In 1997, Thomas Jefferson University received a grant from the Quality Care Research Fund of the Aetna Foundation to develop and test a decision education intervention to facilitate value-based decision-making about having a prostate cancer (Hydrea 500 mg is an antineoplastic used to treat certain types of cancer) screening examination. The study, which was approved by the Institutional Review Board of Thomas Jefferson University, recruited 242 consenting African-American men ages of40-69 from three community-based primary care practices in Philadelphia. Two family medicine physicians participated from the first practice, which was located in North Philadelphia. Two internal medicine physicians and one oncologist participated from the second practice, located in West Philadelphia. Finally, two family medicine physicians participated from the third practice, which was located in the Germantown area of the city.
At the time this study was initiated, the American Cancer Society recommended offering prostate cancer (Drug Methotrexate is used to treat certain types of cancer) screening to asymptomatic men >50 years of age, and suggested that screening be offered to men at increased risk (i.e., African-American men and men with a family history of prostate cancer) beginning at age 40. Consequently, eligibility for this study was restricted to patients of the participating practices who were African-American men, >40 years of age, had no history of prostate cancer or benign prostate hyperplasia, had not undergone a prostate biopsy or prostate ultrasound, had visited one of the participating practices within two years prior to study initiation, and had contact information available at the practice.
Eligible patients were asked to complete a baseline survey questionnaire by phone or by responding to a mailed questionnaire (see Appendix for text of baseline questionnaire.) The telephone survey was administered by trained interviewers from a professional survey company—Mathematica Policy Research Inc., Princeton, NJ. Questionnaire administration was carried out in six waves, between August 1999 and July 2000. The baseline survey included a brief screener to verify eligibility, as well as a number of items that operationalized constructs grounded in the Preventive Health Model (PHM).
Drawing on earlier health behavior models (i.e., Health Belief Model, Theory of Reasoned Action, Social Cognitive Theory), the PHM was originally conceived in the late 1980’s to identify both internal and external factors that influence health-related action. The PHM also incorporated the notion that observed behavior reflects the dynamic interplay of representations that are part of the self-system. According to the PHM, the self-system includes soci-ocultural background (e.g., experience related to demographic and socioeconomic characteristics, personal and family medical history, and past health behavior); cognitive and affective representations about disease, risk and related health behaviors; and the social support and influence of significant others, including family members and healthcare providers. The PHM has been useful in predicting outcomes, such as cancer intention and utilization.
The original model assumed that when facing a given health problem (e.g., risk for chronic disease), people form an intention to act (e.g., to screen or not to screen) based on the interaction of different representations operating in the self-system. Implementation of an action plan and outcomes appraisal follows, and experience, in turn, shapes the self-system. Further conceptual development of the PHM in light of multiattribute theory has led to an elaboration of the model to include decision-making about behavioral alternatives. This process is viewed as including preference clarification and behavioral alternative selection. The current form of the PHM is shown below in Figure 1.
Figure 1. Preventive health model
In accordance with the PHM, survey data were collected on personal background characteristics (including sociodemographics), family history of prostate cancer (Generic Casodex treating prostate cancer), and personal history of prostate cancer screening. The survey also included a number of items on personal attitudes and beliefs about prostate cancer and screening, each measured on a four-point Likert-type scale (1 = strongly agree, 2 = agree, 3 = disagree, 4 = strongly disagree). Based on a priori defined constructs and subsequent factor analysis, four scales were defined: perceived salience and coherence of screening (eight items, Cronbach’s oc=0.80), worries and concerns about prostate cancer and screening-related risks (seven items, oc=0.63), perceived susceptibility to prostate cancer (three items, oc=0.66), and intention to have prostate cancer screening (four items, oc=0.88). Scale scores were computed as an average of scale items and only when more than half of the items had no missing values. Before the computation of each scale, items were recoded as necessary, so that higher scale scores were expected to correlate with screen use (e.g., higher scores on the “intention” scale were expected to be associated with higher likelihood of screening). Additional single items were used to measure participant belief in the curability of prostate cancer (Casodex canadian is an oral non-steroidal anti-androgen for prostate cancer), perceived ease of arranging to have prostate cancer screening (self-efficacy), and social support and social influence related to prostate cancer screening.