Marilyn M. Schapira

Generalizability of the surveillance, epidemiology, and end results registry population: Factors relevant to epidemiologic and health care research

To assess the generalizability of the population included in the Surveillance, Epidemiology, and End Results (SEER) tumor registries to the overall United States population, we compared the population of the 198 SEER counties to the population of the 2882 non-SEER counties regarding sociodemographic factors, physician availability, and availability of pertinent hospital resources. The population residing within the SEER areas is more affluent, has lower unemployment, and is substantially more urban than the remainder of the U.S. population (p < 0.001 for each). The SEER areas have fewer general and family practice physicians, but more total nonfederal physicians, general internists, and specialists relevant to cancer care. SEER areas have fewer Joint Commission on Accreditation of Hospitals accredited hospitals, hospital beds, and hospitals with CT scanners, but more hospitals with bone marrow transplantation. The differences between the SEER population and the remainder of the United States, especially SEERs higher socioeconomic status and more urban population, should be considered when generalizing from SEER to the entire country.

Frequency or Probability? A Qualitative Study of Risk Communication Formats Used in Health Care

Background . The communication of probabilistic outcomes is an essential aspect of shared medical decision making. Methods . The authors conducted a qualitative study using focus groups to evaluate the response of women to various formats used in the communication of breast cancer risk. Findings . Graphic discrete frequency formats using highlighted human figures had greater salience than continuous probability formats using bar graphs. Potential biases in the estimation of risk magnitude were associated with the use of highlighted human figures versus bar graphs and the denominator size in graphics using highlighted human figures. The presentation of uncertainty associated with risk estimates caused some to loose trust in the information, whereas others were accepting of uncertainty in scientific data. Conclusion . The qualitative study identified new constructs with regard to how patients process probabilistic information. Further research in the clinical setting is needed to provide a theoretical justification for the format used when presenting risk information to patients.

Relation between appropriateness of primary therapy for earlystage breast carcinoma and increased use of breast-conserving surgery

BACKGROUND Breast-conserving surgery is a more complex treatment than mastectomy, because a separate incision is needed for axillary lymph-node dissection, and postoperative radiotherapy is necessary. We postulated that adoption of this therapy into clinical practice might have led to discrepancies between the care recommended and that received. METHODS We used records of the US national Surveillance, Epidemiology, and End Results tumour registry to study 144,759 women aged 30 years and older who underwent surgery for early-stage breast cancer between 1983 and 1995. We calculated the proportion undergoing at least the minimum appropriate primary treatment (defined, in accordance with the recommendations of a National Institutes of Health Consensus Conference in 1990, as total mastectomy with axillary node dissection or breast-conserving surgery with axillary node dissection and radiotherapy) during each 3-month period. FINDINGS The proportion of women receiving appropriate primary therapy fell from 88% in 1983-89 to 78% by the end of 1995. This decline was observed in all subgroups of age, race, stage, and population density. Of all women in the cohort, the proportion undergoing an inappropriate form of mastectomy remained stable at about 2.7% throughout the study period. The proportion undergoing an inappropriate form of breast-conserving surgery (omission of radiotherapy, axillary node dissection, or both) increased from 10% in 1989 to 19% at the end of 1995. INTERPRETATION Although most women undergo appropriate care, the appropriateness of care for early-stage breast cancer in the USA declined from 1990 to 1995. Because the proportion of all women who were treated by breast-conserving surgery increased, and because this approach was more likely than was mastectomy to be applied inappropriately, the proportion of all women having inappropriate care increased.

Underutilization of mammography in older breast cancer survivors.

BACKGROUND Annual mammography is recommended for all breast cancer survivors. OBJECTIVES To elucidate mammography use among older survivors of breast cancer and to explore determinants of such use. RESEARCH DESIGN Retrospective cohort study using data from the Surveillance, Epidemiology, and End Results (SEER) registry linked to Medicare claims. SUBJECTS A cohort of 3885 breast cancer survivors aged > or =65 years diagnosed with early-stage breast cancer in the United States in 1991. MEASURES Medicare mammogram claims during the 2-year period following initial breast cancer treatment. RESULTS Overall, 62% of the cohort underwent annual mammography, 23% underwent mammography in 1 of 2 years, and 15% had no mammography claim in the 2 years evaluated. Twenty-two percent of the women who underwent breast-conserving surgery (BCS) without radiotherapy had no mammogram in the 2-year period evaluated, compared with 17% of those who underwent mastectomy and 4% of those who underwent BCS with radiotherapy. In multivariate analyses controlling for age, cancer stage, and other patient factors, the use of annual mammography was significantly lower among women treated with mastectomy or BCS without radiotherapy than among women treated with BCS with radiotherapy. CONCLUSIONS Mammography is underused in the follow-up care of older breast cancer survivors. Underuse is of particular concern in women treated with BCS without radiotherapy because of the high risk of local disease recurrence. It is unknown whether poorer follow-up care contributes to the previously described lower rate of long-term survival among women who received this therapy.

Effect of treatment on quality of life among men with clinically localized prostate cancer.

Background.Quality-of-life outcomes are an important consideration for patients evaluating therapeutic options for localized prostate cancer. Objectives.The objective of this study was to describe the effect of treatment choice on change in health-related quality of life (HRQOL) among men with clinically localized prostate cancer. Research Design.This was a prospective observational study. Subjects.The study subjects were 122 men with clinically localized adenocarcinoma of the prostate. Forty-two subjects (34%) underwent radical prostatectomy, 51 (42%) underwent radiation therapy, and 29 (24%) were followed with e-pectant management. Measures.The University of California at Los Angeles Prostate Cancer Quality of Life Inde- and the Medical Outcomes Study Short Form-36 were administered before and 3 and 12 months after initial treatment. The study used an analysis of covariance model adjusted for baseline differences in clinical and demographic factors. Results.Men who underwent radical prostatectomy e-perienced significant declines in urinary and se-ual function and bother that persisted at 12 months after treatment. Men treated with radiation therapy e-perienced smaller but significant declines in se-ual function and a decline in social function. E-pectant management patients did not have a significant change in disease-targeted or generic HRQOL domains. Differential rates of change in urinary and se-ual function between treatment groups persisted after adjustment for differences in pretreatment clinical and demographic factors. Conclusions.Men undergoing radical prostatectomy have substantial declines in urinary and se-ual function, and men undergoing radiotherapy have declines in se-ual function. Men undergoing e-pectant management have no change in disease-specific or general HRQOL in the first year after treatment.

Treatment of localized prostate cancer in African-American compared with Caucasian men. Less use of aggressive therapy for comparable disease.

The objective of this study is to evaluate differences in treatment for localized prostate cancer in African-American compared with Caucasian men. A cohort of patients was identified from the 1988–1989 Surveillance, Epidemiology, and End Result data base. Data were linked by county of residence to the Area Resource File. The main outcome measures were (1) the receipt of aggressive therapy (radical prostatectomy or external beam radiation) and (2) the treatment modality (radical prostatectomy or external beam radiation). Of 5,786 subjects, 15% were treated with a radical prostatectomy, 29% received external beam radiation, and 56% did not receive aggressive therapy. African Americans were less likely (relative risk 0.46; 95% confidence interval 0.39, 0.54) than Caucasians to receive aggressive therapy. Among those receiving aggressive therapy, African Americans were less likely (relative risk 0.64; 95% confidence interval 0.48, 0.86) than Caucasians to have a radical prostatectomy. Adjustment for patient, disease, and community health care availability characteristics did not change these results. We conclude that African-American and Caucasian men receive different treatments for localized prostate cancer. The reasons for the differences, and their ultimate effect on mortality and quality of life require further study.

The influence of graphic format on breast cancer risk communication.

Graphic displays can enhance quantitative risk communication. However, empiric data regarding the effect of graphic format on risk perception is lacking. We evaluate the effect of graphic format elements on perceptions of risk magnitude and perceived truth of data. Preferences for format also were assessed. Participants (254 female primary care patients) viewed a series of hypothetical risk communications regarding the lifetime risk of breast cancer. Identical numeric risk information was presented using different graphic formats. Risk was perceived to be of lower magnitude when communicated with a bar graph as compared with a pictorial display (p < 0.0001), or with consecutively versus randomly highlighted symbols in a pictorial display (p = 0.0001). Data were perceived to be more true when presented with random versus consecutive highlights in a pictorial display (p < 0.01). A pictorial display was preferred to a bar graph format for the presentation of breast cancer risk estimates alone (p = 0.001). When considering breast cancer risk in comparison to heart disease, stroke, and osteoporosis, however, bar graphs were preferred pictorial displays (p < 0.001). In conclusion, elements of graphic format used to convey quantitative risk information effects key domains of risk perception. One must be cognizant of these effects when designing risk communication strategies.

Predictors of Pessimistic Breast Cancer Risk Perceptions in a Primary Care Population

AbstractOBJECTIVE: To identify sociodemographic characteristics, numeracy level, and breast cancer risk factors that are independently associated with the accuracy of lifetime and 5-year breast cancer risk perceptions. DESIGN: Cross-sectional survey. A probability scale was used to measure lifetime and 5-year risk perceptions. The absolute difference between perceived risk and the Gail model risk of breast cancer was calculated. Linear regression models were built to predict lifetime and 5-year breast cancer risk estimation error. SETTING: Primary care internal medicine practices (N=2). PARTICIPANTS: Two hundred fifty-four women 40 to 85 years of age. RESULTS: The mean lifetime and 5-year calculated breast cancer risk was 8.4% (SD [standard deviation] 6.1) and 1.5% (SD 1.3), respectively. Subjects had a mean estimation error for lifetime and 5-year risk of 29.5% (SD 22.9) and 24.8% (SD 23.9), respectively. In multivariate analyses, lower numeracy scores (0.005), higher number of previous breast biopsies (0.016), and a higher number of first-degree relatives (0.054) were predictive of larger estimation error for lifetime breast cancer risk. White race (0.014), lower educational levels (0.009), higher number of previous breast biopsies (0.008), and higher number of first-degree relatives (0.014) were predictive of larger estimation error for 5-year risk. CONCLUSION: Among a primary care population, breast cancer risk factors may be more consistently associated with pessimistic perceptions of breast cancer risk than other factors studied during a lifetime and 5-year time span. Primary care physicians should consider counseling patients about individual breast cancer risk factors and risk over time.

Enhanced decision-making: the use of a videotape decision-aid for patients with prostate cancer.

We describe the development and evaluation of a videotape decision-aid to assist patients in considering treatment options for clinically localized prostate cancer. The content and form of the videotape were developed utilizing literature review, an expert opinion panel and patient focus groups. Thirty-two men, aged 50-85 years, who did not have prostate cancer underwent pre- and post-videotape testing. A quantitative analysis found the videotape to be moderately effective in improving short-term recall of treatment options and outcomes. Qualitative analysis demonstrate that after viewing the videotape, subjects were more likely to consider an active role in the treatment decision-making process. We conclude that a videotape decision-aid will benefit clinical practice by conveying knowledge to patients regarding treatment options and outcomes and encouraging them to participate with their physicians in medical decision-making.

Breast cancer screening in an era of personalized regimens: A conceptual model and National Cancer Institute initiative for risk-based and preference-based approaches at a population level

Breast cancer screening holds a prominent place in public health, health care delivery, policy, and womens health care decisions. Several factors are driving shifts in how population‐based breast cancer screening is approached, including advanced imaging technologies, health system performance measures, health care reform, concern for “overdiagnosis,” and improved understanding of risk. Maximizing benefits while minimizing the harms of screening requires moving from a “1‐size‐fits‐all” guideline paradigm to more personalized strategies. A refined conceptual model for breast cancer screening is needed to align womens risks and preferences with screening regimens. A conceptual model of personalized breast cancer screening is presented herein that emphasizes key domains and transitions throughout the screening process, as well as multilevel perspectives. The key domains of screening awareness, detection, diagnosis, and treatment and survivorship are conceptualized to function at the level of the patient, provider, facility, health care system, and population/policy arena. Personalized breast cancer screening can be assessed across these domains with both process and outcome measures. Identifying, evaluating, and monitoring process measures in screening is a focus of a National Cancer Institute initiative entitled PROSPR (Population‐based Research Optimizing Screening through Personalized Regimens), which will provide generalizable evidence for a risk‐based model of breast cancer screening, The model presented builds on prior breast cancer screening models and may serve to identify new measures to optimize benefits‐to‐harms tradeoffs in population‐based screening, which is a timely goal in the era of health care reform. Cancer 2014;120:2955–2964.