Presented at the American Society of Clinical Oncology–Genitourinary Symposium, Orlando, Florida, February 2013.
Patients diagnosed with prostate cancer are often counseled about treatment options with the use of terms that are part of the “core vocabulary” of prostate cancer. It is hypothesized that predominantly lower literacy patients would demonstrate a severe lack of comprehension of prostate cancer terms, thus validating the findings of a previous single-institution study.
A previously developed survey was used to evaluate understanding of terms related to urinary, bowel, and sexual function. The survey was administered by trained evaluators at 2 safety net clinics that provide care for low-income, predominantly African American patients. Comprehension was assessed using semiqualitative methods coded by 2 independent investigators. Literacy and numeracy were also evaluated.
Among 109 patients who completed the study, only 5% understood the function of the prostate, and 15%, 29%, and 32% understood the terms “incontinence,” “urinary function,” and “bowel habits,” respectively. Lower levels of comprehension were observed for compound words, such as “vaginal intercourse” (58%), versus single words such as “intercourse” (95%), validating previous work. Median school level was 13 years, yet median literacy level was only ninth grade, and reading level was significantly correlated with comprehension. Only 30% of patients correctly calculated both a fraction and a percent.
Prostate cancer is the most common noncutaneous cancer among men in the United States and the second leading cause of cancer death. The death rate from prostate cancer is relatively low, with 30,000 annual deaths compared with more than 200,000 annual diagnoses. As such, the majority of men diagnosed with prostate cancer are expected to live with treatment-related adverse effects. Although prostate cancer treatment is lifesaving in some cases, a large, randomized study demonstrated that, when compared with watchful waiting, 15 men must be treated with radical prostatectomy to save 1 life. Other studies have demonstrated that the number needed to treat for radical prostatectomy can range from 13 to 42 men, depending on the tumor grade. Thus, selection of a particular therapeutic option should involve careful consideration of both the likelihood of cure and the probability of treatment-related side effects. This necessitates the principle of shared decision-making between the patient and health care provider in the treatment of prostate cancer.
According to the 2010 American Cancer Society guidelines on prostate cancer treatment, physicians should provide patients with the tools necessary to fully participate in informed decision-making. In addition, the National Comprehensive Cancer Network guidelines on prostate cancer state that the multiple aspects of prostate cancer treatment, including outcomes and adverse effects, must be considered by both the physician and the patient. The recent American Society of Clinical Oncology guidelines also stress the necessity of shared and informed decision-making in the screening of prostate cancer. Central to informed decision-making is full comprehension of the risks and benefits of a particular treatment option.
To explain the tradeoffs of different therapies, physicians often employ medical terminology that may or may not be fully understood by their patients. Kilbridge et al studied men attending general medicine clinics in Virginia and found that fewer than 50% understood the word “impotence” and a mere 5% understood the word “incontinence.” Lack of knowledge regarding prostate cancer has been associated with decisional regret, demonstrating the potential harm of such misunderstanding. In addition, lack of formal education has also been associated with decisional regret, with almost twice as many regretful men lacking a college education compared with nonregretful men. In the current study, we sought to independently assess the relation of socioeconomic status and education to the level of comprehension of commonly used prostate cancer terms among a predominantly low-literacy, African American population in a different setting. If the previously reported findings are confirmed, this would indicate that the lack of comprehension of medical terminology in patients with prostate cancer may be widespread, thus emphasizing the need for intervention.
MATERIALS AND METHODS
With Emory University institutional review board approval, we recruited a cohort of patients primarily from 2 safety net clinics—the Urology Clinic and the Radiation Oncology Center—at Grady Memorial Hospital, Atlanta, Ga. Standard written materials concerning prostate cancer and other genitourinary disease processes, including handouts and illustrated charts, were available in the waiting rooms of our clinics, but with a poor degree of patient utilization. Patients were deemed eligible if they were English-speaking men 40 years of age or older, were neither physicians nor registered nurses, and were able to give informed consent. We recruited eligible patients either in person from the waiting rooms or from flyers posted in the same clinics. All patients recruited from the flyers were screened to verify that the above criteria were met. Given the safety net nature of these clinics, patients were seen for a variety of disease types, including both cancer and noncancer diagnoses. We completed 109 interviews between January 25, 2012, and June 13, 2012.
Written informed consent was obtained from each patient after the interviewer reviewed the consent form with the patient and answered any questions. Interviews were conducted face-to-face with full privacy in empty examination rooms and read aloud. Because the initial recruitment rate was not as high as desired, we instituted $25 gift cards to compensate patients for their time. As such, 89 patients received compensation, whereas 20 received none.
We employed the questionnaire and analysis developed by Kilbridge et al. This questionnaire includes 28 technical terms describing urinary, bowel, and sexual function derived from patient education materials and health status questionnaires. The tool assessed patient comprehension of the 28 terms with a sequence of questions that consisted of a general format of: 1) “Is _____ a word that you know?” If the patient responded “yes,” 2 further questions were asked: 2) “What are the other words that you've heard for _____?” 3) “What happens when a man has _____?” In addition, we assessed patient numeracy with the validated Schwartz-Woloshin numeracy test, as well as with the stick figure numeracy test used previously. Knowledge of important anatomic structures was evaluated by asking patients to identify location and function, using 2 male anatomic drawings. The reading level of each patient was assessed with the Rapid Estimate of Adult Literacy in Medicine (REALM). The questionnaire then measured prostate cancer knowledge using questions previously validated in a large study of Canadian men. Finally, demographic information was obtained through interviewers asking patients to self-describe race and provide age, household income, highest grade completed at school, insurance type, living arrangement at home, church attendance, and ownership of common electronics (television, VCR, DVD, and computer), as was done by our group in a previous study on the use of multimedia devices and patient comprehension.
Four trained interviewers, comprising 2 females and 2 males, conducted the interviews. Each interviewer recorded patient responses in detailed written form. A Microsoft Access database was created to store all data, including raw images of patient-performed tasks. Two trained coders, working independently and blinded to each other, determined comprehension of the 28 terms. If a patient responded “no” to the question “Is ______ a word that you know?” he was scored as not understanding the term. Likewise, if a patient skipped a question because he was unsure of the meaning, he was also scored as not understanding the term. Patients who answered “yes” to the question were then scored as either correctly understanding or misinterpreting the term, depending on the patient's answers to the subsequent 2 questions. Scores were then unblinded, and the 2 coders examined any scoring disagreements to reach consensus. A third independent coder reviewed any continuing disagreements to provide a tie-breaking score.
Analyses were performed using SPSS Statistics, version 20.0, software (SPSS Inc., Chicago, Ill). Questions skipped by patients were excluded from analyses. Standard methods based on the binomial distribution were used to calculate confidence intervals for proportions. Spearman's correlation coefficients were used to estimate the association between literacy, as defined by the REALM score, and the proportion of correctly identified terms within each domain (sexual, urinary, and bowel). Fisher z-transformation was used to calculate confidence intervals for correlation coefficients.
Patient characteristics are summarized in Table 1. In total, we interviewed 114 eligible patients, of which 3 did not complete the entire interview, and 2 were identified as repeat interviewees. Patients were 55 years old on average (range, 40-86 years). Most (87%) self-identified their race as African American. The patients' median household income was $14,500, and the majority (84%) were either uninsured or had Medicaid. A total of 80% of patients completed high school or beyond, but only 43% read at a high school level. Only 30% of participants were able to calculate both a fraction and a percent. Fewer than 3% correctly answered all 3 items of the Schwartz-Woloshin numeracy test. Twenty-three men (21%) reported they had been previously diagnosed with prostate cancer, of which 17 reported they had received treatment.
Table 1. Patient Demographics and Study Characteristics (N = 109)
Patient must calculate both 30% of 10 and 30% of 100 correctly.
8.7% illiterate or ≤ third grade reading level, 47.8% can calculate both fraction and percent correctly.
Patient comprehension of medical terms related to prostate cancer is demonstrated in Figure 1. In the urinary function domain, the terms “urine” and “urination” were generally very well understood. However, the composite terms “urinary control” and “urinary hesitancy” were understood by only 51% and 28% of patients, respectively. “Risk factor” was another poorly understood term, with only 13% of patients showing comprehension. In the bowel function domain, approximately half of our patients understood the term “rectum,” but fewer than a fifth of our patients correctly defined the term “rectal urgency.” In the sexual function domain, “intercourse” was correctly understood by a large majority of the patients. The rate of comprehension dropped to fewer than 60% for the compound term “vaginal intercourse.” Comprehension of terms in all 3 functional domains, that is, urinary (P < .001), bowel (P < .001), and sexual (P < .001), was significantly correlated with literacy, measured by the REALM score (Table 2).
Table 2. Correlation Coefficients Between Literacy and Proportion of Correctly Understood Terms in Each Domain
95% Confidence interval
Qualitative analysis revealed that misunderstanding of prostate cancer terminology was associated with confusion of bowel, bladder, and sexual function. In other words, patients commonly provided a response from a different domain than the term in question. The urinary and bowel function domains were the most frequently confused, with 26% of our patients providing a urinary domain definition for a bowel domain term or vice versa. Another 8% of our patients confused the bowel and sexual domains, and 6% of our patients confused the urinary and sexual domains. The following examples illustrate this issue:
Question: What does the prostate do?
Answer: It helps you have a good bowel movement.
Question: What happens when a man has rectal urgency?
Answer: Diabetes—urinary frequently.
Question: What does the rectum do?
Answer: It expands your penis, makes it larger and firm.
Anatomic Identification and Function
Patient knowledge of important anatomic structures relevant to prostate health and their respective functions is presented in Table 3. The majority (60%) of men were unable to locate the prostate, whereas 97% were unable to correctly identify the function of the prostate. In addition, 54% of our patients were unable to locate the bladder and 27% could not locate the bowels; 26% and 37% of our patients were unable to correctly identify the function of the bladder and bowels, respectively. When interviewers pointed to the rectum and asked patients to identify the structure, only 33% correctly identified the structure using the word “rectum,” whereas an additional 14% identified it using correct colloquial terminology.
Misunderstanding of the Prostate and Prostate Cancer
In our study, 106 men (96%) had heard of the prostate. Although 81 (74%) could name something that goes wrong with the prostate, only 5 men could correctly describe the function of the prostate. In addition, overall knowledge of prostate cancer was poor in our patient population, although prostate cancer survivors performed slightly better in this area (Table 4). Whereas nearly half of our overall population could not provide a single correct adverse effect of prostate cancer treatment, two-thirds of the prostate cancer survivors were able to correctly identify an adverse effect. However, overall performance was still deficient within this subset of patients. For example, not a single prostate cancer survivor was able to correctly provide a risk factor for prostate cancer.
Table 4. Prostate Cancer Knowledge as Percentages of Patients Able to Correctly Answer Various Questions Related to Prostate Cancera
All Men (N = 109)
Prostate Cancer (n = 23)
All Men fs(N = 109)
Prostate Cancer (n = 23)
Patient performance subdivided into total study population and patients previously diagnosed with prostate cancer.
Chances of a man (African American) getting prostate cancer in his lifetime (1 in 5)
Risk factors for prostate cancer
Do not know
Treatments for prostate cancer
Adverse effects of prostate cancer treatment
Hormones or chemotherapy adverse effects
Most prostate cancer can be cured if it's caught early enough
Most men who get prostate cancer will die of prostate cancer
A man can have prostate cancer without pain or symptoms
Most prostate cancer tumors are fast growing
Compared to heart disease, do you know more, about the same, or less about prostate cancer? (knew less, %)
Qualitative analysis provided further insight into the confusion regarding the prostate and prostate cancer. One key pattern to emerge was the confusion of prostate and prostate cancer as synonymous terms. For example, when asked about the function of the prostate, one patient responded: “It's something like cancer, I guess it be eatin' you up.” In addition, some prostate cancer survivors exhibited a complete lack of understanding of the organ in which they had developed cancer. When asked about other words he had heard for the prostate, one patient who had been treated for prostate cancer replied: “Intestines. I think the large intestine.” Another patient diagnosed with prostate cancer was asked to provide the function of the prostate and replied: “It helps with poo poo and getting rid of solid waste from the body.”
The Institute of Medicine defines health literacy as “the degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions.” Growing evidence has established health literacy as a problem associated with numerous deleterious effects on a variety of outcomes, including increased rates of hospitalization, decreased rates of screening mammography, and low influenza vaccination coverage.[16-18] Prostate health in particular is an area in which health literacy plays an essential role, because multiple treatment options with numerous potential adverse effects necessitate the use of shared decision-making between patients and physicians.[5, 6, 19] In a study at 2 low-income general medicine clinics in Virginia, Kilbridge et al demonstrated a startling lack of comprehension of medical terms commonly used by physicians and often included in patient-education materials. This lack of comprehension may diminish patients' abilities to obtain, process, and understand basic health information, thus representing a deficiency of health literacy. Although carrying important implications, this study was limited by a small sample size confined to a single geographic location. Thus, we sought to validate these findings in 2 clinics in Atlanta, Georgia.
Our data are consistent with a widespread and severe lack of comprehension of important medical terms related to prostate health. This lack of comprehension, although slightly less pronounced, is similar to that previously demonstrated in the Kilbridge study. Importantly, the terms assessed in both studies comprise a “core vocabulary” frequently employed by physicians when counseling patients and in Web sites and patient education materials. For example, the National Cancer Institute's Web site for patient education discusses adverse effects of prostate cancer treatment, using terms such as “rectum” and “sexual function,” which were only understood by 51% and 36% of our patients, respectively.
Our results also have important implications for the use of numerical information in the communication between patients and physicians. Only 30% of our patients were able to correctly calculate both a fraction and a percentage. In addition, fewer than 3% of our patients were able to correctly answer all 3 items of the Schwartz-Woloshin numeracy test, performance on which is strongly related to accurate perceptions of the benefits of mammography. These data suggest that our patients are unable to perform and process the numerical information that is often used in patient education materials, such as the American Cancer Society Web site, as well as in written screening tools. The significance of this finding is supported by the fact that poorly numerate patients have been shown to misrepresent their American Urological Association symptom scores.
Although very similar, the rates of misunderstanding in our investigation were slightly less pronounced than those found in the study by Kilbridge et al. This difference may be explained by the higher educational level and improved literacy observed in our patient population compared with the population studied by Kilbridge et al, with 80% of our patients completing high school or beyond (Table 1) compared with just 35% of the patients in the Kilbridge study. This also represents a higher educational attainment compared to the average levels reported in African American males (38% completion of high school), according to the US census data. Although not a perfect test, the REALM is a well-accepted tool that adequately measures reading skills, allowing estimation of reading grade levels. The median reading level by REALM among our study participants was seventh to eighth grade, with only 8% reading at a third grade level or lower. This compares favorably to the median reading level of fourth to sixth grade and 27% of patients reading at or below the third grade level in the Kilbridge et al study. Despite being more highly educated and reading at a higher level, our patients still exhibit a troubling lack of comprehension of key medical terms, suggesting that these data are widely generalizable.
Other factors may have also contributed to the differences in misunderstanding between the 2 studies. Although Kilbridge et al examined patients from general medicine clinics, our patient population was recruited from 2 subspecialty clinics. Because one might expect urology patients to be more knowledgeable about the vocabulary used for prostate cancer, this difference in study design highlights that even more-selected patients exhibit a lack of comprehension of this important terminology. In addition, the language criteria prevented the enrollment of many Latino patients, who constitute a significant proportion of the patient population at Grady Memorial Hospital. Ultimately, the mean educational level of our patients does not appear to be representative of the typical patient population at Grady. This suggests that the lack of comprehension exhibited by our patients may be even more prominent if extended to the general population of Grady patients, because health literacy is highly correlated with educational attainment.
The sample size of 109 patients is a relative weakness of our study, although it is a larger population than previously studied. Moreover, although we feel that the diverse educational levels of the patients and the predominance of previously understudied inner-city men in our study are the main strengths, it is important to recognize that our findings may not be generalizable across different subpopulations. In addition, the timing of interviews did not coincide with a diagnosis of prostate cancer, which would be the expected setting for the practical use of prostate health terminology. Perhaps the key point is that physicians mistakenly assume terms such as “incontinence” are straightforward and readily understood at baseline. As such, prostate cancer discussions between patients and their physicians may not achieve fully informed decision-making. The baseline comprehension of these terms before a diagnosis of prostate cancer is of critical importance, because physicians often do not possess sufficient time and/or proficiency to adequately address the gaps in patients' knowledge. Thus, our study evaluates the knowledge of male patients at baseline, precisely the type of patient expected to receive a diagnosis of prostate cancer and undergo the subsequent counseling session.
Our data do reveal a marked misunderstanding of commonly used prostate cancer terminology and validate the findings of Kilbridge et al. However, our study group was a more educated patient sample, which to us, suggests that this lack of comprehension is widespread. Because this lack of comprehension potentially restricts the capacity of patient participation in shared and informed decision-making, our findings support a continued need for more-refined methods of prostate cancer education. Such learning tools may have the potential to educate patients about important relevant terminology before they receive any counseling of treatment options for prostate cancer, thus better equipping them to participate in the decision-making process. Prior work has shown that multimedia may represent a valuable component of such efforts, with the potential to improve knowledge among persons with varying levels of literacy. We previously demonstrated that a multimedia version of the American Urological Association symptom score increased understanding and decreased scoring errors. To further explore the use of multimedia as an approach to patient education, we are currently conducting a pilot study of a patient-centered video-based teaching tool. This tool was developed in multidisciplinary fashion, using iterative inputs from patients, providers, professional animators, and human–computer interaction specialists to produce both 2-dimensional and 3-dimensional animated illustrations combined with narration to help improve patient understanding of medical terms. It is hoped that the lack of comprehension demonstrated in this study will continue to spur the development of other innovative methods of patient education to bridge the communication gap between patient and physician.
Our results show severe lack of patient comprehension of commonly used medical terms related to the diagnosis and treatment of prostate cancer. Our findings validate work done by others, but in a different geographic region, with more selected patients attending specialized clinics, and among a more educated population than in prior reports. Although our study was primarily limited to African American patients of low socioeconomic status seen at urology or radiation–oncology clinics, the wide range of educational levels of our patients and the findings of similar results in other settings suggests that the lack of comprehension of these terms is likely a widespread phenomenon. The results of our study can be used to guide the development of interventions aimed at increasing patient understanding of prostate cancer–related terms and thereby improving shared decision-making.
This work was supported by a Winship Cancer Institute of Emory University Multi-Investigator Research Seed Grant (co–principal investigators: Drs. Master, Jani, and Goodman).