These authors contributed equally to this work.
Several predictors of unfavorable pharmacological treatment response (PTR) in panic disorder (PD) patients have been suggested, such as the duration of the illness, presence of agoraphobia, depression, being a woman, and early trauma. This study aimed to examine whether pathological worry is associated with PTR in PD patients.
This study included 335 PD patients and 418 healthy controls (HCs). The Penn State Worry Questionnaire (PSWQ), the Early Trauma Inventory Self Report-Short Form (ETISR-SF), Beck Depression Inventory (BDI), Panic Disorder Severity Scale (PDSS), and Anxiety Sensitivity Inventory-Revised (ASI-R) were administered. We measured the PTR at 8 weeks and 6 months. Student t-test, chi-square tests, Pearson’s correlation analyses, and binary logistic regression model were used.
Our results showed that the total scores of the PSWQ correlated with the ETISR-SF, BDI, and ASI-R were significantly higher in patients with PD compared with HCs. The PSWQ and BDI could predict unfavorable PTR at 6 months in PD patients.
This is the first study to demonstrate that pathological worry may contribute to poor long-term PTR in PD patients. Therefore, our research suggests that clinicians must be aware of worry to optimize PTR for PD patients.
Generally, worry, which is defined as “a chain of thoughts and images, negatively affect-laden and relatively uncontrollable,” is a cognitive characteristic of anxiety that can be seen in normal individuals [
Panic disorder (PD) is an anxiety disorder whose main symptoms are recurrent and unexpected panic attacks, excessive worry about future panic attacks, and physical symptoms such as the sensation of shortness of breath, palpitation, and fear of dying [
To reduce symptoms like pathological worry in patients with PD, pharmacological treatment and second wave cognitive-behavioral therapy (CBT) have been well-established so far as effective methods [
Several baseline predictors that could contribute to the PTR in patients with PD have been suggested. Previous short-term as well as long-term research has suggested several predictors in PD such as duration of the illness, the symptom severities of PD, presence of agoraphobia, comorbidities with other psychiatric disorders including depressive disorders and personality disorders, and the female gender [
Our study addresses this gap in the literature by examining whether pathological worries are associated with long-term PTR in patients with PD using a relatively larger sample size than those used in previous studies. We hypothesized that patients with PD who experience pathological worry will have an unfavorable response to pharmacological treatments in long-term follow-ups.
Between 2011 and 2021, 335 patients with PD and 418 healthy controls (HCs) participated in this study. We recruited participants from patients with PD who were treated in the Department of Psychiatry at CHA Bundang Medical Center. All participants were Koreans aged between 18 and 70 years old. The participants’ family histories of PD were established through interviews. Participants with PD with or without agoraphobia met the DSM-IV-TR Axis I Disorder, as diagnosed by trained psychiatrists using the Structured Clinical Interview for DSM-IV-TR. Although DSM-5 was published in 2013, we recruited patients with PD with the previous criterion, DSM-IV-TR, to collect data on a consistent basis from 2011 to 2021. Only patients primarily diagnosed as experiencing PD were included, and PD patients with additional major medical comorbidities were excluded. In addition, PD patients who followed up to 6 months after pharmacotherapy were included.
Participants with a primary diagnosis of any schizophrenia spectrum disorders, bipolar disorders, depressive disorders, anxiety disorders including GAD other than PD, substance use disorders, personality disorder, mental retardation, major medical disorders including neurological disorders, and pregnancy were excluded. Furthermore, all patients with PD who had been treated with individual or group psychotherapy such as mindfulness-based cognitive therapy or CBT were excluded. In addition to, HCs with a history of psychiatric disorders were excluded.
All patients with PD had pharmacotherapy with SSRIs such as escitalopram, paroxetine, and sertraline (escitalopram equivalence dosage=9.97±7.50 [mean±SD] mg/day) [
All study procedures complied with to the Institutional Review Board regulations and principles of Good Clinical Practice at the CHA Bundang Medical Center. After participants were provided with an enough explanation of the study process, methods, and purpose, their written informed consent was obtained.
To evaluate pathological worry in patients with PD in clinical settings, the Penn State Worry Questionnaire (PSWQ) is used [
Early trauma was evaluated using the Korean version of the Early Trauma Inventory Self Report-Short Form (ETISR-SF) with a Cronbach’s α of 0.869. ETISR-SF is significantly positively associated with the scores on the Childhood Trauma Questionnaire-Short Form (r=0.691) [
The Korean version of the Anxiety Sensitivity Inventory-Revised (ASI-R) to assess potential anxiety trait markers in patients with PD at baseline [
To measure the clinical severity of the participants’ anxiety and depressive symptoms at baseline in PD, we performed the Panic Disorder Severity Scale (PDSS) [
The PTR was evaluated after a minimum of 8 weeks and 6 months of pharmacotherapy in an adequate dose. It was defined as a total PDSS score reduction of 40% or more from the baseline score after 8 weeks and 6 months of treatment in an adequate dose [
To analyze the sociodemographic characteristics and clinical symptom severities including the PSWQ of patients with PD and HCs, the Student t-test and chi-square tests were performed.
In addition, Pearson’s correlation analyses were applied to determine whether an association existed among continuous variables such as age, ETISR-SF, ASI-R, BDI, PDSS, and PSWQ at baseline.
Further, a binary logistic regression model with treatment response as the dependent variable and with those that can influence the treatment response at 8 weeks and in 6 months as covariates was performed in PD. All statistical analyses used the IBD SPSS Statistics 26.0 software (IBM Corp., Armonk, NY, USA). All reported probability values were two-tailed where p<0.05 was considered statistically significant.
The sociodemographic and clinical characteristics of all study participants are summarized in
The total scores of the PSWQ at baseline were significantly higher in patients with PD compared with HCs {PD: 52.73 (±12.68) (mean [±SD]), HCs: 38.64 (±9.14), t=14.14, p<0.001} (
The mean PSWQ scores of patients with PD did not significantly differ by gender, level of education, monthly income, and marital status (all, p>0.05). Furthermore, there was no statistical difference in the mean PSWQ scores by gender, level of education, and marital status in HCs (all, p>0.05). However, in HCs, the mean PSWQ scores differed significantly in monthly income levels, and the lower the monthly income, the higher the PSWQ scores (Fisher’s exact test, p=0.01).
In patients with PD, the total PSWQ scores at baseline showed significantly positive correlations with the total sum of all subtypes of ETISR-SF, the emotional early trauma scores, ASI-R, BDI, and PDSS at baseline, at 8 weeks, and 6 months (all, p<0.001). However, there were significant negative associations between the PSWQ scores at baseline and age (p<0.01) (
To evaluate which factors contributed to the response of pharmacological treatment at 6 months, a binary logistic regression analysis was performed (
After 6 months, the PSWQ at baseline (B=-0.072, p=0.018) and BDI (B=-0.089, p=0.047) were significantly negatively associated with the response of pharmacological treatment in patients with PD. In particular, higher total scores of PSWQ reduced the possibility of PTR in 6 months (OR=0.930, 95% CI=0.876–0.988). However, the monthly income and the total score of PDSS at baseline were positively significantly associated with the PTR in patients with PD in 6 months. Higher PDSS scores were associated with a significantly better PTR in 6 months (OR=0.284, 95% CI=1.158–1.524).
This research model proved to be significant at p<0.05, with the
This is the first study to demonstrate that pathological worry can influence the poor long-term PTR in patients with PD. The findings of our study suggest that patients with PD with pathological worry are more vulnerable and show more severe symptoms, which may lead to a chronic course of illness. Therefore, our study shows that it is important for clinicians to check for these features at an early stage in patients with PD in initial interviews and to choose appropriate treatments for such individuals.
Previous studies showed the several risk factors contributing to worse PTR of patients with PD [
From a psychological perspective, pathological worry is associated with coping strategies such as attempts to avoid negative events, interpersonal control, and cognitive failures. In the cognitive model of worry process presented by Eysenck [
Our findings have shown a positive association between pathological worry and secondary depressive symptoms. In addition, it has shown that the severity of secondary depressive symptoms in PD were correlated with unfavorable long-term treatment responses, although we excluded participants with primary major depressive disorder consistent with prior literature [
In addition, our findings have shown that high symptom severity of PD at baseline in patients with PD was associated with good long-term PTRs inconsistent with prior literature [
Furthermore, we found that the AS associated with pathological worry in patients with PD does not significantly directly influence the PTRs at long-term follow-ups. Previous studies have suggested AS as a predictor of panic-related pathology, which may affect the frequency of panic attacks [
Notably, previous studies showed an correlation between early trauma and the frequency of panic attacks or the age of onset of disease in patients with PD [
Our findings showed that the higher the patient’s income level, the better the long-term PTR. We assume that individuals with a higher level of income generally tend to show less symptoms and disability, resulting in better PTR after long-term follow-up, which is consistent with a previous study [
In addition, CBT effectively reduced pathological worries with a large overall effect size showing improvement following treatment at the 6-month follow-up in patients with GAD [
Our study has several limitations. First, it is believed that in addition to self-reported evaluation measures such as PSWQ, objective evaluation measures evaluated by clinicians are necessary. Second, recall bias can happen when evaluating early trauma. Third, we recruited patients with PD in an acute care hospital, some of which might have depressive episodes in the long-term follow-up, as happens in the natural course of PD. Therefore, the possibility cannot be ruled out that the sampling bias influenced the response to long-term pharmacological treatment.
In conclusion, the current findings suggest that the symptoms of pathological worries could be associated with poor long-term PTRs in patients with PD. Therefore, our study highlights the need to detect the symptoms of pathological worry early to determine the direction of medical treatment and predict the results of pharmacotherapy for patients with PD. The optimization of pharmacological treatments and CBT will be necessary for individuals with PD in the future.
The datasets generated or analyzed during the study are not publicly available because all participants did not consent their information exposure.
The authors have no potential conflicts of interest to disclose.
Conceptualization: all authors. Data curation: all authors. Formal analysis: all authors. Funding acquisition: Sang-Hyuk Lee. Investigation: all authors. Methodology: all authors. Project administration: all authors. Resources: all authors. Software: all authors. Supervision: Sang-Hyuk Lee. Validation: all authors. Writing—original draft: all authors. Writing—review & editing: all authors.
This work was supported by the Basic Science Research Program through the National Research Foundation of Korea which is funded by the Ministry of Education, Science and Technology, by the Ministry of Science and ICT of the Government of South Korea (2018-R1D1A1B07046978) and (NRF-2019M3C7A1032262) to S.H. Lee.
We thank all members of the Clinical Laboratory for Integrative studies in Mindfulness and Brain imaging for their kindness and its support.
Socio-demographic and clinical characteristics of patients with panic disorder and healthy controls
PD (N=335) | HCs (N=418) | Statistics (t or χ2) | p-value | ||
---|---|---|---|---|---|
Age at baseline (years, mean±SD) | 37.91±11.41 | 36.84±11.31 | 1.29 | 0.199 | |
Gender (male [%]/female) | 149 (44.5)/186 | 195 (46.7)/223 | 0.35 | 0.552 | |
Years of education (high school or less [%]/college or more) | 174 (53.0)/154 | 76 (19.3)/318 | 90.12 | <0.001 |
|
Marital status (living without partner [%]/with partner) | 136 (41.1)/195 | 190 (50.3)/188 | 5.98 | 0.014 |
|
Monthly income (below 1,800 $USD [%]/above 1,800 $USD) | 33 (10.2)/290 | 14 (4.2)/322 | 9.10 | 0.003 |
|
Agoraphobia, yes (%) | 242 (73.8) | - | - | - | |
Duration of illness (months, mean±SD) | 56.05±43.61 | - | - | - | |
Baseline ETISR-SF total score (mean±SD) | 4.96±4.00 | 3.28±3.01 | 4.04 | <0.001 |
|
General trauma sub-scale score | 1.34±1.49 | 0.89±1.15 | 2.89 | <0.001 |
|
Physical trauma sub-scale score | 1.89±1.62 | 1.58±1.65 | 1.63 | 0.104 | |
Emotional trauma sub-scale score | 1.37±1.69 | 0.66±1.19 | 4.21 | <0.001 |
|
Sexual trauma sub-scale score | 0.39±0.88 | 0.16±0.41 | 2.95 | 0.001 |
|
Baseline ASI-R total score (mean±SD) | 52.03±29.79 | 8.23±10.47 | 24.76 | <0.001 |
|
Fear of respiratory symptom | 19.76±11.80 | 1.51±2.82 | 27.24 | <0.001 |
|
Fear of cardiovascular symptom | 14.14±9.74 | 2.80±4.98 | 18.98 | <0.001 |
|
Fear of publicly observable anxiety reaction | 11.06±7.54 | 3.35±4.27 | 16.39 | <0.001 |
|
Fear of cognitive dyscontrol | 6.99±6.97 | 0.84±1.76 | 15.58 | <0.001 |
|
Baseline BDI total score (mean±SD) | 17.77±10.83 | 6.05±6.28 | 17.32 | <0.001 |
|
PDSS total score at baseline (mean±SD) | 12.33±6.33 | - | - | - | |
PDSS total score at 8 weeks (mean±SD) | 9.89±5.01 | - | - | - | |
PDSS total score at 6 months (mean±SD) | 8.85±4.77 | - | - | - | |
Pharmacotherapy, yes (%) | 244 (72.8) | ||||
Kinds of SSRI {escitalopram (N [%])/paroxetine (N [%])} | 71 (29.1)/149 (61.1) | - | - | - | |
SSRI equivalent dosage (mg, mean±SD) |
9.97±7.50 | - | - | - |
p<0.05;
p<0.01;
p<0.001;
the approximate equivalent oral doses to 10 mg Escitalopram are given.
PD, panic disorder; HCs, healthy controls; ETISR-SF, Early Trauma Inventory Self Report-Short Form; ASI-R, Anxiety Sensitivity Inventory-Revised; BDI, Beck Depression Inventory; PDSS, Panic Disorder Severity Scale; SSRI, Selective Serotonin Re-uptake Inhibitor
Pearson’s correlations among continuous variables in healthy controls
Continuous variables | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
---|---|---|---|---|---|---|---|---|---|
1PSWQ at baseline | - | ||||||||
2Age | -0.181 |
- | |||||||
ETISR-SF | |||||||||
3Total sum of all subtypes | 0.179 |
-0.034 | - | ||||||
4General | -0.080 | 0.160 |
0.647 |
- | |||||
5Physical | 0.075 | -0.069 | 0.795 |
0.275 |
- | ||||
6Emotional | 0.367 |
-0.125 | 0.673 |
0.213 |
0.292 |
- | |||
7Sexual | 0.179 |
-0.056 | 0.381 |
0.208 |
0.141 | 0.221 |
- | ||
8ASI-R at baseline | 0.324 |
0.017 | 0.218 |
0.133 | 0.096 | 0.255 |
0.037 | - | |
9BDI at baseline | 0.472 |
0.072 | 0.326 |
0.176 |
0.137 | 0.416 |
0.142 | 0.370 |
- |
p<0.05;
p<0.01.
PSWQ, Penn State Worry Questionnaire; ETISR-SF, Early Trauma Inventory Self Report-Short Form; ASI-R, Anxiety Sensitivity Index-Revised; BDI, Beck Depression Inventory
Pearson’s correlations among continuous variables in patients with panic disorder
Continuous variables | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
---|---|---|---|---|---|---|---|---|---|---|---|---|
1PSWQ at baseline | - | |||||||||||
2Age | -0.178 |
- | ||||||||||
ETISR-SF | ||||||||||||
3Total sum of all subtypes | 0.274 |
-0.313 |
- | |||||||||
4General | 0.108 | -0.153 | 0.708 |
- | ||||||||
5Physical | 0.149 | -0.227 |
0.717 |
0.306 |
- | |||||||
6Emotional | 0.335 |
-0.302 |
0.817 |
0.411 |
0.429 |
- | ||||||
7Sexual | 0.065 | -0.072 | 0.450 |
0.171 |
0.076 | 0.302 |
- | |||||
8ASI-R at baseline | 0.600 |
-0.257 |
0.440 |
0.283 |
0.223 |
0.451 |
0.221 |
- | ||||
9BDI at baseline | 0.682 |
-0.202 |
0.350 |
0.183 |
0.209 |
0.389 |
0.141 | 0.654 |
- | |||
10PDSS at baseline | 0.378 |
-0.235 |
0.170 |
0.148 | 0.131 | 0.153 | -0.025 | 0.579 |
0.538 |
- | ||
11PDSS at 8 weeks | 0.399 |
-0.271 |
0.235 |
0.170 |
0.197 |
0.183 |
0.027 | 0.566 |
0.547 |
0.820 |
- | |
12PDSS at 6 months | 0.417 |
-0.285 |
0.247 |
0.123 | 0.257 |
0.202 |
0.037 | 0.529 |
0.551 |
0.753 |
0.883 |
- |
p<0.05;
p<0.01;
p<0.001.
PSWQ, Penn State Worry Questionnaire; ETISR-SF, Early Trauma Inventory Self Report-Short Form; ASI-R, Anxiety Sensitivity Index-Revised; BDI, Beck Depression Inventory; PDSS, Panic Disorder Severity Scale
Binary logistic regression results predicting the pharmacological treatment response at 6 months in patients with panic disorder
Variables | Pharmacological treatment response at 6 months |
||
---|---|---|---|
B | p-value | Odds ratio (95% CI) | |
Age (years) | 0.018 | 0.467 | 1.018 (0.970–1.069) |
Gender (male) | 0.582 | 0.267 | 1.789 (0.641–4.996) |
Level of education (high school or less) | -0.381 | 0.495 | 0.683 (0.229–2.041) |
Marital status (living without partner) | -0.815 | 0.150 | 0.443 (0.146–1.344) |
Monthly income (below 1,800 $USD) | 3.854 | 0.001 |
47.166 (4.571–486.658) |
PSWQ at baseline (total sum) | -0.072 | 0.018 |
0.930 (0.876–0.988) |
BDI at baseline (total sum) | -0.089 | 0.047 |
0.915 (0.839–0.999) |
PDSS at baseline (total sum) | 0.284 | <0.001 |
1.329 (1.158–1.524) |
ASI-R at baseline (total sum) | 0.017 | 0.283 | 1.017 (0.986–1.050) |
Early trauma (ETISR-SF, total sum of all subtypes) | -0.068 | 0.346 | 0.934 (0.810–1.076) |
-2 log likelihood | 108.163 | ||
χ2 (df) | 43.959 (10) (p<0.001 |
||
Cox & Snell R2 | 0.305 | ||
Overall percentage (%) | 67.8 |
Criteria for response of pharmacological treatment in patients with panic disorder is classified as the total PDSS score of 40% or greater reduction compared to the PDSS total score at baseline.
p<0.05;
p<0.01;
p<0.001.
CI, confidence interval; PSWQ, Penn State Worry Questionnaire; BDI, Beck Depression Inventory; PDSS, Panic Disorder Severity Scale; ASI-R, Anxiety Sensitivity Inventory-Revised; ETISR-SF, Early Trauma Inventory Self Report-Short Form