|Home | About | Journals | Submit | Contact Us | Français|
Men with local stage prostate cancer have access to a large number of information sources about therapy, including print and broadcast media, the Internet, books, and friends with the disease. The purpose of this evaluation is to determine the role that information sources play in the treatment decision process.
Prospective survey of men in three geographically separate regions with local stage prostate cancer. Most men were surveyed following diagnosis but prior to starting therapy.
On average, men consulted nearly five separate sources of information prior to treatment. The most common source of information was the patient’s physician (97%), followed by lay-literature (pamphlets, videos) (76%), other health professionals (71%), friends with prostate cancer (67%), and the Internet (58%). Most men rated the sources they consulted as helpful. Consulting the Internet was associated with considering more treatment options. Several information sources were significantly associated with considering particular treatments, but the magnitude of association was small in relation to patient age, comorbidity, and Gleason score. More than 70% of men stated they were considering or planning only one type of therapy.
Men with local stage prostate cancer consult a wide range of information sources. Non-physician information sources appear to influence their treatment considerations, but to a smaller degree than clinical factors.
Eighty-six percent of 186,000 men diagnosed with prostate cancer in the United States in 2008 will be diagnosed with local stage disease.1,2 Patients with newly diagnosed, local stage prostate cancer are offered several treatment choices, including watchful waiting, surgical resection, brachytherapy, external beam radiation, and hormone therapy.3 Currently no clinical consensus exists regarding the optimal medical management of early stage prostate cancer. With comparative data lacking from controlled studies and divergent clinical opinions about the benefits and harms of each treatment option, men with prostate cancer face difficult choices about their care.
Men with prostate cancer and their families can now readily access an extraordinary amount of information on prostate cancer and its treatment. A recent federally funded study found that patients frequently go to the Internet first to get medical information, even though trust in online material about health has declined.4 Previous studies identified factors that may be influential in the treatment decision5–9 and assessed the information needs of men diagnosed with prostate cancer.10–12 But studies that prospectively examine the range of information sources men use while making treatment decisions are relatively limited. No study has examined relative influence of the variety of information sources men utilize on their treatment decisions.
The goals for this study were to: (1) identify the treatment-related information sources men with newly diagnosed local stage prostate cancer seek out or use, and (2) evaluate the relationship between the information sources consulted and the number of treatment options considered. We hypothesized that the number of different treatment options patients consider is positively correlated with the number of information sources they accessed following diagnosis.
The Prostate CAncer Therapy Selection study (PCATS) was a prospective, multicenter observational study of men’s treatment decision-making process after a diagnosis of local stage prostate cancer. The study was conducted in three regions and settings: community urology practices in Washington State, hospital-based urology clinics affiliated with the University of Southern California as well as at one VA Hospital in the Los Angeles area, and Kaiser Permanente in Northern California. Participating research institutions included the Fred Hutchinson Cancer Research Center (FHCRC), the University of Southern California (USC), the Kaiser Permanente Northern California (KPNC), and the Northern California Cancer Center (NCCC).
Persons eligible for the study included men over age 21 with local stage prostate cancer who were diagnosed between May 1, 2005 and December 31, 2006. Persons were excluded from the study if they had a prior diagnosis of prostate cancer, were unable to read English, or had dementia or other mental illness preventing understanding of written questionnaires.
Study nurses at participating clinics reviewed charts to identify eligible patients. Study staff conducted the informed consent process with interested men either in the clinic (Southern California and Washington) or via telephone follow-up (Washington). Subjects who agreed to participate were sent home with a study packet that included the survey and a postage paid return envelope. Patients who did not return the survey within two weeks were contacted by telephone. In Washington State patients received a small payment of $15 upon completion of the survey.
To identify newly diagnosed prostate cancer patients, electronic medical records were reviewed for positive biopsy results for local stage prostate cancer. Physicians were mailed a letter to confirm the diagnosis and to obtain permission to contact the patient. All patients not withdrawn by their physician received an introduction letter and questionnaire by mail. Return of a completed survey was considered adequate patient consent. No follow-up was conducted of men who did not return a survey.
This study was approved by the Institutional Review Board of each participating center.
The study questionnaire included sociodemographic questions, treatment options, treatments recommended by physicians, information sources consulted, and factors that may be important to prostate cancer patients (e.g., risk of sexual dysfunction). Item responses for available treatment options allowed patients to state whether they were “considering” or “planning” a treatment. Patients could report they were considering or planning more than one therapy. The information sources we asked about in the survey included: specialties of health care providers seen (e.g., urologist, radiologist, oncologist, primary care, nurse); materials provided by physicians (pamphlets, videos, books-referred to as “literature” in the rest of the paper); support groups (in-person or on the Internet); national cancer websites such as the National Cancer Institute (NCI), the Center for Disease Control and Prevention (CDC), or the American Cancer Society (ACS); other Internet resources, media reports (newspapers, magazines, radio); books on prostate cancer, medical journals, and friends or family members diagnosed with prostate cancer. We asked patients to rate each source consulted on a scale ranging from not helpful, somewhat helpful, to very helpful. Clinical data including prostate-specific antigen (PSA) value, stage, and Gleason score was abstracted by clinic staff from the patient’s medical record (USC), collected from a survey of physicians (FHCRC), or extracted from the electronic medical record (KPNC).
Previous research showed younger patients and those with higher education levels more likely to play an active role in the medical decision-making process.13 Furthermore, prostate cancer-specific studies suggest that partners of men diagnosed with prostate cancer are active information seekers and significantly assist their husbands/partners in being informed about cancer and its treatments.14,15 Based on these findings, we postulated that certain patient characteristics would correlate with seeking a greater number of information sources; specifically, higher education level, being married, and diagnosis under age 60.
We computed summary statistics of patient sociodemographic and clinical characteristics and the number of treatment options considered by patients, differentiating between those who stated they were “considering” or “planning” particular treatments. For each information source listed, we report the number responding whether that source was accessed, and for those who reported use, whether the source was rated “somewhat” or “very” helpful.
We compared the number of information sources consulted and the number of treatments considered or planned. Because time available may be related to the number of information sources consulted, we also examined the relationship between time from diagnosis to the time of the survey return (days) and the number of information sources consulted.
To better examine the relative influence of patient sociodemographic factors, clinical factors, and time to access information, we used linear regression analyses to explore factors associated with the number of information sources consulted and the number treatments being considered. Independent variables for the information sources model included age, marital status, employment, education, race (white or non-white), health insurance type, survey site, comorbidity, PSA value and Gleason score. Independent variables for the model evaluating number of treatments considered included those used for the information sources model plus separate indicator variables representing whether each information source was consulted or not.
To evaluate whether patients who consulted a particular information source were more or less likely to consider or plan a particular treatment, we constructed five separate logistic regression models, with dependent variables coded for each treatment considered (including watchful waiting). For example, if the patient indicated surgery as one of their considerations, then the binary outcome for this patient would be 1; if the patient did not indicate surgery in any of their considerations, their outcome would be coded 0. Logistic regression was then carried out for this outcome, using the independent variables noted above for the linear regression model. We coded the remaining four treatment considerations similarly and analyzed them separately in four logistic regression models.
Overall, 1472 subjects were approached for participation. Sixty-nine were later determined to be ineligible due to the following reasons: non-local disease (n=62), missing clinical data (n=5), or receipt of questionnaire more than 8 months after diagnosis (n=2). A total of 804 of 1403 eligible subjects (57%) returned the completed baseline survey. Based on respondent’s reports, Surveillance, Epidemiology and End Results (SEER) records, and California Cancer Registry information, 70% returned the survey prior to initiating treatment. The average age at the time of the questionnaire return was 65.9 (s.d. 8.5) and 43% were under the age of 65; 76% were white, and 22% had an income of under $40,000 (Table 1). With regard to clinical characteristics, 90% had a PSA>4 and 62% had a Gleason score of ≤6 at diagnosis. Twenty-eight percent listed two or more non-cancer comorbidities at diagnosis.
The most common treatment options being considered or planned were surgery (42%) and external radiation (38%) (Table 2). Among those who stated they were considering or planning particular procedure(s), the average number of treatments considered was 1.4, with surgery, brachytherapy, and external radiation being most common. Seventy-one percent of patients who stated they were considering or planning a particular treatment were considering or planning a single treatment (n=563); of those considering planning only one procedure, surgery was the most common (45%), followed by external radiation (27%), brachytherapy (17%), watchful waiting (7%), and hormone therapy (4%). Among those considering or planning two treatment options (n=174), the most common combination was radiation + hormone therapy (34%); among those considering or planning three treatments (n=45), the most common combination was brachytherapy + radiation + hormone therapy (24%). There were 11 people planning or considering four treatment options.
Patients consulted an average of 4.9 (s.d. 1.8) different types of information sources during the decision-making process (median 5, min 0, max 9). The most common sources were in order: doctor(s) who performed the treatment (97%); literature (76%), other health professionals (71%), and family member/friend with prostate cancer (67%) (Table 3). Patients visited an average of 2.3 (s.d. 1.1) physicians to discuss their prostate cancer treatment. The information sources most likely to be rated somewhat or very helpful were the treating physician (96%), literature (96%), books (95%), and the Internet (94%). Face-to-face support groups were the least likely to be rated as somewhat or very helpful (70%).
In the regression model evaluating factors influencing information seeking, men with high school or less education consulted significantly fewer information sources (p<0.01), while younger men consulted significantly more information sources (p<0.01). In the regression model evaluating factors associated with the number of treatment options considered or planned, patients with Gleason scores ≤6 or 7 considered significantly fewer treatment options than those with scores 8–10 (p=0.04 and p=0.03, respectively). Persons under age 60 considered fewer treatment options compared to those over age 75 (p<0.01), although there were no differences in number of treatments considered among patients age 60 and above. Patients who used the Internet considered more options (p=0.04).
Several patient-specific demographic and clinical factors were significantly associated with considering or planning particular treatment options (Table 4). Persons in the youngest age category (<60) were 26 times more likely to consider surgery and much less likely to consider radiation, hormone therapy, or watchful waiting as management options compared to those age 75 and higher. These trends remained but the odds were attenuated for the middle age groups (60–64, 65–69 and 70–74). Married men were about half as likely to consider watchful waiting as unmarried men. Having less comorbidity was associated with a higher likelihood of considering surgery and lower likelihood of considering radiation or hormone therapy. Men with lower Gleason scores (<7) were much less likely to consider radiation or hormones, but were much more likely to consider brachytherapy or watchful waiting. Men enrolled at KPNC were much more likely to consider radiation therapy and less likely to consider surgery.
Among men who consulted particular information sources, those who consulted the media were more likely to consider surgery and watchful waiting. Those who consulted physician-distributed literature were more than twice as likely to consider brachytherapy. Persons who consulted the Internet were more than twice as likely to consider hormone therapy. Men who consulted friends were half as likely to consider watchful waiting. Utilizing Internet support groups or consulting other health professionals were not significantly associated with higher or lower likelihoods of considering any of the treatment modalities.
Prostate cancer patients have access to an extraordinary variety of information sources about treatment options, yet little research has evaluated the sources they utilize and the influence those sources have on treatment preference. We designed a study to examine the relationship between information sources consulted and treatment options considered among men with newly diagnosed local stage prostate cancer who had not yet received therapy for their disease. Although patients consulted nearly five information sources on average, a substantial majority (71%) reported they were considering or planning only one treatment option, primarily surgery or external radiation. In multivariate analysis, younger age at diagnosis was positively associated with consulting a greater number of information sources, while less educated men consulted fewer sources. Men under age 60 and those with more favorable Gleason scores considered fewer treatments; those who consulted the Internet considered more options. Several information sources were significantly associated with considering particular treatments, although the degree of influence was generally much less than age, comorbidity, and Gleason score.
Ready access to a wide variety of information sources may have overcome demographic characteristics that traditionally have been barriers to information seeking during an important decision. The sheer number of sources consulted suggests that men are actively involved in considering the implications of their decision. In multivariate analysis, we found an age and educational disparity in the number of information sources men consulted. It is likely that younger men are more comfortable using the internet, which in turn can direct them to more information sources. Lower educational attainment may restrict knowledge of or access to certain types of information sources (e.g., internet).
The great majority of men who accessed a particular information source rated that source as helpful or very helpful. Face-to-face support groups had the lowest proportion of “helpful” ratings. This may reflect the fact that support groups are more oriented towards helping patients cope with their disease during and after treatment rather than during the decision process.
There was little relationship between the information source consulted and the number of treatments being considered, with the exception of Internet users, who tended to consider more options. Slightly more than half the patients in the sample accessed the Internet to obtain information about prostate cancer. A recent study found that most Internet resources related to prostate cancer required high level reading skills and had limited cultural sensitivity.16 A study of Dutch cancer patients’ Internet use found that cancer-related sites sponsored by pharmaceutical companies were most often mentioned as information sources. Most patients felt that the web sites improved their knowledge about the disease.16,17
Satisfaction with a decision may be influenced by the information source(s) accessed. A prospective study found that age, marital status, the influence of the urologist, and use of the Internet were most strongly associated with satisfaction with men’s decision about therapy for local stage prostate cancer.18
Although patients consulted a relatively large number of information sources, the majority stated they were considering only one treatment. In contrast, several information sources were correlated with particular types of treatment men stated they were considering. Nevertheless, when considering individual treatments, the magnitude of effect of a particular information source for or against a particular treatment was much smaller than demographic and clinical factors, most notably young age (heavily increasing the odds of considering surgery and decreasing the odds of considering watchful waiting), comorbidity (fewer numbers of comorbid conditions increased the odds of considering surgery and decreased the odds of considering radiation), and Gleason score (lower scores heavily increased the odds of considering brachytherapy and watchful waiting and greatly decreased the odds of considering hormone therapy and radiation). Other studies have found that younger men are more likely to seek definitive treatment such as surgery.19,20 Less aggressive pathological subtypes may encourage physicians to steer patients towards watchful waiting rather than definitive therapies.
This study has several limitations. The overall response rate among eligible patients was approximately 57%. Those who did not respond may have been different in some way from responders, such as having problems with literacy or other serious illnesses that prevented them from completing the survey. While all patients were approached prior to treatment and a substantial number of men in our study (70%) returned their surveys prior to starting therapy, nearly 30% appear to have returned their survey after they had started treatment. It is possible that men who receive treatment very quickly after their diagnosis may be different with regard to information seeking and options considered than men who completed the survey prior to starting treatment. This study was conducted in community clinics in Washington State, clinics affiliated with the USC and within Kaiser Permanente. While there was a considerable level of diversity in this sample, it may not be generalizable to other settings. Kaiser Permanente has a particularly active radiation therapy program for patients with local stage prostate cancer. While patients in Washington State were recruited from community urology clinics, the USC academic clinics included patients seeking both primary and secondary opinions. We only asked about the type of information source and whether they found the source helpful in their decision making process. We did not ask men about the content of information obtained from each source, nor did we ask men when they accessed a particular information source and when they decided to consider or plan on a particular treatment. Others have raised questions regarding the lack of accuracy and balance of information that is available to the general public on prostate cancer therapy.21–23
In summary, men with newly diagnosed prostate cancer access many information sources during the time between diagnosis and treatment. More than 7 in 10 patients stated they were considering or planning one treatment. While patients rated most sources as helpful, the source of information has a modest association with treatment preference in comparison to age and cancer-aggressiveness as measured by Gleason score. These are crucial factors in the discussions physicians have with their patients.
This publication was supported by the National Cancer Institute contracts N01-PC-35142, N01-PC-35139, and N01-PC-35136.
This research was also supported by the Cancer Surveillance System of the Fred Hutchinson Cancer Research Center, which is funded by Contract No. N01-PC-35142 from the Surveillance, Epidemiology and End Results (SEER) Program of the National Cancer Institute with additional support from the Fred Hutchinson Cancer Research Center and the State of Washington.
The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the National Cancer Institute.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.