A Multi-modal, Physician-centered Intervention to Improve Guideline-concordant Prostate Cancer Imaging

Status

Active, not recruiting

Phases

Unknown

Study type

Interventional

Source

ClinicalTrials.gov

Registry ID

NCT03445559

Enrollment

Registered

2018-02-26

Start date

2018-04-09

Completion date

2027-02-28

Last updated

2026-03-17

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Prostate Cancer

Keywords

Prostate Cancer, Imaging, Veterans Affairs, Guideline Adherence, Implementation

Brief summary

The primary aim of this study is to determine whether a multi-modal, physician-focused behavioral intervention can improve facility-level guideline-concordant utilization of prostate cancer staging imaging. Other aims of this study include to use mixed methods to explore physician influence on guideline-concordant imaging and to determine the cost and cost impact of a physician-focused behavioral intervention to improve guideline-concordant prostate cancer imaging.

Detailed description

Almost half of Veterans with localized prostate cancer (the most common non-cutaneous malignancy among US men) receive inappropriate, wasteful imaging. The VHA Blueprint for Excellence prioritizes increasing operational effectiveness. Prior studies seeking to limit inappropriate imaging did not assess barriers and achieved mixed results. The investigators' team has explored the causes of guideline-discordant prostate cancer imaging and found that 1) patients with newly diagnosed prostate cancer have little concern for radiographic staging but rather focus on treatment, 2) physician trust imaging guidelines but are apt to follow their own intuition, fear medico-legal consequences, and succumb to influence from colleagues who image frequently. In spite of such discrepant views, most VHA physicians suggested or supported a large-scale effort to improve imaging use across VHA. The investigators propose a multi-site, stepped wedge, cluster-randomized trial to determine the effect of a physician-focused behavioral intervention on VHA prostate cancer imaging use. The multi-level intervention, developed according to the Theoretical Determinants Framework, combines traditional physician behavior change methods with novel methods of communication and data collection. The intervention consists of three components: 1) a system of audit and feedback to clinicians informing individual clinicians and their sites about how their behavior compares to their peers' and to published guidelines 2) a program of academic detailing with the goal to educate providers about prostate cancer imaging, and 3) a CPRS Clinical Order Check for potentially inappropriate imaging. The intervention will be introduced to 10 participating geographically-distributed study sites. The investigators will assess imaging rates 6 months prior to the intervention and 3 months following the intervention. The study's specific aims seek to understand the effects of the intervention on 1) facility-level prostate cancer imaging rates, 2) physician experience with and perceptions of the intervention and its implementation, and 3) the costs of both implementing the intervention and affecting change in imaging use. These aims will support a subsequent intervention to improve guideline-concordant imaging across VHA. Experience gained through this project will be leveraged to improve guideline-concordant care and increase operational effectiveness in other domains.

Interventions

BEHAVIORALClinical Order Check

A Clinical Order Check is an evidence-based, systems-level method to affect significant behavior change. It addresses the intervention functions of education, enablement and incentivization which are effective methods to change behaviors driven by beliefs about capabilities, knowledge, social influences, beliefs about consequences, and environmental context and resources. All VA facilities currently use locally adapted clinical reminders. This strategy is technologically simple, straightforward, and is considered to be a best practice within the VA IT community. The reminder will be self-explanatory and non-intrusive to workflow.

BEHAVIORALAcademic Detailing

Academic detailing is an individual and facility-level intervention consistently shown to improve provider behavior. This strategy addresses the intervention functions of persuasion, coercion, modeling, and education which are effective methods for affecting behaviors driven by beliefs about capabilities, knowledge, social influences, beliefs about consequences, and environmental context and resources. During the meeting, the detailer will follow a script explaining that the visit is part of an experimental program to provide physicians with up-to-date, unbiased information about imaging to stage prostate cancer.

BEHAVIORALAudit and Feedback

Audit and feedback is an effective, individual-level intervention for changing healthcare provider behavior, resulting in small but potentially clinically important benefits. Audit and feedback addresses the intervention functions of education, persuasion and incentivization, all of which are important for addressing beliefs about capabilities and consequences, knowledge, and social influence.

Study design

Allocation

RANDOMIZED

Intervention model

CROSSOVER

Primary purpose

HEALTH_SERVICES_RESEARCH

Masking

SINGLE (Outcomes Assessor)

Intervention model description

The proposed behavioral intervention will be implemented according to a stepped wedge cluster-randomized design. This is a single direction cross-over randomized trial where every site serves, at some point, as both a control and an intervention site. The first time point will be a baseline measurement, where none of the study sites have yet received the intervention. At subsequent time points, study sites initiate the intervention. The time at which each site initiates implementation of the intervention is randomized.

Eligibility

Sex/Gender

ALL

Healthy volunteers

Inclusion criteria

Provider Criteria: * Urology Chiefs and attending urologists employed through the VA (full time, part time) at one of the 10 participating sites * Physician Assistants and Nurse Practitioners employed through the VA at one of the 10 participating sites that work in the respective urology clinics * Providers may be any gender or race/ethnicity Qualitative portion only: * Urology Chiefs and/or frontline staff physicians * participating PAs \& NPs having cared for at least 5 men with incident prostate cancer within the previous 6 months * Patients will not be directly recruited into the study. * The investigators have obtained a waiver of HIPAA authorization and informed consent to analyze electronic health records of patients that are diagnosed with ICD-9 code 185 or ICD-10 code C61 during the study period at the 10 participating sites.

Exclusion criteria

Provider Criteria: * Urology Residents will be excluded. Patients * Patients will be excluded if they have a history of prior malignancy * Are over the age of 85 * Diagnosed at autopsy or by death certificate * Died within 3 months of diagnosis * Not having data on at least one of the following: * PSA * clinical stage * Gleason score

Design outcomes

Primary

Measure	Time frame	Description
Facility-level Inappropriate Prostate Cancer Bone Scan Imaging Rates	Through study completion, an average of 4 years	Facility-level utilization of bone scan imaging among men with newly diagnosed, low-risk prostate cancer. (Inappropriate Imaging according to NCCN guidelines)
Facility-level Appropriate Prostate Cancer Bone Scan Imaging Rates	Through study completion, an average of 4 years	Facility-level utilization of bone scan imaging among men with newly diagnosed, high-risk prostate cancer. (Appropriate Imaging according to NCCN guidelines)

Secondary

Measure	Time frame	Description
Individual-level Inappropriate Prostate Cancer Imaging Rates	Through study completion, an average of 4 years	Provider-level utilization of nuclear medicine bone scan among men with newly diagnosed, low-risk prostate cancer (Inappropriate Imaging). Imaging is not recommended to stage low-risk prostate cancer men with PSA\<10, Gleason\<7, and clinical stage \<T3.
Individual-level Appropriate Prostate Cancer Imaging Rates	Through study completion, an average of 4 years	Provider-level utilization of bone scans among men with newly diagnosed, high-risk prostate cancer. (Appropriate Imaging)
Provider Attitudes Regarding Prostate Cancer Imaging Guidelines and the Behavioral Intervention	Through study completion, an average of 4 years	Qualitative outcome assessed through semi-structured, in-depth interviews with participating providers

Countries

United States

Contacts

PRINCIPAL_INVESTIGATORDanil V Makarov, MD MHS

VA NY Harbor Healthcare System, New York, NY

Participant flow

Recruitment details

Dates: March 2018 - March 2021 Type of location: VA Medical Centers Urology providers (physicians, physician assistants, nurse practitioners) from 10 VA medical centers who treat prostate cancer were enrolled in the study. Prostate cancer patient charts were reviewed but were not directly recruited into the study.

Pre-assignment details

There was a minimum 3 month (maximum 30 month) run-in period prior to provider exposure to the intervention. A total of 58 clinical providers were enrolled at 10 VAMC sites. At the 10 VAMC sites, 2,374 patient charts were reviewed, 2,358 met inclusion criteria and were assigned to an arm. 16 patient charts were excluded at baseline due to * Prior diagnosis of prostate cancer * History of other prior malignancy * Low risk & missing clinical stage or PSA

Participants by arm

Arm	Count
Control - Bone Scan Imaging Not Indicated Patient charts reviewed prior to intervention initiation in a stepped wedge design where a bone scan was not indicated according to National Comprehensive Cancer Network guidelines.	690
Intervention - Bone Scan Imaging Not Indicated Patient charts reviewed after intervention initiation where a bone scan was not indicated according to National Comprehensive Cancer Network guidelines. The intervention is comprised of three components: 1) Clinical Order Check, 2) Academic Detailing, and 3) Audit and Feedback. Clinical Order Check: A Clinical Order Check is an evidence-based, systems-level method to affect significant behavior change. It addresses the intervention functions of education, enablement and incentivization which are effective methods to change behaviors driven by beliefs about capabilities, knowledge, social influences, beliefs about consequences, and environmental context and resources. Academic Detailing: Academic detailing is an individual and facility-level intervention consistently shown to improve provider behavior. This strategy addresses the intervention functions of persuasion, coercion, modeling, and education which are effective methods for affecting behaviors driven by beliefs about capabilities, knowledge, social influences, beliefs about consequences, and environmental context and resources. Audit and Feedback: Audit and feedback is an effective, individual-level intervention for changing healthcare provider behavior. Audit and feedback addresses the intervention functions of education, persuasion and incentivization, all of which are important for addressing beliefs about capabilities and consequences, knowledge, and social influence.	734
Control - Bone Scan Imaging Indicated Patient charts reviewed prior to intervention initiation in a stepped wedge design where a bone scan was indicated according to National Comprehensive Cancer Network guidelines.	425
Intervention - Bone Scan Imaging Indicated Patient charts reviewed after intervention initiation where a bone scan was indicated according to National Comprehensive Cancer Network guidelines.	453
Intervention - Providers Enrolled Urology Providers (urologists, Physician Assistants, Nurse Practitioners) who treat prostate cancer patients at the 10 participating VA Medical Centers. The following baseline characteristics were not collected from enrolled providers: 1) Reside in Zip Code with High Education, 2) Reside in Zip Code with High Broadband Access, and 3) Marital Status.	58
Total	2,360

Baseline characteristics

Characteristic	Control - Bone Scan Imaging Not Indicated	Intervention - Providers	Total	Intervention - Bone Scan Imaging Not Indicated	Control - Bone Scan Imaging Indicated	Intervention - Bone Scan Imaging Indicated
Age, Continuous	64.9 Years STANDARD_DEVIATION 12.6	49.71 Years STANDARD_DEVIATION 5.09	67.3 Years STANDARD_DEVIATION 7.7	66.7 Years STANDARD_DEVIATION 6.8	68.0 Years STANDARD_DEVIATION 10.3	67.2 Years STANDARD_DEVIATION 10.5
Marital Status Married	372 Participants	—	1104 Participants	345 Participants	198 Participants	189 Participants
Marital Status Not currently married	307 Participants	—	1099 Participants	361 Participants	215 Participants	216 Participants
Marital Status Not reported	11 Participants	—	99 Participants	28 Participants	12 Participants	48 Participants
Race/Ethnicity, Customized Black	116 Participants	3 Participants	397 Participants	138 Participants	56 Participants	84 Participants
Race/Ethnicity, Customized Non-Black	531 Participants	55 Participants	1754 Participants	538 Participants	329 Participants	301 Participants
Race/Ethnicity, Customized Not reported	43 Participants	0 Participants	209 Participants	58 Participants	40 Participants	68 Participants
Region of Enrollment United States	690 Participants	58 Participants	2360 Participants	734 Participants	425 Participants	453 Participants
Reside in Zip Code with High Broadband Access	616 Participants	—	1938 Participants	604 Participants	368 Participants	350 Participants
Reside in Zip Code with High Education	485 Participants	—	1595 Participants	524 Participants	306 Participants	280 Participants
Sex: Female, Male Female	0 Participants	10 Participants	10 Participants	0 Participants	0 Participants	0 Participants
Sex: Female, Male Male	690 Participants	44 Participants	2346 Participants	734 Participants	425 Participants	453 Participants

Adverse events

Event type	EG000 affected / at risk	EG001 affected / at risk	EG002 affected / at risk
deaths Total, all-cause mortality	1 / 1,149	1 / 1,267	0 / 58
other Total, other adverse events	0 / 1,149	0 / 1,267	0 / 58
serious Total, serious adverse events	0 / 1,149	0 / 1,267	0 / 58

Outcome results

Primary

Facility-level Appropriate Prostate Cancer Bone Scan Imaging Rates

Facility-level utilization of bone scan imaging among men with newly diagnosed, high-risk prostate cancer. (Appropriate Imaging according to NCCN guidelines)