JMIR Publications

ABSTRACT

Background:

Internet searching has become an increasingly useful tool for seeking health information. However, it is largely textual in nature and written for high literacy populations. Therefore, low literacy Latinas are at a disadvantage when searching online for health-related answers as well as processing online resources. Internet searching presents a problem, as low literacy populations tend to make mistakes while typing, ending up with misleading search results lists. In addition, even if an appropriate result list is displayed, Low literacy Latinas may be challenged in their ability to process the information contained in any individual article from that list.

Objective:

The objective of this observational study is to explore and describe the search behaviors of Latina breast cancer survivors and their caregivers. We explore their behaviors with different modalities of querying (voice/text), as well as other factors that present difficulties when searching for, and processing health information online.

Methods:

We recruited 7 Latina breast cancer survivors, of which 3 brought their caregivers, totaling 10 women. We administered the Health LiTT health literacy test, a demographic survey, and a breast cancer knowledge assessment to each participant. Next, we trained the participants on searching online with either a keyboard or voice. Then, we asked participants to find information about (a) maintaining good spirits after breast cancer; (b) affordability of breast cancer treatment; (c) general information about breast cancer and treatment options; and (d) any lingering issues they had around survivorship. Participants were allowed to search in English and/or Spanish. We video and audio recorded the computer activity of each participant/patient-caregiver (dyad) and coded it for search behaviors.

Results:

Consistent with prior research, we found web articles to be written for a grade level of 11.33 in English and 7.15 in Spanish. We also found that, although most of our participants preferred searching using speech, they struggled with this modality. Speech based search brought up new problems such as incomplete search queries, as the computer considered pauses as an indication of the end of a query. At other times, the participants made side comments, or the computer was still reading results when a new search was started; those comments were then, taken to be part of the search. We also found that participants formulated a query that was very general with the hope that the results would yield useful information about a more specific topic they had in mind. Also, in several occasions participants took notes based on information from the snippets on the result lists, without clicking on any results to read more. Lastly, participants that spent more time reviewing articles were those that scored higher in the health literacy test.

Conclusions:

Despite the problems of searching using speech, we found a preference for this search modality, which may indicate a need to compensate for potential misspellings and grammatical errors that could appear in written queries. We also found the use of general questions in the hopes that detailed answers will provide information about more specific concerns. Understanding search behaviors and information evaluation strategies of low literacy Latina breast cancer survivors is fundamental for designing useful search interfaces for them to find relevant and reliable information online.