AYAs' online information and eHealth needs: A comparison with healthcare professionals' perceptions

Abstract Background Adolescents and young adults (AYAs) diagnosed with cancer fulfill their cancer‐related information needs often via the Internet. Healthcare professionals (HCPs) have a crucial role in guiding patients in finding appropriate online information and eHealth sources, a role that is often overlooked. Misperceptions of AYAs' needs by HCPs may lead to suboptimal guidance. We aimed to examine the extent to which AYAs' online information and eHealth needs corresponded with HCPs' perceptions of these needs. Methods Two cross‐sectional online surveys (AYAs, n = 299; HCP, n = 80) on online information and eHealth needs were conducted. HCPs provided indications of their perceptions of AYA's needs. Results AYAs reported significantly more online information needs compared with HCPs' perceptions regarding: survival rates (AYA = 69%, HCP = 35%, p < 0.001), treatment guidelines (AYA = 65%, HCP = 41%, p < 0.001), return of cancer (AYA = 76%, HCP = 59%, p = 0.004), “what can I do myself” (AYA = 68%, HCP = 54%, p = 0.029), and metastases (AYA = 64%, HCP = 50%, p = 0.040). Significantly more unmet eHealth needs were reported by AYAs compared with HCPs relating to access to own test results (AYA = 25, HCP = 0%, p < 0.001), request tests (AYA = 30%, HCP = 7%, p < 0.001), medical information (AYA = 22%, HCP = 0%, p = 0.001), e‐consult with nurses (AYA = 30%, HCP = 10%, p < 0.001), e‐consult with physicians (AYA = 38%, HCP = 13%, p = 0.001), and request prescriptions (AYA = 33%, HCP = 21%, p = 0.009). Conclusion AYAs' online information and eHealth needs are partially discrepant with the impression HCPs have, which could result in insufficient guidance related to AYAs' needs. AYAs and HCPs should get guidance regarding where to find optimal information in a language they understand. This may contribute to AYAs' access, understanding, and satisfaction regarding online information and eHealth.


| INTRODUCTION
Every year about 3800 adolescents and young adults (AYAs), aged 18-39 years, are newly diagnosed with any type of cancer in the Netherlands. 1 The relative 5-year survival rate for AYAs with cancer in the Netherlands is currently over 80% and is still rising. Even though AYAs are (much) younger than most cancer survivors, they are commonly treated in general oncology departments with little attention to their age-specific questions. 2,3 However, research shows that AYAs have different needs related to their developmental phase of life such as individualized information, advice, communication, services, counseling (emotional and psychological), social support, and social relationships compared with older cancer survivors (e.g., study, work, fertility, family). [4][5][6] AYAs are digital natives, which mean that they are native speakers of the digital language and pervasive users of technology. 7 Next that, the amount of technology and Internet use, in general, has been rising (e.g., Skype, Instagram, WhatsApp). 8 Therefore, their information needs often relate to the Internet, in addition to receiving information from healthcare professionals (HCP) directly. 9,10 However, previous research has shown that 50% of AYAs' online information needs are unmet, which negatively influences psychological distress, anxiety, and, in turn, quality of life, [11][12][13][14] but, importantly, it is unclear to what extent HCPs have an adequate perception of these needs. Information provision is of great importance in cancer care as fulfilled information needs may support adjustment to cancer, satisfaction, understanding, knowledge, patient participation, shared decision-making, sense of disease-control, acceptance, increasing compliance, rational expectations, selfcare, long-term well-being, and coping. 9, [15][16][17][18][19] The way in which patients use the Internet has been described by Eysenbach 20 and is, despite major changes in Internet usage, still perceived as a highly suitable model. 21 The model includes the following four aspects of Internet usage: (1) content (i.e., online information), (2) communication, (3) communities, and (4) e-commerce. However, since Eysenbach developed his model, e-commerce (in a health setting) has been superseded by the broader concept of eHealth, which is the term Van Eenbergen et al.
(2020) preferred, which facilitates self-management. 21 Self-management refers to the use of electronic health records (including personal health information, medical history, medication, allergies, and hospital consult reports) 22 or electronic diaries (e-diary) in which patients are, for example, able to monitor pain intensity and medication adherence. 23 However, in this study, we only address online information and eHealth since in this research, we only study self-management without interactions with others.
Conclusion: AYAs' online information and eHealth needs are partially discrepant with the impression HCPs have, which could result in insufficient guidance related to AYAs' needs. AYAs and HCPs should get guidance regarding where to find optimal information in a language they understand. This may contribute to AYAs' access, understanding, and satisfaction regarding online information and eHealth.

K E Y W O R D S
adolescent, AYA, cancer, eHealth, healthcare professional, online information needs, young adult may not be well aware either. This can lead to inappropriate or insufficient information provided to patients. 17,24,36 Although various studies have examined AYAs' needs for online information and eHealth, little attention has been paid to a comparison of HCPs' perception of these needs. This is crucial as realistic perceptions of these needs contribute to appropriate guidance from HCPs. We, therefore, aim to examine (1) AYAs' online information seeking behavior and online information and eHealth needs; (2) HCPs' perceptions of AYAs' online information seeking behavior and online information and eHealth needs; and (3) the extent to which AYAs' indications and HCPs' perceptions correspond.

| Participants
Two cross-sectional surveys on online information and eHealth were administered as follows: one among AYAs and one among HCPs. A flow chart of both data collections is presented in Figure 1. Data collection among AYAs was performed within PROFILES Registry (Patient Reported Outcomes Following Initial Treatment and Long Term Evaluation of Survivorship). 37 AYAs were recruited through the distribution of the AYA questionnaire via Kanker.nl (the leading online platform in the Netherlands with information about cancer), newsletters of the Dutch Federation of Cancer Patient Organizations (NFK) and other patient organizations, social media, and three hospitals. Patients could participate if they were diagnosed with cancer during the AYA age. However, patients were excluded if cancer was diagnosed before 2010, as the Internet was less accessible and less common to use before 2010, and Internet use has been increasing since 2010. 38 HCPs who were approached were HCPs from hospitals that the AYA care network in the Netherlands maintained contact with. Thus, these HCPs either had some form of AYA care or were willing to provide it. Participants received no compensation for participation. The questionnaire for HCPs was programmed in Qualtrics. HCPs were recruited via a mailing from the AYA healthcare network. The anonymity of patients and HCPs was guaranteed. This study was approved by the Research Ethics Review Committee of Tilburg School of Humanities and Digital Sciences (internal code: REDC 2019.104).

| Measures
The AYA questionnaire was based on an existing questionnaire from the previous research. 21 This questionnaire was adapted based on feedback from 6 AYAs (6 female, 26-33 years, diagnosed with cancer 0-6 years earlier) and AYA project leaders. A list of 67 items was generated, which was discussed with both AYAs and HCPs, after which adjustments were made. Adjustments included adding and removing questions, changing the order of questions, and adding minor nuances to questions.
Demographics including age, gender, education, year of diagnosis, type of cancer, and treatment modality were collected for AYAs. HCPs' demographics included gender, age at the time of questionnaire, profession, and employment or activity in the AYA expertise center or team.
We asked AYAs to indicate their online informationseeking behavior across disease phases (e.g., how often did you search for information about cancer on the Internet in the period that I had to wait for surgery/treatment?). HCPs had to indicate their perceptions of AYAs' online information seeking behavior across disease phases (i.e., can you estimate how many of the AYAs search for information on the Internet?). Online information seeking behavior across the disease phases (1 = just before diagnosis, 2 = right after diagnosis, 3 = during treatment, and 4 = follow-up) was measured with different values in HCPs (1 = all of them/ many, 3 = none) compared with AYAs (1 = daily/several times a week, 3 = never). However, these questionnaires were compared in the analyses since HCPs are not able to estimate how often AYAs seek information. In this way, it was still possible to see to what extent the perceptions of HCPs correspond with the actual search behavior of patients.
Furthermore, needs regarding online information (e.g., did you search for information about fertility and wanting children after cancer during your illness and recovery period?) were assessed (0 = no, 1 = yes). HCPs were asked to make estimations about AYAs' online information needs (i.e., please mark which topics you think are important to AYAs, e.g., fertility and childbearing after cancer) (0 = no, 1 = yes).
Finally, eHealth needs (0 = no, 1 = yes) and possibilities (0 = no, 1 = sometimes, 2 = yes, and 3 = do not know) (e.g., what online options did you have and what wishes do you have in being able to request results of examinations?) were examined. HCPs were asked to indicate AYAs' eHealth needs (0 = no, 1 = yes) and possibilities (0 = no, 1 = yes, and 2 = do not know) (i.e., on each line, tick whether something is possible and whether you think AYAs would also find it desirable) (e.g., requesting test results).
Online information needs were considered as "unmet" when certain online information was indicated as needed but no possibilities were available. When it was indicated that no possibilities were available, this means that the person was not able to find certain online information. This does not automatically mean that the online information was not available.

| Statistical analyses
Statistical analyses were conducted using SAS version 9.4 (SAS Institute, Cary, NC, 1999). The sociodemographic and clinical characteristics of the AYAS and HCPs were described as percentages or means and standard deviations. Searched topics on the Internet and eHealth needs were described as percentages. The searched topics on Internet were compared between AYAs and HCPs using chi-square analyses (or Fisher's exact tests when sample sizes are small). Two-sided p values of <0.05 were considered statistically significant. We decided not to conduct Bonferroni correction because of the exploratory nature of the study.

| AYAs
Just before diagnosis, 45% searched the Internet for information about cancer daily to several times a week, whereas 48% never did. Directly after diagnosis, the number of patients who searched for information daily to several times a week increased to 71%, while those who never searched the Internet decreased to 17%. During treatment, 65% of patients searched the Internet daily to several times a week. During follow-up, only 15% searched daily to several times a week and 40% did never search for information about cancer. Figure 2 shows when and how often AYAs searched the internet for information regarding cancer.

| HCPs
According to HCPs, AYAs searched the Internet mostly relating fertility and child wish (100%), late effects of treatment (88%), consequences of treatment in general (87%), treatments (85%), and involvement in treatment decision (84%). Regarding psychosocial (age-specific) information the most frequently searched topics by AYAs according to HCPs were consequences for a young family (98%), insurance and/or mortgage (95%), sexuality and intimacy (89%), return to work and/or study (89%) and meeting possibilities fellow sufferers (81%). HCP indications of their perceptions of AYAs' online information seeking regarding behavior regarding medical and psychosocial (age-specific) topics are shown in Table 2 and 2B, respectively.

| Comparison of AYAs and HCPs
Furthermore, AYAs' eHealth wishes and needs were examined, as well as the relating perceptions of HCPs, of which the significant differences are shown in Figure 4A. AYAs reported significantly more unmet eHealth needs regarding access medical information

| DISCUSSION
This study examined AYAs' online information and eHealth needs, HCPs' perceptions of these needs, and the extent to which AYAs' indications and HCPs' perceptions correspond. Results showed that AYAs mostly sought information directly after diagnosis and during the treatment phase, whereas HCPs expected AYAs to seek information in similar frequencies across the different phases. Furthermore, our results showed that HCPs underestimate online information seeking of AYAs regarding several medical (e.g., return of cancer, treatment guidelines, metastases) and psychological (i.e., what can I do myself) topics. HCPs also underestimated several eHealth needs of AYAs, such as assessing test results, e-consults with physicians or nurses, and requesting tests or prescriptions. However, it was shown that HCP's overestimate eHealth needs regarding peer contact. Such underestimations and overestimations could result in T A B L E 2 (A/B) Medical and psychosocial (age-specific) topics searched for on Internet during and after treatment by AYAs and HCPS perspective. T A B L E 2 (Continued) a mismatch in online information and eHealth guidance and needs between HCPs and AYAs. As a result, HCPs may provide (1) too little information about certain topics because they think that AYAs read about them online and (2) too few suggestions where this kind of information can be found online. This is undesirable since meeting information needs is crucial for several health outcomes of patients with cancer (e.g., knowledge, understanding, shared decisionmaking, long-term well-being, coping). 9,[15][16][17][18] It has been shown that there are inconsistencies between AYAs' needs for online information and eHealth and how HCPs perceive these needs, which may explain why the provision of information from HCPs to patients often does not match the needs of the patient. 24,39,40 This could be explained by HCPs providing information that they think is important, instead of checking what the AYA finds important. 41 In addition, this problem may be greater in this specific patient group due to the age difference between AYAs and HCPs (range 24-62 years), which appears to complicate communication between them. 42,43 Furthermore, previous research has shown that many HCPs experience difficulties in communicating and connecting with AYAs, 44 which could indicate that during the consultations, the needed topics are therefore not discussed. It is important that HCPs are attentive to the patient's preferences and that they realize the urgency to guide AYAs so that AYAs are satisfied with the usage of appropriate online information and eHealth possibilities. 45 To achieve this, AYAs and HCPs should get guidance regarding where to find optimal information in a language they understand. Guidance is essential in optimal online information use, to ensure that AYAs find reliable and complete information. One of the treating HCPs should refer to one place on the Internet, where all age-tailored services and information are presented. In the Netherlands, this is currently being developed as part of the COMPRAYA study. 46 Next to that, HCPs could refer patients more frequently to electronic health records. When necessary, HCPs should provide guidance in finding and using these sources to encourage optimal usage.
In addition, the type of online information needs to be indicated by AYAs and HCPs differs. AYAs seem to be mainly concerned about the cancer diagnosis and treatments they are currently in, including topics such as treatment, survival rates, lifestyle and nutrition, and activities and sports during treatment. This means that AYAs are primarily focused on survival and (re)starting life. Meanwhile, HCPs are also concerned about the future of AYAs (i.e. after treatment), which may include topics such as children, sexuality, finances, and resume work. It is important to emphasize that HCPs should actively contribute to highlighting this long-term topics. 47 Interestingly, AYAs report a high level of need for online and offline peer contact. AYAs report relatively high peer contact needs (almost 60%) compared with the average cancer population (around 10%-20%). 21,48 Therefore, it would also be advisable to allow for the possibility to meet peers online for information. This recognized relevance of peer support seems to have increased significantly in recent years from an HCPs' perspective, as it was previously seen as less relevant. 49 To our knowledge, there is no research that has studied this. However, over the past 10 years, patient organizations in the Netherlands have been visibly active with topics such as cocreation, patient journeys, and shared decision-making. The healthcare sector, and thus HCPs, is giving more substance to these topics in the last 10 years The growth in recognized relevance could be explained by the increasing relevance of patient advocacy. The opportunities for peer support that the Internet, and specifically social media, offer can also play a role in this. The fact that care providers are now well aware of the need for and importance of contact with peers may improve guidance for opportunities in peer contact.

| Limitations
Several limitations of this study need to be acknowledged. First, since nearly 80% of the patient sample is female, this is not an adequate representation of the AYA patient group. This could possibly be explained by the high proportion of breast cancer patients. However, this may have led to partly nongeneralizable results for the general AYA population, since women report more information needs and seeking behavior compared with men. 36,50 Gender also influences the type of online information needs, as, for example, women have more interpersonal and emotional information needs, whereas men have more information needs regarding sexual function and fertility. 51 The results must therefore be interpreted with caution. Second, this study mostly includes respondents with higher levels of education. Higher educated patients tend to have more information needs compared with lower educated patients. [52][53][54][55] Highly educated people are more self-reliant and are therefore more likely to gather online information to help them take care of themselves, while patients with lower levels of education have higher support service needs, which means that they are more likely to use specific AYA care. 54 Third, this study includes a native respondent group. This group may be more likely to search online for cancer information than ethnic minorities since people from ethnic minorities generally experience more difficulties and less trust in sources during online information seeking. 56 Specific attention should be paid to this group because they are less proficient in the language. Therefore, careful and culturally appropriate consideration should be given to guidance regarding online information. 56 Fourth, no comparison was made between the outcomes of HCPs who do and do not work in an AYA expertise center or AYA team. However, an increasing amount of AYA expertise centers emerge in the Netherlands, making it increasingly relevant.

| CONCLUSION
It can be concluded that HCPs both overestimate and underestimate online information and eHealth needs of AYAs. This indicates that information that is provided does not always meet the patient's needs, while it is crucial to meet their needs since adequate information is vital for coping with cancer. It is advisable that AYAs and HCPs should get guidance regarding where to find optimal information in a language they understand. This may contribute to AYAs' access to, understanding of, and satisfaction with online information and eHealth.