CSIRO NBN trial shows big savings with at-home monitoring

At-home monitoring of vital signs in older patients with chronic diseases can lead to significant savings over time in terms of MBS costs, PBS dispensing costs, costs associated with hospitalisations and average length of stay, and it can also reduce mortality rates by up to 37 per cent, new CSIRO data shows.

The data from the CSIRO's NBN-enabled telehealth project also shows that home telehealth services on an annual subscription model, including internet connection, home telehealth device and nurse wages, can be viably offered for about $2400 a year.

With annual savings of about $8000 in MBS, PBS, hospitalisation and visiting nurse costs, this represents an estimated return on investment of between two and three, the figures show.

Lead researcher Branko Celler, formerly principal scientist at the CSIRO Centre for Computational Informatics who has been heading up CSIRO's NBN Telehealth Research Group, told the Health Informatics Conference (HIC 2015) in Brisbane last week that the results will make a significant contribution to government understanding on how to nationally deploy telehealth services and how to develop cost-effective and sustainable models to fund them.

Professor Celler's team was one of the winning bidders for the NBN-enabled Telehealth Pilots Program announced in 2013. The $3 million randomised control trial recruited hundreds of patients in several states who were provided with a free internet connection and Telemedcare device, which allows for video conferencing and vital signs monitoring through peripherals including blood pressure monitors, pulse oximeters and ECGs.

The participants had an average age of 71, had one or more chronic illnesses such as congestive heart failure, chronic obstructive pulmonary disease, diabetes, hypertension and coronary heart disease, and had been hospitalised twice in the preceding year.

They were asked to take their measurements once a day, with a compliance rate of 68.2 per cent, and over the course of the trial about 111,000 measurements were taken, including 9000 ECGs, 20,000 blood pressure measurements and 20,000 pulse oximetries.

Healthcare costs for this cohort of patients can add up very quickly. The CSIRO trial found that over a 100-day period, patients visited the GP 5.7 times at a cost of $245; they had 0.6 visits from allied health, which cost them $30; they had 1.6 visits to specialists for a cost of $159; and they had 28 medications dispensed over the period at a cost of $959.

“The total cost of procedures and tests was $625,” Professor Celler said. “When you look at a total cost of PBS and MBS items, it's about $2029, and they spent $44.40 travelling to see the GP.”

While there is plenty of international evidence that telehealth might be able to reduce some of these costs, there has not been enough local evidence to sway the government to seriously invest in and fund in-home monitoring for chronic illness.

However, the evidence from this trial is very persuasive. It shows that costs to the MBS for chronically ill patients over the age of 71 can be reduced by about $800 or up to 31 per cent per year, PBS dispensing costs by $200 or 10 per cent per year and substantial reductions in the avoidable costs of hospitalisations and length of stay.

Reductions in mortality rates for this cohort were comparable to those from international studies such as the UK's Whole System Demonstrator, Professor Celler said.

“We compared the mortality between our test and our control patients and the calculation is that the probability of death for our test patients was 3.75 per cent and the probability of death for our control patients was 5.95 per cent,” he said.

“This means that telehealth reduced the mortality by 37 per cent against the control [group].”

In terms of hospitalisations, there was a reduction of about 0.6 to 0.7 presentations over a year, for a saving of about $3000 on average. Acute care beds can cost well over $1000 a day.

There was also a significant change in length of stay for those who did end up in hospital, Professor Celler said. “Patients that were being monitored at home and that did go to hospital spent less time in hospital by about a day, or a day and a bit. These data are statistically significant.”

He described the MBS and PBS data as “very gratifying”.

“This has never been shown before, that telehealth actually has an impact on MBS expenditure, which is all procedures, visits to GPs, visits to specialists,” he said. “It does reduce it.”

And the return on investment is likely to be between two and three, he said. The initial estimate of cost for a subscription-based telehealth model was about $12 a day, which includes $7 a day for video conferencing, messaging and a comprehensive suite of instruments, $1 a day for the internet and $4 a day for monitoring and triage.

“Telehealth, to be run professionally, had to be around $12 per day on a subscription based model,” Professor Celler said.

“However, we followed that discussion further and at the end of the trial we re-evaluated the likely costs. The cost per month for a chronically ill patient of some significant severity, we believe the monitoring, hosting and maintenance cost will be that of $150 per month. The nurse coordination, at 0.5 hours per week per patient, it is about $50 per month.

“And that means that the cost of running a telehealth service ought to be around the $2400 mark. The annual savings estimates, the savings in PBS dispensing that were shown to be $200, the savings in the MBS items of $800 … even if you consider the rate of hospitalisation to be only half per annum, for chronically ill patients that’s a saving of $3000 per year, because for chronically ill people, hospitalisation are typically around $6700 per year.

“And the reduced demand on community nursing. This is based around a saving of five per cent [full-time equivalent] per patient for a community nurse, based on the fact that a community nurse would ... have a caseload of about 10 patients, and this could be increased [using telehealth].

“So that is a total saving of about $8000 and that gives you an estimated return on investment of about three. We believe this is a fairly realistic figure, although quite preliminary still. We are doing some of the final calculations.”

Posted in Australian eHealth

Comments   

# Phil 2015-08-14 16:06
Finally, an Australian business case for RPM. Well done Branko and team.

Now we need the private health funds to adopt in the first instance and then the government for their Medicare frequent flyers.
# Georgia 2015-08-17 13:02
This is very exciting. Is there a report being published, or has it been published were we can examine how the study was put together and the results . This may have a significant impact on health resourcing as long as the government recognizes the true impact of remote monitoring.
# Sandy 2015-08-18 10:59
As with above comment, is this study published and if so where and when. Thanks
# Kate McDonald 2015-08-18 11:12
Hi Georgia and Sandy

The trial only finished in December and Professor Celler said his team was still analysing the data, of which there was an enormous amount. There will be numerous papers, I suspect.

Prof Celler's presentation slides with details about the study are here: http://www.hisa.org.au/slides/mon/HIC2015_CSIRO_Celler.pdf

Pulse+IT has also written a few background stories:

http://www.pulseitmagazine.com.au/index.php?option=com_content&view=article&id=2003:telehealth-in-the-home-nbn-remains-the-gold-standard&catid=67:aged-care&Itemid=332

http://www.pulseitmagazine.com.au/index.php?option=com_content&view=article&id=1421:csiro-to-demonstrate-evidence-for-telehealth-under-nbn-project&catid=16:australian-ehealth&Itemid=328
# Tim Chambers 2016-09-15 09:20
Some numbers don't match the final report. Great significance has been applied to some results where confidence intervals are overlapping ie. results are not significant and cannot be relied upon for the calculations made in the final report. The total costs, $/day are different. The benefits could be overstated due to insignificant results. http://www.agedcare2015conference.com/wp-content/uploads/2015/10/Branko-Celler.pdf
https://www.csiro.au/~/media/BF/Files/Telehealth-Trial-Final-Report-May-2016_3-Final.pdf
About 50% of potential test subjects did not consent or dropped out after start. Compliance with the expected actions sometimes below 50% of the frequency. MBS savings are using asumptions of compounding reductions ie. a downward slope implying ever increasing savings towards costing $0/month if continued long enough. If implemented, Telehealth may cost twice the amount to get the saving and cannot be used by 100% of expected target population. Remember that they used ADSL and mention data volumes that do not require an NBN connection.
# Kate McDonald 2016-09-15 09:31
Hi Tim. This article is as you will see from August last year. Branko and his team have since revised the figures based on the cost of a hospital bed per day for the final report.

I covered the final report in May: https://www.pulseitmagazine.com.au/news/australian-ehealth/3111-csiro-study-shows-massive-return-on-investment-for-at-home-telemonitoring?highlight=WyJicmFua28iXQ==

And then did an update when the CSIRO actually made the report publicly available last month: https://www.pulseitmagazine.com.au/news/australian-ehealth/3111-csiro-study-shows-massive-return-on-investment-for-at-home-telemonitoring?highlight=WyJicmFua28iXQ==

You need to log in to post comments. If you don't have a Pulse+IT website account, click here to subscribe.

Sign up for Pulse+IT eNewsletters

Sign up for Pulse+IT website access

For more information, click here.

Copyright © 2017 Pulse+IT Magazine
No content published on this website can be reproduced by any person for any reason without the prior written permission of the publisher.