Approximate Bayesian inference for random effects meta-analysis.

K Abrams; B Sansó

doi:10.1002/(sici)1097-0258(19980130)17:2<201::aid-sim736>3.0.co;2-9

Home/Archive/PubMed

PubMed · 9483729

Approximate Bayesian inference for random effects meta-analysis.

K Abrams·B Sansó

View original source ↗Export metadata ↓

Abstract

Whilst meta-analysis is becoming a more commonplace statistical technique, Bayesian inference in meta-analysis requires complex computational techniques to be routinely applied. We consider simple approximations for the first and second moments of the parameters of a Bayesian random effects model for meta-analysis. These computationally inexpensive methods are based on simple analytical formulae that provide an efficient tool for a qualitative analysis and a quick numerical estimation of posterior quantities. They are shown to lead to sensible approximations in two examples of meta-analyses and to be in broad agreement with the more computationally intensive Gibbs sampling.

Explore related subjects

Antibiotic Prophylaxis Bayes Theorem Dental Caries Dentifrices Digestive System Humans Logistic Models Meta-Analysis as Topic Models, Statistical Odds Ratio Journal Article Research Support, Non-U.S. Gov't

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

K Abrams, B Sansó. 1998-01-30. Approximate Bayesian inference for random effects meta-analysis.. https://doi.org/10.1002/(sici)1097-0258(19980130)17%3A2%3C201%3A%3Aaid-sim736%3E3.0.co%3B2-9

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2001

Routine prophylactic antibiotic use in the management of snakebite.

BACKGROUND: Routine antibiotic prophylaxis following snakebite is not recommended but evidence suggests that it may be common practice in Zimbabwe. This study set out to determine and describe the extent of this practice at Parirenyatwa Hospital, a large teaching hospital in Zimbabwe METHODS: A retrospective case review (1996 to 1999 inclusive) of all cases of snakebite was undertaken at Parirenyatwa Hospital. Cases with a diagnosis of snakebite, presenting within 24 hours of the bite and with no complications or concurrent illness were defined as "routine prophylactic antibiotic use". RESULTS: From 78 cases which satisfied the inclusion criteria, 69 (88.5%) received antibiotics. Ten different antibiotics from 6 different classes were used with penicillins the most commonly prescribed (benzylpenicillin in 29% of cases, alone or in combination). Over 40% of antibiotics were given parenterally although all patients were conscious on admission. The total cost of antibiotics used was estimated at US$522.98. CONCLUSION: Routine prophylactic use of antibiotics in snakebite at Parirenyatwa Hospital is common practice. This may highlight the lack of a clearly defined policy leading to wasteful inappropriate antibiotic use which is costly and may promote bacterial antibiotic resistance. Further work is required to investigate the reasons for this practice and to design appropriate interventions to counter it.

Antibiotic Prophylaxis↗

PubMed2001

Long-term infection rates associated with the pectoral versus abdominal approach to cardioverter- defibrillator implants.

Infection is an uncommon (0% to 6.7%) but serious complication after implantable cardioverter-defibrillator (ICD) implantation. All ICD primary implants, replacements, or revisions performed at the Massachusetts General Hospital between April 1983 and May 1999 were reviewed. A total of 21 ICD-related infections (1.2%) were identified among 1,700 procedures affecting 1.8% of the 1,170 patients who underwent a primary implant, a generator change, or a revision of their systems. The mean follow-up time was 35 +/- 33 months. Of the 959 patients with long-term follow-up, 19 of the 584 patients (3.2%) with abdominal and 2 of the 375 patients (0.5%) with pectoral systems developed ICD-related infections (p = 0.03). There was no significant difference between the infection rate among the 959 primary ICD implants and the 447 replacements or system revisions. Only 5 of the patients (24%) had systemic signs of infection, including fever (T>100.5) and elevated white blood count >12,000. Cultures from the wound revealed staphylococcal species in 16 patients (76%). Nineteen patients were treated with removal of the entire ICD system in addition to intravenous antibiotics for 2 to 4 weeks. A decrease in the incidence of ICD-related infection has occurred since the advent of transvenous pectoral systems. The main organism responsible for ICD infection is Staphylococcus. The mainstay of ICD infection management consists of complete removal of the entire implanted system.

Antibiotic Prophylaxis↗

PubMed2001

['Evidence-based' perioperative antibiotic prophylaxis policy in Belgian hospitals after a change in the reimbursement system].

The costs of antibiotics in Belgian hospitals are nearly fully reimbursed by the health insurance. Such a situation is not conductive to rational drug use. A new reimbursement system for perioperatively-administered antibiotics in Belgian hospitals was implemented in May 1997 by Royal Decree. A reimbursement code for antibiotic use was linked to the reimbursement of surgical interventions. This code represents a reimbursement which covers 75% of the cost of perioperative prophylaxis based on optimal indication, dose, and duration as recommended by international and Belgian consensus guidelines. The actual antibiotic prescribed during the 72-hour perioperative period (the day before, during and after surgery) is reimbursed at only 25% of its full cost. Thus, if the perioperative prophylactic antibiotic regimen complies with the evidence-based guidelines, the costs of antibiotic prescribing will be fully reimbursed by the health insurance (75% of the standard +25% of the actual costs). The new reimbursement system does not apply to antibiotics which are prescribed for treatment of intercurrent infections; these antibiotics continue to be fully reimbursed. Annual expenditures for antibiotics, for both antibiotic treatment and prophylaxis, nationwide and per hospital, have shown marked improvements in perioperative antibiotic use after the decree was implemented. Surgeons' adherence to the evidence-based standard of prophylactic antibiotic use has improved over time. In conclusion, rapid implementation of the perioperative antibiotic prophylaxis policy was achieved through changes in the reimbursement of antibiotics for surgery patients.

Antibiotic Prophylaxis↗