The analysis was based on a decision tree model with a time horizon of the lifetime of the newborn. The authors stated a societal perspective.
A literature search of electronic databases was carried out to identify relevant sources of evidence. English-language literature from 1993 to October 2011 was searched and 28 full-text articles were identified. The level of quality of each study was evaluated and the most relevant studies were selected. Model probabilities came from several sources that included systematic reviews of clinical trials, cohort studies and other observational studies. No further details on these sources were given. Development of GBS sepsis 48 hours after birth and within seven days of birth was the primary input of the analysis.
Monetary benefit and utility valuations:
Utility valuations for most health states were derived from published studies relative to individuals with cerebral palsy.
Measure of benefit:
Quality-adjusted life-years (QALYs) were used as the summary benefit measure and were discounted at an annual rate of 3%.
The economic analysis included costs of an additional 24-hour hospital observation, an emergency department visit for newborns developing early onset GBS (EOGBS) while not under in-patient observation, EOGBS sepsis treatment, the lifetime cost of long-term sequelae resulting from EOGBS sepsis and productivity losses based on average earnings of workers in USA. Costs were mainly taken from previously published cost studies and official databases on hospital and physician services. Utility costs associated with EOGBS sepsis sequelae were based on patients with cerebral palsy. Costs were in US dollars ($). The price year was 2010. Costs were discounted at an annual rate of 3%.
Analysis of uncertainty:
A multivariate sensitivity analysis was carried out using a Monte Carlo simulation. Pre-specified probability distributions were selected for model inputs using a Program Evaluation and Review Technique (PERT) distribution. Cost-effectiveness acceptability curves were generated. Productivity losses were excluded in an alternative scenario. Discount rates were also varied.