User login
, according to recent research published in the journal Headache.
“[T]he burden and impact of migraine on the individual both during and between attacks were identified through supervised machine learning models to be strongly associated with seeking care,” Sait Ashina, MD, of the department of neurology at Harvard Medical School in Boston, and colleagues wrote in their study.
Dr. Ashina and colleagues performed a cross-sectional study of 61,826 patients from the web-based ObserVational survey of the Epidemiology, tReatment and Care Of MigrainE (OVERCOME) study with migraine who visited a primary care, specialty care, or urgent care, or emergency setting for headache between 2018 and 2020.
The patients recruited for OBSERVE were a mean of 41.7 years old and had experienced migraines for an average of 19.0 years; 59.4% had between 0 and 3 average headache days per month, 74.5% were women, 78.8% were White, and 85.4% had health insurance; and they were demographically representative of the US population.
Researchers used a machine learning model, which consisted of random forest and least absolute shrinkage and selection operator (LASSO) algorithms, to identify the relationship between patients who sought care for migraine and 54 different clinical, sociodemographic, and migraine-associated factors, which included age, years with migraine, symptom scores, pain intensity scores, disability score, comorbidities, vomiting, presence and severity of allodynia, and other factors.
The results showed 31,529 patients (51.0%) had an in-person or e-visit encounter with a primary care, specialty care, or urgent care, or emergency care location within 12 months of the survey, and were mostly White (76.5%) women (73.3%) with health insurance (88.9%). Of the patients who sought care, 52.8% had severe interictal burden measured by Migraine Interictal Burden Scale-4 score, compared with 23.1% of patients who did not seek care. Compared with patients who did not seek care, those who did visit a health care setting for migraine had a higher percentage of severe migraine-related disability as measured by the Migraine Disability Assessment Scale (36.7% vs 14.6%) and severe ictal cutaneous allodynia as measured by the Allodynia Symptom Checklist (21.0% vs 7.4%).
In a multivariable logistic regression model analysis, Dr. Ashina and colleagues said the factors most associated with seeking care included severe interictal burden (odds ratio [OR], 2.64; 95% confidence interval [CI], 2.5-2.8), severe migraine-related disability (OR, 2.2; 95% CI, 2.0-2.3), and severe ictal allodynia (OR, 1.7; 95% CI, 1.6-1.8), compared with less severe factors.
The researchers said their results have “significant implications for public health and advocacy efforts.”
“As seen through three decades of epidemiological research in the United States, rates of care-seeking have not improved dramatically over time despite significant additions to scientific knowledge and the therapeutic armamentarium, leaving a significant unmet need. This is also important from a clinical perspective,” they explained. “Health care professionals in primary care and internal medicine most likely see patients with migraine who do not discuss it during visits. This underscores the importance of maintaining vigilance for migraine, especially among those who may experience greater disability, impact, and interictal burden.”
Asking the Right Questions
Asked to comment on the research, Robert P. Cowan, MD, a neurologist and professor in the Stanford University School of Medicine department of neurology and neurological sciences in Palo Alto, California, said in an interview that the value of the paper is in what it does not say about the main reasons patients seek care.
“Most clinicians readily acknowledge that the average number of migraine headache days per month is, at best, a weak predictor of which patients seek care and when,” he said.
Dr. Cowan said that most patients are referred to him by other providers, and when he asks them why they did not seek care for migraine sooner, the answer is usually because the migraine was not severe enough or because over-the-counter medication had previously worked for them. He noted that change in frequency is, in his experience, a primary reason why patients will seek care. “[F]or new (or increasing) headache, it is the concern that the headaches are something more ‘serious,’ and once that is ruled out, the conversation often stops,” he said. “For long-standing migraine sufferers, it is the perception that the headache is a ‘fact of life’ and does not rise to the bar of seeking medical advice.”
The questions a survey or a provider asks matters, Dr. Cowan said. “Often, when we ask a patient how many headache (or migraine) days per month, the answer is in single digits. But if we follow-up with a question about the number of headache-free days [per] month, the answer is ‘never’ or ‘hardly ever,’” he explained. “The point here is that what questions a survey (or a provider) asks introduces a clear bias. The use of machine learning instruments, especially when utilizing supervised learning, only reinforces and amplifies the bias of the designers of the categories.”
Epidemiologic studies are interesting but “often ask the wrong questions,” Dr. Cowan said. “I am less worried about the ... 49% of migraine or possible migraine patients who do not seek care and do [not] progress to more disabling ‘chronic’ migraine than I am with identifying the subpopulations of migraine patients who seek care from providers who do not have adequate tools to match patients to the best treatments.”
The authors reported personal and institutional relationships in the form of advisory board memberships, consultancies, employment, honoraria, research support, speakers bureau positions, stock ownership, and teaching services with AbbVie, Aeon, Alder, Allay Lamp, Allergan, Amgen, Axon, Biohaven Pharmaceuticals, Collegium, CoolTech, Currax, Dr. Reddy’s Laboratories (Promius), electroCore, GlaxoSmithKline, Impel NeuroPharma, Informa, Eli Lilly and Company, Lundbeck, Mainistee, Merck, National Headache Foundation, National Institutes of Health, Novartis, Pfizer, Satsuma, Supernus, Percept, Teva, Theranica, UpsherSmith, the US Food and Drug Administration, Vector, Vedanta Research, and Wolff’s Headache. The study was supported by Eli Lilly. Dr. Cowan reports no relevant conflicts of interest.
, according to recent research published in the journal Headache.
“[T]he burden and impact of migraine on the individual both during and between attacks were identified through supervised machine learning models to be strongly associated with seeking care,” Sait Ashina, MD, of the department of neurology at Harvard Medical School in Boston, and colleagues wrote in their study.
Dr. Ashina and colleagues performed a cross-sectional study of 61,826 patients from the web-based ObserVational survey of the Epidemiology, tReatment and Care Of MigrainE (OVERCOME) study with migraine who visited a primary care, specialty care, or urgent care, or emergency setting for headache between 2018 and 2020.
The patients recruited for OBSERVE were a mean of 41.7 years old and had experienced migraines for an average of 19.0 years; 59.4% had between 0 and 3 average headache days per month, 74.5% were women, 78.8% were White, and 85.4% had health insurance; and they were demographically representative of the US population.
Researchers used a machine learning model, which consisted of random forest and least absolute shrinkage and selection operator (LASSO) algorithms, to identify the relationship between patients who sought care for migraine and 54 different clinical, sociodemographic, and migraine-associated factors, which included age, years with migraine, symptom scores, pain intensity scores, disability score, comorbidities, vomiting, presence and severity of allodynia, and other factors.
The results showed 31,529 patients (51.0%) had an in-person or e-visit encounter with a primary care, specialty care, or urgent care, or emergency care location within 12 months of the survey, and were mostly White (76.5%) women (73.3%) with health insurance (88.9%). Of the patients who sought care, 52.8% had severe interictal burden measured by Migraine Interictal Burden Scale-4 score, compared with 23.1% of patients who did not seek care. Compared with patients who did not seek care, those who did visit a health care setting for migraine had a higher percentage of severe migraine-related disability as measured by the Migraine Disability Assessment Scale (36.7% vs 14.6%) and severe ictal cutaneous allodynia as measured by the Allodynia Symptom Checklist (21.0% vs 7.4%).
In a multivariable logistic regression model analysis, Dr. Ashina and colleagues said the factors most associated with seeking care included severe interictal burden (odds ratio [OR], 2.64; 95% confidence interval [CI], 2.5-2.8), severe migraine-related disability (OR, 2.2; 95% CI, 2.0-2.3), and severe ictal allodynia (OR, 1.7; 95% CI, 1.6-1.8), compared with less severe factors.
The researchers said their results have “significant implications for public health and advocacy efforts.”
“As seen through three decades of epidemiological research in the United States, rates of care-seeking have not improved dramatically over time despite significant additions to scientific knowledge and the therapeutic armamentarium, leaving a significant unmet need. This is also important from a clinical perspective,” they explained. “Health care professionals in primary care and internal medicine most likely see patients with migraine who do not discuss it during visits. This underscores the importance of maintaining vigilance for migraine, especially among those who may experience greater disability, impact, and interictal burden.”
Asking the Right Questions
Asked to comment on the research, Robert P. Cowan, MD, a neurologist and professor in the Stanford University School of Medicine department of neurology and neurological sciences in Palo Alto, California, said in an interview that the value of the paper is in what it does not say about the main reasons patients seek care.
“Most clinicians readily acknowledge that the average number of migraine headache days per month is, at best, a weak predictor of which patients seek care and when,” he said.
Dr. Cowan said that most patients are referred to him by other providers, and when he asks them why they did not seek care for migraine sooner, the answer is usually because the migraine was not severe enough or because over-the-counter medication had previously worked for them. He noted that change in frequency is, in his experience, a primary reason why patients will seek care. “[F]or new (or increasing) headache, it is the concern that the headaches are something more ‘serious,’ and once that is ruled out, the conversation often stops,” he said. “For long-standing migraine sufferers, it is the perception that the headache is a ‘fact of life’ and does not rise to the bar of seeking medical advice.”
The questions a survey or a provider asks matters, Dr. Cowan said. “Often, when we ask a patient how many headache (or migraine) days per month, the answer is in single digits. But if we follow-up with a question about the number of headache-free days [per] month, the answer is ‘never’ or ‘hardly ever,’” he explained. “The point here is that what questions a survey (or a provider) asks introduces a clear bias. The use of machine learning instruments, especially when utilizing supervised learning, only reinforces and amplifies the bias of the designers of the categories.”
Epidemiologic studies are interesting but “often ask the wrong questions,” Dr. Cowan said. “I am less worried about the ... 49% of migraine or possible migraine patients who do not seek care and do [not] progress to more disabling ‘chronic’ migraine than I am with identifying the subpopulations of migraine patients who seek care from providers who do not have adequate tools to match patients to the best treatments.”
The authors reported personal and institutional relationships in the form of advisory board memberships, consultancies, employment, honoraria, research support, speakers bureau positions, stock ownership, and teaching services with AbbVie, Aeon, Alder, Allay Lamp, Allergan, Amgen, Axon, Biohaven Pharmaceuticals, Collegium, CoolTech, Currax, Dr. Reddy’s Laboratories (Promius), electroCore, GlaxoSmithKline, Impel NeuroPharma, Informa, Eli Lilly and Company, Lundbeck, Mainistee, Merck, National Headache Foundation, National Institutes of Health, Novartis, Pfizer, Satsuma, Supernus, Percept, Teva, Theranica, UpsherSmith, the US Food and Drug Administration, Vector, Vedanta Research, and Wolff’s Headache. The study was supported by Eli Lilly. Dr. Cowan reports no relevant conflicts of interest.
, according to recent research published in the journal Headache.
“[T]he burden and impact of migraine on the individual both during and between attacks were identified through supervised machine learning models to be strongly associated with seeking care,” Sait Ashina, MD, of the department of neurology at Harvard Medical School in Boston, and colleagues wrote in their study.
Dr. Ashina and colleagues performed a cross-sectional study of 61,826 patients from the web-based ObserVational survey of the Epidemiology, tReatment and Care Of MigrainE (OVERCOME) study with migraine who visited a primary care, specialty care, or urgent care, or emergency setting for headache between 2018 and 2020.
The patients recruited for OBSERVE were a mean of 41.7 years old and had experienced migraines for an average of 19.0 years; 59.4% had between 0 and 3 average headache days per month, 74.5% were women, 78.8% were White, and 85.4% had health insurance; and they were demographically representative of the US population.
Researchers used a machine learning model, which consisted of random forest and least absolute shrinkage and selection operator (LASSO) algorithms, to identify the relationship between patients who sought care for migraine and 54 different clinical, sociodemographic, and migraine-associated factors, which included age, years with migraine, symptom scores, pain intensity scores, disability score, comorbidities, vomiting, presence and severity of allodynia, and other factors.
The results showed 31,529 patients (51.0%) had an in-person or e-visit encounter with a primary care, specialty care, or urgent care, or emergency care location within 12 months of the survey, and were mostly White (76.5%) women (73.3%) with health insurance (88.9%). Of the patients who sought care, 52.8% had severe interictal burden measured by Migraine Interictal Burden Scale-4 score, compared with 23.1% of patients who did not seek care. Compared with patients who did not seek care, those who did visit a health care setting for migraine had a higher percentage of severe migraine-related disability as measured by the Migraine Disability Assessment Scale (36.7% vs 14.6%) and severe ictal cutaneous allodynia as measured by the Allodynia Symptom Checklist (21.0% vs 7.4%).
In a multivariable logistic regression model analysis, Dr. Ashina and colleagues said the factors most associated with seeking care included severe interictal burden (odds ratio [OR], 2.64; 95% confidence interval [CI], 2.5-2.8), severe migraine-related disability (OR, 2.2; 95% CI, 2.0-2.3), and severe ictal allodynia (OR, 1.7; 95% CI, 1.6-1.8), compared with less severe factors.
The researchers said their results have “significant implications for public health and advocacy efforts.”
“As seen through three decades of epidemiological research in the United States, rates of care-seeking have not improved dramatically over time despite significant additions to scientific knowledge and the therapeutic armamentarium, leaving a significant unmet need. This is also important from a clinical perspective,” they explained. “Health care professionals in primary care and internal medicine most likely see patients with migraine who do not discuss it during visits. This underscores the importance of maintaining vigilance for migraine, especially among those who may experience greater disability, impact, and interictal burden.”
Asking the Right Questions
Asked to comment on the research, Robert P. Cowan, MD, a neurologist and professor in the Stanford University School of Medicine department of neurology and neurological sciences in Palo Alto, California, said in an interview that the value of the paper is in what it does not say about the main reasons patients seek care.
“Most clinicians readily acknowledge that the average number of migraine headache days per month is, at best, a weak predictor of which patients seek care and when,” he said.
Dr. Cowan said that most patients are referred to him by other providers, and when he asks them why they did not seek care for migraine sooner, the answer is usually because the migraine was not severe enough or because over-the-counter medication had previously worked for them. He noted that change in frequency is, in his experience, a primary reason why patients will seek care. “[F]or new (or increasing) headache, it is the concern that the headaches are something more ‘serious,’ and once that is ruled out, the conversation often stops,” he said. “For long-standing migraine sufferers, it is the perception that the headache is a ‘fact of life’ and does not rise to the bar of seeking medical advice.”
The questions a survey or a provider asks matters, Dr. Cowan said. “Often, when we ask a patient how many headache (or migraine) days per month, the answer is in single digits. But if we follow-up with a question about the number of headache-free days [per] month, the answer is ‘never’ or ‘hardly ever,’” he explained. “The point here is that what questions a survey (or a provider) asks introduces a clear bias. The use of machine learning instruments, especially when utilizing supervised learning, only reinforces and amplifies the bias of the designers of the categories.”
Epidemiologic studies are interesting but “often ask the wrong questions,” Dr. Cowan said. “I am less worried about the ... 49% of migraine or possible migraine patients who do not seek care and do [not] progress to more disabling ‘chronic’ migraine than I am with identifying the subpopulations of migraine patients who seek care from providers who do not have adequate tools to match patients to the best treatments.”
The authors reported personal and institutional relationships in the form of advisory board memberships, consultancies, employment, honoraria, research support, speakers bureau positions, stock ownership, and teaching services with AbbVie, Aeon, Alder, Allay Lamp, Allergan, Amgen, Axon, Biohaven Pharmaceuticals, Collegium, CoolTech, Currax, Dr. Reddy’s Laboratories (Promius), electroCore, GlaxoSmithKline, Impel NeuroPharma, Informa, Eli Lilly and Company, Lundbeck, Mainistee, Merck, National Headache Foundation, National Institutes of Health, Novartis, Pfizer, Satsuma, Supernus, Percept, Teva, Theranica, UpsherSmith, the US Food and Drug Administration, Vector, Vedanta Research, and Wolff’s Headache. The study was supported by Eli Lilly. Dr. Cowan reports no relevant conflicts of interest.
FROM HEADACHE