| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Division of General Internal Medicine, Northwestern University Medical School,
675 N. St. Clair, Ste. 18-200, Chicago, IL 60611.
2
Department of Ambulatory Oncology, Northwestern Memorial Hospital, 675 N. St.
Clair, 3rd Fl., Chicago, IL 60611.
3
Department of Radiology, Section of Breast Imaging, Methodist Hospital, Baylor
College of Medicine, 6550 Fannin St., 7th Fl., Ste. 701, Houston, TX
77030.
Received May 24, 2001;
accepted after revision August 20, 2001.
Presented in part at the annual meeting of the Radiological Society of
North America, Chicago, November 2000.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We used a telephone survey with sampling that was stratified to reflect patients with normal and abnormal findings who had screening and diagnostic mammograms. Patients with visits before the mandate became effective (April 1999, n = 298) and after (January 2000, n = 316) were interviewed about the average time to receive results, satisfaction with communication about results, anxiety, and perceived follow-up recommendations. Multiple logistic regression was used to test the association of time period with patient dissatisfaction, controlling for age, anxiety level (considerable or extreme vs none or moderate), examination type (screening vs diagnostic), and examination result (normal vs abnormal findings).
RESULTS. No significant difference was found between periods in anxiety about results or agreement with documented radiology recommendations, but we found a substantial increase in the number of screening patients who reported timely receipt of results. Significantly fewer patients were dissatisfied with mammographic results communication after the mandate (multivariable odds ratio = 0.46, p = 0.006). Screening examination patients and patients who reported considerable or extreme anxiety about test results were more likely to be dissatisfied in both periods.
CONCLUSION. By standardizing results notification, the Mammography Quality Standards Reauthorization Act improved patient satisfaction and reporting timeliness among screening examination patients, but did little to improve anxiety or recollection of recommendations. Future quality improvement efforts should focus on improving patients' understanding of follow-up recommendations.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The objectives of this study were to evaluate whether implementation of the mandate was associated with improved timeliness of the reporting of results, changes in anxiety about test results, changes in patient understanding of results and recommendations, and differences in satisfaction with how results were communicated. Although satisfaction encompasses many dimensions of the mammography experience such as the facility, performance of the procedure by the staff, and patient discomfort with the procedure, few studies have evaluated various aspects of patient satisfaction with communication of mammographic results [5,6,7,8,9,10,11,12]. Our study is among the first to directly examine satisfaction with the communication of results, particularly among patients undergoing screening mammography.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Study Setting and Results Reporting Practices
Approximately 17,000 screening mammograms, 14,500 diagnostic mammograms,
and 4,500 breast sonograms are obtained each year at the study center.
Screening mammography and diagnostic mammography are performed at separate
locations with central radiology interpretation and reporting of results.
After the first study period, the diagnostic mammography center was moved to
another site on the same campus; the screening mammography center remained the
same. The mammography center is part of a tertiary care referral center that
is affiliated with a large regional cancer center, with traditionally high
abnormal screening rates
[15].
Before the written notification mandate took effect on April 28, 1999, the mammography center mailed written mammographic results and recommendations to the referring physician. It was assumed that women with screening mammograms were generally notified of the results by their referring physician. Women with abnormal screening mammograms requiring additional views or immediate follow-up were also called directly by the mammography center to schedule a diagnostic examination. After 30 days, lack of response to repeated telephone messages would then prompt reminder letters from the mammography center. Except under unusual circumstances, women who had diagnostic appointments were informed of their results and recommendations directly at the mammography center at the time of their examinations.
After April 28, 1999, the mammography center mailed a standardized letter in lay language to all patients who underwent screening mammography with their results and recommendations, usually within a week of the visit. If additional views or immediate follow-up were indicated, a call was also still made, and lack of response after 30 days still prompted a reminder letter. For patients who underwent diagnostic mammography, in addition to results being given verbally, a written recommendation checklist was given to the patient at the time the mammogram was discussed.
Survey Sampling and Interview Procedure
Potential study participants were accrued during a 1.5-week period before
the mandate was implemented and during a 1.5-week period 8 months after the
mandate was implemented. Receptionists at the study site gave all women
presenting for mammograms during these times an informational letter approved
by the institutional review board that described the study. The patients were
asked to indicate refusal to participate by checking off a box on the
informational letter. Telephone numbers of women accured during the two
periods were collected from a computerized mammography examination registry.
From these lists, sampling was conducted randomly among patients in each of
four categories: normal screening, abnormal screening, normal diagnostic
examination, and abnormal diagnostic examination.
Calls were made sequentially down each sampling list until a target of at least 60 interviews in each examination type and result category was reached. Interviewers called additional women who were randomly selected to replace those who could not be contacted after five attempts. This sample size provided 80% power to detect a difference between 10% and 30% in the proportion of patients reporting ratings of fair or poor in overall satisfaction with the communication of mammographic results in the period before versus after the mandate (p = 0.05, in two-tailed test).
The telephone interviews for both periods were conducted by eight women who had been given a 1.5-hr training session. The interviewers were trained to specifically focus women's recall about results and recommendations on a specified examination date. Although interviewers were provided with each respondent's age and examination category, they were unaware of the respondents' test results.
Telephone interviews were conducted 2 weeks after the examination date for patients undergoing diagnostic examinations, 3 weeks after examination for patients with abnormal screening examinations, and 4 weeks after examination for patients with normal screening examinations. The different intervals were chosen to allow ample time for women to receive their results, thus minimizing any anxiety potentially created by calling women who had not yet received their results, while maximizing recall of recommendations.
For this study, the Breast Imaging Reporting and Data System (BI-RADS) categories for mammography interpretation were used to define normal versus abnormal results [16]. Abnormal results were defined as mammograms that required additional evaluation: results suggestive of malignancy, recommendations for additional follow-up in 6 months or less, additional views, clinical or surgical consultation, sonography, or biopsy (BI-RADS categories 0, 3, 4, and 5). Normal results were defined as negative or benign results with recommendations for follow-up in 1 year or at 40 years old (BI-RADS categories 1 and 2).
Survey Items
The survey measured patient self-reported receipt of results, mean number
of days to receive results after the examination date, anxiety level about
test results, recollection of results and follow-up recommendations, overall
satisfaction with communication of test results, and specific dimensions of
satisfaction with communication. Responses were then compared with documented
radiology results and recommendations obtained using RADNET radiology results
software (version 7.2; Cerner Products, Kansas City, MO). Because of the
differences in mammography center test reporting practices based on
examination type, surveys for diagnostic and screening patients had distinct
items (e.g., surveys for patients with screening examinations contained a
question about the amount of time to receive results). Surveys administered to
the January 2000 sample contained an additional question about how the written
materials affected the patient's anxiety.
The initial survey was reviewed for content validity by radiologists and internal medicine physician investigators. A revised version was tested on six past and current mammography patients. The final survey for patients with both normal and abnormal screening examinations contained 17 items, and the final survey for both normal and abnormal diagnostic examinations contained 12 items. Both versions were approved by the institutional review board. Interviews averaged approximately 10 min in duration.
Comparative Study End Points
Timeliness was measured by the proportion of patients with normal screening
examinations who had not received results by 4 weeks after the examination
date, the proportion of patients with abnormal screening examinations who had
not received results by 3 weeks after the examination date, and the mean
number of days to notification after the screening examination date among
those who had received results. Anxiety level about test results was measured
by the proportion of patients who reported experiencing no anxiety, or
moderate, considerable, or extreme anxiety about test results. The percentage
of agreement between documented radiology recommendations and patient
self-reported recommendations was compared for each period. The proportion of
patients who were dissatisfied overall with results of communication was
compared between time periods. Finally, Likert-scale ratings of specific
dimensions of communication for screening patients, such as clarity and the
amount of time to receive results, were also compared.
Statistical Analysis
Likert-scale ratings of satisfaction items were trichotomized (poor or
fair, good or very good, excellent) and chi-square tests were used to test the
significance of differences in ratings between time periods. A t test
was used to compare periods for mean numbers of days to receipt of screening
results. Kappa tests of agreement were performed for each time period to
calculate chance-adjusted agreement between radiology recommendations and
patient-reported recommendations. A z test was used to detect
differences in kappa scores between time periods.
Multiple logistic regression was then used to test the association of time period (before vs after the written-notification mandate) with the ratings of overall dissatisfaction with reporting practices. Dissatisfaction was defined as a fair or poor overall rating. The regression model tests the significance of differences in dissatisfaction between periods and controls for the effects of patient age over 50 [13], anxiety level about test results (considerable and extreme anxiety vs no or moderate anxiety), examination type (screening vs diagnostic), and examination result (normal vs abnormal mammography findings).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Receipt of Results After Screening Mammography
Table 1 presents the
proportion of screening patients who reported that they had received their
results by the time of the interview. All patients who underwent diagnostic
examination reported having received results. By the interview date, the mean
number of days to notification for normal screening patients was
nonsignificantly lower after the mandate (13.2 ± 7.3 days before vs
11.1 ± 6.5 days after the mandate, p = 0.09). Among patients
with abnormal screening examinations, the mean number of days to notification
was significantly lower after the mandate (10.3 ± 4.0 days before vs
5.0 ± 3.6 days after the mandate, p = 0.001).
|
Anxiety About Results
No significant differences were found in anxiety levels among any of the
four examination type and result categories between the two periods (data not
shown). As expected, compared with normal results, a higher proportion of
patients with abnormal results for both screening and diagnostic mammograms
reported moderate, considerable, or extreme anxiety. Patients with abnormal
screening examinations reported the highest anxiety levels during both
periods. After the mandate, a higher proportion of patients with abnormal
diagnostic examinations reported extreme anxiety (12.5% before vs 16.1% after
the mandate), but a lower proportion of patients with abnormal screening
examinations reported extreme anxiety about test results (12.5% before vs 7.9%
after the mandate). Among all respondents in the January 2000 sample who were
asked to rate how the written results affected their anxiety, 38% reported
that the written results decreased their anxiety, 47.4% reported no change,
and 14.6% reported that the written results increased their anxiety.
Agreement Between Documented and Patient-Reported
Recommendations
Table 2 presents agreement
rates before and after the mandate between four categories of documented
radiology recommendations and patient-reported recommendations. The kappa
statistic for overall agreement before the mandate was 0.75; after the
mandate, the kappa was 0.71 (z test, p = 0.4 for differences
between time periods). Agreement rates were highest for patients with normal
examinations and recommendations for routine 1-year follow-up (90% before vs
87% after the mandate). For patients with surgical consultation or biopsy
recommendations, agreement rates were lower (68% before vs 70% after the
mandate).
|
Satisfaction with Reporting of Results
Table 3 presents ratings of
satisfaction with communication of results for patients undergoing screening
examinations. The greatest difference in overall satisfaction ratings between
periods was for patients with normal screening examinations (17.9% poor or
fair ratings before vs 6.5% after mandate, p = 0.03). Overall satisfaction
results were mirrored in ratings of poor or fair for clarity of results (20.4%
before vs 6.4% after the mandate, p = 0.002) and ratings of
timeliness of results (34.6% before vs 15.2% after the mandate, p =
0.01) for patients with normal screening examinations. Although similar
decreases in dissatisfaction were noted among women whose screening
examinations had abnormal findings, the results were not statistically
significant.
|
Differences in overall satisfaction ratings by diagnostic examination patients were all nonsignificant (data not shown). The proportion of diagnostic examination patients with abnormal findings who reported poor or fair ratings decreased slightly after the mandate (12.3% before vs 6.8% after the mandate). Despite improvements in overall satisfaction for this group, ratings given by women with abnormal diagnostic examinations for the amount of time spent communicating results were nonsignificantly worse after the mandate (13.7% poor or fair ratings before vs 17.2% after the mandate).
Predictors of Dissatisfaction
Table 4 presents the
logistic regression model of overall dissatisfaction with communication of
mammographic results and recommendations. Having a mammogram in the time
period after the mandate was associated with about half the likelihood of
dissatisfaction (postmandate vs premandate odds ratio = 0.46, p =
0.004). The likelihood of dissatisfaction was also associated with undergoing
a screening examination (screening vs diagnostic examination type, odds ratio
= 2.1, p = 0.006), and reporting considerable or extreme anxiety
about test results (odds ratio = 3.4, p < 0.001). Younger age and
a normal versus abnormal mammography result were not significantly associated
with dissatisfaction.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The MQSRA has been criticized as yet another unfunded federal government mandate that imposes considerable costs on mammography providers with little evidence of potential benefits. Legislators hoped that the MQSRA would improve the clarity of recommendations for follow-up and reduce unnecessary anxiety caused by long waiting times for test results. A recent study that examined women's preferences for communication of diagnostic mammographic results found that the majority preferred direct communication from radiologists for both normal and abnormal results at the time of the mammogram rather than waiting for referring physicians to disclose the results at a later time [11]. A survey among mammographers reported that approximately three quarters of respondents agreed with this notification strategy [12]. However, the mailing of many thousands of written results and recommendations itself posed a potential communication hazard and could have actually increased anxiety and confusion among patients, particularly for those mammography centers who already had an effective results reporting system.
In this light, implementation of mailed notification at this study center may be viewed as a qualified success. Although anxiety and concordance measures remained unchanged, many patients, with both normal and abnormal results, appeared to benefit from better and more timely communication. This appears to be largely the result of substituting a standardized reporting protocol for what had previously been a fairly complex effort to reach patients either directly or through a multitude of referring physicians. However, these results indicate that a large potential still exists for improving communication with patients about results and recommendations.
Previous research suggests that the psychological consequences of an abnormal screening study may persist despite subsequently learning of normal findings in an initially suspicious lesion [17]. In addition to experiencing anxiety about abnormal test results, women with abnormal screening mammograms may also feel anxiety about delays in finding out about the need for a further work-up. It may thus be surprising that, despite halving the mean number of days to receipt of results, there was no difference in anxiety level between the two periods among patients with abnormal results. However, our survey specifically asked patients about anxiety related to their actual results, and not about the overall experience of anxiety during the entire follow-up period after their initial screening visit. Shorter time to receipt of results may in fact have reduced the overall burden of the breast imaging for our patients.
Ideally, written notification could serve to reinforce follow-up recommendations for patients. However, in this study, we found little change in agreement between documented results and recommendations and those reported by the patient. Although the agreement rates were fairly high (87-90%) for women needing follow-up in 1 year, the agreement rates for those with mammograms requiring surgical consultation or biopsy were much lower. This finding is likely the result of confusion about initial versus subsequent results, but it may reflect a more serious problem. For centers with similarly discordant results, this is probably the most important area for quality improvement in communication with patients.
This study has many limitations. First, it is uncertain whether written notification or other temporal trends may have caused improvements in satisfaction. Second, no assessment was made of traditional patient satisfaction measurements, such as discomfort with the mammography procedure, interactions with the staff, or waiting time for appointments. Third, although efforts were made to focus patients' recall about results and recommendations for the relevant examination date, agreement between documented and patient-reported recommendations may be less accurate for patients with multiple recommendations and examinations. Fourth, the diagnostic mammography center changed sites after the first study period. However, the move is unlikely to have influenced outcomes, because the primary focus of the study was on patients undergoing screening examinations, and that site remained unchanged. Finally, this study was based on the screening process used at one institution and may not reflect test reporting practices at other centers.
In conclusion, implementation of systematic results notification as mandated by the MQSRA was associated with an improvement in receipt of results, timeliness of reporting, and patient satisfaction with the communication of results. We found no change in patients' understanding of results or anxiety about results. Future research on patient communication should focus on methods of improving understanding of follow-up recommendations and allaying anxiety.
Acknowledgments
We acknowledge the research assistance of Matthew Weinberg and Asta
Sorensen of the Division of General Internal Medicine at Northwestern
University Medical School, and Rebecca Rogers, Robin Loebach, and Nikki Kay of
the Northwestern Memorial Hospital Lynn Sage Comprehensive Breast Center.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
