A Journal of Hospital Infection special edition focusses on the cost of HCAI, publishing the results of the Scottish from the Evaluation of Cost of Nosocomial Infection (ECONI) Study. 

• The first article provides a point prevalence study on the epidemiology of HCAI in a general teaching hospital in Scotland, which informed the economic analyses in the other studies in the special issue. The overall rate of HCAI was 250 per 100,000 occupied bed days, with UTIs the highest rate at 51 per 100,000 occupied bed days, followed by BSIs at 45 per 100,000 bed days. I found it quite frustrating that the HCAI data were not presented as a % of inpatients with an HCAI, which seems to be the common way to present PPS HCAI data – this made comparison with other HCAI PPS studies difficult. However, whilst UTI are often most common HCAI (as was the case in the most recent national HCAI PPS in Scotland), it’s uncommon to see BSI second most common, and respiratory tract infection only 6^th most common. I guess this illustrates the challenges of a single centre vs. larger multi-centre study.
• The next article is all about HCAI and length of stay (LOS). The study concluded that the excess LOS attributable to HCAI was 7.8 days. The article highlighted the importance of accounting for time-dependent bias (essentially the fact that patients who have HCAI have other co-morbidities that make them incomparable to most hospital admissions with relatively short inpatient stays); the crude difference in length of stay between patients with/without and HCAI was 27 days (compared with the adjusted figure of 7.8 days). Pneumonia was associated with the greatest increase in LOS – but is probably the “Cinderella” HCAI in terms of prevention focus!
• The next study looks at the cost of HCAI. The national estimate of HCAI costs in Scotland based on this study was almost £50m (with an estimate range of £19m to £129m), and for the UK was £774m (range £328m–£2,192m). Most of the cost of HCAI is driven by excess bed days. Only 2.4% of the increased cost associated with HCAI was cash expenditure. This is important, because if we are asked to say exactly where these large costs of HCAI will appear as revenue in hospital budgets as a result of the improvement initiatives we are pitching (and I have been asked this question more than once!), we can’t answer that. So, the impact of these figures depends on your perspective. If you are only interested in cash flow, you have bigger priorities than HCAI. But if you are interested in quality of care and bed closures, then HCAI become increasingly important. 
• The next study evaluated post-discharge costs associated with HCAI. As you might imagine, patients who developed an HCAI as an inpatient had a more complicated journey post discharge, and this was associated with almost £1,500 excess cost per HCAI in the 90 days after discharge. Patients who had developed an HCAI were also more likely to be readmitted and to be prescribed antibiotics in the community. As the authors reflect, post-discharge costs related to HCAI are rarely quantified, so this is a really helpful study.
• The next study identified individualised risk factors for subsequent HCAI at the time of admission. Factors associated with an increased risk of HCAI were admission to a teaching hospital; prior surgery; emergency admission; cancer, cardiovascular disease, chronic renal failure, diabetes; critical care stay or surgery during the episode of care; and a long LOS. Whilst the authors suggest that these findings could be used as a prediction tool to identify those at a high risk of HCAI, we’d have to think carefully about this because a huge proportion of the hospitalised population would have quite a few of these risk factors. And then if we could come up with a model identifying a manageable number of patients, what would we do differently? We already try to prevent HCAI in all patients – so would we really try to prevent HCAI in these patients?

Overall, this series of publications are incredibly helpful and take forward our understand on the increased costs of HCAI. A limitation is that all of the interest cost and impact modelling is based on a single centre point prevalence estimate so may not be generalisable. I really liked the inclusion of post-discharge costs, and this is something that should become standardised in economic evaluations of HCAI.