School of Psychology FACULTY OF MEDICINE AND HEALTH Improving health information to promote health literacy Factors influencing the perception of side effect risk information – reflections and prospects for research Peter Gardner Senior Lecturer and Head of School School of Psychology [email protected]
23
Embed
Factors influencing the perception of side-effect risk information - reflections and prospects for research
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
School of PsychologyFACULTY OF MEDICINE AND HEALTH
Improving health information to promote health literacy
Factors influencing the perception of side effect risk information – reflections and
Knapp, Raynor, Berry (2004) Quality & Safety in Health Care.
Berry, Knapp, Raynor (2002) Lancet.
What about online Information?
Using a pop-up on a medicine page of Cancerhelp.org.uk
(now ‘About Cancer’ pages)
• 8 on-line studies since 2004• Publications in British Journal of Health Psychology, Drug Safety, Patient Education
and Counseling, Health Expectations.
• Tested alternative formats for presenting side effect risk on perceptions of risk
• Controlled design with random allocation
• Taxol, Ibuprofen, Tamoxifen
• Approximately 15 participants per month
• Participants more likely to have a personal interest
Cancer Research UK studiesBackground
• Broadly, the suite of studies have tested the effect of stating the risk of side effects as verbal descriptors, percentages and frequency statements and combinations of these formats.
• Started off with the premise that ‘Natural Frequencies’ are more concrete and would probably work better
• e.g. If 100 people took this medicine, 3 would get constipation.
Cancer Research UK studiesFindings
• The evidence suggests that verbal descriptors (“common”/ “rare”) on their own produce markedly less accurate estimations of risk
• Percentages generally perform well, but some evidence that these are not so good for low risks
• More specific statements about frequency show some superiority over frequency bands (“affects more than 1 in 10 patients”) but may not be so feasible
• People prefer combined statements (“affects 1 in 500 people (0.2%)”) but are no more accurate in their estimation of risk
Cancer Research UK studiesImpact
Cancer Research UK studiesLatest study
Testing EMA recommendations
• EMA requires a combination of words and frequency bands
Common: may affect up to 1 in 10 people
• NICE suggest that verbal descriptors should not be used without numerical information
• It could be argued that these recommendations are on the basis of consensus, not evidence
Testing EMA recommendations
• Two questions:• Does the use of the verbal descriptor ‘frame’ patients’
understanding leading to overestimation of risk?
• Does the use of two indicators of uncertainty lead to confusion?
‘may affect’ and ‘up to 1 in 10’
• Compare with
‘will affect’ and ‘up to 1 in 10’
Testing EMA recommendations
• 2 x 2 factorial design• Numerical format vs. Combined verbal and numerical
• ‘will affect’ vs. ‘may affect’
• Presented with information on 10 potential side effects; 2 each from the 5 EC frequency bands
• Asked to provide risk estimates for 5 of the side effects; 1 from each of the bandings
Scenario
Testing EMA recommendations
• 339 participants• Numerical + may = 91
• Combined + may = 85
• Numerical + will = 77
• Combined + will = 86
Participants
Testing EMA recommendations
• Asked to provide risk estimates for 5 of the side effects; 1 from each of the bandings and a rating of the perceived risk of experiencing ANY of the side effects
• Also completed 5 Likert scales regarding:• Satisfaction with the information
• Severity of the side effects
• Likelihood of experiencing a side effect
• General risk to health
• Effect on decision to continue treatment
ResultsNumerical vs. Combined
Results
• No differences between ‘may’ and ‘will’ conditions on perceptions of side effect risk
• Little difference in Likert ratings between the conditions• Higher perceived likelihood of experiencing side effects in the
combined condition
• No evidence of interactions between the variables
Results‘May’ vs. ‘Will’
Conclusions
• Possible framing effect of verbal descriptors
• Variability of response suggests lack of shared understanding of risk descriptions
• Problem of overestimation still prevalent in ‘numerical only’ condition, but…• Low incidence rates mean that people are more likely to
overestimate than underestimate
• Heterogeneity of risk perception responses may mean it is impossible to get shared understanding
• Precise risk information is difficult to obtain anyway
Future research
• Need for replication of findings with different medicines
• Are there other factors involved?• Numeracy (see Gardner et al, 2011)
• Other individual differences
• Emotional reaction to different types of treatment
• Need to investigate the effect on actual behaviour
References
Berry, D.C., Knapp, P. and Raynor, D.K. (2002). Provision of information about drug side effects to patients. Lancet, 359, 853-4.
Gardner, P.H., McMillan, B.R.W., Raynor, D.K., Woolf, E., Knapp, P. (2011). The effect of numeracy on the comprehension of information about medicines in users of a patient information website. Patient Education and Counseling, 83, 398-403.
Knapp, P., Gardner P.H., Carrigan N., Raynor, D.K. and Woolf, E. (2009). Perceived risk of medicine side effects in users of a patient information website: a study of the use of verbal descriptors, percentages and natural frequencies. British Journal of Health Psychology,14, 579-94.
Knapp, P., Gardner, P.H., Raynor, D.K., Woolf, E. and McMillan, B.R.W. (2010). Perceived risk of Tamoxifen side effects: a study of the use of absolute frequencies or frequency bands, with or without verbal descriptors. Patient Education and Counseling, 79(2), 267-271.
Knapp, P.R., Gardner, P.H. and Woolf, E. (2015). Combined verbal and numerical expressions increase perceived risk of medicine side-effects: a randomized controlled trial of EMA recommendations. Health Expectations, published online 26 Jan 2015, doi: 10.1111/hex.12344.
Knapp, P., Raynor, D.K. and Berry, D.C. (2004). Comparison of two methods of presenting risk information to patients about the side effects of medicines. Quality and Safety in Health Care,13, 176-80.
Knapp, P., Raynor, D.K., Woolf, E., Gardner, P.H., Carrigan, N. and McMillan, B.R.W. (2009). Communicating the risk of side effects to patients: an evaluation of UK regulatory recommendations. Drug Safety, 32(10), 837-849.