Peer influence as a potential magnifier of ADHD diagnosis
Introduction
The prevalence of Attention Deficit and Hyperactivity Disorder (ADHD), a mental health disorder characterized by inattention, hyperactivity, and impulsiveness, has climbed over the past few decades; however, the cause of ADHD's expansion is not well-understood (American Psychiatric Association, 2013). Environmental factors, such as exposure to tobacco or other chemicals, can influence ADHD diagnosis, but there is little evidence to suggest that exposure to such biological factors is increasing or causing increased rates of ADHD (Braun et al., 2006, Nomura et al., 2010). Academics attribute much of ADHD's surge to ADHD's poor diagnostic reliability and changes in the DSM, which have broadened its diagnostic criteria (Conrad and Potter, 2000, Cuffe, 2005). However, previous research has been unable to identify why ADHD's poor diagnostic reliability should lead to increased rates of prevalence. Even studies that have used a single conservative diagnostic criteria for ADHD have found increased population rates of ADHD (Getahun et al., 2013). In other words, the cause of ADHD's increasing prevalence is still unknown.
To begin filling this gap, I explore whether peer influence confounds ADHD diagnosis. Although ADHD is an inherited disorder caused by genetic factors (Faraone et al., 2005), there are several ways that peer influence could affect a person's propensity to be diagnosed with ADHD. For example, a person that has friends who have ADHD may become more aware of ADHD and consider himself ADHD. Similarly, a person with ADHD friends might inadvertently learn behaviors that are conducive to an ADHD diagnosis, such as a tendency to interrupt others. For the same reasons, frequent interaction with highly mindful, organized, or supportive individuals may have a reverse effect on a person's propensity to be diagnosed with ADHD. If behaviors that are conducive to an ADHD diagnosis can spread through a social network, it could lead to poor diagnostic reliability, and, through varying mechanisms, could potentially raise population rates of ADHD.
Using longitudinal data of students and friendship networks within schools, I investigate whether self-reports of inattention are subject to peer influence. To test this proposition, I use a recently developed statistical model that jointly measures the evolution of social networks and behaviors (Snijders, 2001, Steglich et al., 2010). This is the first paper to study whether core ADHD traits are subject to peer influence, and how peer influence might lead to magnified rates of ADHD in America.
ADHD's poor diagnostic reliability is largely due to its vague diagnostic criteria. ADHD maps onto a large set of criteria, but lacks any conclusive, measureable indicators. Current standards advocate a multi-method approach to diagnosing ADHD where comorbid mental health disorders, family problems, school performativity, and social outcomes are factored into diagnosis (Escobar, 2005, Hoza, 2007, Nijmeijer et al., 2010). The DSM-5 also requires ADHD problems to be observed in multiple settings (American Psychiatric Association, 2013), so therapists typically ask for input from parents and teachers regarding a child's ADHD behaviors. However, this approach to diagnosing ADHD is particularly susceptible to social confounders. Over half of ADHD diagnoses are first suggested by children's teachers, but teachers are often poorly educated about ADHD, and are often targeted and influenced by pharmaceutical marketing efforts to refer children to doctors about ADHD (Edwards and Sigel, 2015, Loe and Feldman, 2007, Phillips, 2006, Sax, 2003). Social stigmatization of ADHD can influence individuals to mask their ADHD features during counseling (Canu et al., 2007). Social support and high intelligence can significantly improve an ADHD child's ability to function, masking ADHD's performativity-based diagnostic criteria (Antshel et al., 2009, George, 2012, Katz et al., 1997). In fact, a recent study has found that ADHD symptoms in young adults may be very context-dependent (Lasky et al., 2016). Social norms about demographic characteristics can also confound ADHD diagnosis, for example, doctors often consider it less problematic for a woman to act absent-minded - a trait often associated with ADHD - contributing to lower rates of ADHD diagnosis in women (Bruchmüller et al., 2012).
Although few say so directly, it is often implied by scholars and medical professionals that ADHD's increasing prevalence may be related to its poor diagnostic reliability (Moncrieff and Timimi, 2010, Rafalovich, 2005). Depending on the diagnostic criteria used, estimates of the prevalence of ADHD in America range from 1.5% to 19.9% (Cuffe, 2005). Slight differences in diagnostic criteria across countries causes significant variation in population rates of ADHD (Conrad and Bergey, 2014, Polanczyk et al., 2007). Although the majority of epidemiological studies of ADHD estimate prevalence at 3%–7%, this still suggests that rates of false positive and false negative diagnoses are high (Goodman, 2001). For many scholars, ADHD's poor diagnostic reliability is perceived as evidence that ADHD diagnosis is invalid. Instead, scholars argue that ADHD symptoms have become medicalized, a process where non-medical problems come to be seen as medical (Conrad, 2008, Singh et al., 2013), and increased rates of diagnosis is simply a result of changing cultural norms (Canino and Alegría, 2008, Timimi and Leo, 2009). ADHD's poor diagnostic reliability has also led to public and medical uncertainty regarding whether ADHD is being over-diagnosed (Rafalovich, 2005), or if ADHD is even a valid mental health diagnosis (Moncrieff and Timimi, 2010, Timimi and Leo, 2009). Nonetheless, researchers are still unclear about how ADHD's poor diagnostic consistency and susceptibility to social confounders could lead to increasing rates of ADHD diagnosis.
Although ADHD is an inherited disorder caused by genetic factors, peer influence could confound and influence rates of ADHD diagnosis through two related social processes: Spurious contagion and diagnosis-conversion. Spurious contagion arises when behaviors that are conducive to a diagnosis are socially influenced. For example, the DSM-5 lists fidgeting, excessive talking, difficulty waiting turns, and interrupting others as key hyperactive-impulsive symptoms of ADHD (American Psychiatric Association, 2013). These behaviors are typically perceived as impulsive, but they are also conversational styles, which could be socially learned (Sacks et al., 1974). Consequently, a person with impulsive friends might adopt their friends' impulsive habits, increasing their odds of becoming perceived and diagnosed as impulsive by a mental health professional. Likewise, spurious contagion could dampen diagnostic rates of ADHD if people with calm or highly attentive friends were more likely to learn habits that decreased their odds of becoming perceived as impulsive by a mental health professional. Such habits could be learned unconsciously, or through more direct functional support. Diagnosis-conversion occurs when a person's increased awareness of a disease causes him to report having that disease. For example, when a person's friends report having problematic ADHD behaviors, these ADHD traits might become more salient to him. Consequently, he may become more likely to report having ADHD traits, and seek treatment.
If peer influence confounds ADHD diagnosis, it could contribute to ADHD's increased rates of diagnosis. Diseases with diagnostic criteria that are confounded by peer influence sometimes spread across social networks, magnifying rates of diagnosis. For example, Liu et al. (2010) found evidence that diagnosis-conversion could have been an important factor in the recent spike of autism in California; living close to another child with autism increased a child's odds to become diagnosed with autism in the following year. Other behaviors that are susceptible to peer influence can spread when there are systematic changes in how people socialize. Since individuals need to interact and be exposed to each other's behaviors for behaviors to diffuse across individuals, social settings that effect whom individuals interact with can mediate behavior diffusion. For example, cross-school friendship is rare for young children, since students from different schools tend to have fewer opportunities to meet, converse, and forge friendships, therefore, the diffusion of behaviors among children from different schools is also rare. Even within a classroom; children who are seated close to each other have an easier time conversing, are more likely to view each other positively, and are more likely to become friends (Frank et al., 2013, van den Berg et al., 2012). When social settings facilitate friendships among individuals with a particular behavior, it can reinforce the frequency that individuals engage in that behavior; likewise, social settings that discourage interactions between individuals with differential behaviors can discourage the correction of such behaviors. Several policies that have been gaining traction in recent years may have had this effect on friendships patterns among students with and without ADHD, such as educational tracking and behavioral classroom management techniques (Evertson and Weinstein, 2013, Loveless, 2013).
Section snippets
Methods
Data were primarily drawn from three waves of the National Longitudinal Study of Adolescent to Adult Health (Add Health), a school-based network of adolescents (Harris, 2009). I used Add Health's initial in-school interview, collected September 1994–April 1995, as the first time period in the sample, Add Health's Wave 1 in-home interview, collected April 1995–December 1995 as the second time period in the sample, and Add Health's Wave 2 in-home interview, collected April 1996–December 1996 as
Results
Table 1 provides descriptive statistics for the Add Health sample by wave and by school. The meaning of the network statistics are as follows: Reciprocity refers to the rate that students reciprocate nominations; Transitivity refers to the rate of triadic closures – the rate that friends nominate friends of friends; Network degree refers to the average number of friends students nominate across each wave; Jaccard index refers to the relative fraction of stable friendships in the network.
Discussion
The frequency of ADHD diagnosis has been increasing in America, but the reason for this increase is unknown. Previous research suggests that ADHD's diagnostic ambiguity has played a role in its increasing pervasiveness, but has been unable to specify diagnostic confounders that could substantially increase rates of ADHD diagnosis. In this paper, I connected ADHD's diagnostic ambiguity with a concrete social mechanism - peer influence. I examined friendship networks and self-reports of
Conclusion
This article demonstrated that peer influence can confound measures of inattention, a core diagnostic trait for ADHD. The current study does not indicate whether peer influence has affected rates of ADHD diagnosis, but it opens up new avenues for research regarding this matter. Future work should examine whether institutional or other societal changes over the past two decades have facilitated the spread of inattentive behavior via peer influence, and whether peer influence has had a direct
Acknowledgements
This research uses data from Add Health, a program project directed by Kathleen Mullan Harris and designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris at the University of North Carolina at Chapel Hill, and funded by grant P01-HD31921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, with cooperative funding from 23 other Federal Agencies and Foundations. Special acknowledgment is due Ronald R. Rindfuss and Barbara Entwisle for
References (57)
- et al.
Knowledge and information about ADHD: evidence of cultural differences among African-American and white parents
Soc. Sci. Med.
(1998) - et al.
The impending globalization of ADHD: notes on the expansion and growth of a medicalized disorder
Soc. Sci. Med.
(2014) - et al.
Molecular genetics of attention-deficit/hyperactivity disorder
Biol. Psychiatry
(2005) Psychometric properties of the strengths and difficulties questionnaire
J. Am. Acad. Child Adolesc. Psychiatry
(2001)- et al.
ADHD in context: young adults' reports of the impact of occupational environment on the manifestation of ADHD
Soc. Sci. Med.
(2016) - et al.
Mental health and social networks in early adolescence: a dynamic study of objectively-measured social interaction behaviors
Soc. Sci. Med.
(2015) - et al.
Misuse and diversion of stimulants prescribed for ADHD: a systematic review of the literature
J. Am. Acad. Child Adolesc. Psychiatry
(2008) - et al.
Leveraging social influence to address overweight and obesity using agent-based models: the role of adolescent social networks
Soc. Sci. Med.
(2015) - et al.
Adult ADHD Self-report Scale-v1
(2003) Diagnostic and Statistical Manual of Mental Disorders:: DSM-5: ManMag
(2013)