Abstract
Purpose:
Incidental adrenal masses (IAMs) are common but rarely evaluated. To improve this, we developed a standardized radiology report recommendation template and investigated its implementation and effectiveness.
Methods:
We prospectively studied implementation of a standardized IAM reporting template as part of an ongoing quality improvement initiative, which also included primary care provider (PCP) notifications and a straightforward clinical algorithm. Data were obtained via medical record review and a survey of radiologists. Outcomes included template adoption rates and acceptability (implementation measures), as well as the proportion of patients evaluated and time to follow-up (effectiveness outcomes).
Results:
Of 4995 imaging studies, 200 (4.0%) detected a new IAM. The standardized template was utilized in 54 reports (27.0%). All radiologists surveyed were aware of the template and 91% affirmed that standardized recommendations are useful. Patients whose reports included the template were more likely to have PCP follow-up after IAM discovery compared to those with no template (53.7% vs. 36.3%, p=0.03). After adjusting for sex, current/prior malignancy, and provider ordering the initial imaging (PCP, other outpatient provider, or ED/inpatient provider), odds of PCP follow-up remained 2.0 times higher (95%CI 1.02–3.9). Patients whose reports included the template had a shorter time to PCP follow-up (log-rank p=0.018). PCPs ultimately placed orders for biochemical testing (35.2% vs. 18.5%, p=0.01), follow-up imaging (40.7% vs. 23.3%, p=0.02), and specialist referral (22.2% vs. 4.8%, p<0.01) for a higher proportion of patients who received the template compared to those who did not.
Conclusions:
Use of a standardized template to communicate IAM recommendations was associated with improved IAM evaluation. Our template demonstrated high acceptability, but additional strategies are necessary to optimize adoption.
Keywords: Incidental findings, adrenal masses, standardized recommendations, implementation science
INTRODUCTION
There is a clear need for process improvement in the management of incidental radiographic findings of the adrenal glands and other organs. Adrenal masses are identified on 2–8% of radiographic studies performed for clinically unrelated reasons1–3 and are termed “incidental adrenal masses” (IAMs) or “adrenal incidentalomas” when they are 1 cm or larger.4 Current guidelines from major American and European specialty societies typically recommend biochemical testing and follow-up imaging after identification of an IAM to detect masses that cause hormonal syndromes or are cancerous.2,5 However, studies suggest that biochemical evaluation of IAMs is performed only 10–30% of the time, and radiographic follow-up occurs in just 20–50% of cases.6–8
Though many IAMs are clinically insignificant, 10–25% represent disease-causing lesions such as functional adenomas, pheochromocytomas, primary adrenocortical carcinoma, or metastases.5,9 The pathway from IAM detection to diagnosis involves several steps, and successful clinical follow-up depends on patient factors, provider practices, and lesion characteristics.10 There is a limited understanding of which modifiable factors drive adherence to published IAM guidelines. Prior work focused on lung nodules and other incidentalomas has shown that clinicians may be unaware of the finding or recommended follow-up11–13 and believe that recommendations in the radiology report are helpful.14 We investigated whether a standardized statement of IAM significance and concrete recommendations in the radiology report would be associated with an increase in appropriate and timely IAM evaluation. We studied the acceptability, adoption, and effectiveness of this intervention to identify factors driving or limiting its success and best focus future efforts.
METHODS
Setting and Context:
This hybrid effectiveness/implementation study was prospectively conducted in a large safety net academic medical center. In a prior study, we found that our medical center annually identifies >200 new IAMs, but that only a minority were appropriately evaluated,7 which is consistent with other centers in the U.S. and internationally.6,8,15,16 In light of this finding, we created a multidisciplinary quality improvement (QI) team to improve IAM management. Introduction of our standardized radiology report template was nested within this larger IAM QI initiative. The overall initiative also included (1) a centralized identification scheme to collate all patients with IAMs into a common database, (2) specific notifications to primary care providers (PCPs) via electronic medical record (EMR) message and email, (3) access to a web-based, concise, easy-to-use clinical evaluation algorithm, and (4) streamlined access to a multi-specialty adrenal clinic. Patients were included in this QI initiative and PCPs contacted in a prospective manner as each IAM was detected. Patients’ PCPs were identified through the EMR; patients without a PCP were called by a QI coordinator and offered a referral. If no IAM evaluation occurred by 3 months after detection, a reminder notification was sent to PCPs.
We specifically investigated the standardized radiology template’s implementation and effectiveness because of the importance of the radiology report in incidentaloma evaluation.10,16–18 Although all PCPs and patients with IAMs received the other aspects of our QI initiative, inclusion of the template was ultimately at the discretion of the reading radiologist. Because not all patients’ reports included the template, we were therefore able to estimate the association between template use and IAM evaluation independently of the other aspects of the QI initiative.
Standardized IAM Radiology Template Intervention:
Prior to the intervention, radiologists reported IAMs according to individualized practice. We hypothesized that a standardized statement of IAM significance and recommendations within the “Impression” section of the radiology report would increase appropriate and timely follow-up. The standardized text (Figure 1) included a link to a clinical management algorithm for IAMs that was developed by our QI team using a combination of guidelines from American and European endocrinology and endocrine surgery societies (Supplemental Figure 1).2,5 Radiologists could insert the template into the radiology report through dictation. In January 2019, the standardized template was introduced to radiologists via email and in-person meeting. Email reminders about template usage were issued in June, October, and December 2019.
Figure 1.
Standardized radiology template text
Inclusion Criteria:
Potentially eligible patients underwent imaging between January 1 and December 31, 2019. Eligible imaging modalities included computed tomography (CT), positron emission tomography/CT, or magnetic resonance imaging if the abdomen, chest, or spine were imaged. Radiology reports describing an IAM were provisionally identified using an automated search algorithm that examined reports for keywords related to IAMs, including “adrenal mass,” “adrenal nodule,” “adrenal lesion,” and “adrenal adenoma.” Initial automated search results were reviewed by a clinician. We excluded patients <18 years of age, those whose IAM had been detected previously, those who had imaging specifically for adrenal-related conditions, and 26 patients who were otherwise eligible but were deceased at 6 months of follow-up. A post hoc analysis including these deceased patients confirmed no change in our primary outcome. We included patients with a history of cancer. All IAMs were ≥1 cm unless a sub-centimeter mass was specifically described as concerning by the radiologist (e.g. rapid growth or characteristic of malignancy). All included patients were then prospectively followed.
Data Collection:
Data extracted from the EMR included patient demographics, comorbidities, PCP (if any), index imaging date and modality, IAM characteristics, and terminology used to describe the IAM. For outpatient imaging, our radiology department utilizes a “yellow alert” system to indicate that special attention is required; all yellow alerts were captured. Data regarding IAM follow-up care and final diagnosis (if available) was collected at standardized intervals to a minimum of six months follow-up; data collection concluded on June 30, 2020. All data extraction was performed by trained research assistants with access to clinician investigators for questions. Data were collected and managed in Research Electronic Data Capture (REDCap).19,20
Implementation Outcomes:
We measured two implementation outcomes: (1) adoption of the standardized template and (2) radiologists’ perceived acceptability of the standardized template.21 To assess adoption we calculated the percentage of imaging reports that deployed the template out of all eligible reports noting a new IAM. We also collected self-reported adoption as part of an 11-question anonymous survey of our radiologists. The survey was developed de novo by our study team, which included both perceived radiologists and non-radiologists. To assess acceptability, we surveyed radiologists about their perceived utility of and satisfaction with the template. Additional open format questions allowed radiologists to discuss suggestions or concerns. Survey data were collected between February 12 and March 28, 2020, after patient enrollment was complete.
Effectiveness Outcomes:
The primary effectiveness outcomes were (1) the proportion of patients receiving PCP follow-up before the end of study follow-up and (2) time to PCP follow-up. We compared the group of patients who received the standardized template to a contemporaneous group of patients who were eligible for the template but whose radiology reports did not include it. “Follow-up” consisted of a PCP visit after the discovery of the IAM, either with or without explicit documentation that the IAM was addressed or that the PCP was aware of the IAM. Time to PCP follow-up from the date of index imaging was also assessed. We used PCP visits as a primary outcome because chart review may miss discussion of IAM findings that was not documented and did not result in orders for further workup. The proportions of patients whose IAMs were addressed by PCPs is also reported as a secondary outcome. We defined “addressing an IAM” as any of the following: (1) documentation of history focused on symptoms or known diagnoses suggestive of adrenal masses/conditions (e.g. heart palpitations, stretch marks, history of cancer), (2) order for labs to evaluate biochemical functionality, (3) order for follow-up imaging, (4) referral to a relevant specialist. Though a visit was not required to have occurred to satisfy this outcome, these actions typically occurred in concert with a PCP visit.
Statistical Analysis:
We used descriptive statistics to report survey and EMR data. Fisher’s exact test was used to compare categorical variables, and t-tests were used to compare continuous variables. The Cochrane-Armitage test for trend was used to evaluate adoption over time. Time to follow-up after IAM diagnosis was analyzed using the Kaplan-Meier method and log-rank test. We performed a multivariable logistic regression to adjust for patient sex, current or prior diagnosis of malignancy, and initial imaging ordering provider category. Sex and malignancy were selected on the basis of differential template use and plausible association with the PCP follow-up visit outcome. We expected the initial imaging setting to influence template use and patient follow-up based on our prior findings.7 Because this analysis focused on PCPs, we split initial imaging setting into 3 groups: (1) outpatient setting, ordered by PCP, (2) outpatient setting, ordered by non-PCP provider, and (3) inpatient or emergency department provider. Jamovi version 1.2 (The Jamovi Project, Sydney, Australia), SAS version 9.4 (SAS Institute Inc., Cary, NC, USA), and Microsoft Excel version 2111 (Microsoft Corp., Redmond, WA, USA) were used for statistical analysis.
Human Subjects Protections and Reporting Guidelines:
The Boston University Medical Campus Institutional Review Board (IRB) determined this research to be exempt. This work is reported using the Standards for Quality Improvement Reporting Excellence (SQUIRE) 2.0 guidelines.22
RESULTS
Patient Characteristics
During the study period, 4,995 relevant imaging studies were performed. Through the automated keyword search of radiology reports followed by manual screening, 200 patients with new IAMs (4.0% of all studies) were prospectively included. Patient baseline characteristics are shown in Table 1. There were fewer patients with current or prior malignancy in the template group than in the non-template group (5.6% vs. 26.0%, p<0.01). Other demographics and comorbid conditions were relatively balanced.
Table 1:
Patient and selected imaging characteristics by report type
| Reports with template | Reports without template | Total reports | p-value | |
|---|---|---|---|---|
| Number of patients | 54 | 146 | 200 | |
| Sex, n (%) | ||||
| Female | 36 (66.7) | 74 (50.7) | 110 (55.0) | 0.05 |
| BMI (kg/m 2 ), mean (SD) | 29.6 (7.2) | 31.0 (8.5) | 30.6 (8.2) | 0.33 |
| Age (years), mean (SD) | 60.6 (12.9) | 59.8 (13.1) | 60.0 (13.0) | 0.73 |
| Primary language, n (%) | ||||
| English | 34 (62.9) | 101 (69.2) | 135 (67.5) | 0.40 |
| Race, n (%) | ||||
| Caucasian | 17 (31.4) | 58 (39.7) | 75 (37.5) | 0.53 |
| Black | 22 (40.7) | 52 (35.6) | 74 (37.0) | |
| Asian | 0 (0.0) | 4 (2.7) | 4 (2.0) | |
| Hispanic/Latino | 5 (9.3) | 14 (9.6) | 19 (9.5) | |
| Unknown/Not reported | 10 (18.5) | 18 (12.3) | 28 (14.0) | |
| Ethnicity, n (%) | ||||
| Not Hispanic or Latino | 39 (72.2) | 115 (78.8) | 154 (77.0) | 0.54 |
| Hispanic or Latino | 14 (25.9) | 27 (18.5) | 41 (20.5) | |
| Other/Not reported | 1 (1.9) | 4 (2.7) | 5 (2.5) | |
| Insurance status, n (%) | ||||
| Free care/none | 2 (3.7) | 1 (0.7) | 3 (1.5) | 0.21 |
| Private | 16 (29.6) | 36 (24.7) | 52 (26.0) | |
| Medicaid | 15 (27.8) | 28 (19.2) | 43 (21.5) | |
| Medicare | 11 (20.3) | 38 (26.0) | 49 (24.5) | |
| BMC Healthnet | 7 (13.0) | 35 (24.0) | 42 (21.0) | |
| Unknown | 3 (5.6) | 8 (5.5) | 11 (5.5) | |
| Current/prior malignancy, n (%) | 3 (5.6) | 38 (26.0) | 41 (20.5) | <0.01 |
| Hypertension, n, (%) | 30 (55.6) | 88 (60.3) | 118 (59.0) | 0.63 |
| Diabetes, n (%) | 13 (24.1) | 48 (32.9) | 61 (30.5) | 0.30 |
| Have an assigned PCP, n (%) | 51 (94.4) | 139 (95.2) | 190 (95.0) | 0.73 |
| PCP location, n (%) | ||||
| BMC + BMC Healthnet | 17 (31.4) | 50 (34.2) | 67 (33.5) | 0.39 |
| Boston, non-BMC | 29 (53.7) | 63 (43.2) | 92 (46.0) | |
| Non-Boston | 5 (9.3) | 26 (17.8) | 31 (15.5) | |
| No PCP | 3 (5.6) | 7 (4.8) | 10 (5.0) | |
| Pre-existing endocrinologist, n (%) | 10 (18.5) | 25 (17.1) | 35 (17.5) | 0.84 |
| Imaging setting, n (%) | ||||
| Inpatient/emergency dept. | 31 (57.4) | 73 (50.0) | 104 (52.0) | 0.43 |
| Outpatient - PCP | 3 (5.6) | 17 (11.6) | 20 (10.0) | |
| Outpatient - other | 20 (37.0 | 56 (38.4) | 76 (38.0) | |
| Indication for imaging, n (%) | ||||
| Trauma | 6 (11.1) | 9 (6.2) | 15 (7.5) | <0.01 |
| Cancer staging/surveillance | 1 (1.9) | 36 (24.7) | 37 (18.5) | |
| Abdominal/pelvic pain | 26 (48.1) | 50 (34.2) | 76 (38.0) | |
| Chest pain or shortness of breath | 7 (13.0) | 20 (13.7) | 27 (13.5) | |
| Vascular | 0 (0.0) | 1 (0.7) | 1 (0.50) | |
| Other | 14 (24.1) | 30 (20.5) | 44 (22.0) | |
| Type of imaging, n (%) | ||||
| CT abdomen | 1 (1.9) | 1 (0.7) | 2 (1.0) | 0.25 |
| CT chest | 7 (13.0) | 34 (23.3) | 41 (20.5) | |
| CT abdomen/pelvis | 29 (53.7) | 54 (37.0) | 83 (41.5) | |
| CT chest/abdomen/pelvis | 8 (14.8) | 23 (15.8) | 31 (15.5) | |
| PET-CT | 1 (1.9) | 6 (4.1) | 7 (3.5) | |
| MRI | 2 (3.7) | 3 (2.1) | 5 (2.5) | |
| Other | 6 (11.1) | 25 (17.1) | 31 (15.5) |
BMI = body mass index, BMC = Boston Medical Center, PCP = primary care provider
Most imaging and report characteristics were similar across the template and non-template groups (Table 1, Supplemental Table 1). The most common imaging indication was abdominal/pelvic pain (38.0%). Essentially none of the reports describing cancer staging or surveillance imaging used the template, contributing to a difference between groups in distribution of imaging indication. Most IAMs were discovered on CT chest, abdomen, pelvis or a combination thereof (78.5%). IAMs were most often solitary (88.5%), and <2 cm (67.5%). Neither size nor imaging modality were associated with template usage. Radiologists overwhelmingly described IAM characteristics as indeterminate or not defined (94.0%), rather than benign or concerning for malignancy. The “yellow alert” flagging system for critical findings was used similarly across template and non-template scans (15.2% vs. 19.4%, p=0.80).
Adoption and Acceptability of the Standardized Radiology Template
The adoption and acceptability survey was completed by 22 of 51 eligible radiologists for a 43.1% response rate (Table 3). All radiologists taking the survey reported awareness of the template. Most respondents (86.4%) reported using the standardized template at least once, and 72.7% reported using it for either 75% or 100% of IAMs. However, actual adoption appeared lower: the template was included in 54 of 200 reports (27.0%). There was no trend toward increasing adoption despite successive reminders (p=0.87).
Table 3.
Evaluation of IAMs by template usage group
| Report With Template (n = 54) | Report Without Template (n = 146) | P Value | |
|---|---|---|---|
|
| |||
| Follow-up recommended in discharge summary, n [inpatient subtotal (%)] | |||
| Yes | 12 [21 (57.1)] | 10 [43 (23.3)] | .01 |
| Patient had any PCP follow-up, n (%) | |||
| Yes | 29 (53.7) | 53 (36.3) | .03 |
| PCP addresses IAM*, n (%) | 30 (55.6) | 47 (32.2) | <.01 |
| Workup conducted, n (%) | |||
| By PCP | 19 (35.2) | 14 (9.6) | <.01 (global) |
| By specialist | 6 (11.1) | 25 (17.1) | <.01 (PCP vs non-PCP) |
| No workup | 29 (53.7) | 107 (73.3) | |
| Repeat imaging ordered, n (%) | |||
| Yes | 22 (40.7) | 34 (23.3) | .02 |
| PCP orders any biochemical evaluation, n (%) | 19 (35.2) | 27 (18.5) | .01 |
| PCP orders evaluation for primary aldosteronism | |||
| Full cohort, n (%) | 13 (24.1) | 20 (13.7) | .09 |
| Patients with hypertension, n [subtotal (%)] | 10 [30 (33.3)] | 15 [88 (17.0)] | .07 |
| PCP orders evaluation for pheochromocytoma, n (%) | |||
| Yes | 15 (27.8) | 20 (13.7) | .03 |
| PCP orders evaluation for autonomous cortisol excess, n (%) | |||
| Yes | 16 (29.6) | 21 (14.4) | .02 |
| Completes evaluation for autonomous cortisol excess (any orderer), n (%) | |||
| Any cortisol level testing | 23 (42.6) | 35 (24.0) | .01 (any vs no cortisol level) |
| Dexamethasone suppression test | 9 (16.7) | 13 (8.9) | |
| Night-time salivary cortisol | 0 (0.0) | 0 (0.0) | |
| Serum cortisol level obtained | 14 (25.9) | 22 (15.1) | |
| PCP orders biopsy, n (%) | |||
| Yes | 0 (0.0) | 2 (1.4) | >.99 |
| PCP referral to specialist, n (%) | |||
| None | 42 (77.8) | 139 (95.2) | <.01 |
| Endocrinologist | 7 (13.0) | 6 (4.1) | |
| Endocrine surgeon | 4 (7.4) | 0 (0.0) | |
| Endocrinologist and surgeon | 1 (1.9) | 1 (0.7) | |
| Adrenal surgery, n (%) | |||
| Yes | 0 (0.0) | 2 (1.4) | >.99 |
| Final diagnosis, n (%) | |||
| No evaluation | 31 (57.4) | 90 (61.6) | <.01 |
| Benign adenoma or non-functional | 19 (35.2) | 23 (15.8) | |
| Radiographically benign, no laboratory studies | 3 (5.6) | 8 (5.5) | |
| Adenoma: primary hyperaldosteronism | 0 (0.0) | 0 (0.0) | |
| Adenoma: Cushing’s syndrome or cortisol excess | 0 (0.0) | 0 (0.0) | |
| Pheochromocytoma | 0 (0.0) | 3 (2.1) | |
| Adrenocortical cancer | 0 (0.0) | 0 (0.0) | |
| Adrenal metastasis | 0 (0.0) | 10 (6.8) | |
| Other | 0 (0.0) | 9 (6.2) | |
| Unknown follow-up | 1 (1.9) | 3 (2.1) | |
IAM = incidental adrenal mass; PCP = primary care provider.
Addressing an IAM was defined as at least one of the following: (1) documentation of history focused on symptoms or past medical diagnoses suggestive of adrenal masses or conditions (eg, symptoms of heart palpitations, easy bruising, stretch marks; personal history of cancer, diabetes, osteoporosis, hypertension), (2) order(s) placed for laboratory studies to evaluate biochemical functionality, (3) order placed for follow-up imaging, (4) referral placed to endocrinology or endocrine surgery. These actions typically occurred during or after a PCP visit but a visit was not required to have occurred.
Acceptability of the standardized template was high. The vast majority (90.9%) of respondents agreed that providing recommendations about IAM workup was within their scope of practice. Moreover, 90.9% felt that standardized reporting language is useful. In open-ended survey questions, respondents recommended developing additional decision-support statements for incidental findings in other organ systems and confirmed that evidence-based dictation “macros” were a useful mechanism to recommend appropriate clinical follow-up.
Effectiveness of the Standardized Radiology Template in IAM Evaluation
Because of its standardized language, 100% of the patients whose radiology reports included the template had IAM-specific recommendations. By contrast, IAM-specific recommendations were made by radiologists for only 50 patients (34.2%, p<0.01) among those whose reports did not include the standardized template (Supplemental Table 1). Among patients whose IAM was discovered during an inpatient hospitalization, outpatient evaluation was recommended within the discharge summary over twice as often if the template was used (57.1% vs. 23.3%, p=0.01; Table 2). When the template was used, more patients also had follow-up with their PCPs within 6 months (53.7% vs. 36.3% without template use, p=0.03). There was a shorter time from index imaging to first PCP follow-up if the template was used (Figure 2; log-rank p=0.018). Median time to follow-up was 118 days in the template group; because fewer than 50% of patients had follow-up in the non-template group, a median could not be calculated. After adjustment for patient sex, diagnosis of malignancy, and provider type ordering the initial imaging, patients whose reports included the template had 2.0 (95% CI 1.02–3.9) times higher adjusted odds of PCP follow-up. Template use was also associated with the secondary outcome of whether PCPs explicitly addressed an IAM through chart documentation, orders, or referral (55.6% vs. 32.2% without template use, p<0.01).
Table 2:
Evaluation of IAMs by template usage group
| Report with template (n=54) | Report without template (n=146) | p-value | |
|---|---|---|---|
| Follow-up recommended in discharge summary, n [inpatient subtotal, (%)] | |||
| Yes | 12 [21, (57.1)] | 10 [43, (23.3)] | 0.01 |
| Patient had any PCP follow-up, n (%) | |||
| Yes | 29 (53.7) | 53 (36.3) | 0.03 |
| PCP addresses IAM * , n (%) | 30 (55.6) | 47 (32.2) | <0.01 |
| Workup conducted, n (%) | |||
| By PCP | 19 (35.2) | 14(9.6) | <0.01 (global) |
| By specialist | 6 (11.1) | 25 (17.1) | <0.01 (PCP |
| No workup | 29 (53.7) | 107 (73.3) | vs. non-PCP) |
| Repeat imaging ordered, n (%) | |||
| Yes | 22 (40.7) | 34 (23.3) | 0.02 |
| PCP orders any biochemical evaluation, n (%) | 19 (35.2) | 27 (18.5) | 0.01 |
| PCP orders eval. for primary aldosteronism | |||
| Full cohort, n (%) | 13 (24.1) | 20 (13.7) | 0.09 |
| Patients with hypertension, n [subtotal, (%)] | 10 [30, (33.3)] | 15 [88, (17.0)] | 0.07 |
| PCP orders eval. for pheochromocytoma, n (%) | |||
| Yes | 15 (27.8) | 20 (13.7) | 0.03 |
| PCP orders eval. for autonomous cortisol excess, n (%) | |||
| Yes | 16 (29.6) | 21 (14.4) | 0.02 |
| Completes evaluation for autonomous cortisol excess (any orderer), n (%) | |||
| ANY CORTISOL LEVEL TESTING | 23 (42.6) | 35 (24.0) | 0.01 (any vs. no cortisol |
| Dexamethasone suppression test | 9 (16.7) | 13 (8.9) | level) |
| Night-time salivary cortisol | 0 (0.0) | 0 (0.0) | |
| Serum cortisol level obtained | 14 (25.9) | 22 (15.1) | |
| PCP orders biopsy, n (%) | |||
| Yes | 0 (0.0) | 2 (1.4) | 1.00 |
| PCP referral to specialist, n (%) | |||
| None | 42 (77.8) | 139 (95.2) | <0.01 |
| Endocrinologist | 7(13.0) | 6(4.1) | |
| Endocrine Surgeon | 4 (7.4) | 0 (0.0) | |
| Endocrinologist and surgeon | 1 (1.9) | 1 (0.7) | |
| Adrenal surgery, n (%) | |||
| Yes | 0 (0.0) | 2 (1.4) | 1.00 |
| Final diagnosis, n (%) | |||
| No evaluation | 31 (57.4) | 90 (61.6) | <0.01 |
| Benign adenoma/non-functional | 19 (35.2) | 23 (15.8) | |
| Radiographically benign, no labs | 3(5.6) | 8(5.5) | |
| Adenoma: primary hyperaldosteronism | 0 (0.0) | 0 (0.0) | |
| Adenoma: Cushing’s syndrome/cortisol excess | 0 (0.0) | 0 (0.0) | |
| Pheochromocytoma | 0 (0.0) | 3 (2.1) | |
| Adrenocortical cancer (ACC) | 0 (0.0) | 0 (0.0) | |
| Adrenal metastasis | 0 (0.0) | 10(6.8) | |
| Other | 0 (0.0) | 9(6.2) | |
| Unknown follow-up | 1 (1.9) | 3 (2.1) |
Addressing an IAM was defined as at least one of the following: (1) documentation of history focused on symptoms or past medical diagnoses suggestive of adrenal masses/conditions (e.g. symptoms of heart palpitations, easy bruising, stretch marks; personal history of cancer, diabetes, osteoporosis, hypertension), (2) order(s) placed for labs to evaluate biochemical functionality, (3) order placed for follow-up imaging, (4) referral placed to endocrinology or endocrine surgery. These actions typically occurred during or after a PCP visit but a visit was not required to have occurred. IAM = incidental adrenal mass, PCP = primary care provider.
Figure 2.
Time to PCP follow-up from IAM discovery for patients with and without the template
When the standardized template was used, there were differences in what type of provider, if any, initiated IAM workup (p<0.01 for global comparison). In the template group, PCPs initiated IAM workup prior to or concomitant with specialist referral for 35.2% of patients, compared to 9.6% in the non-template group. Template group patients were more likely to have PCPs order follow-up imaging (40.7% vs. 23.3%, p=0.02) as well as place orders for biochemical testing for autonomous cortisol excess, pheochromocytoma, and/or primary aldosteronism (35.2% vs. 18.5%, p=0.01). There were also differences in specialist referral between groups; 22.2% in the template group received a referral to endocrinology or endocrine surgery, compared to 4.8% in the non-template group (p<0.01).
There were significant differences in the distribution of final diagnoses across groups (Table 2; p<0.01). More than double the proportion of patients in the template group was ultimately diagnosed with a benign, non-functional adenoma (35.2% in the template vs. 15.8% in the non-template group). Three pheochromocytomas were detected, all in the non-template group. No primary adrenocortical carcinomas were detected. There were 10 diagnoses of adrenal metastasis, all in the non-template group. A substantial portion of patients did not undergo an evaluation that allowed for determination of final diagnosis, including 57.4% in the template group and 61.6% in the non-template group.
DISCUSSION
This study investigated the adoption, acceptability, and effectiveness of a standardized radiology template designed to improve IAM evaluation. Although 73% of radiologists reported using the template all or most of the time, actual adoption was 27% in its first year of implementation. Despite suboptimal adoption, use of our concise, standardized template within the “Impression” section of the radiology report was associated with a nearly 50% increase in PCP follow-up after IAM detection. Time to follow-up was significantly shorter if the template was used, and we also found improvements in radiographic testing and biochemical assessments. Overall, these findings suggest that efficient and timely communication that includes specific next steps may lead to improved evaluation of IAMs.
The success of our standardized template adds to a small body of work showing that standardized reporting templates are an effective and evidence-based intervention.16,17 Like many studies seeking to improve incidentaloma management, we targeted radiologists’ behavior.23 Information conveyed in the radiology report is a crucial factor in whether patients receive follow-up, and use of specific, proscriptive language has been shown to be beneficial.7,24 A recent retrospective study also found that radiology reports with specific diagnostic concerns or recommendations were associated with increased follow-up, including in a multivariable model adjusted for mass size, patient age, and managing clinical team.10 Our study was similar to a study by Eldeiry et al. which incorporated standardized radiology report language that linked to a clinical IAM evaluation algorithm, though they excluded patients with known extra-adrenal malignancies.18 Of nearly 900 patients, 20% had a clinical management algorithm directly incorporated into or linked to from the Impression section of their radiology report. Eldeiry et al. found no significant increase in follow-up imaging overall but did see an increase in biochemical testing. By contrast, we found significant increases in both imaging and biochemical testing with our template.
Although we found that template usage was associated with significant increases in PCP visits and several other IAM evaluation outcomes, only 2 patients underwent adrenal surgery by the end of follow-up. First, though it is important to identify patients with clinically relevant IAMs who would benefit from resection, 75% or more of IAMs do not represent biochemically functional lesions or malignancy, and not all functional lesions necessitate short-term resection.8,25 In addition, our cohort size was relatively small, the outpatient evaluation of IAMs often takes place over several months or more, and the COVID-19 pandemic disrupted non-emergent care during a key portion of our follow-up period. We nevertheless hypothesize that our standardized radiology template would ultimately be associated with an increase in appropriate adrenalectomies, given additional follow-up time and enrollment.
As noted above, adoption of the standardized template occurred in only 27% of eligible reports. Though we used multiple, temporally-spaced email reminders to encourage template use, we did not detect increased usage over time. However, generating behavior change remains challenging and our results are not unique. In Grimshaw et al.’s seminal review of knowledge translation, studies focused on professional behavior change showed generally similar and modest improvements regardless of specific implementation strategy.26 Notably, our survey results revealed a disconnect between radiologists’ perception of their template use and actual adoption. This result may reflect social desirability bias or selection bias; however, it also suggests important opportunities for improvement. If radiologists view the intervention as acceptable, refining implementation may lead to better outcomes. Other implementation strategies could include periodic audit and feedback for individual radiologists, reporting benchmarks set by high-adopters, or financial incentives. Periodic audits for IAM reporting have been explored previously, though initial adoption of an IAM evaluation and referral pathway remained low.27 Other studies have described gradually increasing adoption over a multi-year time period, suggesting adoption might continue to improve beyond our study timeline.28 Finally, improving feasibility, such as with automatic prompts that employ a natural language processing approach, might also enhance template adoption.
Our study has several limitations. It was conducted at a single center and transportability may therefore be limited. Additionally, some patients had as little as 6 months of study follow-up and may have undergone evaluation beyond that time period, especially given the impact of the COVID-19 pandemic in 2020. Our primary outcome required only a PCP follow-up visit, regardless of whether any explicit documentation about the IAM occurred. However, we also found consistently positive associations between template usage and IAM workup across several secondary outcomes. Importantly, the template was not utilized at random, which could introduce bias. We noted statistically significant or nearly significant differences in template usage on the basis of sex, current or prior malignancy, and imaging indication. Imaging indication was itself most notable for differences in the proportion of patients undergoing cancer staging or surveillance. We hypothesize that template usage may have been especially less likely among patients with a cancer history, as any newly discovered adrenal masses may have been interpreted as suspicious for metastasis rather than truly incidental. Though our cohort size prevented adjustment for all potentially relevant confounding variables, we employed a logistic regression analysis to adjust for three important potential confounders and found that standardized template use remained significantly associated with PCP follow-up. Our template was introduced in the context of an ongoing QI initiative; however, the other components were implemented for all patients in this study and should not bias our results. Finally, the acceptability component of our study was limited by the small number of survey respondents. These limitations notwithstanding, to our knowledge, this work represents the only prospective examination of the implementation, physician perceptions, and effectiveness of a standardized reporting template for IAMs.
If our effectiveness results are replicated, future work should examine ways to facilitate adoption of standardized reporting templates for incidental radiographic findings. Development of successful strategies for IAM evaluation has implications for incidentalomas in other organ systems, and given their high frequency,29,30 appropriate incidentaloma management continues to represent an important healthcare quality and patient safety goal.
Supplementary Material
Supplemental Figure 1. IAM clinical managementalgorithm
This clinical algorithm was provided to patients’ primary care providers at the time of IAM discovery. Both the basic and detailed versions of this algorithm were publicly available on the institution’s website. IAM = incidental adrenal mass.
Supplemental Table 1: Additional imaging report characteristics
IV = intravenous, CT = computed tomography, PET = positron emission tomography, MRI = magnetic resonance imaging
TAKE HOME POINTS.
Use of a concise standardized reporting template within the “Impression” section of the radiology report was associated with a nearly 50% increase in PCP follow-up after IAM detection
More patients in the template group underwent radiographic workup (41% vs. 23% in the non-template group) and biochemical testing (35% vs. 18%)
91% of radiologists agreed that a standardized IAM radiology reporting template was useful
Nearly three-fourths of radiologists surveyed reported using the template 75% or 100% of the time, but actual adoption occurred in only 27% of opportunities
Even with imperfect adoption, clear and specific radiology report recommendations can lead clinicians to initiate appropriate incidental adrenal mass evaluation
SUMMARY SENTENCE.
Use of our concise, standardized template within the “Impression” section of the radiology report was associated with a nearly 50% increase in PCP follow-up after IAM detection.
Funding:
This project was supported by a Boston University grant supporting REDCap data capture, Clinical & Translational Science Institute #1UL1TR001430. Alison Woods is supported by a grant from the National Institutes of Health (National Cancer Institute award #T32CA126607). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH and supporting institutions.
Abbreviations
- CT
computed tomography
- EMR
electronic medical record
- IAM
incidental adrenal mass
- IRB
institutional review board
- PCP
primary care provider
- QI
quality improvement
- SQUIRE
Standards for Quality Improvement Reporting Excellence
Footnotes
Conflicts of Interest: The authors declare no conflicts of interest.
Approval for >7 Authors: Permission requested via email by Alison Woods on 7/12/22. Notified of approval on behalf of the editor via email by Kelly Church on 7/14/22.
Leadership Roles
Avneesh Gupta – Abdominal Imaging Fellowship Director, Department of Radiology Information Technology Officer (Boston Medical Center)
Allan Walkey – Co-Director of the Evans Center for Implementation and Improvement Sciences (Boston University)
Mari-Lynn Drainoni – Co-Director of the Evans Center for Implementation and Improvement Sciences (Boston University)
David McAneny – Chief Medical Officer, Senior Vice President of Medical Affairs, Associate Dean for Clinical Affairs (Boston Medical Center)
F. Thurston Drake – Chief of Endocrine Surgery (Boston Medical Center)
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Data Statement:
The authors declare that they had full access to all of the data in this study and the author take complete responsibility for the integrity of the data and the accuracy of the data analysis.
REFERENCES
- 1.Bovio S, Cataldi A, Reimondo G, et al. Prevalence of adrenal incidentaloma in a contemporary computerized tomography series. J Endocrinol Invest. 2006;29(4):298–302. doi: 10.1007/BF03344099 [DOI] [PubMed] [Google Scholar]
- 2.Zeiger MA, Thompson GB, Duh QY, et al. The American Association of Clinical Endocrinologists and American Association of Endocrine Surgeons medical guidelines for the management of adrenal incidentalomas. Endocr Pract Off J Am Coll Endocrinol Am Assoc Clin Endocrinol. 2009;15 Suppl 1:1–20. doi: 10.4158/EP.15.S1.1 [DOI] [PubMed] [Google Scholar]
- 3.Barzon L, Sonino N, Fallo F, Palu G, Boscaro M. Prevalence and natural history of adrenal incidentalomas. Eur J Endocrinol. 2003;149(4):273–285. doi: 10.1530/eje.0.1490273 [DOI] [PubMed] [Google Scholar]
- 4.Young WF MANAGEMENT APPROACHES TO ADRENAL INCIDENTALOMAS: A View from Rochester, Minnesota. Endocrinol Metab Clin North Am. 2000;29(1):159–185. doi: 10.1016/S08898529-(05)70122-5 [DOI] [PubMed] [Google Scholar]
- 5.Fassnacht M, Arlt W, Bancos I, et al. Management of adrenal incidentalomas: European Society of Endocrinology Clinical Practice Guideline in collaboration with the European Network for the Study of Adrenal Tumors. Eur J Endocrinol. 2016;175(2):G1–G34. doi: 10.1530/EJE-16-0467 [DOI] [PubMed] [Google Scholar]
- 6.Sahni P, Trivedi A, Omer A, Trivedi N. Adrenal incidentalomas: are they being worked up appropriately? J Community Hosp Intern Med Perspect. 2016;6(5):32913. doi: 10.3402/jchimp.v6.32913 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Feeney T, Talutis S, Janeway M, et al. Evaluation of incidental adrenal masses at a tertiary referral and trauma center. Surgery. 2020;167(5):868–875. doi: 10.1016/j.surg.2019.07.034 [DOI] [PubMed] [Google Scholar]
- 8.Feeney T, Madiedo A, Knapp PE, Gupta A, McAneny D, Drake FT. Incidental Adrenal Masses: Adherence to Guidelines and Methods to Improve Initial Follow-Up: A Systematic Review. J Surg Res. 2022;269:18–27. doi: 10.1016/j.jss.2021.07.041 [DOI] [PubMed] [Google Scholar]
- 9.Sherlock M, Scarsbrook A, Abbas A, et al. Adrenal Incidentaloma. Endocr Rev. 2020;41(6):775–820. doi: 10.1210/endrev/bnaa008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Maher DI, Williams E, Grodski S, Serpell JW, Lee JC. Adrenal incidentaloma follow-up is influenced by patient, radiologic, and medical provider factors: A review of 804 cases. Surgery. 2018;164(6):1360–1365. doi: 10.1016/j.surg.2018.07.011 [DOI] [PubMed] [Google Scholar]
- 11.Golden SE, Wiener RS, Sullivan D, Ganzini L, Slatore CG. Primary Care Providers and a System Problem: A Qualitative Study of Clinicians Caring for Patients With Incidental Pulmonary Nodules. Chest. 2015;148(6):1422–1429. doi: 10.1378/chest.14-2938 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Blagev DP, Lloyd JF, Conner K, et al. Follow-up of Incidental Pulmonary Nodules and the Radiology Report. J Am Coll Radiol. 2014;11(4):378–383. doi: 10.1016/j.jacr.2013.08.003 [DOI] [PubMed] [Google Scholar]
- 13.Bates R, Plooster C, Croghan I, Schroeder D, McCoy C. Incidental Pulmonary Nodules Reported on CT Abdominal Imaging: Frequency and Factors Affecting Inclusion in the Hospital Discharge Summary. J Hosp Med. 2017;12(6):454–457. doi: 10.12788/jhm.2757 [DOI] [PubMed] [Google Scholar]
- 14.Zafar HM, Bugos EK, Langlotz CP, Frasso R. “Chasing a Ghost”: Factors that Influence Primary Care Physicians to Follow Up on Incidental Imaging Findings. Radiology. 2016;281(2):567–573. doi: 10.1148/radiol.2016152188 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wickramarachchi BN, Meyer-Rochow GY, McAnulty K, Conaglen JV, Elston MS. Adherence to adrenal incidentaloma guidelines is influenced by radiology report recommendations. ANZ J Surg. 2016;86(6):483–486. doi: 10.1111/ans.12799 [DOI] [PubMed] [Google Scholar]
- 16.Corwin MT, Arora A, Loehfelm TW, Fananapazir G, Campbell MJ. Adherence to guidelines for hormonal evaluation in patients with incidentally detected adrenal nodules: effects of radiology report wording and standardized reporting. Abdom Radiol N Y. 2020;45(9):2910–2915. doi: 10.1007/s00261-020-02517-3 [DOI] [PubMed] [Google Scholar]
- 17.Almeida RR, Bizzo BC, Singh R, Andriole KP, Alkasab TK. Computer-assisted Reporting and Decision Support Increases Compliance with Follow-up Imaging and Hormonal Screening of Adrenal Incidentalomas. Acad Radiol. Published online February 12, 2021:S1076–6332(21)00028–3. doi: 10.1016/j.acra.2021.01.019 [DOI] [PubMed] [Google Scholar]
- 18.Eldeiry LS, Alfisher MM, Callahan CF, Hanna NN, Garbe JR The impact of an adrenal incidentaloma algorithm on the evaluation of adrenal nodules. J Clin Transl Endocrinol. 2018;13:39–45. doi: 10.1016/j.jcte.2018.07.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377–381. doi: 10.1016/j.jbi.2008.08.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Harris PA, Taylor R, Minor BL, et al. The REDCap consortium: Building an international community of software platform partners. J Biomed Inform. 2019;95:103208. doi: 10.1016/j.jbi.2019.103208 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Proctor E, Silmere H, Raghavan R, et al. Outcomes for Implementation Research: Conceptual Distinctions, Measurement Challenges, and Research Agenda. Adm Policy Ment Health. 2011;38(2):65–76. doi: 10.1007/s10488-010-0319-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.SQUIRE | SQUIRE 2.0 Guidelines. Accessed August 29, 2021. http://www.squire-statement.org/index.cfm?fuseaction=page.viewpage&pageid=471
- 23.Crable EL, Feeney T, Harvey J, et al. Management Strategies to Promote Follow-Up Care for Incidental Findings: A Scoping Review. J Am Coll Radiol JACR. 2021;18(4):566–579. doi: 10.1016/j.jacr.2020.11.006 [DOI] [PubMed] [Google Scholar]
- 24.de Haan RR, Schreuder MJ, Pons E, Visser JJ. Adrenal Incidentaloma and Adherence to International Guidelines for Workup Based on a Retrospective Review of the Type of Language Used in the Radiology Report. J Am Coll Radiol JACR. 2019;16(1):50–55. doi: 10.1016/j.jacr.2018.08.011 [DOI] [PubMed] [Google Scholar]
- 25.Kahramangil B, Kose E, Remer EM, et al. A Modern Assessment of Cancer Risk in Adrenal Incidentalomas: Analysis of 2219 Patients. Ann Surg. 2022;275(1):e238–e244. doi: 10.1097/SLA.0000000000004048 [DOI] [PubMed] [Google Scholar]
- 26.Grimshaw JM, Eccles MP, Lavis JN, Hill SJ, Squires JE. Knowledge translation of research findings. Implement Sci. 2012;7(1):50. doi: 10.1186/1748-5908-7-50 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Paterson F, Theodoraki A, Amajuoyi A, Bouloux PM, Maclachlan J, Khoo B. Radiology reporting of adrenal incidentalomas – who requires further testing? Clin Med. 2014;14(1):16–21. 392 doi: 10.7861/clinmedicine.14-1-16 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.McDonald JS, Koo CW, White D, Hartman TE, Bender CE, Sykes AMG. Addition of the Fleischner Society Guidelines to Chest CT Examination Interpretive Reports Improves Adherence to Recommended Follow-up Care for Incidental Pulmonary Nodules. Acad Radiol. 2017;24(3):337–344.doi: 10.1016/j.acra.2016.08.026 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.O’Sullivan JW, Muntinga T, Grigg S, Ioannidis JPA. Prevalence and outcomes of incidental imaging findings: umbrella review. BMJ. 2018;361:k2387. doi: 10.1136/bmj.k2387 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Hitzeman N, Cotton E. Incidentalomas: Initial Management. Am Fam Physician. 2014;90(11):784–789. [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Figure 1. IAM clinical managementalgorithm
This clinical algorithm was provided to patients’ primary care providers at the time of IAM discovery. Both the basic and detailed versions of this algorithm were publicly available on the institution’s website. IAM = incidental adrenal mass.
Supplemental Table 1: Additional imaging report characteristics
IV = intravenous, CT = computed tomography, PET = positron emission tomography, MRI = magnetic resonance imaging
Data Availability Statement
The authors declare that they had full access to all of the data in this study and the author take complete responsibility for the integrity of the data and the accuracy of the data analysis.