Unrealistic optimism about treatment risks for acute appendicitis

Abstract

Unrealistic optimism is a cognitive bias that causes people to think they are at lower risk of a negative outcome than they actually are. This study identified significant unrealistic optimism towards the risks of appendicitis treatment complications.

Introduction

Patients must understand treatment risks to make good healthcare decisions. However, patients’ risk perceptions can be affected by cognitive biases, ‘mental shortcuts’ that simplify how people process complex information that can lead to suboptimal decisions¹. A highly prevalent cognitive bias is unrealistic optimism—the belief that one is less likely than others to experience negative outcomes, and more likely than others to experience positive outcomes^2–4.

Although unrealistic optimism has been shown to influence risk perceptions and health decisions^5–7, it has not been studied for surgical disease. This is a knowledge gap given the complexity and risks associated with surgical treatments and non-surgical alternatives. To address this, the present study evaluated whether unrealistic optimism existed and affected risk perceptions about surgical and non-surgical treatments for appendicitis. It was hypothesized that unrealistic optimism would be present and associated with lower personal risk perceptions; that is, individuals with unrealistic optimism would rate appendicitis treatments as being lower risk, and the magnitude of unrealistic optimism would vary between surgical and non-surgical treatments, and between methods used to communicate risk.

Methods

Full methods are available in Appendix S1 and the full survey text in Appendix S2. The University of Washington institutional review board exempted this study (STUDY00012210), which followed the American Association for Public Opinion Research reporting guidelines where applicable (Table S1).

Survey design and administration

Adults (aged at least 18 years) in the USA were recruited from Amazon Mechanical Turk, which has been used to evaluate healthcare decisions in previous studies^8–10. Surveys had three sections. The first comprised vignettes describing surgery and antibiotic treatment of appendicitis, and the risk of treatment complications. One-third of respondents were presented with risk information using qualitative terms (such as common, uncommon), one-third using point estimates (for example, 3 per cent risk), and one-third using ranges (such as between 25 and 35 per cent). This was based on recognized variation in interpretation based on risk descriptors, and done to reflect variation in how risk is communicated in clinical practice^11,12.

The second section assessed unrealistic optimism using a direct comparative measure^3,4,6,13. After each vignette, participants rated their chance of developing each complication compared with ‘an average person’ with appendicitis, on a seven-point scale ranging from ‘much less than average’ to ‘much more than average’.

The third section captured participants’ personal risk perceptions (chance of experiencing any postoperative or antibiotic-related complication if they developed appendicitis in the future; 7-point scale, ‘no chance’ to ‘extremely high chance’); perceptions about their own health relative to that of an average person with appendicitis (7-point scale; ‘much less healthy’ to ‘much more healthy’); and demographic information.

Statistical analysis

For each complication, unrealistic optimism was measured on a seven-point scale from −3 to +3, with positive values denoting unrealistic optimism. The mean response was calculated and compared with the expected value of 0 using a one-sample t test^4,6. Multivariable linear regression models with robust standard errors were used to evaluate whether unrealistic optimism was associated with risk perceptions. Unrealistic optimism for surgical versus non-surgical treatments was compared using a paired-sample t test, and risk communication methods by three-level one-way ANOVA.

Results

The sample consisted of 601 individuals (Fig. S1), with a mean(s.d.) age of 38.1(12.7) years and 248 (41.3 per cent) identifying as male (Table S2). Few patients (28 of 601, 4.7 per cent ) reported previous appendicitis (most treated by appendicectomy). Nearly half of participants (288 of 601, 48 per cent) felt they were healthier than ‘a typical person with appendicitis’.

Unrealistic optimism was observed for all complications across most risk-description conditions (Table S3 and Fig. 1). The magnitude of unrealistic optimism was greatest for deep space infections after surgery and deep space infections after antibiotic treatment, and lowest for antibiotic-associated diarrhoea and treatment failure on antibiotics.

Fig. 1.

Unrealistic optimism by complication and risk description

Values are mean levels of unrealistic optimism, on a seven-point Likert scale with 0 reflecting no optimism, for each surgical and antibiotic complication queried and different methods of risk description. More positive values reflect higher levels of unrealistic optimism. Error bars represent standard errors. *P < 0.001, †P = 0.002 (ANOVA).

Unrealistic optimism was greater for surgical, compared to antibiotic-associated complications: mean Likert scale rating (M) 0.94 (95 per cent c.i. 0.84 to 1.04) versus 0.58 (0.50 to 0.66); P < 0.001). For nearly all complications, unrealistic optimism was lower when risks were communicated using qualitative descriptors (Fig. 1). For instance, there was significantly more unrealistic optimism towards the risk of deep space infection after surgery when risk was described with a point estimate of 3 per cent (M 1.30, 1.12 to 1.48) or range of 1–5 per cent (M 1.20, 1.01 to 1.39) compared with the qualitative descriptor ‘uncommon’ (M 0.89, 0.72 to 1.06) (P = 0.002).

In regression analysis, unrealistic optimism was associated with lower personal risk perceptions for both surgery (P < 0.001) and antibiotics (P < 0.001) (Tables S4 and S5). Fig. 2 shows the relationship between unrealistic optimism and perceived risk with an overlaid adjusted regression line.

Fig. 2.

Unrealistic optimism and perceived risk of complications

Relationship between unrealistic optimism and participants’ perceived risk of a complication if they were to undergo a surgery or b antibiotic treatment for appendicitis, with an overlying adjusted linear regression. A jitter plot is used to allow visualization of data density.

Discussion

In this study, unrealistic optimism was observed about appendicitis treatment complications, the degree of which varied depending on treatment type, complication type, and method of risk communication. Unrealistic optimism was also associated with personal risk perceptions. The results underscore the need for surgeons to monitor for potential overoptimism when discussing treatment risks with patients. In this study, individuals displaying unrealistic optimism also perceived an absolute lower risk of appendicitis treatment complications for themselves—evidence of a potentially problematic connection between the notion that ‘I am lower risk than the average’ and the more general notion that ‘I am low risk’¹⁴. The consequences of such overoptimism may include less appreciation of one’s lack of knowledge about risks¹³, and more severe negative emotional consequences from falling short of expectations^4,15. Such dynamics have been observed in studies evaluating cancer treatment⁶, and could be highly detrimental if patients similarly misinterpreted the risks and made uninformed decisions about treatments for appendicitis and other surgical conditions.

Surgeons can use several strategies to combat unrealistic optimism. One is to help patients understand how self-assessment of risk factors compares with population baselines¹⁶. Another is to guide patients toward closer comparisons by choosing more relatable and identifiable comparators¹⁷. However, unrealistic optimism can be difficult to eliminate, and few interventions have been shown to consistently curb overoptimism across settings^7,13,18.

The present findings have also highlighted how surgeons could benefit from anticipating several sources of variation in overoptimism. Individuals exhibited higher levels of unrealistic optimism about surgical compared with antibiotic-related complications, and were more overly-optimistic about low-probability but more serious complications, such as deep space infections. This is consistent with existing evidence from non-surgical settings that unrealistic optimism tends to be greater for low-probability, high-consequence, and less familiar outcomes^3,4,19.

The magnitude of unrealistic optimism in the present study also differed based on how risk information was communicated. The present study may be the first to describe this relationship, and the fact that overoptimism was lower when risk was communicated qualitatively is not necessarily reassuring. Although ubiquitous in clinical medicine, qualitative descriptions of probabilities are not recommended in a number of other fields because they are context-dependent and can result in a wide range of interpretations^11,12. These inaccuracies can impede informed decision-making¹¹.

This study has limitations. It used a sample that differed from the US population with respect to several characteristics²⁰. The use of standardized vignettes and risk estimates does not capture aspects of real-world informed consent discussions. Finally, self-reported survey information may differ from the perspectives of patients facing an acute illness. Nonetheless, this study provides novel and, to the authors’ knowledge, the first dedicated evaluation of unrealistic optimism in decision-making for surgical disease.

Funding

J.E.R. is supported by a training grant from the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health (T32DK070555). This work was funded in part by a generous gift from Marty and Linda Ellison. The funding sources had no role in study design, implementation, analysis, or reporting.

Supplementary material

Supplementary material is available at BJS online.