Methods

This Markov cohort transition state model utilized a 1-year time horizon, a cycle length of 1-month, and was conducted from the societal perspective. It included 5 mutually exclusive permanent health states, 4 related to angina frequency (no, monthly, weekly, and daily angina symptoms) and the absorbing health state of death ( Figure 1 ). This model was built primarily using efficacy and tolerability data from the Efficacy of Ranolazine in Chronic Angina (ERICA) trial, which randomly assigned 565 patients with stable coronary disease experiencing ≥3 angina attacks/week (i.e., 5.6 ± 0.18 episodes/week and consuming 4.7 ± 0.21 nitroglycerin tablets/week) to receive ranolazine (500 mg twice daily for the first week followed by 1,000 mg twice daily thereafter) or placebo in addition to SoC antianginal therapy including a maximal dose of amlodipine (10 mg once daily) in all patients, 45% and 52% long-acting nitrate and angiotensin-converting enzyme inhibitor use, and no β-blocker use. Therefore, patients entering the model started in 1 of 3 of the 4 angina frequency health states (no patients started in the “no angina” state) based upon Seattle Angina Questionnaire angina frequency (SAQAF) domain scores calculated at randomization in the ERICA trial. For the purposes of this model, patients scoring 100 points on the SAQAF were deemed to have no angina symptoms, whereas scores of 61 to 99, 31 to 60, and 0 to 30 represented monthly, weekly, and daily angina symptoms, respectively. The SAQAF was utilized to define our model’s health states because it was a key patient-reported outcome measure utilized in the ERICA trial and has been used in the other angina clinical trials and previous stable angina epidemiologic and cost-of-illness analyses.

Schematic representation of the Markov model. The model was used to separately determine the total cost of treatment and QALYs accrued by the patients with stable angina receiving and not receiving ranolazine. Regardless of treatment assignment, patients entered the model in 1 of the 3 angina frequency health states based upon SAQAF scores (100 = no, 61 to 99 = monthly, 31 to 60 = weekly, and 0 to 30 = daily angina; no patients started in “no” angina) and were allowed to transition between states in the first month based upon treatment-specific probabilities derived from the ERICA trial and other studies. Patients not responding to ranolazine in month 1 (i.e., not improving ≥1 SAQAF health state) or experiencing an adverse event requiring discontinuation were assumed to stop taking ranolazine and behave like SoC (plus placebo) patients. Only patients assigned to receive ranolazine at the initiation of the model could discontinue therapy (for lack of efficacy or adverse drug events), and discontinuation could only occur during the first cycle. Patients randomized to SoC (plus placebo) started and had to remain “off drug.” In the second through twelfth month, all patients were assumed to stay in the same angina frequency health state for the remainder of the model’s time horizon or until death. Transition to death could occur during any cycle. M = Markov node.

The model allowed patients to transition between the 4 above-mentioned angina frequency health states and the death state, with transitions occurring only once per each 1-month cycle. A 1-month cycle length was chosen because ranolazine’s benefit typically occurs in the first few weeks in those whom respond, and 1 month correlates to 1 prescription dispensed. The model’s first month’s (cycle’s) transition probabilities for movement through the angina frequency health states were derived from the ERICA trial. For patients not receiving ranolazine, the probability of moving from 1 angina frequency health state to another was calculated based upon those observed in the SoC arm of the ERICA trial ( Table 1 ). Transition probabilities for ranolazine patients achieving adequate efficacy on-treatment, defined as improving by at least 1 angina frequency health state (e.g., transitioning from daily to weekly angina symptoms), were calculated based upon rates observed in corresponding ERICA patients ( Table 2 ). Patients receiving ranolazine could also discontinue treatment due to adverse drug reactions or lack of efficacy during, and only during, the first month of treatment. This assumption was based upon the reasoning that patients reporting a lack of efficacy or adverse reactions requiring discontinuation of therapy would most likely to do so in the first month. The rates of ranolazine discontinuation due to adverse reactions and lack of efficacy were derived from the ERICA trial ( Table 3 ). For those patients discontinuing ranolazine for any reason, transition probabilities were assumed to follow the same pattern as SoC (plus placebo) patients. In the second month (cycle 2) and onward, all patients were assumed to stay in the same angina frequency health state for the remainder of the model’s time horizon or until death. Thus, no loss or additional efficacy in either treatment group could occur.

Table 3

Other base-case variables and ranges used in sensitivity analysis

Variable	Base Case	Range	Reference
SAQAF classification definition
No	SAQAF = 100	NA	2, 9
Monthly	SAQAF = 61–99	NA	2, 9
Weekly	SAQAF = 31–60	NA	2, 9
Daily	SAQAF = 0–30	NA	2, 9
SAQAF classification at baseline			1, 9
No	0%	NA	1, 9
Monthly	6.1%	100%	1, 9
Weekly	71.0%	100%	1, 9
Daily	22.9%	100%	1, 9
Definition of SAQAF responder	Improvement of ≥1 SAQAF classification	20-point change in SAQAF	8, 9
Ranolazine nonresponse	48% during first 4 weeks	42.2%–53.9%	1
Ranolazine discontinuation due to adverse event	1.1% during first 4 weeks	0.37%–6%	1
All-cause mortality by angina frequency
No	4.6%/yr	3.8%–5.5%	8
Monthly	4.8%/yr	3.8%–6.1%	8
Weekly	8.1%/yr	6.1%–10.8%	8
Daily	10.9%/yr	7.5%–15.4%	8
All-cause mortality for all patients with angina	5.8%/yr	NA	8
Angina frequency utility (using EOT data)			1, 7
No	0.87	0.84–0.90	1, 7
Monthly	0.76	0.75–0.77	1, 7
Weekly	0.65	0.64–0.66	1, 7
Daily	0.54	0.52–0.56	1, 7
Cost of ranolazine 500 mg twice daily	$242/month	$124–$315	10, survey
Cost of ranolazine 1,000 mg twice daily	$397/month	$248–$525	10, survey
Stable angina direct treatment costs/year (not including ranolazine)			9
No	$5,235	$4,074–$4,710	9
Monthly	$6,989	$5,297–$6,253	9
Weekly	$8,149	$5,925–$7,749	9
Daily	$12,422	$8,407–$12,437	9
Stable angina indirect costs/year
No	$3,504	$1,500–5,004	6
Monthly	$5,952	$3,996–$8,004	6
Weekly	$6,336	$3,996–$8,004	6
Daily	$12,156	$8,004–$15,996	6

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. Usefulness of the SYNTAX Score to Predict Acute Kidney Injury After Percutaneous Coronary Intervention (from the Acute Catheterization and Urgent Intervention Triage Strategy Trial)
2. Impact of High-Dose Atorvastatin Therapy and Clinical Risk Factors on Incident Aortic Valve Stenosis in Patients With Cardiovascular Disease (from TNT, IDEAL, and SPARCL)
3. Left Ventricular Systolic Function Following Alcohol Septal Ablation for Symptomatic Hypertrophic Cardiomyopathy
4. Cardiotoxicity Research in Breast Cancer Patients: Past and Future

Stay updated, free articles. Join our Telegram channel

Join