Positive attitudes and failed queries: an exploration of the conundrums of consumer health information retrieval

https://doi.org/10.1016/j.ijmedinf.2003.12.015Get rights and content

Abstract

Several studies have found that consumers report a high level of satisfaction with the Internet as a health information resource. Belied by this positive attitude, however, are other studies reporting that consumers were often unsuccessful in searching for health information.

In this paper, we present an interview and observation study in which we asked health consumers to search for health information on the Internet after first stating their search goals. Upon the conclusion of the session they were asked to evaluate their searches.

We found that many consumers were unable to find satisfactory information when performing a specific query, while in general the group viewed health information retrieval (HIR) on the Internet in a positive light.

We analyzed the observed search sessions to determine what factors accounted for the failure of specific searches and positive attitudes, and also discussed potential informatics solutions.

Introduction

Consumers and patients are confronted with a plethora of information about health and health care, especially through the proliferation of Web resources [1]. They often have difficulty, however, identifying and accessing answers to their idiosyncratic health questions using standard information retrieval techniques [2], [3], [4]. This paper focuses on health information retrieval (HIR) via the Internet that is initiated by consumers to meet their specific information needs—as opposed to educational information prescribed and delivered to consumers over the Internet by health care providers.

Several survey studies on consumer HIR patterns [5], [6], [7] confirm that HIR on the Internet is widespread and that consumers rely on it and benefit from it in many ways. Particularly worth noting is the series of interviews and surveys conducted by the Pew Internet & American Life Project during the past few years [5], [8]. The Pew studies reveal that millions of consumers seek health information for themselves and others and that the retrieved information is helpful and greatly influences their decision making. This picture of satisfied users has been confirmed by other surveys, for example, the visitor profile survey of MEDLINEplus (http://www.nlm.nih.gov/medlineplus/survey2003/) conducted in February 2003.

At the Decision Systems Group we have conducted a few pilot studies on consumer HIR through analyzing log files of searching activity and through observing and interviewing subjects. Our studies reveal that consumers encounter a number of difficulties in HIR [9], [10], [11]. In contrast to the successful and empowered consumers reported in the Pew and other surveys, the consumers in our observational and log analysis studies were not very successful at the HIR tasks they performed. This raised some questions: was the contrast caused by the methodology difference (self-reporting versus observation/log analysis)? Is the population of the Pew studies different from the population of our studies? Do consumers hold generally positive views toward HIR in spite of having low success rates at the query level?

In order to answer these questions, and also further develop informatics strategies to support consumer HIR, we conducted a formal study of 97 health care consumers using one-on-one interviews. During the study consumers were asked to explain their information needs, perform searches on a large consumer health information website (MEDLINEplus) and other sites of their choice, and comment on their search results. This report describes the study design, data analysis, and subjects, and also discusses the informatics challenges and opportunities that were identified.

The study confirmed that HIR provides an important source of health information for consumers and that it has become an integral part of how consumers take care of themselves and others. The study also showed that although the consumers’ attitudes in our study were generally positive about HIR, the specific searches of the consumers were often neither straightforward nor successful. To address this problem, further informatics research is needed in the areas of helping consumers construct effective queries, identifying appropriate information sources and search spaces, and filtering out repetitive or outdated information.

Section snippets

Main research questions

The questions that concerned us most in this study were as follows:

  • 1.

    Do consumers who report positive attitudes toward HIR, nevertheless, have low success rates when performing specific HIR tasks?

  • 2.

    What are the major factors that contribute to failure when consumers perform a specific HIR task?

Patient information retrieval

In the past, most health informatics research and development has emphasized clinicians’ needs, primarily focusing on hospital and office-oriented activities. Recently, the focus of health information systems has shifted to include support for the needs of patients and consumers. For example, Brigham and Women’s Hospital developed a website in the mid-1990s to better serve the general public (“consumers”) and its own patient community by providing information about and access to services [12],

Methods

We recruited and interviewed 97 consumers to learn more about how they search for health information and services, what their goals are, and if they are satisfied with the results they get.

Demographics

Of the 97 consumers who participated in the test, 45 were male and 52 female, with a mean age of 49 (51.91 years for males, 47.07 for females). Most (83%1) of the subjects had an education level of 2 years of college or above. Most (84%) of the subjects were white, not of Hispanic origin. Although Brigham and Women’s Hospital is a large urban hospital

Significance

The study we have conducted is an interview and observational study. Compared to survey studies and self-report questionnaires, our study provided a more intimate and in-depth look into the problems consumers encounter during HIR as well as the causes. This knowledge is of particular importance and interest for health informatics research because consumer HIR is an area where information technology has made a huge impact on how patients take care of themselves and can be expected to have an

Acknowledgements

This work was funded by grant R01 LM07222 from the National Library of Medicine. We thank Ms. Audrey Kalmus for her work in data collection.

References (19)

  • R.J Cline et al.

    Consumer health information seeking on the Internet: the state of the art

    Health Educ. Res.

    (2001)
  • R Milne et al.

    Consumer health information needs to be rigorous, complete, and relevant

    BMJ

    (2000)
  • R Kiley

    Consumer health information on the Internet

    J. R. Soc. Med.

    (1998)
  • Proceed with Caution: A Report on Quality of Health Information on the Internet, in: RAND Health and California Health...
  • S. Fox, L. Rainie, J. Horrigan, A. Lenhart, T. Spooner, M. Burke, et al., The online health care revolution: how the...
  • HarrisInteractive, eHealth Traffic Critically Dependent on Search Engines and Portals, 23 April 2001,...
  • P Williams et al.

    Surfing for health: user evaluation of a health information website. Part two: fieldwork

    Health Info. Libr. J.

    (2002)
  • S. Fox, D. Fallows, Health searches and email have become more commonplace, but there is room for improvement in...
  • Q Zeng et al.

    Characteristics of consumer terminology for health information retrieval

    Methods Info. Med.

    (2002)
There are more references available in the full text version of this article.

Cited by (0)

View full text