Cookies on this website
We use cookies to ensure that we give you the best experience on our website. If you click 'Continue' we'll assume that you are happy to receive all cookies and you won't see this message again. Click 'Find out more' for information on how to change your cookie settings.

Economic evaluation of a targeted occupational therapy and physiotherapy intervention for people with stroke living in nursing and residential homes in Oxfordshire

Dates: 2003-2006
Funding: PPP Health Foundation
Collaborators: School of Health Sciences, University of Birmingham
Information: Emma McIntosh

An economic evaluation was undertaken as part of a cluster randomised assessment of targeted occupational therapy (OT)/Physiotherapy (PT) for people with stroke living in nursing homes. The objective of the economic evaluation was to assess the cost and cost effectiveness of providing intensive physiotherapy (PT) and occupational therapy (OT) for patients in nursing homes. This involved the collection of both resource use data and quality of life data alongside the RCT.

Economic evaluation methods

The cost component included collection of data on the costs of providing the intensive occupational and physiotherapy intervention, hospital costs incurred (re-admissions) and community services. The aim of collecting these cost data were to identify both the costs of the intervention and any the potential associated cost implications arising as a consequence of the intervention, e.g. reduced hospital re-admissions. The economic evaluation followed the guidelines established by the National Institute for Clinical Excellence ( and the US panel on cost-effectiveness (Gold et al.1996).

Measurement of economic resource use data

Resource use was measured in naturally occurring units. For example, OT/PT staff time was measured in terms of the number of visits and length of each visit.

Occupational therapy and physiotherapy costs included: staff time, equipment and travel costs. An OT/PT log was developed and completed for both the immediate and delayed intervention arms. The OT/PT log recorded detailed information on the duration of each visit (including face to face contact with patients/carers and other time related to travelling to the nursing/residential home or contact with staff). These data were used to calculate the costs of the intervention.

The logs were also used to record any equipment provided for each patient. Some equipment and consumables used by the OT/PT were purchased directly via the trial budget such as chair raisers. Further equipment was ordered from the local social services. In order to compare the number of OT/PT visits in the delayed intervention these data were collected through prospective three monthly diaries.

Table 1: Summary of data collection methods for resource use

Cost Item resource use data capture
OT/PT staff time with patients Staff time: Number of visits/duration OT/PT log
OT/PT staff time with carer Staff time: Number of visits, duration of visits OT/PT log
OT/PT aids Description Local data provided by social services/price for trial equipment
Dependency in nursing home Nursing home staff time Adapted version of Northwick Park Dependency Scores
Hospital cost
Speciality/length of stay/reason for admission e.g. falls/discharge destination CHRU 3/6 month diaries checked with PAS data if required
GP visit
Other OT*
Other PT*
Clinical Psychologist
Hearing Services
Number of visits in the previous 3 months Home notes, diaries, supplemented with letters to GP and PAS data

In order to determine whether there were resource implications associated with differences in dependency scores following the intervention additional questions were asked alongside the Barthel questionnaire. These questions were based on an adapted version of the Northwick Park Care Needs Assessment, which was designed assess care needs in the community. The objective of these questions was to identify the amount of staff time and number of staff involved in different activities. For example, questions were asked to measure any changes in the number of staff needed for transfers.

Community and hospital resource use

Community and hospital services resource were collected prospectively. Community costs include: General Practitioner (GP), Nurse, OT, PT social worker, Community Psychiatric Nurse (CPN), Chiropodist, clinical psychologists, optician, dentist, and hearing service visits. Hospital data including reasons for admission (e.g. fall), speciality, discharge destination and length of stay.

Initial piloting work revealed that the quality of note keeping varied between the nursing/residential homes. Therefore to supplement these records prospective diaries were designed and inserted into the notes for completion at each home. Despite the best efforts of the research assistants there were still some homes where the records were incomplete. Therefore in order to supplement these data it was necessary to write to some GPs to collect data on community and hospital resource use. Two reminder letters were administered and then any further outstanding missing data was checked against the local hospital Patient Administration System (PAS). The process of collecting the GP data and hospital data was extremely time-consuming as further ethical approval was required for each of these steps.

Valuation of economic resource use data

All prices for the trial were valued at 2006 prices. Where required prices were adjusted using the hospital and community service price index. Unit costs for the items of community care such as OT visits, practice nurse and GP visits were obtained from a national unit cost manual. In the baseline analysis, the national unit cost for an OT visit was used to avoid bias between the immediate intervention arm and the delayed intervention arm. It was assumed that a community PT visit cost the same cost as an OT as these data were not provided in the national unit cost manual.

Potentially unit costs in the trial could differ from the national unit costs for OT and PT service provision. For example, the national unit cost for an OT is based on an average visit length of 40 minutes. Local unit costs were estimated using staff times observed in the trial. Where consumables or equipment were purchased via the trial these prices were used. The unit costs of items of equipment or consumables provided from the local social services were obtained directly from the service provider. All equipment costs were adjusted to a 6-month time horizon based on an assumed usable life of 2-5 years using the equivalent annual cost method.

Re-admission to hospital and duration of hospital stay were valued using readily available NHS unit cost information. These hospital cost data were valued according to the particular specialty e.g. Neurology/Geriatrics.


The costs of occupational therapy and physiotherapy were carefully logged in this analysis and the results show that it is possible to provide this intervention over a three-month period at between £115 and £120 per patient.

The results of the analysis of community and health care resource use indicate that these costs were significantly higher in the intervention group than in the delayed intervention group. One possible explanation of this is that unmet health care need was identified by the physiotherapist and or occupational therapists and hence these costs increased. However, these results should be treated cautiously given the difference at baseline in Barthel scores between the intervention and delayed intervention. For example it may be that the intervention group were in poorer health generally and that is why the average hospital use is higher.

Again the dependency scores (designed to reflect nursing home staff time) were also higher initially in the intervention group (although only borderline significant). At 3 months dependency scores were lower in both groups however, by 3 months dependency scores were significantly lower for some activities in the arm that had not received the intervention. Therefore these results do not support a conclusion that the intervention had lowered dependency in terms of nursing staff time.

Given the conclusions of the clinical study that there was no identifiable improvement of health outcomes with targeted physiotherapy and occupational therapy intervention and given the additional cost of this intervention it would seem that this intervention may not be a cost-effective use of resources. It may be better however to direct physiotherapy and occupational health resources to specific populations where improvements in health outcomes following the intervention can be identified.


For a copy of the final report please contact Emma McIntosh.