JMIR Publications

ABSTRACT

Background:

Over 20,000 parents in the United States face the extraordinary challenge of participating in decisions about whether to use life support for their infant(s) born on the cusp of viability. Clinicians must help these families grasp complex medical information about their baby’s immediate prognosis as well as the risk for significant long-term morbidity. Patients faced with this decision desire supplemental information and frequently seek medical information on the internet. Empiric evidence about the quality of websites, however, is lacking.

Objective:

We sought to evaluate the quality of online information available about periviable birth and treatment options for infants born at the cusp of viability.

Methods:

We read a counseling script that included information typically provided by obstetrical and neonatal providers when periviable birth is imminent to twenty pregnant women. The women were then asked to list terms they would use to search the internet if they wanted additional information. Using these search terms, two reviewers evaluated the content of websites obtained via a GoogleTM search. We used two metrics to assess the quality of websites. The first was the DISCERN instrument, a validated questionnaire designed to assess the quality of patient-targeted health information for treatment choices. The second metric was the Essential Content Tool (ECT), a tool designed to address all the key components of counseling around periviable birth as outlined by professional organizations. Scores ≥6 using ECT and ≥4 using DISCERN were considered "high quality”. Inter-reviewer agreement was assessed by calculated kappa statistic.

Results:

A total of 97 websites were reviewed. Over half (58%) were for-profit sites, new stories or personal blogs, 28% were government or medical sites, and 13% were non-profit or advocacy sites. The majority of sites scored poorly in DISCERN questions designed to assess the reliability of information presented. For example only 27% (n=26) presented a range of treatment options for infants born during periviable period and over half (n=62) did not cite the sources that informed their website. Only 10% of websites were “high quality” as defined by the DISCERN tool. The majority of sites did not address the essential content defined by the ECT. Importantly, only 18% of websites (n=17) indicated that there are often a number of reasonable approaches to newborn care when faced with periviable birth. Agreement was strong with ᴋ ranging from 0.72-0.91.

Conclusions:

Most information about periviable birth found on the internet using common search strategies is of low quality. News stories highlighting positive outcomes are disproportionately represented. Few websites discuss comfort-care or how treatment decisions impact quality of life.