Quality and accuracy of patient-oriented web-based information on tooth avulsion

Background/Aim Tooth avulsion (TA)
is a severe form of traumatic dental injury. Treatment outcome depends on the
immediate measures taken. Usually, first responders are laypeople and internet
has become a favored platform for them to access health-related information. The
aim of this study was to evaluate the accuracy, quality, readability and
popularity of patient-oriented web information on emergency management of TA.

Methods Three search engines
(Google, Bing and Yahoo.com) were browsed with six VPNs using ‘knocked out
tooth’, ‘fallen out tooth’ and ‘tooth avulsion’ as keywords. Top 20 results for
each were evaluated. Webpages were included if they had information oriented to
caregivers. They were analyzed for accuracy (cut-off ≥ 95%). Quality of the web
pages was assessed using DISCERN and Journal of the American Medical
Association (JAMA) benchmark. Readability was evaluated using: Flesch-Kincaid
Grade Level (FKGL), Gunning Fog (FOG), Simple Measure of Gobbledygook (SMOG),
and Coleman-Liau Index (CLI). Popularity was assessed by Alexa Popularity Rank
(APR). Accurate and inaccurate webpages were compared using Chi-square analysis
and Mann-Whitney U test. Spearman's correlation between accuracy and popularity
was established. p<0.05 was considered significant.

Results Seventy-two webpages were
included. 23 (32%) were accurate (k = 0.94). Median DISCERN rating for accurate
and inaccurate webpages was 3.0 [2.5–3.0] and 3.0 [2.0–3.0] respectively (p= 0.331).
10(44%) of accurate and 12(24.5%) of the inaccurate webpages had high-quality
JAMA (p=0.089). Median
readability scores for accurate: FKGL (8[6.75-9.2]), FOG (10.6 [9.55-12.3]), CLI (7[6-7]), SMOG (6.7
[6.15-7.5]); and
inaccurate webpages: FKGL (7.85
[6.275-8.525]), FOG (10.55 [8.875-11.425]), CLI (7 [7-8.25]), SMOG (7
[6.2-7.925]). 68(94.5%) and 4(5.5%) webpages had average and difficult readability,
respectively with a significantly higher CLI score in inaccurate webpages
(p=.0035). Median APR was 1,02,538 [26852-1443755] for accurate and 7,63,190 [219799-2992067]
for inaccurate webpages (p=0.163), with no correlation between accuracy and
APR, Spearman rho = -0.241, p (2-tailed) = 0.0704.

Conclusion Relevant webpages were
limited and had low accuracy. This can impact the emergency
management by laypeople and adversely affect the outcome of TA.