As a first step, an environmental scan was conducted consisting of three components: a review of published literature; a review of gray literature (unpublished literature from relevant Internet sources); and interviews with healthcare providers, researchers, patient advocates, and developers of other online patient education tools. The purpose of the environmental scan was to identify risk factors for developing neutropenia among patients with cancer, effective infection control practices to be implemented by patients with cancer, current practices of healthcare providers on infection control and patient education, and barriers and facilitators to healthcare providers’ use of an online education program for infection control with their patients. The environmental scan also identified previously developed online education programs for patients with cancer and potential challenges in the development of the programs.

Findings from the environmental scan informed the development of a logic model (www.preventcancerinfections.org/sites/default/files/LogicModel-FINAL.pdf) that guided the development and implementation of the website. The logic model was based on the understanding of information that patients and caregivers need, including awareness and knowledge about neutropenia and subsequent risk of infection, knowledge and skills about how to prevent infections, and knowledge and skills about how to quickly recognize and control infections when they do occur.

Following the formative research, activities focused on website design and functionality and the creation of a risk assessment algorithm for the website. The environmental scan revealed certain factors (e.g., age, cancer type, chemotherapy regimen, the Eastern Cooperative Oncology Group performance status). The risk factors were used to develop an algorithm estimating a patient’s risk for developing neutropenia while receiving chemotherapy. The algorithm determines risk by assigning scores to responses to questions related to demographic and clinical characteristics (see Figure 1). Based on the answers, respondents are placed in either a low-risk or high-risk category and, according to this classification, provided with evidence-based information to reduce their risk for developing infections during chemotherapy treatment. The messages help patients and caregivers recognize the signs and symptoms of an infection and what to do if they develop any of these signs and symptoms.

In addition, the CDC assembled an EAB representing the fields of oncology and infection control, nursing, and patient advocacy. The role of the EAB was to provide input during the development of the website and to help with its dissemination into the patient and provider community. From October 2011 to January 2012, 11 EAB members (selected by the CDC based on expertise) contributed 2–3 hours per month completing assigned tasks and received no monetary compensation for their contributions to the project. Two EAB members still serve in an advisory capacity when needed.

During this phase, the EAB reviewed findings from the environmental scan and provided input into the development of the algorithm for assessing a patient’s risk for developing neutropenia. During later phases of the project, EAB members reviewed focus group findings, evidence-based content for the website, branding concepts, a pilot testing plan, and dissemination strategies. Members of the EAB also assisted in other specific tasks. For example, the CDC conducted a site visit to one of the member’s outpatient clinics to better understand potential obstacles and facilitators in incorporating a web-based educational tool into a practice. In addition, several EAB members contacted representatives from their respective professional organizations to increase awareness, dissemination, and use of the website.

A webinar was conducted with five researchers and health-care providers (including urban and rural oncology nurses, an oncologist, and an infectious disease physician) to further inform development of the website. Webinar participants were selected based on their experience working with patients with cancer or their expertise in infection control. The main objectives of the webinar were to confirm the findings of the environmental scan (i.e., the risk factors for developing neutropenia among patients with cancer), reach a consensus on questions for the website’s algorithm, review the website’s draft content, and obtain input on the delivery, use, and limitations of the website.

After the webinar was completed, 11 focus groups were conducted with a total of 64 cancer survivors, caregivers, and healthcare providers to better understand the perspectives of these audiences on preventing infections during cancer treatment and an online patient education tool (see Table 1). The focus groups were an important component of the concept validation phase in that they allowed direct interaction and conversation with the audiences for whom the website is intended.

Recruitment for all groups was conducted through local cancer centers, support groups, professional medical associations, and patient advocacy organizations. Content from the focus groups was transcribed and reviewed to conduct a content analysis of themes across all 11 focus groups that informed development of the website. As an example, to increase user acceptance, three preliminary website designs were evaluated by the patient, caregiver, and healthcare provider focus groups. Participants were asked to provide feedback on colors, images, graphic elements, structure, and content. One design was selected most frequently across all focus group participants and was ultimately chosen, with participants’ suggested revisions incorporated.

A web developer and database programmer used HTML and Microsoft.NET^™ technologies to create the site. Many of the development tasks centered on integrating website design specifications, algorithm logic, and tailored infection control and prevention messages. The environmental scan revealed that the Internet has become a leading source for health information. However, to reach targeted audiences and ensure adoption of recommended practices and behaviors, information must be appealing, easy to understand, and educational (Bensley et al., 2004). During the development phase, the team was mindful of maintaining readability of the materials using plain language tactics to ensure a user-friendly website, such as incorporating bullet lists and short paragraphs and sentences (Plain Language.gov, 2011). The project team aimed to keep the website content to a sixth grade reading level; however, the necessary use of medical terminology increased the content to a seventh grade level. As a result, a glossary was added to provide users with definitions of this terminology in lay terms.

A development approach was employed that enables future modifications with minimal effort, therefore providing maximum flexibility for updates. The development approach also included the use of the software development life cycle method to guide the process and ensure the website met industry standards (Conger, 1993). Once a preliminary version of the website was complete, pilot testing was conducted with 24 respondents (physicians, nurses, caregivers, and patients) who had previously participated in the focus groups. Data were collected through a short, web-based survey. Information gathered from pilot testing, including quantitative satisfaction results and qualitative comments, were summarized and minor edits were made to clarify glossary terminology and risk assessment completion instructions.

Following these website revisions, promotional activities began by targeting key partner groups such as patient advocacy organizations, federal agencies, state and national organizations, as well as oncology, nursing, and infectious disease associations and societies. To garner support for the website and inspire key groups to promote and embed the website into their organizations’ existing websites, social media channels, and email blasts/communications to their memberships, the CDC conducted three teleconferences with these various partner organizations the week before the website was launched. The partners received a variety of resources, including the embargoed press release, fact sheets, posters, key messages, web buttons, tweets, e-cards, and an educational video explaining how to use the website and its benefits. Many partners posted this information on their organizations’ websites, sent out a separate email to their membership, and/or posted one of the web badges (directly linking the user to www.preventcancerinfections.org) on their own organization’s site. Partner efforts helped to drive traffic to the website and added to its exposure.

Using multiple marketing and dissemination channels to reach a diverse target audience was critical to the launch of the website on October 25, 2011. In addition to partners, the development group targeted traditional print and Internet-based media to ensure appropriate saturation for maximum audience exposure. Online and print coverage alone reached more than 12 million readers and helped drive the public to the site. In addition, a satellite media tour was conducted that had the potential to reach 2.9 million people. The developers also used social media outlets, including Facebook and Twitter, which garnered 357 total tweets, reaching more than 885,000 people. By using this multifaceted approach, the launch of resources alone reached more than 168 million impressions and more than 273 million impressions as of May 2013.

A maintenance phase has been in place since the launch. The goal is to track and document user feedback, monitor risk assessment questionnaire data, and track site usage through web analytics. Although users do not receive responses to their comments, the feedback feature provides a mechanism for capturing accessibility, content, and/or general user concerns. In addition, data gathered from completed risk assessment questionnaires provide a snapshot of the demographic and clinical characteristics of website users. Web analytics data (e.g., site usage, sources of site traffic, average time on site, most frequently accessed content) also are analyzed and reported weekly (see Table 2).

Initial evaluation activities also were conducted during this phase to expand the understanding of how the website is used and incorporated into outpatient clinical settings. Site visits to two outpatient cancer treatment facilities were conducted and included in-depth interviews with staff and patients. Both facilities were identified and selected based on their staff’s involvement in the EAB and their use of the website within their facilities. Findings from these visits provided a concrete understanding of how the website and its collateral materials are used in the outpatient clinical setting (see Figure 2). Summary reports were developed, providing key information regarding the availability of process (e.g., the various ways the website is used by providers and patients) and outcome data (e.g., infection rates) that may help to inform the design of a more rigorous evaluation of the website.

As a final evaluation activity, interviews were conducted to gather input and lessons learned from two websites with similar content, purpose, and audiences. Telephone interviews with individuals who were responsible for the development, marketing, and maintenance of these websites were conducted strictly for information-gathering purposes. All qualitative and statistical data captured during this phase will inform future website enhancements, as well as marketing, website optimization, and evaluation activities.