Association of aphthous ulcers with self-reported symptoms of depression in a sample of smartphone users
Yale University, Child Study Center, New Haven, Connecticut, USA
Texas Tech University Health Sciences Center, Lubbock, Texas, USA
Department of Pharmacy Practice and Science, University of Iowa, Iowa City, Iowa, USACarol S. North, MD, MPH
The VA North Texas Health Care System, University of Texas Southwestern Medical Center, Departments of Psychiatry and Surgery/Division of Emergency Medicine, Dallas, Texas, USA
BACKGROUND: Our goal is to examine the association of recurrent aphthous stomatitis (RAS) with symptoms of depression using a smartphone-based questionnaire survey.
METHODS: An electronic survey was administered through a smartphone app asking respondents about current depressive symptoms using the Quick Inventory of Depressive Symptoms (QIDS), and asking whether they had ever or recently experienced RAS. Multivariate logistic regression analysis was used to determine associations.
RESULTS: A total of 478 individuals completed the survey, with 64% reporting a lifetime prevalence of RAS, and 21% experiencing an aphthous ulcer within the last month. RAS was significantly associated with increased sleep, decreased appetite, low energy, and feeling sluggish. RAS was not associated with overall depression severity as measured by total QIDS score, or with cardinal features of depression such as sadness, insomnia, impaired concentration, self-blame, thoughts of death, or anhedonia. Prevalence of RAS did not differ by age, sex, or smoking status, but was less likely in blacks and Asians compared with whites.
CONCLUSIONS: RAS was a common phenomenon in this sample of mostly depressed individuals, and was associated with some neurovegetative symptoms of depression, but not depression severity.
KEYWORDS: aphthous ulcer, depression, RAS
ANNALS OF CLINICAL PSYCHIATRY 2013;25(4):266-270
Recurrent apthous stomatitis (RAS) is characterized by episodes of painful, shallow ulcerations on the oral mucosa. Although anecdotally it appears to be a common condition, studies of RAS have examined widely divergent populations and prevalence estimates vary from 20% to 50%.1-5
The etiology of RAS is unclear, with trauma, food hypersensitivity, infectious agents, and metabolic derangements appearing to contribute.6 An association between active RAS and stress was first noted by dental clinicians.7,8 Subsequent studies have demonstrated this association,9-12 also noting that biochemical correlates of stress such as salivary cortisol are higher in individuals with active ulcers.11
Stress is common among individuals with psychiatric disorders, and the mental and emotional manifestations of stress were noted to be more closely associated with RAS compared with the physical manifestations.12 Anxiety symptoms have a particularly strong correlation with active RAS and may account for much of the link to stress.13-15
Less is known about the association between RAS and other psychiatric disorders such as depression. Those experiencing from RAS are more likely to have a particular polymorphism in the serotonin transporter (5-HTTLPR).16 This polymorphism has attracted considerable interest for its link to depression,17 suggesting a potential linkage of RAS with depression. Earlier reports did not find an association between RAS and depression,15,18 although sample sizes were small. These studies did not stratify severity of depression and odds of RAS, and did not examine the association of RAS with individual symptoms of depression.
In this study, we employed a smartphone-based survey to measure user-reported depressive and RAS symptoms, and examined whether the severity of depressive symptoms was associated with risk of lifetime or current RAS.
After obtaining university institutional review board approval, an anonymous, confidential survey was developed as a smartphone application, available for download and use on the Google™ marketplace for individuals age ≥18. A total of 603 respondents initiated the survey, and 478 usable responses were obtained (79% response rate), which we defined as those who answered question items regarding their depressive symptoms and whether they had ever or currently were experiencing aphthous ulcers. Those not answering these questions or giving a response of “unknown” were excluded from the analysis.
The survey consisted of multiple-choice and free-form questions, and respondents were asked demographic questions about their age, sex, and race. They also were asked about current depressive symptoms via the Quick Inventory of Depressive Symptomatology (QIDS),19 which is comprised of 16 multiple-choice questions that encompass various subjective mood and somatic symptoms of depression.
To assess the prevalence of lifetime or current RAS, users were shown a photograph of an aphthous ulcer with text description of an ulcer that clarified how they look, and where in the mouth they tend to occur. Users selected via multiple choice whether they had ever experienced an aphthous ulcer, and if so, whether they had experienced ≥1 in the past month.
Participants could download the survey from the internet via smartphone. To minimize recruitment bias towards those with aphthous ulcers participating in this study, we constructed the survey to eliminate keyword searches directly related to aphthous ulcers, so that users specifically seeking apps related to aphthous ulcers would not find it on a search. The app was listed by keywords according to its primary purpose, which was to measure depressive symptoms. The Google Play™ app market, where the survey was hosted, allows analysis of what users are searching for when downloading a certain app, giving an estimation of original user intent in finding it. The 5 most common apps viewed and installed by users of our app were variants of depression and anxiety surveys, and none related to aphthous ulcers or other general medical conditions.
Data were transmitted wirelessly to a remote, secure database for analysis (Starfield Technologies, Scottsdale, AZ). Item-level responses in the survey occasionally were lost because of user error or loss of connection to the remote database, resulting in missing data points in some of the surveys. Therefore, the total responses to individual items in TABLE 1 does not add up to the total number of respondents (478 individuals).
Prevalence of aphthous ulcers by respondent characteristics
| Male (n=133)
| Female (n=238)
| White (n=243)
| Black (n=25)
| Hispanic (n=35)
| Asian (n=48)
| Non-smoker (n=171)
| Lifetime smoker (n=203)
| Current smoker (n=125)
| Light smoker (n=78)
| Heavy smoker (n=47)
| Yes (n=175)
| No (n=198)
| 0 to 9 (n=39)
| 10 to 14 (n=99)
| 15 to 20 (n=180)
| 21 to 27 (n=58)
Prevalence of RAS was calculated by dividing the number of respondents reporting aphthous ulcers by the total number of valid responses to that question. We used 2 outcome measures: whether someone had ever experienced RAS, and whether they had experienced an ulcer in the past month. We assessed potential risk factors for aphthous ulcers in a logistic regression model adjusting for depression severity (using QIDS score cutoffs of 0 to 9, 10 to 15, 16 to 20, and ≥21), age, sex, race, and smoking status.
The association between the 16 individual question items from the QIDS and likelihood of lifetime or current RAS was examined by comparing an item-level response of no symptoms (a score of 0) vs any report of that depressive symptom regardless of severity (score of 1 through 3). Question items 1 through 16 in TABLE 2 correspond exactly to the 16 question items asked (in the same order) in the QIDS-SR16.19 We assessed these associations using a logistic regression model adjusting for depression status, age, sex, race, and smoking status. Two-tailed statistical significance was defined as P ≤ .05. All statistical analyses were performed using STATA MP Version 12.0 (STATA Corp, College Station, Texas, USA).
Odds ratios of aphthous ulcers by specific QIDS question
|QIDS question item
|(1) Initial insomnia
||1.1 (0.7 to 1.9)
||0.8 (0.4 to 1.6)
|(2) Middle insomnia
||1.2 (0.7 to 2.0)
||1.2 (0.6 to 2.3)
|(3) Terminal insomnia
||1.0 (0.6 to 1.4)
||1.2 (0.7 to 2.0)
|(4) Sleeping too much
||1.4 (0.9 to 2.1)
||1.7a (1.0 to 2.8)
|(5) Feeling sad
||0.9 (0.4 to 2.5)
||1.9 (0.4 to 9.3)
|(6) Decreased appetite
||2.1a (1.1 to 4.1)
||2.4 (0.9 to 6.2)
|(7) Increased appetite
||0.8 (0.5 to 1.4)
||1.1 (0.5 to 2.2)
|(8) Decreased weight
||1.3 (0.8 to 2.3)
||1.4 (0.7 to 2.9)
|(9) Increased weight
||0.9 (0.5 to 1.5)
||0.8 (0.4 to 1.6)
|(10) Impaired concentration
||1.0 (0.4 to 2.2)
||1.3 (0.4 to 4.4)
||1.2 (0.7 to 2.2)
||1.5 (0.6 to 3.7)
|(12) Thoughts of death
||1.5 (1.0 to 2.3)
||1.4 (0.8 to 2.4)
||1.1 (0.6 to 2.1)
||1.6 (0.6 to 4.1)
|(14) Energy level
||2.1a (1.2 to 3.9)
||3.0a (1.1 to 7.7)
|(15) Feeling sluggish
||1.8b (1.2 to 2.8)
||2.7b (1.4 to 5.0)
|(16) Feeling restless
||1.2 (0.8 to 1.9)
||1.4 (0.7 to 2.6)
Android smartphone application and respondent characteristics
Our smartphone application was downloaded and used by individuals worldwide. The majority of respondents were from the United States (57%), followed by United Kingdom (11%), Australia (7%), and Canada (6%). The final respondent sample had an average age of 30, was otherwise predominantly male, white, and slightly more likely to have smoked than not (TABLE 1).
Prevalence of aphthous ulcers
Overall, 64% of respondents reported a lifetime history of RAS, and 21% reported having had an active ulcer in the past month. Aphthous ulcer prevalence was found to be lower in respondents from the United Kingdom (odds ratio [OR]=0.4, P < .05) compared with the United States, although controlling for country of origin did not affect any other measured outcomes (data not shown). Blacks and Asians were less likely to report RAS (OR=0.4 and 0.5 respectively, P < .05). The prevalence of RAS did not vary by age, sex, or smoking status. In 1 specification, we also assessed whether heavy smoking (ie, >10 cigarettes per day) correlated with changes in the prevalence of aphthous ulcers, but none was found.
Depression and aphthous ulcers
Ninety percent of subjects reported QIDS symptoms of ≥10, which indicates at least mild to moderate depression. Having this degree of depressive symptoms was not associated with the likelihood of lifetime or current RAS (OR=1.25, P=.52; OR=1.22, P=.67, respectively). Similarly, reporting a history of previously diagnosed depression did not associate with lifetime or current RAS (OR=1.23, P=.67; OR=1.25, P=.30, respectively). As noted in TABLE 3, no association was found between RAS and severity of depression symptoms. The 16 item-level responses to the QIDS were examined individually (TABLE 2) to measure association between specific symptoms of depression and likelihood of lifetime or active RAS. Significant associations were found with symptoms of sleeping too much, decreased appetite, decreased energy, and feeling sluggish but not with other key features of depression such as sadness, insomnia, impaired concentration, self-blame, thoughts of death, and anhedonia.
Prevalence of aphthous ulcers by QIDS severity
|0 to 9
|10 to 14
||1.1 (0.5 to 2.2)
||1.0 (0.4 to 2.7)
|15 to 20
||1.5 (0.8 to 3.1)
||1.4 (0.5 to 3.6)
|21 to 27
||1.2 (0.6 to 2.7)
||2.0 (0.7 to 5.7)
The primary objective of this research was to examine the association between active symptoms of depression and the likelihood of reporting lifetime or currently active RAS. Overall, we found that the likelihood of lifetime or current RAS was not associated with having at least moderate symptoms of depression, nor was it associated with the severity of depression symptoms, at least as measured by the total self-reported QIDS score. This is in agreement with prior work that similarly did not find an association between depression and RAS.15,18
Certain individual symptoms in the QIDS depression questionnaire were associated with increased likelihood of RAS, such as increased sleep, decreased appetite, low energy, and feeling sluggish. Although these particular symptoms are features of major depression, they are not specific, and are common in those who are anxious or under stress. It is notable that other key symptoms that help define major depression, such as sadness and anhedonia, were not associated with RAS. From this data, we do not view RAS as being specifically associated with major depression per se, but instead more likely to associate with a general stress phenotype, as has been noted previously.9-12
In this sample, lifetime prevalence of aphthous ulcers (64%) was higher than most earlier reports from various populations, although prevalence rates vary considerably.1-3,5 This survey selected for individuals who were depressed and likely seeking to measure their depression via a self-administered survey, and therefore the prevalence results cannot be generalized to a non-clinical population. We note again that the prevalence was similarly high among those reporting fewer symptoms of depression.
Web-based surveys hold great promise for reaching large numbers of respondents in a convenient manner. Nonetheless, there are limitations to the method. As with any convenience sample, selection bias is a serious concern. Almost all respondents to this survey endorsed depressive symptoms, making comparison with healthy controls not possible, although still allowing for analysis of association with depression severity. It also is possible that users were biased to take the study based on their history of RAS, however, we were able to monitor user intent in finding the survey, which suggested that users were not looking for medical apps related to general health or RAS, but were searching for apps related to depression. Also, our results are in line with the prior literature, which also did not find a link between diagnosis of depression and RAS, and now merely extend the findings with a larger sample size to show that a certain subset of symptoms does associate.
Truthful responses to a survey are a concern with methods that do not have in-person oversight, however, this is a common problem to any anonymous survey, and not specific to web-based instruments. We further note that few respondents endorsed unusual response patterns, such as giving unlikely ages (eg, age >100), or symptom response clusters (eg, reporting the maximum severity to all questions). Nonetheless, the association we found between certain vegetative symptoms and RAS should be replicated in other settings before conclusions can be drawn.
The prevalence of RAS did not correspond to depression severity in this sample, but was associated with some vegetative symptoms of depression that also overlap with stress, anxiety, and changes in cortisol levels. Given how prevalent RAS is in this depressed sample and its known association to various somatic illnesses, it may be of future interest to investigate whether RAS may serve as a biomarker for identifying subtypes of depression and anxiety, and whether this may help predict response to treatment.
DISCLOSURES: Dr. Schroeder receives grant/research support from the Agency for Healthcare Research and Quality and the National Institutes of Health. Dr. Webb, Mr. Webb, and Dr. North report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.
- Axéll T, Henricsson V. The occurrence of recurrent aphthous ulcers in an adult Swedish population. Acta Odontol Scand. 1985;43:121–125.
- Rivera-Hidalgo F, Shulman JD, Beach MM. The association of tobacco and other factors with recurrent aphthous stomatitis in an US adult population. Oral Dis. 2004;10:335–345.
- Axéll T, Zain RB, Siwamogstham P, et al. Prevalence of oral soft tissue lesions in out-patients at two Malaysian and Thai dental schools. Community Dent Oral Epidemiol. 1990;18:95–99.
- Reichart PA. Oral mucosal lesions in a representative cross-sectional study of aging Germans. Community Dent Oral Epidemiol. 2000;28:390–398.
- Kleinman DV, Swango PA, Pindborg JJ. Epidemiology of oral mucosal lesions in United States schoolchildren: 1986-87. Community Dent Oral Epidemiol. 1994;22: 243–253.
- Natah SS, Konttinen YT, Enattah NS, et al. Recurrent aphthous ulcers today: a review of the growing knowledge. Int J Oral Maxillofac Surg. 2004;33:221–234.
- Kaufman AY. Aphthous stomatitis as a featuring syndrome of emotional stress in dental treatment. Quintessence Int Dent Dig. 1976;7:75–78.
- Larato DC. Stress and aphthous ulcers. J Acad Gen Dent. 1972;20:25–26.
- Pedersen A. Psychologic stress and recurrent aphthous ulceration. J Oral Pathol Medicine. 1989; 18:119–122.
- Gallo Cde B, Mimura MA, Sugaya NN. Psychological stress and recurrent aphthous stomatitis. Clinics (Sao Paulo). 2009;64:645–648.
- Albanidou-Farmaki E, Poulopoulos AK, Epivatianos A, et al. Increased anxiety level and high salivary and serum cortisol concentrations in patients with recurrent aphthous stomatitis. Tohoku J Exp Med. 2008;214:291–296.
- Huling LB, Baccaglini L, Choquette L, et al. Effect of stressful life events on the onset and duration of recurrent aphthous stomatitis. J Oral Pathol Med. 2012;41:149–152.
- McCartan BE, Lamey PJ, Wallace AM. Salivary cortisol and anxiety in recurrent aphthous stomatitis. J Oral Pathol Med. 1996;25:357–359.
- Zadik Y, Levin L, Shmuly T, et al. Recurrent aphthous stomatitis: stress, trait anger and anxiety of patients. J Calif Dent Assoc. 2012;40:879–883.
- Al-Omiri MK, Karasneh J, Lynch E. Psychological profiles in patients with recurrent aphthous ulcers. Int J Oral Maxillofac Surg. 2012;41:384–388.
- Victoria JM, Correia-Silva Jde F, Pimenta FJ, et al. Serotonin transporter gene polymorphism (5-HTTLPR) in patients with recurrent aphthous stomatitis. J Oral Pathol Med. 2005;34:494–497.
- Willeit M, Praschak-Rieder N, Neumeister A, et al. A polymorphism (5-HTTLPR) in the serotonin transporter promoter gene is associated with DSM-IV depression subtypes in seasonal affective disorder. Mol Psychiatry. 2003;8:942–946.
- Picek P, Buljan D, Rogulj AA, et al. Psychological status and recurrent aphthous ulceration. Coll Antropol. 2012;36:157–159.
- Rush AJ, Trivedi MH, Ibrahim HM, et al. The 16-Item Quick Inventory of Depressive Symptomatology (QIDS), clinician rating (QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients with chronic major depression. Biol Psychiatry. 2003;54:573–583.
CORRESPONDENCE: Jadon R. Webb Yale University, Child Study Center 230 S. Frontage Road New Haven, CT 06520 USA E-MAIL: Jadon.Webb@yale.edu
Annals of Clinical Psychiatry ©2013 Quadrant HealthCom Inc.