Dear authors,
Thank you for the work you have done here on HAM.
Hospital acquired malnutrition should be a never event in our healthcare systems, so I am happy to see more research being undertaken to measure this issue. There are a plethora of studies investigating the prevalence of malnutrition, but very few on the prevalence of HAM. So, whilst I applaud the intent of this study, I have considerable feedback on the execution and interpretation of findings.

Overarchingly, the sample size was too small, despite the authors referencing the identified prevalence rate of HAM being similar to other studies. With the knowledge of previous studies, where HAM prevalence is found to be <5%, a much larger sample should have been planned to study. The resultant finding that 5 or 6 (?) patients had HAM made attempts to then identify descriptors that could identify risk of HAM generally invalid.

The large number of patients who were unable to consent or declined is of concern. As correctly pointed out in the limitations, the HAM prevalence (and indeed malnutrition prevalence) is likely under reported, as many participants who declined or were unable to consent were likely very unwell or cognitively impaired, where the risk of malnutrition is known to be greater.

With respect to the factors associated with HAM, as already noted above, the sample size is too small to make these assertions. For example, the reference to over-representation of Indigenous patients cannot be made, when this is based on 2/5 patients. The statement that 'patients with HAM remained in hospital on average 72 days longer than those with malnutrition on POA, implies a causation effect of HAM on LOS, when in fact, more obviously patients who stay in hospital longer are at greater risk of HAM. This is also why patients with HAM were more likely to be rehabilitation or sub-acute wards.

The biggest issue with developing malnutrition (in hospital) is poor nutritional intake relative to requirements, yet there is no reference to the importance of monitoring nutrition intake symptoms and intake in long stay patients, which is the most useful screening process to identify risk of HAM.

The references to the Nutrition Dashboard in this article demonstrate its usefulness in presenting daily cross sectional data on nutrition status, and potentially other useful nutrition indicators. But the information in the dashboard is only as good as the information put in by clinicians. If the dashboard was able to flag patients at risk of HAM eg. from LOS and nutrition impact factors this would be very useful.

thank you again for your efforts.