Changes in antibiotic prescription following an education strategy for acute respiratory infections

[Design of a predictive score table for peritonsillar infection based on signs and symptoms]

BACKGROUND

Peritonsillar infection (PTI) is a reason for urgent consultation due to intense throat discomfort. A delayed or inaccurate diagnosis can jeopardize the upper aerodigestive tract and be fatal in its evolution. Our objective was to develop a predictive model for the presence of IPA helping in its rapid detection.

PATIENTS AND METHODS

A 66-month retrospective observational study from 2017 was carried out in a county and tertiary referral hospitals, registering data from all patients diagnosed with PTI and a proportional volume of subjects with pharyngeal symptoms without PTI. Collection of clinical, exploratory and demographic data among participants. Their higher relative risk of PTI presence allowed them to be considered as variables to be tested. Development of a scoring scale for the probability of suffering from it and logistic regression analysis, obtaining the ROC curve with the best diagnostic correlation. Internal validation and estimation of the predictive values of the model.

RESULTS

On 348 cases of PTI, the assessment scale scored the presence of six variables: trismus (3), unilateral dysphagia-odynophagia (2), velar bulging (2), reflex otalgia (1), pharyngolalia (1) and age between 16 and 46 years (1). With a range of 0-10, a cut-off ≥6 offered a sensitivity of 96.1%, a specificity of 93.9%, and an efficiency of 94.9%. The area under the ROC curve was 0.979.

CONCLUSIONS

The internal validation of this model based on signs and symptoms makes it a very useful tool for early detection of PTI in otorhinolaryngology and primary care.

Effects of a feedback intervention on antibiotic prescription control in primary care institutions based on a Health Information System: a cluster randomized cross-over controlled trial

OBJECTIVES

Overuse and misuse of antibiotics are major factors in the development of antibiotic resistance in primary care institutions of rural China. In this study, the effectiveness of a Health Information System-based, automatic, and confidential antibiotic feedback intervention was evaluated.

METHODS

A randomized, cross-over, cluster-controlled trial was conducted in primary care institutions. All institutions were randomly divided into two groups and given either a three-month intervention followed by a three-month period without any intervention or vice versa. The intervention consisted of three feedback measures: a real-time pop-up warning message of inappropriate antibiotic prescriptions on the prescribing physician's computer screen, a 10-day antibiotic prescription summary, and distribution of educational manuals. The primary outcome was the 10-day inappropriate antibiotic prescription rate.

RESULTS

There were no significant differences in inappropriate antibiotic prescription rates (69.1% vs. 72.0%) between two groups at baseline (P = 0.072). After three months (cross-over point), inappropriate antibiotic prescription rates decreased significantly faster in group A (12.3%, P < 0.001) compared to group B (4.4%, P < 0.001). At the end point, the inappropriate antibiotic prescription rates decreased in group B (15.1%, P < 0.001) while the rates increased in group A (7.2%, P < 0.001). The characteristics of physicians did not significantly affect the rate of antibiotic or inappropriate antibiotic prescription rates.

CONCLUSION

A Health Information System-based, real-time pop-up warnings, a 10-day prescription summary, and the distribution of educational manuals, can effectively reduce the rates of antibiotic and inappropriate antibiotic prescriptions.