Interpretation of prostate cancer screening events and outcomes: a focus group study

https://doi.org/10.1016/S0738-3991(02)00180-5Get rights and content

Abstract

Professional guidelines call for physicians to provide patients with information to permit informed decision making. We conducted focus groups to obtain reactions to numerical information about events and outcomes related to prostate cancer screening (prevalence, natural history, accuracy of screening, and treatment outcomes). The focus groups were used to help develop a balance sheet, a decision aid that explicitly compares likelihood and value of outcomes. In all, 90 persons participated in 12 focus groups homogeneous in ethnicity (African American, White, Hispanic) and gender. Discussions were transcribed and analyzed using qualitative methods. The view of screening derived by participants from the numerical information was less positive than initial opinions based on the media and confidence in medicine. Participants suggested shorter screening intervals, regular screening, and different treatment methods. No single topic was seen as sufficient for the screening decision. The balance sheet should cover prevalence and outcomes of screening and treatment.

Introduction

Prostate cancer screening is controversial because both society and individuals face the dilemma of uncertain benefits and large costs. Prostate cancer screening tests are not accurate, and it is not clear that prostate cancer screening will reduce mortality and morbidity from prostate cancer [1]. Treatments for prostate cancer are associated with significant complications having adverse quality of life consequences [2]. Professional guidelines for prostate cancer screening vary. The American Urological Association and American Cancer Society support screening for prostate cancer [3], [4]. The American Task Force for Preventive Services, American College of Physicians, American College of Preventive Medicine, and Academy of Family Physicians do not recommend it [5], [6], [7], [8]. Organizations on both sides of this question call for physicians to describe advantages and disadvantages of prostate cancer screening to permit informed decision making.

Those making the call for mass screening of asymptomatic persons must consider the accuracy of the test, the availability of effective treatments, the prevalence and seriousness of the condition, and costs [9]. Prostate cancer is sufficiently prevalent—the most common major cancer for men—and serious—second leading cause of cancer mortality in men—to justify screening [10]. In the United States, cancer incidence and mortality are higher for African American men than for Whites [11]. Characteristics of the screening test should also be considered. The two screening methods used are digital rectal exams (DRE) and prostate specific antigen (PSA) tests. The DRE requires subjective judgment and can assess only a limited portion of the prostate [12]. The PSA has higher sensitivity than the DRE, but has low predictive value and specificity [1]. So, many men without disease face additional invasive testing. Finally, prostate cancer screening would be clearly called for if it improved survival through identification of prostate cancer in an early stage. The contribution of early treatment to survival is under dispute in part because many prostate cancers are slow growing and will not threaten life. The complex interplay of positive and negative aspects of prostate cancer screening contributes to disagreement concerning guidelines for screening, but supports the importance of providing information to support individuals in making decisions about screening.

The informed decision maker is presumed to base action on the anticipated consequences of the actions, taking into account how likely it is that those consequences will happen. These value expectancy principles are embodied in many models of health behavior, for example in the Health Belief Model [13] and in the image of decisional balance or weighing of pros and cons for an action in the Transtheoretical Model [14]. To support patients’ informed decision making, physicians must provide information about advantages and disadvantages of the options, and the opportunity to integrate this information with patients’ personal values. Informed decision making is particularly important for decisions for which personal values vary and where decisions have serious or long lasting consequences.

We conducted this research to develop a balance sheet decision aid for prostate cancer screening. Decision aids may be developed for complex decisions, such as prostate cancer screening, in which the evidence for outcomes is uncertain or personal values may vary [15]. The balance sheet explicitly compares one or more courses of action (screening versus not screening) by providing information about how likely it is that good and bad outcomes would follow. It also provides information about how good or bad are the outcomes. When applied to health decisions, balance sheets are an approach to patient education. Balance sheets have been suggested as a way to present evidence for practice guidelines [16], [17]. Eddy [18], for example, outlined a proposed balance sheet for colorectal cancer screening in which the columns would correspond to the options (various screening modalities and not screening) and the rows contain the probabilities that beneficial or harmful outcomes would occur or the costs associated with options.

We conducted 12 focus groups to determine participant reactions to outcomes associated with prostate cancer screening associated events. The purpose of this paper is to summarize reactions to information about events and outcomes related to prostate cancer screening and to suggest ways to construct a decision aid for prostate cancer screening. The systematic and extensive use of numerical information about cancer screening is relatively novel in patient education. Thus, it is important to gain information about patient reactions to numerical information in making decisions. There are significant questions about how best to present such information. For example, benefits and risks may be presented in relative or absolute terms [19], and probabilistic information may be better understood when presented as frequencies rather than conditional probabilities [20].

Section snippets

Sample and recruitment

In all, 12 focus groups were conducted with 90 participants. Eligibility criteria included race/ethnicity (Hispanic, non-Hispanic White, and African American) and age (50–70 years for Hispanic and non-Hispanic White men and 40–70 years for African American men). A wider age range was used for African American men because they have higher prostate cancer incidence and mortality and some professional groups recommend initiation of prostate screening at the age of 40 years for them. We excluded

Characteristics of participants

Table 1 summarizes characteristics of participants by race/ethnicity and gender. There were 90 participants, 48 men and 42 women. Group size ranged from 4 to 11 members. The average age was in the 50 years for all groups (range 50.9–56.4 years). Education level varied by race/ethnicity, with Hispanic and African Americans having lower educational levels. About two-thirds of African American and non-Hispanic White males reported prior prostate cancer screening compared to only one-third of

Discussion

We conducted 12 focus groups to develop a balance sheet decision aid for prostate cancer screening. In this paper, we have examined reactions to numerical information related to screening events and outcomes. The discussions were organized around a structured sequence of topics identified by Chan and Sulmasy [22] as important for the prostate cancer screening decision. The topics were prevalence of prostate cancer, natural history, screening accuracy, and treatment benefits and complications.

Acknowledgements

The research was supported by the Centers for Disease Control and Association of Teachers of Preventive Medicine through Cooperative Agreement U50/CCU300860 TS-281. We also acknowledge the contributions of Keri Pierce, M.P.H. in arranging focus groups in Oklahoma City and coding data. Thanks to Robert J. Volk, Ph.D.; Juana Gonzales; and Armadina Garza who conducted the groups in Houston. The display materials were reviewed by Dewey C. Scheid, M.D.; James L. Mold, M.D.; and Donald Elmajian, M.D.

References (37)

  • U.S. Preventive Services Task Force. Guide to clinical preventive services. 2nd ed. Baltimore: Williams & Wilkins;...
  • Woolf SH, Jonas S, Lawrence RS, editors. Health promotion and disease prevention in clinical practice. Baltimore:...
  • R.T. Greenlee et al.

    Cancer statistics, 2000

    CA Cancer J. Clin.

    (2000)
  • American Cancer Society. Cancer facts and figures for African Americans, 1998–1999. American Cancer Society, Atlanta,...
  • A. Hoogendam et al.

    The diagnostic value of digital rectal examination in primary care screening for prostate cancer: a meta-analysis

    Fam. Pract.

    (1999)
  • Rosenstock IM. The health belief model: explaining health behavior through expectancies. In: Glanz K, Frances ML, Rimer...
  • Prochaska JO, Redding CA, Evers KE. The transtheoretical model and stages of change. In: Glanz K, Lewis FM, Rimer BK,...
  • A.M. O’Connor et al.

    Decision aids for patients facing health treatment or screening decisions: systematic review

    Br. Med. J.

    (1999)
  • Cited by (19)

    View all citing articles on Scopus
    View full text