A Study of Gruesome and Benign Pictorial Health Warning Labels Among Georgian Adults: For Whom Are They Effective? Open Access

Lingwall, Cailyn (Spring 2018)

Permanent URL: https://etd.library.emory.edu/concern/etds/2f75r804x?locale=en


BACKGROUND: Research has documented the effectiveness of pictorial health warning labels (HWLs) for reducing population level cigarette smoking, and may be particularly effective in low-and middle-income countries (LMICs). Additional research is needed on LMICs disproportionately impacted by tobacco-related diseases and death to examine who may find different messages to aid in smoking prevention or cessation. Thus, we 1.) compared perceived effectiveness of pictorial vs. text-only HWLs; 2.) examined themes emerging from pictorial HWLs; and 3.) examined correlates of perceived effectiveness of different pictorial HWL themes among adults in the Republic of Georgia.

METHODS: We analyzed a cross-sectional national household survey of Georgian adults conducted in 2014 (n=1,163). Participants were randomized to evaluate the perceived effectiveness (on a 9-point scale) of either Set A or Set B HWLs, with each set containing half of the HWLs presented pictorially and half with text-only. Bivariate analyses compared the perceived effectiveness of pictorial vs. text-only HWLs. We then conducted factor analyses to determine themes of pictorial HWLs, identifying gruesome and benign themes. Finally, regressions were conducted to identify sociodemographic and tobacco use related correlates of perceived effectiveness of the pictorial HWL themes.

RESULTS: Factor analyses identified one factor among Set A HWLs, labeled “benign”; Set B yielded two factors – “benign” and “gruesome.” All gruesome HWLs were perceived as more effective than text only; however, 2 of the 9 benign HWLs were perceived as no more effective than text only. Among Set A HWLs, correlates of greater perceived effectiveness for all participants included: being female (p<.001), living in a rural setting (p=.001), not having children in the home (p=.038), and being a nonsmoker (p=.005); for non-smokers, being female (p=.016), and living in a rural setting (p=.017); and for current smokers, being female (p=.014), not married/living with a partner (p=.045), having more close friends who smoke (p=.013), and rating quitting smoking as more important (p=.009). Among Set B HWLs, 43.8% rated gruesome HWLs more effective, 43.4% rated gruesome and benign equally effective, and 12.9% rated benign more effective. Correlates of benign HWL effectiveness included having fewer friends who smoked (p=.019), and a higher household income (p=.031); and for smokers, fewer friends who smoked (p=.013) and perceiving quitting smoking to be important (p=.006). Finally, a lower household income (p=.003) predicted gruesome HWLs effectiveness.  

CONCLUSIONS: While pictorial HWLs are largely perceived as more effective, gruesome and only some benign HWLs outperform text-only. Some benign HWLs are perceived as equally effective, and a minority found them more effective than gruesome HWLs. Some benign HWLs may be more effective for smokers rating quitting as important. Social factors are critical in understanding the effectiveness and impact of pictorial HWLs.  

Table of Contents



Global Tobacco Use and Regulation.. 4

FCTC Article 11: Health Warning Label Policy.. 5

Health Warning Label Messaging Strategies.. 6

Reactions to Pictorial Health Warning Labels by Sociodemographic Factors.. 7

Tobacco Use and HWLs in the Republic of Georgia.. 9

Theoretical Framework.. 11

Gaps in the Literature and Significance.. 12

Research Aim.. 13


Study Protocol and Participants.. 14

Measures.. 15

Data Analyses.. 16


Study Participants.. 19

Aim 1. Perceived Effectiveness of Pictorial versus Text-only HWLs.. 19

Aim 2: Factor Analysis Examining Themes Emerging from Pictorial HWLs.. 19

Aim 3: Correlates of Perceived Effectiveness of Benign HWLs (Set A).. 20

Aim 3: Correlates of Perceived Effectiveness of Gruesome Versus Benign HWLs (Set B).. 20


Perceived Effectiveness of Pictorial versus Text-only HWLs.. 22

Correlates of Perceived Effectiveness of Benign HWLs (Set A) .. 23

Correlates of Perceived Effectiveness of Gruesome Versus Benign HWLs (Set B).. 25

Study Significance.. 27

Future Implications for Research and Practice.. 27

Limitations.. 28

Conclusions.. 28




Table 1.. 30

Table 2.. 31

Table 3.. 32

Table 4.. 33


About this Master's Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
  • English
Research Field
Committee Chair / Thesis Advisor
Committee Members
Partnering Agencies
Last modified

Primary PDF

Supplemental Files