Treatment Persistence and Switching in Triptan Users
According to the gathered data, between 25–60% of patients never refill their index triptan. The main reasons for this lack of refill appear to be lack of efficacy and adverse events. These data are in line with that of studies on within-patient consistency of response (three out of three attacks) to different orally administered triptans, in which an average of less than half achieve a pain-free response in three out of three treated migraine attacks. This leads to the question of how overall efficacy as measured by within-patient consistency of response (three out of three attacks) could be increased, resulting in a more ideal clinical outcome for the patient and best value for money spent. Variability in attack characteristics may place a ceiling on within-patient consistency, but approaches to consider to improve efficacy include taking a more complete history in terms of previous treatment successes and failures and responder rates, optimization of dose by the dose response curve using an appropriate formulation given the clinical setting, early treatment during an attack, possible combination with synergistic antimigraine medications, and consideration of cost.
Efficacy may be inadequately evaluated and considered by care provider and patient alike. Several attacks are often needed to be sure a patient is a non-responder, so that both clinicians and patients may not be doing adequate trials before switching or abandoning a given triptan. Education in regard to optimal dose and formulation, early intervention, and several therapeutic trials may help in improving persistence and outcomes. It is also the responsibility of the care provider to consider cost and payer for each patient before prescribing.
And the tougher the clinical situation, the more that a lack of persistence may manifest. Patients with medication overuse headache, in which acute treatments may be less effective, may abandon acute treatments faster than those with episodic migraine or those more naïve to initial treatment.
The degree of agreement was high among studies measuring a given outcome on a similar time scale, despite the diversity of study designs and populations. Three studies from three different countries reported an annual rate of discontinuation among a general population of triptan users between 30% and 60%. Eight studies using data from several different countries all reported a rate of one-time triptan filling between 40% and 65%. In addition to all of the patients who discontinue before ever refilling their triptan, two studies reported the rate of immediate switching (switching to a new triptan before refilling the originally prescribed triptan) to be between 4% and 7%.
One interesting issue was that the agreement was lacking, concerning the relative frequency of within-class switching, between-class switching, and discontinuation of all migraine pharmacotherapy. Katić et al and Ng-Mak et al were the only studies that reported all three rates after index triptan fill. Katić et al found that switching to a new medication class was the most common option: −4% of patients switched to a different triptan, 25.5% of patients gave up migraine medications entirely, and 36.1% of patients switched to another class of medications. However, Ng-Mak et al found that switching to a new class was the least common option and complete discontinuation of migraine medications was significantly more common: −2.3% to 4% switched to a new medication class, 4.9–6.8% switched to a new triptan, and 48.5–54.9% discontinued all migraine pharmacotherapy. In both cases, switching to a new triptan was relatively uncommon.
Although not the primary focus of this review, differences between persistent and nonpersistent patients, as well as the most common reasons for nonpersistency, are important to note. One recent study from the US found that sustained users (defined as those patients who filled their triptan at least once per year) were significantly more satisfied with the medication and more confident in the medication's ability to control headache, compared with lapsed users (those patients who went 1 year or more without filling). Sustained users also reported control of migraine with fewer doses of their medication and greater benefit from triptans (able to restore normal daily function), compared with lapsed users. Other studies have asked patients who switched between triptans to report their reasons for switching. Results were varied, but "incomplete or no relief" and "side effects" were the most commonly reported causes of switching. One study demonstrated that taking sumatriptan 50 mg vs 100 mg was associated with a higher likelihood of one-time filling, suggesting links between incorrect dosing, poor clinical response, and lack of persistence. Taken together, these studies indicate that problematic persistence to triptan therapy is significant because it likely represents unmet therapeutic needs.
The results of this review are concerning given the fact that nonpersistence with triptan therapy and switching to other acute medications (opioids, barbituates, combination analgesics, NSAIDs) has been associated with increased primary care costs and risk of progression to chronic migraine. Chronic migraine is associated with worse quality of life, decreased productivity, and increased health care costs compared with episodic migraine. Further research is needed to better understand all of the consequences of nonpersistency and switching, as well as the best methods of communicating this risk to patients.
This literature review is subject to several limitations. Among the 12 selected studies, half of the studies were (partly) supported by pharmaceutical companies, which could hypothetically result in a bias to present more favorable persistence to triptans. However, given the results of this review, which indicate a lack of persistence to triptan therapy, this bias does not appear to be significant. We sought to focus on real-world adherence/persistence by limiting included studies to observational designs, rather than randomized trials. However, this may result in publication bias if reports of adequate persistence are less likely to be published than studies reporting more worrisome results. The data reviewed span 15 years during which the availability of different triptans has increased dramatically. The results of earlier studies when fewer alternatives existed may not be directly comparable with more recent research. This review was conducted from an international perspective, and drawing direct comparisons between countries has the added complication of differing health care systems, costs, prescription coverage, national formularies (where applied), and practice patterns. The extent to which differences in health care systems are reflected in adherence patterns is outside the scope of this review. However, even in countries where triptans are fully reimbursable to the patient (eg, the Netherlands), 38% were found to have filled only one prescription.
The majority of studies included in this analysis were based on claims data, which have limitations that can make it difficult to determine whether a patient was actually persistent. Although a prescription was filled, it is unknown whether and when a patient takes it, and if it is taken appropriately (eg, with timing of administration). However, similarities in outcomes suggest similarities of patient behavior internationally. Finally, the assessment of persistence using claims data is generally unable to include medication fills paid by the patient without insurance coverage, but we do not expect this limitation to significantly affect the persistence results.
Discussion
According to the gathered data, between 25–60% of patients never refill their index triptan. The main reasons for this lack of refill appear to be lack of efficacy and adverse events. These data are in line with that of studies on within-patient consistency of response (three out of three attacks) to different orally administered triptans, in which an average of less than half achieve a pain-free response in three out of three treated migraine attacks. This leads to the question of how overall efficacy as measured by within-patient consistency of response (three out of three attacks) could be increased, resulting in a more ideal clinical outcome for the patient and best value for money spent. Variability in attack characteristics may place a ceiling on within-patient consistency, but approaches to consider to improve efficacy include taking a more complete history in terms of previous treatment successes and failures and responder rates, optimization of dose by the dose response curve using an appropriate formulation given the clinical setting, early treatment during an attack, possible combination with synergistic antimigraine medications, and consideration of cost.
Efficacy may be inadequately evaluated and considered by care provider and patient alike. Several attacks are often needed to be sure a patient is a non-responder, so that both clinicians and patients may not be doing adequate trials before switching or abandoning a given triptan. Education in regard to optimal dose and formulation, early intervention, and several therapeutic trials may help in improving persistence and outcomes. It is also the responsibility of the care provider to consider cost and payer for each patient before prescribing.
And the tougher the clinical situation, the more that a lack of persistence may manifest. Patients with medication overuse headache, in which acute treatments may be less effective, may abandon acute treatments faster than those with episodic migraine or those more naïve to initial treatment.
The degree of agreement was high among studies measuring a given outcome on a similar time scale, despite the diversity of study designs and populations. Three studies from three different countries reported an annual rate of discontinuation among a general population of triptan users between 30% and 60%. Eight studies using data from several different countries all reported a rate of one-time triptan filling between 40% and 65%. In addition to all of the patients who discontinue before ever refilling their triptan, two studies reported the rate of immediate switching (switching to a new triptan before refilling the originally prescribed triptan) to be between 4% and 7%.
One interesting issue was that the agreement was lacking, concerning the relative frequency of within-class switching, between-class switching, and discontinuation of all migraine pharmacotherapy. Katić et al and Ng-Mak et al were the only studies that reported all three rates after index triptan fill. Katić et al found that switching to a new medication class was the most common option: −4% of patients switched to a different triptan, 25.5% of patients gave up migraine medications entirely, and 36.1% of patients switched to another class of medications. However, Ng-Mak et al found that switching to a new class was the least common option and complete discontinuation of migraine medications was significantly more common: −2.3% to 4% switched to a new medication class, 4.9–6.8% switched to a new triptan, and 48.5–54.9% discontinued all migraine pharmacotherapy. In both cases, switching to a new triptan was relatively uncommon.
Although not the primary focus of this review, differences between persistent and nonpersistent patients, as well as the most common reasons for nonpersistency, are important to note. One recent study from the US found that sustained users (defined as those patients who filled their triptan at least once per year) were significantly more satisfied with the medication and more confident in the medication's ability to control headache, compared with lapsed users (those patients who went 1 year or more without filling). Sustained users also reported control of migraine with fewer doses of their medication and greater benefit from triptans (able to restore normal daily function), compared with lapsed users. Other studies have asked patients who switched between triptans to report their reasons for switching. Results were varied, but "incomplete or no relief" and "side effects" were the most commonly reported causes of switching. One study demonstrated that taking sumatriptan 50 mg vs 100 mg was associated with a higher likelihood of one-time filling, suggesting links between incorrect dosing, poor clinical response, and lack of persistence. Taken together, these studies indicate that problematic persistence to triptan therapy is significant because it likely represents unmet therapeutic needs.
The results of this review are concerning given the fact that nonpersistence with triptan therapy and switching to other acute medications (opioids, barbituates, combination analgesics, NSAIDs) has been associated with increased primary care costs and risk of progression to chronic migraine. Chronic migraine is associated with worse quality of life, decreased productivity, and increased health care costs compared with episodic migraine. Further research is needed to better understand all of the consequences of nonpersistency and switching, as well as the best methods of communicating this risk to patients.
Limitations
This literature review is subject to several limitations. Among the 12 selected studies, half of the studies were (partly) supported by pharmaceutical companies, which could hypothetically result in a bias to present more favorable persistence to triptans. However, given the results of this review, which indicate a lack of persistence to triptan therapy, this bias does not appear to be significant. We sought to focus on real-world adherence/persistence by limiting included studies to observational designs, rather than randomized trials. However, this may result in publication bias if reports of adequate persistence are less likely to be published than studies reporting more worrisome results. The data reviewed span 15 years during which the availability of different triptans has increased dramatically. The results of earlier studies when fewer alternatives existed may not be directly comparable with more recent research. This review was conducted from an international perspective, and drawing direct comparisons between countries has the added complication of differing health care systems, costs, prescription coverage, national formularies (where applied), and practice patterns. The extent to which differences in health care systems are reflected in adherence patterns is outside the scope of this review. However, even in countries where triptans are fully reimbursable to the patient (eg, the Netherlands), 38% were found to have filled only one prescription.
The majority of studies included in this analysis were based on claims data, which have limitations that can make it difficult to determine whether a patient was actually persistent. Although a prescription was filled, it is unknown whether and when a patient takes it, and if it is taken appropriately (eg, with timing of administration). However, similarities in outcomes suggest similarities of patient behavior internationally. Finally, the assessment of persistence using claims data is generally unable to include medication fills paid by the patient without insurance coverage, but we do not expect this limitation to significantly affect the persistence results.
SHARE
