|Home | About | Journals | Submit | Contact Us | Français|
Prognostic awareness, or knowing that one has a life-ending disease, is associated with a better end-of-life experience, including less depression and anxiety. We sought to determine whether reliable sources on the Internet contained helpful prognostic information about advanced cancer.
We played the role of a 62-year-old person with stage IV incurable cancer and accessed four commonly used Web sites for the 10 most common causes of cancer death (American Cancer Society, ASCO, National Cancer Institute, Up To Date), as well as disease-specific Web sites.
Approximately half the Web sites (26 of 50; 52%) had some notation of 5-year survival. Only four of 50 (8%) gave any average or median survival. Only 13 of 50 (26%) noted that stage IV cancer was a serious and usually life-ending illness. Nearly all had some information about hospice and palliative care.
Information that can help with patient prognostic awareness is not currently found on cancer-related Web sites. Oncologists should be aware that their patients will not find estimates of survival or treatment effect on the Internet. This may contribute to overoptimistic estimates of survival and subsequent aggressive end-of-life care.
People with advanced cancer overestimate their prognosis and the effects of treatment, often dramatically. For instance, nearly three quarters of contemporary patients with incurable colorectal and lung cancer thought it was possible to be cured with chemotherapy1 or radiation therapy2 when there was no possibility of cure and the average survival was short. It can be difficult for patients to find truthful and helpful information about advanced cancer. Oncologists can find it hard to share bad prognoses3–5 and provide overestimates of survival of at least 30%.6 In addition, most information during the visit is related to treatment.7 We have listed some of the most telling communication miscues, including the fact that only 17% of patients with advanced lung and colorectal cancer can identify their prognosis (Table 1).
If a person does not get information from their oncologists, they will often resort to the Internet. Nearly 90% of Americans and 59% of seniors routinely use the Internet, and 72% of Internet users looked online for health information. Half of the users are searching on behalf of someone else such as a parent or spouse.8 Can a person with advanced cancer learn what will happen to them, what can be done about it, and what to expect? We sought to determine what a seriously ill patient with stage IV cancer could learn from the most commonly used Web sites (ie, the “Big Four”) of the American Cancer Society, the National Cancer Institute, ASCO (eg, lung cancer9), and Up to Date (eg, breast cancer10), as well as disease-specific Web sites.
We (a second-year medical student with a prior MPH, and a 62-year-old medical oncologist with 33 years of practice experience) searched the Internet in the guise of a 62-year-old person with an advanced (stage IV) case of one of the 10 cancers most likely to cause death in the United States. We accessed Web sites and searched for the information that might be important to most patients. We focused on information that was directed to patients and excluded content that was meant for health care providers and researchers. Web sites were accessed between June 27, 2014, and July 7, 2014, and rechecked in February 2015. We viewed five different Web site for each of the top-10 cancers in the United States (50 Web sites total), and searched for information that would be important to most patients in making informed decisions about their care (giving the diagnosis a name, the natural history of the illness, the likely prognosis, common symptoms, and what caused it11). Review included the following five items: (1) whether there was any mention of this disease being a life-threatening or terminal illness, (2) presence of survival rates (1 and 5 years) and median survival times, (3) any mention of advanced planning, (4) any mention of hospice care, and (5) any mention of palliative care.
Results are shown in Table 2. For the 10 most lethal cancers in the United States, only 13 Web sites stated specifically that this was a serious or life-ending illness. A person searching specifically for 5-year survival figures for stage IV cancers could find some notation of survival in only 26 Web sites (52%). Only four Web sites (8%), two of which came from UpToDate (UpToDate: Breast Cancer, and UpToDate: Non-Hodgkin's Lymphoma), contained any information about the average or median survival, the most meaningful statistic to someone asking “What will most likely happen to me?” For example, the patient information on Up to Date states explicitly that the average person with metastatic breast cancer lives “… 18 to 24 months, although the range in survival spans between a few months to many years and depends very much on the type if breast cancer the patient has.” Currently, approximately 80% of Web sites have links to sites related to hospice and palliative care, but none explicitly recommended concurrent anti-cancer and palliative care, as suggested by national guidelines.12
We found that the majority of Web sites now mention hospice (39 of 50; 78%) and palliative care (37 of 50; 74%). Approximately half (26 of 50; 52%) give 5-year survival rates. However, only a quarter (13 of 50; 25%) mention that the disease under discussion—for example stage IV pancreatic cancer—is a serious, life-threatening, or even life-ending disease. Finally, only four of 50 (8%) give the statistic most meaningful for treatment and advance care planning: how long will the average person with this serious cancer live? This information can be provided on Web sites, as noted above for the publically available Up To Date and ASCO lung cancer decision aids.13 A similar publically available tool for heart failure patients gives actual survival graphs and encourages the patient to print and bring to his or her provider.13
This lack of relevant information has an impact on what oncology patients know and understand, and bring to the patient-oncologist encounter. We do know that patients who over-estimate their survival do not live longer but are more likely to receive aggressive end-of-life care including chemotherapy in the last 30 days of life, intensive care unit use, resuscitation, and less use of hospice.14,15 Care of Medicare-age patients with cancer is also becoming more aggressive over time, with more patients admitted to hospitals and intensive care units, and fewer days in hospice.16 We know that the great majority of patients want to know their prognosis, even if it is bad.17 We also know that people who actually discuss their end-of-life care wishes with a physician are twice as likely to get care consistent with their preferences18 such as dying at their preferred place.19 Patients who have “prognostic awareness” – who know and admit that they have an incurable disease – paradoxically have better mood, less depression and anxiety, and better quality of care at the end of life, in addition to living at least as long if not longer.20–22
The high percentage of Web sites giving information on hospice and palliative care, coupled with the low percentage of information regarding survival time highlights a common challenge in our approach to the terminally ill patient. Data consistently show that actual patients want such prognostic and treatment information,23 use it for advance care and life planning, and will access a patient-friendly Web site,24 but we now know they will not find realistic information on the Internet. Realistic prognostic awareness is also strongly countered by commercial media campaigns of “Cancer Center ______, where hope lives,” or “Cancer Center ______, where we will never give up on you.” Cancer center advertisements concentrate on hope, while rarely mentioning survival, risks, or benefits of treatment.25
Strengths of this report include that we looked at Web sites for the top ten different cancers and found consistent patterns. Limitations for this study include the fact that Web site content can be interpreted differently by patients based on their medical literacy, culture, and support. Also, even though many Web sites had the information on palliative care, hospice care, and advance care planning, we did not compare the quality of the content and the kinds of information being presented.
Some will argue that giving patients such information over the Internet is somehow dangerous and unethical. We read that oncologists should not say “You have [insert number here] months to live” because patients will take it literally, and no doctor knows for certain.26 We counter that patients understand that if the average survival is 12 months, half will be above and half below—not all can be above average, the Lake Wobegon effect27—and that physician estimates are in fact accurate, if a little overestimated.28 The current data suggest that patients are not hearing us, are not accepting the facts, or we are not telling them effectively—or all three—and that they are unable to get such information from the next source they will access, the Internet.
One way to increase prognostic awareness is for patients and their families to have access to truthful, helpful prognostic information, for those who want it. National societies of health professionals caring for patients with cancer have the responsibility to provide in their patent-oriented Web sites descriptive prognostic information that uses the technology of decision aides, which we know are effective.29
Supported by a Medical Student Training in Aging Research Program grant to Johns Hopkins Department of Geriatrics (I.C.), Grant No. P30 CA 006973 to the Sidney Kimmel Comprehensive Cancer Center, and the California HealthCare Foundation.
Disclosures provided by the authors are available with this article at jop.ascopubs.org.
Conception and design: Thomas J. Smith
Financial support: Thomas J. Smith
Administrative support: Thomas J. Smith
Provision of study materials or patients: Thomas J. Smith
Collection and assembly of data: All authors
Data analysis and interpretation: All authors
Manuscript writing: All authors
Final approval of manuscript: All authors
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or jop.ascopubs.org/site/misc/ifc.xhtml.
No relationship to disclose
Stock or Other Ownership: United Healthcare