Beta-lactams (BLs), especially penicillins, are the most commonly used antibiotics, particularly in primary care, and one of the most reported drug allergies. Fearing cross-reactivity, clinicians refrain from prescribing another BL (e.g., cephalosporin or carbapenem) to penicillin-allergic patients. This can have significant consequences for the patients and the health-care system (e.g., exposure to broad-spectrum antibiotics, increased risk adverse effects, and increased healthcare costs).

To assess the absolute cross-reactivity risk, two systematic reviews with meta-analysis were conducted. Then, an approach based on a knowledge mobilization framework considering scientific, contextual and experiential evidences was used. Focus groups with stakeholders, including primary care clinicians, pediatricians, infectious disease specialists and allergists/immunologists, were also held to meet the needs of all actors concerned.

Following this work, it appears that true allergies to penicillin are very rare. Indeed, in patients with a history of penicillin allergy, very few are truly allergic and thus the risk of cross-reaction with another BL is even lower, varying according to structural and physicochemical similarities with alleged-penicillin. Moreover, the risk of having an anaphylactic reaction after penicillin exposure is very low, especially among children. As well, in patients with confirmed penicillin allergy, the observed reactions are usually delayed non-severe skin reactions. However, with a confirmed penicillin allergy, it is important to remain cautious when administering a new BL, especially if the initial reaction was serious or severe. Based on these key messages, a decision aid including an algorithm was developed. Likewise, individualized algorithms for common infections met in primary care were produced.

From this work, health professionals non-specialized in allergology should be able to better manage the risks attributed to penicillin allergies. Therefore, patients should receive the most effective and safe antibiotics to treat their clinical conditions in primary care.

Antibiotic prophylaxis with a single dose of doxycycline after a tick bite is one of the tools for preventing Lyme disease, which is becoming increasingly prevalent in Quebec. The aim of this work was to revisit this practice in adults and children younger than 8 years of age.

To assess the safety and absolute risk reduction (ARR) of doxycycline for preventing Lyme disease in contraindicated populations, two systematic reviews were conducted with a re-analysis of the original efficacy data. A knowledge mobilization framework was used to consider the scientific, contextual, and experiential evidence, taking into account information on patients’ and clinicians’ experiences.

A single dose of doxycycline prescribed within 72 hours of being bitten by a tick (Ixodes scapularis) could prevent cutaneous manifestation of Lyme disease (ARR -2.8%, 95% confidence interval: -11.7–6.1; p = 0.06), without serious side effects, provided that the bite occurred in a geographical region where at least 25 percent of nymph and 50 percent of adult ticks are infected with the disease. However, the level of evidence was low and its generalizability to other contexts was doubtful. The decision to prescribe antibiotic prophylaxis may be based more on the fear of Lyme disease, rather than on effectiveness data and the real risk of contracting Lyme disease.

It may be challenging for clinicians to discuss Lyme disease prophylaxis with patients and their families in contexts where people are fearful of the disease, and the risk of contracting it from a tick bite is uncertain. Decision aids that provide scientific evidence on the real risk of developing Lyme disease after a tick bite, particularly in Quebec, can promote informed decisions based on patient preferences and values by supporting discussion between clinicians and patients.

Opioids are being used increasingly to treat chronic noncancer pain despite the uncertainty regarding its long-term benefits. This study served to determine if problems are associated with opioid use in Québec for new users from 2006 to 2013 without history of cancer.

A retrospective longitudinal cohort study was conducted using administrative databases stored at the Régie de l'assurance maladie du Québec (RAMQ) to describe the annual proportion of new users to whom at least one of the five indicators of potentially inappropriate opioid use applied was estimated. These indicators are (i) overlapping opioid prescriptions, (ii) overlapping opioid and benzodiazepine prescriptions, (iii) the use of long-acting opioids at the start of treatment, (iv) a high mean daily dose, and (v) a rapid increase in the opioid dose.

The annual proportion of new users to whom at least one of the five indicators of potentially inappropriate opioid use applied decreased from 15.4 percent in 2006 to 12.3 percent in 2013. It was mainly the following three indicators that contributed the most to these proportions in 2013: (i) overlapping opioid prescriptions (5.8 percent), (ii) overlapping opioid and benzodiazepine prescriptions (8.2 percent), and (iii) the use of long-acting opioids at the start of treatment (1.8 percent).

The vast majority of new users with no history of diagnosed cancer used opioids adequately according to the five indicators of potentially inappropriate opioid use applied. Improvement could still be made to decrease mainly overlapping opioid prescriptions and overlapping opioid and benzodiazepine prescriptions.