Fining councils for delayed discharges has almost halved the extra time that older people spend in hospital, but up to half of those patients are back in hospital within three months, a report by the Commission for Social Care Inspection revealed last week..

The research, which looked at the effects of delayed discharge reimbursement, also suggests that some councils are turning to residential care as a quick fix to avoid fines - at the expense of intensive home care and rehabilitation services.

Researchers looked at 151 case files at seven local authorities, and interviewed 70 older people about their experiences of discharge in the first three months of the reimbursement system's operation.

Although the CSCI was unable to find an instance of the threat of fines leading to a patient being discharged prematurely or into unacceptable care, it says that something is clearly going wrong for patients.

"Reviewers did see some examples of people with chronic or complex needs being ping-ponged around the system," says the report.

One older person recalled being talked into leaving their sheltered housing for a nursing home while in hospital. "They said it was for me to decide, but with the drugs I was on I was in no fit state to do so."

The controversial policy, introduced in October 2003, allows NHS trusts to levy fines of £100 (£120 in London) for every extra day a patient spends in hospital due to delays in organising social care.

On one level, it has been successful. Figures for England show that in the first six months of the policy, the number of extra hospital days that patients spent waiting for social services assessments fell by two-thirds.

But it has prompted concerns that decisions made in haste to avoid fines are driving up hospital readmission rates. The Healthcare Commission reported earlier this year that the proportion of people over 75 who needed emergency readmission to hospital within 28 days had shot up from 7.1 per cent to 8.2 per cent during the period when delayed discharges had fallen the fastest.

At one council examined by the CSCI, more than half of older patients were readmitted to hospital within three months of their discharge. In the best-performing authority this figure was 8 per cent.

In a few cases, older people's fear of NHS continuing care costs may be responsible, the report suggests. Some people with a clear need refused all further help once the six-week no-charge period had elapsed.

"Most of these people worried about the cost and preferred to struggle on alone," says the report. "It is quite possible this could accelerate the need for hospital readmission."

There is also evidence that some councils are using residential care homes as a quick fix to avoid fines.

In one authority one in three older people went directly into residential care after hospital discharge, compared with one in 25 in the authority that made least use of them. In three of the seven authorities studied, admissions to residential and nursing homes increased by more than the amount forecast.

Worryingly, older people are often pressurised to make life-changing decisions about their future care while still in hospital - a time when they are most vulnerable. In some councils one in three people needing social care after discharge reported this experience.

David Behan, the CSCI's chief inspector, warns of the dangers of rushing people into life-defining decisions from their hospital bed.

"We found a number of cases where people were being pushed into long-term residential care when they could have been supported to live in their own homes."

But equally, the report cites cases where the new notification system, where doctors inform social services within three days of a patient's discharge, enabled social workers to challenge these decisions if the person was not mobile or lacking a mental health or continuing care assessment.

Is an over-reliance on institutional care linked to the rise in revolving door admissions? The CSCI does not address this question, but has promised to revisit the issue next year, perhaps giving a clearer picture.

Frank Ursell, chief executive of the Registered Nursing Home Association, says that a correlation is unlikely as studies show that residential and nursing homes have the lowest rate of hospital readmission.

Ursell believes the difficulties of getting funded NHS continuing care are a more likely explanation. "A big problem with readmissions is that a large number of people who should go into nursing homes are going into care homes."

But the report does suggest that councils are more likely to use residential care where community-based services are lacking. "Councils with delayed discharge problems also had underdeveloped community provision," it says. "There is some evidence that these councils may be turning to institutional options as a quick solution to discharge pressures."

The CSCI noted more "fragility" in intensive home care services in many authorities it investigated. In four councils there was a decline in people receiving intensive home care services, while use of non-residential intermediate care declined at three authorities. The proportion of people receiving rehabilitation services ranged from none in one authority to a third in another.

Incomplete assessments were a particular problem in one area. In the worst-performing authority, 12 out of 20 delayed discharge days in one month were attributed to incomplete assessments.

Equally significant were the difficulties people encountered in accessing community health provision, sheltered housing and transport.

Glenys Jones, Sunderland's social services director and co-chair of the Association of Directors of Social Services older people's committee, says high readmission rates are "extremely worrying".

"In some areas intermediate care has seemed like an alternative to community rehabilitation. It is not. It should be placed alongside. We need much greater focus on risk identification and proactive engagement," she says.

Although delayed discharge fines superficially appear to be a roaring success, the jury is still out on them. Discharge rates were improving before the system was introduced anyway. The fines policy may have accelerated that improvement, but it may have come at the expense of care in the community solutions, such as rapid response and 24-hour crisis teams.

"There is some risk that the spotlight on speed of discharge may take effort away from extending community capacity to prevent hospital admissions in the very councils where this development is most fragile," the CSCI report notes. "Localities with less well-developed community provision know this is where investment is needed but are pulled towards securing quick fixes to support discharge."

Instead of letting health and social care get on with developing intermediate care and rehabilitation, has the system just introduced conflict and a drain on resources?

Much of this investment has had benefits, and there are plenty of examples of good practice. Authorities which invested in additional managers to co-ordinate discharges - Cornwall, for instance - have had the best results for patients. Better discharge liaison between health and social care has also improved communications between the NHS and social services rather than pulling them apart.

But Ray Jones, Wiltshire's social services director, disagrees. "The concept of fining wasn't what has got us there but a combination of performance indicators and earmarked funding to invest in the system. I wouldn't want to have to have those fining discussions again."

Many senior figures are now hostile to extending the fine system to mental health and intermediate care patients.

"We have made dramatic improvements in spite of fines and it has cost us £250,000 to administer the process," says David Munro, Surrey executive member for social services. "We have done it by ourselves, with health, and I don't want a fining system ever again."