IJE Advance Access originally published online on May 30, 2008
International Journal of Epidemiology 2008 37(6):1326-1332; doi:10.1093/ije/dyn090
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Use of breast cancer screening and treatment services by Australian women aged 25–44 years following Kylie Minogue's breast cancer diagnosis
1 Centre for Health Policy, Programs and Economics, School of Population Health, University of Melbourne, Victoria, Australia.
2 St Vincent's Breast Screen, St Vincent's Hospital, University of Melbourne, Victoria, Australia.
3 Department of Surgery, University of Melbourne and Royal Melbourne Hospital, Victoria, Australia.
4 Key Centre for Women's Health in Society, University of Melbourne, Victoria, Australia.
5 Melbourne Law School, University of Melbourne, Victoria, Australia.
* Corresponding author. Centre for Health Policy, University of Melbourne, 207 Bouverie Street, Carlton, Victoria 3010, Australia. E-mail: mkelaher{at}unimelb.edu.au
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionACKNOWLEDGEMENTSREFERENCES |
|---|
Objective To examine the effects of the publicity surrounding Kylie Minogue's diagnosis with breast cancer on doctor-referred breast imaging, image-guided biopsy, and cancer excisions among a low-risk population of women in Australia.
Method We examine changes in unilateral and bilateral breast imaging, image-guided breast biopsies, and surgical excisions of breast cancer before and after the announcement of Kylie Minogue's diagnosis with breast cancer in May 2005. The study included procedures provided through the Australian public health system to women aged 25–44 years from October 2004 and June 2006.
Results The odds of women aged 25–44 years undergoing imaging procedures increased by 20% in the first and second quarters after the Minogue publicity, compared to the preceding two quarters. The volume of biopsies als increased but the biopsy rate, measured as a proportion of imaging procedures, did not change among women aged 25–34 years and decreased among women aged 35–44 years. The volume of operations to excise breast cancers did not change for either age group. Compared to the 6 month period before the publicity, there was a large and significant decrease in the odds that an excision would follow biopsy (25–34 years: OR 95% CI=0.69, 0.48–0.98; 35–44 years: OR 95% CI=0.83, 0.72–0.95).
Conclusions High-publicised illnesses may affect both consumer and provider behaviour. Although they present opportunities to improve public health, they also have the potential to adversely impact the appropriateness and cost-effectiveness of service delivery.
Keywords Mass media, health promotion, practice guidelines
Accepted 29 April 2008
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionACKNOWLEDGEMENTSREFERENCES |
|---|
We live in an age of celebrity worship. The daily activities of the public's favourite actors, singers, royals and athletes shape consumer behaviour. The influence extends to health care.
Celebrity illness and treatment may trigger intense media coverage, raise public awareness of the relevant condition and impact care-seeking behaviour and treatment choices at the population level.1–15 Such behaviour change has been linked to diagnoses of cancer among high-profile celebrities,1–3,5,6,8,10–12 particularly breast cancer.3,10,11 Previous research has tended to emphasize the potential for the media exposure to deliver public health benefits3,10,11—for example, promotion of clinically appropriate screening practices. The potential for reactions that are undesirable from a public health standpoint has received relatively little attention.
On May 17, 2005, Australian and international news agencies began reporting that the popular Australian singer-actress, Kylie Minogue, had been diagnosed of breast cancer. A wave of headline stories followed in the print and electronic media. Chapman and colleagues3 found a 20-fold increase in media coverage of breast cancer in Australia in the 2 weeks after the story first broke. Bookings for mammography at BreastScreen Australia, a national programme aimed primarily at women aged 50–69 years, soared by 40% among previously screened women and by 101% among previously unscreened women.
In this study, we examine effects on actual service delivery in a low-risk population. Using national data, we compare the frequency of breast imaging, image-guided breast biopsies and operations to excise breast cancers performed through Medicare before and after the publicity surrounding Minogue's diagnosis. We focus the analysis on women aged 25–44 years for two reasons. First, the impact of celebrity illness is believed to be greatest among subpopulations most demographically similar to the affected celebrity.9,11 (Minogue was 36 years of age in mid-May 2005.) Second, in the absence of family history or specific symptoms, routine screening is not recommended for women in these age groups.16 Investigating service use in this age group may thus shed light on potentially negative public health effects of publicity-inspired spikes in screening activity.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionACKNOWLEDGEMENTSREFERENCES |
|---|
Clinical process
Access to publicly funded breast imaging in Australia occurs either directly through the BreastScreen program or by referral from a doctor, usually a general practitioner. BreastScreen services are not available to women under 40 years. Recommendations state that referral for imaging in this age group is indicated if there is a personal or family history of breast cancer, or symptoms or clinical indications of possible breast cancer.17 The decision resides with referring doctor.18
Figure 1 illustrates the screening and treatment pathway for women aged 25–44 years.
|
The approach chosen for imaging is age- and symptom-dependent.19 Among women who present because of symptoms, ultrasound is the recommended imaging procedure for those younger than 35 years and mammography is recommended for those 35 years and older. Some symptomatic women may then undergo further imaging, typically with the other modality, to clarify the initial investigations.
Asymptomatic women referred for screening generally undergo bilateral mammography. However, because mammography is less sensitive in younger women,20 women in their 20s and early 30s will sometimes have bilateral ultrasound (instead of or in addition to bilateral mammography). It would be highly unusual for a woman of any age to have repeat bilateral imaging as part of a single diagnostic course. If symptoms are detected or require further investigation following imaging then biopsy is the next stage in the process.19
Because our dataset consists of procedure counts, not counts of individual women, we focused on bilateral imaging procedures to obtain imaging rates. This strategy minimizes the chances of double-counting women. However, it opens up the possibility of some undercounting among symptomatic women, some of whom will have received unilateral procedures only. The effect of such undercounting would be to render any increases in imaging observed in the wake of the Minogue publicity underestimates of the true increases.
Data
Medicare Australia, the country's universal health insurance programme, provides partial reimbursement for all services related to the detection and treatment of breast cancer. All Australian citizens and permanent residents are eligible for Medicare coverage.
We obtained data from Medicare Australia on the use of doctor-referred breast imaging (bilateral mammography and ultrasound), image-guided biopsy (fine needle and core breast biopsy and wire localization of breast lesions) and surgical excision of breast cancers (open surgical biopsy of malignant tumour; complete local excision of malignant tumour; total mastectomy and subcutaneous mastectomy), respectively, for the period January 1, 2004 to September 30, 2006.21 The extracted data were arrayed by age group (25–34 years, 35–44 years) and calendar quarter. Quarterly estimates by age of the population of Australian women were obtained from the Australian Bureau of Statistics.21
Medicare data comprehensively capture imaging, biopsies and excisions among women under 40 years of age. Women who are 40 years and older, however, have another option: they may obtain these services through BreastScreen, which would not be observable in Medicare data. Consequently, our analysis only partially captures the service use of women over 40 in the 35–44 year age group. The potential implications of this issue for our analysis are addressed subsequently.
Procedure rates
We calculated quarterly rates for each of the services of interest. Imaging rates were calculated at the population-level, by dividing the sum of bilateral breast mammograms and bilateral breast ultrasound procedures by the number of Australian women in the relevant age group. In addition to this overall imaging rate, we calculated subrates for bilateral ultrasound and bilateral mammography, which were also denominated by the population.
The rates of image-guided biopsy (hereafter, ‘biopsy’) and cancer excision were both calculated as conditional rates, denominated by the number of ‘eligible’ women from the preceding stage. The biopsy rate was the number of biopsies divided by the total number of bilateral breast imaging procedures. The cancer excision rate was the number of excisions divided by the number of biopsies.
Analysis
We used logistic regression analysis to test for temporal changes in imaging rates. Rates in the four successive quarters following the Minogue publicity (July 1, 2005 to June 30, 2006) were compared with rates during the 6-month period leading up to the event (October 1, 2004 to March 31, 2005). We excluded the quarter in which media coverage of Minogue's diagnosis began (April 1 to June 30, 2005) because it began mid-quarter. The same approach was used to test for changes in rates of biopsy and cancer excisions. All analyses were conducted separately for women aged 25–34 years and women aged 35–44 years.
Previous research has suggested increasing rates of breast imaging over time.22 To examine whether any changes observed in this study may have been amplified by secular changes in imaging rates, we used polynomial trend contrasts to test for trends by quarter during the four quarters preceding the Minogue publicity (first quarter 2004–first quarter 2005).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionACKNOWLEDGEMENTSREFERENCES |
|---|
Imaging
There was a sharp increase in rates of doctor-referred breast imaging in Australia among women aged 25–44 years following media coverage of Kylie Minogue's diagnosis with breast cancer (Figure 2). Service use peaked in the first and second quarters after the first media reports, and then began to decline in the third and fourth quarters, although it did not settle back to earlier levels by the end of the study period. These trends were evident in both the younger and older age groups.
|
In the two quarters preceding the publicity, imaging rates averaged 8/1000 women per quarter in the 25- to 34-year-old age group and 21/1000 women per quarter in the 35- to 44-year-old age group. In the two quarters after the publicity, the corresponding rates were 12 and 28. These represent increases of 33 and 25%, respectively.
Although the percentage increase in imaging volume was larger for the younger age group than the older age group, the latter underwent more imaging. The younger age group accounted for an average of 2332 imaging procedures per quarter in two pre-event quarters and 3411 in the two post-event quarters. The older age group accounted for an average of 5411 imaging procedures per quarter in the same pre-event quarters and 7464 in the in the same post-event quarters.
There was no evidence of increasing linear trends (odds ratio 0.96, 95% CI 0.95–0.97) or quadratic trends (odds ratio 0.92, 95% CI 0.91–0.93) in breast imaging rates in the six quarters prior to the Minogue publicity.
The odds ratios for changes in imaging rates are shown in Table 1. Compared with the 6 months before publicity of the Minogue diagnosis, the odds that a 25- to 44-year-old woman would undergo breast imaging in the two quarters afterwards increased by 
20%. The increase was largest for mammography among women aged 25–34 years in the first post-quarter (odds ratio 1.33, 95% CI 1.30–1.37) and second post-quarter (odds ratio 1.26, 95% CI 1.22–1.30).
|
Biopsies and cancer excisions
The absolute number of breast biopsies increased in both age groups following the Minogue publicity, and did not return to earlier levels over the following year. Among women aged 25–34 years, there was no short-term change in the biopsy rate, measured as a proportion of imaging procedures, essentially because increases in the volume of biopsies kept pace with increases in imaging procedures. Among women aged 35–44 years there was a small decrease in the biopsy rate in the second quarter after the publicity (Table 2).
|
There was no clear trend in the absolute number of cancer excisions during the period of observation. However, both age groups experienced a large decrease in the proportion of biopsies that led to excisions (Table 2). In other words, there was a much lower rate of breast cancer detection within the larger volume of biopsies that followed the Minogue publicity. The decrease was especially pronounced for 25- to 34-year-old women, with the odds that biopsy led to a diagnosis of a cancer decreasing 31 and 41% in the first and second post-quarters, respectively; for 35- to 44-year-olds, the odds decreased 17 and 21% in the first and second post-quarters, respectively. Finally, we checked for long-term trends in excisions and found no increase the 2 years following the Minogue publicity.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionACKNOWLEDGEMENTSREFERENCES |
|---|
This study found large increases in the frequency of bilateral breast imaging procedures among women in Australia aged 25–44 years following the intense media coverage of Kylie Minogue's diagnosis of breast cancer in mid-2005. The frequency of breast biopsies also increased. However, the number of confirmed breast cancers necessitating surgical excision changed little and their rate, calculated as a proportion of biopsies completed, dropped fairly dramatically.
Because we relied on aggregate procedure counts, not patient-level analyses sensitive to the clinical aspects of individual cases, we cannot draw definitive conclusions about the progression of individual women through the stages of the diagnostic process. In addition, our research approach—simple juxtaposition of these aggregates—does not allow solid estimates of ‘excess’ procedure use or how many additional breast cancer diagnoses were attributable to the Minogue publicity. Nonetheless, the findings raise serious questions about whether the thousands of additional imaging procedures and biopsies ordered in Australia in the wake of Minogue's highly publicized diagnosis resulted in improved breast cancer detection among young women.
More generally, our findings reinforce existing evidence that both patients and doctors are susceptible to behaviour change in response to celebrity health events.6 The consumer phenomenon has been reasonably well documented.1,3,5,6,9–12,14 Its impact on clinical decision making is less well understood.
Minimal change in the frequency of excisions, despite substantial spikes in screening activity point to two possible upstream developments: (i) different casemix in the introduced population and (ii) changes in clinical decision making about referral for imaging. We did not observe casemix and so cannot assess the clinical appropriateness of referrals for mammography and ultrasounds. The fact that the proportion of imaging procedures that led to biopsies did not decrease, even increased slightly, implies some constancy in the proportion of patients with troubling or ambiguous symptoms at initial presentation. However, casemix changes are implausible as a full explanation. The reduced positive predictive value of image-guided biopsy suggests that the Minogue publicity, the influx itself or both, also lowered doctors’ threshold for referral to breast imaging services. The changed environment may have raised doctors own perceptions of breast cancer risk among young women; it may also have elevated their apprehension, including medico-legal fears, about missing breast cancer diagnoses in this population. These concerns emphasize the need to increase awareness of evidence-based guidelines among clinicians, both in response to current events that may implicate them and more generally.
In the furore that surrounds a celebrity illness, holding the line on recommended care may be difficult. Chapman and colleagues’3 analysis of the media surrounding the Minogue diagnosis found that one in six articles reporting on Minogue's illness criticized the government's decision not to allow women aged under 40 years open access to screening,3 a sentiment echoed by the shadow health minister.3 The coverage appears to have included very little discussion of the evidence basis for current screening policy, which may have further undermined adherence.3 Add the demands of anxious patients, and the pressures on doctors to screen may be considerable.
Patients must weigh the risk of a late cancer diagnosis against the time, discomfort, exposure to radiation, potential for false positives associated with an unnecessary investigation. Doctors, on the other hand, from a personal standpoint, face mostly downside risk if patients’ requests for screening are denied. At a health system level, there is a potentially serious organization and cost issue: an influx of lower risk women may reduce the capacity of services to deal with higher risk women, particularly if this occurs suddenly and there is no time to plan a response.
There are a number of ways celebrities, the media, clinical experts and public health agencies could work more closely together to ‘manage the message’. Although public health agencies may be taken by surprise, the initial releases are often carefully orchestrated. Consultation between a celebrity's public relations team and a public health agency on how to shape and disseminate the release could help create a message with the best chances of furthering quality-of-care and sound public health practice. In addition to ensuring the information reaches its target, care should be taken to avoid stirring inappropriate demand.
The study has several limitations in addition to those already mentioned. First, our analyses show associations between the Minogue publicity and increased use of screening services. The relationship may not be causal, but the intensity of media coverage, the timing and magnitude of the changes and the lack of an upward trend prior to the publicity all lend weight to a causal assumption. Second, the increase in demand for imaging was greatest in the first two quarters after the announcement of Kylie Minogue's diagnosis and subsequently subsided. A small part of the short-term effect may have been due to women moving forward previously scheduled appointments. Third, delays in the filing of Medicare claims may have affected some of the quarterly rates, although the expected effect of the resulting misclassification of procedure times would be to dissipate the temporal effect observed. Fourth, these data only capture bilateral breast imaging, making our estimates a lower bound on the true impact. Fifth, because of the BreastScreen option for women 40 years and older, we only partially observed the service usage of women in the 35–44 year age group. It is unlikely that the Minogue publicity substantially altered the ratio of 40- to 44-year-olds obtaining screening inside and outside BreastScreen in a manner that would undercut our main findings. On the contrary, available evidence suggests a substantial hike in BreastScreen bookings following the Minogue publicity3 that would complement the rise in procedure use we saw outside Breastscreen, which would again tend our results toward an underestimate of the publicity's overall impact on screening rates. Finally, these results relate to an extremely high profile diagnosis in a generally low-risk population. They may not, therefore, be generalizable to other celebrity diagnoses.
The visibility of a celebrity's illness presents an opportunity to improve public health; it also has the potential to undercut the appropriateness and cost-effectiveness of service delivery. A free press rejects central control of how stories are reported. However, responsible journalism, coupled with releases of the information that help promote evidence-based care, will help maximize the chances of a productive impact. Since celebrities often cite a desire to help others as their motivation for going public, strong partnerships between the media, public health experts and celebrities’ public relations teams should represent a win for all sides.
|
ACKNOWLEDGEMENTS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionACKNOWLEDGEMENTSREFERENCES |
|---|
These analyses are based on data provided by Medicare Australia. We would also like to thank Lyle Gurrin for his statistical advice. M.K. is supported by an Australian National Health and Medical Research Council Career Development award and the Victorian Health Promotion Foundation. D.S. is supported by Federation Fellowship from the Australian Research Council.
Conflict of interest: None declared.
KEY MESSAGES
|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionACKNOWLEDGEMENTSREFERENCES |
|---|
1 Boudioni M, Mossman J, Jones AL, Leydon G, McPherson K. Celebrity's death from cancer resulted in increased calls to CancerBACUP. Br Med J (1998) 317:1016.
2 Brown ML, Potosky AL. The presidential effect – the public-health response to media coverage about Reagan, Ronald Colon Cancer Episode. Public Opin Quart (1990) 54:317–29.
3 Chapman S, McLeod K, Wakefield M, Holding S. Impact of news of celebrity illness on breast cancer screening: Kylie Minogue's breast cancer diagnosis. Med J Aust (2005) 183:247–50.[Web of Science][Medline]
4 Clarke JN, Binns J. The portrayal of heart disease in mass print magazines, 1991–2001. Health Commun (2006) 19:39–48.[CrossRef][Web of Science][Medline]
5 Cram P, Fendrick AM, Inadomi J, Cowen ME, Carpenter D, Vijan S. The impact of a celebrity promotional campaign on the use of colon cancer screening – The Katie Couric effect. Arch Intern Med (2003) 163:1601–05.
6 Du X, Freeman DH Jr, Syblik DA. What drove changes in the use of breast conserving surgery since the early 1980s? The role of the clinical trial, celebrity action and an NIH consensus statement. Breast Cancer Res Treat (2000) 62:71–79.[CrossRef][Web of Science][Medline]
7 Ernst E, Pittler MH. Celebrity-based medicine. Med J Aust (2006) 185:680–81.[Web of Science][Medline]
8 Howe A, Owen-Smith V, Richardson J. The impact of a television soap opera on the NHS Cervical Screening Programme in the North West of England. J Public Health Med (2002) 24:299–304.
9 Kalichman SC, Russell RL, Hunter TL, Sarwer DB. Johnsons, Earvin, Magic Hiv Serostatus Disclosure – effects on mens perceptions of aids. J Consult Clin Psychol (1993) 61:887–91.[CrossRef][Web of Science][Medline]
10 Lane DS, Polednak AP, Burg MA. The impact of media coverage of reagan, nancy experience on breast-cancer screening. Am J Public Health (1989) 79:1551–52.
11 Larson RJ, Woloshin S, Schwartz LM, Welch HG. Celebrity endorsements of cancer screening. J Natl Cancer Inst (2005) 97:693–95.
12 Nattinger AB, Hoffmann RG, Howell-Pelz A, Goodwin JS. Effect of Nancy Reagan's mastectomy on choice of surgery for breast cancer by US women. JAMA (1998) 279:762–66.
13 Stein K, Lewendon G, Jenkins R, Davis C. Improving uptake of cervical cancer screening in women with prolonged history of non-attendance for screening: a randomized trial of enhanced invitation methods. J Med Screen (2005) 12:185–89.
14 Wiggins MA. Celebrity and community spirit combine for successful campaign. Healthtexas (1989) 45:17.[Medline]
15 Yip PSF, Fu KW, Yang KCT, et al. The effects of a celebrity suicide on suicide rates in Hong Kong. J Affect Disord (2006) 93:245–52.[CrossRef][Web of Science][Medline]
16 National Breast Cancer Centre. In: Clinical Practice Guildines for the Support and Management of Young Women with Breast Cancer. (2004) Sydney: National Breast Cancer Centre.
17 BreastScreen Victoria. BreastScreen assessment procedures. (2007) (Accessed on March 2007). Available at: http://wwwbreastscreenorgau/Bsassessmenthtm,.
18 National Breast Cancer Centre. In: National Breast Cancer Centre Breast Survey Techincal Report. (2005) Sydney: National Breast Cancer Centre.
19 National Breast Cancer Centre. In: Breast Imaging: A Guide for Practice. (2002) Sydney: National Breast Cancer Centre.
20 International Agency for Research on Cancer. In: IARC Working Group on the Evaluation of Cancer Prevention Strategies. (2003) Lyon: World Health Organization.
21 Australian Bureau of Statistics. In: Population Projections, Australia, 2004 to 2101. Catalogue 3222.0. (2006) Canberra: Australian Bureau of Statistics.
22 Kelaher M, Stellman JM. The impact of medicare funding on the use of mammography among older women: implications for improving access to screening. Prev Med (2000) 31:658–64.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
Related articles in Int. J. Epidemiol.:
- A temporal decline in asthma but not eczema prevalence from 2000 to 2005 at school entry in the Australian Capital Territory with further consideration of country of birth
- Anne-Louise Ponsonby, Nicholas Glasgow, Angela Pezic, Terence Dwyer, Karen Ciszek, and Marjan Kljakovic
Int. J. Epidemiol. 2008 37: 559-569.[Abstract] [FREE Full Text]
This article has been cited by other articles:
|
|
 |
|
  S. Ebrahim Media hype: good or bad for patients and the health care system? Int. J. Epidemiol., December 1, 2008; 37(6): 1199 - 1200. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||