Oregon Health Go Local: A Retrospective Look

Abstract

MedlinePlus® Go Local is the National Library of Medicine and National Institutes of Health's Internet resource that allows individuals to find health services in their local communities. Oregon Health Go Local recently joined the network of active Go Local projects. This paper describes the planning and development of Oregon Health Go Local, with a discussion of the success and challenges of initiating such a project. Funding, marketing, volunteer coordination, data acquisition, and taxonomy are discussed.

INTRODUCTION

With 34 active projects and three in development, MedlinePlus® Go Local is an online resource that a growing number of Internet users can use to locate health-related services in their local communities.¹ Go Local allows users of the MedlinePlus health topics pages to link to community health services that are directly related to their health topic of interest. In spring of 2009, after more than a year of development, in collaboration with the National Library of Medicine Oregon Health & Science University (OHSU) Library, launched Oregon Health Go Local and joined the ranks of other states offering a Go Local service.

PROJECT BACKGROUND AND HISTORY

In 2006, Pew's Online Health Search document indicated that 80% of American Internet users went online for health information. The same report suggested that three quarters of health seekers did not check the source and date of health information they found online. Alarmingly, nearly 60% in this group indicated that the information they found in their last search affected a decision about how to treat an illness or condition.² At the time, an emerging tenant of librarianship was to “go where your users are” and provide context-sensitive help at the point of need.³ Because online health information consumers make medical decisions without adequate evaluation of information quality, there is an opportunity for libraries to help direct them to resources and services that would be likely to positively impact medical decision making and ultimately their health. Initiating a MedlinePlus Go Local project represented a unique opportunity to help make this happen. The Head of OHSU Library's Instruction, Research and Outreach Department was an early champion of Go Local and sought opportunities for getting the project funded. Under her guidance in fall of 2007, the library submitted two grant proposals that, if funded, would help Oregon Health Go Local get off the ground.

Funding

One of the initial barriers to implementing Oregon Health Go Local was finding adequate funding. Each year the Oregon State Library allocates Library Services and Technology Act (LSTA) funds to library projects that offer widespread essential improvements to library services in Oregon. The OHSU Library targeted LSTA funds in combination with funds from the National Library of Medicine as a viable way to initiate the project. In spring of 2008, the OHSU Library learned that Oregon Health Go Local would receive nearly $25,000 from the National Library of Medicine and almost $95,000 in LSTA funds. This funding allowed the OHSU Library to begin development of Oregon Health Go Local by covering the cost of project management, marketing, and data entry.

Work Model

Grant funding allowed the OHSU Library to hire a full-time Project Director who worked exclusively on Oregon Health Go Local for one year. She worked in tandem with the project's Principal Investigator to coordinate planning and implementation of the project. The OHSU Library also funded a half-time project assistant student position about midway through the one-year grant cycle. Go Local received strong early support from 211info and various public library systems across the state. These public libraries are important collaborators and promoters of the project, and several of their staff members offered to volunteer for the Go Local project. When planning Oregon Health Go Local the intent was to rely heavily on the contributions of volunteers working within their communities to locate and enter data into the Oregon Health Go Local database. This volunteer-centric approach was adopted for two reasons. First, when seeking to acquire project data, utilizing community volunteers who are likely to be familiar with some of the health services in their local counties seemed advantageous. Second, volunteer involvement would result in project buy-in and foster grassroots promotion of the project from within Oregon communities. Volunteer recruitment and management was coordinated by the Project Director.

PUBLIC RELATIONS AND MARKETING

Public relations and marketing was a major undertaking for Oregon Health Go Local. At the outset the team knew that Oregon Health Go Local's overall success would lie in communication, public relations, and how the project was marketed. Much of this effort began during the grant-writing process, when the Principal Investigator sought letters of support from libraries and other agencies. It continued when the Project Director was hired.

The first item of import for Oregon Health Go Local was to identify a title, tagline, and mission statement for the project. These three items are fundamental to have a foundation upon which to build a solid identity and brand. As literature has shown, a cohesive identity or brand assists in the public's recognition of projects and their authority. ^{4 5}

Adopting “Oregon Health Go Local” as the project's title indicated the locality, the health connection, and the “go local” portion of the project. Additionally, creating “talking points” about Oregon Health Go Local to be used when informing people at conferences, when meeting with stake holders, and when communicating with funders and the public at large, enabled the project to stay true to the Oregon Health Go Local brand. By using these tactics, Oregon Health Go Local attempted to provide cohesive information about the project and produce greater and quicker buy-in with stakeholders and the public.

The next step was to create the visual identity for Oregon Health Go Local. Working with a professional designer who donated hours of work saving approximately $2,500, Oregon Health Go Local obtained a logo for all printed marketing materials, displays, giveaways, and the Web site itself. Once Oregon Health Go Local had this design, other promotional items such as a table cloth, a display, brochures, and giveaways were purchased.

With an identity to promote, Oregon Health Go Local was ready to begin getting the word out to organizations, individuals, and the public at large: Oregon Health Go Local was underway. Attending the Oregon Library Association annual conference, the Oregon Rural Health Conference, and other events throughout the year enabled the Principal Investigator and Project Director to promote this brand and identity within the community.

During these events, the Principal Investigator and Project Director began collecting contact information of interested individuals and organizations. Because promotion of the project began almost a year before launch, it was important to be able to give stakeholders updates about Oregon Health Go Local's progress. Using an e-mail list, the Project Director disseminated bi-monthly newsletters to report progress and gently remind people about its existence. Each newsletter was branded with the Oregon Health Go Local logo and provided a consistent Oregon Health Go Local message. This contact list became critical when the Project Director started to organize travel and events to promote Oregon Health Go Local throughout the state. Curating stakeholder relationships with representatives from Commissions on Families and Children, state public health workers, librarians, clinicians, and other stakeholders before the project went live, helped to foster word of mouth promotion and maintain project momentum.

One of the key groups of stakeholders for Oregon Health Go Local was libraries and librarians. Librarians know their communities throughout the state and are trusted providers of quality information. Their ties within their local communities to use and promote Oregon Health Go Local could be the deal breaker.

The challenge of public relations and marketing is twofold. How does one communicate the value of the product to the public, and how does one approach all stakeholders? Despite the many groups Oregon Health Go Local reached, the Principal Investigator and Project Director are still grappling with how to reach a wider array of clinicians and the public. How does one reach people who don't attend community fairs? How does one ask busy medical professionals to use a new resource and remember it along with all of the other things that they have to know and remember on a regular basis?

ACCOMPLISHING IT ALL

Style Guide

Before any data was received or input into the Go Local system, the Project Director created style guidelines that would help inform the process of data entry. The goals for the style guide were to create a consistent form for data input, to enable remote volunteers to work independently, to provide resources and documentation, and to promote and enable consistency over the life of the project. Using style guides from other Go Local projects as examples and an assessment of the strengths and weaknesses of the data input system informed the creation of Oregon Health Go Local's style guidelines. Oregon Health Go Local's guidelines were specific enough to be a strong foundation to start indexing records in the Go Local input system.

The document, A Field-by-Field Guide to Style, defined each field in the data input system, provided authority files where appropriate, and included numerous examples. For example, the phone number field in the input system is one long field that allows 200 characters. It is not masked to create uniform punctuation (e.g., parenthesis, periods, or dashes between parts of the number). What's more, there is only one phone number field in the database, so multiple phone numbers, including toll free and crisis numbers, must be included in one field. Because Oregon Health Go Local is a project that relies on remote server access at the National Library of Medicine, there is no way to add additional phone number fields in the database. Instead, guidelines help to ensure data input consistency and integrity.

Phone

Best Practice: Include area code as well as the 7-digit number. For phone numbers that are toll free, crisis lines, etc., indicate “toll free” or “crisis line” before the number. For records containing more than one phone number, include the local number first and the toll free number second. Or, if two local numbers exist, provide the most general for information first, then subsequent numbers.

Style: Separate area code, prefix, and number with hyphens. Do not use parentheses for the area code.

• Authority: n/a

• Example: 503-292-8888

• Example: Toll free: 800-262-9874

• Example: 503-292-8888 Toll free: 800-623-9887

• Example: Crisis line: 800-222-9999 Business line: 541-566-9874

Creating a style guide presented some challenges and opportunities for decision making. By creating guidelines before acquiring any data, some guidelines did not complement the nuances of the input system. Additionally, many style inconsistencies and mistakes were not discovered until the National Library of Medicine reviewed the database and content a few weeks prior to Oregon Health Go Local's scheduled site launch. A good example is the formatting in the site name field: how does one handle abbreviations for words like Saint and Mount? Should the abbreviation use a period or not? Reviewers at the National Library of Medicine brought this inconsistency to Oregon Health Go Local's attention. Since the style guidelines did not address this issue and many proverbial cooks had been stirring the broth, it was inconsistent. To address this issue, the Project Director modified the style guidelines. However, previously approved records were not edited to comply with the new guideline. When the records are audited, stylistic inconsistencies such as these will be identified and rectified. Had the style guidelines accounted for such instances and had reviewers paid attention to such a guideline, the inconsistency issue of record formatting could have been avoided.

Volunteers

As mentioned previously, the model of Oregon Health Go Local has varied quite a bit from other Go Local projects. Because Oregon Health Go Local relied heavily on volunteers for data input and management, the Project Director needed to provide volunteer training. The model for training involved two training sessions. The first session acquainted volunteers with Oregon Health Go Local, style guidelines, and indexing; it also included some light practice with the input system. After the first session, trainers asked volunteers to use the system and return to the second training session with their questions. The second training session, usually held a week later, was intended to be a two-hour hands-on workshop for practice with the input system. Unfortunately, few people had time to work with the system, or they found it frustrating and did not practice during the week between training sessions. From observation the Principal Investigator and Project Director learned that if volunteers did not practice regularly after their training, they would become frustrated and disoriented with it and eventually abandon the project. Because of the steep learning curve to use the input system, along with other factors, the attrition rate for Oregon Health Go Local volunteers ran and continues to run above 50%.

Wiki

One of the attractive things about Oregon Health Go Local is that volunteers can volunteer on their own time, in the comfort of their home or local coffee shop. However, working with remote volunteers can pose some challenges. How will volunteers be able to bounce ideas off of one another? How will they have access to a volunteer coordinator, to support their needs and questions? One of the ways Oregon Health Go Local decided to provide remote volunteers with resources was to create an online repository of resources. Using a wiki to host this information (see Figure 1) makes perfect sense. Volunteers can edit pages if they needed, and they can interact with other volunteers in one online place. Moreover, a free online service WetPaint©, <http://www.wetpaint.com>, meets Oregon Health Go Local's needs. It has discussion board functionality, easy WYSIWYG (what you see is what you get), editing tools (because wiki formatting is notoriously difficult if you've never used it) and can be accessed from the Web. See Figure 2 to see how the wiki editing interface looks in WetPaint.

FIGURE 1.

Oregon Health Go Local Wiki Home Page

FIGURE 2.

Editing the Oregon Health Go Local Wiki

Rather than sending volunteers paper copies of documents or overloading their e-mail in-boxes with document attachments, the wiki proved to be a good solution. Documents and information on the wiki are iterative. Instead of re-publishing a PDF and e-mailing volunteers each time the Project Director updated the style guidelines, changes could simply be made in the wiki. If volunteers subscribed to wiki e-mail notifications, they would automatically be notified of the change. Otherwise, changes simply happened “behind the scenes” and the risk of using out-of-date documents drastically lowered. From the wiki, Oregon Health Go Local was also able to link to training videos, Web sites, and other resources that might have been of use to volunteers. See Figure 3 for an example. Generally using the volunteer wiki enabled Oregon Health Go Local to function as a volunteer-driven project and provide resources to volunteers at their point of need.

FIGURE 3.

Volunteer Training Videos on the Oregon Health Go Local Wiki

When setting up the WetPaint wiki, the Principal Investigator and the Project Director thought it would be a good spot to try and create some community around Oregon Health Go Local. In addition to the discussion board, as shown in Figure 4, the site enables users to create member profiles, to have friends, and to send virtual compliments to community members.

FIGURE 4.

The Oregon Health Go Local Wiki Discussion Board

These features, while not central to informing volunteers, provided an opportunity for community to grow. However, in the long run the wiki was used more as a resource portal than as an online community. You will notice that Figure 4 shows the last discussion thread started in December 2008. For context, Oregon Health Go Local did not go live until May 2009. Ultimately volunteers did not engage in the online community, nor was there much interest in it.

Data Curation

Data curation was another significant project activity. At the outset the Project Director and Principal Investigator spent quite a bit of time meeting with individuals who represented governmental and non-profit organizations throughout the state, with the aim of identifying datasets that could be donated to Oregon Health Go Local. Soliciting letters of support for the grant from 211info, Oregon's local information and referral service, and some Oregon libraries, enabled the Principal Investigator to start identifying data sources even before the project received funding. The OHSU Library began to cultivate relationships with organizations such as 211info, the Portland, Oregon area's 211 project. There was even a representative from 211info on the hiring committee for the Project Director! This relationship and others provided Oregon Health Go Local the opportunity to obtain 211info data and use it as a foundation upon which to build.

The many relationships created resulted in donated data from numerous groups including: Cascade AIDS Project, Oregon AIDS Hotline, Oregon Hospice Association, Oregon Office of Rural Health, 211Lane (Lane County, Oregon's 211 project), and the Malheur County Commission on Families and Children. Additionally, the project obtained some very low cost data sets from medical licensing agencies including the Oregon Board of Naturopathic Examiners, and a free download of the licensed database provided by the Oregon Board of Chiropractic Examiners. These particular sets of data allowed Oregon Health Go Local to include clinics not typically included in Go Local projects. Oregon's statewide policies and regulation for this kind of medical care is one of the reasons this data was attainable. Oregon is only one of 15 states in the country that licenses naturopathic physicians as primary care providers.⁶

Although the Project Director encountered some challenges when the data received was not in a format that could be used or was out of date, it was still a starting place. Despite the challenge of having to potentially “clean up” imported datasets, the ability to incorporate data en masse was preferable to gathering small bits of information here and there.

Taxonomy

The Go Local taxonomy is the power behind providing users context-sensitive information. It is this taxonomy that enables Go Local databases to “talk” to MedlinePlus and provide valuable linking between health literacy and education about where services can be obtained.

Each Go Local project utilizes taxonomy differently. Some projects use all taxonomy terms, whereas others opt to use more generalized terms for certain services. Projects need to make decisions whether they will assign all appropriate terms, only major terms, or if they will assign terms that are more “user friendly” than others. After doing some research about how to best utilize a taxonomy for information and referral, the Project Director created indexing guidelines to assist volunteers in their work and to help ensure consistency in indexing. See the Appendix to view the Indexing Guidelines.

Even though these indexing guidelines existed, many issues arose that required the Principal Investigator and the Project Director to make decisions about how to best implement the Go Local taxonomy. One of the issues that arose was surrounding the taxonomy term “gynecologist.” Some health clinics, such as community and Planned Parenthood clinics, may not necessarily employ gynecologists. However, they do provide gynecologic care. The question remains: What terms will users use to find where they can receive gynecologic care? Will they look for “nurse practitioners?” The Principal investigator and the Project Director didn't think so. Oregon Health Go Local decided to use the taxonomy term “gynecologist” for these clinics, knowing that some of them may not offer services from a board certified gynecologist. This decision was made in order to improve the “findability” of a service using the Go Local taxonomy.

Another challenge is that some of the taxonomy terms can be confusing to users. The Health Literacy of America's Adults: Results from the 2003 National Assessment of Adult Literacy indicates 37% of those assessed by the study have basic or below basic health literacy.⁷ Unfortunately the Go Local taxonomy can, at times, be difficult for users with even intermediate skills to know to what a term refers. For example, what does “ambulatory care services” mean? A member of the general public might be looking for “urgent care” or “walk-in care” facilities and not know that “ambulatory care services” is the category under which she will find what she needs. Scope notes for the NLM taxonomy are not made available to the public. There is no definition, thesaurus, or embedded instructional content in the Go Local system; such tools would help assist the public in trying to discover what a particular term means.

Another example of a health literacy challenge is the use of the term “pregnancy termination services” over “abortion.” Any person looking for abortion might not look under “p” to find the service they want. In this instance, there is a cross-reference from “abortion” to “pregnancy termination services.”

Many terms in the taxonomy are user friendly. Taxonomists at NLM used the term “Ear/Nose/Throat Specialists” over “Otolaryngologist,” and combined common names for some terms. For example: “Physical Medicine/Rehabilitation Specialists.” Despite the challenges posed by taxonomy, the taxonomists at NLM are very receptive to taxonomy changes, but since the taxonomies are not tailored for each project, any changes need to reflect utility for the majority of Go Local projects’ needs.

The Principal Investigator and the Project Director also made distinct decisions surrounding some taxonomy terms. Oregon Health Go Local used “faith-based services” to indicate that a service offered by an organization either had a religious stance or requested that people receiving services adhere to certain religious rules.

Another example of a taxonomy challenge, which has been the subject of discussion among Go Local projects, is the ability to create a taxonomy crosswalk between the AIRS taxonomy, which is used by 211 databases, and the Go Local taxonomy. Many Go Local projects have acquired data from 211 projects but have struggled to easily incorporate the AIRS taxonomy in the Go Local system. In Oregon the decision was made to use the AIRS taxonomy that came with the 211info records simply as informational. By not attempting to import that taxonomy into the Go Local input system, the Project Director saved herself (and NLM) the headache of creating a crosswalk. However, each of these 211info records demanded more work time to apply the appropriate Go Local taxonomy terms.

EVALUATION

In Oregon Health Go Local's LSTA funding application, initial and final evaluation periods were written into the grant. Early on in the project a survey was posted on the state wide library listserv asking librarians and library staff to take an awareness survey. The survey measured awareness of MedlinePlus, Go Local projects, and Oregon Health Go Local. At the end of the grant period and after the launch of the site, the Project Director reposted the same survey to evaluate the change in librarians’ and library staff awareness about MedlinePlus, Go Local, and Oregon Health Go Local over the year. Comparing results from the first and second survey shows successful efforts for Oregon Health Go Local's marketing and public relations. A little more than fifty percent (50.7%) of respondents had not heard of MedlinePlus Go Local or Oregon Health Go Local in May 2008. In August 2009 only 20% of respondents had not heard of either resource. And almost fifty-four percent (53.8%) of respondents had heard of both MedlinePlus Go Local and Oregon Health Go Local, whereas in May 2008 only 29.7% claimed to have heard of these tools. Additionally, data from the follow-up survey indicates that an overwhelming percentage of respondents had heard of Oregon Health Go Local and MedlinePlus Go Local within one to six months prior to the survey.

Another piece of evaluation regarding Oregon Health Go Local is part of Oregon's LSTA process. It requires a review of each project be conducted by an external reviewer. Being funded by LSTA meant that Oregon Health Go Local had an external review, and a substantial report was submitted to the Oregon State Library by this reviewer.

Finally, Oregon Health Go Local continues to collect user feedback through NLM-provided avenues – the “contact us” and “suggest a resource” forms within the Oregon Health Go Local Web site and the monthly Web statistics reports generated by the Go Local system.

CONCLUSION

Oregon Health Go Local embodied a different model than many other existing Go Local projects. Spending a large amount of energy and time cultivating community relationships and marketing the project has proven successful in providing local information to consumers. While Oregon Health Go Local's work model revealed some unexpected challenges, Oregon Health Go Local chose to view them as learning and decision making opportunities. In the end, the hope is that this model of work and the challenges Oregon Health Go Local faced will help inform others who adopt similar consumer health related projects in their libraries, hospitals, and communities.