Panic disorder in primary care: Comorbid psychiatric disorders and their persistence

Abstract

Objective. Although 70–80% of panic disorder patients use primary care to obtain mental health services, relatively few studies have examined panic patients in this setting. This study aimed to examine both the lifetime and current comorbid psychiatric disorders associated with panic disorder in primary care, the duration and severity of the disorder, and the sociodemographic factors associated with it. Design. Patients were screened for panic disorder. Panic disorder and the comorbid disorders were determined using the Structured Clinical Interview for DSM-IV Axis I and II. Setting. Eight different health care centers in primary care in the city of Espoo. Subjects. Finnish-speaking, between 18 and 65 years of age. Main outcome measures. Comorbid psychiatric disorders, the duration and severity of the disorder, and the sociodemographic factors. Results. A sample of 49 panic disorder patients and 44 patients with no current psychiatric diagnosis were identified; 98% of panic disorder patients had at least one comorbid lifetime DSM-IV Axis I disorder. Major depressive disorder and other anxiety disorders were most common comorbid disorders. Lifetime alcohol use disorders also showed marked frequency. Interestingly, the remission rates of alcohol use disorders were notable. The panic symptoms appeared to persist for years. Panic disorder was associated with low education and relatively low probability of working full time. Conclusions. Also in primary care panic disorder is comorbid, chronic, and disabling. It is important to recognize the comorbid disorders. High remission rates of comorbid alcohol use disorders encourage active treatment of patients also suffering from these disorders.

Key Words::

• Comorbid disorders
• Finland
• general practice
• panic disorder
• primary care

There are only a few studies considering the comorbid psychiatric disorders associated with panic disorder in primary care and no earlier study had examined the prevalence of all other psychiatric disorders, both lifetime and current, using a structured diagnostic interview method.

• In this study 98% of panic disorder patients had at least one comorbid lifetime psychiatric disorder.

• Major depressive disorder, other anxiety disorders, and alcohol use disorders were the most common comorbid disorders.

• The panic symptoms appeared to persist for years.

Introduction

The course, psychiatric comorbidity, and disability associated with panic disorder have been examined extensively in large epidemiological community studies [1] and in psychiatric settings [2–4]. However, 70–80% of panic patients use primary care to obtain mental health services [5]. Considering this, relatively few studies have examined panic patients in primary care settings.

The psychiatric comorbidity associated with panic disorder in primary care has been examined in only a few studies. Most of the studies [6–10] have examined only either lifetime or current disorders and the Structured Clinical Interview for DSM-IV (SCID) or parts of it has been used in only a couple of studies [8,10]. There is no earlier study using the whole Structured Clinical Interview for DSM-IV and reporting both the lifetime and current comorbid psychiatric disorders. Also the comorbid personality disorders have not been studied in primary care settings. However, in the earlier studies the majority of panic disorder patients have had at least one comorbid psychiatric disorder reflecting the importance of comorbidity in this setting also [6,7,9,10,11].

There are also only a few studies considering the sociodemographic factors associated with panic disorder and the severity and course of the disorder in primary care. In these studies panic disorder patients had significantly less education and were significantly less likely to be employed than controls [11]. The panic disorder severity was rather mild [7]. Panic disorder with agoraphobia was characterized by a chronic course of illness with very few episodes of recovery [10]. Panic disorder without agoraphobia had quite high rates of recovery, but the course of the disease was fluctuating with high recurrence rates [10].

The aim of this study was to examine the comorbid psychiatric disorders associated with panic disorder in primary care using the Structured Clinical Interview for DSM-IV Axis I and II (SCID I and II) [12,13] and determining both the lifetime and current disorders to evaluate the permanence of the comorbid psychiatric disorders. We also aimed to examine the duration and severity of the disorder and the sociodemographic factors associated with it.

Material and methods

Participants and procedure

The study was conducted in primary care in the city of Espoo, at eight different health care centers (all health care centers in Espoo except one, which was currently under renovation), where all the physicians participated in the recruitment of the patients. The study protocol was approved by the Ethical Committee of the Helsinki and Uusimaa Hospital District and the Social and Health Services of the city of Espoo. The participants were Finnish-speaking, between 18 and 65 years of age, and scheduled for a medical appointment on the day of recruitment. The informed consent was obtained from all. The recruitment period lasted for 10 weeks. The recruitment process is described in detail in our earlier paper concerning testing and validating the screens [14]. A sample of 333 subjects completed the screening methods: the Autonomic Nervous System Questionnaire (ANS) [15], the Brief Patient Health Questionnaire (BPHQ) [16], and the General Health Questionnaire-12 (GHQ-12) [17]. The screening positivity was defined based on the recommendations given in earlier papers about testing the screens in primary health care [15,16]. All patients screening positive for panic disorder (n = 141), participated in a confirmatory diagnostic telephone interview within two weeks after screening: the panic disorder module of the SCID I. A total of 93 patients proved to have a lifetime panic disorder. As we aimed to study those patients with current panic disorder receiving psychiatric treatment in primary health care we excluded 32 patients from the study sample, because they had either received treatment in specialized mental health care during past year or had been in remission for one year or more. The remaining 61 patients agreed to participate in a face-to-face diagnostic interview consisting of SCID I and II. We excluded one patient because she was diagnosed with schizophrenia. Eleven patients repeatedly cancelled or failed to show up for their appointments. This led to a sample of 49 patients with current panic disorder: 20 patients with panic disorder only (PD) and 29 patients with panic disorder and agoraphobia (PD+ AG).

We managed to contact 84 of those 192 patients who screened negative for panic disorder: 25 declined to participate and six patients had lifetime panic disorder. The exclusion criterion for the comparison group with no psychiatric disorders was scoring 4 or more on the GHQ-12, which Holi et al. found to be the optimal cut off-point for psychiatric illness when testing the screen in a Finnish population [18]; scoring 15 or more (men) or 13 or more (women) on the Alcohol Use Disorders Identification Test (AUDIT), as Rubinsky et al. found that over 75% of primary care patients with these AUDIT scores met standardized interview criteria for past-year alcohol dependence [19]; and scoring 14 or more on the Beck Depression Inventory (BDI), which is the recommended cut off-point for screening for major depressive disorder [20]. We excluded six patients and three patients failed to return the measures used in the study. This led to a comparison group of 44 patients. The comparison group was needed to test the differences in sociodemographic characteristics between the panic disorder patients and primary care patients with no current psychiatric disorders. The mean age of the whole sample (n = 93) was 43.4 years (SD = 14.7). About 75% of the subjects were female.

Measures

The diagnostic instrument used to determine panic disorder and the comorbid psychiatric disorders was the Structured Clinical Interview for DSM-IV Axis I and II (SCID I and II). The Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I) is a semi-structured interview for making the major DSM-IV Axis I diagnoses. The SCID-II is a semi-structured interview for making DSM-IV Axis II: Personality Disorder diagnoses. The instrument is designed to be administered by a clinician or trained mental health professional. It is widely tested and considered to be a valid and reliable instrument in diagnosing psychiatric disorders. We determined the age at onset of panic attacks during the interview. The measures used to determine panic disorder severity were the Beck Anxiety Inventory (BAI), which is a self-administered measure for anxiety severity, with 21 symptoms of anxiety each with a severity scale between 0 and 3 [21]; and the Panic Disorder Severity Scale (PDSS), a seven-item interview-based scale, which was developed to evaluate the panic disorder severity on the basis of a systematic evaluation of DSM-IV criteria for panic disorder and has good reliability and validity [22]. The information concerning the sociodemographic factors was collected using a self-report questionnaire developed for this study.

Statistical analyses

Statistical analyses were performed using PASW Statistics 18.0 software (SPSS Inc., Somers, NY, USA). We used cross-tabulation and chi-squared analyses, Related-Samples McNemar Test and Independent Samples t-test, when appropriate. We analyzed the changes in the frequencies of the comorbid psychiatric disorders and thus the persistence of the disorders using the Related-Samples McNemar Test.

Results

Comorbid psychiatric disorders

Some 98% of the panic disorder patients had at least one comorbid lifetime DSM-IV Axis I disorder (Table I). Major depressive disorder was the most common comorbid psychiatric disorder. Over 70% of panic disorder patients had another lifetime anxiety disorder and over 40% had lifetime alcohol use disorder. Other lifetime substance use disorders had a notable frequency (21%) in the PD+ AG-group.

Table I. Lifetime and current DSM-IV comorbid disorders.

	Lifetime comorbid disorders			Current comorbid disorders
	Panic disorder without agoraphobia (n = 20) n (%)	Panic disorder with agoraphobia (n = 29) n (%)	All (n = 49) n (%)	Panic disorder without agoraphobia (n = 20) n (%)	Panic disorder with agoraphobia (n = 29) n (%)	All (n = 49) n (%)
Any other psychiatric disorder	19 (95.0)	29 (100)	48 (98)	13 (65.0)	25 (86.2)	38 (77.6)
Major depressive disorder	11 (55.0)	20 (69.0)	31 (63.3)	1 (5.0)^1	8 (27.6)^1	9 (18.4)^1
Dysthymia	–	–	–	0 (0)	1 (3.4)	1 (2.0)
Other depressive disorder	1 (5.0)	4 (13.8)	5 (10.2)	0 (0)	0 (0)	0 (0)
Bipolar disorder	2 (10.0)	2 (6.9)	4 (8.2)	2 (10.0)	2 (6.9)	4 (8.2)
Social phobia	3 (15.0)	17 (58.6)	20 (40.8)	3 (15.0)	13 (44.8)	16 (32.7)
GAD	–	–	–	4 (20.0)	3 (10.3)	7 (14.3)
OCD	1 (5.0)	6 (20.7)	7 (14.3)	0 (0)	4 (13.8)	4 (8.2)
PTSD	1 (5.0)	6 (20.7)	7 (14.3)	0 (0)	3 (10.3)	3 (6.1)
Specific phobia	5 (25.0)	15 (51.7)	20 (40.8)	5 (25.0)	14 (48.3)	19 (38.8)
Any other anxiety disorder	10 (50.0)	25 (86.2)	35 (71.4)	9 (45.0)	21 (72.4)	30 (61.2)
Somatoform disorders	1 (5.0)	3 (10.3)	4 (8.2)	1 (5.0)	2 (6.9)	3 (6.1)
Eating disorders	1 (5.0)	1 (3.4)	2 (4.1)	0 (0)	0 (0)	0 (0)
Psychotic disorders	0 (0)	1 (3.4)	1 (2.0)	0 (0)	0 (0)	0 (0)
Alcohol disorders	7 (35.0)	14 (48.3)	21 (42.9)	0 (0)^2	3 (10.3)^1	3 (6.1)^1
Other substance use disorders	1 (5.0)	6 (20.7)	7 (14.3)	1 (5.0)	2 (6.9)^3	3 (6.1)^3

Notes: Statistical significance of the change in the frequency of the comorbid disorder:

¹p < 0.005; ²p < 0.05; ³p = 0.125.

Other anxiety disorders were found to be the most persistent comorbid psychiatric disorders: none of them showed a statistically significant decrease in frequency (see Table I). Over two-thirds of the patients with a lifetime major depressive disorder were currently in remission. The remission rate of alcohol use disorders was also notable. Also the frequency of other substance use disorders showed some decrease in the PD+ AG-group; however, the change was not statistically significant.

In total, 15% of patients in the PD group and 24.1% in the PD+ AG group had comorbid personality disorders. The most common personality disorders were antisocial and borderline personality disorders (Table II).

Table II. Comorbid personality disorders.

	Panic disorder without agoraphobia (n = 20) n (%)	Panic disorder with agoraphobia (n = 29) n (%)	All (n = 49) n (%)
Any personality disorder	3 (15.0)	7 (24.1)	10 (20.4)
Antisocial	1 (5.0)	2 (6.9)	3 (6.1)
Borderline	1 (5.0)	2 (6.9)	3 (6.1)
Avoidant	0 (0)	1 (3.4)	1 (2.0)
Dependent	0 (0)	1 (3.4)	1 (2.0)
Obsessive-compulsive	1 (5.0)	0 (0)	1 (2.0)
Mixed (antisocial + borderline)	1 (5.0)	1 (3.4)	2 (4.1)

Panic disorder duration and severity

The mean age at onset of panic disorder was 29.2 in the PD group and 22.4 in the PD+ AG group. The difference was statistically significant (p < 0.05). The mean duration of the disorder at the study point was 8.9 years in the PD group and 21.1 years in the PD+ AG group. Here also the difference between the groups was statistically significant (p < 0.005).

The BAI total score in both subgroups reflected moderate severity of anxiety, whereas the PDSS composite scores reflected mild severity of the disorder for the PD group and moderate severity for the PD+ AG group (Table III). Both subgroups had mild panic frequency (less than one full-blown attack per week and maximum one limited attack per day). However, the distress during the attacks was severe (loses concentration and has to cease activity) in both subgroups. The agoraphobic fear and avoidance was moderate in the PD+ AG groups, as also was work impairment and distress.

Table III. Clinical severity of the disorder (Panic Disorder Severity Scale and Beck Anxiety Inventory).

	Panic disorder without agoraphobia (n = 20) mean	Panic disorder with agoraphobia (n = 29) mean	All (n = 29) mean
Panic frequency	1.30 (mild)	1.17 (mild)	1.22 (mild)
Panic distress	2.90 (severe)	2.62 (severe)	2.73 (severe)
Anticipatory anxiety	1.15 (mild)	1.45 (mild)	1.33 (mild)
Agoraphobic fear/avoidance	1.10 (mild)	1.69 (moderate)	1.45 (mild)
Interoceptive fear/avoidance	0.80 (mild)	1.14 (mild)	1.00 (mild)
Work impairment/distress	1.05 (mild)	1.52 (moderate)	1.33 (mild)
Social impairment/distress	0.60 (mild)	1.34 (mild)	1.04 (mild)
Total score	8.75	10.93	10.04
Composite (average)	1.27 (mild)	1.56 (moderate)	1.44 (mild)
BAI (total)	21.12 (moderate)	25.46 (moderate)	23.74 (moderate)

Sociodemographic characteristics

The three groups (PD, PD+ AG, and no current diagnosis) did not differ in age or gender. The patients with panic disorder were significantly less educated (Table IV). Only about one-third of panic disorder patients had finished high school, whereas about 70% of the patients with no current diagnosis had. Some 40% of patients with PD and over half of patients with PD+ AG had no professional education, whereas almost 90% of patients in the comparison group were professionally qualified. The difference in the percentile of patients working or studying full time was also statistically significant.

Table IV. Sociodemographic data.

	Panic disorder without agoraphobia (n = 20)	Panic disorder with agoraphobia (n = 29)	No current diagnosis (n = 44)	chi-square	p
Age, years	38.10	43.52	45.77	–	–
Female	13 (65.0)	20 (69.9)	38 (86.4)	4.745	0.093
Married/cohabiting	10 (50.0)	15 (51.7)	32 (72.7)	2.975	0.226
Completed high school	6 (30.0)	11 (37.9)	31 (70.5)	9.284	0.010
Professionally qualified	12 (60.0)	14 (48.3)	39 (88.6)	10.660	0.005
Working or studying full time	9 (45.0)	10 (34.5)	29 (65.9)	6.141	0.046

Discussion

This study shows that panic disorder is also a comorbid and chronic disorder in primary care. The remission rates for some comorbid disorders (especially major depressive disorder and alcohol use disorders) are marked; the comorbid anxiety disorders are persistent. Panic disorder is associated with low education and relatively low probability of working full time.

The strength of our study is utilization of the Structured Clinical Interview for DSM-IV Axis I and II for both lifetime and current comorbid psychiatric disorders. To our knowledge this is the first study to report both lifetime and current comorbid psychiatric disorders and also the personality disorders associated with panic disorder in primary care. The weakness of our study is the medium sample size, which weakens the generalizability of the results. As the primary care physicians recruited the patients to the study, we do not know the exact number of patients who were initially contacted and who refused to take part in the study, and were not able to analyze the reasons for the refusal. Thus it is not possible to definitively exclude a bias caused by the recruitment process. However, the essential findings (namely, the high percentile of comorbid psychiatric disorders and the order of the most common comorbid disorders) are in line with previous studies [6,7,9,10,11], which is suggestive of more general importance of the findings.

To our knowledge, this is the only study so far to evaluate the persistence of comorbid psychiatric disorders. One other primary care study reported the frequencies of both lifetime and current comorbid major depressive disorder; there was also a clear decrease observed [11]. This is expected considering the typically remitting and relapsing course of major depressive disorder. The most persistent comorbid disorders in this study were other anxiety disorders. This is also in line with the chronic course observed for anxiety disorders in follow-up studies [2,3]. The high remission rate we found for alcohol use disorders is notable, as in clinical practice the patients with comorbid alcohol use disorders are easily regarded as having a poorer prognosis.

In this study 15.0% of patients with panic disorder and 24.1% of patients with panic disorder and agoraphobia had at least one personality disorder. This rate is lower than reported in earlier studies evaluating the frequencies of personality disorders associated with panic disorder [4,23]. However, these studies have been carried out in specialized mental health care settings, where more severely disturbed patients are typically referred.

Panic disorder had a long-lasting course in this study. We do not know the possible fluctuation during the course of the disorder but all the patients still fulfilled the DSM-IV criteria for panic disorder at intake. The chronic course of the disorder, and specifically of PD+ AG, is in line with previous studies. In the only follow-up study in primary care so far, only one-quarter of patients with PD+ AG had remission during the three-year follow-up. Three-quarters of the patients with PD had remission; however, there was no difference in the likelihood of the disorders in experiencing recurrence [10]. The follow-up studies carried out in specialized mental health care settings [2,3] have also shown a chronic clinical course of panic disorder.

There is one other study reporting the Panic Disorder Severity Scale scores of primary care patients [7]. The total scores measured are pretty close to the scores in our study, reflecting similar severity. In both studies panic attack frequency was rather low, but the attack intensity was severe. The rather mild severity of the disorder may also reflect its long duration: during the course of the disorder the patients find their own ways to adapt to and cope with the disorder.

In our study the patients with panic disorder had significantly less education and were significantly less likely to work full time than the controls. Two earlier studies report similar findings [10,11]. Considering the age at onset of the disorder one could appraise the lower education and lower probability of working full time rather as the consequence than as the predecessor of the disorder.

This study highlights the importance of recognition of the comorbid psychiatric disorders associated with panic disorder in primary care. High remission rates of comorbid alcohol use disorders encourage active treatment of patients suffering from these disorders also. Based on this study, panic disorder could be seen as an easily recognized marker of an individual's vulnerability and predisposition to a wide range of mental disorders. Panic disorder and the comorbid psychiatric disorders also seem to have a vast impact on the affected individual's social abilities. Thus the early recognition of panic disorder is particularly important. The chronic disease self-management approach recommended for enhancing the treatment of panic disorder in primary care [24] is warranted based on this study too. The high numbers of comorbid disorders, which have also been found with depressive disorders in primary care [25], emphasize the importance of cooperation and stepped care with specialized mental health care.

Acknowledgements

This work was financially supported by the Finnish Psychiatric Association, Yrjo and Tuulikki Ilvonen Foundation, Oy H. Lundbeck Ab and the Graduate School of Psychiatry. The funders had no involvement in the conduct or reporting the study.

Declaration of interest

Dr Suominen has received honoraria for lecturing and participating in planning of educational meetings sponsored by the pharmaceutical industry (AstraZeneca, BristolMyersSquibb, Pfizer, Lundbeck), has been a member of the Advisory Boards of AstraZeneca, GlaxoSmithKline, and Servier in Finland, has received reimbursement for attending conferences from BristolMyersSquibb, Janssen-Cilag, Lundbeck, Pfizer, and AstraZeneca, and has participated as an investigator in clinical drug trials sponsored by AstraZeneca and Pfizer. Dr Karlsson has received honoraria for lecturing and participating in planning of an educational meeting sponsored by the pharmaceutical industry (Lundbeck). Dr Tilli has no conflict of interest.