A total of 1172 women with either breast or gynecologic cancer (or both) were assessed in a cross-sectional approach. Patients were recruited to answer an electronic online survey (Questback) through social media, special patient Internet platforms, self-help group leaders, patient networks (eg, Breast Cancer Aid Germany or BRCA Network), and cancer counseling centers. Duplicate entries were avoided by preventing users with the same IP address further access after completion of the survey. Furthermore, consecutive inpatients and outpatients were asked whether they would like to participate in the study (paper-and-pencil questionnaire in the department of gynecology at the University Hospital Tuebingen, Germany). The self-reported paper-and-pencil or self-reported online questionnaires took participants an average of 20 minutes to complete. Eligibility criteria were defined as an adult (age 18 years and older) with breast cancer, gynecologic cancer, or both with sufficient language skills to complete a set of questionnaires. Participation was voluntary and anonymous; no personal identifying information was collected from the patients. The beginning of the questionnaire included the consenting page. Of the 1172 participants assessed, 41 did not meet the eligibility criteria because of another cancer diagnosis. Incomplete datasets (less than 80% response rate) were excluded, resulting in a final dataset of 716 participants, with 581 surveys completed online and 135 surveys completed as paper-and-pencil questionnaires. The local ethics committee of the University Hospital Tuebingen approved the study protocol.

Descriptive statistics, mean and standard deviation, frequencies, percentages, and chi-square statistics for categorical variables were performed using SPSS 21 for Windows (IBM Corp). Statistical analysis was performed to search for a relationship between patient preferences for psycho-oncological eHealth care and psychological burden with distress, QoL, and need for psychosocial support. Data were normally distributed. Chi-square statistics were used to examine the data for associations between the psychological burden and the preference for psycho-oncological content topics and ePOAs. We computed for distress (highly distressed, distressed, not distressed) and QoL (low, middle, high) in a 3×2 distribution table and for HSI (needing psychosocial support vs not needing psychosocial support) in a 2×2 distribution table. For this purpose, the responses of items were dichotomized (preferences: important vs nonimportant; eHealth applications: suitable vs nonsuitable). Missing data only occurred for the patient preference surveys. The overall mean of missing values was estimated as 2.075%. Missing values were considered only if at least 80% of each of the questionnaires had been completed. Using the Little missing completely at random test, it was confirmed that the data were missing randomly. The expectation-maximization algorithm was used to input the missing data [48]. All of the statistical tests were 2-sided, and P<.05 was considered statistically significant.

Of the 1172 patients who responded, 716 (61.09%) datasets met the inclusion criteria, showed acceptable quality, and were included in the study. The mean age of participants was 50.2 (SD 10.3) years (range 25-83 years). Nearly 80.4% (576/716) of the patients were primarily diagnosed with cancer, and 12.2% (87/716) of participants were diagnosed with metastasis; 11.0% (79/716) were experiencing a recurrence of the past cancer diagnosis. The frequencies of other disease-related and demographic variables and mean values and standard deviations of the Distress Thermometer, EQ-5D-3L, and HSI questionnaires are presented in Table 1.

The 19 content topics for a psycho-oncological intervention were rated by the patients (see Figure 1). The highest rates reached anxiety (675/697, 96.8%), followed by ability to cope (674/696, 96.8%), QoL (657/696, 94.4%), depressive feelings (655/695, 94.2%), and adjustment to new life situation (654/700, 93.4%).

Lower preference rates were reached by spirituality (308/692, 44.5%), sense-making (632/692, 66.0%), and sexuality (495/692, 71.5%).

Almost all of the eHealth applications for conveying the content topics were considered by more than 50% as suitable. Web-based application/info home page (540/695, 77.7%) was considered to be the most suitable compared to other eHealth applications. Blogs or chats were considered suitable or very suitable (470/694, 67.7%). More than half considered the eHealth applications info email (387/694, 55.8%) and consultation hotline (361/694, 52.0%) suitable or very suitable. Videoconference was judged by the least number of patients (285/687, 41.5%) as suitable for psycho-oncological care (see Figure 2).

Preferences for all content topics were equally distributed in the subgroups distress and QoL. Preferences were not dependent on high, middle, or low distress or high, middle, or low QoL. Also, in the context of needing psychosocial support, patients preferred the same content topics for a psycho-oncological intervention. We found no differences between participants with different levels of distress or QoL concerning the preferred content of eHealth interventions. Furthermore, time since diagnosis or prognosis as well as recruitment (eg, hospital vs Facebook) had no influence on needs (data not shown).

Our survey explored for the first time the perceived demands and needs for a psycho-oncological eHealth intervention among women with breast and gynecological cancer. Furthermore, we investigated whether there is a relationship between psychological burden (distress, QoL, and need for psychosocial support) and content topic relevance. In this sample of 716 cancer-affected women, we found distinctively relevant content topics for eHealth interventions. The content topics of ability to cope, anxiety, depressive feelings, or adjustment to new life situations have a high impact on eHealth interventions and, in turn, reflect the needs and demands for psycho-oncological care of cancer-affected women. Spirituality, sense-making, and dealing with children had no high relevance for eHealth interventions in our sample. Furthermore, Web-based application/info home page, info email, and chats and blogs were identified as very suitable and suitable for conveying psycho-oncological content topics to the patients. We found that preferences for specific content topics and eHealth applications were equal between patients with high and low burden (experienced distress and QoL). In addition, the need for psychosocial support did not influence the demands and needs of the patients. To summarize, interestingly, women with cancer experience—independently of their psychological burden—have the same demands and needs for psycho-oncological content topics. Furthermore, they expressed the same demands for eHealth applications in psycho-oncological care.

We found in our survey that the most preferred type of psycho-oncological eHealth interventions are Web-based applications or info emails. These results are in line with previous studies that also identified high preferences for eHealth applications [10,33,34,49]. Moreover, we found that eHealth applications may have potential beneficial effects for specific psycho-oncological content topics like anxiety, coping, depressive feelings, self-esteem, or adjustment to new life situations (see Figure 1). It seems that ePOAs have the potential to support cancer-affected women in the context of delivering information, feelings of control, self-efficacy, or self-management during the time of dealing psychologically with the disease. Jansen et al [50] also found an overall positive attitude toward self-management and eHealth among different cancer survivors. Furthermore, they determined that men were more likely to report supportive care needs regarding healthy lifestyle programs, and they are in general highly interested in eHealth. Børøsund et al [51] described the use of patterns in Web-based illness management support among prostate and breast cancer patients. Regarding the use of Web-based support applications, they determined that lower levels of social support and higher depression scores were more influential among women with breast cancer, and symptom distress was more influential for men with prostate cancer. It seems that cancer-affected men are more likely to participate in eHealth applications that offer lifestyle elements, and women with cancer are more willing to participate in eHealth applications that also contain psycho-oncological elements.

The psycho-oncological content topics that were evaluated for eHealth applications (see Figure 2) are similar to the topics discovered in other studies. Different researchers discovered high unmet psychological and psychosocial needs in cancer patients and reported that it is urgently necessary to further evaluate and address these specific demands and needs in future and modern eHealth intervention [9,52,53]. In our survey, sexuality, sense-making, and spirituality were considered to be less important, a finding which diverges from other research findings [54,55]. It could be that these content topics are more suitable for face-to-face interventions and are not suitable for eHealth interventions. The content topic self-esteem was rated extremely high and considered to be very important by the patients for an eHealth intervention. In previous psycho-oncological interventions, this content topic was mostly neglected and not taken into account, especially in eHealth interventions. Furthermore, we have found that more than 50% of the patients reported a high preference for ePOAs independently of the experienced psychological burden. This is in line with the findings of Jansen et al [50] who reported that the perceived needs for supportive care, including healthy lifestyle programs, were highly accepted, and in general, cancer survivors had a positive attitude toward eHealth. Different from our findings, Jansen and colleagues [50] found that the attitude was associated with QoL [50]. An et al [30] demonstrated that cancer patients perceived more social support from the Internet when they actively posted or shared contents than when they used the Internet solely as an informational resource [30]. We also determined that chats and blogs were highly accepted by our patients. In addition, studies reported that future psycho-oncological interventions should consider daily practice and the local accessibility as ePOAs have the potential to close this gap [24,56,57]. Additionally, cancer survivors were positive toward ePOAs that enable them to enhance their own QoL and support them in finding tailored supportive care [57]. It was shown that eHealth applications are well accepted for therapy assistance in general (like patient-physician communication) and eHealth programs as a part of usual health care may be promising [29,50,57]. A promising result of our study is that a substantial group of participants in need of supportive care prefer ePOAs for the delivery of adequate psychosocial care. Furthermore, they rate eHealth as adequate for specific psycho-oncological content topics like anxiety, ability to cope, depressive feelings, self-esteem, or adjustment to new life situations.

Our survey study was based on a large sample (N=716) of patients diagnosed with breast cancer, gynecologic cancer, or both. Our use of various and novel recruitment strategies (Internet links, Facebook, blogs, flyers) led to a large proportion being included through the online questionnaire (n=581). This shows that eHealth is especially targeting patients with eHealth literacy, and therefore our results can be considered as representative of these patients [58]. Nevertheless, there are limitations in the sample selection and generalizability of this survey. Our survey cohort was homogenous, mostly younger, white, and highly distressed. Furthermore, it is important to note that mainly women with breast cancer (91.1%) participated in our survey. Therefore, a recruitment bias can be assumed in our study. However, this trend has been observed in similar studies, and it also reflects reality [51]. Breast cancer has the highest tumor prevalence among women, and various studies show that this patient group suffers mostly under high psychological distress [4,5] and younger patients prefer eHealth applications more than older patients. Further studies including other tumor entities and male patients are needed. Our results agree with findings of former studies [50,51]. The lack of diversity also does not allow extrapolation of study results to statements concerning other tumor entities and men. Our self-generated patient preference survey has not undergone formal reliability and validity testing. Hence, a validated questionnaire for these research questions did not exist when the study was performed.

Our findings show high preference rates for eHealth applications independently of experienced psychological burden among women with cancer. Furthermore, ePOAs may have a high benefit for women with cancer, especially when they address psycho-oncological content topics like anxiety, ability to cope, depressive feelings, self-esteem, or adjustment to new life situations. ePOAs have the potential to help patients overcome disease-related burden and reduce barriers in psychosocial care. However, they can encourage patients who they believe are not sufficiently burdened to participate in common psycho-oncological interventions [24,59]. Our findings encourage the development of further innovative ePOAs that specifically focus on the evaluated psycho-oncological content topics (see Figure 1). In summary, the findings of this study indicate that ePOAs are a promising medium to improve psychosocial care and enhance individual disease management among women with cancer.