Applied Psychophysiology and Biofeedback

https://doi.org/10.1007/s10484-020-09465-0

Evaluating the Impact of Freespira on Panic Disorder Patients’ Health

Outcomes and Healthcare Costs within the Allegheny Health Network
Alicia Kaplan1 · Anthony P. Mannarino1 · P. V. Nickell1

© Springer Science+Business Media, LLC, part of Springer Nature 2020

Abstract
Panic disorder (PD) is a debilitating condition that drives medical spending at least twice as high as medically matched
controls. Excessive utilization of healthcare resources comes from emergency department (ED), medications, diagnostic test-
ing, and physician visits. Freespira is an FDA-cleared digital therapeutic that treats PD and panic attacks (PA) by correcting
underlying abnormal respiratory physiology. Efficacy of Freespira has been established in prior studies. This paper reports
on a quality improvement program that investigated whether treating PD patients with Freespira would reduce medical costs
and improve outcomes over 12-months. Panic symptoms were assessed using the Panic Disorder Severity Scale (PDSS).
Pre-and post-treatment insurance claims determined costs. At baseline, mean Clinician Global Impression (CGI-S) was 4.4
(moderately/markedly ill), mean PDSS was 14.4 and mean PA frequency/week was 2 (range 0–5). Immediately post-treatment
(week 5) mean CGI-S, PDSS and weekly PA frequency declined to 2.8 (borderline/mildly ill, 4.9 (remission) and 0.2 (range
0–2) respectively, p < 0.001. 82% reported PDSS decrease of ≥ 40% (clinically significant), 86% were PA-free. One-year post
treatment mean CGI-S, PDSS and PA remained low at 2.1, 4.4, and 0.3 (range 0–1) respectively. 91% had PDSS decrease
of ≥ 40%, 73% were PA-free. The majority of patients were panic attack free and/or reduced their symptoms and avoidance
behaviors 1-year post Freespira treatment. Mean overall medical costs were reduced by 35% from $548 to $358 PMPM (per
member per month) or an annual reduction of $2280. at 12 months post-treatment. There was a 65% reduction in ED costs
from $87 to $30 PMPM. Median pharmacy costs were reduced by 68% from $73 to $23 PMPM.

Keywords  Panic disorder · Panic attacks · Hyperventilation · Freespira · CGRI

Introduction and differentiates it from the other anxiety disorders. The

cardio-respiratory symptoms of panic mimic heart disease,
Epidemiologic data suggest that 3–4 percent of American and often lead to care-seeking in emergency departments
adults suffer from panic disorder (PD), an anxiety disor- (Fleet 1996, 2003) and cardiology settings (Dammen et al.
der associated with marked impairment in social and occu- 1999). Deacon et al (2008) reported that patients with PD
pational functioning, significant impact on quality of life, visited family medicine and cardiology practices, and emer-
and high utilization of health care services (Deacon et al gency departments with greater frequency than those with
2008). Fearful interpretation of bodily symptoms such as other anxiety disorders. Barsky et al. (1999) reported that
tachycardia, shortness of breath, chest tightness, and diz- patients with PD averaged 10.6 physician visits in 1 year
ziness with catastrophic beliefs is the core of the diagnosis versus 4.4 visits for patients without PD. Data from the
National Comorbidity Survey Replication (Kessler et al.
2006) showed that 28.3% of respondents reported having
An interim analysis with fewer subjects was previously presented had at least one panic attack in their lifetime, and 11.2%
by Kaplan at the Annual Conference of Anxiety and Depression reported a panic attack in the prior year. The increased medi-
Association of America in April 2018. cal utilization rates noted in patients with PD bring about a
* Alicia Kaplan high financial burden associated with the condition. Shirme-
[email protected] shan et al. (2013) estimated that the cost of ambulatory care
of patients with anxiety disorders at $33 billion, with PD
1
Department of Psychiatry, Allegheny Health Network, 4 patients being included in this analysis. Retrospective claims
Allegheny Center 8th Floor, Pittsburgh, PA 15212, USA

13
Vol.:(0123456789)
 Applied Psychophysiology and Biofeedback

analyses by Health Lumen and Highmark Health found that recognition in PD sufferers of subtle respiratory irregulari-
patients with PD represented 2–5 times higher costs than ties associated with hyperventilation, and carbon dioxide
matched controls with PD (Highmark Health 2015). sensitivity, Meuret et al. (2008) developed a breathing inter-
Treatments for PD include pharmacotherapy, psychother- vention focused on normalizing both exhaled carbon dioxide
apy, relaxation techniques, and breathing training. First-line levels ­(ETCO2) and respiratory rate. The protocol provided
treatments for PD include the selective serotonin reuptake breath-to-breath feedback of ­ETCO2, while modeling paced
inhibitors (SSRIs) fluoxetine, paroxetine, and sertraline; the breathing at four different respiratory rates. Administered
serotonin norepinephrine reuptake inhibitor (SNRI) ven- as twice daily, 17-min sessions over a 4-week period, the
lafaxine; and cognitive behavioral therapy (CBT) which authors reported that by study end, 86% of subjects reported
focuses on desensitizing patients to feared stimuli and bod- zero weekly panic attacks; improvement that was durable
ily sensations. For more refractory patients, second-line over time, as 73% reported zero weekly attacks 1-year
medications may include Tricyclic antidepressants such as post-treatment.
imipramine and clomipramine, which often have a higher Freespira, a digital therapeutic developed and manufac-
adverse effect burden than the SSRIs and SNRI; and high tured by Palo Alto Health Sciences, Inc. (PAHS, Kirkland,
potency benzodiazepines (alprazolam, and clonazepam), WA), incorporates the protocol described above. Freespira
which carry risks of sedation, abuse and dependence. For includes a proprietary portable sensor that measures respira-
patients treated by primary care physicians, less than 25% of tory rate and exhaled C ­ O2 and a dedicated software App
diagnosed individuals received evidence-based psychother- (on a tablet) that guides the patient through the standard-
apy or medications (Roy-Byrne et al. 1999). Pharmacother- ized 17-min protocol. Freespira received FDA clearance in
apy often involves the need for continuous treatment, with 2013 for the treatment of PD, panic attack, and symptoms
high relapse rates following discontinuation. (Nardi et al. of panic.
2015), reporting on the effectiveness of pharmacotherapy Tolin et al. (2017) used the same twice-daily, 4-week pro-
for panic, noted cumulative relapse rates of 41%, 77%, and tocol and described the treatment as Capnometry Guided
94% at 1, 4, and 6 years respectively following medication Respiratory Intervention (CGRI). Patients were treated in
discontinuation (clonazepam or paroxetine) in patients who four non-academic outpatient clinics. This trial reported that
were panic free at least 1 year preceding tapered drug with- 71% of participants who completed treatment were panic
drawal. The tolerability of SSRI/SNRI’s can be problematic, attack free immediately post treatment, and 79% were panic
with discontinuation rates of 47% within 180 days of initi- attack free at 1-year post-treatment. At 1-year post treatment,
ating drug therapy (Vlahiotis et al. 2011). Shankman et al. 82% of participants achieved treatment response (defined
(2017) reported elevated side effect frequency and severity as at least 40% reduction in Panic Disorder Severity Scale
in patients with co-morbid depression and PD, related to scores). Tolin et al. additionally benchmarked this study
somatic hypersensitivity. Access to evidence-based CBT can against Meuret’s et al. (2008) and found similar reductions
be limited. Goddard (2017) reported that approximately one on the Anxiety Sensitivity Index and Sheehan Disability
third of panic sufferers gain long-term relief from symp- Scale. Both papers reported favorable side effect profiles,
toms, while the remaining majority have either chronic or limited primarily to transitory sensations of light-headedness
frequently relapsing conditions. and dizziness.
The limitations described above make it clear that there Allegheny Health Network (AHN) is a large multi-
remains a need for new interventions to treat PD that are specialty health care system that works with Highmark
more effective, better tolerated, and more readily available. Health (HH) As part of an effort to facilitate the early use
Numerous theories of the etiology of PD exist, with pro- of interventions that have proven to be safe and effective,
posed pathways leading to effective treatments (Meuret et al. have received FDA clearance, but are not yet covered by
2010). Tolin et al. (2017) observed that many patients with most commercial insurers, HH initiated the VITAL Plat-
PD chronically hyperventilate. Ley (1985) and Klein (1993) form. Given the unmet need for additional treatments for
hypothesized that hyperventilation is central to the etiol- PD and high health care utilization and costs for patients
ogy and maintenance of the condition. Klein’s “suffocation with PD, we wanted to study Freespira through the VITAL
alarm hypothesis” suggests that the respiratory abnormali- program to assess its effectiveness in our setting, and to
ties associated with PD may be due to a hypersensitivity see if the treatment reduced associated health care costs.
to ­CO2. Physical symptoms of acute hyperventilation and Prior to commencing the program, HH performed a claims
of associated compensatory mechanisms include gastroin- analysis to determine the total costs of care for members
testinal distress, cold sensations, fatigue, rapid or irregular with and without PD. Claims analyzed from members with
heartbeat, chest pain, impaired breathing, muscle tension, PD showed that members with PD had higher utilization
and paresthesia, all sensations which are similar to those and greater than twice the health costs of control members
seen in anxiety and panic (Tolin et al. 2017). Based on the matched on age, gender, and the status of certain high-cost

13
Applied Psychophysiology and Biofeedback

medical conditions (Diabetes, CHF, Thyroid condition and also uploaded to a secure server after each session for review
Asthma). (Highmark Health 2015). by the therapist (password protected access).
The primary study goal of this quality improvement pro- Patients were instructed how to operate Freespira, ori-
gram (QIP) was to determine if treatment with Freespira ented to the purpose of modifying breathing style, and
of PD patients would significantly reduce the cost of care observed in a single clinic visit by a behavioral health clini-
of these patients in the 12 months following treatment. A cian. Brief (20–30 min) weekly follow up clinic visits were
secondary study goal was to assess the efficacy of Freespira provided for the 4-week treatment episode. Visit content
in our setting at Allegheny Health Network. included review of progress and adjustment of target res-
piratory rate as per protocol.

Outcome Measures
Methods
Clinical outcome measures were collected pre- and post-
Patients with health insurance including medical and Freespira treatment at week 5, and at 6 and 12-month follow-
pharmacy coverage from Highmark BCBS of Western up assessments. The primary cost endpoint was total allowed
Pennsylvania were recruited for this QIP. All patients pro- medical costs. emergency department (ED) costs, and medi-
vided written consent to participate in the QIP. Eligibility cation costs were also determined. The primary clinical out-
criteria included a primary diagnosis of Panic Disorder, come measure was the Severity Scale (PDSS) (Shear et al.
age > 18 years old, and a Clinician Global Impression (CGI- 2001). Using this scale, remission was considered a PDSS
S) score of 4 or more, which is at the moderate level. The score of 5 or less. Response was defined as a 40% reduction
Panic Disorder Severity Scale (PDSS) (Shear et al. 1997; in PDSS scores from baseline (Furukawa et al. 2009). Sec-
Furukawa et al. 2009) served as the primary clinical end- ondary clinical endpoints included the Sheehan Disability
point for the QIP. Patients could be taking an SSRI and/ Survey (SDS), Clinical Global Impression—Severity scale
or benzodiazepines (including as needed benzodiazepines), (CGI-S), panic attacks (weekly count), patient satisfaction
but were not to change medications immediately prior to, survey (post-treatment, 6 and 12-months), and physiologic
during, or immediately after the 4 week Freespira treatment. data (respiratory rate and exhaled ­CO2 levels). Physiologic
Additional DSM-5 (APA 2013) disorders were acceptable changes were determined by comparing the mean exhaled
and were tracked. Exclusionary criteria included pregnancy ­CO2 (mmHg) and mean respiratory rate (breaths-per-minute,
or planning to become pregnant, Obesity Hypoventilation bpm) in the 2-min baseline stages of the first at-home session
Syndrome (OHS), evidence of organic mental disorder, sig- and the final completed session. Adherence to the Freespira
nificant hearing or sight impairment, suicidality, or the pres- 4-week twice-daily treatment protocol was defined as the
ence of any psychotic disorder or unstable bipolar disorder, percentage of the protocol-specified 56 sessions completed.
or if the interviewer considered Bipolar Disorder to be the
patient’s primary diagnosis. Analytic Strategies
Additional exclusionary criteria included pulmonary dis-
ease, such as emphysema or COPD, a score of ≥ 10 on the For patients treated with Freespira, allowed costs by month
COPD assessment (if administered), ­ETCO2 of ≥ 48 mmHg for the 12 months preceding and 12 months following treat-
at first treatment visit, epilepsy or seizures, or inability to ment were collected and the mean or median per-member-
understand or comply with the project’s procedures. per-month (PMPM) cost determined. The percent change
Fifty-two adult patients with PD were consented and from pre-treatment to 12 months post-treatment was then
enrolled, of which fifty-one commenced treatment with calculated.
Freespira in Allegheny Health Network Behavioral Health Clinical outcomes were compared between baseline and
clinics. Treatment was provided by eight behavioral health the post-treatment follow-up and the means compared by
clinicians trained to deliver the protocol. t-test. The % of patients achieving clinically significant
Initiation of Freespira treatment followed assessment for PDSS changes was also determined.
panic symptoms by an AHN Freespira-trained behavioral
health clinician, based on interview plus administration of
PDSS. For the 4-week duration of home device use, each Results
patient was instructed to perform a Freespira session twice
daily, spaced at least 6 h apart, with weekly check-in visits Fifty-two patients with Highmark insurance with PD
to their therapist. were enrolled in this Quality Improvement Program. The
All physiologic session data are stored on the tablet for mean age of participants was 39 years (Table 1), and 67%
viewing by the therapist and patient. De-identified data are were female. Average duration of PD at enrollment was

13
 Applied Psychophysiology and Biofeedback

Table 1  Baseline study demographics were in remission, while 68% of subjects were in remis-
AHN/HH QIP
sion at 12 months (remission defined PDSS ≤ 5, Furukawa
et al. 2009). 82% of participants had clinically significant
Enrolled N 52 symptom reduction immediately post-treatment, while
Mean Age (range) 39 (18–63) 91% of subjects had significant symptom reduction at
Female % 67% 12 months post-treatment (“clinically significant symp-
African-American or Hispanic % 6% tom reduction”, Response, is defined as at least a 40%
Duration of PD, mean years (range) 10 (0.67–33) reduction in PDSS, Furukawa et al. 2009). All measures
# Panic attacks/mean per week 2 of PDSS reduction were statistically significant (p < 0.01,
t-test) at post-treatment, 6-months and 12-months. Mean
weekly panic attack frequency fell from 2 panic attacks per
10.1 years, and subjects reported averaging 2 panic attacks week at baseline to 0.2 immediately post-treatment and
per week. Baseline panic severity and functional disability 0.3 at 12 months (p < 0.01, t-test) and 73% reported being
were in the moderate range (CGI-S mean = 4.4, mean PDSS panic-attack free at 12-months. From baseline, scores on
score at baseline was 14.4). the Sheehan Disability Scale declined from 14.7 (sd = 6.5)
to 5.0 (sd = 5.6) at 12 months (p < 0.01, t-test).
Attrition & Adherence Measures of respiratory physiology shifted in the course
of treatment and follow-up. Initial mean baseline ­ETCO2
Of the 52 enrolled patients, 45 completed 15 or more ses- was 36.6 (sd = 3.6) and rose to 39.1 (sd = 3.5) at the final
sions (13% attrition) and were included in the adherence session. Mean respiratory rate at baseline decreased from
analysis. Consistent with Tolin, patients completing > 56 15.2 (sd = 3.9) to 11.7 (sd = 4.1) at completion (p < 0.01
sessions (as specified in the protocol) were coded to 100%. on both measures).
Overall, participants completed an average of 47 sessions Patient satisfaction (Table 3) remained at greater than
(84%) of the target 56. 90% from post-treatment thru 12-months.
Clinical measures were available at end of treatment on
44 of the treatment-completers. Six-month measures were
available on 27 participants and 12-month measures were
available on 22 participants.
Table 3  Patient satisfaction AHN Patient Survey: would
Clinical & Physiologic Impact of Freespira Treatment definitely/would recommend
Freespira
Table  2 shows symptom reduction data. The cohort’s Post-Tx 6-Months 12-Months
PDSS score fell from baseline 14.4 (sd = 3.8) to 4.9
90% 93% 100%
(sd = 3.4) immediately post-treatment and 4.4 (sd = 4.5)
n = 40 n = 27 n = 21
at 12 months. Immediately post-treatment, 48% of subjects

Table 2  Baseline and post- Measure Baseline Post-tx Month-6 Month-12

treatment outcomes
n 50 44 27 22
Mean PDSS Score (SD) 14.4 (3.8) 4.9 (3.4) 4.1 (4.3) 4.4 (4.5)
Mean PDSS Score decrease from enrollment N/A 66% 72% 69%
%Pts w/40% PDSS Score Reduction N/A 82% 85% 91%
%Pts w/PDSS ≤ 5 Remission 0% 48% 74% 68%
Mean weekly PA (range) 2 (0–5) 0.2 (0–2) 0.6 (0–7) 0.3 (0–1)
%Pts who were PA free 22% 86% 81% 73%
Sheehan Disability Scale (SD) 14.7(6.5) 5.9 (5.0) 3.8 (5.8) 5.0 (5.6)
Clinician Global Impression (CGI-S) 4.4 (0.7) 2.8 (0.8) 2.1 (1.0) 2.1 (0.9)
Exhaled C02 mmHg (SD) 36.6 (3.6) 39.1 (3.5) N/A N/A
(completers n = 45)
Respiration rate BPM (SD) 15.2 (3.9) 11.7 (4.1) N/A N/A
[completers n = 45]

PDSS Panic Disorder Severity Scale, CGI-S clinician’s global impression-severity, SD Sheehan Disability
Scale; PA panic attacks in past week

13
Applied Psychophysiology and Biofeedback

Cost Impact of Freespira Treatment increase in ED spend at 10–12 months post-treatment was

due to one patient with multiple visits in month-10.
Monthly claims data was analyzed from a total of 389 mem-
ber-months within the year pre-treatment and 430 member-
months within the year post-treatment. Discussion
Table 4 shows the per month allowed medical cost post-
Freespira treatment was reduced by 35% compared to the The analysis of health care claims by Highmark Health prior
pre-treatment period. This included all medical claims for to initiation of the Freespira QIP confirmed prior studies
physician visits, ED visits, and diagnostic tests. Mean Emer- documenting that patients with PD use more healthcare
gency Department costs were reduced by 65% at 12 months resources compared to matched controls without PD. In this
post-Freespira. Panic patients tend to use a variety of medi- initial HH analysis, overall health spending was significantly
cations, both to address panic symptoms, but also due to elevated in patients with PD. PD patients also demonstrated
their misinterpretation of panic symptoms as medical prob- higher emergency department utilization and pharmacy
lems (i.e. GI, IBS, cardiac) this is reflected in a reduction in costs. (Highmark Health 2015).
median overall medication costs (per member per month) of The results of this Freespira QIP are consistent with pre-
68% post-Freespira. (Due to the effect of a outliers in both vious reports that individuals with PD evidence increased
groups, mean pharmacy costs were reduced by only 16%). rates of medical utilization (Deacon et al 2008; Barsky et al.
Figure 1 shows per member per month (PMPM) costs at 1999; Health Lumen 2016; Highmark Health 2015).
3, 6, 9, and 12 months pre and post Freespira treatment. Each This Quality Improvement Program showed clinically and
point represents a 3-month weighted average PMPM. The statistically significant reductions in panic symptoms at the
completion of a 4-week protocol. The 68% remission and
91% improvement rates at 1-year post-treatment are similar
to outcomes from previously published trials (Meuret et al.
Table 4  Cost impact 2008; Tolin et al. 2017). As in previous trials, significant
PMPM 12-Months pre 12-Months post % Savings improvement in panic symptoms were observed at the com-
pletion of the treatment and maintained for the year follow-
Overall medical $548 $358 35 ing treatment. Reductions in the Sheehan Disability Scale
(mean)
reflect significant improvements in quality of life, suggest-
ED claims (mean) $87 $30 65
ing that improvements in panic frequency and severity, as
Pharmacy (median) $73 $23 68
well as reduced avoidance, accompany subjective clinical
PMPM per member per month improvement and reduce health care expenditures.

Fig. 1  Per member per month (PMPM) costs at 3, 6, 9, and costs, and emergency costs are displayed, with the 4-week Freespira
12 months pre and post Freespira treatment. Each point represents a treatment window indicated
3-month weighted average PMPM. Overall medical costs, pharmacy

13
 Applied Psychophysiology and Biofeedback

This is the first published report to examine the trajec- prevent the development of secondary safety behaviors
tory of medical spending in the first year following a short- or agoraphobia. Such early intervention would further
duration, technology-enabled treatment for PD. The 35% improve overall patient quality of life and reduce the costs
reduction in overall medical costs, 68% reduction in medi- of medical care. Additional studies should investigate the
cation costs and 65% reduction in ED costs indicate that potential benefit of Freespira treatment of other pathologi-
medical-care utilization was sharply reduced following the cal anxiety conditions.
intervention, paralleling the reduction in the physical symp-
toms of panic. Acknowledgements  We thank the research coordinators from AHN,
especially Ben Kelly for the work on this QIP. We also thank Rajesh
As typical panic symptoms of gastrointestinal distress, Chitturi (Highmark Health) and Simon Thomas (PAHS) for data man-
cold sensations, fatigue, rapid or irregular heartbeat, chest agement and analysis.
pain, impaired breathing, muscle tension, and paresthesias
are frequently misinterpreted by patients with PD as car- Funding  The VITAL Innovation Program at Highmark Heath provided
diac, GI, and respiratory concerns, the significant decrease funding for the study and extracted and analyzed the claims data to
obtain cost information. VITAL is a test and learn platform that helps
in medication costs after Freespira may reflect a decrease in to propel new health solutions into the marketplace by facilitating a
medications targeted at these specific somatic symptoms. real-world test and generating clinical, cost, and experience outcomes.
The 13% attrition rate and 84% compliance rates sug- PAHS provided study-related materials and equipment. The authors
gest that the Freespira treatment is well tolerated and has a take responsibility for the integrity of the data and the accuracy of the
data analysis.
favorable side effect profile, consistent with earlier reports
(Tolin et al. 2017).
We note some limitations to this QIP. Patients lost to
Compliance with Ethical Standards 
follow-up at the 6 and 12-month time points create some Conflict of interest  The authors declare that they have no conflict of
uncertainty as to whether data loss differentially affected interest.
treatment responders or non-responders. However, access to
cost data was independent of follow-up visit symptom meas-
urement, and the reduction in cost data across participants
suggests a positive trajectory of overall health in all QIP References
participants, both those assessed at follow-up, and those not
assessed. Additionally, the remission and improvement rates American Psychiatric Association (APA). (2013). Diagnostic and
in this study closely resemble the findings of the prior trials. statistical manual of mental disorders (5th ed.). Arlington, Vir-
ginia: American Psychiatric Association.
Barsky, A., Delamater, B., & Orav, J. (1999). Panic disorder patients
and their medical care. Psychosomatics, 40(1), 50–56.
Conclusions Dammen, T., Arnesen, H., Ekeberg, O., Husebye, T., & Friis, S.
(1999). Panic disorder in chest pain patients referred for cardio-
logical outpatient investigation. Journal of Internal Medicine,
PD is associated with significant impairment in life function 245, 497–507.
in general and quality of life in particular, as well as high Deacon, B., Lickel, J., et al. (2008). Medical utilization across the anxi-
rates of co-morbidity with other medical and psychiatric dis- ety disorders. Journal of Anxiety Disorders, 22, 344–350.
orders and sustained and excess healthcare utilization and Fleet, R., Dupuis, G., Marchand, A., Burelle, D., Arsenault, A., &
Beitman, B. (1996). Panic disorder in emergency room chest pain
spending. patients: Prevalence, comorbidity, suicidal ideation, and physician
This report documents significant decreases in panic recognition. American Journal of Medicine, 101, 371–380.
symptoms, functional impairment, and overall medical Fleet, P., Lavoie, K., Martel, J., Dupuis, G., Marchand, A., & Beit-
spending in a commercially insured cohort of panic sufferers man, B. (2003). Two-year follow-up of emergency department
patients with chest pain: Was it panic disorder? Canadian Journal
treated with Freespira. With the clinical efficacy reported in of Emergency Medicine, 5(4), 247–254.
this and prior published reports, strong adherence, favora- Furukawa, T., Shear, K., Barlow, D., Gorman, J., Woods, S., Money, R.,
ble side effect profile, and now documented reductions in et al. (2009). Evidence-based guidelines for interpretation of the
medical costs, Freespira represents an important addition to Panic Disorder Severity Scale. Depress Anxiety, 26(10), 922–929.
https​://doi.org/10.1002/da.20532​.
treatment options for panic conditions. Goddard, A. (2017). Treatment-resistant panic disorder. Psychiatric
This is the first published report to show a clear asso- Times, 34(11).
ciation between the treatment of underlying respiratory Highmark Health Panic Disorder Prevalence Analysis: High Level Cost
dysfunction associated with PD and significant health care and Count Overview & Matched Cohort Analysis. Unpublished
Study 2015.
claim reduction in patients with PD. This further confirms Kessler, R., Chiu, W., Jin, R., Ruscion, A., Shear, K., & Walters, E.
the relationship of respiratory dysfunction to panic attacks (2006). The epidemiology of panic attacks, panic disorder, and
and PD. Further, offering a highly effective, well-tolerated agoraphobia in the National Comorbidity Survey Replication.
treatment by patients early in their course of illness may Archives of General Psychiatry, 63(40), 415–424.

13
Applied Psychophysiology and Biofeedback

Klein, D. (1993). False suffocation alarms, spontaneous panics, and Shear, K., Brown, T., Barlow, D., Money, R., Sholomskas, D., Woods,
related conditions. An integrative hypothesis. Archives of General S., et al. (1997). Multicenter collaborative panic disorder severity
Psychiatry., 50, 306–317. scale. The American Journal of Psychiatry, 154(11), 1571–1575.
Ley, R. (1985). Blood, breath and fears: A hyperventilation theory of Shear, K., Rucci, P., Williams, J., Frank, E., Grochocinski, V., Vander
panic attacks and agoraphobia. Clinical Psychology Review, 5, Bilt, J., et al. (2001). Reliability and validity of the Panic Disorder
271–285. Severity Scale: Replication and extension. Journal of Psychiatric
Meuret, A., Rosenfield, D., Seidel, A., Bhaskara, L., & Hofmann, S. Research, 35, 293–296.
(2010). Respiratory and cognitive mediators of treatment change Shirmeshan, E., Bailey, J., Relyea, G., et al. (2013). Incremental direct
in panic disorder: Evidence for intervention specificity. Journal medical expenditures associated with anxiety disorders for the
of Consulting and Clinical Psychology, 78(5), 691–704. https​:// U.S. adult population: Evidence from the Medical Expenditure
doi.org/10.1037/a0019​552. Panel Survey. Journal of Anxiety Disorders, 27(7), 720–727.
Meuret, A. E., Wilhelm, F. H., & Roth, W. T. (2008). Feedback of end- Tolin, D., McGrath, P., Hale, L., Weiner, D., & Gueorguieva, R. (2017).
tidal ­pCO2 as a therapeutic approach for panic disorder. Journal A multisite benchmarking trial of capnometry guided respiratory
of Psychiatric Research, 42, 560–568. intervention for panic disorder in naturalistic treatment settings.
Nardi, A., Mochcovitch, M., Freire, R., Machado, S., Silva, A., Dias, Applied Psychophysiology and Biofeedback, 42(1), 51–58.
G., et al. (2015). Prospective, open, randomized 3-years long-term Vlahiotis, A., Devine, S., Eichholz, J., & Kautzner, A. (2011). Dis-
treatment of panic disorder with clonazepam, paroxetine, or their continuation rates and health care costs in adult patients starting
combination and follow-up during additional 6 years. European generic versus brand SSRI or SNRI antidepressants in commer-
Psychiatry, 30(Supplement 1), 170. cial health plans. Journal of Managed Care Pharmacy, 17(2),
Palo Alto Health Sciences Panic Disorder Claims Analysis. Health 123–132.
Lumen Healthcare Analytics. Unpublished Study 2016.
Roy-Byrne, P., Stein, M., Mercier, E., Thomas, R., McQuaid, J., Publisher’s Note Springer Nature remains neutral with regard to
Katon, W., et al. (1999). Panic disorder in the primary care set- jurisdictional claims in published maps and institutional affiliations.
ting: comorbidity, disability, service utilization, and treatment.
Journal of Clinical Psychiatry, 60(7), 492–499.
Shankman, S., Gorka, S., Katz, A., Klein, D., Markowitz, A., Man-
ber, R., et al. (2017). Side effects to antidepressant treatment in
patients with depression and comorbid panic disorder. Journal of
Clinical Psychiatry, 78(4), 433–440.

13