Research published in the BMJ found that, compared with usual care, using telehealth devices to monitor patients' health cut hospital admissions by 18% and deaths by 46%.
But there was no evidence that savings from reduced hospital admissions could offset the cost of the devices.
Researchers said bed spaces not taken up by people using the devices may simply be filled by other patients, meaning no cash savings are released for commissioners to use elsewhere.
In March, care services minister Paul Burstow said telecare and telehealth devices could save the NHS up to £1.2bn over five years.
This claim is central to DH plans to roll out the technology to three million people by 2016. The DH signed a 'concordat' with industry in January, before full results of the trial were released.
Now, researchers have finally published the first analysis of the DH-backed Whole System Demonstrator project, the world’s largest telehealth RCT.
Researchers randomised 3,230 patients with diabetes, COPD or heart failure to using telehealth or usual care for over one year, with 179 practices across Cornwall, Kent and Newham in London taking part.
Patients took regular readings using a range of devices including a pulse oximeter for COPD, a glucometer for diabetes and weighing scales for heart failure.
Researchers found that, over 12 months, 43% of telehealth users were admitted to hospital compared with 48% of control patients, a relative difference of 18%.
In total, 4.6% of patients using telehealth died in this period compared with 8.3% of controls: equal to 46% relative reduction or around 60 lives a year.
No evidence of savings
But researchers found no difference in the number of elective admissions, outpatient attendances or emergency department visits between groups.
Furthermore, hospital admissions among controls rose steeply at the start of the study period.
Excluding the first three months of data from the trial removed all differences in overall activity levels between groups, casting doubts over the longevity of benefits.
Authors concluded telehealth may help patients better manage their condition and avoid worsening symptoms that can lead to hospital admission.
But researchers said differences in hospital costs between groups were 'relatively small' at around £188 per patient over 12 months, 'especially compared with the potentially high costs of these types of telehealth intervention', and may have been due to chance.
As such, researchers said they ‘cannot conclude that telehealth reduces secondary care costs over 12 months’.
Researchers also questioned how much of any savings GP commissioners would see, as telehealth expert David Barret of Hull University recently told GP.
Authors said: 'For commissioners of care services, there are questions about whether any reduction in hospital use for patients receiving telehealth translates to an overall change at the organisational level.
'Any bed days released as a result could be filled with non-study patients rather than released as cash savings.'
Full scale roll-out 'unwarranted'
In an editorial, Dr Josip Car, director of the Global eHealth Unit at Imperial College London and colleagues, said the evidence doesn’t warrant full scale roll-out, as planned by the DH, and telehealth should be targeted at specific groups instead.
They wrote: 'Does the demonstrator trial provide convincing evidence for commissioning a national roll-out of telehealth? The findings reported to date suggest not, although we recommend caution until the full data are released.'
However, the 'wealth' of data generated by rolling out telemonitoring could support future research into the devices' effectiveness, they said.
Study authors writing in a Nuffield Trust summary of the findings said many NHS quality, innovation, productivity and prevention (QIPP) efficiency plans assume financial savings from investment in telehealth.
'Commissioners, providers and policy-makers keen on introducing similar telehealth technologies should look carefully at this trial, both at the intervention and at how it was targeted, and consider whether it is necessary to refine the approach,' they said.
Lead author Adam Steventon of the Nuffield Trust told GP the study raised 'all sorts of questions' about how long the effects of telehealth would last. 'If it is appropriate for patient to receive telehealth, then how long for? Three months or several years?'
He added that the spike in admissions among the control group was unexpected. 'It does raise questions about whether the findings we saw in emergency admissions are attributable to telehealth or some sort of effect in the way the trial was operated,' he said.
Responding to the research, care services minister Paul Burstow said: 'This paper shows that telehealth can help to reduce emergency hospital admissions by 20% and mortality by 45% for patients with long-term conditions. To realise these benefits we need service transformation - and telehealth can help deliver that change.
'We are working closely with industry, the NHS and social care organisations to make progress through the "3millionlives" initiative to develop flexible funding models with a reduced price point, which will achieve the economies of scale needed to make telehealth a success.'
The BMJ paper is the first of five studies to be published over the coming months and years. Other analyses will examine telehealth's effect on QoL, cost effectiveness and patient, professional and organisational impact of rolling out the technology nationally.
The DH initially refused to explain how it had calculated the £1.2bn figure, but relented after an appeal by GP.
This revealed that the department predicts as many as two million patients will use the technology, saving the NHS £70 to £543 each per year, mainly from fewer hospital admissions.
