120
ASAA
Santos ERR, Oliveira DA, Leite AFB, Rocha Filho PAS, Valença MM
Headache catastrophization and its relationship with disability, depression, anxiety, stress and sleep quality
indicates a more signicant impact of the headache on
daily living. The results were classied as little or no impact:
38 to 49 points; some impact: 50 to 55 points; substantial
impact: 56-59 points; very severe impact: 60 and over.
9,10
We used the criteria of the International Classication of
Headache Disorders (ICHD)
11
to classify the headache
types; individuals with non-migraine headache and
migraine were divided into two groups and denominated:
Group A (individuals with migraine) and B (students with
other primary headaches).
The Beck Depression Inventory (BDI) is a 21-item
psychometric self-assessment tool and was used to track
the depression symptoms. Each category has four or ve
alternatives, which express the severity levels of depressive
symptoms. The score ranges from zero to three, where
zero is a lack of symptoms and three evidences the
most intense ones. Scores up to 9 points suggest lack of
depression or minimal symptoms; 10 to 18 points, mild to
moderate depression; 19 to 29 points, moderate to severe
depression; 30 to 63 points, severe depression. A score of
21 or more in the inventory is usually representative of a
clinically signicant depression.
12,13
We used the 21-item Beck Anxiety Inventory (BAI) to
elucidate anxiety symptoms. For each question, the
individual should choose one of four levels of anxiety on
a scale ranging from zero to three. The items added can
achieve a total score ranging from 0 to 63, generated from
the sum of the scores of individual items, which allows
classication of anxiety intensity levels as mild, moderate,
or severe.
14
The Perceived Stress Scale (PSS) was used to evaluate
stress, from 14 questions as answer alternatives, with a
score ranging from 0 to 4 points. The scale total is the
sum of the scores, and scores range from 0 to 56, and the
higher the score, the greater the stress.
15,16
The Pittsburgh Sleep Quality Index was applied to assess
the sleep quality of the previous month. The global score
ranges from 0 to 21 points, and the highest the score, the
worse the quality of sleep. This scale classies participants
as the highest the score, the greater the tendency of bad
sleep quality.
17
Statistical Analysis
We performed analyses with measures and structuring
elements, differences of means by t-test, chi-square, and
odds ratio, as well as other appropriate statistical tests.
Concerning statistical inferential analyses, simple linear
regression, and generalized multivariate logistic regression
techniques, as well as their analytical protocols
11,18
were
used, evaluating the dependent variables: the presence of
migraine or another type of headache, correlated with the
HIT-6; and independent variables, such as age, sex, and
PCTS, BDI, BAI, PSS and Pittsburgh scales.
Results
The sample was stratied according to the International
Classication of Headache Disorders criteria, third
edition
11
, based on the denition of two groups. The rst,
classifying the participants with non-migraine headache,
and the second, classifying those with migraine (Figure 1).
Of the 360 participants, 20 losses were recorded due
to inadequate completion of scales, leaving out 340
participants. Of these, 52/340 (15.25%) did not suffer
from headaches. Finally, 288/340 (84.7%) participants
that reported headaches were included in this study, of
which 133/288 (46.2%) had migraine, and 155/288
(53.8%) had non-migraine headache (Figure 1).
Women who had migraine attacks were those with the
most intense anxious symptoms, worse sleep quality, and
catastrophization (Table 1).
Table 2 shows the logistic regression for the occurrence
of migraine between headache-affected individuals.
It is observed that, among all variables in the model,
catastrophization has a more signicant impact on the
changing values of probability of migraine, with an increase
of 5.78 percentage points at every unit variance, with the
other covariates remaining constant, with a statistically
robust estimated association (odds ratio) with migraine.
Depression and perceived stress were covariates that had
a positive association with the occurrence of migraine. On
the other hand, the interaction between catastrophization
and perceived stress expressed a negative coefcient for
a higher probability of occurrence of migraine.
The linear regression in Table 3 shows the association
between the predictive variable, catastrophization, and
other covariates, considering the absence and presence
of migraine. It is noted that each unit variation in the
catastrophization scale (PCTS) increases the impact of
headache (HIT-6), the intensity of depressive (BDI) and
anxiety symptoms (BAI), perceived stress (PSS) and sleep
quality deterioration (Pittsburg Scale). Graphically, when
the adjusted regression lines between migraine and non-
migraine individuals are observed, differences are noted
between the two situations, showing a more signicant