ISSN 0017-8748

doi: 10.1111/head.12450
Published by Wiley Periodicals, Inc.

Headache
2014 American Headache Society

Research Submission
Persistent Frequent Nausea Is Associated With Progression to
Chronic Migraine: AMPP Study Results
Michael L. Reed, PhD; Kristina M. Fanning, PhD; Daniel Serrano, PhD; Dawn C. Buse, PhD;
Richard B. Lipton, MD

Background.Though nausea is a cardinal feature of migraine, its influence on migraine progression has not been
evaluated. This article aims to evaluate persistent frequent headache-related nausea (PFN) in persons with episodic migraine
(EM) as a predictor of new onset chronic migraine (CM).
Methods.This prospective cohort study uses data from the 2007 and 2008 American Migraine Prevalence and Prevention
study surveys to identify subgroups with episodic International Classification of Headache Disorders, 2nd edition defined
migraine and either PFN or no or low frequency nausea (NLFN). PFN was defined by the presence of nausea half the time
in both 2007 and 2008. NLFN was defined by nausea that was present < half the time, rarely or never in both years. Persons were
considered CM in 2009 if they met symptom criteria for migraine with headaches 15 days per month over the preceding 3
months. Univariate differences in demographics for PFN and NLFN were evaluated with chi-square. Binary logistic regressions
were performed hierarchically to assess progression to CM in 2009 as a function of nausea status in 2007 and 2008. The initial
model included sociodemographic variables only. Subsequent models added the following variables in a hierarchical manner:
migraine symptom severity composite score (to control for the impact of other headache features), headache-related disability,
depression, opioid use, and an interaction term for nausea status and opioid use. Odds ratios (OR) and 95% confidence intervals
(CI) contrasted PFN and NLFN on the rate of progression to CM in 2009.
Results.There were 3182 respondents with headache symptom and frequency data available for all 3 years of the analysis.
PFN was found in 43.7% (1389) of respondents, and 3.4% (47) progressed to CM. NLFN was seen in 27.6% (877) of the EM
group, and 1.5% progressed to CM. In comparison with the NLFN group, PFN was more common in females (P < .001) and
Caucasians (P < .06). PFN was associated with a doubling of the risk of progression to CM after adjusting for sociodemographic
variables (OR 2.09, 95% CI 1.11-3.91, P = .022). Adding the symptom composite score and headache-related disability
covariates to the model attenuated the association slightly (OR 2.00, 95% CI 1.03-3.87, P = .04). With the addition of depression,
the association fell just below statistical significance but progression risk with PFN remained at nearly two-fold that of the
NLFN group (OR 1.90, 95% CI 0.98-3.71, P = .059). When opioid use and the nausea by opioid use interaction are added to the
final model, the significant effect and the doubling of CM progression risk for those with PFN was retained (OR 2.24, 95% CI
1.07-4.70, P = .033).
Conclusions.Persistent frequent nausea is common (43.7%) among persons with episodic migraine. After controlling for
sociodemographics, migraine symptom severity, headache-related disability, depression, and opioid medication use, migraineurs
with frequent nausea that persisted for 2 years of study were twice as likely to progress to CM compared to those with no or low
frequency nausea. The study is limited by self-reports of symptom and headache frequency data and the use of modified
diagnostic criteria. Additional prospective research is needed to confirm study findings. Persistent frequent nausea could be a
marker for the risk of progression to CM or it could be in the causal pathway.
Key words: migraine, chronic migraine, nausea, severity, impact, progression

From Vedanta Research, Chapel Hill, NC, USA (M.L. Reed, K.M. Fanning, and D. Serrano); Montefiore Medical Center, Albert
Einstein College of Medicine, Bronx, NY, USA (D.C. Buse and R.B. Lipton).
Address all correspondence to D.C. Buse, Montefiore Medical Center, Albert Einstein College of Medicine, 1575 Blondell Avenue,
Suite #225, Bronx, NY 10461, USA, email: dbuse@montefiore.org
Accepted for publication August 05, 2014.

76

Headache

77

Abbreviations: AMPP American Migraine Prevalence and Prevention, CI, confidence interval, CM chronic migraine,
EM episodic migraine, PFN persistent frequent nausea, NLFN no or low frequency nausea, OR odds ratio,
ICHD-2 International Classification of Headache Disorders, 2nd edition, MIDAS Migraine Disability Assessment Scale, PHQ-9 Patient Health Questionnaire, 9-item depression module, PFN persistent frequent nausea,
PRIME-MD Primary Care Evaluation of Mental Disorders, TNS Taylor Nelson Sofres
(Headache 2015;55:76-87)

Symptoms other than headache pain can be significant factors in the disability associated with
migraine.1 Nausea is a common migraine-related
symptom, reported by 70% to 90% of persons with
migraine; it is associated with impaired functioning
and decreased quality of life in persons with frequent
migraine attacks.2,3 In addition, frequent nausea is
both a challenge in treatment and a target for treatment. Up to a third of migraineurs avoid or delay
taking oral medications due to nausea as reported in
surveys of patient practices during migraine attacks.4,5
Treatment guidelines suggest that in the presence of
severe nausea, non-oral treatments should be considered.6 Therefore, better characterization of the patterns of nausea in individuals with episodic migraine
may help improve migraine treatment and outcomes.
Clinical observation suggests that patterns of
nausea vary widely from person to person and within
an individual from attack to attack. The American
Migraine Prevalence and Prevention (AMPP) Study

provides cross-sectional and longitudinal data on

headache epidemiology, disease-related impact,
health care related resource use, and outcomes
among persons with migraine in the United States.7-13
Previous AMPP cross-sectional data indicated that
nearly half (49.5%) of persons with episodic migraine
(EM) experienced nausea half the time or more often
with their headaches.14 Other work reported on the
demographic patterns of frequent headache-related
nausea, the associated increases in other headache
symptoms and disease-related burden15 as well as
direct health care-related costs.15 Herein, we use data
from the AMPP Study to characterize the frequency
of persistent headache-related nausea and ask the
question: Is persistent headache-related nausea associated with a greater risk of progression to chronic
migraine (CM)7. Our analysis contrasts those persistent frequent nausea (PFN, nausea occurring more
than half of the time with headache for
2 consecutive years) and those with no or low

Conflicts of Interest: Michael L. Reed: Dr. Reed has received funding for research and data analysis from Allergan Inc., CoLucid,
Endo Pharmaceuticals, GlaxoSmithKline, MAP Pharmaceuticals, Merck & Co., Inc., Novartis, NuPathe, Ortho-McNeil, and the
National Headache Foundation. Kristina M. Fanning: Dr. Fanning has received research support from Allergan Inc., CoLucid, Endo
Pharmaceuticals, GlaxoSmithKline, MAP Pharmaceuticals, Merck & Co., Inc., NuPathe, Novartis, Ortho-McNeil, and the National
Headache Foundation. Daniel Serrano: Dr. Serrano has received research support from Allergan Inc., CoLucid, Endo Pharmaceuticals, GlaxoSmithKline, MAP Pharmaceuticals, Merck & Co., Inc., NuPathe, Novartis, Ortho-McNeil, and the National Headache
Foundation. Dawn C. Buse: Dr. Buse has received grant support and honoraria from Allergan Inc., MAP Pharmaceuticals, Novartis,
and Zogenix. Richard B. Lipton: Dr. Lipton receives research support from the NIH (PO1 AG03949 [Program Director, Project and
Core Leader], RO1AG025119 [Investigator], RO1AG022374-06A2 [Investigator], RO1AG034119 [Investigator], RO1AG12101
[Investigator], K23AG030857 [Mentor], K23NS05140901A1 [Mentor], and K23NS47256 [Mentor]), the National Headache Foundation, and the Migraine Research Fund; serves on the editorial board of Neurology, has reviewed for the NIA and NINDS, holds
stock options in eNeura Therapeutics; serves as consultant, advisory board member, or has received honoraria from: Allergan Inc.,
American Headache Society, Autonomic Technologies, Boehringer-Ingelheim Pharmaceuticals, Boston Scientific, Bristol Myers
Squibb, Cognimed, Colucid, Eli Lilly, ENDO, eNeura Therapeutics, GlaxoSmithKline, Merck & Co., Inc., Novartis, NuPathe, Pfizer
and Vedanta Research.
Financial Support: The American Migraine Prevalence and Prevention Study was funded through a research grant to the National
Headache Foundation from McNeil-Janssen Scientific Affairs LLC, Raritan, NJ. The AMPP Study database was donated by
McNeil-Janssen Scientific Affairs LLC to the National Headache Foundation for use in various projects. Additional analyses and
manuscript preparation were supported by a grant from NuPathe Inc., a wholly owned subsidiary of Teva Pharmaceuticals USA,
Inc., to the National Headache Foundation.

78
frequency nausea (NLFN) during headache, on rates
of progression from EM to the more severe state of
CM at follow-up.

METHODS
The AMPP is a 2-phase longitudinal populationbased study whose methods have been described
in detail previously.16,17 The study was modeled after
study designs used in the American Migraine Study I
and II.1,18 In brief, persons with self-reported severe
headache were selected from a representative
sample of the general US population. In phase I, a
self-administered screening questionnaire was mailed
to 120,000 households encompassing 257,339 household members drawn from a 600,000-household
nationwide panel maintained by Taylor Nelson Sofres
(TNS, formerly National Family Opinion Inc.).
Surveys were returned by 77,879 households (64.9%
response rate) yielding data for 162,756 household
members aged 12 years or older. A total of 30,721
respondents reported severe headache, and usable
data were obtained from 28,261 who reported severe
headache in the preceding year.
Phase II of AMPP is a longitudinal study initiated in 2005. Annual surveys assessed headache
frequency and symptoms, headache-related impairment, and health care resource utilization among
other variables. Surveys also contain a diagnostic
module, which was developed and validated as part
of the American Migraine Study.19,20 It has been
shown to have sensitivity of 100% and specificity of
82% for the diagnosis of migraine, and sensitivity of
91% and specificity of 80% for the diagnosis of CM.
Reasons for nonparticipation were not assessed
during screening or at annual follow-up. The current
analysis is based on data collected from the 2007,
2008, and 2009 AMPP study survey mailings. This
study was reviewed and approved by the Albert
Einstein College of Medicine Institutional Review
Board.
Sample Population.The present analysis is
based on the most recent AMPP Study years where
surveys were mailed to n = 20,489 individuals with
headache in 2007, n = 17,892 in 2008, and n = 16,983
in 2009. Surveys were returned by n = 14,062,
n = 10,720, and n = 11,799 individuals aged 18 years

January 2015
or older, respectively, in each of the three years.
International Classification of Headache Disorders
2nd Edition (ICHD-2) criteria21 were used to identify individuals with EM in the 2007 and 2008 survey
years. (ICHD-3 beta criteria have been developed
since this study was completed; however, the criteria
for diagnosing migraine did not change.) The sample
was then stratified into a group with either PFN or
NLFN. Nausea frequency patterns were determined
based on responses to the headache symptom item
You feel nauseated or sick to your stomach which
was reported using prelisted response options of [1]
never, [2] rarely, [3] less than half the time, and [4]
half the time or more. PFN was defined by the presence of nausea half the time or more with headache
in both the 2007 and 2008 surveys. NLFN was
defined by nausea less than half the time, rarely or
never with headache in both survey years. The
AMPP Study survey structure provided three opportunities to report on headache symptom frequency:
either for the most severe type of headache, the
second most severe type of headache, or for other
type of headache. The symptom frequency scores
used for nausea group assignment and for computing the symptom composite score (see below) were
acquired from the headache classified as migraine
(regardless of self-reported severity). Persons were
considered to have CM in 2009 if they met
Silberstein and Lipton symptom criteria22 with headache day frequency of 15 days over the preceding 3
months. The rate of progression to CM in 2009 was
measured and compared between groups. The reference group for analysis was survey respondents who
remained EM in 2009 or met criteria for other
non-CM headache types.
Regression Modeling and Covariates.The impact
of nausea on headache frequency and progression to
CM could be affected by a number of variables
working alone or in combination with nausea. To
adjust for the influence of variables other than nausea
on the risk of progression, our analysis included
covariates where links to CM progression have been
found previously. This included sociodemographic
characteristics,
headache
symptom
severity,
headache-related disability, depression, and opioid
medication use.

Headache
Sociodemographic Variables: Age and sex were
obtained for each respondent via self-report in the
survey.Data for annual household income,race,Hispanic
origin, and household size were provided by TNS from
household records that were updated annually.
Symptom Composite Score: A symptom composite
score was included in the modeling to control for the
potential influence of other headache symptoms and
overall headache severity on the rate of progression to
CM. The score is computed by summing symptom frequency categories [0] never, [1] rarely, [2] less than half
the time,[3] half the time or more for each of six migraine
headache features: unilateral pain, pounding, pulsating
or throbbing pain, pain made worse by routine activities,
photophobia, and phonophobia.
Headache-Related Disability: Headache-related
disability was assessed with the Migraine Disability
Assessment Scale (MIDAS),23 which is a validated selfadministered questionnaire consisting of 5 items that
assess days of missed activity or substantially reduced
activity over the preceding 3 months due to headache.
MIDAS includes domains for schoolwork or paid
employment, household work or chores, and nonwork
(family, social, and leisure) activities. Scale scores were
treated as continuous variables in calculating descriptive
statistics; however, for modeling CM onset, we recoded
MIDAS to 10-point intervals to facilitate interpretation
of odds ratio (OR) calculations. The modeled effect is
interpreted as the increase in odds of CM onset for every
10-point change in MIDAS score.
Depression: Depressive symptomology was
obtained, and a clinical diagnosis was made using the
9-item Patient Health Questionnaire (PHQ-9), which is
part of the Primary Care Evaluation of Mental Disorders
(PRIME-MD).24 The PHQ-9 is a validated measure of
major depressive disorder based on the Diagnostic and
Statistical Manual of Mental Disorders Fourth Edition
criteria.Each of the 9 items is rated on a 4-point scale that
assesses symptoms and functional impairment over the
preceding 2 weeks. Depression was coded as a dichotomous variable using a cut score of 10 which indicated
moderate, moderately severe, or severe depressive
symptomology.
Opioid Medication Use for Headache: Prescription
medication use was obtained by asking respondents to
report the medication you currently use to treat your

79
most severe type of headaches.Opioid containing medication endorsements were captured via precoded and
open-ended questioning. Generic and brand name mentions of the same medication were combined.The list of
reported opioid containing medications included:
Darvocet (Eli Lilly and Company, Indianapolis, IN,
USA),Tylenol with codeine (McNeil Consumer Healthcare, Fort Washington, PA, USA), Vicodin (Abbott
Laboratories, Abbott Park, IL, USA), Vicoprofen
(Abbott Laboratories), Demerol (Sanofi-Aventis US
LLC, Bridgewater, NJ, USA; oral or injected meperidine), Stadol/butorphanol nasal spray (Apothecon,
Bristol-Myers Squibb, Princeton, NJ, USA), Fentanyl
(Janssen Healthcare,Titusville, NJ, USA), hydrocodone,
morphine, oxycodone, hydromorphone, and pentazocine. Barbiturate medication use has also been found
associated with increased risk of CM progression,25 but
the number of respondents reporting barbiturate use in
the current sample was insufficient for inclusion in the
regression modeling as a covariate.
Statistical Analysis.Statistical analyses were performed using SPSS (IBM SPSS Statistics for Windows,
Version 20.0,IBM Corp.,Armonk,NY,USA).A value of
.05 was used to identify statistically significant differences. Descriptive statistics were generated on NLFN
and PFN group members for sociodemographic variables and for the other covariates. Chi-square (chi) was
used to assess differences on binary and categorical variables (sex, income, race, Hispanic origin, household size,
opioid use, and depression) and ANOVA was used to
assess differences between groups on continuous variables (age, MIDAS). Binary logistic regressions were
performed hierarchically to assess progression to CM in
2009 as a function of nausea status in 2007 and 2008.The
initial regression model included sociodemographic
covariates only. The variables age, sex, and race were
retained in subsequent models; however, income, household size, and Hispanic origin were removed because
they did not contribute additional predictive value to the
models. Modeling progressed sequentially adding the
symptom composite score, MIDAS headache-related
disability, PHQ-9 depression, and then opioid medication use. To adjust for the risk of confounding arising
from the association between opioid use and CM onset26
and opioid use and treatment emergent nausea,26 we
added the baseline interaction between persistent

80

January 2015
nausea and opioid use at baseline) to the model. For all
models, ORs and 95% confidence intervals (CI) contrasted the NLFN and PFN groups on progression from
EM in 2007 and 2008 to CM from 2009.
To determine the possible impact of sample attrition on the baseline assessment of CM, target respondents with complete data from the 2007 and 2008
surveys were contrasted with those who were lost to
follow-up between 2007 and 2008. Because our
outcome was assessed in 2009, a second analysis was
completed for those lost to follow-up between 2008
and 2009. ORs and CIs were derived from binary
logistic regression models of gender, age, income,
household size, race, ethnicity, headache day frequency, symptom composite score, and nausea.

Figure.Data collection, sample flow, and attrition across 3

years of data collection (2007, 2008, 2009). *The usable sample
in each year were subjects who met ICHD-2 criteria for episodic migraine (2007 and 2008), provided nausea symptom
data, and who could be demographically matched (on age and
gender) to prior year (2008 and 2009) data.

frequent nausea and opioid use (along with the

commensurate main effects of persistent frequent

RESULTS
The Figure provides data collection results,
sample flow, and respondents lost to follow-up over
the 3 years of data collection in 2007, 2008, and 2009.
Survey respondents considered usable in the analysis
met ICHD-2 symptom criteria for EM (in 2007 and
2008), provided nausea symptom data, and could be
matched on subject ID, age, and gender (in 2008 and
2009) to their survey in the prior year. There were
2665 subjects lost to follow-up between 2007 and 2008
and 684 subjects lost to follow-up between 2008 and
2009. A total of 3182 subjects provided usable surveys
for all 3 years of data collection. There were 1389

Table 1.Distribution of Headache-Related Nausea Frequency for Survey Years 2007 and 2008 Among Respondents Providing
Data in All Three Time Periods

Nausea With Headache Frequency in 2008 Questionnaire

Nausea With Headache

Frequency in 2007 Questionnaire

Never

Rarely

Less Than
Half the Time

Half the
Time or More

Total (n)

Never
Rarely
Less than half the time
Half the time or more
Total (n)

1.1%
1.0%
0.8%
0.4%
103

0.9%
4.5%
3.9%
2.3%
367

0.3%
3.5%
11.7%
11.5%
859

0.6%
2.7%
11.3%
43.7%
1853

92
370
879
1841
3182

NLFN (877, 27.7% with no or low frequency nausea with headache).

PFN (1,389, 43.7% with persistent frequent nausea with headache).
The balance of the sample (916, 28.8%) provided inconsistent data on nausea frequency.

Headache

81

eligible respondents with PFN (43.7%) and 877 with

NLFN (27.7%) in both 2007 and 2008 (Table 1).
The balance of the sample (916, 28.8%) provided
inconsistent nausea frequency data in 2007 and 2008.
In 2009, 3.4% of those with PFN in 2007 and 2008 had
progressed to CM, and 1.5% of those with NLFN had
done so.
Females were overrepresented in the PFN group
(86.2%) compared to the NLFN group (78.8%,
P = .000), as were Caucasians (P = .006) (Table 2).
There were no significant differences between nausea
groups with respect to age, income, Hispanic origin, or
household size (Table 2). However, those with PFN vs
NLFN were more likely to report more frequent
headache symptoms (P = .001), greater headacherelated disability (P = .001), higher rates of depression (PHQ-9 cut score 10, P = .002), and more
frequent opioid use (P = .001).

Results from logistic regression modeling

(Table 3) indicated that progression to CM was
roughly twice as likely for the PFN group compared
to NLFN group members after adjusting for sociodemographic features (OR 2.09, 95% CI 1.11-3.91,
P = .022). With the addition to the model of the
symptom composite score and headache-related disability, the OR was slightly attenuated but remained
significant (OR 2.00, 95% CI 1.03-3.87, P = .04). Statistical significance was lost with the addition of
PHQ-9 depression covariate; however, the nearly
2-fold increase in risk of CM progression for PFN
respondents remained (OR 1.90, 95% CI 0.98-3.71,
P = .059). In the final model, which included nausea
covariates, opioid medication use and the nausea by
opioid use interaction, the PFN group is at increased
risk for progression to CM (OR 2.24, 95% CI 1.074.70, P = .033).

Table 2.Sociodemographic Features, Headache Characteristics, and Comorbidities by Headache-Related Nausea Frequency
Group Defined by Nausea Status in Both 2007 and 2008

Nausea With Headache Frequency Group

Variable

Sex
Mean age (SD)
Annual household income

Hispanic origin
Race
Household size

Symptom composite score mean (SD)

MIDAS Mean (SD)
Opioid use for headache (Yes)
PHQ-9 score (Depression = no)
PHQ-9 score (Depression = yes)

Male
Female
<$30,000
$30,000 to 49,999
$50,000 to 74,999
$75,000
Hispanic
Non-Hispanic
Caucasian
Other
1 member
2 members
3 members
4 members
5 members

<10
10

NLFN
(n = 877)

PFN
(n = 1389)

Estimate

P value

186 (21.2%)
691 (78.8%)
50.6 (12.2)
237 (27.0%)
206 (23.5%)
187 (21.3%)
247 (28.2%)
30 (3.6%)
812 (96.4%)
770 (90.4%)
82 (9.6%)
182 (20.8%)
298 (34.0%)
165 (18.8%)
134 (15.3%)
98 (11.2%)
11.5 (2.4)
6.3 (11.3)
77 (8.8%)
739 (85.0%)
130 (15.0%)

192 (13.8%)
1197 (86.2%)
50.2 (11.4)
375 (27.0%)
323 (23.3%)
310 (22.3%)
381 (27.4%)
34 (2.5%)
1,315 (97.5%)
1265 (93.6%)
87 (6.4%)
246 (17.7%)
535 (38.5%)
239 (17.2%)
231 (16.6%)
138 (9.9%)
13.2 (2.0)
9.7 (14.7)
212 (15.3%)
1102 (79.9%)
277 (20.1%)

21.10

.000

0.60
0.36

.438
.948

1.99

.159

7.51

.006

7.83

.098

329.23
33.51
20.30
9.45

.001
.001
.001
.002

chi-square test statistic.

ANOVA F statistic.
MIDAS = Migraine Disability Assessment Scale; NLFN = no or low frequency nausea; PFN = persistent frequent nausea; PHQ9 = Patient Health Questionnaire, 9-item depression schedule.

82

January 2015
Table 3.Predictors of CM Onset in 2009 Based on Nausea Status in 2007 and 2008 and 2007 Covariates:
Binary Logistic Models

95% CI

Model 1 Sociodemographics only

No or low frequency nausea (NLFN) vs persistent frequent nausea (PFN)
Age
Sex (Female)
Income
Household size
Hispanic origin (Yes)
Race (Caucasian)

OR

Lower

Upper

Significance

2.09
0.99
1.23
0.89
1.00
0.80
1.45

1.11
0.96
0.55
0.71
0.80
0.11
0.44

3.91
1.01
2.76
1.13
1.26
5.98
4.76

.022
.337
.609
.330
.974
.824
.541

3.87
1.01
2.92
1.17
1.16
1.42

.040
.466
.526
.557
.821
.000

Model 2 Sociodemographics, symptom composite score, and headache-related disability (MIDAS)

NLFN vs PFN
2.00
1.03
Age
0.99
0.97
Sex (Female)
1.30
0.58
Income
0.93
0.74
Symptom composite score
1.02
0.89
Headache-related disability (MIDAS)
1.28
1.16

Model 3 Sociodemographics, symptom composite score, headache-related disability (MIDAS) and PHQ-9 depression
NLFN vs PFN
1.90
0.98
3.71
Age
0.99
0.97
1.01
Sex (Female)
1.37
0.61
3.08
Income
1.02
0.81
1.30
Symptom composite score
1.00
0.87
1.14
Headache-related disability (MIDAS)
1.23
1.11
1.37
Depression (PHQ-9)
2.87
1.63
5.05

.059
.397
.451
.858
.964
.000
.000

Model 4 Sociodemographics, symptom composite score, headache-related disability (MIDAS), depression (PHQ-9)
and opioid use
NLFN vs PFN
2.24
1.07
4.70
Age
0.99
0.97
1.01
Sex (Female)
1.36
0.60
3.08
Income
1.02
0.81
1.30
Symptom composite score
1.00
0.88
1.14
Headache-related disability (MIDAS)
1.23
1.10
1.37
Depression (PHQ-9)
2.88
1.63
5.08
Opioid use
2.51
0.66
9.51
Nausea by opioid use interaction
0.36
0.08
1.69

.033
.373
.456
.848
.998
.000
.000
.176
.198

Household size, race, and Hispanic origin did not contribute to the initial model and were removed from subsequent models.

Of 7694 study respondents in 2007, 2665 EM

respondents were lost to follow-up in 2008, and the
analysis of attrition indicated that attrition was less
likely in Caucasian respondents (OR 0.78, 95% CI
0.63-0.97) and in those with greater symptom severity
(OR 0.94, 95% CI 0.91-0.96), and more frequent in
those with nausea symptoms (OR 0.89, 95% CI 0.830.96). Of the 3888 eligible cases in 2008, 684 were lost
to follow-up in 2009. The only significant predictor
associated with attrition between 2008 and 2009 was

age (OR 0.98, 95% CI 0.97-0.98) with older people

less likely to be lost. This is consistent with prior findings on research participation and attrition, and
because headache frequency is not associated with
attrition, we can conclude that our outcome was likely
not impacted by attrition (See Table 4).

DISCUSSION
We assessed the frequency and persistence of
headache-related nausea in the 2007 and 2008

Headache

83

Table 4.Binary Logistic Models of the Impact of Attrition on Exposure During Baseline (2007 and 2008) and on the CM
Outcome Measure in 2009

Exposure
Exposure Attrition (n = 2665)
vs Final Sample (n = 2266)
Predictors Are From 2007

Outcome
Outcome Attrition (n = 684)
vs Final Sample (n = 2266)
Predictors Are From 2008

95% CI

Sex (male is REF)

Age
Income (<$30,000 is REF)
$30,000-$49,000
$50,000-$74,999
$75,000
Household size
Race Caucasian vs Other (other is REF)
Hispanic (non-Hispanic origin is REF)
HA Days 0-4 (REF)
HA Days 5-9
HA Days 10-14
Symptom composite score
Nausea with headache

95% CI

OR

Lower

Upper

OR

Lower

Upper

0.86
0.995

0.74
0.99

1.01
1.00

1.05
0.98

0.82
0.97

1.35
0.98

1.01
0.94
0.99
1.00
0.78
0.96

0.86
0.79
0.85
0.95
0.63
0.66

1.20
1.11
1.16
1.05
0.97
1.41

0.96
0.88
0.97
1.07
0.79
0.98

0.74
0.68
0.76
0.99
0.57
0.56

1.24
1.15
1.23
1.15
1.08
1.70

0.94
1.12
0.94
0.89

0.79
0.88
0.91
0.83

1.12
1.44
0.96
0.96

0.97
1.16
0.98
1.09

0.74
0.74
0.94
0.97

1.26
1.81
1.02
1.22

CI = confidence interval; HA = headache; OR = odds ratio; REF = reference.

Confidence intervals that do not include 1.0 are significant, P < .05.

AMPP survey as a predictor of progression to CM

in 2009. The results show that persistent frequent
nausea (half the time or more in both 2007 and
2008) occurs in about half (43.7%) the episodic
migraine population. Women and Caucasians were
overrepresented in the group with PFN. Higher
rates of nausea have previously been reported in
women,27 and lower rates of nausea have previously
been reported in African Americans.28 The reasons
for different symptom profiles in men and women29
or among racial subgroups are not fully understood
although biological, genetic, and psychosocial
hypotheses have been proposed.30-32
A prior cross-sectional analysis of data from the
AMPP Study demonstrated that frequent headacherelated nausea is associated with higher rates of
moderate to severe headache pain, greater headacherelated disability and impact, and higher rates of
depression compared to nausea free migraineurs.14
Episodic migraineurs with frequent headache-related

nausea also had annual per-person direct care utilization costs that were 1.7 times higher for primary care/
obstetricsgynecology utilization, 2.1-times higher
for neurology and headache specialist utilization,
5.7-times higher for emergency department and
urgent care utilization, and 8.3-times higher for overnight hospital stays.15
This prior work demonstrates that migraine sufferers with nausea have more severe disease and use
more resources than migraine sufferers free of
nausea. The longitudinal analyses presented herein
show that when frequent nausea persists, over time it
is associated with a doubling of the risk of progression
to CM compared to those with no or low frequency
nausea (3.4% vs 1.5%). The 2-fold increase seen
in the odds of progressing to CM persists after
controlling for sociodemographic features, migraine
symptom severity, headache-related disability,
depression, and opioid medication use. The interaction between baseline nausea status and opioid

84
medication use was included in the model to control
for potential confounding by indication. Given that
PFN is associated with the onset of CM, and that
opioid use is associated with both treatment emergent nausea and CM onset, the interaction controls
for the possibility that the association between PFN
and CM onset found in the earlier models was due to
the confounding influence of opioid medication use
rather than PFN. Because the main effect of PFN on
CM onset remained significant when the interaction
is included in the model and the interaction itself is
not significant, it can be concluded that effect is due to
PFN and not the result of confounding.
Nausea has also been identified as a significant
predictor (P < .05) of chronic migraine in tensiontype headache based on pooled univariate data analysis from a population based study.33 Some risk factors
for progression to chronic migraine can be modified
by health or lifestyle changes (eg, obesity) or changes
in treatment (medication use), but others cannot
readily be impacted (eg, sex, race).33,34 Chronic
migraine is associated with greater physical and
social burden and less satisfaction with care.35 Other
studies have shown that individuals with chronic
migraine have greater headache-related disability36
and headache impact,12 lower socioeconomic status,37
lower health-related quality of life,38,39 higher rates
of comorbid medical conditions, depression, anxiety,
and chronic pain,40 lower productivity at work,41,42 and
increased health care resource utilization and costs.43
These data indicate that persistent frequent
nausea may directly worsen migraine-associated outcomes. Moreover, a history of persistent frequent
nausea may help to identify migraineurs with severe
and debilitating disease as well as those who are at
greater risk of progression to CM. It is not known
whether persistent frequent nausea is a marker for
the risk of CM progression or in the causal pathway.
The fact that PFN remains significant after adjusting
for disability, migraine symptom severity, depression,
and opioid use suggests that PFN may be more than a
marker of severity. Clinical management approaches
that consider nausea as well as pain may improve
outcomes for people with migraine. Clearly, reducing
the risk of progression to CM should be a primary aim
of therapeutic interventions.

January 2015
This study has several limitations. First, these data
were obtained by respondent self-report and were
not verified by health care providers, medical records,
or medication databases. However, this is a common
approach in large-scale population-based studies. A
further limitation is that our diagnostic criteria for
CM used the Silberstein and Lipton criteria,22 instead
of those informed by stricter ICHD-2 or ICHD-3 beta
criteria, which are difficult to implement in a large
population study. Among female respondents, a
potential study confounder is the presence of nausea
associated with pregnancy. However, nausea would
be unlikely to persist over 2 consecutive years, and in
fact only two females reported pregnancy in both
2007 and 2008 where exposure was being assessed.
Strengths of this study include a large populationbased sample, longitudinal study design with 3 consecutive years of data, the ability to assess the effects
of sampling attrition on study outcomes, the use of
validated survey tools, the assignment of migraine
diagnosis based on a validated diagnostic module, and
the ability to adjust for multiple confounders. Our
modeling included covariates found in prior studies
to be more common among those with CM or implicated in the progression to CM among persons with
migraine, but still the effect of persistent headacherelated nausea remained statistically significant and
indicated a 2-fold greater risk of progressing
from EM to CM for PFN respondents. Despite
the robust nature of these findings there may be a
latent predictor or other causal factors yet to be
identified. We cannot definitively establish the causal
relationship between persistent frequent nausea in
EM and CM progression. Additional prospective
studies are needed to better understand the relationship between nausea persistence and migraine progression and outcomes and the potential benefit
of anti-nausea treatment in preventing disease
progression.
Frequent headache-related nausea among episodic migraineurs is common and associated with
increased headache symptom severity and frequency,
increases in headache-related disability, depression,
and health care resource utilization and costs. When
frequent nausea persists over time, the risk of
progression to CM increases 2-fold. Migraine is a

Headache
constellation of symptoms, and these data suggest
that headache-related nausea frequency warrants
monitoring and treatment in concert with the management of headache pain. Since persistent nausea
may impact migraine medication use and effectiveness,5 effective management of nausea symptoms
could improve headache treatment and outcomes
among those with episodic migraine.

STATEMENT OF AUTHORSHIP
Category 1
(a) Conception and Design
Michael L. Reed; Dawn C. Buse; Richard B.
Lipton
(b) Acquisition of Data
Michael L. Reed
(c) Analysis and Interpretation of Data
Michael L. Reed; Kristina M. Fanning; Daniel
Serrano; Dawn C. Buse; Richard B. Lipton
Category 2
(a) Drafting the Manuscript
Michael L. Reed; Kristina M. Fanning; Daniel
Serrano; Dawn C. Buse; Richard B. Lipton
(b) Revising It for Intellectual Content
Michael L. Reed; Kristina M. Fanning; Daniel
Serrano; Dawn C. Buse; Richard B. Lipton
Category 3
(a) Final Approval of the Completed Manuscript
Michael L. Reed; Kristina M. Fanning; Daniel
Serrano; Dawn C. Buse; Richard B. Lipton

REFERENCES
1. Lipton RB, Stewart WF, Diamond S, Diamond ML,
Reed M. Prevalence and burden of migraine in the
United States: Data from the American Migraine
Study II. Headache. 2001;41:646-657.
2. Holroyd KA, Drew JB, Cottrell CK, Romanek KM,
Heh V. Impaired functioning and quality of life in
severe migraine: The role of catastrophizing and
associated symptoms. Cephalalgia. 2007;27:11561165.
3. Silberstein SD. Migraine symptoms: Results of a
survey of self-reported migraineurs. Headache.
1995;35:387-396.

85
4. Pryse-Phillips W, Aub M, Bailey P, et al. A clinical
study of migraine evolution. Headache. 2006;46:
1480-1486.
5. National Headache Foundation. Migraine patient
survey.
2010.
http://www.headaches.org/NHF
_Programs/NHF_Web_Surveys. Accessed June 14,
2012.
6. Silberstein SD. Practice parameter: Evidence-based
guidelines for migraine headache (an evidencebased review): Report of the Quality Standards Subcommittee of the American Academy of Neurology.
Neurology. 2000;55:754-762.
7. Diamond S, Bigal ME, Silberstein S, Loder E, Reed
ML, Lipton RB. Patterns of diagnosis and acute and
preventive treatment for migraine in the United
States: Results from the American Migraine Prevalence and Prevention Study. Headache. 2007;47:355363.
8. Lipton RB, Bigal ME, Diamond M, Freitag F, Reed
ML, Stewart WF. Migraine prevalence, disease
burden, and the need for preventive therapy. Neurology. 2007;68:343-349.
9. Stewart WF, Wood GC, Razzaghi H, Reed ML,
Lipton RB. Work impact of migraine headaches.
J Occup Environ Med. 2008;50:736-745.
10. Ashina S, Serrano D, Lipton RB, et al. Depression
and risk of transformation of episodic to chronic
migraine. J Headache Pain. 2012;13:615-624.
11. Lipton RB, Serrano D, Holland S, Fanning
KM, Reed ML, Buse DC. Barriers to the diagnosis and treatment of migraine: Effects of sex,
income, and headache features. Headache. 2013;
53:81-92.
12. Buse D, Manack A, Serrano D, et al. Headache impact of chronic and episodic migraine:
Results from the American Migraine Prevalence and Prevention study. Headache. 2012;52:
3-17.
13. Holland S, Fanning KM, Serrano D, Buse DC, Reed
ML, Lipton RB. Rates and reasons for discontinuation of triptans and opioids in episodic migraine:
Results from the American Migraine Prevalence
and Prevention (AMPP) study. J Neurol Sci.
2013;326:10-17.
14. Lipton RB, Buse DC, Saires J, Fanning KM, Serrano
D, Reed ML. Frequency and burden of headacherelated nausea: Results from the American Migraine
Prevalence and Prevention (AMPP) study. Headache. 2013;53:93-103.

86
15. Lipton RB, Buse DC, Saires J, Serrano D, Reed ML.
Healthcare resource utilization and direct costs
associated with frequent nausea in episodic
migraine: results from the American Migraine
Prevalence and Prevention (AMPP) study. J Med
Econ. 2013;16:1-10.
16. Lipton RB, Bigal ME, Diamond M, Freitag F, Reed
ML, Stewart WF. Migraine prevalence, disease
burden and the need for preventive therapy. Neurology. 2007;68:343-349.
17. Buse DC, Manack AN, Fanning K, et al. Chronic
migraine prevalence, disability, and sociodemographic factors: Results from the American
Migraine Prevalence and Prevention Study. Headache. 2012;52:1456-1470.
18. Stewart WF, Lipton RB, Celentano DD, Reed
ML. Prevalence of migraine headache in the
United States. Relation to age, income, race, and
other sociodemographic factors. JAMA. 1992;267:
64-69.
19. Lipton RB, Diamond S, Diamond M, Reed M,
Stewart WF. Migraine diagnosis and treatment:
Results from the American Migraine Study II. Headache. 2001;41:638-645.
20. Liebenstein M, Bigal M, Sheftell F, Tepper S,
Rapoport A, Lipton RB. Validation of the Chronic
Daily Headache Questionnaire (CDH-Q). Headache. 2007;47:760-761.
21. Headache Classification Subcommittee of the
International Headache Society. The international
classification of headache disorders: 2nd edition.
Cephalalgia. 2004;24(Suppl. 1):9-160.
22. Silberstein SD, Lipton RB, Sliwinski M. Classification of daily and near-daily headaches: Field
trial of revised IHS criteria. Neurology. 1996;47:871875.
23. Stewart WF, Lipton RB, Kolodner KB, Sawyer J,
Lee C, Liberman JN. Validity of the Migraine Disability Assessment (MIDAS) score in comparison to
a diary-based measure in a population sample of
migraine sufferers. Pain. 2000;88:41-52.
24. Kroenke K, Spitzer RL, Williams JB. The PHQ-9:
Validity of a brief depression severity measure.
J Gen Intern Med. 2001;16:606-613.
25. Bigal ME, Serrano D, Buse D, Scher A, Stewart WF,
Lipton RB. Acute migraine medications and evolution from episodic to chronic migraine: A longitudinal population-based study. Headache. 2008;48:
1157-1168.

January 2015
26. Buse DC, Pearlman SH, Reed ML, Serrano D,
Ng-Mak DS, Lipton RB. Opioid use and dependence among persons with migraine: Results of the
AMPP study. Headache. 2012;52:18-36.
27. Stewart WF, Wood C, Reed ML, Roy J, Lipton RB.
Cumulative lifetime migraine incidence in women
and men. Cephalalgia. 2008;28:1170-1178.
28. Stewart WF, Lipton RB, Liberman J. Variation in
migraine prevalence by race. Neurology. 1996;47:5257.
29. Buse DC, Loder EW, Gorman JA, et al. Sex differences in the prevalence, symptoms, and associated
features of migraine, probable migraine and other
severe headache: Results of the American Migraine
Prevalence and Prevention (AMPP) study. Headache. 2013;53:1278-1299.
30. MacGregor EA, Rosenberg JD, Kurth T. Sexrelated differences in epidemiological and clinicbased headache studies. Headache. 2011;51:843-859.
31. Smitherman TA, Burch R, Sheikh H, Loder E. The
prevalence, impact, and treatment of migraine and
severe headaches in the United States: A review of
statistics from national surveillance studies. Headache. 2013;53:427-436.
32. Merikangas KR. Contributions of epidemiology to
our understanding of migraine. Headache. 2013;53:
230-246.
33. Ashina S, Lyngberg A, Jensen R. Headache characteristics and chronification of migraine and
tension-type headache: A population-based study.
Cephalalgia. 2010;30:943-952.
34. Bigal ME, Lipton RB. Modifiable risk factors for
migraine progression. Headache. 2006;46:1334-1343.
35. Bigal ME, Serrano D, Reed M, Lipton RB. Chronic
migraine in the population: Burden, diagnosis, and
satisfaction with treatment. Neurology. 2008;71:559566.
36. Bigal ME, Rapoport AM, Lipton RB, Tepper SJ,
Sheftell FD. Assessment of migraine disability
using the migraine disability assessment (MIDAS)
questionnaire: A comparison of chronic migraine
with episodic migraine. Headache. 2003;43:
336-342.
37. Buse DC, Manack A, Serrano D, Turkel C, Lipton
RB. Sociodemographic and comorbidity profiles of
chronic migraine and episodic migraine sufferers.
J Neurol Neurosurg Psychiatry. 2010;81:428-432.
38. Blumenfeld AM, Varon SF, Wilcox TK, et al. Disability, HRQoL and resource use among chronic and

Headache
episodic migraineurs: Results from the International
Burden of Migraine Study (IBMS). Cephalalgia.
2010;31:301-315.
39. Meletiche DM, Lofland JH, Young WB. Quality-oflife differences between patients with episodic and
transformed migraine. Headache. 2001;41:573-578.
40. Buse DC, Silberstein SD, Manack AN,
Papapetropoulos S, Lipton RB. Psychiatric
comorbidities of episodic and chronic migraine.
J Neurol. 2013;260:1960-1969.
41. Stewart WF, Wood GC, Manack A, Varon SF, Buse
DC, Lipton RB. Employment and work impact of
chronic migraine and episodic migraine. J Occup
Environ Med. 2010;52:8-14.

87
42. Serrano S, Manack AN, Reed ML, Buse DC, Varon
SF, Lipton RB. Cost and predictors of lost productive time in chronic migraine and episodic migraine:
Results from the American Migraine Prevalence
and Prevention (AMPP) study. Value Health. 2013;
16:31-38.
43. Stokes M, Becker WJ, Lipton RB, et al. Cost of
health care among patients with chronic and episodic migraine in Canada and the USA: Results
from the International Burden of Migraine Study
(IBMS). Headache. 2011;51:1058-1077.