Internal Medicine residents use heuristics to estimate disease probability.

BACKGROUND: Training in Bayesian reasoning may have limited impact on accuracy of probability estimates. In this study, our goal was to explore whether residents previously exposed to Bayesian reasoning use heuristics rather than Bayesian reasoning to estimate disease probabilities. We predicted that if residents use heuristics then post-test probability estimates would be increased by non-discriminating clinical features or a high anchor for a target condition. METHOD: We randomized 55 Internal Medicine residents to different versions of four clinical vignettes and asked them to estimate probabilities of target conditions. We manipulated the clinical data for each vignette to be consistent with either 1) using a representative heuristic, by adding non-discriminating prototypical clinical features of the target condition, or 2) using anchoring with adjustment heuristic, by providing a high or low anchor for the target condition. RESULTS: When presented with additional non-discriminating data the odds of diagnosing the target condition were increased (odds ratio (OR) 2.83, 95% confidence interval [1.30, 6.15], p = 0.009). Similarly, the odds of diagnosing the target condition were increased when a high anchor preceded the vignette (OR 2.04, [1.09, 3.81], p = 0.025). CONCLUSIONS: Our findings suggest that despite previous exposure to the use of Bayesian reasoning, residents use heuristics, such as the representative heuristic and anchoring with adjustment, to estimate probabilities. Potential reasons for attribute substitution include the relative cognitive ease of heuristics vs. Bayesian reasoning or perhaps residents in their clinical practice use gist traces rather than precise probability estimates when diagnosing.

Estimation of post-test probabilities by residents: Bayesian reasoning versus heuristics?

Although the process of diagnosing invariably begins with a heuristic, we encourage our learners to support their diagnoses by analytical cognitive processes, such as Bayesian reasoning, in an attempt to mitigate the effects of heuristics on diagnosing. There are, however, limited data on the use ± impact of Bayesian reasoning on the accuracy of disease probability estimates. In this study our objective was to explore whether Internal Medicine residents use a Bayesian process to estimate disease probabilities by comparing their disease probability estimates to literature-derived Bayesian post-test probabilities. We gave 35 Internal Medicine residents four clinical vignettes in the form of a referral letter and asked them to estimate the post-test probability of the target condition in each case. We then compared these to literature-derived probabilities. For each vignette the estimated probability was significantly different from the literature-derived probability. For the two cases with low literature-derived probability our participants significantly overestimated the probability of these target conditions being the correct diagnosis, whereas for the two cases with high literature-derived probability the estimated probability was significantly lower than the calculated value. Our results suggest that residents generate inaccurate post-test probability estimates. Possible explanations for this include ineffective application of Bayesian reasoning, attribute substitution whereby a complex cognitive task is replaced by an easier one (e.g., a heuristic), or systematic rater bias, such as central tendency bias. Further studies are needed to identify the reasons for inaccuracy of disease probability estimates and to explore ways of improving accuracy.

Expression patterns and role of prostaglandin-endoperoxide synthases, prostaglandin E synthases, prostacyclin synthase, prostacyclin receptor, peroxisome proliferator-activated receptor delta and retinoid x receptor alpha in rat endometrium during artificially-induced decidualization.

To determine if changes in endometrial expression of the enzymes and receptors involved in prostaglandin (PG) synthesis and action might provide insights into the PGs involved in the initiation of decidualization, ovariectomized steroid-treated rats at the equivalent of day 5 of pseudopregnancy were given a deciduogenic stimulus and killed at various times up to 32 h thereafter. The expression of PG-endoperoxide synthases (PTGS1 and PTGS2), microsomal PGE synthases (PTGES and PTGES2), cytosolic PGE synthase (PTGES3), prostacyclin synthase (PTGIS), prostacyclin receptor, peroxisome proliferator-activated receptor delta (PPARD) and retinoid x receptor alpha (RXRA) in endometrium was assessed by semiquantitative RT-PCR, western blot analyses and immunohistochemistry. In addition, to determine which PG is involved in mediating decidualization, we compared the ability of PGE(2), stable analogues of PGI(2), L165041 (an agonist of PPARD), and docasahexanoic acid (an agonist of RXRA) to increase endometrial vascular permeability (EVP, an early event in decidualization), and decidualization when infused into the uterine horns of rats sensitized for the decidual cell reaction (DCR). EVP was assessed by uterine concentrations of Evans blue 10 h after initiation of infusions. DCR was assessed by the uterine mass 5 days after the initiation of the infusions. Because enzymes associated with the synthesis of PGE(2), including PTGS2, are up-regulated in response to a deciduogenic stimulus and because PGE(2) was more effective than the PGI(2) analogues and PPARD and RXRA agonists in increasing EVP and inducing decidualization, we suggest that PGE(2) is most likely the PG involved in the initiation of decidualization in the rat.

Prostaglandins and the initiation of blastocyst implantation and decidualization.

The process of blastocyst implantation in mammals is remarkably variable, especially in the extent of trophoblast invasion of the endometrium. In most species studied, the earliest macroscopically identifiable sign of blastocyst implantation is an increase in endometrial vascular permeability in areas adjacent to the blastocysts. This is followed in species with invasive implantation by decidualization, again localized to areas adjacent to the blastocysts. In some species, the application of a stimulus to the endometrium can result in increased endometrial vascular permeability and decidualization. Based initially on studies utilizing inhibitors of prostaglandin (PG) synthesis and more recently on studies using the techniques of transgenics, considerable evidence has accumulated indicating that PGs have an important role in the early events of implantation and artificially induced decidualization. However, which PGs are involved remains controversial. There may be differences between species, and different PGs may be involved at different times.