Abstract

Cancer education commonly focuses on communicating objective risk information. Because education and behavior change have the potential to reduce the incidence of colon cancer, knowledge of colon cancer risk factors is integral to individual risk reduction efforts. The purpose of this study was to evaluate a 16-item questionnaire devised to assess knowledge of colon cancer risk factors. Items were developed based on risk factors described in the literature. Respondents were asked to indicate whether each of 16 items increase, decrease, or is unrelated to risk. A "don't know" option was provided. Items were scored as correct or incorrect with higher scores indicating more accurate knowledge. Data were collected from 609 female participants in an ovarian cancer screening program, 102 matched controls, and 98 male spouses of the screening group. Participants were predominantly Caucasian (98%), married (66%), high school educated (38%), and were an average age of 60 years (SD = 11). Results of principle components analysis suggested a three-component solution explaining 55% of the variance in scores. The three components included: (1) familial/health risk factors; (2) behavioral risk factors; and (3) factors unrelated to risk. Estimates of internal consistency for the three subscales were reasonable for dichotomous items (alpha's .89, .60, and .76, respectively). Respondents were comparatively more knowledgeable of familial/health risks for colon cancer than behavioral or unrelated risks. Data suggest that this measure may be useful in assessing general colon cancer risk knowledge and evaluating the efficacy of colon cancer risk education.

Introduction

Study Aims
• The primary aim of the current investigation is to evaluate a 16-item questionnaire devised to assess knowledge of colon cancer risk factors.
• The secondary aim is to assess the accuracy of respondents’ knowledge of ovarian cancer risk factors.

Method

Study Design
• Sample
– 609 women prior to transvaginal ultrasound screening
– 102 sex, age and education matched controls (T1,T2)
– 98 male spouses of screening group members
• Questionnaire Survey
– T1: 25-45 minute initial survey administration
– T2 (controls only): 10-20 minute telephone follow-up

Sample Description
Screening Group (n = 609)
Age (Mean Years): 61.2 ± 10.4
Education (Mean Years): 13.7 ± 2.7
% Married: 69
% Caucasian: 98
% Employed: 34

Control Group (n = 102)
Age (Mean Years): 53.5 ± 14.1
Education (Mean Years): 15.2 ± 2.6
% Married: 54
% Caucasian: 94
% Employed: 62

Male Group (n = 98)
Age (Mean Years): 61.8 ± 10.7
Education (Mean Years): 14.3 ± 3.5
% Married: 99
% Caucasian: 100
% Employed: 33

Description of Measure
• 16 items assessed knowledge of colon cancer risk factors based on the epidemiological literature.
• Respondents were provided 4 response options: (1) increase risk; (2) decrease risk; (3) unrelated
to risk; or (4) don’t know.
• Items scored dichotomously (correct vs. incorrect).

Analyses
• Exploratory Factor Analysis
– Principle Components Analysis
– Varimax Rotation
• Reliability of Derived Subscales
– Internal Consistency (Alpha)
– Test - Retest (1 month interval)
• Accuracy of Risk Factor Knowledge

Results

Exploratory Factor Analysis
Colon Cancer Risk Factor Knowledge Test

Factors Loadings

Factor 1: Familial/Health
(item 2) Father w/ colon Ca: .88
(item 5) Personal Hx of polyps: .66
(item 6) Personal Hx of IBD: .55
(item 7) Mother w/ colon Ca: .84
(item 9) Grandfather w/ colon Ca: .84
(item 16) Brother w/ colon Ca: .87

Variance Explained = 24.0%

Factor 2: Behavioral

(item 1) High fiber, low fat diet: .57
(item 8) Smoking cigarettes: .61
(item 11) Not eating produce: .65
(item 15) Vigorous exercise: .73

Variance Explained = 17.4%

Factor 3: Unrelated

(item 3) Being Caucasian: .74
(item 4) Sister w/ skin Ca: .66
(item 10) Vitamin C use: .53
(item 12) Mother w/ lung Ca: .63
(item 13) Alcohol consumption: .69
(item 14) Being Non-Caucasian: .78

Variance Explained = 13.1%

Total Variance Explained = 54.6%

Results:

(1) Results demonstrate a three factor structure
(2) Each item loaded on only one factor
(3) All 16 items were retained in the analysis

Factor Reliability
Time 1 Time 2
Factor Alpha Alpha
Factor: Familial/Health (6 items)
Time 1 Alpha: .89
Time 2 Alpha: .91
1-Month Test-Retest: .52

Factor: Behavioral (4 items)
Time 1 Alpha: .60
Time 2 Alpha: .48
1-Month Test-Retest: .53

Factor: Unrelated (6 items)
Time 1 Alpha: .76
Time 2 Alpha: .56
1-Month Test-Retest: .43

Overall (16 items)
Time 1 Alpha: .81
Time 2 Alpha: .72
1-Month Test-Retest: .56

Mean Correct Responses

Familial/Health (6 items)
Time 1 Screening Group Correct: 4.93
Time 1 Control Group Correct: 5.41
Time 1 Male Group Correct: 4.50

Behavioral (4 items)
Time 1 Screening Group Correct: 2.64
Time 1 Control Group Correct: 2.87
Time 1 Male Group Correct: 2.55

Unrelated (6 items)
Time 1 Screening Group Correct: 1.43
Time 1 Control Group Correct: 1.54
Time 1 Male Group Correct: 1.68

Overall (16 items)
Time 1 Screening Group Correct: 9.00
Time 1 Control Group Correct: 9.82
Time 1 Male Group Correct: 8.73

Conclusions

• The Colon Cancer Risk Factor Knowledge Test was designed to assess awareness of risk factors for colon cancer.
• Factor analysis resulted in a 3-factor structure including familial/health risks, behavioral risks, and unrelated risks factors.
• Respondents maintained more accurate information about familial/health risk factors than behavioral or unrelated risk factors for colon cancer.
• This scale may be helpful in evaluating the efficacy of cancer risk communication interventions or exploring the impact of risk factor knowledge on adherence to cancer screening.