Associations between Primary Hyperhidrosis and Psychiatric Conditions: A Matched Case-Control Study of 8,508 Patients

Abstract

Introduction

Primary hyperhidrosis (PHH) is defined by excess sweating not attributable to medications or underlying medical conditions. It is associated with significant burden of subsequent psychiatric comorbidities, but psychiatric conditions preceding PHH diagnosis are not well characterized. We assessed associations of PHH with preceding psychiatric conditions using a matched case-control approach.

Methods

Overall, 1,418 participants with PHH and 7,090 matched controls were assessed using the All of Us Research Program database. Multivariable logistic regressions calculated odds ratios and 95% confidence intervals, with Benjamini-Hochberg correction for multiple testing.

Results

Subjects with PHH vs. controls had higher odds of prior diagnosis with obsessive-compulsive disorder (OCD) (OR: 1.97, 95% CI: 1.29–2.94, corrected p = 0.007), restless legs syndrome (RLS) (OR: 1.63, 95% CI: 1.22–2.15, corrected p = 0.007), attention-deficit/hyperactivity disorder (ADHD) (OR: 1.35, 95% CI: 1.07–1.69, corrected p = 0.035), and insomnia (OR: 1.22, 95% CI: 1.04–1.44, corrected p = 0.037).

Conclusion

We found significant associations with subsequent PHH diagnosis in participants with OCD, ADHD, RLS, and insomnia. Further research into the etiologies of psychiatric conditions and PHH may better elucidate their relationship.

Introduction

Primary hyperhidrosis (PHH) is defined by excess sweating not attributable to any medications or underlying medical conditions and is almost exclusively focal. PHH accounts for approximately 93% of hyperhidrosis cases and significantly detracts from quality of life, causing physical discomfort and impairments in psychological, social, and occupational functioning [1–6]. PHH is widely prevalent, with a 2016 survey reporting that 4.8% of Americans have hyperhidrosis, and questionnaires in Vancouver, Shanghai, and Japan identified regional prevalence rates of up to 14.5% [7–9].

Prior research has suggested that autonomic nervous system dysfunction and aberrant emotional control are possible causative factors in PHH and demonstrated that PHH significantly increases the risk of later developing psychiatric comorbidities, including anxiety disorders, depressive disorders, and sleep disorders [1, 10–12]. However, associations between PHH and pre-existing psychiatric conditions are not well characterized. Therefore, our objective was to evaluate associations between psychiatric conditions and the development of PHH using a matched case-control approach.

Methods

A matched case-control study was performed on subjects ≥18 years old diagnosed with PHH using the All of Us Research Program (AoURP), a National Institutes of Health program that prioritizes participant recruitment from demographic groups historically under-represented in research. Systematized Nomenclature of Medicine (SNOMED) diagnostic codes were used to identify participants diagnosed with focal PHH (SNOMED codes 427794001, 303089000, 303090009, 403375001, 403380005, 10677591000119103, 10677631000119103, and 10677671000119100). All assessed PHH were focal, corresponding to “primary focal” hyperhidrosis (n = 1,410) (ex. “primary focal hyperhidrosis”), or focal hyperhidrosis (n = 8) (ex. “hyperhidrosis of axilla”) that were not coded as secondary. SNOMED codes were used to retrieve participant information on the following psychiatric comorbidities: generalized anxiety disorder (GAD), obsessive-compulsive disorder (OCD), major depressive disorder (MDD), bipolar disorder, schizophrenia, insomnia, restless legs syndrome (RLS), attention-deficit/hyperactivity disorder (ADHD), and posttraumatic stress disorder (PTSD) (respective SNOMED codes 21897009, 191736004, 370143000, 13746004, 58214004, 193462001, 32914008, 406506008, 47505003, and all descendant codes for each). Subjects with PHH were matched 1:5 to controls based on self-reported age, sex, and race using nearest neighbor propensity score matching and inter-group differences were assessed with Fisher’s exact test and Welch two-sample t test. Multivariable logistic regressions calculated odds ratios (ORs) and 95% confidence intervals (CIs), using the Benjamini-Hochberg procedure to correct for multiple testing. Age, sex, and race were used as covariates. Conditions corresponding to a false discovery rate of less than 0.05 were considered significant and corrected p values were reported. Temporality was established by considering only psychiatric diagnoses made before the time of PHH diagnosis. Assessed studies in the psychiatric literature consider subjects presenting with primary, local, or essential hyperhidrosis, to correspond with our sample.

Results

A total of 1,418 participants with PHH and 7,090 matched controls were included in the final analysis (Table 1). Mean age of PHH patients was 51 years, with 71% females and 65% White patients. Fisher’s exact test and Welch two-sample t test indicated no significant inter-group demographic differences (age: p = 0.993, sex: p = 0.264, race: p = 0.999) (Table 1). Subjects with PHH vs. controls had higher odds of previous diagnosis of OCD (OR: 1.97, 95% CI: 1.29–2.94, p = 0.007), RLS (OR: 1.63, 95% CI: 1.22–2.15, p = 0.007), ADHD (OR: 1.35, 95% CI: 1.07–1.69, p = 0.035), and insomnia (OR: 1.22, 95% CI: 1.04–1.44, p = 0.037), with no significant associations with GAD, MDD, bipolar disorder, schizophrenia, or PTSD (Table 2).

Table 1.

Demographics of PHH patients and controls

​	PHH (n = 1,418)	Controls (n = 7,090)	p value
Age, mean (SD)	50.94 (16.50)	50.93 (16.51)	0.993
Sex at birth (%)	​	​	0.264
Male	400 (28.21)	2,007 (28.31)
Female	1,011 (71.30)	5,058 (71.34)
No response	7 (0.49)	25 (0.35)
Race (%)	​	​	0.999
White	916 (64.60)	4,583 (64.64)
Black	171 (12.06)	856 (12.07)
Asian	34 (2.40)	173 (2.44)
Other	30 (2.12)	159 (2.24)
No response	267 (18.83)	1,319 (18.60)

PHH, primary hyperhidrosis; SD, standard deviation.

Table 2.

Multivariable logistic regression of factors associated with PHH diagnosis

​	PHH (n = 1,418), n (%)	Controls (n = 7,090), n (%)	OR (95% CI)	p value
OCD	36 (2.54)	84 (1.18)	1.97 (1.29–2.94)	0.007
Restless legs syndrome	73 (5.15)	215 (3.03)	1.63 (1.22–2.15)	0.007
ADHD	114 (8.04)	413 (5.83)	1.35 (1.07–1.69)	0.035
Insomnia	268 (18.90)	1,084 (15.29)	1.22 (1.04–1.44)	0.037
PTSD	111 (7.83)	492 (6.94)	1.05 (0.82–1.32)	0.810
MDD	387 (27.29)	1,838 (25.92)	0.96 (0.83–1.12)	0.810
GAD	192 (13.54)	888 (12.52)	0.96 (0.80–1.16)	0.810
Schizophrenia	17 (1.20)	92 (1.30)	0.91 (0.51–1.52)	0.810
Bipolar disorder	100 (7.05)	506 (7.14)	0.86 (0.67–1.10)	0.481

OCD, obsessive-compulsive disorder; ADHD, attention-deficit/hyperactivity disorder; PTSD, posttraumatic stress disorder; MDD, major depressive disorder; GAD, generalized anxiety disorder.

Boldface in p value column indicates significance (p < 0.05). p values were corrected for multiple testing using the Benjamini-Hochberg procedure; corrected p values are shown.

Groups were matched on age, sex at birth, and race. These variables were also retained as covariates in the multivariable regression models.

Discussion

Our study assessed focal PHH, which is etiologically unrelated to secondary and generalized hyperhidrosis. We found that PHH was associated with sleep-wake psychiatric disorders, including RLS and insomnia. This is concordant with prior studies which demonstrate that sleep-wake disorders are associated with autonomic dysfunction, which is implicated in PHH [10–14]. For example, a questionnaire-based study of 409 patients with untreated RLS and 331 controls found that autonomic dysfunction, as measured by the SCOPA-AUT questionnaire, was greater in patients with RLS in all domains assessed (p < 0.0001) except sexual dysfunction in men [13]. In addition, a case-control sleep laboratory study of 32 patients with chronic insomnia and 19 controls reported increased autonomic activity in every sleep stage in patients with insomnia, with higher heart rate and lower heart rate variability [14].

Notably, RLS and insomnia may share common mechanisms, including dysregulated dopaminergic and glutamatergic neurotransmission and brain iron deficiency despite normal serum iron, which may also underlie PHH [15, 16]. A cross-sectional study on blood samples from 42 PHH patients and 37 controls demonstrated that patients with PHH had altered iron metabolism, with atypically high erythrocyte iron despite normal serum iron [17]. Furthermore, a case study described a focal PHH patient who was sporadically anemic due to excess iron excretion in sweat [18]. It is not clear if mechanisms of iron dysregulation in RLS and PHH are similar. No studies on brain iron in PHH have been conducted, and iron dysregulation in RLS is heterogeneous [15]. Nonetheless, our findings of associations between sleep disorders and autonomic dysfunction may suggest that altered iron metabolism underlies pathogenesis of some PHH cases.

We also found that PHH was associated with prior ADHD diagnosis. Related to our finding, a retrospective chart review of 500 PHH patients at a single site found higher rates of ADHD among PHH patients compared to the general population baseline (6.4% vs. 4.8%, no p value reported) but did not use a timing variable [19]. In addition, a systematic review of 55 studies of autonomic function in ADHD found that ADHD was associated with autonomic dysregulation, although hypo-arousal was more common than hyperarousal [20]. Both this review and another systematic review found that stimulant ADHD medications can cause autonomic hyperarousal [20, 21]. While stimulant medications cause secondary hyperhidrosis via autonomic hyperarousal, our study identified an association between ADHD and focal, PHH, suggesting an underlying link distinct from medication side effects [22].

We also found that PHH was associated with OCD, which is novel. A cross-sectional comparative study of 17 PHH and 29 control outpatients found no link between PHH and Obsessive-Compulsive Inventory-Revised total scores, but assessed only 6 participants with OCD, only one of whom also had PHH [23]. Although PHH-related stress, paranoia regarding bodily secretions, or preoccupations with odors have been speculated to contribute to OCD development, no study has yet identified a link between OCD and PHH [24]. Additionally, no study has yet assessed the relationship between OCD and PHH with a timing variable. Our study suggests that OCD either increases the likelihood of developing PHH or modifies behavior in a manner that increases its likelihood of diagnosis.

Our finding that PHH is not associated with prior GAD, MDD, bipolar disorder, schizophrenia, or PTSD is partially consistent with prior research. The etiology of PHH is hypothesized to involve abnormal central control of emotions. A Danish hospital-based national cohort study on 7,191,519 patients, of which 8,758 had localized hyperhidrosis and 1,102 had generalized hyperhidrosis, found that pre-existing depression was more common in patients with localized hyperhidrosis (n = 349, p < 0.001) and generalized hyperhidrosis (n = 69, p < 0.001) than controls [25]. The authors of the hospital-based study maintain that, as 96.6% of patients with PHH have focal sweating, the localized hyperhidrosis patients studied represent a very close proxy for PHH [5, 25]. Furthermore, in a systematic review of PHH, atypical central control of emotion has been proposed as a contributing factor to PHH development. A case-control study of 56 PHH and 49 control outpatients found differences in DSM-IV Axis I and Temperament and Character Inventory profiles [26]. Additionally, a case-control study of 20 PHH and 20 control outpatients found elevated Eysenck Personality Questionnaire (EPQ) scores for neuroticism in PHH patients (F = 6.352, p < 0.01) [27]. Nonetheless, a cross-sectional case series of 97 PHH outpatients found EPQ scores in PHH patients to be within Korean norms [28]. A case-control study of 26 patients with “essential hyperhidrosis” (a term synonymous with PHH) and 22 control outpatients also reported no difference in underlying anxious vs. risk-taking personality traits on the Temperament and Character Inventory (19.1 ± 6.7 vs. 17.3 ± 3.3 points) [29]. Furthermore, a cross-sectional study of 42 presurgical PHH patients found low rates of overt psychopathology (no elevations in 88% of MMPI-2 profiles and 86% of STAI State and Trait Anxiety scales), suggesting that pre-existing psychological distress does not predispose to PHH [30, 31]. Thus, a consistent, well-supported link between these disorders and the development of PHH has not yet been found.

Limitations of our study include lack of information about psychiatric disease severity, PHH severity, psychiatric medications or other treatments, or adherence to treatments. Miscoding of diagnoses and delays between the development of assessed conditions and their diagnoses are both possible. Factors lengthening PHH time-to-diagnosis relative to psychiatric conditions, such as poor access to dermatologists or delays in seeking dermatological care, would limit assessment for temporality. Detection bias due to differences in healthcare-seeking habits or healthcare access is possible. Medications used to treat ADHD, OCD, and insomnia, such as stimulants, SSRIs, and SNRIs, are associated with secondary hyperhidrosis and would confound results if hyperhidrosis type was miscoded [22, 32]. Medication exposure data are available on All of Us, and thus, future studies may assess medication-PHH relationships or miscoding. Caution should be taken when comparing our findings to studies describing “hyperhidrosis,” as this includes secondary and general hyperhidrosis etiologically unrelated to PHH. Identified effect sizes are modest (all OR <2). Nonetheless, given the large proportion of patients with PHH or implicated psychiatric comorbidities, the clinical relevance of such counseling could be meaningful at the population level.

These findings suggest meaningful opportunities for patient counseling and multidisciplinary management. Psychiatrists may consider cautioning patients diagnosed with OCD, ADHD, RLS, and insomnia about the possibility of developing PHH. Dermatologists may consider inquiring about past psychiatric diagnoses in PHH patients to identify possibly untreated comorbidities. Given the effective treatments available for PHH and the drawbacks of undertreating psychiatric disorders, inter-disciplinary referrals are important.

In summary, in this matched case-control study, we found significant associations with subsequent PHH diagnosis in participants with OCD, ADHD, RLS, and insomnia. Further research into psychiatric conditions, iron dysregulation in PHH, and the links between psychiatric medications and PHH, with and without timing variables, may better elucidate PHH disease mechanisms.

Acknowledgments

The All of Us Research Program is supported by the National Institutes of Health, Office of the Director: Regional Medical Centers: 1 OT2 OD026549; 1 OT2 OD026554; 1 OT2 OD026557; 1 OT2 OD026556; 1 OT2 OD026550; 1 OT2 OD026552; 1 OT2 OD026553; 1 OT2 OD026548; 1 OT2 OD026551; 1 OT2 OD026555; IAA #: AOD 16037; Federally Qualified Health Centers: HHSN 263201600085U; Data and Research Center: 5 U2C OD023196; Biobank: 1 U24 OD023121; The Participant Center: U24 OD023176; Participant Technology Systems Center: 1 U24 OD023163; Communications and Engagement: 3 OT2 OD023205; 3 OT2 OD023206; and Community Partners: 1 OT2 OD025277; 3 OT2 OD025315; 1 OT2 OD025337; 1 OT2 OD025276. We also thank all participants in the All of Us Research Program for contributing their time and health information, without which this study would not have been possible.

Statement of Ethics

Per the All of Us Research Hub: “as a single IRB, the All of Us IRB is charged with reviewing the protocol, informed consent, and other participant-facing materials for the All of Us Research Program. The IRB follows the regulations and guidance of the Office for Human Research Protections for all studies, ensuring that the rights and welfare of research participants are overseen and protected uniformly. The Researcher Workbench employs a data passport model, through which authorized users do not need IRB review for each research project. Most authorized users will not be conducting human subjects research with All of Us data for 2 reasons: (1) the research will not directly involve participants, only their data, and (2) the data available in the Researcher Workbench have been carefully checked and altered to remove identifying information while preserving their scientific utility.” The Weill Cornell Medicine Institutional Review Board reviewed this protocol and waived the need for informed consent.

Conflict of Interest Statement

Mr. Zeldin has no conflicts of interest. Dr. Lipner has served as a consultant for BelleTorus Corporation and Moberg Pharmaceuticals.

Funding Sources

This study was not supported by any sponsor or funder.

Author Contributions

All authors contributed significantly to the production of this manuscript. S.D.Z. performed data collection, statistical analysis, and manuscript drafting and editing. S.R.L. performed manuscript drafting and editing and is the corresponding author.

Funding Statement

This study was not supported by any sponsor or funder.

Data Availability Statement

The authors confirm that data supporting the findings of this study are freely available from All of Us with registration and approval. Further inquiry can be directed to the corresponding author.