C

Patient Anxiety Before andImmediately After Imaging-Guided

Breast Biopsy Procedures: Impact ofRadiologist-Patient Communication

Lauren S. Miller, MDa, Rebecca A. Shelby, PhDb, Melissa Hayes Balmadrid, MDc,Sora Yoon, MDd, Jay A. Baker, MDd, Liz Wildermanne, Mary Scott Soo, MDd

Purpose: The aim of this study was to evaluate patient anxiety and its association with perceived radiologist-patient communication in the setting of imaging-guided breast biopsy.

Methods: After informed consent was obtained, 138 women recommended for imaging-guided breastprocedures completed questionnaires immediately before and after biopsies, measuring state anxiety using theState-Trait Anxiety Inventory (range, 20-80). Before biopsies, women also completed questionnaires regardingtheir perceived communication with the radiologists recommending the procedures (modified Questionnaireon the Quality of Physician-Patient Interaction), demographic characteristics, and medical history; immedi-ately after the biopsies, they completed a measure of perceived communication with the radiologists performingthe biopsies. Experience levels (eg, attending radiologist, fellow) of the radiologists recommending and per-forming the biopsies were recorded. Data were analyzed using paired and independent t tests, one-way analysisof variance, Pearson’s correlations, and multiple linear regression analyses.

Results: Average prebiopsy anxiety was 44.5 � 12.4 (range, 20-77) on a scale ranging from 20 to 80 points.Perceived communication with radiologists recommending biopsies averaged 52.4 � 11.5 (range, 18-65).Better communication with radiologists recommending biopsies was significantly associated with lower levels ofprebiopsy anxiety (r � �0.22, P � .01). After the biopsies, women’s anxiety significantly decreased (paired t ��7.32, P � .001). Better communication with radiologists performing biopsies (mean, 57.8 � 8.4; range,32-65) was associated with lower postbiopsy anxiety after accounting for patients’ baseline anxiety levels (� ��0.17, P � .04). White women reported higher prebiopsy and postbiopsy anxiety; nonwhite women reportedpoorer communication with recommending radiologists.

Conclusions: Patients’ perceptions of better communication with radiologists were associated with lowerlevels of anxiety before and after biopsies. These results have implications for radiologist training and adherenceto mammographic screening.

Key Words: Breast needle biopsy, anxiety, communication

J Am Coll Radiol 2013;10:423-431. Copyright © 2013 American College of Radiology

mia

INTRODUCTIONAnxiety in women undergoing breast imaging studies canbe experienced at multiple time points (eg, during mam-

aRiverside Radiology and Interventional Associates, Columbus, Ohio.bDepartment of Psychiatry and Behavioral Sciences, Duke University Schoolof Medicine, Durham, North Carolina.cSeattle, Washington.dDepartment of Radiology, Duke University Medical Center, Durham, North

arolina.eDepartment of Surgery, Duke University Medical Center, Durham, North Carolina.

Corresponding author and reprints: Mary Scott Soo, MD, Duke University

Medical Center, Department of Radiology, Box 3808, Room 24244B, 2nd

© 2013 American College of Radiology0091-2182/13/$36.00 ● http://dx.doi.org/10.1016/j.jacr.2012.11.005

ographic screening, when recalled for additional imag-ng, during diagnostic workup and biopsy, whilewaiting results). Previous studies have shown that

Floor Red Zone, Duke South Clinic, Trent Drive, Durham, NC 27710;e-mail: soo00002@mc.duke.edu.

This study was supported in part by pilot grant 12111 from the JohnTempleton Foundation through the Center for Spirituality, Theology andHealth at Duke University Medical Center.

Dr Baker is a consultant for and receives payment for lectures fromSiemens Medical Solutions USA, Inc (Mountain View, California). Healso reviews medicolegal cases for Dickie, McCarney & Chilcote (Pitts-

burgh, Pennsylvania).

423

ntposeac

424 Journal of the American College of Radiology/Vol. 10 No. 6 June 2013

women undergoing core-needle breast biopsy report highlevels of anxiety before and during the procedure and upto 3 years after receiving benign results [1-11]. Althoughfor some women, recall and biopsy can improve compli-ance [12-14], in other women, persistent anxiety caninterfere with motivation to adhere to follow-up recom-mendations [7,15-19]. Long-term high anxiety levels arealso associated with other negative health effects, includ-ing increased risk for heart disease, immune system sup-pression, and lower health-related quality of life [20-22].

Anxiety associated with abnormal mammographic re-sults and subsequent breast biopsy also has importantpublic health policy implications. Anxiety was cited asone of the harms of mammographic screening that influ-enced the 2009 US Preventive Services Task Force’s rec-ommendation to limit mammographic screening [23].The predominant reaction from the medical communityand general public to this change in screening recom-mendations prompted the medical community’s reeval-uation of the data, so that now, most major medicalorganizations still support the previously establishedscreening guidelines. However, the task force’s view thatsignificant anxiety can be experienced by women withabnormal biopsy results remains a factor to be addressed.It seems feasible that anxiety can be reduced for womenundergoing breast imaging without affecting screeningintervals or compromising the success of screening pro-grams, while potentially improving adherence to fol-low-up recommendations and reducing the negativeimpact of anxiety on other health-related issues.

Previous studies have examined patient demographicand medical characteristics associated with patient anxi-ety in the setting of abnormal mammographic findingsand percutaneous breast biopsy [1-9,24]; however, dataexamining the relationship between radiologist-patientcommunication and anxiety are lacking. Studies have,however, described important implications for both phy-sicians and patients related to physician-patient commu-nication. Not only has physician-patient communicationbeen linked to patient satisfaction [25,26], but poorcommunication has been associated with more frequentmalpractice litigation [27-30]. Outside of breast imag-ing, physician-patient communication has been shown toinfluence cancer treatment, patient outcomes [26,31-34],and psychological symptoms (eg, anxiety, depression)[4,10,25]. It follows that radiologist-patient communi-cation could have an impact on patients’ anxiety andhealth-related issues, given the challenging nature of dis-cussions around need for breast biopsy and potentialimplications of the results. The ACR and the Society ofBreast Imaging have listed communication issues andproficiency in interacting with patients regarding how torecommend a biopsy, explain a cancer diagnosis, anddevelop sensitivity to patients’ emotions during this pro-cess within their educational curriculum for residency

and fellowship training in breast imaging [35]. Given the

eed to address issues of patient anxiety resulting fromhe biopsy recommendation and surrounding the biopsyrocedure, and to better understand the impact of radi-logist-patient communication during this process, weought to (1) evaluate patient anxiety in this setting, (2)xamine patient characteristics associated with anxiety,nd (3) examine whether perceived radiologist-patientommunication is associated with patient anxiety.

METHODS

ParticipantsFrom August 2010 through February 2011, 207 womenundergoing ultrasound-guided or stereotactic-guidedcore-needle breast biopsies or ultrasound-guided diag-nostic cyst aspiration were invited to participate in thisprospective study on the day of their procedures. Womenmet the following inclusion criteria: (1) �21 years of age,(2) presented to our breast biopsy center for percutane-ous imaging-guided diagnostic procedure, (3) were ableto speak and read English, and (4) were able to providewritten informed consent. Women who had undergoneimaging-guided breast biopsy in the previous 6 monthswere excluded. The study was performed under an insti-tutional review board-approved protocol and wasHIPAA compliant. Data presented in this paper werecollected as part of a larger longitudinal study of adher-ence to recommended care that follows participants for3 years.

One hundred fifty-two women completed informedconsent (participation rate, 73%), and these patients re-ceived parking passes for their participation. Fourteenparticipants were excluded from data analyses; biopsyprocedures were not completed in 4 women, 1 womanwas found to have cognitive impairment interfering withquestionnaire completion, and 9 women did not com-plete the study measures. A total of 138 women wereincluded in this sample.

ProceduresStudy participants completed written questionnaires inthe biopsy clinic immediately before the biopsies (prebi-opsy assessment) and immediately afterward (postbiopsyassessment). Prebiopsy questionnaires included measuresof anxiety and communication with the radiologist whorecommended the biopsy. Immediately after their biop-sies and before leaving the biopsy clinic, women againcompleted measures of anxiety and assessed communica-tion with the radiologists who performed the biopsies.Participants also completed questionnaires assessing so-ciodemographic and medical characteristics. The type ofpercutaneous imaging-guided procedure (stereotactic-guided core biopsy, ultrasound-guided core biopsy, orultrasound-guided diagnostic aspiration) was recorded,along with who recommended the biopsy, the experiencelevel of the radiologist(s) performing the biopsy (attend-

ing radiologist alone, fellow alone, attending radiologist

t

es

Pciiqnmii“wseheeintss0

gpflatIaitvcbpof

Dnwsleca

Miller et al/Patient Anxiety and Breast Biopsy 425

with fellow, or attending radiologist with resident), andwhether the radiologist recommending and performingthe biopsy was the same individual.

Before study entry, all women received a recommen-dation for biopsy either from 1 of 7 breast imaging at-tending radiologists or 1 of 3 breast imaging fellows inour diagnostic breast imaging clinic or from a physicianor staff member outside our clinic. For patients undergo-ing diagnostic evaluations in our clinic who had suspi-cious imaging findings (BI-RADS® final assessmentcategory 4 or 5), our radiologists consulted with them inperson immediately after the examination, informingthem of the biopsy recommendation. At that time, abiopsy appointment was scheduled for the patient, thebiopsy procedure was described briefly, a pamphlet withinformation about the biopsy was provided to the pa-tient, and the patient had an opportunity to ask ques-tions. The extent to which the radiologist described thelikelihood of malignancy varied among these interac-tions, depending on the radiologist’s discernment of thepatients’ desire and ability to receive the information atthat time.

If the patient had undergone diagnostic imaging andbiopsy recommendation at an outside facility and pre-sented to our institution for biopsy or consultation witha surgeon, the outside radiologist, outside physician, orour institution’s referring physician discussed the biopsyrecommendation with the patient apart from our clinic.Ultrasound-guided and prone-table stereotactic-guidedlarge-core biopsies and ultrasound-guided diagnostic as-pirations were performed in standard fashion. Histologicresults from the biopsies were not available to the patientsat the time of the biopsies or when completing the post-biopsy questionnaires; results were given by phone to thepatients when the pathology reports became availableover the next several days.

Measures

Anxiety. The State Anxiety Scale of the State-Trait Anx-iety Inventory was used to assess prebiopsy and postbi-opsy anxiety [36]. The State Anxiety Scale is a 20-itemself-report scale that assesses how much anxiety a personfeels at that moment (eg, “I feel nervous”). The 4-pointresponse format assesses the intensity of feelings, withchoices ranging from 1 (“not at all”) to 4 (“very much”).Items are summed to create a total score (possible range,20-80), with higher scores indicating higher anxiety.This scale had high reliability in this sample (Cronbach’s� � 0.91 to 0.92). Scale reliability (ie, internal consis-ency) was measured using Cronbach’s �, which in-

creases as the intercorrelations among scale itemsincrease. For Cronbach’s �, a value �0.90 is consideredxcellent, �0.80 is considered good, and �0.70 is con-

idered acceptable [37].

erceived Communication. Patients’ perceptions ofommunication with the radiologist were measured us-ng a modified version of the Questionnaire on the Qual-ty of Physician-Patient Interaction [38]. The 14-itemuestionnaire was modified by dropping 1 item that wasot pertinent to the biopsy scenario (“The physician gavee a thorough examination”) and by revising the word-

ng of the remaining 13 items to refer to “the radiologist”nstead of using the more general term “physician” (eg,The radiologist’s/health professional’s explanationsere easy to understand”; “The radiologist/health profes-

ional spoke to me in detail about the risks and sideffects of the proposed procedure”; “The radiologist/ealth professional did all he/she could to put me atase”; “The radiologist/health professional gave menough time to ask questions”). Participants rated eachtem using a 5-point response scale ranging from 1 (“I doot agree”) to 5 (“I strongly agree”). Items were summedo create a total score (possible range, 13-65), with highercores indicating better perceived communication. Thiscale had high reliability in this sample (Cronbach’s � �.93 to 0.95).To assess perceived communication with the radiolo-

ist who recommended the biopsy, before the biopsy, theatient filled out a questionnaire requesting that she re-ect on her perceived communication during the inter-ction with that radiologist, which may have occurred onhe same day as the biopsy or in the preceding days.mmediately after the biopsy, a similar questionnaire wasdministered, requesting that the patient reflect on hernteraction with the radiologist who had just performedhe biopsy. The preprocedural and postproceduralersions of the Questionnaire on the Quality of Physi-ian-Patient Interaction were similar, except that the pre-iopsy questionnaire referred to the radiologist or healthrofessional recommending the biopsy, and the postbi-psy questionnaire referred to the radiologist who per-ormed the biopsy.

emographic and Medical Information. A question-aire assessing demographic and medical characteristicsas completed by participants on the day of their biop-

ies (either before or after the procedures as time al-owed). Demographic information included age, race,ducation, and marital status. Medical information in-luded history of depression or anxiety, cancer history,nd history of breast biopsy or surgery.

Statistical AnalysisPaired t tests were conducted to examine changes inanxiety from before to after the biopsy. Bivariate analyses(independent t tests, one-way analysis of variance, orPearson’s correlations as appropriate) were conducted forprebiopsy and postbiopsy measures of anxiety and com-munication to examine associations between these studyvariables and participant characteristics. We conducted

multiple linear regression analyses to test whether prebi-

te

iaifm(oadc4p(g

426 Journal of the American College of Radiology/Vol. 10 No. 6 June 2013

opsy and postbiopsy anxiety was associated with radiol-ogist-patient communication and radiologist experiencelevel. Separate regression models were conducted for pre-biopsy and postbiopsy measures of anxiety. On the basisof bivariate analyses (P � .05), we included age, educa-tion, race, history of anxiety or depression, and previousbreast biopsy in each prebiopsy regression model as con-trol variables. For each postbiopsy regression model, pre-biopsy anxiety, education, race, and type of biopsy wereincluded as control variables. Multilevel linear mixedmodels were also performed to compare patient anxietyscores and communication scores across individual radi-ologists. These models accounted for the nesting ofscores within patients (prebiopsy and postbiopsy ratings)and the nesting of patients within radiologists (ie, eachradiologist treated multiple patients).

RESULTS

Participant CharacteristicsTable 1 displays participant characteristics and descrip-ive statistics for measures of communication and anxi-ty. The average patient age was 52.1 � 12.9 years, and

73.2% (n � 101) of the patients were white. The major-ty of women (91.3%) underwent diagnostic breast im-ging evaluation at our clinic, with attending breastmaging radiologists (69.6% [n � 96]) or breast imagingellows (21.7% [n � 30]) discussing the biopsy recom-endations with the patient. The remaining women

8.7% [n � 12]) presented for biopsies after physiciansr medical professionals from outside radiology or fromn outside facility recommended the biopsies. One hun-red two (73.9%) ultrasound-guided procedures (89ore-needle biopsies and 13 diagnostic aspirations) and1 (29.7%) stereotactic-guided core needle biopsies wereerformed by either an attending radiologist alone38.4%), a fellow alone (8.7%), or an attending radiolo-ist with a fellow or resident (52.9%).

Descriptive Statistics and Comparison of AnxietyFrom Before to After BiopsyThe average prebiopsy State Anxiety Scale score was 44.5 �12.42 (range, 21-77); scores of 39 and 40 are often usedto indicate elevated levels of anxiety [36]. This decreasedafter the biopsies to an average of 37.2 � 6.3 (range,20-71). Perceived communication scores for the recom-mending radiologists were on average 52.4 � 11.48(range, 18-65) and for the radiologists performing thebiopsy 57.8 � 8.43 (range, 32-65).

Paired t tests were conducted to examine whether levelsof anxiety changed from before to after the biopsies. Anxietyscores significantly decreased, with an average decrease of7.25 � 11.64 points (t[137] � �7.32, P � .001).

Associations With Patient CharacteristicsAssociations between participant characteristics and pre-

biopsy (Table 2) and postbiopsy (Table 3) measures of

anxiety and communication were examined. Comparedwith white participants, nonwhite participants reportedpoorer prebiopsy perceived communication (mean,53.71 vs 48.99; t[136] � �2.17; P � .03) and loweranxiety (mean, 45.98 vs 40.43; t[136] � �2.37; P �.02). At postbiopsy, compared with white participants,nonwhite participants reported lower postbiopsy anxiety(mean, 39.01 vs 32.39; t[136] � �3.05; P � .003). Racewas not associated with perceived radiologist communi-

Table 1. Sample characteristics (n � 138)Variable Value

Age (y) 52.12 � 12.93 (21-87)Education (y) 15.61 � 3.26 (7-25)Married 58.0% (80)Race

White 73.2% (101)Black 23.9% (33)Asian 2.2% (3)American Indian 0.7% (1)

History of anxiety or depression 21.7% (30)Personal history of cancer 26.1% (36)Personal history of breast cancer 18.8% (26)First-degree relative with breast

cancer21.0% (29)

Previous breast biopsy 46.4% (64)Previous breast surgery 34.8% (48)Provider recommending biopsy

Attending radiologist 69.6% (96)Radiology fellow 21.7% (30)Outside referral 8.7% (12)

Experience level of radiologistperforming biopsy

Attending radiologist alone 38.4% (53)Fellow alone 8.7% (12)Attending radiologist with

fellow or resident52.9% (73)

Number of biopsy proceduresperformed

1 83.3% (115)2 12.3% (17)3 4.3% (6)

Type of procedure�

Ultrasound-guided core biopsy 64.5% (89)Ultrasound-guided diagnostic

aspiration9.4% (13)

Stereotactic 29.7% (41)Communication with

recommending radiologist52.44 � 11.48 (18-65)

Prebiopsy anxiety (STAI StateAnxiety Scale score)

44.49 � 12.42 (20-77)

Communication with radiologistconducting biopsy

57.84 � 8.43 (32-65)

Postbiopsy anxiety (STAI StateAnxiety Scale score)

37.24 � 11.63 (20-71)

Note: Data are expressed as mean � SD (range) or as percentage(number). STAI � State-Trait Anxiety Inventory.�Total is �100% because some women had more than one typeof biopsy procedure.

cation during biopsy (P � .69).

‡Possible range of scores for postbiopsy anxiety, 20 to 80.

Miller et al/Patient Anxiety and Breast Biopsy 427

Before the biopsy, better perceived communicationwith the recommending radiologist or provider was asso-ciated with lower anxiety (r � �0.22, P � .01). Bettercommunication with the recommending radiologist wasassociated with better communication with the radiolo-gist performing the biopsy (r � 0.40, P � .001).

Table 2. Bivariate analyses examining associationsbetween patient characteristics and prebiopsycommunication and anxiety (n � 138)

Variable

CommunicationWith

RecommendingRadiologist†

PrebiopsyAnxiety‡

Age (y) r � 0.16 r � �0.15Education (y) r � �0.05 r � 0.08Marital status

Unmarried 51.15 � 12.55 42.55 � 11.04Married 52.99 � 10.94 45.69 � 12.78

RaceBlack, Asian, or

American Indian48.99 � 12.05� 40.43 � 10.89�

White 53.71 � 11.05 45.98 � 12.66History of anxiety or

depressionAbsent 52.08 � 11.36 43.57 � 12.07Present 53.74 � 11.99 47.80 � 13.29

Cancer historyNo history of cancer 52.12 � 11.85 45.05 � 12.07Personal history of

cancer53.36 � 10.44 42.90 � 13.41

Breast cancer historyNo history of breast

cancer52.19 � 11.80 44.74 � 12.07

Personal history ofbreast cancer

53.52 � 10.12 43.40 � 14.03

First-degree relative withbreast cancer

None 51.62 � 11.93 44.91 � 12.08�1 55.54 � 9.12 42.92 � 13.75

Previous breast biopsyNone 50.19 � 12.27� 46.15 � 12.45�1 55.06 � 9.96 42.57 � 12.19

Previous breast surgeryNone 52.75 � 11.82 45.38 � 11.95�1 51.86 � 10.91 42.83 � 13.23

Provider recommendingbiopsy

Attending radiologist 53.28 � 11.60 44.87 � 12.40Radiology fellow 48.98 � 10.71 43.73 � 13.24Outside physician or

health professional54.41 � 11.55 43.39 � 11.32

Note: Data are expressed as mean � SD. Pearson’s correlations (a measureof the linear relationship between two variables, denoted as r and rangingfrom �1 to �1) were conducted for continuous demographic variables(ie, age and education), a one-way analysis of variance was conductedfor the 3-level variable for provider recommending biopsy, and theremaining 2-level variables were examined using independent t tests.�P � .05.†Possible range of scores for communication with recommend-ing radiologist, 13 to 65.

‡Possible range of scores for prebiopsy anxiety, 20 to 80.

Table 3. Bivariate analyses examining associationsbetween patient characteristics and postbiopsycommunication and anxiety (n � 138)

Variable

CommunicationWith BiopsyRadiologist†

PostbiopsyAnxiety‡

Age (y) r � 0.05 r � �0.06Education (y) r � 0.05 r � 0.05Marital status

Unmarried 57.72 � 8.37 36.19 � 10.35Married 58.12 � 8.40 37.57 � 12.27

RaceBlack, Asian, or other

race57.36 � 8.99 32.39 � 7.89�

White 58.01 � 8.25 39.01 � 12.29History of anxiety or

depressionAbsent 57.59 � 8.60 36.85 � 11.13Present 58.73 � 7.83 38.65 � 13.40

Cancer historyNo history of cancer 58.45 � 7.72 37.06 � 10.81Personal history of

cancer56.09 � 10.08 37.73 � 13.85

Breast cancer historyNo history of breast

cancer58.19 � 8.24 36.92 � 11.04

Personal history ofbreast cancer

56.32 � 9.19 38.59 � 14.04

First-degree relativewith breast cancer

None 58.03 � 8.43 37.20 � 10.98�1 57.08 � 8.51 37.39 � 14.01

Previous breast biopsyNone 57.46 � 8.22 37.80 � 11.07�1 58.27 � 8.71 36.59 � 12.29

Previous breast surgeryNone 58.61 � 7.85 37.35 � 11.07�1 56.38 � 9.33 37.02 � 12.72

Experience level ofbiopsy radiologist

Attending radiologistor fellow alone

59.05 � 8.55 36.24 � 11.06

Attending radiologistwith fellow orresident

56.75 � 8.22 38.13 � 12.12

Type of biopsyUltrasound guided 58.21 � 8.18 36.03 � 10.72Stereotactic 56.96 � 9.03 40.10 � 13.24

Number of biopsyprocedures

1 58.13 � 8.46 37.06 � 11.49�2 56.35 � 8.28 38.14 � 12.55

Note: Data are expressed as mean � SD. Pearson’s correlations (a measureof the linear relationship between two variables, denoted as r and rangingfrom �1 to �1) were conducted for continuous demographic variables(ie, age and education), and the remaining two-level variables wereexamined using independent t tests.�P � .05.†Possible range of scores for communication with biopsy radi-ologist, 13 to 65.

pw(rw

fFnl.a

afstn

428 Journal of the American College of Radiology/Vol. 10 No. 6 June 2013

Multiple Regression AnalysesTable 4 displays the results of multiple regression analy-ses examining whether perceived radiologist-patientcommunication and radiologist experience level wereassociated with prebiopsy and postbiopsy anxiety. Theregression model for prebiopsy anxiety was significant(total R2 � 0.15, F[8, 129] � 2.80, P � .007). Better

erceived recommending radiologist communicationas significantly associated with lower prebiopsy anxiety

� � �0.24, t � �2.74, P � .007). Recommendingadiologist or provider experience level was not associated

Table 4. Multiple regression analysis (n � 138)Variable Final Standardized �

Outcome: prebiopsy anxiety (totalR2 � 0.15)

Age �0.14Education �0.04Race† 0.25�

History of depression oranxiety‡

0.11

Previous biopsy§ �0.09Recommending radiologist

fellow/resident vs attendingradiologist�

�0.05

Recommendingphysician/provider outsideradiology vs attendingradiologist

0.01

Recommending radiologistcommunication

�0.24�

Outcome: postbiopsy anxiety(total R2 � 0.35)

Prebiopsy anxiety 0.52�

Education �0.01Race 0.15Type of biopsy (ultrasound vs

stereotactic)¶0.08

Biopsy radiologist experiencelevel#

0.03

Biopsy radiologistcommunication

�0.17�

�P � .05.†Race: 0 � black/Asian/American Indian, 1 � white.‡History of depression or anxiety: 0 � no, 1 � yes.§Previous biopsy: 0 � none, 1 � one or more.�As recommended for variables with three categories (ie, at-tending radiologist, fellow or resident, and outside physician orprovider), two dummy-coded variables were included in theregression model. The first variable compared fellows or resi-dents (coded 1) with attending radiologists (coded 0). Thesecond variable compared outside physicians or providers(coded 1) with attending radiologists (coded 0). Because it wasthe largest group, attending radiologists were used as thecomparison group for fellows or residents and outside physi-cians or providers.¶Type of biopsy: 0 � ultrasound guided, 1 � stereotactic.#Biopsy radiologist experience level: 0 � attending radiologistor fellow alone, 1 � attending radiologist with a fellow orresident.

ith prebiopsy anxiety (P � .30). The regression model

or postbiopsy anxiety was significant (total R2 � 0.35,[6, 131] � 11.60, P � .001]. Better perceived commu-ication during biopsy was significantly associated with

ower postbiopsy anxiety (� � �0.17, t � �2.36, P �02). Biopsy radiologist experience level was not associ-ted with postbiopsy anxiety (P � .68).

Differences Across RadiologistsIn 28 of the 138 study participants (20%), the sameradiologist recommended biopsies and performed theprocedures. Using independent t tests, there were nosignificant differences in anxiety between women whohad the same radiologist recommending and performingtheir biopsies and those who had different radiologists foreach (P � .30 for postbiopsy anxiety). However, a signif-icant difference was noted in communication scoresamong these groups (t � 3.05, P � .003). Women whohad the same radiologist recommending and performingtheir biopsies reported higher communication scores af-ter biopsy (mean, 62.3) than those who had differentradiologists (mean, 56.9).

Multilevel linear mixed models were used to comparepatients’ reports of anxiety and communication by indi-vidual radiologists. These models accounted for repeatedmeasurements from patients (eg, anxiety assessed beforeand after biopsies) and for multiple patients having thesame radiologists. All anxiety and communication scoresrelating to each individual radiologist (including scoresfor recommending biopsy and for performing biopsy)were included in one model. These analyses showed nosignificant differences in patient anxiety (P � .25) acrossradiologists. However, for communication scores, therewas a significant effect by individual radiologist (F[8,242] � 2.56, P � .01). Looking at pairwise comparisonsmong radiologists using Bonferroni’s correction, oneellow-in-training with the lowest mean communicationcore (47 � 10.8) had significantly lower scores (P � .05)han two attending radiologists with the highest commu-ication scores (58.7 � 8.2 and 57.4 � 7.2).

DISCUSSIONAs demonstrated in previous reports [1,2,9,24,39], ourstudy confirms elevated anxiety at the time of imaging-guided breast biopsies and has identified several associ-ated factors. The average prebiopsy State Anxiety Scalescore was 44.5 in our population of mean age 54 years,which is �1 standard deviation above the mean norms(32.2 � 8.67) for women of that age [36]. In our study,white women demonstrated higher levels of anxiety be-fore and immediately after biopsies compared with non-white women. This suggests that white women eitherneed greater emotional and psychological supportthroughout the biopsy process or have a reporting stylethat leads to higher scores; additional studies would beneeded to examine this issue. Although our patients’

mean anxiety score decreased from 44.5 before biopsies

lsitc

taithsnccicpmtctbb

tvftbttpmiertwfmTigmapSolahls

Miller et al/Patient Anxiety and Breast Biopsy 429

to 37.2 immediately after the procedures, suggesting thata component of prebiopsy anxiety is related to concernabout the procedure itself, this level of anxiety remainshigher than the normative sample mean, likely because ofuncertainty regarding biopsy results [1,11,39]. Patientswho reported higher levels of anxiety before their biopsiestended to remain more anxious after the procedures;these women might benefit from additional support, par-ticularly given that high levels of patient anxiety cannegatively influence adherence to future screening rec-ommendations [15-19].

Before considering anxiety-reducing interventions,however, it is important to understand how patients per-ceive communication with radiologists recommendingand performing breast biopsy procedures and what im-pact this communication has on anxiety. Breast imagingradiologists often engage in challenging and potentiallystressful conversations with patients when delivering the“bad news” of abnormal mammographic results, describ-ing the need for biopsy and the likelihood of malignancy,when obtaining consent from the patients for the proce-dures, during the biopsies themselves, and when notify-ing patients of malignant biopsy results [27,40,41].Given the challenging and sometimes life-altering con-tent of these discussions, patients’ perceptions of thesecommunications could greatly influence their psycholog-ical states [10,40,42,43].

Our study demonstrates that better perceived commu-nication with the recommending radiologist was associ-ated with lower prebiopsy anxiety. Likewise, betterperceived communication with the radiologist perform-ing the biopsy was associated with significantly loweranxiety afterward. These effects became stronger whencontrolling for baseline levels of anxiety, indicating thatradiologist-patient communication is a unique factor con-tributing to anxiety.

The experience level of the recommending radiologist orthe presence of a radiology resident or fellow during thebiopsy was not correlated with perceived radiologist-patientcommunication. There was no significant correlation be-tween biopsy type and perceived radiologist-patient com-munication or patient anxiety levels. Nonwhite womenreported poorer communication with radiologists at thetime of biopsy recommendation but also lower anxietycompared with white women. In the presence of theseother factors, the impact of this poorer communicationon nonwhite women is uncertain, given that communi-cation styles and patient responses are known to vary byethnicity [40]. Women with histories of breast biopsy, onthe other hand, reported better communication with therecommending radiologists, perhaps because previousbreast biopsy experience led to a better understanding ofwhat was being recommended. In this same subset ofpatients, however, previous breast biopsy history was notsignificantly associated with lower anxiety. Although com-

munication and anxiety were shown to be inversely corre- a

ated overall in the biopsy recommendation setting, thisubset of patients may have been more aware of the potentialmplications of breast biopsy (ie, possible cancer diagnosis),riggering anxiety that was not offset by the benefits of betterommunication.

Our finding of a significant inverse correlation be-ween radiologist-patient communication and patientnxiety scores represents a salient initial investigationnto the importance of radiologist-patient communica-ion both before and during breast biopsy procedures;owever, this relationship certainly deserves further con-ideration. First, although patient anxiety and commu-ication were correlated, our observational study designould not determine a causal relationship. These findingsould relate to other patient-specific factors not discern-ble in our study. Alternatively, lower patient anxietyould have been a direct result of better radiologist-atient communication, reflecting the radiologists’ com-unication skills. If lower anxiety scores indeed reflected

he radiologist’s skill in communication, a factor thatould be potentially enhanced by the radiologist’s effort,hen certainly facilitating better communication duringiopsy recommendation and the procedure itself shoulde emphasized.Although radiologists commonly deliver bad news in

he breast imaging setting, Adler et al [27] showed thatery few radiologists have had training in this importantorm of communication, and there are few data outlininghe necessary skills or the impact of applying them inreast imaging [4,40,44,45]. Our study measured pa-ients’ perceptions of radiologist-patient communica-ion, which is likely a critical factor in their subsequentsychological adjustment [10]; however, evaluation ofore objective indices of the radiologist’s behavior dur-

ng the interaction is also warranted, as investigated inarlier studies of other physicians delivering breast biopsyesults [10,25,42,43]. We found that at the conclusion ofhe biopsies, patients perceived better communicationith the radiologists who both recommended and per-

ormed the biopsies compared with women who com-unicated with a different radiologist at each time point.his might relate to receiving a consistent message dur-

ng biopsy discussions. Overall, having the same radiolo-ist recommend and perform the biopsy whenever possibleay be beneficial to patients in terms of communication

nd may also reduce the time required to perform therocedure, given the radiologist’s familiarity with the case.urprisingly, however, having the same radiologist rec-mmend and perform the biopsy did not seem to corre-ate with lower patient anxiety. We receive frequentnecdotal comments from patients expressing comfort inaving familiar individuals perform their biopsies, but

ower anxiety was not identified in our observationaltudy.

We also studied radiologists’ levels of experience (eg,

ttending radiologists vs fellows) and found no differ-

it

tmdd

430 Journal of the American College of Radiology/Vol. 10 No. 6 June 2013

ences in patients’ perceptions of radiologist-patient com-munication. However, other radiologist-related factorsthat might contribute to a communicator’s skill (eg, de-gree of formal training or interpersonal skills) were notinvestigated. Future studies are necessary to evaluate ra-diologists’ individual behaviors (eg, specific radiologistcommunication styles, amount of time a radiologistspends with individual patients) that could affect com-munications and therefore anxiety [10,25,42,43]. Wedid find that patients perceived significant differences incommunication among individual radiologists, suggest-ing that attention to specific radiologist factors in futureanalyses may be important for communication skills devel-opment in breast imaging training programs and for imple-menting anxiety-reducing measures for biopsy patients.

We recognize other limitations of our study. Therewas a 27% decline rate of invited participants, with somepatients indicating that they were too anxious to partic-ipate; lack of input from these patients could have biasedour results. Also, we invited patients into the study on theday they presented for biopsy to our outpatient breastinterventional facility, a separate facility from our diag-nostic breast imaging clinic. Prebiopsy questionnairesconcerning interactions with the radiologist recom-mending biopsy were administered at that time, ratherthan immediately after the patient communicated withthe recommending radiologist at our diagnostic clinic.Given the time interval (often �1 day) to reflect on thatcommunication, patients’ perceptions of communica-tion with the radiologists recommending biopsies couldhave changed from their original impressions and couldhave introduced bias into our investigation. Addition-ally, at the time of recommending biopsy, BI-RADS -Mammography, fourth ed., subcategories were not re-corded in all cases, and the manner and degree to whichthe radiologist informed the patient of the risk for cancerwas not assessed; this information could also influencepatient anxiety. Future studies could record radiologist-patient interactions to evaluate the impact of communi-cating cancer risk and determine the relationshipbetween BI-RADS 4 subcategories and anxiety. Finally,although we report on the significant anxiety patientsexperience after receiving biopsy recommendations andundergoing biopsy procedures, significant distress canalso occur after biopsies, during the time of uncertaintywhile patients await their results [39]. This period ofuncertainty is known to induce levels of physiologicaldistress in some women that are similar to receiving acancer diagnosis [39]. Our study did not address thessues of anxiety during that later time period; this will behe subject of future analysis.

CONCLUSIONSBetter radiologist-patient communication is a uniquefactor contributing to lower patient anxiety during the

breast biopsy process, with implications for radiologist

raining and efforts to improve patient adherence toammographic screening. Further investigation into ra-

iologist-patient communication for patient anxiety re-uction is warranted.

TAKE-HOME POINTS

● High levels of patient anxiety occur in women beforeimage-guided breast biopsy procedures, and althoughmean scores for anxiety decrease immediately after bi-opsy, anxiety persists at an elevated level above normsfor the population age group.

● White women report significantly higher levels of pre-biopsy and postbiopsy anxiety compared with non-white women in the breast biopsy setting, whereasnonwhite women report poorer communication withradiologists.

● Better perceived communication with the radiologistrecommending imaging-guided breast biopsy is asso-ciated with significantly lower patient anxiety beforethe procedure; likewise, better perceived communica-tion with the radiologist performing the imaging-guided breast biopsy is associated with significantlylower anxiety immediately after the biopsy.

● Given that elevated patient anxiety related to breastbiopsy can persist for years and has affected screeningmammography policy recommendations, these resultshighlighting the association between better radiolo-gist-patient communication and lower patient anxietyhave implications for efforts aimed at improving pa-tient adherence to mammographic screening.

● Radiologists’ recognition of factors associated withhigher patient anxiety in the setting of breast biopsycould lead to the identification of patients who mightbenefit from better efforts in communication.

● Awareness of these results could influence radiologisttraining in patient communication in breast imaging,as radiologist-initiated efforts for improved communi-cation might help reduce patient anxiety in the settingof breast biopsy.

REFERENCES

1. Flory N, Lang EV. Distress in the radiology waiting room. Radiology2011;260:166-73.

2. Maxwell JR, Bugbee ME, Wellisch D, Shalmon A, Sayre J, Bassett LW.Imaging-guided core needle biopsy of the breast: study of psychologicaloutcomes. Breast J 2000;6:53-61.

3. Liao MN, Chen MF, Chen SC, Chen PL. Uncertainty and anxiety duringthe diagnostic period for women with suspected breast cancer. CancerNurs 2008;31:274-83.

4. Deane KA, Degner LF. Information needs, uncertainty, and anxiety inwomen who had a breast biopsy with benign outcome. Cancer Nurs1998;21:117-26.

5. Lindfors KK, O’Connor J, Acredolo CR, Liston SE. Short-interval fol-low-up mammography versus immediate core biopsy of benign breastlesions: assessment of patient stress. AJR Am J Roentgenol 1998;171:

55-8.

Miller et al/Patient Anxiety and Breast Biopsy 431

6. Lindfors KK, O’Connor J, Parker RA. False-positive screening mammo-grams: effect of immediate versus later work-up on patient stress. Radiol-ogy 2001;218:247-53.

7. Brett J, Austoker J. Women who are recalled for further investigation forbreast screening: psychological consequences 3 years after recall and fac-tors affecting re-attendance. J Public Health Med 2001;23:292-300.

8. Brett J, Austoker J, Ong G. Do women who undergo further investigationfor breast screening suffer adverse psychological consequences? A multi-centre follow-up study comparing different breast screening result groupsfive months after their last breast screening appointment. J Public HealthMed 1998;20:396-403.

9. Lang EV, Berbaum KS, Faintuch S, et al. Adjunctive self-hypnotic relax-ation for outpatient medical procedures: a prospective randomized trialwith women undergoing large core breast biopsy. Pain 2006;126:155-64.

10. Mager WM, Andrykowski MA. Communication in the cancer “badnews” consultation: patient perceptions and psychological adjustment.Psychooncology 2002;11:35-46.

11. Montgomery M. Uncertainty during breast diagnostic evaluation: state ofthe science. Oncol Nurs Forum 2010;37:77-83.

12. Pinckney RG, Geller BM, Burman M, Littenberg B. Effect of false-positive mammograms on return for subsequent screening mammogra-phy. Am J Med 2003;114:120-5.

13. Lipkus IM, Halabi S, Strigo TS, Rimer BK. The impact of abnormalmammograms on psychosocial outcomes and subsequent screening. Psy-chooncology 2000;9:402-10.

14. Burman ML, Taplin SH, Herta DF, Elmore JG. Effect of false-positivemammograms on interval breast cancer screening in a health maintenanceorganization. Ann Intern Med 1999;131:1-6.

15. Roman R, Sala M, De La Veg M, et al. Effect of false-positives andwomen’s characteristics on long-term adherence to breast cancer screen-ing. Breast Cancer Res Treat 2011;130:543-52.

16. Chiarelli AM, Moravan V, Halapy E, Majpruz V, Mai V, Tatla RK. Falsepositive result and reattendance in the Ontario Breast Screening Program.J Med Sreen 2003;10:129-33.

17. Johnson MM, Hislop TG, Kan L, Coldman AJ, Lai A. Compliance withthe screening mammography program of British Columbia: will she re-turn? Can J Public Health 1996;87:176-80.

18. McCann J, Stockton D, Godward S. Impact of false-positive mammog-raphy on subsequent screening attendance and risk of cancer. BreastCancer Res 2002;4:R11.

19. Lerman C, Rimer B, Trock B, Balshem A, Engstrom PF. Factors associ-ated with repeat adherence to breast cancer screening. Prev Med 1990;19:279-90.

20. Roest AM, Martens EJ, de Jonge P, Denollet J. Anxiety and risk ofincident coronary heart disease: a meta-analysis. J Am Coll Cardiol 2010;56:38-46.

21. O’Leary A. Stress, emotion, and human immune function. Psychol Bull1990;108:363-82.

22. Stein MC, Roy-Byrne PP, Craske MG, et al. Functional impact andhealth utility of anxiety disorders in primary care outpatients. Med Care2005;43:1164-70.

23. US Preventive Services Task Force. Screening for breast cancer: U.S.Preventive Services Task Force recommendation statement. Ann InternMed 2009;151:716-26.

24. Novy DM, Price M, Huynh PT, Schuetz A. Percutaneous core biopsy ofthe breast: correlates of anxiety. Acad Radiol 2001;8:467-72.

25. Roberts CS, Cox CE, Reintgen DS, Baile WF, Gibertini, M. Influence of

physician communication on newly diagnosed breast patients’ psycho-logic adjustment and decision-making. Cancer 1994;74:336-41.

26. Liang W, Burnett CB, Rowland JH, et al. Communication betweenphysicians and older women with localized breast cancer: implications fortreatment and patient satisfaction. J Clin Oncol 2002;20:1008-16.

27. Adler DD, Riba MB, Eggly S. Breaking bad news in the breast imagingsetting. Acad Radiol 2009;16:130-5.

28. Beckman HB, Markakis KM, Suchman AL, Frankel RM. The doctor-patient relationship and malpractice. Lessons from plaintiff depositions.Arch Intern Med 1994;154:1365-70.

29. Berlin L. Communicating findings of radiologic examinations: whithergoest the radiologist’s duty? AJR Am J Roentgenol 2002;178:809-15.

30. Ambady N, Laplante D, Nguyen T, Rosenthal R, Chaumeton N, Le-visnson W. Surgeon’s tone of voice: a clue to malpractice history. Surgery2002;132:5-9.

31. Fallowfield L, Jenkins V. Effective communication skills are the key togood cancer care. Eur J Cancer 1999;35:1592-7.

32. D’Angelica M, Hirsch K, Ross H, Passik S, Brennan MF. Surgeon-patientcommunication in the treatment of pancreatic cancer. Arch Surg 1998;133:962-6.

33. Quirt CF, Mackillop WJ, Ginsburg AD, et al. Do doctors know whentheir patients don’t? A survey of doctor-patient communication in lungcancer. Lung Cancer 1997;18:1-20.

34. Silliman RA, Dukes KA, Sullivan LM, Kaplan SH. Breast cancer care inolder women: sources of information, social support, and emotionalhealth outcomes. Cancer 1998;83:706-11.

35. American College of Radiology. ACR, SBI breast imaging training cur-riculum. Available at: http://gm.acr.org/SecondaryMainMenuCategories/NewsPublications/FeaturedCategories/CurrentACRNews/archive/SBIACRBreastImagingTrainingCurriculum.aspx. Accessed August 7, 2012.

36. Spielberger CD, Gorsuch RL, Lushene PR, Vagg PR, Jacobs AG. Manualfor the State-Trait Anxiety Inventory (Form Y). Palo Alto, California:Consulting Psychologists Press; 1983.

37. Kline P. The handbook of psychological testing. 2nd ed. London: Rout-ledge; 1996.

38. Bieber C, Muller KG, Nicolai J, Hartmann M, Eich W. How does yourdoctor talk with you? Preliminary validation of a brief patient self-reportquestionnaire on the quality of physician-patient interaction. J Clin Psy-chol Med Settings 2010;17:125-36.

39. Lang EV, Berbaum KS, Lutgendorf SK. Large-core breast biopsy: abnor-mal salivary cortisol profiles associated with uncertainty of diagnosis.Radiology 2009;250:631-7.

40. Harvey JA, Cohen MA, Brenin DR, Nicholson BT, Adams RB. Breakingbad news: a primer for radiologists in breast imaging. J Am Coll Radiol2007;4:800-8.

41. Brandon CJ, Mullan PB. Patients’ perception of care during image-guided breast biopsy in a rural community breast center. J Canc Edu2011;26:156-60.

42. Loge JH, Kaasa S, Hytten K. Disclosing the cancer diagnosis: the patients’experiences. Eur J Cancer 1997;33:878-82.

43. Butow PN, Kazemi JN, Beeney LJ, Griffin A-M, Dunn SM, TattersallMHN. When the diagnosis is cancer: patient communication experiencesand preferences. Cancer 1996;77:2630-7.

44. Sasson JP, Zand T, Lown BA. Communication in the diagnostic mam-mography suite: implications for practice and training. Acad Radiol 2008;15:417-24.

45. Lown BA, Sasson JP, Hinrichs P. Patients as partners in radiology educa-

tion: an innovative approach to teaching and assessing patient-centeredcommunication. Acad Radiol 2008;15:425-32.