Methods

A decision tree was constructed ( Figure 1 ) to ensure that all permutations of outcome probabilities, complication probabilities, procedural and medication probabilities, and medication complication probabilities were accounted for in the cost calculations.

Decision tree modeling the presented clinical scenarios. The tree shows the first open square node. This node represents a decision to be made among different management options. All subsequent nodes are probability nodes ( empty circle ), describing the possible outcomes after the initial decision. Markov branches for all the iterations not depicted are identical to the PFO closure iteration shown. These iterations are represented by the explode nodes ( squares containing crosses ). Each possible cost for the management strategy used, complications encountered, medical therapy received, medical therapy associated complications encountered, and outcome encountered were calculated. ASA = aspirin; OAC = oral anticoagulant.

End point and complication costs were calculated assuming an even distribution of event risk across all patients, and on the basis of the previously reported end point event rates, pooled weighted probabilities for stroke, TIA, and death were annualized. Each annualized end point event probability ( Table 1 ) was multiplied by the cost of that particular outcome to determine the annual end point event cost per patient. The annualized cost of complications (atrial fibrillation, major bleeding, and vascular complications) per patient was also estimated in a similar manner. Vascular complications were assumed to occur only in the immediate and reported follow-up period. Assuming an even distribution of complication risk across all patients, and on the basis of the previously reported complication event rates, pooled and weighted probabilities for atrial fibrillation, major bleeding, and vascular complications were annualized. Each annualized complication event probability ( Table 1 ) was multiplied by the cost of that particular complication to determine the annual complication cost per patient.

Procedural costs were calculated assuming the Medicare cost for diagnosis-related group 251, for “percutaneous cardiac procedure without PCI,” for fiscal year 2011 in a patient aged 45 to 64 years, given that the average patient age in the meta-analysis of the RCTs was 45.7 years.

Medication costs were calculated on the basis of the RCT-reported rates of specific medication regimens and assuming an even distribution of medication probability across all patients. The pooled and weighted daily probabilities were calculated for a patient to receive a specific medication regimen for 1 to 30, 31 to 90, 91 to 180, and >180 days. These probabilities were multiplied by the number of days they were prescribed, the daily medication cost, and the associated costs for that medication regimen. The annual medication cost was determined from the sum of these medication costs over the length of follow-up.

Medication-associated major bleeding complications were assumed to begin after the conclusion of study reported follow-up, because each study reported the complications that occurred during the follow-up period. Individual medication bleeding costs were determined as the sum of the products of each medication regimen: the probability of bleeding multiplied by the probability of a patient receiving the specific medication regimen multiplied by the cost of major bleeding. The annual probability of major bleeding on the basis of the medication regimen was determined from previously published rates of major bleeding for the each individual medication regimen.

Total costs were estimated by summing the procedural costs, end point event costs, complication costs, medication costs, and medication-associated major bleeding costs over the duration of follow-up.

Next, an incremental cost per life-year analysis for PFO closure was performed using the average patient in the RCTs : a 45.7-year-old man. The presumed life expectancy for this 46-year-old man was determined from 2009 Social Security tables ( Table 2 ). The presumed life expectancy was used to calculate life-years lost because of death. The life-years lost because of stroke and TIA were determined by multiplying the annual probability of death after TIA or stroke by the presumed life expectancy ( Table 2 ). The annual incremental risk for death, TIA, and stroke was calculated between PFO closure and medical therapy and multiplied by each respective life-year lost and then summed to determine the annual life-years gained by a patient who underwent PFO closure. To determine the cost per life-year gained, the incremental cost increase per PFO closure for the length of follow-up was divided by the life-years gained by closure for the length of follow-up. To determine the cost per quality-adjusted life-year (QALY) gained, the incremental cost increase per PFO closure for the length of follow-up was divided by the QALYs gained by closure for the length of follow-up. QALYs gained were calculated from previously published QALY adjustments: 0.87 for TIA and 0.52 for stroke. A willingness-to-pay threshold of <$50,000/QALY gained was used to define incremental cost-effectiveness.

The Amplatzer device (St. Jude Medical, St. Paul, Minnesota) has been shown to result in a statistically significant reduction in stroke. Therefore, a sensitivity analysis using the same modeling was also completed for only those RCTs in which the Amplatzer device was implanted.

Results

The pooled length of study follow-up was 2.6 years. When modeled to study conclusion (2.6 years), the per patient costs associated with PFO closure were more than for medical therapy alone, primarily as a result of PFO closure procedural costs ( Table 3 ). The expenditure to prevent 1 combined end point (TIA, stroke, or death) at 2.6 years was in excess of $1.0 million. At study conclusion, the incremental cost per life-year gained was slightly greater than $100,000.

At study conclusion, medicine costs and outcome costs were greater in the medical therapy group. However, complication costs and procedural costs were greater in the PFO closure group ( Table 4 ). The complication costs were higher in the PFO closure group for atrial fibrillation, vascular complications, and major bleeding. When costs were extrapolated to 5 years, the medication cost remained higher in the medical group and the difference in outcomes costs for medical therapy group compared with closure became larger. At 5 years, the major bleeding cost associated with PFO closure was smaller than that for medical therapy; however, the cost from atrial fibrillation remained higher in the PFO closure group ( Table 4 ).

Table 4

Annualized costs per patient and modeled costs per patient ∗

	PFO Closure	Medical Therapy
PFO Closure Procedure Cost †	$17,403.00	$0.00
Outcomes
Death	$69.13	$112.43
TIA	$53.45	$62.45
Stroke	$181.96	$259.43
Annual Sum	$304.54	$434.31
Complications
Atrial Fibrillation	$115.60	$23.72
Major Bleeding	$40.67	$21.13
Major vascular complication	$27.04	$0.00
Annual Sum	$183.31	$44.85
Medical therapy
Mean Weighted Cost (1st 30 days)	$178.51	$48.32
Mean Weighted Cost (30–180 days)	$440.03	$241.61
Mean Weighted Cost (after 180 days)	$33.37	$719.11
Annual Cost (1st 365 days)	$651.90	$1009.05
Annual Cost (after 1st 365 days)	$26.03	$560.85
Follow-up (years)	2.6	2.6
Costs at study completion (2.6 years)
Procedural cost	$17,403.00	$0.00
Outcomes	$791.81	$1129.21
Complications	$476.61	$116.62
Medical regimen	$693.55	$1906.41
Total	$19,364.97	$3152.23
Annual cost of complications after follow-up (after 2.6 years)
Atrial Fibrillation	$115.60	$23.72
Major Bleeding	$47.97	$62.68
Major vascular complication	$0.00	$0.00
Annual Sum	$163.56	$86.40
Modeling to 5.0 years (after 2.6 years)
Procedural cost	$0.00	$0.00
Outcomes	$730.90	$1042.35
Complications	$392.55	$207.36
Medical regimen	$62.46	$1346.03
Total (costs after 2.6 years)	$1185.91	$2595.74
Costs at study completion (up to 2.6 years)	$19,364.97	$3152.23
Total cost modeled to 5.0 years	$20,550.88	$5747.97

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Coronary Atheroma Regression and Plaque Characteristics Assessed by Grayscale and Radiofrequency Intravascular Ultrasound After Aerobic Exercise
2. Analysis of Early Out-of Hospital Mortality After Transcatheter Aortic Valve Implantation Among Patients With Aortic Stenosis Successfully Discharged from the Hospital and Alive at 30 Days (from the Placement of Aortic Transcatheter Valves Trial)
3. Aldosterone-Cortisol Imbalance Immediately After Fontan Operation With Implications for Abnormal Fluid Homeostasis
4. Gluten Sensitivity: New Epidemic or New Myth? Every Major Change in Our Diet Carries With It the Possibility of Unforeseen Risks

Stay updated, free articles. Join our Telegram channel

Join