Contributions to the literature
-
We systematically reviewed the literature on the de-implementation of low-value services in cancer care delivery.
-
Auditing and providing feedback, having a clinical champion, educating clinicians through developing and disseminating new guidelines, and developing a decision support tool that is often integrated within the electronic health record system are the common components of the de-implementation interventions.
-
Our findings highlight the need for moving from passive de-implementation (i.e., clinicians voluntarily follow the new guidelines and decide to change the way they practice) to active de-implementation (i.e., organizations initiating interventions aimed at reducing the low-value care).
Background
Methods
Study inclusion and exclusion criteria
Study characteristic | Inclusion criteria | Exclusion criteria |
---|---|---|
Population | • Hospitals/clinics • Inpatient units • Outpatient general medical settings (e.g., primary care, urgent care, private offices) • Cancer centers • Emergency departments • Managed care organizations | • Health insurance • Free standing EDs • Nursing home |
Intervention | • Interventions that purposefully developed to removea, replaceb, reducec, restrictd, reverse, de-implement, de-adopt, disinvest, decrease in use, discontinue, abandon, reassess, obsolete, withdraw, contradict, refute, delist, substitute, exnovate, cease, or end an established low-value practice | • Changes in clinicians’ practice pattern over time in response to educational campaigns, guidelines, or dissemination of scientific publications without active effort to de-implement an established low value practice • Quality improvement interventions without a de-implementation component |
Reasonse | • Low value practicese (e.g., ineffectivef, contradictedg, mixedh, and untestedi interventions) | |
Outcome | • De-implementation determinants (i.e., factors influence de-implementation outcomes such as incentives and resources) • De-implementation process (i.e., process of reducing, replacing, or stopping low-value services) • De-implementation outcome (e.g., effectiveness, volume of procedures, cost saving, quality) | • Any outcomes not listed |
Study design | • Randomized trials • Quasi-experiment studies • Cross-sectional • Qualitative studies • Case reports and case studies • Interrupted time-series studies or repeated measures studies • Prospective and retrospective observational studies (i.e., cohort studies, case control studies) | • Descriptive studies with no outcomes data • Modeling studies that used simulated data • Not a clinical study (e.g., editorial, nonsystematic review, letter to the editor) • Prospective and retrospective observational studies • Clinical guidelines • Measurement or validation studies • Pilot studies without adequate power to assess impact of intervention on outcomes. |
Publication types | • Full publication in a peer-reviewed journal • English-language publications • 1990 to current date | • Non-English language • Not a full publication in a peer-reviewed journal • Letters, editorials, reviews, dissertations, meeting abstracts, protocols without results |
Information sources and search strategy
Study selection process
Study quality assessment
Data extraction and analysis
Results
Study selection
Characteristics of included studies
Citation | Design | Country | Setting | Participants | Outcomes | Low-value care | Study primary objective |
---|---|---|---|---|---|---|---|
Durieux et al. (2003) [38] | Interrupted time series | France | An academic medical center | Patients with GI tumors | The number of tumor markers ordered by physicians and the number of admissions | Inappropriate ordering of three tumor markers (carcinoembryonic antigen, alpha-fetoprotein, carbohydrate antigen 19-9) | To evaluate the long-term impact of an intervention designed to reduce the ordering of three tumor markers frequently prescribed for gastroenterological diseases (carcinoembryonic antigen, alpha-fetoprotein, carbohydrate antigen 19-9). |
Miller et al. (2011) [43] | Before and after study | USA | Multi-site urology practices | Patients with prostate cancer | Use of bone scans and computerized tomography across prostate cancer risk strata | Imaging in patients with low-risk prostate cancer | To describe findings from a Urological Surgery Quality Collaborative project focused on improving the use of radiographic staging in men with newly diagnosed prostate cancer. |
Butler et al. (2015) [37] | Before and after study | UK | An academic medical center | Patients with hematologic cancers | The proportion of noncompliant transfusions received above the recommended triggers. The total number of RBCs and PLTs received during the study period, proportion of patients transfused, mean pretransfusion Hb level and PLT count, mean post-transfusion Hb level and PLT count, and time delay between pre- and post-transfusion full blood count and the receipt of blood products. | Unnecessary blood transfusion | To assess the impact of a clinical decision support system for blood product ordering in patients with hematologic disease. |
Ross et al. (2015) [39] | Before and after study | USA | Urology practices | Patients with prostate cancer | The number of bone scan and CT scans | Imaging in patients with low-risk prostate cancer | To determine whether collaborative-wide data review and performance feedback would decrease the imaging rate in men with low-risk prostate cancer. |
Shelton et al. (2015) [40] | Interrupted time series | USA | Outpatient clinics, academic and ambulatory care centers (VA Medical Centers) | Patients with prostate cancer | Monthly PSA-based prostate cancer screening rate in unique patients who had a visit to any primary care clinic. | PSA-based screening for prostate cancer in men aged 75 years and older | To determine whether a highly specific computerized clinical decision support alert to remind providers, at the moment of PSA screening order entry, of the current guidelines and institutional policy would reduce the use of inappropriate PSA-based prostate cancer screening among men aged 75 and over. |
Martin Goodman et al. (2016) [42] | Before and after study | USA | A comprehensive cancer center | Patients with non–small-cell lung cancer | Patients with non–small-cell lung cancer who received pegylated granulocyte colony-stimulating factor (pGCSF) for low- or intermediate-risk febrile neutropenia chemotherapy regimens | Inappropriate use of prophylactic pegylated granulocyte colony-stimulating factor in patients with less than 10% risk of neutropenic fever | To examine the baseline rate of primary prophylactic pGCSF administration for patients with non–small-cell lung cancer, increase provider awareness of appropriate pGCSF use, and minimize the prescription of primary prophylactic pGCSF for patients with lung cancer who are treated with low-risk chemotherapy regimens, without a negative impact on patient safety. |
Sheridan et al. (2016) [32] | Randomized clinical trial | USA | Community-based practices | Patients with prostate or colorectal cancer | The change in intention to accept screening. General and disease-specific knowledge, perceived risk and consequences of disease, screening attitudes, perceived net benefit of screening, values clarity, and self-efficacy for screening. | Prostate cancer screening in men ages 50–69 years and colorectal cancer screening in men and women ages 76–85 years | To examine the comparative effectiveness of 4 alternate formats for presenting benefits and harms information in reducing intentions for screening and changing secondary behavioral and decision-making outcomes for patients eligible for 1 of 3 low-value or potentially low-value screening services. |
Hill et al. (2018) [33] | Before and after study | USA | A comprehensive interdisciplinary breast center | Patients with breast cancer | Frequency of ordering CBC and LFTs (overall and per provider), subsequent testing prompted by abnormal results, and overall compliance with guidelines. | Ordering complete blood cell count and liver function tests in patients with early breast cancer | To measure compliance with guidelines for ordering complete blood cell count (CBC) and liver function tests (LFT) before and after the calendar date when the guidelines transitioned from routine to unnecessary. |
Gob et al. (2019) [41] | Before and after study | UK | A tertiary care unit | Patients admitted to oncology unit | The percentage of one-unit red cell transfusion orders (aggregated monthly). | Two-unit red cell transfusion orders. | Assess the proportion of one-unit red cell transfusion orders on the oncology ward |
Hoque et al. (2020) [36] | Observational cohort study | USA | VA Medical Centers | Patients with colorectal, breast and non-small cell lung cancer | Erythropoisis stimulating agent treatment use and transfusion, and venous thromboembolism occurrence and mortality | ESA treatment use and transfusion | Evaluate the influence of FDA black box warnings and risk evaluation monitoring strategies on use of erythropoiesis stimulating agent in Veterans Administration cancer patients with chemotherapy induced anemia. |
Ciprut et al. (2020) [35] | Before and after study | USA | VA Medical Center | Patients with prostate cancer | Effectiveness (number of imaging) and acceptability of an EMR-based Clinical Reminder Order Check intervention | Imaging in patients with low-risk prostate cancer | To understand how to potentially improve inappropriate prostate cancer imaging rates. |
Laan et al. (2020) [34] | Interrupted time series | Holland | University and general hospitals | Patients admitted to oncology unit | Percentages of short peripheral intravenous catheters, catheter-related infections and other complications, catheter reinsertion rate, use of antibiotics, hospital length of stay (and ICU), and mortality | Inappropriate use of peripheral intravenous and urinary catheters | To reduce inappropriate use of catheters to reduce health care-associated infections. |
Quality assessment of studies
De-implementation interventions’ characteristics
Citation | De-implementation intervention description | The effects of the de-implementation intervention | Determinants of the use of the de-implementation intervention |
---|---|---|---|
Durieux et al. (2003) [38] | A specific laboratory order form with clinical guidelines to improve appropriate test orders. | The number of tumor markers prescribed, and the ratio tumor markers/admissions decreased in the hospital (p < .0001), and in the Departments of Gastroenterology (p < .0001) and Internal Medicine (p < .01). | Local adaptation of guidelines by those who are going to use them, implementation strategy for guidelines, and scientific knowledge concerning the utility of different markers. |
Miller et al. (2011) [43] | A multistep intervention including (1) audit and comparative performance feedback, (2) having a clinical champion, (3) dissemination of clinical guidelines, and (4) establishing the Urological Surgery Quality Collaborative as an infrastructure for physician led, collaborative quality improvement in urology. | Compared with baseline practice patterns (31% bone scans, 28% computerized tomography), urologists in Urological Surgery Quality Collaborative practices ordered fewer bone and computerized tomography scans in post-intervention phases 2 (23%, 21%) and 3 (16%, 13%) of data collection (p < 0.01), including a significant reduction in the use of these studies in patients with low and intermediate risk cancer (p < 0.05). | Not reported |
Butler et al. (2015) [37] | Computerized physician order entry systems have been integrated with a clinical decision support system software to improve compliance with restrictive blood management protocols. Such systems require physicians to specify the indication for blood product transfusion and highlight to the clinician the requests that lie outside prespecified guidelines for transfusion by linking them to the most recent laboratory results. In addition, extensive, real-time education, support, and feedback were provided to clinicians. | There was no significant difference in (1) the mean number of transfusions per patient, (2) the proportion of patients transfused, (3) post-transfusion hemoglobin (Hb), and (4) pre- and post-transfusion PLT count, although mean pretransfusion Hb decreased. The proportion of noncompliant RBC and PLT transfusion requests improved from baseline to CDSS2 (69.0 to 43.4% p < 0.005 for RBCs, and 41.9 to 31.2%, p = 0.16 for PLT). | The amount of time and human resources required to provide monitoring, analysis, and feedback, and provider reluctance. |
Ross et al. (2015) [39] | A multistep intervention including (1) audit and performance feedback, (2) having a clinical champion. | Bone scan decreased from 3.7 to 1.3% (p = 0.03), and computerized tomography decreased from 5.2 to 3.2% (p = 0.17). | Not reported |
Shelton et al. (2015) [40] | A clinical computerized decision support (CCDS) tool to remind providers of current recommendations against PSA-based prostate cancer screening for men 75 and older. A pop-up message to alert providers ordering a screening PSA test in a patient 75 years of age or older. When triggered, a brief interruptive educational message was shown on the ordering screen. | The mean monthly screening rate decreased from 8.3 to 4.6%. The screening rate declined by 38% during the baseline period and by 40% and 30%, respectively, during the two periods when the CCDS tool was turned on. The screening rate ratios for the baseline and two periods when the CCDS tool was on were 0.97, 0.78, and 0.90, respectively, with a significant difference between baseline and the first CCDS-on period (p < 0.0001), and a trend toward a difference between baseline and the second CCDS-on period (p = 0.056). | The alert fatigue, difficulty in changing providers’ behavior, and the rotation of resident physician staff |
Martin Goodman et al. (2016) [42] | Three Plan-Do-Study-Act cycles, educated providers about the appropriate use and cost of pGCSF, developed the Cleveland Clinic consensus guidelines, removed primary prophylactic pGCSF from LRCR EMR orders. | The percentage of patients who received inappropriate primary prophylactic pGCSF and the number of doses per patient decreased significantly. Cost analysis showed an average 86% decrease in billed charges per month, which would result in $408,000 in annual savings based on the current CMS allowable payment per dose. | Not reported |
Sheridan et al. (2016) [32] | One-page, written evidence-based decision support sheet. | Intentions to accept screening were high before the intervention and change in intentions did not differ across intervention arms (words, − 0.07; numbers, − 0.05; numbers plus narrative, − 0.12; numbers plus framed presentation, − 0.02; P = .57 for all comparisons). Change in other outcomes also showed no difference across intervention arms. | Not reported |
Hill et al. (2018) [33] | A planned implementation strategy using levels of the National Quality Strategy including (1) learning and technical assistance, (2) measurement and feedback, (3) certification, accreditation, and regulation, (4) innovation and diffusion (of quality improvement strategies), (5) workforce development, (6) patient education, (7) reward providers, and (8) modify the existing electronic medical record synoptic documentation template. | The overall rate of compliance with guidelines for ordering a CBC and LFT was 82% and 87%, respectively. Segregated by the pre- and post-guideline change time period, the compliance rates for ordering a CBC and LFT were 78% and 87% (P = 0.076). | (1) Integrated health care systems, (2) resource availability (e.g., electronic medical records and funding for academic research assistants) |
Gob et al. (2019) [41] | A multistep intervention including (1) a root cause analysis targeted at discovering contributing factors to two-unit transfusion orders, including a retrospective audit of the previous month’s two-unit transfusions, structured brain-storming by the study authors, and focused interviews with house staff and attending physicians. (2) An educational campaign with an educational email, and a Grand Rounds presentation focusing on improving awareness of the Choosing Wisely guidelines. (3) A real-time audit and feedback, (4) focused oncologist interviews, (5) modify the transfusion orders setting. | Modifying the transfusion orders templates was the only intervention that resulted in an immediate and sustained change to the system. Post-intervention, the mean proportion of one-unit transfusions rose to 86.0% and was sustained for the 17 months of ongoing data collection. | Bias inertia toward low-value care (i.e., status quo bias) |
Hoque et al. (2020) [36] | FDA black box warnings and risk evaluation monitoring strategies | ESA use for epoetin fell from 22 to 1%, and for darbepoetin fell from 11 to 1% (p < 0.01). Mean hematocrit levels at ESA initiation decreased from 30 to 21% (p < 0.01). | National policies and regulatory decisions, patient consent |
Ciprut et al. (2020) [35] | Using the National Comprehensive Cancer Network’s guidelines, a Clinical Reminder Order Check (CROC) that alerts ordering providers of potentially inappropriate imaging orders in real-time based on patient features of men diagnosed with low-risk prostate cancer | The percentage of the men who were staged according to guidelines increased from 65 to 81%. Inappropriate imaging of men with low-risk prostate cancer was reduced by 16%. | Not reported |
Laan et al. (2020) [34] | A tailored multi- faceted intervention incudes an assessment of determinants of practice for inappropriate catheter use nurse education, physician champion, empowerment of nurses depending on the local situation of the participating hospital, audit and feedback, and additional interventions such as smart phrase for the daily patient report in electronic health records. | Inappropriate use of peripheral intravenous catheters decreased from 22.0 to 14.4% (incidence rate ratio [IRR] 0.65, 95% CI 0·56 to 0.77, p < 0·0001). An absolute reduction in inappropriate use of peripheral intravenous catheters from baseline to intervention periods of 6.65% (95% CI 2.47 to 10.82, p = 0·011). Inappropriate use of urinary catheters decreased from 32.4 to 24.1% (IRR 0.74, 95% CI 0.56 to 0.98, p = 0·013). An absolute reduction in inappropriate use of urinary catheters of 6.34% (95% CI – 12.46 to 25.13, p = 0·524). | Piloting the intervention, developing evidence-based and consensus-driven criteria for appropriate use of peripheral intravenous catheters |