Online cancer-related support is an under-studied resource that may serve an important function in the information seeking, care, and support of cancer patients and their families. With over 9.8 million cancer survivors (defined as anyone living with cancer) in the United States alone and the number growing worldwide, it is important to understand how they seek and use online resources to obtain the information they need, when they need it, and in a form and manner appropriate to them. These are stated cancer communication goals of the US National Cancer Institute.
Our purposes are to (1) present background information about online mailing lists and electronic support groups, (2) describe the rationale and methodology for the Health eCommunities (HeC) study, and (3) present preliminary baseline data on new subscribers to cancer-related mailing lists. In particular, we describe subscribers' use of mailing lists, their reasons for using them, and their reactions to participating shortly after joining the lists.
From April to August 2004, we invited all new subscribers to 10 Association of Cancer Online Resources mailing lists to complete Web-based surveys. We analyzed baseline data from the respondents to examine their perceptions about cancer-related mailing lists and to describe how cancer patients and survivors used these lists in the period shortly after joining them.
Cumulative email invitations were sent to 1368 new mailing list subscribers; 293 Web surveys were completed within the allotted time frame (21.4% response rate). Most respondents were over age 50 (n = 203, 72%), white (n = 286, 98%), college graduates (n = 161, 55%), and had health insurance (n = 283, 97%). About 41% (n = 116) of new subscribers reported spending 1 to 3 hours per day reading and responding to list messages. They used the mailing lists for several reasons. Among the most frequently reported, 62% (n = 179) strongly agreed they used mailing lists to obtain information on how to deal with cancer, 42% (n = 121) strongly agreed they used mailing lists for support, and 37% (n = 109) strongly agreed that they were on the mailing lists to help others. Smaller proportions of new subscribers strongly agreed that others on the mailing lists had similar cancer experiences (n = 23, 9%), that they could relate to the experiences of others on the lists (n = 66, 27%), and that others on the list gave them good ideas about how to cope with cancer (n = 66, 27%).
Cancer-related online mailing lists appear to be an important resource, especially for information seeking but also for support of cancer survivors. These were the primary motivators most members reported for joining mailing lists. The modest proportion of subscribers who strongly agreed that they could relate to others' cancer experiences (as well as similar responses to other process questions) is undoubtedly due at least in part to the short duration that these subscribers were involved with the mailing lists. Emerging data, including our own, suggest that mailing lists are perhaps under-used by minority patients/survivors. These preliminary data add to a growing body of research on health-related online communities, of which online mailing lists are one variant.
Recent data indicate that 65% of men and 61% of women in the United States go online [
By 2001, there were approximately 9.8 million cancer survivors in the United States [
The purposes of this paper are as follows: (1) to present background information about online mailing lists and electronic support groups (ESGs), (2) to describe the rationale and methodology for the Health eCommunities (HeC) study, and (3) to present preliminary baseline data on new subscribers to cancer-related mailing lists who chose to respond to an online survey. In particular, we describe subscribers' use of mailing lists, their reasons for using them, and their reactions to participating shortly after joining.
The rise in Internet use has led to an increased number of people who seek support and information online. Some sources estimate that as many as 90 million Americans have participated in online support groups and that 1 in 4 people seeking disease information join online discussion groups [
Electronic support groups (ESGs), which include mailing lists, are much like self-help groups in that they are composed of members who share a common condition, situation, heritage, symptom, or experience [
Currently, most ESGs appear to be unmoderated and are more like self-help or mutual help groups than face-to-face support groups [
Patterson et al [
Cancer patients use mailing lists and other Internet resources for many reasons. These include seeking and obtaining information and support, seeking second opinions, and getting information needed to interpret information from health providers [
ESGs may offer privacy and convenience, and people who do not feel well can participate from home. Moreover, people can communicate on the basis of shared experiences and concerns, not shared social characteristics (such as age, race, or gender) [
Participation in ESGs may help patients be more involved in their care, find information, obtain support, and formulate questions to ask health providers [
Access to health information through ESGs may serve an especially important communication function (eg, enhancing confidence in asking questions of one's physicians) [
Online support groups have many of the same potential disadvantages as other forms of Internet communication, such as email. For example, there may be more hostile messages or “flaming” than would occur outside the Internet [
The Science Panel on Interactive Communication and Health [
Overall, Lieberman and Russo [
Health eCommunities (HeC), funded through the Robert Wood Johnson Foundation's Health eTechnologies Program, is assessing the impact on cancer survivors and their caregivers of participating in mailing lists sponsored by the ACOR (
Why do new mailing list subscribers join cancer-related mailing lists?
What are new subscribers' expectations regarding mailing lists?
How frequently do new mailing list subscribers use the mailing lists?
How do they assess the lists shortly after joining, in terms of the similarity of their experiences to those of others and on several aspects of information seeking and social support?
The preliminary baseline data reported here were collected as part of a larger impact evaluation. Understanding why people join these mailing lists and how they use them will enhance what is known about online health support.
ACOR home page
Our approach to understanding cancer survivors' use of cancer-related mailing lists is informed by theories on stress and coping [
Heightened perceptions of risk can cause distress, disengagement, and avoidance behaviors [
We are using a longitudinal cohort design to assess the impact of ACOR mailing lists on selected outcomes for new subscribers; however, we only report cross-sectional baseline data in this preliminary paper. For the pilot study, new members were recruited to participate about a week after they subscribed to one of 10 ACOR mailing lists. (However, some subscribers may have waited several days to respond or may never have responded.) Invitations to participate were sent to new ACOR members via email. Willing participants could either follow hyperlinks to Web-based surveys or request that they be interviewed via telephone. As less than 1% of respondents requested telephone interviews, we will not discuss telephone interviews further in this paper. Up to three post-notification contacts were made to each non-respondent by email to increase responses [
For analyses reported here, we focused on information seeking and the processes by which new subscribers used the mailing lists. We also examined how new subscribers responded to the surveys through analysis of time-stamp data (described below). Relevant survey items are summarized as follows:
Throughout the pilot phase, we collected data on survey usage patterns through time stamps. The online survey took a “stamp” of the time when respondents continued to another page of the survey, saved their progress, and completed the survey. See
Time-stamp data for three respondents
|
|
|
|
|
1 | 6/17/04 5:42 | New Record | 0:00:00 | |
2 | 6/17/04 5:43 | 0:00:51 | 0:00:51 | |
3 | 6/17/04 5:44 | 0:00:21 | 0:01:12 | |
4 | 6/17/04 5:44 | 0:00:08 | 0:01:20 | |
5 | 6/17/04 5:44 | 0:00:23 | 0:01:43 | |
6 | 6/17/04 5:45 | 0:00:52 | 0:02:35 | |
7 | 6/17/04 5:46 | 0:00:54 | 0:03:29 | |
8 | 6/17/04 5:46 | 0:00:32 | 0:04:01 | |
9 | 6/17/04 5:47 | 0:00:21 | 0:04:22 | |
10 | 6/17/04 5:47 | 0:00:16 | 0:04:38 | |
11 | 6/17/04 5:47 | 0:00:27 | 0:05:05 | |
12 | 6/17/04 5:48 | 0:00:28 | 0:05:33 | |
13 | 6/17/04 5:49 | 0:00:46 | 0:06:19 | |
16 | 6/17/04 5:50 | 0:01:32 | 0:07:51 | |
17 | 6/17/04 5:51 | 0:00:47 | 0:08:38 | |
18 | 6/17/04 5:52 | 0:00:30 | 0:09:08 | |
19 | 6/17/04 5:53 | 0:01:15 | 0:10:23 | |
20 | 6/17/04 5:54 | 0:01:05 | 0:11:28 | |
21 | 6/17/04 5:54 | 0:00:16 | 0:11:44 | |
22 | 6/17/04 5:55 | 0:00:31 | 0:12:15 | |
23 | 6/17/04 5:55 | 0:00:09 | 0:12:24 | Completed |
|
||||
1 | 6/17/04 7:43 | New Record | 0:00:00 | Break-off at page 1 |
|
||||
1 | 6/17/04 8:02 | New Record | 0:00:00 | |
2 | 6/17/04 8:03 | 0:01:11 | 0:01:11 | |
3 | 6/17/04 8:04 | 0:00:18 | 0:01:29 | |
4 | 6/17/04 8:04 | 0:00:12 | 0:01:41 | |
5 | 6/17/04 8:05 | 0:01:10 | 0:02:51 | Break-off at page 5 |
* One respondent completed the survey in 12:24 minutes, and two respondents broke off (one at less than 1 minute and one at 2:51 minutes).
Cumulative email invitations were sent to 1368 new mailing list subscribers; 293 Web surveys were completed (21.4% response rate).
Project flow
As
Sociodemographic characteristics of respondents (N = 293)
|
|
|
|
Excellent | 40 (14) |
Very good | 90 (31) |
Good | 89 (30) |
Fair | 50 (17) |
Poor | 23 (8) |
|
|
< 40 | 21 (7) |
40s | 60 (21) |
50s | 105 (37) |
60s | 71 (25) |
≥ 70 | 27 (10) |
|
|
Male | 149 (51) |
Female | 144 (49) |
|
|
Employed for wages | 103 (35) |
Self-employed | 30 (10) |
Out of work/unable to work | 57 (20) |
Homemaker/student/retired | 101 (35) |
|
|
Married or living as married | 233 (80) |
Divorced/separated/widowed/never married | 58 (20) |
|
|
American Indian or Alaska native | 2 (1) |
Asian | 3 (1) |
Black or African American | 1 (0) |
Native Hawaiian or other Pacific Islander | 0 (0) |
White | 286 (98) |
|
|
High school (grade 12), GED, or less | 42 (14) |
College, 1 year to 3 years (some college or technical school) | 90 (31) |
College, 4 years or more (college graduate) | 161 (55) |
|
|
Yes | 283 (97) |
No | 9 (3) |
* Percentages may not add to 100% due to rounding.
We asked why respondents used the Internet over the past 30 days (
Purpose of Internet use over past 30-day period (N = 293)
|
||||
|
|
|
|
|
Find out more about cancer | 11 (4) | 13 (4) | 91 (31) | 176 (61) |
Find information on general health issues | 33 (11) | 60 (21) | 123 (42) | 76 (26) |
Communicate with other people who have the same condition | 68 (24) | 49 (17) | 111 (38) | 61 (21) |
Find information on prescription drugs | 75 (26) | 69 (24) | 97 (33) | 49 (17) |
Find information on health-related products such as herbal remedies and vitamins | 105 (36) | 82 (28) | 70 (24) | 35 (12) |
Communicate with doctors or other health professionals (including email) | 162 (56) | 64 (22) | 45 (15) | 20 (7) |
* Rows may not add to 100% due to rounding.
New mailing list members said they used ACOR mailing lists for a variety of purposes (
Reasons for using mailing lists (N = 293)
|
||||
|
|
|
|
|
Find out about the latest treatments for cancer | 6 (2) | 9 (3) | 89 (31) | 184 (64) |
Get information about how to deal with cancer | 1 (0) | 6 (2) | 105 (36) | 179 (62) |
Find out how to deal with the side effects of cancer treatments | 1 (0) | 17 (6) | 118 (40) | 156 (53) |
Get information about treatment options | 6 (2) | 11 (4) | 120 (41) | 153 (53) |
See how other patients with my cancer are doing | 3 (1) | 12 (4) | 137 (47) | 138 (48) |
Get support from other people with my cancer | 6 (2) | 30 (10) | 133 (46) | 121 (42) |
Help me make decisions about what is the best cancer treatment for me | 12 (4) | 26 (9) | 134 (46) | 117 (41) |
Help reduce my uncertainty about which treatments are best for me | 12 (4) | 32 (11) | 131 (45) | 114 (40) |
Help others | 3 (1) | 26 (9) | 153 (53) | 109 (37) |
Get ideas about how to talk with my doctor about my illness | 11 (4) | 41 (14) | 135 (47) | 103 (35) |
Feel less lonely | 28 (10) | 73 (25) | 100 (35) | 87 (30) |
* Rows may not add to 100% due to rounding.
We asked these new subscribers to specific ACOR mailing lists how they used the mailing lists in the past seven days since they had subscribed. This information was intended to serve as the baseline for subsequent comparison: 78% said they checked their email for messages four or more times in the last seven days; 41% of new subscribers reported spending 1 to 3 hours per day reading and responding to mailing list messages; and 30% said they exchanged private emails with 1 to 3 subscribers (
We also assessed reactions to group processes that took place on the mailing lists. Of new ACOR subscribers, 9% strongly agreed (62% agreed) that others on the list had similar experiences, 31% strongly agreed that they could express their opinions on the mailing lists, 27% strongly agreed that they could relate to other members' cancer experiences and that others on the mailing list gave them good ideas about how to cope with cancer, and 12% strongly agreed that they could disagree with other members' statements (
Mailing list use (N = 292)
|
|
||||
|
|
|
|
|
|
Number of times subscribers checked their mailing list email | 41 (14) | 23 (8) | 60 (20) | 87 (30) | 81 (28) |
Hours spent each day reading and responding to messages from mailing list | 155 (54) | 116 (41) | 9 (3) | 3 (1) | 4 (1) |
Number of different mailing list members with whom subscribers exchanged private emails | 155 (54) | 87 (30) | 33 (12) | 5 (2) | 6 (2) |
Number of times subscribers exchanged private emails with other mailing list members | 177 (62) | 65 (23) | 27 (9) | 7 (2) | 12 (4) |
Number of mailing list members subscribers called on the phone | 269 (93) | 18 (6) | 1 (1) | 0 (0) | 0 (0) |
* Number may vary slightly due to skip patterns.
Evaluation of mailing list experiences (N = 252)
|
||||
|
|
|
|
|
Overall, my experiences were similar to those of other members. | 7 (3) | 63 (26) | 154 (62) | 23 (9) |
I could relate to other members' experiences about cancer. | 3 (1) | 19 (8) | 160 (65) | 66 (27) |
Other people on the mailing list gave me good ideas about how to cope with cancer. | 3 (1) | 33 (13) | 147 (59) | 66 (27) |
I felt it was OK to express my opinions in the group. | 2 (1) | 18 (7) | 149 (61) | 77 (31) |
I felt it was OK to disagree with other members' statements. | 4 (2) | 48 (20) | 160 (66) | 30 (12) |
* Number may vary slightly due to skip patterns.
Help provided by the mailing lists (N = 276)
|
|||||
|
|
|
|
|
|
How much did the listowner (or listowners) give information that group members need? | 18 (7) | 27 (11) | 80 (32) | 80 (32) | 46 (18) |
How much did the listowner (or listowners) help the discussion? | 31 (13) | 27 (11) | 83 (33) | 71 (29) | 36 (14) |
How much help did other mailing |
27 (11) | 32 (12) | 97 (38) | 67 (26) | 33 (13) |
How much did the listowner (or listowners) help group members resolve conflicts? | 65 (28) | 23 (10) | 74 (32) | 50 (21) | 21 (9) |
How much help did you give to |
125 (49) | 48 (19) | 63 (25) | 16 (6) | 3 (1) |
* Number may vary slightly due to skip patterns.
The picture that emerges from Web surveys completed by cancer survivors who were new subscribers and received invitations to participate within about a week after subscribing to ACOR mailing lists is one of people who turned to the mailing lists for information and support, especially information. They particularly were looking for information about treatment, coping with side effects, and treatment options, with half or more of new subscribers providing these as reasons for joining mailing lists. Over 40% also cited support as one of the reasons they joined the lists. Although only 37% of respondents noted helping others as one of the reasons they joined the list, it is striking that even this proportion of people saw helping others as a reason for joining mailing lists. It is possible that helping others and receiving support become more important over time as the lists become more useful and central to subscribers when their tenure on it increases. Reading and responding to messages accounted for one to two hours a day, on average, for survey respondents. Follow-up survey data will help to put these numbers in context. We cannot predict whether the amount of time will increase, decrease, or stay the same. Also, we do not know the accuracy with which people report time spent online.
In view of their newness to the lists, it would be expected that many new subscribers would watch and wait, perhaps even
Although our full study will examine changes over time in subscribers' use of mailing lists and other outcomes of interest, here we focused only on the preliminary baseline survey, which would have the same strengths and weaknesses as similar cross-sectional surveys. For example, we can only report what subscribers said and cannot infer causal relationships.
It is unfortunate that so few minorities completed surveys. We do not know if that is a reflection of a small number of minorities using mailing lists, their lesser inclination to complete this survey, or both. Recent data indicate that 46% of African American adults are online versus 64% of white and 63% of Hispanic Americans [
Only 14% of the respondents in Schmidt and Andrykowski's study of Internet support for breast cancer were African American [
More research is needed to understand the many issues involved in asking patients to complete Web surveys, from assessing the physical, psychological, and cognitive demands of different question formats to examining ways to estimate true response rates. Online and other survey methods are substantially different. The lessons of one method cannot be transferred to others without more research [
Our conservatively calculated response rate of 21.4% is less than one would expect based on other survey formats. However, as Kraut et al [
Finally, although there is much that is challenging about Internet research, especially when conducted among cancer survivors, the Internet also offers tools not available in other modes. For example, we found time-stamp data extremely valuable in overcoming break-offs by permitting us to pinpoint areas in the survey that respondents had trouble answering. Moreover, the potential to include people from around the world is an attractive aspect of Web surveys.
While our data do not yet answer Eysenbach's questions about the impact of online communities [
This research was funded by a Robert Wood Johnson Foundation (RWJF) Health e-Technologies Initiative grant. This manuscript was supported by grant # 49152. This report does not represent the views of the RWJF. The authors would like to thank Gilles Frydman, President of ACOR, Charles Rawls, Chief Technical Officer of ACOR, and the members of the online survey programming team: Edward Maas, Ilya Zherebetskiy, Anthony Volodkin, Raj Goel, Bridget Rawls, Alanna Rawls, and Jack Brooks. This research could not have been done without ACOR's role in identifying lists and listowners, sending invitations to subscribers, tracking survey data, programming the survey, and communicating with listowners. A number of ACOR listowners have been important partners. We are grateful to Tara Strigo, MPH, and Rachel Mandal, MSc, at UNC for their roles in managing the survey, as well as Bernard Glassman, MA, 3AM Communications. In addition, we thank Dr. Neeraj Arora, National Cancer Institute, and Dr. Lari Wenzel, University of California at Irvine, for their contributions to survey development and Martha Songayllo for graphics help. Finally, we are grateful to all the ACOR mailing list subscribers who completed surveys.
None declared.
Association of Cancer Online Resources
Computer Health Enhancement Support System
electronic support groups
Health eCommunities
Robert Wood Johnson Foundation