Correspondence to: Robert M. Arnold, MD, Division of General Internal Medicine, Section of Palliative Care and Medical Ethics, Montefiore University Hospital, 3459 Fifth Avenue Suite 932 West, Pittsburgh, PA 15213; telephone: (412) 692-4810; Fax: (412) 692-4892
Following a diagnostic or screening mammogram, patients with breast lesions are often referred for a biopsy.1 Time between the procedure and the notification of results is anxiety-provoking as women wait to find out if the lesion is malignant or benign.2
The preferences of women in this scenario regarding the method of communication and the provider who gives results are unknown. Providers try to balance different priorities: getting the information to the woman quickly,,  and  providing as much information as possible,, , ,  and  having the person who talks to the woman be someone they know and trust,8 and giving the information in person rather than on the phone.3 One may not be able to maximize all of the competing variables. If the sole priority is speed, then one would develop a system where the radiologist or nurse calls as soon as the biopsy result comes back; if complete information is most important, having someone from an oncologist's office call about positive results may be best. The problem is that there is a lack of published data on women's preferences, leading different programs to be based on assumptions rather than evidence.
Complicating this problem is that women likely have different preferences when the results are benign versus when they are malignant. To our knowledge, the communication preferences of these two groups of patients have never been compared. With evidence-based data about what women prefer, programs can develop more patient-centered programs to communicate biopsy results.
The goal of this study was to ascertain how women who have had a breast biopsy prefer to receive their test results. We also wanted to determine their satisfaction with the way they did receive their biopsy results at our cancer center and whether satisfaction differed based on patient age, race, or biopsy results. It is hoped that these data will help other programs develop communication policies.
Materials and Methods
Study Setting and Patient Recruitment
This study was based on a telephone survey of all radiologic breast biopsy patients seen at a large urban academic breast center between June 1, 2008, and July 31, 2008. The study was approved by the University of Pittsburgh Institutional Review Board. Study participants were adult females receiving a minimally invasive radiologic breast biopsy who spoke English and had a working phone. All patients fitting the inclusion criteria were approached before their procedure and given the option to participate in the study.
The center performed over 3,500 breast biopsies in 2009. At the center, two nurses gave results to an average of 30 patients a day. In addition to making calls, the nurses are responsible for admitting and discharging patients and work on rotation in order to call patients they have personally met. They make notes in patients' charts about their demeanor and concerns to help them tailor the phone calls to the patients' personalities. When patients leave after the biopsy, the nurses discuss the results call and ask the patients if they would prefer the nurse to call them or if they would like to call the nurse on their own time. No option is available for an in-person results visit except by the physician ordering the biopsy.
Study participants received biopsy results within 4 business days. Information about positive and negative results is communicated in different ways. While all results are faxed to the referring physician, positive results have a cover sheet indicating the urgency of the information. If results are positive or require further surgical biopsy, the nurses call in a room with a closed door and a “do not disturb” sign, to minimize interruptions. Nurses provide information using a standardized script that describes the result and their implications. Patients with a malignant biopsy are given a phone number to make a breast magnetic resonance imaging appointment. They are also told that they need to make a surgical appointment, but the choice of surgeon is left up to the patient and the referring physician. Patients are given time to ask questions and the direct phone number if they wish to speak to the nurse again.
Phone Survey Procedure
Two weeks were allotted between receipt of results and the study interview to allow time for patients to understand their diagnosis and seek follow-up care as necessary. After the 2-week waiting period, a study staff member contacted patients by telephone. Calls were conducted in the order in which patients received their results, and four attempts were made to call each participant, with a message left each time.
The phone survey consisted of four sections: an informational section, which collected data about how the patient received the results; a communication skills section, which assessed patient impressions of the person giving results; an improvement section, which assessed patient views about how to improve the communication of results; and finally, a communication priorities section, which assessed the relative importance of four distinct aspects of communication (Table 1). In addition, patients were asked “What did you like best about how you were told your results?” and “What can we do to make the process of giving results better?” The communication skills and improvement sections were scored on a five-point Likert scale, and the communication priorities section was scored on a rank scale from most to least important. (The survey is available on request.) Demographic information as well as the number of previous biopsies the woman had were also collected.
|Receiving the results of the biopsy as soon as possible|
|Being told by a person who knows the most about what the results mean|
|Being told the results in person|
|Being told by your primary care provider|
Survey statistics were analyzed using IBM SPSS Statistics software (SPSS, Inc., Chicago, IL). A one-sample Kolmogorov-Smirnov test was used to test variable normality. As all Likert-scaled survey variables were not normally distributed, a Mann-Whitney U-test was used to compare Likert scores between cancer and benign groups as well as first-time biopsy and repeat biopsy groups. Ordinal regression was used to evaluate the effects of age on Likert-scaled variables.
We screened 133 patients, and of these, 131 patients consented to participate in the study. Of these, 64 could not be reached during follow-up and one patient withdrew from the study, for a total of 66 patients completing the study. The overall response rate was 50.4%. Of the patients who completed the telephone interview, 39 had benign biopsies and 27 had cancer. Of the patients who did not complete the survey, 10 had cancer and 55 were benign (P = 0.004). Other demographic data from the survey cohort are illustrated in Table 2. As the vast majority of patients were white, a comparison between different races could not be performed. Age did not have any significant effect on any of study variables.
Of all patients in the study, 53 (80.3%) were contacted by a nurse from the breast center. The other patients were contacted by their primary care provider first (n = 5) or a radiologist (n = 3) or did not know who they were contacted by (n = 5). Forty-one patients (62.1%) recalled meeting the provider they spoke with, while 15 patients reported they did not meet the person who contacted them and 10 were not sure. Sixty-three patients (95.5%) were told their results over the phone, two were told in person, and one person did not respond.
Mean Likert scales are reported in Table 3. There were no significant differences in the patients' assessments based on demographic or clinical variables. Overall, patients rated the communication skills of the person who gave their results very positively.
|SURVEY ITEM||MEAN LIKERT SCORE||STANDARD DEVIATION||% OF PATIENTS REPORTING “AGREE” OR “STRONGLY AGREE”|
|You were given the diagnosis in a timely fashion||4.35||0.76||93.2%|
|The person who gave you the diagnosis was considerate and tactful||4.60||0.63||93.2%|
|The person who told you the diagnosis was honest||4.46||0.50||100%|
|The person told you the results in a way you could understand||4.31||0.77||94.9%|
|The person who told you the results did not rush||4.08||0.85||86.4%|
|The person who told you the results gave you the opportunity to ask questions||4.13||0.77||87.9%|
|The person who told you the results was sensitive to your emotional reaction||4.26||0.81||87.5%|
|You were satisfied with hearing your results by the method you did (eg, over the phone)||3.98||0.98||79.7%|
|You were satisfied with hearing the results from the person you heard from||4.11||0.95||86.4%|
Areas for Improvement
The proportions of patients responding “agree” or “strongly agree” to each item are reported in Figure 1 and are compared between several patient groups in Table 4. Patients were more likely to want additional materials to help them understand their diagnosis. This was significantly more common among patients having a first biopsy and patients who had cancer (P < 0.05). For example, 65.4% of cancer patients wanted more information versus 43.5% of benign patients. Also, 60.5% of patients having a first biopsy wanted additional information versus 37.0% of patients having a repeat biopsy. For all other items, less than 50% of patients answered “agree” or “strongly agree,” and there were no significant trends based on clinical or sociodemographic variables.
|SURVEY ITEM||MEAN LIKERTa ± SD||P||MEAN LIKERTa ± SD||P|
|CANCER||BENIGN||FIRST BIOPSY||NOT FIRST BIOPSY|
|You would have preferred additional materials to help you understand the diagnosis||3.50 ± 0.99||2.82 ± 1.14||0.018||3.34 ± 1.05||2.74 ± 1.16||0.036|
|You would have preferred to talk to someone beforehand to discuss how much you wanted to know about your results||2.78 ± 1.09||2.77 ± 1.16||0.977||2.97 ± 1.08||2.50 ± 1.13||0.068|
|You would have preferred more assistance making follow-up appointments||2.19 ± 0.62||2.58 ± 1.22||0.370||2.57 ± 1.02||2.21 ± 1.03||0.088|
|You would have preferred to choose who gave you the results||2.54 ± 1.07||2.54 ± 1.07||0.896||2.55 ± 1.00||2.46 ± 1.07||0.618|
|You would have preferred to receive the results faster||2.85 ± 1.20||2.85 ± 1.20||0.428||3.00 ± 1.27||2.86 ± 1.17||0.638|
a 1, strongly disagree; 2, disagree; 3, neutral; 4, agree; 5, strongly agree
Patient Priorities for Receiving Biopsy Results
For patients with benign and those with malignant disease, receiving results quickly was the most important factor, followed by being told by a person who knows the most about what the results mean (Figure 2). Hearing from a primary care provider and hearing in person were of much lower priority. Cancer patients ranked “Being told by a person who knows the most about what the results mean” significantly higher in priority than benign patients (P < 0.05).
In our study population of women who had a breast biopsy, the number one priority was receiving the results as soon as possible. We found that women were generally satisfied with learning about their results from a nurse over the phone. However, a majority of patients said they would prefer additional materials to help them understand their diagnosis. This trend was more pronounced in women with cancer or those for whom this was the first biopsy. Previous literature has shown that both doctors and patients agree that potentially bad news should be given in a quiet, uninterrupted, face-to-face conversation3 and that it should be given by a provider they know well.8 However, it has also been suggested that many women want to hear test results as quickly as possible, even if that means they will receive them from a provider they do not know well.,  and  This study is unique in that it forced women to choose which of these aspects of communication were most important as we believe that rapid results and hearing in person are often mutually exclusive. Women in this study clearly preferred to hear quickly more than in person, whether they were given a diagnosis of cancer or not.
Some studies suggest that providing written information along with verbal communication would be beneficial. For example, in a 2004 study by Lobb et al,9 genetic counselors who added a summary letter after communicating breast cancer risks to a patient significantly increased realistic risk assessments in patients, as well as lowering anxiety. Our study found that women would prefer additional materials. Since most communication in this study was over the phone, women may not have had time to fully process all the information given and therefore wanted information supplements as well. In women with cancer or those receiving a first biopsy, information and follow-up instructions are even more complex or overwhelming; and it follows that they are more likely to want informational materials than other patients.
As expected, patients who are diagnosed with cancer want to talk to someone with more knowledge of their specific condition. While patients in this study were not asked specifically about the apparent knowledge base of the nurse who gave results, their overall satisfaction (86.4%) seems to suggest that our system with specific scripts for each diagnosis was adequate.
This study is limited in several ways. First, as a single-center satisfaction study, we are only measuring the experience of patients with a particular system of communication. While these data reflect the general system in place at the breast center, they are also contingent on the specific providers communicating with patients. However, we do believe our results are reproducible at other centers as scripts are often used for sharing patient results. This also means that these data are a reflection of a population that was contacted by telephone. We cannot make any assumptions about how these preferences may differ from those of patients contacted in person. Second, only general data were gathered about the interaction of the nurse with each woman. It is not known what was specifically said to each patient, so we can only report the patient's view of the discussion. However, since responses are scripted at the breast center, we assume that the communication was relatively standardized for all conversations. Third, patients in the study were more likely to have a cancer diagnosis than the total population of biopsy patients. This may represent a greater commitment by cancer patients to aid in the improvement of communication. It also may indicate that women with a benign biopsy were less opinionated about their communication in general and less likely to want to express their opinions in the study.
In conclusion, patients generally prefer to hear breast biopsy results quickly over other factors, including hearing in person or from a more experienced practitioner. Therefore, a program similar to the one at our center meets most patient needs; it minimizes wait time by calling patients as soon as biopsy results are in, utilizes nurses to facilitate the large amount of calls that must be made every day, and ensures patients are contacted by a provider they personally met during their biopsy. A majority of patients desired additional materials to supplement phone communication. We highly recommend providing a variety of materials, including both written and Web-based, to address this need. Further research is necessary to determine the effects of these interventions on patient understanding and long-term emotional outcomes.