Use of the Internet is common among patients with cancer and their companions. However, little is known about patterns of use of print or telephone-based resources amidst growing Internet utilization, nor is it known whether different types of information are sought from electronic compared with print media. It is not clear as to whether patients and their companions differ in their patterns of content seeking.
A survey was developed to evaluate the use of electronic and nonelectronic informational resources by patients and their companions. During a 10-week period, this questionnaire was administered to 443 outpatients and 124 paired companions attending an urban academic cancer center.
In this cohort, 64% of patients and 76% of companions were computer owners, with home Internet access indicated by 58% and 68%, respectively. Use of the Internet to obtain cancer-related information was reported by 44% of patients and 60% of companions. Print resources were used by 79% of patients and 83% of companions, with telephone resources used by 22% and 23%, respectively. The majority of Internet users also read print content (85%), whereas one-half of print users did not access data electronically (52%). Topic areas sought via print and the Internet were similar, with the exception of nutrition-related information, which was more commonly sought in print texts. There was a high rate of concordance between patient and companion use of both electronic and nonelectronic resources.
Patients with cancer and their caregivers frequently seek information about disease management, prognosis, and therapeutic alternatives.1–5 Use of the Internet by this group is common, with estimated use by 39–58% of patients in the United States.6–9 Usage rates range from 42% to 49% among patients with breast carcinoma,10–13 from 32% to 45% among patients with prostate carcinoma,14, 15 and from 16% to 18% among patients with lung or head and neck malignancies.16, 17 Clinical oncologists have estimated that approximately 30% of their patients obtain cancer-related information online.18
Amidst increasing rates of Internet searching by patients, there has been limited study of the frequency of use of nonelectronic media resources. We conducted a survey of patients with cancer and their companions to investigate the degree to which Internet use has replaced or supplemented more traditional information resources such as books, pamphlets, and telephone hotlines. Moreover, we examined which types of media are used to gather various categories of information, whether particular patient characteristics are associated with types of media sought, and whether resource use differs between patients and their companions. We hypothesized that although rates of Internet use and online health care seeking are increasing,19–21 patients continue to access nonelectronic resources.
MATERIALS AND METHODS
English-speaking adult patients with cancer who reported to the general surgical and gastrointestinal oncology outpatient clinics at Memorial Sloan-Kettering Cancer Center (MSKCC; New York, NY) between December 14, 1999, and February 24, 2000, were invited to participate in the current study. After clinic check-in, a single research assistant invited patients to complete a written 15-minute questionnaire regarding their use of cancer information resources. Patients were informed that results would not be linked to individual medical records or made available to clinical staff. Each questionnaire was coded with a unique identification number to ensure confidentiality. Consenting patients accepted and returned the study questionnaire, either directly to the research assistant or to boxes at the clinic check out desk.
Companions accompanying patients to the clinic were invited to complete a similar questionnaire regarding their use of information resources to learn about cancer on behalf of the patient. When more than one person accompanied a patient, a single individual, designated by the patient, completed the companion questionnaire. In the few instances where multiple companions elected to complete questionnaires, their responses were weighted in data analysis. When both patient and companion responded to the questionnaire, their unique identification numbers allowed linkage during analysis.
The initial questionnaire was developed based on review of the medical literature, targeted Internet browsing, and suggestions made by nurse educators at MSKCC. The majority of the items were constructed specifically for the purposes of the current study based on published guidelines for survey development or were derived from previously validated instruments.22 A pilot version of the survey was administered orally by the research study assistant to 17 patients and 15 companions in the clinic waiting area. Results were used to shorten and revise the instrument to enhance comprehensibility. A second pilot survey was administered by the research assistant to 18 patients and 9 companions; the results of this second pilot trial indicated that the feasibility, willingness of individuals to respond, and survey content were acceptable.
The finalized questionnaire assessed whether patients and companions had used telephone, print, and/or Internet resources to obtain information related to the diagnosis of cancer. For example, participants were asked whether they had used any telephone resource, including the National Cancer Institute (NCI) hotline and the American Cancer Society (ACS) hotline, to obtain cancer-related information. Participants estimated the amount of total time spent using each resource type since their diagnosis and indicated whether they considered each to be not helpful, helpful, or very helpful. Similarly, for print and Internet resources, participants estimated the time dedicated to research in particular content areas such as nutrition, complementary and alternative medicine, physician information, and cancer diagnosis. Internet users were asked to indicate whether they had accessed any of 21 individual websites, selected based on clinician feedback at our institution as well as U.S. usage statistics published by Harris Interactive23 and Jupiter Media Metrix,24 and to estimate the time spent using each particular site. The list of sites included large general medical sites, such as WebMD (available from URL: http://www.webmd.com), as well as cancer-specific sites, such as those of the ACS (available from URL: http://www.cancer.org) and the NCI (available from URL: http://www.cancer.gov). The questionnaire provided blank space for Internet users to identify the three websites that they found to be most helpful, including sites not included in the provided list. Computer ownership and the availability of Internet access at home were also assessed.
To ensure confidentiality, survey responses contained no information that could be linked to an individual's medical record. However, participants did provide basic demographic information, such as age, gender, and level of education, as well as clinical information, including cancer type and date of diagnosis. Companions indicated their relationship to the patient and whether or not they lived in the same household.
Summary statistics were generated based on the results of the questionnaire. Differences between information resource users and nonusers with respect to demographic and clinical characteristics were examined using chi-square or t tests, depending on whether the variable measures were categoric (e.g., highest level of education), dichotomous (e.g., computer owner/nonowner), or continuous (e.g., number of months since diagnosis).
In addition, several hypotheses were tested. Based on available survey data regarding the use of Internet health care information among the general population,19, 20 we expected that younger age, higher education level, and more recent cancer diagnosis would be associated with higher levels of Internet use. We further hypothesized that these variables, as well as computer ownership and the availability of Internet access, would predict Internet use. Variables with significance levels of P < 0.20 on univariate analysis were considered to be candidates for entry into multivariate logistic regression models used to assess correlations with information resource use. The multivariate models yielded adjusted odds ratios and 95% confidence intervals for each variable while controlling for the simultaneous effects of other variables. Variables not contributing substantially to the model were systematically removed in a backward stepwise regression procedure using the likelihood ratio test as the criterion for removal. SAS software (Version 8.0; SAS Institute, Cary, NC), was used to perform all analyses. All P values were two sided.
Of the 625 individuals approached, 443 (71%; 223 patients and 220 companions) completed and returned questionnaires. From these 443 questionnaires, 124 patient questionnaires were paired with companion questionnaires. No information was available on nonrespondents. Table 1 summarizes the characteristics of patients and companions in the current cohort. On average, companions were 6 years younger than patients. Both groups of respondents were relatively well educated, with the majority of individuals having received college education or more. Furthermore, 64% of patients and 76% of companions were computer owners. Internet access at home was also common, as it was reported by 58% of patients and 68% of companions in the current cohort (Table 1).
Table 1. Characteristics of Survey Respondents
% of patients (n = 223)
% of companions (n = 220)
SD: standard deviation.
Patients who had participated in a clinical trial were not asked to respond.
Thirty-six patients received both surgical and medical evaluations.
Internet Use and Characteristics Associated with Use
Among enrollees, 44% of patients and 60% of companions reported use of the Internet to obtain cancer-related information. On multivariate analysis, several factors were found to be significantly associated with increased Internet use among both patients and companions, including younger age (P = 0.03), computer ownership (P < 0.05), Internet access at home (P = 0.002), and higher education level (P < 0.05). Adjusting for these predictors, we identified an interaction between gender and respondent type (Table 2). Female companions were the most likely Internet users, whereas female patients were the least likely users (P = 0.03). A small percentage of patients (7%) who did not specify whether they used Internet resources were treated as having provided negative responses.
Table 2. Respondent Characteristics and Internet Use
Comparison of Internet, Print, and Telephone Resource Use
To place Internet usage rates in context, we surveyed respondents regarding their use of noncomputer content. We found that print resources, such as books and pamphlets, were used more often (patients: 79%; companions: 83%) than the Internet. However, telephone resources were used substantially less often (patients: 22%; companions: 23%) than either the Internet or print resources (Table 3).
Table 3. Resource Use Overall and By Topic
% of patients (n = 223)
% of companions (n = 220)
ACS: American Cancer Society.
Diagnosis and Treatment
Drugs and side effects
Diagnosis and treatment
Diagnosis and treatment
≥ $25 spent on print resources
ACS information line
Physician/hospital referral services
To derive a more complete picture of respondents' information resource use patterns, we assessed their tendencies to rely on more than one resource. We found that 86% of patients and 92% of companions who used the Internet also used print resources. Of the patients and companions who used telephone resources, 100% also used print or both print and Internet resources. Only 14% of patients and 12% of companions denied use of any of the three resource types (Fig. 1).
Patterns of Use by Topic Area and Specific Source
To evaluate respondents' patterns of information gathering, specific questions were asked regarding the type of information sought from print and Internet resources, as well as the particular sources that were accessed (e.g., websites, bookstores, specific hotlines). Most Internet users sought information on diagnosis and treatment, and many inquired about drugs, treatment side effects, physicians, and hospitals (Table 3). Government-sponsored websites (e.g., the National Institutes of Health [available from URL: http://www.nih.gov] and NCI websites) were more commonly used than were commercial sites (Table 4). The majority of users described the Internet as helpful or very helpful and spent ≥ 1 hour using it (Fig. 2), while 41% spent > 5 hours using it.
Table 4. Specific Websites Visited by Respondents Reporting Internet Use
% of patients (n = 99)
% of companions (n = 131)
MSKCC: Memorial Sloan-Kettering Cancer Center; NCI: National Cancer Institute; NIH: National Institutes of Health; ACS: American Cancer Society; PDQ: Physician Data Query; ASCO: American Society of Clinical Oncology.
Print resource users also most commonly sought information on cancer diagnosis and treatment, but these individuals were more likely to research nutrition than were Internet users. Like Internet users, many print resource users investigated physicians, cancer prevention, and complementary/alternative medicine. Use of a library to obtain printed texts was reported by 19% of patients and 30% of companions. Similarly, 24% of patients and 29% of companions indicated use of a bookstore. Only approximately one-fourth of respondents reported spending ≥ $25 on print products (Table 3). Among print resource users, 99% reported ≥ 1 hour of use, with 23% reporting > 5 hours.
Telephone use was evaluated by hotline type. We found that a majority of telephone users called 1-800-4CANCER and/or the ACS information hotline. A telephone was used for physician and hospital referral services by 32% of patients and 37% of companions (Table 3).
Comparison of Patient and Companion Resource Use
There was a high rate of concordance between patient and companion resource use (Fig. 3). The rate of concordance with respect to Internet use between patients and companions was 0.67 (kappa coefficient = 0.35). Similarly, the concordance rate for printed resource use was 0.75 (kappa coefficient = 0.20), and the concordance rate for telephone resource use was 0.75 (kappa coefficient = 0.31). For each resource type, and for resource use overall, companions whose patient counterparts denied use were more likely to report use than were patients whose companions denied use.
Use of the Internet to seek health care content is common, particularly among patients with life-threatening or chronic diseases. Surveys suggest that two-thirds of the U.S. population actively uses the Internet,19, 20 whereas > 90 million have sought health information online.21 More than one-third of patients ask clinicians about websites during visits25 and believe that use of the Internet enhances the relationship between patient and physician.26, 27
Multiple surveys reflect the growing prevalence of Internet use by patients with cancer, but little is known about the balance between information gathering on the web and other available resources, such as books, magazines, and telephone hotlines. In the context of increasing concern regarding the quality of Internet health-related content, patient reliance on nonelectronic resources or government-sponsored websites may reassure clinicians.26, 28–35 However, similar concerns regarding quality of content are likely to exist, regardless of the medium of information delivery.
The current study suggests that although the Internet is popular among patients with cancer and their companions as a vehicle for gathering information, print publications remain the most commonly used resources. Furthermore, print products are sought by the majority of Internet users, whereas ≤ 30% of print users do not access the Internet for health-related purposes. These findings suggest a need to focus increased scrutiny on print publications aimed at patients with cancer by broadening recent efforts targeting the quality of Internet-based information.
With the exception of nutrition information, which was gathered more frequently from print resources than from the Internet, topics of inquiry were similar for print and electronic resources. The most common topic area sought by both print and Internet users was the diagnosis and treatment of specific cancer types, including chemotherapy agents. Many also sought information on physicians, prevention, and complementary/alternative medicine. Better data are needed about the specific print resources being used by patients, especially regarding how electronic and print products complement each other. For example, Internet sites can be updated in a more timely manner and may be more useful for tracking information on new therapeutic developments, clinical trial availability, or physician profiles.
Telephone resources were less commonly accessed. However, with the advent of new mobile technologies such as wireless web browsing, short messaging systems, or interactive telephone programs, there may be an opportunity for increased use of telephone-based health care applications in the future.36–42
The majority of respondents had computers with Internet access in their homes. Supplemental print resources were retrieved from both libraries and bookstores. For those who sought information on the web, government-hosted sites were generally preferred over commercial sites. A majority of users found these resources to be satisfactory, describing them as helpful or very helpful, and were willing to devote multiple hours to them.
Very few individuals in either group denied using any of the resource types presented. There was a high rate of concordance with respect to resource use between patients and companions. We confirmed that higher education level was associated with Internet use, which is consistent with other published data.19, 20 Youth was also a factor in Internet use. Less of a factor was the length of time since diagnosis, suggesting that patients and their companions continue to have informational needs over the course of the disease.
The current study has several limitations. The cohort was derived from patients treated at an urban academic cancer center and probably does not represent the general population of individuals with cancer. Nonetheless, although overall rates of resource use may differ in the general population, it is likely that the observed relations between variables would remain unchanged. The generalizability of our results also is limited by the lack of information regarding nonparticipants. Furthermore, patients and companions may have influenced each other's responses, accounting for the high level of response concordance between groups.
Patterns of Internet use among patients with cancer may have changed since the current survey was conducted, and the balance between use of electronic and print resources may have shifted as a result of overall increased Internet use in the United States. However, high levels of home computer and Internet access, consistent with present-day levels of use in the U.S., were observed in the current cohort. Therefore, it is likely that the observed resource use characteristics continue to hold true. Finally, although we have no information regarding any correlation between information use and clinical decision-making or satisfaction with care, this area merits future investigation.