Diagnosis of specific phobia

The Diagnostic and Statistical Manual of Mental Disorders (APA, 2000) establishes as the main criterion for specific phobia a “marked and persistent fear that is excessive or unreasonable, cued by the presence or anticipation of a specific object or situation” (criterion A), where an anxiety response is triggered by the phobic stimulus (criterion B). Although the fear is identified as excessive or unreasonable (criterion C), the person avoids the phobic stimulus or deals with it under anxiety or distress (criterion D), which interferes significantly with functioning (criterion E). Specific phobias are grouped under four main types: animal (e.g. dogs, spiders), natural environment (e.g. heights, water), blood-injection-injury, and situational (e.g. airplanes, elevators). Prevalence of specific phobia subtypes might fluctuate throughout the lifespan. For example, fear of animals is thought to decrease with age, whereas situational fears occur more often later in life (Fredrikson et al., Reference Fredrikson, Annas, Fischer and Wik1996). A recent study found that the overall prevalence of specific phobias in a sample of 462 persons aged 65 years or older was 11.5% (Kirmizioglu et al., Reference Kirmizioglu, Dogan, Kugu and Akyuz2009). The same prevalence for specific phobias (11.5%) was found in a primary care setting, in a group of 303 Puerto Rican participants aged 50 years and older (Tolin et al., Reference Tolin, Robison, Gaztambide and Blank2005). These results are not conclusive but highlight the need to address specific phobias as a multiple entity, with heterogeneous age of onset, genotype and evolutionary underpinnings (Fredrikson et al., Reference Fredrikson, Annas, Fischer and Wik1996; Himle et al., Reference Himle, McPhee, Cameron and Curtis1989; Kendler et al., Reference Kendler, Karkowski and Prescott1999), as well as diverse developmental characteristics across the lifespan.

Several instruments exist either for diagnosing or for screening specific phobias. The Structured Clinical Interview for DSM-IV (SCID; First et al., Reference First, Spitzer, Gibbon and Williams1996) is a semi-structured interview, developed to diagnose Axis I disorders, both current episode and lifetime prevalence. The Anxiety Disorders Interview Schedule (ADIS-R; Di Nardo and Barlow, Reference Di Nardo and Barlow1988) was developed specifically to diagnose anxiety disorders, further focusing on differential diagnosis. Both interviews should be administered by expert clinicians and display adequate psychometric properties when used with older adults.

Notwithstanding the usefulness of diagnostic interviews, general screening instruments for specific phobias add additional value in signalling potential problem areas and age-specific issues. Within the specific phobias, the Fear Survey Schedule-II is frequently used in research (FSS-II; Geer, Reference Geer1965), listing 51 stimuli and situations that potentially evoke fear. More recently, Kogan and Edelstein (Reference Kogan and Edelstein2004) administered this instrument to a sample of older adults, along with an open-ended question about specific objects and situations feared. Through the analysis of the open-ended questions the authors found that the most common fears were age specific (e.g. mental decline, poor well-being of loved ones) and, as such, were not even included in the FSS-II. Aiming to overcome these limitations, the authors developed the FSS-II for older adults (FSS-II-OA; Kogan and Edelstein, Reference Kogan and Edelstein2004), which lists 22 fears, 11 of which are age related. Additionally, there are several screening instruments that tap specific phobias, such as spiders (Fear of Spiders Questionnaire; Szymanski and Odonohue, Reference Szymanski and Odonohue1995) and heights (Acrophobia Questionnaire; Cohen, Reference Cohen1977). Instruments such as these are not yet tested in older adults, and additional research that addresses this specific issue is required.

Differential diagnosis of specific phobia and panic disorder with agoraphobia

The misdiagnosis of panic disorders and phobias is common, as subjects with specific phobias frequently experience similar symptoms to those experienced in panic attacks (Craske, Reference Craske1991; Craske et al., Reference Craske, Widiger, Frances, Pincus, Ross, First and Davis1996), and in clinical settings panic disorder with agoraphobia is frequently associated with several subtypes of specific phobia (Starcevic and Bogojevic, Reference Starcevic and Bogojevic1997). Evidence suggests that the physiological symptoms are similar for both groups, with the exception of the blood-injuries-injections phobias (Vogele et al., Reference Vogele, Coles, Wardle and Steptoe2003). These symptoms are usually described as a fight or flight response, and include tachycardia and sweating. Furthermore, both disorders have common features. For instance, driving, flying, traveling, heights, and confinement to a small space are commonly circumscribed phobic situations, and are also commonly referenced as potential triggers for panic (Craske et al., Reference Craske, Widiger, Frances, Pincus, Ross, First and Davis1996). Despite the common features between both diagnoses, two important differences can be highlighted. First, persons with specific phobias do not display noticeable trait anxiety (Marks, Reference Marks1987), as their fear is limited to a specific set of objects or situations (APA, 2000). Second, the primary cause of the global state of anxiety usually manifested in panic disorder is the fact that subjects are unsure about the specific moment in which they will experience a panic attack, despite being aware of the triggering contexts (Craske et al., Reference Craske, Widiger, Frances, Pincus, Ross, First and Davis1996). In contrast, in specific phobia an external circumscribed stimulus is required to trigger the anxious reaction. In this way, predictability seems to be the most significant variable in the differential diagnosis between a specific phobia and panic disorder, since the physiological aspects of each condition overlap significantly.

Certain characteristics – such as severity, age of onset and prevalence – differentiate panic from other anxiety disorders. Late onset panic disorder in particular is less severe (Deer and Calamari, Reference Deer and Calamari1998), possibly due to biological factors (Flint et al., Reference Flint, Koszycki, Vaccarino, Cadieux, Boulenger and Bradwejn1998). Isolated panic attacks or subclinical panic symptoms are less frequent in older adults when compared with their younger counterparts (Depp et al., Reference Depp, Woodruff-Borden, Meeks, Gretarsdottir and DeKryger2005), and are usually not accompanied by significant family history of panic (Segui et al., Reference Segui, Salvador-Carulla, Marquez, Garcia, Canet and Ortiz2000), supporting the idea that exposure and conditioning might be privileged etiologic mechanisms in panic disorder in older cohorts.

Epidemiological studies indicate that panic disorder is rare in older adults (Bryant et al., Reference Bryant, Jackson and Ames2008). Low prevalence estimates of panic disorder were reported by Corna and colleagues (Reference Corna, Cairney, Herrmann, Veldhuizen, McCabe and Streiner2007), with a prevalence of 0.82% for 12 months and 2.45% for lifetime in a sample of 12,792 adults who were aged 55 years or older. However, the authors recognized that recall bias may undermine the validity of such low estimates (Corna et al., Reference Corna, Cairney, Herrmann, Veldhuizen, McCabe and Streiner2007). Participants who experienced panic attacks earlier in life might not be very accurate in reporting the frequency and intensity of the events. Additionally, a diagnostic shift towards dementia may result in a significant amount of variability (Krasucki et al., Reference Krasucki, Howard and Mann1998), as the decrease in the prevalence of anxiety disorders occurs together with an increase in cognitive decline. As dementia can be accompanied by anxiety symptoms, as well as agitation, clinicians may be more likely to attribute anxiety symptoms to cognitive impairment in older adults.

Contrary to the low prevalence of panic disorder found among people over 65 years of age, specific phobia is highly prevalent in this group. Similar to younger cohorts, older adults with specific phobias usually do not manifest generalized anxiety. Manela et al. (Reference Manela, Katona and Livingston1996), after assessing 694 older adults, proposed that phobic disorder in old age should be regarded as distinct from the generalized anxiety spectrum. Predictability seems to be once again the significant variable in the differential diagnosis between specific phobia and panic disorder. Nonetheless, to better distinguish specific phobias from panic attacks in older adults it is essential to determine the physiological responses to specific triggers and examine if these responses are also diminished in this age group. This was suggested by Flint and colleagues (Reference Flint, Koszycki, Vaccarino, Cadieux, Boulenger and Bradwejn1998), who found that CCK-4, a panicogenic agent, induced fewer and less enduring panic symptoms in those aged 65 years old or older.

Notwithstanding the potentially low prevalence of late onset panic disorder, risk factors exist for being diagnosed for the first time in old age. The main factors are female gender, low educational levels, external locus of control, adverse life experiences, having a chronic disease, and losing a partner (Beekman et al., Reference Beekman1998). Several measures exist to aid the assessment of panic disorder, apart from the diagnostic interviews mentioned above. The Anxiety Sensitivity Index (ASI; Reiss et al., Reference Reiss, Peterson, Gursky and McNally1986) is a 16-item questionnaire that assesses not only anxiety symptoms but also their consequences, a mechanism usually present in panic disorder. A confirmatory factor analysis performed with a sample of 322 participants aged 65 years or older indicated that the psychometric properties of the ASI were satisfactory (Mohlman and Zinbarg, Reference Mohlman and Zinbarg2000).

Differential diagnosis of specific phobia and agoraphobia

Avoidance of numerous situations is the most useful criterion in differentiating between specific phobias and agoraphobic avoidance (Craske et al., Reference Craske, Widiger, Frances, Pincus, Ross, First and Davis1996). In individuals with agoraphobia, apprehension occurs even when the subject is not expecting to be exposed to or to be in contact with a feared object or situation, producing a constant state of anxiety. Therefore, if avoidance is related to a specific incident or stimulus rather than to the more general fear of having a panic attack, then an additional diagnosis of specific phobia can be made (APA, 2000). Typically, specific phobias encompass harm, danger or contamination related to a very specific stimulus or situation. On the other hand, agoraphobia implies an avoidance of several situations in which a panic attack can potentially occur. Nonetheless, the generalized anxious anticipation that characterizes a panic attack, if circumscribed, can be part of a specific phobia (APA, 2000). The DSM-IV-TR (APA, 2000) highlights this important aspect of differential diagnosis using as an example a case study regarding a person who was not afraid of using elevators. After having experienced a panic attack while using an elevator, the subject developed a fear of this particular situation. The diagnosis of specific phobia (of elevators) remains in this case, even if the focus of apprehension is the possibility of having a panic attack rather than of falling or being locked in. According to the DSM criteria, if panic attacks are restricted to elevators, even when the source of fear is about having a panic attack, specific phobia is the most suitable diagnosis. However, if the person experiences panic attacks in other situations which they subsequently avoid, then the diagnosis of panic disorder linked with agoraphobia is more appropriate (APA, 2000). As such, the concreteness and specificity of the eliciting trigger is the most important feature when making this diagnosis. In this way, uniqueness as opposed to commonness of the avoidance behavior is the most significant variable in the differential diagnosis between a specific phobia and agoraphobia.

A recent study postulated that the prevalence of the subtypes of specific phobias differs between younger and older cohorts. Cohen et al. (Reference Cohen, Magai, Yaffee, Huangthaisong and Walcott-Brown2006) found that almost 9% of their sample, comprising 1074 participants over 55 years of age, met criteria for a current phobia, with the most prevalent subtypes being enclosed spaces (2.2%), any kind of travel (1.8%), going out of the house (1.1%), being home alone (1.1%), and special fears, including spiders or blood (3.8%). These results clearly indicate a tendency toward agoraphobic avoidance in this population, suggesting a special need to address this problem.

Factors specifically related to aging pose significant problems when diagnosing agoraphobia in the older population. Fears related to severe falls in older adults are one example of a complication that may be present with this particular differential diagnosis. After three years of follow-up, almost half of the 1282 older women assessed in a community study stated that fear of falling constrained their daily activities, and contributed to social isolation (Austin et al., Reference Austin, Devine, Dick, Prince and Bruce2007). Agoraphobia in later life is frequently seen as related to specific problems that arise with aging. Fall history appears to be an important contributor to the fear of falling (Howland et al., 1998) and this fear, in many cases, is not an anxiety disorder but a rational fear once age-related changes to gait and balance increase the probability of falling.

Fear of falling is already considered a subtype of specific phobia, designated within “space phobia” (APA, 2000), which is related to a very specific stimulus or situation. Though the final result of this fear may be apparent agoraphobia – that is, confinement to the house – an extreme fear of falling is better classified as a specific phobia. Aged persons usually have no fear of having a panic attack but rather fear the real injuries that can arise from a fall. The correct perspective for such points remains somewhat hard to pin down. It is essential that the phobic individual understands that the fear is irrational and disproportionate to the danger imposed by the object or situation. However, in some cases, reduced levels of physical activity, gait problems, impairments in balance and mobility, and obesity (see Austin et al., Reference Austin, Devine, Dick, Prince and Bruce2007), medication (Stolze et al., Reference Stolze, Klebe, Zechlin, Baecker, Friege and Deuschl2004) and osteoporosis (Iglesias et al., Reference Iglesias, Manca and Torgerson2009) and atherosclerosis (van Balkom et al., Reference van Balkom, Beekman, de Beurs, Deeg, van Dyck and van Tilburg2000) can create genuine challenges to an older adult. These specific situations should ideally be evaluated to differentiate specific phobia related to falls and fear of falling; agoraphobia in which the older patient has fear of falling as a consequence of a panic attack; and natural fear of falling due to balance and gait impairments that can generate a real, dangerous situation. In the latter case, improvement of balance and mobility problems is more appropriate than any anxiety-related intervention (Hatzitaki et al., Reference Hatzitaki, Amiridis, Nikodelis and Spiliopoulou2009).

There are several scales and questionnaires specifically developed for screening for agoraphobia. However, to the best of our knowledge, none of these measures has been validated with older cohorts. A viable alternative to screen agoraphobia in older adults is the Hopkins Symptom Checklist (HSCL; Derogatis et al., Reference Derogatis, Lipman, Rickels, Uhlenhut and Covi1974), a self-report inventory composed of five factors, such as somatization, interpersonal sensitivity, and anxiety. Although not a measure of agoraphobia, this scale has been used as an outcome measure for intervention in agoraphobia (Cox et al., Reference Cox, Ballenger, Laraia, Hobbs, Peterson and Hucek1988) and it is well accepted for use with older adults (Beck and Stanley, Reference Beck, Stanley, Antony, Orsillo and Roemer2001).

Differential diagnosis of specific phobia and post-traumatic stress disorder

From a behavioral perspective, post-traumatic stress disorder (PTSD) may seem to represent a stereotypical example of a specific phobia, in terms of etiology and symptoms, as there is an identified traumatic stimulus that triggers fear reactions whenever the person is faced with similar or associated situations. However, several differences exist (Foa et al., Reference Foa, Steketee and Rothbaum1989). First, whereas both anxiety responses and avoidance of specific situations or stimuli are prerequisite criteria for the diagnosis of specific phobia, they form only part of the more complex constellation of symptoms in PTSD cases. Second, it is not always possible to identify a traumatic incident related to a specific phobia, whereas the diagnosis of PTSD requires the presence of an identifiable trauma. Finally, there are several distinguishable symptoms of PTSD that are usually absent in specific phobias, such as nightmares, flashbacks, intrusive memories, exaggerated startle responses, sleep disorders, memory and concentration deficits, and chronic feelings of detachment (APA, 2000). Another factor that distinguishes these disorders is that feared situations or objects are extremely circumscribed and predictable in specific phobias, allowing the person to avoid them successfully, thus permitting anxiety control (Coelho and Purkis, Reference Coelho and Purkis2009). In contrast, circumstances that trigger fear within PTSD cases are rather pervasive and unpredictable and, as a consequence, the person is unable to avoid them to control their anxiety. From an etiologic perspective, both entities are also disparate, as in PTSD the precipitant event has a life-threatening dimension, which violates basic principles of safety and survival. Consequently, stimuli that were previously assessed as safe now have the potential to be associated with danger, causing the person to chronically fear the environment.

Despite these distinctions, it can be misleading to make a differential diagnosis that is based merely on the nature of the precipitant stressor. It is preferable to assess comprehensively all of the variables present at the time of the event. If the precipitating situation occurs in a familiar and safe context, the development of PTSD is more likely. With this in mind, it seems quite credible that civil wars are considered to be especially gruesome (Pizarro et al., Reference Pizarro, Silver and Prause2006) given their occurrence on familiar ground. Specific phobia and PTSD should thus be differentiated beyond an identified traumatic stimulus. Key features are: an identifiable trauma; several distinguishable symptoms (e.g. nightmares and flashbacks); and again, similar to agoraphobia, unpredictability that in this case is related to the environment, and subsequently, chronic fear of the environment.

Owing to increasing physical frailty, older adults may have greater exposure to life threatening situations than younger adults, thus being at additional risk for the development of PTSD (Beyer, Reference Beyer, Blazer, Steffens and Busse2004). Etiological mechanisms for post-traumatic stress disorder in older adults might also be dissimilar to those encountered in other life stages. Epidemiological data available for PTSD in older adults, mainly collected from war veterans and war survivors, has yielded prevalence figures from 0.9% up to 46% (Bryant et al., Reference Bryant, Jackson and Ames2008), a range so wide that it is difficult to interpret. Symptomatic patterns are expected to be similar throughout the lifespan, although it is possible that older adults report fewer symptoms than their younger counterparts (Weintraub and Ruskin, Reference Weintraub and Ruskin1999).

Risk factors for the development of PTSD within specific groups, such as Holocaust survivors, are female gender and low self-assessed health, whereas adverse events in early childhood and neuroticism are predictors for PTSD in older adults (van Zelst et al., Reference van Zelst, de Beurs, Beekman, Deeg and van Dyck2003b). This may suggest that once certain vulnerabilities are acquired they are difficult to overcome, even at a later age. Some life events associated with aging, such as retirement or loneliness (Macleod, Reference Macleod1994) may work as triggers for previous episodes of PTSD, reactivating the disorder. This is nonetheless an area that requires further investigation. The majority of the existing instruments to assess PTSD in older adults were developed with war veterans (for a review of these instruments, see Cook and O'Donnell, Reference Cook and O'Donnell2005). Recently, van Zelst and colleagues (Reference van Zelst, de Beurs, Beekman, Deeg, Bramsen and van Dyck2003a) validated the Self-rating Inventory for Posttraumatic Stress Disorder (SRIP) for community living older adults (Hovens et al., Reference Hovens, Vanderploeg, Bramsen, Klaarenbeek, Schreuder and Rivero1994), a 47-item scale based on DSM criteria for PTSD. Using the comprehensive international diagnostic interview as a gold standard, the SRIP had an area under the curve of 0.85, with a suggested cut-off point of 39 (van Zelst et al., Reference van Zelst, de Beurs, Beekman, Deeg, Bramsen and van Dyck2003a).

Differential diagnosis of specific phobia and obsessive-compulsive disorder

Both specific phobias and obsessive-compulsive disorder (OCD) are characterized by anxiety towards particular stimuli or situations and a consequent avoidance of related anxiety triggers. Similarly, both conditions respond well to exposure treatments (Marks, Reference Marks1987), and have high comorbidity. Recently, Mohammadi et al. (Reference Mohammadi, Ghanizadeh and Moini2007) suggested that approximately one-sixth of those diagnosed with specific phobia had comorbid OCD. Despite some resemblance, there are adequate distinctions between these disorders. According to the DSM IV-TR (APA, 2000), type of avoidance is the best criterion to distinguish specific phobia from OCD. For instance, OCD avoidance is related to the content of the obsession, such as contamination, whereas in phobias, avoidance is related to the stimulus itself. Another significant point of distinction is the associated thoughts experienced, as persons diagnosed with OCD commonly have recurrent thoughts and actions that require response prevention therapy to achieve the attenuation of related symptoms.

The triggers that initiate a fear response can be similar between these two conditions. Cleanliness, contamination and physical damage are common themes within OCD, but these can also be present in specific phobia diagnoses. Whereas OCD is characterized by disgust and discomfort, specific phobias are associated with fear and panic. An important and distinguishing feature between OCD and specific phobia is the amount of comfort experienced by subjects once they are no longer faced with the stimulus (Oldham et al., 1996). People with OCD usually feel uncomfortable for a long period of time after a real or imagined encounter with the feared stimulus or situation, in many instances ruminating and ritualizing for several hours before feeling comfortable and safe. As triggers can be more abstract in OCD (e.g. dirt) when compared with specific phobias (e.g. snake) consequent rituals and avoidance are more widespread in OCD. Additionally, it is possible to differentiate OCD and specific phobias based on the prospect of encountering the phobic stimulus. Whereas specific phobias are characterized by an anxious anticipation of coming across the stimulus or feared situation, subjects diagnosed with OCD mainly fear the imagined potential consequences of the stimulus (Marks, Reference Marks1987). In this way, it is possible that one who is diagnosed with OCD can eventually bring him or herself to contact the feared stimulus, for instance dirt, if he or she is subsequently able to reduce the discomfort, for example through repeated washing. Finally, it is possible to distinguish both disorders based on the complexity of the underlying cognitions and beliefs (Marks, Reference Marks1987). Commonly, the beliefs associated with specific phobias are quite straightforward when compared with the more elaborate and complex cognitions of OCD, for instance regarding the spread of infection and transmission of diseases, or situations of possible injury. In summary, compared with specific phobia, OCD is related to a higher magnitude of discomfort experienced after the stimulus has passed or has been avoided. In addition, OCD involves greater complexity of cognition, particularly as related to rumination and consequent compulsions.

Little is known about the progression of OCD in later life (Calamari et al., Reference Calamari, Janeck, Deer, Frost and Steketee2002). Epidemiological data obtained in community samples establish a prevalence of OCD in people over 65 years old between 0 and 2.44% (Bryant et al., Reference Bryant, Jackson and Ames2008). Bland et al. (Reference Bland, Newman and Orn1988), for instance, found a 6-month prevalence rate of OCD of 1.5% among persons aged over 65. Documentation of the onset of OCD for the first time in old age is often limited to case reports related to cerebral abnormalities (e.g. Weiss and Jenike, Reference Weiss and Jenike2000), although stressful life events at any time in the lifespan can in theory precipitate OCD (e.g. Petrikis et al., Reference Petrikis, Andreou, Pitsavas and Garyfallos2004; Velayuldhan and Katz, Reference Velayuldhan and Katz2006).

The phenomenological characteristics of non-organic OCD in older adults seem similar to those verified in younger populations (Teachman, Reference Teachman2007), albeit with a few differences, such as higher rates of obsessive thoughts regarding sins, more compulsions related to washing rituals, and less intrusive thoughts about symmetry (Kohn et al., Reference Kohn, Westlake, Rasmussen, Marsland and Norman1997). There are also some indications that OCD symptoms, such as hoarding (Grisham et al., Reference Grisham, Frost, Steketee, Kim and Hood2006; Ayers et al., Reference Ayers, Saxena, Golshan and Wetherell2009), may be particularly serious among older adults. It is not uncommon for older adults suffering from OCD to be afraid of hospitals due to the risk of contamination. In this case an additional diagnosis of specific phobia would not be appropriate, as the reason for the fear and avoidance of the hospital is the fear of contamination. In a situation where avoidance is exclusive to hospitals, however, illness phobia may be a necessary differential diagnosis, as it is possible that avoidance of physicians is due to the fear that some disease might be discovered (Desai et al., Reference Desai, Bruce and Kasl1999; Noyes et al., Reference Noyes, Carney and Langbehn2004).

Older adults might be more reluctant to admit certain intrusive obsessions, such as thoughts of a sexual nature, in which case assessment might be more successful when employing alternative methods, such as written rather than oral reports (Carmin et al., Reference Carmin, Pollard and Ownby1999). Although there are no published norms for older samples, some screening instruments for OCD have been used in studies with this age cohort. The Yale Brown Obsessive-Compulsive Scale (YBOCS; Goodman et al., Reference Goodman1989) is a measure of severity of OCD. In a study that compared symptoms between younger and older adults diagnosed with OCD, Kohn and colleagues (Reference Kohn, Westlake, Rasmussen, Marsland and Norman1997) found that the instrument held similar psychometric properties for both age groups, proving it adequate for older cohorts. However, more work both in the development of norms for existing instruments for use with older adults, as well as the important task of developing specific instruments for use with OCD in older adults, needs to be carried out.