chart review	63,222 orders	Hospital	Clinical decision support system	Several	No significant difference
Levitt et al. (2015) [120]	USA	Before-and-after study	415 patients	Hospital	Decision support and education	Stress echocardiography	12%-points reduction in use	*
Lu et al. (2012) [85]	USA	Retrospective chart review	267 patients	Hospital	Importing images from other institutions into PACS	Repeat imaging	61%-points reduction in use
Luther et al. (2019) [86]	USA	Retrospective chart review	273 patients	Hospital	Standardized clinical assessment and management plans	Wrist XR	60% reduction in use	*
Masood et al. (2020) [87]	Canada	Evaluation of a quality improvement project	Parents in the emergency department	Tertiary care center	Education, guideline and checklist, patient handouts, and feedback	Head CT in adults	7–14% reduction in use
McGrew et al. (2018) [88]	USA	Retrospective chart review	1934 children	Pediatric Level 2 Trauma Center	Guideline	Head and abdomen/pelvis CT pediatric trauma	11.5–18.8% reduction in use
Mills et al. (2018) [121]	USA	Before-and-after study	7987 patients	Emergency department	Decision support	Chest CTA	2.5% increased diagnostic yield	*
Min et al. (2017) [89]	Canada	Retrospective chart review	4562 patients	Emergency department	Checklist	Imaging in low back pain	22% reduction in use
Mittal et al. (2014) [91]	USA	Before and after design	3641 patients	Tertiary care hospital	Clinical practice guidelines	Chest XR	14.6–20%-points reduction in use
Moriarity et al. (2015) [91]	USA	Retrospective chart review	33,311 patients	Hospital	Clinical decision support	Inpatient MR, CT, and NM	No significant difference
Mulders et al. (2020) [92]	The Netherlands	Before-and-after comparative prospective cohort study	1261 patients	Emergency department	Amsterdam Wrist Rules	Wrist XR	15%-points reduction in use
Mäenpää et al. (2011) [93]	Finland	Retrospective, longitudinal study	1 regional hospital	Hospital	Regional Health InformationExchange	Several	16.4% reduction in use	*
Nigrovic et al. (2015) [94]	USA	Multifaceted quality improvement initiative	Children < 21 years	Urban tertiary care academic center	Guideline development, feedback, and education	Head CT	6%-points reduction in use	*
O’Connor et al. (2014) [95]	USA	Prospective before and after study	28,420 CT orders	Tertiaryacademic medical center	Requiring a clinical justification to override a repeat CT alert	Several	Prevented 1 in 13 scans	*
Ong et al. (2013) [122]	USA	Before-and-after study	471 patients	Hospital	Algorithm and guideline	Chest CTA	26% reduction in use	*
Ostby et al. (2020) [96]	USA	Evaluation of a quality improvement project	235 patients	Emergency department	Specialist consolation before imaging	CT of gynecological cancer patients	54%-points reduction in use	*
Palen et al. (2019) [97]	USA	Stepped-wedge study	31,426 orders	several	Check boxes	Several	Modest increase in appropriateness
Parikh et al. (2016) [98]	USA	Retrospective cohort study	220,539 patients	Hospital	Guideline	Chest XR	6.4% significant reduction in use	*
Poeran et al. (2019) [99]	USA	Retrospective interrupted time series	27,549 orders	Emergency department	Clinical decision support	Low appropriateness imaging	9%-points reduction in use	*
Prevedello et al. (2013) [100]	USA	retrospective cohort study	2891 patients	Emergency department	Alerts	Chest CTA	2.2/1000 reduction in use	*
Puffenbarger et al. (2019) [123]	USA	Retrospective chart review	556 visits	Emergency department	Education, guideline, and hand-outs	Head CT	21.6%-points reduction in use	*
Pugel et al. (2018) [101]	USA	Retrospective interrupted time series	213,532 consultations	Ambulatory care	Education, feedback and guidelines	DEXA and head CT	23.4% reduction in use
Raja et al. (2012) [103]	USA	Retrospective cohort study	6838 patients	Quaternary care institution	Clinical decision support system	Chest CTA	20.1% reduction in use	*
Raja et al. (2015) [102]	USA	Prospective randomized controlled trial	2167 patients	Urban level 1 adult trauma center	Feedback	Chest CTA	2/1000 reduction in use	*
Reiter et al. (2018) [104]	Israel	Prospective cohort study	544 children	Pediatric Emergency department	Education, guideline cards at computers	Chest XR for Bronchiolitis	20%-points reduction in use	*
Rezaii et al. (2020) [105]	USA	Cohort study	27 practices, 4601 cases	Academic and private practices	Online educational material and feedback	Chest CTA, advanced imaging of low back pain, follow-up of adnexal cysts	3% reduction in use	*
Rosati et al. (2015) [106]	USA	Retrospective review	233 children	Level I trauma center	Guideline	C-spine CT	23% reduction in use
Sclafani et al. (2010) [107]	USA	Retrospective chart review	1092 patients	Hospital	Education	Head CT, brain MRI and carotid US of syncope patients	No significant difference
Shah et al. (2016) [108]	USA	Chart review	824 children	Emergency department	Diagnostic algorithm	Abdominal CT appendicitis	51.2%-points reduction in use
Shelton et al. (2015) [131]	USA	Before-and-after study	2001 patients	Hospital	Feedback	Several	38% reduction in use	*
Singer et al. (2014) [109]	USA	Before and after trial design	34,961 children	Academic medical center	Opening a dedicated pediatric ED	Several	3.2% reduction in use
Sodickson et al. (2011) [110]	USA	Retrospective chart review	1487 patients	Tertiary care, level I trauma center	Health InformationExchange	Several	16–18% reduction in use
Sy et al. (2016) [111]	Canada	Evaluation of a quality improvement project	1492 patients	Intensive care unit	Education, posters, and change in order system	Chest XR	26% reduction in use	*
Tajmir et al. (2017) [124]	USA	Randomized controlled trail	613 patients	Hospital	Decision support	Ankle XR	Modest change in diagnostic yield
Tyler et al. (2018) [112]	USA	Evaluation of a quality improvement project	2211 patients	Children’s hospital	Education, feedback, sign pledge, visualizing algorithm, guideline	Chest XR	12.1%-points reduction in use
Vartanians et al. (2010) [113]	USA	Retrospective study	118,975 orders	Several	Change in ordering system	Several	3.5%-points reduction in use	*
Walen et al. (2016) [125]	USA	Prospective observational study	250 patients	Hospital	Wells-score documentation	Chest CTA	6.6% increase in diagnostic yield
Walker et al. (2020) [114]	Canada	Retrospective chart review	302 Consultations	Primary care	Electronic communication system	Abdominal, musculoskeletal, neuro, and thoracic imaging	28% reduction in use
Wang et al. (2018) [115]	USA	Retrospective chart review	3 clinics	Primary care	On-site and online education and feedback	Lumbar MRI	3.7 MRIs reduced per month
Wu et al. (2020) [116]	USA	Evaluation of a quality improvement project	6441 Chest XRs	Medical Intensive care unit	Survey, journal club, discussions, posters, alerts in electronic referral system, education, pocket cards, electronic decision tool	Chest XR	36.1% reduction in use
Xu et al. (2020) [117]	Canada	Retrospective review	400 referrals	Tertiary care center	Mandatory check list	Knee MRI	48% reduction in use	*
Zafar et al. (2019) [118]	USA	Randomized cohort study	54 providers	Tertiary academic health system	Algorithm and alerts	Low back imaging	No significant difference	*
Zamora-Flores et al. (2015) [119]	USA	Retrospective chart review	322 children	Rural community hospital	Guideline	Chest XR	No significant difference

CT computed tomography, CTA CT angiography, DEXA dual-energy x-ray absorptiometry, MMAT Mixed Methods Appraisal Tool, MRI magnetic resonance imaging, MUCG micturating cystourethrogram, NM nuclear medicine, PET positron emission tomography, US ultrasound, XR X-ray
* Fulfill all MMAT criteria
Italic = Mixed results, Bold = Reduction in use/increase in rate of diagnostic yield, Normal = no significant difference/increase

Back to article page

ISSN: 1472-6963

Contact us

General enquiries: journalsubmissions@springernature.com