The cost of management of patients with atrial fibrillation: An observational study in UK NHS primary care

Dr George Kassianos, The Ringmead Medical Practice, Bracknell; Dr Ahmet Fuat, Carmel Medical Practice, Darlington and Centre for Integrated Health Care Research, Durham University; Dr Chris Arden, Park Surgery, Chandlers Ford; Dr Simon Hogan, Sanofi, Guildford; Laura Baldock, pH Associates, Marlow

Introduction

The management of atrial fibrillation (AF) represents a significant burden on the UK National Health Service (NHS), both in primary and secondary care1

With the incidence and prevalence of AF predicted to rise significantly in the coming years2, this economic burden has the potential to increase unless management efficiencies are made

A clear understanding of the resource use and costs associated with the current management of AF is important in informing future planning and policy development

Methods

Design:

A retrospective, observational research study undertaken in 8 UK primary care practices, three of which provided their own anticoagulation services (i.e. patient consultations for warfarin management and regular INR monitoring)

Review of physicians’ routine clinical and prescribing records, conducted between March and August 2010, according to a standardised protocol and data collection form

No change to the management of patients for the purposes of any part of the review

Research ethics and local research and development (R&D) approval obtained in each participating primary care trust

Study patients and data collection:

825 adult patients with AF (≥18 years at diagnosis), providing written informed consent for researcher access to their medical records

Patients diagnosed less than 12 weeks before data collection, those with secondary AF and those with no diagnosis date were excluded

For the purposes of this study the first 12 weeks of management following diagnosis of AF was defined as the ‘initiation phase’. The period from week 12 onwards was referred to as the ‘maintenance phase’

For patients who had been recently diagnosed with AF (<9 months before data collection), data were collected on the initiation phase only

For patients diagnosed more than 3 years before data collection, data were collected on the most recent 3 years of management (i.e. the maintenance phase only)

Patients diagnosed between 9 months and 3 years before data collection provided data on both the initiation and maintenance phases of management

Data analysis:

Costs (Great British Pounds, GBP) were assigned to AF-related healthcare resource use (medications, primary care visits, secondary care attendances [emergency department, outpatient and daycare], hospitalisations, investigations and blood tests) using published NHS reference costs3-5. These were used to calculate a total cost per patient for the initiation and maintenance phases of management (as applicable). The cost of primary care anticoagulation visits (for centres providing anticoagulation services) was not included in the total costs, as details of anticoagulation visits for patients in other practices (which would have taken place in secondary care) were not available in the primary care records and so were outside the scope of data collection

Analysis included stratification of costs by type of centre (i.e. practices with / without their own anticoagulation service)

Multiple regression analysis was performed to determine which combination of variables contributed most to total costs during both the initiation and maintenance phases of management

Where costs are presented in Euros (€) or US Dollars ($), conversion from Great British Pounds (£) was based on 2010 exchange rates

Demographics and sample characteristics

Acknowledgements & Conflict of Interests This study was sponsored by Sanofi. A. Fuat and C. Arden have no conflict of interest. G Kassianos has received honorarium from Sanofi. S. Hogan is an employee of Sanofi and L. Baldock is an employee of an agency funded by Sanofi.

Presented at the ISPOR 15th Annual European Congress, Berlin 3rd-7th Nov 2012. Abstract no. 39106

Objective

To describe the National Health Service (NHS) costs associated with the management of Atrial Fibrillation in routine UK clinical practice

The mean total cost of AF management was £947 (€1,153/$1,476) per patient in the initiation phase and £469 (€571/$731) per patient year in the maintenance phase.

Inpatient admissions and secondary care attendances accounted for 83% of total initiation phase and 64% of total maintenance phase costs (Table 1).

References: 1. The Office of Health Economics. Estimating the direct costs of atrial fibrillation to the NHS in the constituent countries of the

UK and at SHA level in England, 2008. November 2009, London 2. Miyasaka Y, Barnes M, Gersh BJ, Cha SS, Bailey KR, Abhayaratna WP, et al. Secular trends in incidence of atrial fibrillation in

Olmsted County, Minnesota, 1980-2000, and implications on the predictions for future prevalence. Circulation 2006;114:119- 25

3. Department of Health. National Schedule of NHS Reference Costs 2009/10. Available at http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_123459

4. Personal Social Services Research Unit (PSSRU) Unit costs of Health and Social Care 2010. Available at http://www.pssru.ac.uk/uc/uc2010contents.htm

5. Joint Formulary Committee. British National Formulary. Edition 61. London: BMJ Group and Pharmaceutical Press; 2011

Results

Cost per patient (initiation phase, n=310)

Cost per patient year (maintenance phase, n=769)

Component of AF management

Mean SD*1 % of total

cost*2 Mean SD % of total

cost

Investigations £62 £64 6.5% £51 £34 10.9%

Blood testing £6 £10 <1% £9 £13 1.9%

AF medications £11 £11 1.2% £61 £61 13.0%

Primary care visits*3 £78 £69 8.2% £49 £58 10.4%

Secondary care visits £166 £201 17.5% £128 £206 27.3%

Inpatient admissions £624 £1,006 65.9% £170 £448 36.2%

Total cost £947 £1,098 - £469 £597 -

Table 1: Total costs and cost of each component of AF management

Figure 2: Distribution of total costs per patient in the initiation and maintenance phases of management

The between-patient range of costs was high, with the care of most patients (57% in the initiation phase and 72% in the maintenance phase) costing less than £500 (€609/$779) per patient / per patient year. Thirty three percent of patients in the initiation phase and 12% in the maintenance phase had

care costs of more than £1,000 (€1,218/$1,559) per patient / per patient year (Figure 2).

*1 SD = Standard deviation

*2 Mean cost for each component in initiation or maintenance phase, divided by the mean total cost for that period

*3 Excluding anticoagulation visits

Data were collected on a total of 825 patients. Data were available on the initiation phase from 310 patients and from 769 patients on the maintenance phase.

Four hundred and sixty two patients (56%) were male ; the mean age at diagnosis of AF was 70.5 years (range 22.4-95.7 years).

Figure 3: Total costs, stratified by type of centre

The mean maintenance phase cost per patient year was significantly higher for practices providing anticoagulation services (£555/€676/$865) than for practices without these services (£421/€513/$656), p=0.002 (Figure 3), even though primary care visits directly attributable to anticoagulation were excluded from the analysis.

The higher maintenance phase costs in these centres results from significantly higher costs for investigations (p<0.0001), blood testing (p<0.0001), primary care visits (p<0.0001) and secondary care attendances (p=0.008). There was no significant difference between the two types of centre in the maintenance phase costs for AF medications (p=0.337) or inpatient admissions (p=0.141).

Table 2: Multivariate analysis – variables contributing to total cost in the initiation and maintenance phases of management

Regression coefficient (£)*1

95% confidence interval (£) (+/-) P value

a) Patient variables contributing to total cost in the initiation phase

Age at diagnosis (per year) -16.18 11.59 0.01

Presence of hypertension 457.47 247.11 <0.001

b) Patient / initiation phase variables contributing to total cost (per year) in maintenance phase

Presence of congestive heart disease 151.92 106.21 0.01

Presence of structural heart disease 161.12 110.09 0.004

Presence of diabetes 182.20 113.87 0.002

Presence of dyslipidaemia -94.42 88.75 0.04

No. hospitalisations in initiation phase*2 370.05 93.67 <0.0001

No. ECGs in initiation phase*2 102.48 80.32 0.01

No. daycare attendances in initiation phase*2 832.87 291.54 <0.0001

*1 Regression coefficients show the added/reduced initiation/maintenance phase cost (£) associated with each variable

*2 Added cost per hospitalisation, ECG or daycare attendance in the initiation phase

In the initiation phase, the variables contributing most to total cost were age at diagnosis (reducing costs by £16 for each increasing year of age) and hypertension (adding £457 to total costs) (Table 2 part a). However, these variables combined, only explained 5% of the variability in initiation phase costs (r=0.05).

In the maintenance phase, the patient variables contributing most to total cost were the presence of congestive heart disease, structural heart disease and diabetes, which added £152, £161 and £182 (respectively) to the total cost per year in the maintenance phase (Table 2 part b). The presence of dyslipidaemia reduced maintenance phase costs by £94 per year. The number of hospitalisations, ECGs and daycare attendances in the initiation phase were also found to contribute to increased maintenance phase costs. All of these patient and initiation phase variables combined, explained 18% of the variability in maintenance phase costs (r=0.18).

P=0.24 P=0.002

Conclusions:

• This study confirms that inpatient admissions and secondary care attendances contribute most to total AF management costs. Future work should focus on how to safely reduce avoidable hospital admissions.

• The between-patient range in costs was high, with a small number of patients at the high end of the cost distribution. Although we identified a number of significant variables predictive of high care costs, none of those analyzed accounted for much variability in the total cost of AF management. This suggests that it is often not possible to predict which patients will be high NHS resource users

• Maintenance phase costs were significantly higher for patients managed by practices providing anticoagulation services than patients for whom anticoagulation was managed in secondary care, even though the cost of primary care anticoagulation visits was excluded from this analysis. This may be due to the increased patient contact for anticoagulation, generating further healthcare professional activity, but this cannot be determined from the data collected.

Figure 1: Distribution of recorded AF type at date of data collection