Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

The cost-effectiveness of outreach respiratory care for COPD patients

  • Comment

Adam Gordois, MSc (Health Economics).

Research Fellow

Inpatient treatment, where potentially avoidable, is an unnecessary drain on health service resources. At the margin, the decision to treat a patient in secondary care prevents access for others requiring hospital treatment. Raising the number of emergency admissions places pressure on both traditional emergency beds and beds the hospital would rather protect for elective admissions.

Inpatient treatment, where potentially avoidable, is an unnecessary drain on health service resources. At the margin, the decision to treat a patient in secondary care prevents access for others requiring hospital treatment. Raising the number of emergency admissions places pressure on both traditional emergency beds and beds the hospital would rather protect for elective admissions.

Such problems may be reduced if remedial care can be administered in a domiciliary setting. Intermediate care is a modern approach to patient management that includes services such as ‘hospital at home’, where therapy, assessment, rehabilitation and support are administered in the patient’s own home rather than in a hospital ward.

Particular demands are placed on resources at South Tyneside District Hospital because of the high admission rate for patients with chronic obstructive pulmonary disease (COPD). For example, in 1998-1999 there were 39.2 admissions per 10 000 population in South Tyneside, compared with an average of 26.6 in England (D. Gibbons, personal communication; DoH, 2000a; Office for National Statistics, 2000). The relatively high prevalence of COPD in the region may be due to the history of employment in heavy industries such as shipbuilding and coal mining (Osman et al, 1993). There is a higher incidence of COPD in lower socio-economic groups and among smokers (Weidzicher, 1995). A local survey indicated high prevalences of social deprivation and smoking in the local population. Thirty-five per cent of respondents smoked, while 72% described themselves as ex-smokers (Taylor et al, 1993). Only 4% of COPD patients admitted during the ARAS project had never smoked (D. Gibbons, personal communication).

Past experience
Traditionally, patients admitted to South Tyneside hospital suffering from exacerbations of COPD due to a chest infection, were treated in the hospital until their condition was relieved sufficiently for discharge. As well as being costly, lengthy hospital stays increase the risk of hospital-acquired secondary infection, and hence the likelihood of escalating treatment costs. As a result, South Tyneside was under pressure to develop a better, and potentially cost-saving, strategy for the treatment of patients with COPD.

An acute respiratory assessment service (ARAS) was established at the district hospital in June 1999. The benefits that the ARAS team have delivered have been reported previously (Gibbons et al, 2001). Although the team believed their service to be cost-effective this was in relation to absolute cost-savings on certain treatments. A formal cost comparison of the strategy with previous practice was not undertaken. This paper describes the economic evaluation of the scheme, which enabled a formal analysis of ARAS’s cost-effectiveness to be presented in the request for ongoing funding.

Benefits of an outreach service
We undertook an economic evaluation of ARAS compared with previous practice. This required the key differences in resource use and costs between the two strategies to be identified. Patient treatments, throughput and resource use were recorded by the ARAS team during the initial seven-month project (August 1999 to March 2000).

Box 1 gives a breakdown of the ARAS team. Before ARAS, general hospital staff treated patients with acute exacerbations of uncomplicated COPD.

The ARAS service, available daily from 9am to 5pm, accepts referrals from GPs, accident and emergency, the acute medical admissions unit and medical wards. In addition, patients can self-refer if they have previously used the service. Patients with uncomplicated acute exacerbations of COPD are appropriate for ARAS care. Of the 583 referrals to the team during the project phase, 307 were deemed appropriate. A number of these were multiple referrals (that is, the same patient was referred on more than one occasion) and, in total, 218 patients were treated by ARAS during the project.

The team aims to prevent unnecessary admissions and facilitate early discharge of patients by a thorough patient assessment consisting of medical, nursing and clinical examinations (Box 2).

Streamlined decision-making regarding whether to admit patients or to provide treatment at home follows patient assessment. The team’s guidelines for admission and treatment are based on published clinical evidence (British Thoracic Society, 1997; Callaghan, 1997; Conway, 1998; Gravil, 1998). Where admission is required, the team review the patient daily until he or she meets these discharge criteria. Previously, all patients would have been admitted to hospital and, due to the volume of admissions, not all of the patients would have had input from the respiratory nurse specialist for assessment, education and support.

As the location of care shifts from the hospital to the home, treatment protocols may alter to fit the new context. For ARAS this has meant regular home visits by a dedicated specialist nursing team. At each visit patients undergo a clinical assessment to evaluate their response to treatment. Thereafter, the nursing team makes decisions regarding ongoing care, with reference to evidence-based clinical management protocols. On discharge from the service a pulmonary rehabilitation programme acts as an adjunct to medical treatment to improve patients’ ability to cope with their symptoms. Despite the fact that all patients are offered this treatment, only 30% wished to enrol on the programme. This may reflect the considerable amount of patient time and motivation required, since participation entails twice-weekly visits for eight weeks.

As a result of changing treatment practice, further reductions in treatment cost have been realised through changing or discontinuing medications. Seventy-four patients assessed during the ARAS project phase had been admitted during the previous year. Of these, 40 were on long-term home nebuliser therapy. Under ARAS protocols, 15 were successfully changed to more clinically appropriate and less expensive high-dose inhaler therapy. Investigations by the team indicated that another three patients had been inappropriately prescribed an oxygen concentrator. With education and support from the team, this therapy was successfully discontinued.

For ARAS patients who had been admitted during the previous year, 20% of admissions had resulted in an inappropriate prescription of antibiotics. Although antibiotic therapy is routinely used for COPD patients, not all exacerbations are due to bacterial infections. Therefore, a prescription of amoxycillin or erythromycin may have little impact upon the outcome of patient care. Under ARAS, the consistent use of objective diagnostic investigations such as full blood count, sputum culture and patient temperature allows for more appropriate prescription of antibiotics than reliance on subjective clinical history.

In addition to cost reductions, an improvement in patient outcomes was noted in the decreased rate of readmissions, from 18% to 16% under ARAS. This lower readmission rate may reflect patient education on breathing exercises and chest clearance, leading to fewer chest infections and, hence, greater clinical effectiveness of the new strategy.

Cost-effectiveness analysis
The findings of the ARAS team form the evidence base for the cost-effectiveness analysis of the service. Standard cost-effectiveness analyses synthesise the incremental health benefits and costs accrued from using a new intervention compared with an alternative. However, in the absence of explicit data on health status, patient outcomes were assumed to be at least as good under ARAS as they were under previous practice. This is acceptable in cost-benefit analyses where, in the absence of effectiveness data, the conservative view is taken that a study intervention is no better than the comparator.

A cost-minimisation analysis of the service is then conducted by determining whether the new scheme is less expensive than the alternative strategy. On demonstrating ARAS to be a more costly strategy, the exact degree of additional health benefits it provides would be of importance. Decision-makers would require reference to the additional costs that are required to reap the benefits of a domiciliary service. In the event we found ARAS to be less costly; for the service to be a superior strategy it matters only that ARAS is not clinically inferior to previous practice.

The project phase of ARAS was evaluated against a suitable comparator: the hypothetical scenario of treating the same patients in the absence of the scheme. This is standard practice where data has been collected specifically as an evidence base for a new scheme, since it is unlikely that a dataset of similar quality is available for previous practice. Therefore, it is necessary to hypothesise how the patient group studied would have been treated in the absence of the new service. Because health outcomes are assumed to be the same under ARAS and in the absence of ARAS, the strategy with the lowest cost is deemed the cost-effective option.

Since the results of the evaluation were to be used to inform local budget-holders, the analysis was conducted from the perspective of South Tyneside District Hospital: only costs accruing to the hospital were included. Hence, the analysis accounts for bed days, staff, major therapeutic interventions, patient assessments and home visits. Costs borne by patients, such as travel expenses, were omitted.

The limited nature of data collection within hospitals means that it is usually not possible to calculate the absolute costs incurred under both strategies. The logistical impossibilities of accounting for all resources used in patient treatment and being able to calculate precise resource costs means that estimations are inevitable. The analysis here is limited to key cost components, and either their estimated total usage or the difference in usage between ARAS and the previous practice. Where necessary, these cost components are valued using standard national costs from published sources.

Initial set-up costs for the scheme included the purchase of a computer and database package, two pulse oximeters, two portable spirometers and 25 portable nebulisers, together totalling just under £9000. These initial set-up costs may be regarded as ‘sunk costs’: expenses that would not be incurred consistently during the operation of ARAS, which are excluded from the analysis. The main interest lies in the marginal costs (the costs incurred from treating one additional patient) and an estimate of cost savings that can be generalised over time.

Costs examined were those incurred consistently during the operation of ARAS. The additional resources utilised are summarised in Table 1.

ARAS requires a dedicated team of six staff (4.75 WTE). Staffing costs are estimated as the seven-month joint salary of the team members (£58 936).

During the project phase the ARAS team made 1275 home visits. The cost of these visits is estimated by the specific budget allocated to travel expenses scaled down to a seven-month figure (£1850). The nursing team use their own private vehicles for patient home visits and are reimbursed from this budget on a per-mile basis.

An additional spirometry/reversibility test is required as part of a patient’s assessment. In total, 583 tests were performed during the period. Since there is no published data on the cost of laboratory tests, the cost of a spirometry/reversibility test is estimated as 20 minutes of a grade-G nurse’s salary. This equals £4.39 and represents the cost of staff time taken to perform the test.

Cost savings are achieved through the reduction in total bed days required for COPD patients. With ARAS, fewer bed days were required in the treatment of COPD: 18% of appropriate referrals were discharged on assessment while, for those admitted, the average length of stay fell by four days, from 7.8 to 3.8 days. Bed days saved for inpatients are estimated as the number of ward admissions by ARAS (252 admissions) multiplied by the reduction in the average length of stay achieved under the service (four days). Bed days saved by averting admissions are estimated as the number of referrals immediately discharged (55 referrals) multiplied by the average length of stay before ARAS (7.8 days). We estimate a total saving of 1437 bed days. The cost of a medical inpatient bed day at the hospital is £146.

Data for a sample of patients admitted before the establishment of ARAS showed that 19% were inappropriately prescribed long-term nebuliser therapy. Therefore it was estimated that, during the ARAS project phase, 57 prescriptions of long-term nebuliser therapy were avoided (for 19% of 307 referrals), totalling 427 months of long-term nebuliser therapy over the study period. It was assumed that these patients were assigned to long-term high-dose inhaler therapy by the team. Data also showed that 20% of previous admissions had received inappropriate antibiotics. Therefore it was estimated that 63 courses of antibiotics (that is for 20% of 307 referrals) were avoided during the ARAS project phase. A course of antibiotics usually comprises either 21 amoxycillin capsules (250mg) or 28 erythromycin tablets (500mg). We costed a course as an average of these two prescriptions costs (£5.97), weighted by the relative total private expenditure on the two drugs in England during 1999 (DoH, 2000b). Therapy costs were estimated by the hospital’s pharmacy and the British National Formulary (March 2001).

The additional costs of ARAS were found to be more than offset by the value of resources saved under the new treatment protocols (Table 2). Over the seven months, an additional cost of £63 345 was incurred. However, a gross cost saving of £219 017 was realised. Therefore, a net cost saving of £155 672 was achieved with ARAS.

It is estimated that a cost saving of 40% was achieved over the initial project phase. This was largely driven by a 60% decrease in the total cost of bed days, achieved via reductions in both the average length of inpatient stay due to swifter discharge, and averting admissions when appropriate.

If the perspective of the study was extended, benefits of the ARAS scheme could be realised outside of the hospital. The number of GP contacts made by COPD patients has fallen under ARAS: 31 patients were able to refer themselves back to the scheme under the new protocol, thus saving additional GP consultations. The cost of a GP consultation is estimated to be £18 (Netten, 2000). Further cost savings to GPs were realised by averting an estimated 86 patient months of inappropriate oxygen therapy (based on previous practice). These quantifiable knock-on effects of intermediate care have enabled additional cost savings, totalling £5286, to be realised at the wider local health authority level.

ARAS has been shown to be a cost-saving service in South Tyneside. However, there are some limitations to the generalisability of these results over time and between settings.

First, the epidemiology of COPD may be such that greater or lower prevalences of the disease may be present in different communities and the incidence may change over time. As such, different referral rates to the scheme would be expected. While not affecting the ultimate decision of whether ARAS is cost-effective compared with previous practice, our estimate of the cost-savings would change.

Second, hospital managers may wish to expand the ARAS service. This would incur additional costs over the current scheme. A future consideration for the team is to encompass other respiratory disorders, for example asthma, into ARAS care. This will ensure equity of service for respiratory patients and an improvement in quality of care.

The ARAS service is continually being developed and the team is now a substantive part of the trust. Due to the numbers of patients using the service, the G-grade sisters’ hours have been increased to three WTE. The team has also extended its remit to take on a caseload of 10 chronic patients who have had multiple admissions due to the severity of their illness despite regular input from ARAS. Maintenance home visits are now provided on a weekly or fortnightly basis to enable early detection of deterioration in condition. It is hoped that this will lead to further reductions in hospital admissions and GP visits.

Evidence from a patient-satisfaction survey has indicated that 95% of users would prefer to be cared for by the ARAS team in the future as opposed to the standard medical and nursing care previously received (D. Gibbons, personal communication).

There are several lessons to be learned from South Tyneside. Shifting the location of care from the hospital to the home has allowed for greater cost-efficiency in the treatment of COPD. Less expensive treatment and an emphasis on early discharge has led to a fall in both readmissions and numbers of patients on long-term medication, rather than poor care and resulting morbidity.

Meanwhile, the economic analysis of ARAS sets an example for tackling scepticism towards intermediate and outreach care strategies. The nursing team now have formal evidence of the benefits they provide for the hospital and the region.



British Thoracic Society. (1997)Guidelines for the management of chronic obstructive pulmonary disease. Thorax 52: (suppl 5).

Callaghan, S. (1998)The RITE stuff. Nursing Times 94: 49, 42-43.

Conway, A. (1998)Breathing life into an idea. Nursing Times 94: 39, 72-73.

Department of Health. (2000a)NHS Hospital In-patients: hospital episodes statistics 1998-1999. London: The Stationery Office.

Department of Health. (2000b)Prescription Cost Analysis: England 1999. London: The Stationery Office.

Gibbons, D., Hamilton, J., Maw, G., Telford, J. (2001)Developing a nurse-led service for COPD patients. Professional Nurse 16: 4, 1035-1037.

Gravil, J.H., Al Rawas, O.A., Cotton, M.M. et al. (1998)Home treatment of exacerbations of chronic obstructive pulmonary disease by an acute respiratory assessment service. Lancet 351: 1853-1855.

Netten, A., Curtis, L. (eds). (2000)Unit Costs of Health and Social Care. London: Personal Social Services Research Unit.

Office for National Statistics. (2000)Population: Subnational - Government office regions of England 1971 onwards: Population trends 104. London: The Stationery Office.

Osman, A.D., Muir, D.C., Shennon, H.S. et al. (1993)Occupational dust exposure and chronic obstructive pulmonary disease: a systematic review of the evidence. American Journal of Respiratory Disorders 148: 1, 38-39.

Taylor, D., Stephens, C., Blackman, T., Bellingham, M.J. (eds). (1993)South Tyneside Health and Lifestyle Survey. South Shields: South Tyneside Health Authority.

Weidzicher, W. (1995)Chronic Obstructive Pulmonary Disease. Ingelheim, Germany: Boehringer Ingelheim.


  • Comment

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Links may be included in your comments but HTML is not permitted.

Related Jobs