The impact of parenteral nutrition preparation on bloodstream infection risk and costs.

BACKGROUND: Catheter-related bloodstream infections (BSIs) are a serious problem leading to increased morbidity, longer hospital stay, and hence, additional costs. This study evaluated the risk of BSI and the cost of parenteral nutrition (PN) in Germany. SUBJECTS/METHODS: A retrospective observational chart review of patients hospitalized from October 2009 to April 2011 and receiving PN via ready-to-use three-chamber bag (MCB), single bottle (SB) or hospital compounded admixture (CPN) was conducted across Germany. Propensity score-adjusted models were used to evaluate the association between the type of PN, BSI (Cox Proportional Hazards) and hospitalization cost (generalized linear models) within a subgroup receiving all three macronutrients (lipids, amino acids, glucose). RESULTS: Of the 1995 patient records reviewed (MCB=816; CPN=584; SB=595), 1457 patients received all three macronutrients. After adjustment, SB was associated with an increased hazard of BSI, vs MCB without additions (hazard ratio (HR) (95% confidence interval (CI))=2.53 (1.66-3.86)) in the total cohort. Adding supplements to MCB on the ward also increased the BSI risk in both total and subgroup analyses. In patients receiving all three macronutrients, adjusted total costs were MCB (no additions): \[euro]6,572 (95% CI: \[euro]6,896-6263); CPN: \[euro]6,869 (\[euro]7,283-6479); SB: \[euro]6,872 (\[euro]7,242-6521); MCB (ward additions): \[euro]7,402 (\[euro]7,878-6955); P<0.001; P<0.001. CONCLUSION: Use of MCB does not appear to increase treatment costs, possibly by reducing the risk of infection. This study identified several PN preparation methods associated with a significantly increased hazard for BSI; definitive CPN findings are limited by our inability to distinguish automated from manual pharmacy compounding.

Obtaining disability weights in rural Burkina Faso using a culturally adapted visual analogue scale.

Burden of disease (BOD) estimates used to foster local health policy require disability weights which represent local preferences for different health states. The global burden of disease (GBD) study presumes that disability weights are universal and equal across countries and cultures, but this is questionable. This indicates the need to measure local disability weights across nations and/or cultures. We developed a culturally adapted version of the visual analogue scale (VAS) for a setting in rural Burkina Faso. Using an anthropologic approach, BOD-relevant health states were translated into culturally meaningful disability scenarios. The scaling procedure was adapted using a locally relevant scale. Nine hypothetical health states were evaluated by seven panels of in total 39 lay individuals and 17 health professionals. Results show that health professionals' rankings and valuations of health states matched those of lay people to a certain extent. In comparison to that of the lay people, health professionals rated seven out of nine health states as slightly to moderately less severe. The instrument scored well on inter-panel and test-retest reliability and construct validity. Our research shows the feasibility of eliciting disability weights in a rural African setting using a culturally adapted VAS. Moreover, the results of the present study suggest that it might be possible to use health professionals' preferences on disability weights as a proxy for lay people's preferences.