Importance
Patterns and trends of subthreshold DSM-IV mental health diagnoses for youth within US community treatment settings merit systematic research.
Objective
To quantify and assess temporal patterns of DSM-IV diagnoses not otherwise specified (NOS) among youth during physician office visits.
Design, Setting, and Participants
We conducted a retrospective study using psychiatric diagnostic data from the US National Ambulatory Medical Care Survey and the National Hospital Ambulatory Medical Care Survey (n = 16 295) from 1999 through 2010, combined in 4-year intervals. Using diagnoses from visits to physicians, we compared trends of the proportional distribution of the major psychiatric diagnoses with subthreshold criteria (coded as NOS) with proportions of diagnoses reaching full criteria.
Main Outcomes and Measures
Specific common psychiatric diagnoses NOS compared with full-criteria psychiatric diagnoses.
Results
Between the 1999-2002 and 2007-2010 periods, the proportion of US medical visits reporting DSM-IV NOS psychiatric diagnoses compared with the proportion reporting full psychiatric diagnostic criteria for youth aged 2 to 19 years rose prominently for major mood diagnostic subtypes. Among all visits for mood disorders, NOS visits grew proportionally 1.5-fold from 45.3% in the 1999-2002 period to 68.8% in the 2007-2010 period (P < .001). Among visits for bipolar disorder, NOS visits increased more than 18-fold, from 3.6% in the 1999-2002 period to 72.6% in the 2007-2010 period (P < .001). In addition, anxiety disorder NOS increased from 44.6% in the 1999-2002 period to 58.1% in the 2007-2010 period. Overall, NOS visits constituted 35.0% of the total psychiatric visits in 2007-2010 but represented 55.9% when attention-deficit/hyperactivity disorder codes were excluded.
Conclusions and Relevance
The expansion of subthreshold (NOS) DSM-IV diagnoses of mood disorder, bipolar disorder, and anxiety disorder in youth that has occurred since 1999 in all likelihood will continue in the DSM-5 era unless administrative efforts are made to alter this practice. Unspecified diagnoses lack research reliability and potentially increase the likelihood of off-label prescribing of psychotropic medication.
Substantial research evidence indicates that the overall prevalence of emotional and behavioral disorders in research-assessed populations has not increased during the past 2 to 3 decades. This conclusion may not apply strictly to certain specific disorders, such as bulimia and marijuana abuse,1,2 but has been reported consistently in research samples of youth and adults in the United States,3-7 Canada,8 the Netherlands,9 and England.10 Nevertheless, population surveys during this period have shown clearly that many more community-treated persons have been diagnosed with and treated for 1 or more psychiatric disorders than occurred in previous decades.5,11-15
Such increases in psychiatric diagnosis and treatment in the United States can be explained partially by the following factors. First, an increasing number of individuals have been diagnosed as having expanded, less severe, and broader categories of psychiatric disorders. In clinical assessment studies using DSM-III-R and DSM-IV criteria, for example, 40% to 50% of those interviewed and given a diagnosis of a DSM psychiatric disorder were judged to have a mild disorder (rated as the lowest severity of 5 gradations).16 Second, marketing by drug companies stressing the merits of medication treatment has a profound effect.17 Third, the availability of insurance coverage and outpatient treatment have increased in the United States. Finally, pharmacologic treatments for emotional disorders are increasingly accepted by the US public. Whether the increased diagnostic trends apply equally to all subtypes of a particular psychiatric diagnosis in children and adolescents is unclear.
This research report will focus on the frequency of DSM-IV psychiatric diagnoses not otherwise specified (NOS) that were recorded during US physician office visits for youth from 1999 through 2010. The NOS is a DSM-IV category that is characterized by consistently (>95%) meeting subthreshold but not full diagnostic criteria and by causing distress or impairment. Providers and investigators generally agree that NOS is coded when (1) diagnostic criteria are subthreshold, (2) there is uncertainty about the etiology of the diagnosis, (3) some relevant symptoms fall outside the listed criteria, and (4) insufficient diagnostic criteria are obtained during the examination.18-21
Initially, NOS codes were included in the DSM-I (4 unspecified diagnoses) and were expanded thereafter as NOS, although the terminology has been slightly modified in the DSM-5 to unspecified or other specified. Twenty NOS codes were included in the DSM-II and 30 in the DSM-IV; at present, 31 unspecified diagnostic codes are included in the DSM-5. Also of note is that mood disorder NOS and pervasive developmental disorder (PDD) NOS from the DSM-IV are not included in the DSM-5.
Specifically, this study will describe national trends in major psychiatric NOS diagnoses given to youth during community physician office visits during the calendar years 1999 through 2010. In the analyses, we compare the proportion of NOS diagnoses for specific common psychiatric disorders given during these visits over time with the proportion of full-criteria (FC) diagnoses recorded in the same category to document NOS diagnostic trends statistically. The discussion that follows will assess the benefits and drawbacks of including an NOS category in the nomenclature.
Study Design and Data Source
Because the data set consisted of deidentified data and was publicly available, the study was exempt from institutional review board review at the University of Maryland, Baltimore. In a cross-sectional design comparing visit data from annual surveys, this retrospective study analyzed data derived from the National Ambulatory Medical Care Survey (NAMCS) and the National Hospital Ambulatory Medical Care Survey (NHAMCS) for youth (aged 2-19 years) receiving ambulatory care during the calendar years 1999 through 2010 (total visits, 120 795; total psychiatric visits, 16 295). The NAMCS and NHAMCS are annual surveys conducted by the National Center for Health Statistics on a nationally representative sample of visits to physicians in office-based practices and hospital outpatient departments.22 The NAMCS and NHAMCS use a multistage probability design, which samples visits from primary sampling units (ie, a county, a group of adjacent counties, or a standard metropolitan statistical area) from physician practices according to specialty within these units and from patient visits within these practices. The NHAMCS uses outpatient visits in hospital settings within primary sampling units, in clinic outpatient departments, in emergency service areas within these hospitals, and in patient visits to these clinics. During the survey years, survey response rates varied from 58.3% to 70.4%, with a median response rate of 62.5% for the NAMCS. For the NHAMCS, the median response rate was 90.5% (range, 71.0%-95.6%). Each visit is assigned a visit weight as well as values for clustering and stratification to account for the complex multistage probability survey design.23 Physician and staff members provided sociodemographic data and diagnoses.
Main Independent Variable
To evaluate changes in patterns for psychiatric diagnoses, the primary independent variable was the period. Following the guidelines of the National Center for Health Statistics, we combined NAMCS and NHAMCS medical visit data from contiguous survey years to arrive at stable estimates. Medical visits were grouped into periods from 1999 to 2002, 2003 to 2006, and 2007 to 2010.
Psychiatric Diagnostic Codes
The primary dependent variable was the proportion of youth visits that had a psychiatric DSM-IV NOS diagnostic code compared with the proportion of youth visits with FC psychiatric diagnostic codes within the category. Psychiatric diagnoses were assessed using codes from the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). Diagnoses recorded by physicians using DSM-IV codes were matched to the ICD-9-CM codes. As many as 3 diagnoses were recorded for each visit.
The FC diagnostic codes for bipolar disorder included the following: 296.0, 296.1, and 296.4 through 296.7 for bipolar I disorder; 296.89 for bipolar II disorder; and 301.13 for cyclothymic disorder. The codes for anxiety disorders included 293.84, 300.2, 300.3, 300.8, 309.21, 309.81, 313.0, 313.2, and 313.89. The codes for depressive disorders included 296.2 through 296.3 for major depressive disorder (MDD) and 300.4 for dysthymic disorder. The codes for learning disorder included 315.0 through 315.2. The codes for PDD included 299.1, 299.9, and 299.00 for autistic disorder. The codes for tic disorders included 307.21 through 307.23. The codes for mental retardation included 317 and 318. The codes for adjustment disorders included 309.0, 309.23, 309.28, 309.3, and 309.4. The codes for disruptive behavior disorders included 312.81, 312.82, 312.89, and 313.81. The composite mood disorder category included all depressive and bipolar disorders. The codes for attention-deficit/hyperactivity disorder (ADHD) included 314.00 and 314.01.
Based on the total number of visits recorded for each diagnosis, we selected 10 NOS diagnostic categories by rank order that met National Center for Health Statistics reliability standards (≥30 visits per diagnosis). These diagnostic categories included 311 for depressive disorder NOS, 300.00 for anxiety disorder NOS, 312.9 for disruptive behavior disorder NOS, 296.90 for mood disorder NOS, 296.80 for bipolar disorder NOS, 299.80 for PDD NOS, 309.9 for adjustment disorder NOS, 315.9 for learning disorder NOS, 307.20 for tic disorder NOS, and 319 for mental retardation unspecified. The code for ADHD NOS was 314.9.
Other study covariates included sociodemographic and administrative characteristics of youth visits. For this study, age categories were adapted from US Census groups to cluster ages 2 to 9, 10 to 14, and 15 to 19 years so as to improve statistical power. Race/ethnicity was categorized as white or nonwhite (black, Hispanic, Native American, Pacific Islander, and Asian or >1 race). The data regarding payment source were collapsed into the mutually exclusive categories of private insurance (including self-payment) and public insurance (including Medicare, Medicaid, other government insurance, no charge, and unknown payment source). Other variables used in the analyses include sex, region of the country (Northeast, South, Midwest, and West-Pacific), and geographical area (metropolitan and nonmetropolitan). The term NOS is used interchangeably with subthreshold because both terms ultimately indicate not meeting full diagnostic criteria; furthermore, both terms routinely overlap in the psychiatric literature.24-27 However, 2 DSM-IV depressive disorder NOS diagnoses were not clearly identified as subthreshold. These disorders are termed depressive features in postpsychotic disorder of schizophrenia and disorders wherein the medical etiology of depressive symptoms is unclear.
All statistical analyses were performed at the visit level using commercially available software (SAS, version 9.3; SAS Institute Inc). To produce nationally representative estimates, we applied visit-specific weights to all youth visits. To produce reliable variance estimates, visit-specific clustering and stratification variables were used. Estimates based on fewer than 30 weighted visits are unreliable.23
To describe dynamic changes in the patterns of all youth visits over time, we compared differences in sociodemographic and administrative characteristics of youth visits across 3 study periods (1999-2002, 2003-2006, and 2007-2010) by population-weighted χ2 analyses. Among youth mental health visits (ie, youth visits with clinician-reported psychiatric diagnoses), we analyzed trends in psychiatric diagnostic groups across 12 years by population-weighted χ2 analyses. To elucidate NOS diagnostic trends across 12 years, we compared the proportion of NOS diagnoses given for selected psychiatric diagnostic groups over time with the proportion of FC diagnoses recorded for the same diagnostic groups.
Characteristics of Physician Office–Based Youth Visits Across 12 Years
Total physician office–based youth visits (N = 120 795) were analyzed for the study years 1999 through 2010. As a proportion of all youth visits, visits to physician offices for any psychiatric diagnosis increased from 7.7% in the 1999-2002 period to 8.3% in the 2007-2010 period. Table 1 describes demographic and administrative characteristics of youth with physician office visits. Across the 12 years, young children (aged 2-9 years) represented nearly half of the youth visits (48.3%). Most of the visits were made by youth who were white (64.7%-71.5%), privately insured (64.1%-72.9%), and residing in metropolitan areas (85.0%-88.0%). Compared with the 1999-2002 period, the proportion of physician office visits in the 2007-2010 period by youth who were nonwhite (35.3% vs 28.5%) and not privately insured (35.9% vs 27.1%) increased significantly. The pattern of youth visits did not differ appreciably by US region across the periods (Table 1).
NOS as a Proportion of Total Psychiatric Visits
We ranked the NOS diagnoses as weighted proportions of all psychiatric visits. In the 2007-2010 period, the most frequent DSM-IV subthreshold psychiatric diagnoses were depressive disorder NOS (8.1%), anxiety disorder NOS (7.3%), mood disorder NOS (3.9%), bipolar disorder NOS (3.6%), and disruptive behavior disorder NOS (2.2%). Following in rank order were PDD NOS (2.0%), learning disorder NOS (1.1%), tic disorder NOS (0.7%), psychotic disorder NOS (0.7%), and adjustment disorder NOS (0.6%).
Proportions of FC vs NOS Diagnoses
Table 2 conveys physician office visit data with their weighted percentages in relation to DSM-IV diagnoses and trends over time. We compared data for FC diagnostic patterns proportionally with NOS diagnostic patterns. The distinctive findings reveal that (1) FC ADHD represented approximately half of all the diagnoses recorded in psychiatric visits (48.3%); (2) the number of ADHD NOS visits was miniscule; (3) the composite FC mood disorder visits (including all FC depressive and bipolar disorder visits) decreased profoundly as the composite mood disorder NOS visits proportionally increased; (4) FC bipolar disorder visits dramatically decreased as bipolar disorder NOS visits expanded; (5) FC anxiety disorders experienced the same pattern of NOS increase; (6) depressive disorder NOS increased proportionally to 58.1% of the diagnostic category total; (7) PDD NOS decreased prominently as FC autistic disorder increased; and (8) learning disorder NOS rose from 15.1% of the learning disorder total in the 1999-2002 period to 60.3% in the 2007-2010 period.
NOS Diagnostic Trends Over Time
A marked shift occurred in the proportional distribution of FC diagnoses vs NOS diagnoses for certain psychiatric disorders (Figure 1 and Figure 2). Figure 1A reveals that, during the study period, the proportion of visits for composite FC mood disorders (combining all FC depressive, bipolar, and cyclothymic disorder visits) declined from 54.7% to 31.2%, whereas the proportion of visits for composite mood disorder NOS (combining visits for mood disorder NOS, bipolar disorder NOS, and depressive disorder NOS) rose from 45.3% to 68.8% (P < .001). Independently, visits for mood disorder NOS increased from 1.1% to 4.5% of the total number of psychiatric visits during the 12-year study period (χ22 = 17.79; P < .001).
Figure 1B shows that the largest temporal increase in NOS diagnoses was in the visits reporting a bipolar disorder diagnosis. Bipolar disorder NOS diagnoses rose more than 18-fold during the 12-year period from 3.6% to 72.6% (P < .001), whereas the diagnostic visits of FC bipolar I and II disorders decreased nearly 5-fold.
Figure 2A shows the decline in visits for FC anxiety disorder during the 12-year study period compared with an increase in visits for anxiety disorder NOS. The NOS proportion rose during the study period from 44.6% to 58.1% (P = .09).
Figure 2B shows the increase in visits for depressive disorders NOS during the study period. The proportion of visits for depressive disorders NOS rose from 52.5% in the 1999-2002 period to 58.1% in the 2007-2010 period. This increase was not statistically significant, but it should be noted that, over time, an increased majority (58.1%) of depressive disorder visit diagnoses became NOS.
Proportional Changes in NOS Diagnoses
In a number of disorders, we found no significant trends in the proportion of NOS diagnoses compared with the proportion of FC diagnoses between the 1999-2002 and 2007-2010 periods. These diagnoses included adjustment disorder NOS (42.8% to 34.8%), mental retardation unspecified (41.8% to 43.8%), and disruptive behavior disorder NOS (38.3% to 32.6%). Pervasive developmental disorder NOS had a significant decline from 52.4% to 37.8% (P = .02). Learning disorder NOS had a significant increase from 15.1% to 60.3% (P < .001), and tic disorder NOS showed an increase from 40.9% to 65.0% that approached statistical significance (P = .08).
Excluding FC ADHD Diagnoses
By far the most common FC psychiatric diagnosis was ADHD, representing 48.3% of all psychiatric diagnoses in the 2007-2010 period; however, its DSM-IV NOS diagnosis (314.9) was 0.4% of all ADHD (code 314) diagnostic visits (Table 2). Thus, the very low number of ADHD NOS diagnostic visits and the high number of ADHD FC visits profoundly altered the overall NOS distribution. From 1999 through 2010, when ADHD and all other FC diagnoses were combined, the NOS proportion of all psychiatric diagnostic visits increased from 31.2% to 35.0% of all psychiatric visits (P = .25). Excluding ADHD visits from total psychiatric visits resulted in a proportional increase of all NOS diagnostic visits from 48.7% to 55.9%, representing a significant (P = .02) rise during the 12-year study period.
A major finding in this national study is that the proportion of psychiatric visits for youth who received a DSM-IV diagnosis coded as NOS (compared with the proportion of visits coded with full diagnostic criteria) dramatically increased in internalizing diagnostic categories between the 1999-2002 and the 2007-2010 periods. This change was most apparent in the comparison between the composite mood disorders NOS and the composite FC mood disorders (Figure 1) and specifically in the diagnostic categories of bipolar disorder NOS, mood disorder NOS, anxiety disorder NOS, and depressive disorder NOS. During the 12-year study period, proportional increases in NOS diagnoses of these mood and anxiety categories rose significantly from 3.6% to 52.5% in the 1999-2002 period to 58.1% to 72.6% in the 2007-2010 period (Figures 1 and 2).
Two child psychiatry diagnostic categories did not experience a major NOS increase during the study period. The ADHD NOS category remained extremely small, with a diagnostic proportion of less than 0.4% over the 12-year assessment. However, ADHD subthreshold diagnoses have been found to be very common in community research studies.28-30 One suggested explanation for the almost exclusive use of FC to diagnose youth with features of ADHD is up-coding.14,31 This possibility is based largely on the explanation by Blader and Carlson14 for the profound recent diagnostic increases in bipolar disorder in youth.
In the PDD diagnostic category, youth were diagnosed increasingly as having autistic disorder (DSM-IV code 299.00) rather than PDD NOS (DSM-IV code 299.8) during the 12-year study period (47.6% to 62.2%; P = .02). This increase occurred in all likelihood because of the advantages of the more serious diagnosis for special education placement.32 As Judith Rapoport, MD, has noted in this respect: “We’ll call a kid a zebra if he needs to be called a zebra to get the educational and other services that he needs and deserves.”33(p69) Further evidence clarifying this pragmatic diagnostic shift to autistic disorders consists of research studies revealing that, unlike prevalence study results based on special education searches, diagnostic prevalence findings using community assessments consistently identified PDD NOS at least twice as often as autistic disorder.34-36
Is Use of NOS DSM Diagnoses Justified?
For diagnostic precision, it is important to document the fact that many youth with psychiatric symptoms do not meet full DSM criteria. This fact has been reported to be particularly frequent in primary care settings,37-40 in part because the more serious cases are referred to specialists. Similarly, in this physician office visit–based study, nonpsychiatric medical practitioners made proportionally more NOS internalizing diagnoses than psychiatrists (P < .001).
For clinicians, NOS categories serve a number of useful functions. The NOS categories “provide indispensable placeholders when patients definitely need …[diagnoses that] don’t fit existing molds.”41(p10) The categories apply also when “there is an appreciable level of diagnostic uncertainty—a useful thing when the simple fast answer is so often wrong and harmful.”41(p10-11) The NOS categories also provide the clinician time before committing to a firm diagnosis.42
For researchers, NOS diagnostic categories represent subthreshold levels of psychopathology that produce problematic symptoms and often predict future similar FC disorders.43 Nonetheless, numerous critics find that DSM NOS diagnoses have major limitations. Their concerns and criticisms follow.
How Precise and Useful Are NOS Diagnoses?
Rutter and Uher44(p523) critically concluded that the NOS diagnosis was a “rather unhelpful ‘rubbish basket’ diagnosis.” Paris45(p70) similarly termed the NOS code a “waste basket for the patient who does not meet criteria for any specific diagnosis.” First19(p468) termed the NOS diagnosis a “catch-all” category that results in “a loss of important clinical information.”
Do We Know Why Physicians Are Diagnosing NOS So Frequently?
Surveys of physician practice are uncommon. Psychiatric researchers in this field ascribe the large number of NOS diagnoses reported by primary care practitioners to the following: pediatricians are very busy, with a mean of 13 minutes per case46; they seldom review the diagnostic codes before recording a psychiatric diagnosis; they rarely report psychiatric comorbidities; and they commonly refer cases that have any complicated psychiatric feature.47-52 Mojtabai53 reported that more than 60% of community-treated adult patients who use antidepressants did not meet DSM-IV criteria. The author hypothesized that a major cause of increased antidepressant use related to “clinical uncertainty about diagnostic criteria and ambiguity regarding subthreshold syndromes.”53(p165)
Will Subthreshold DSM Diagnoses Increase Further?
In all likelihood, subthreshold mental health diagnoses will increase in the future. In their follow-up study of 1704 high school students, Lewinsohn et al54 found that 53% of these adolescents had 1 or more subthreshold DSM psychiatric disorders by young adulthood. Paris45(p15) added: “Admitting subclinical phenomena into a diagnostic classification is a very slippery slope. The lifetime prevalence of mental disorders could easily come to approach 100%.” Such an expansion of psychiatric diagnoses would be more likely to occur within spectrum disorder categories. This possibility led McClellan55(p643) to write: “Labeling wide swaths of the population as spectrum cases creates vast heterogeneity and confounds biological and intervention research.”
Effect of an Increase in Subthreshold Diagnoses on Off-Label Treatment
In depressive disorders, the only diagnostic indication for psychotropic medication treatment by the US Food and Drug Administration is MDD. Because only about one-third of youth with depressive disorder diagnoses were given an MDD diagnosis in this and other national data sets,56 those treated with psychotropic medication for other DSM-IV depression diagnoses would have been given off-label treatment. This consequence also holds true for bipolar diagnoses other than bipolar I disorder. On the effect of such treatment, Frances57(p2) wrote: “None of the suggested subthreshold disorders has a proven effective treatment. In fact, there is good reason to doubt whether currently available [NOS] treatments will have any specific positive effect.” Furthermore, “the cost, side effects, and complications [of subthreshold medication treatment] are considerable.”57
Subthreshold diagnoses also tend to increase prevalence rates of disorders and comorbidity, whereas FC diagnoses decrease these rates.58 Treating subthreshold psychiatric disorders can also result in the misallocation of needed treatment services away from those with chronic ailments,59 although—for some patients—treatment of subthreshold disorders may limit subsequent psychopathology.
Should Subthreshold Categories Become Part of a Spectrum Disorder?
Although the DSM-IV included bipolar disorder NOS as a subthreshold diagnostic category, some clinical investigators60-63 have established specific guidelines for this diagnosis in youth. These authors report that youth who meet their bipolar disorder NOS criteria are similar to and are as treatment responsive as youth who meet full diagnostic criteria for bipolar I disorder. Consequently, they combine bipolar I and II disorders and bipolar disorder NOS as a bipolar spectrum in their diagnostic and treatment analyses.
Other investigators64 find that the arbitrary criteria modifications and the shortening of the symptom duration required for bipolar disorder NOS described by the previous investigators60-63 is nonspecific and misleading. In fact, 11 major US investigators achieved an interclass correlation agreement of only 0.03 for bipolar disorder NOS in a recent attempt to measure the specificity of bipolar disorder diagnoses in youth.65 (A European reliability study showed similar low specificity.66) In addition, diagnostic criteria described by bipolar spectrum researchers are quite different from bipolar spectrum criteria used by community child psychiatrists,67 and interview measures used to support a bipolar disorder diagnosis in youth are mixed and inconsistent.68
System Suggestions About the Frequent Use of NOS Categories
Academicians increasingly are making major suggestions concerning subthreshold diagnoses. One suggestion is to merge related diagnostic features into a graded spectrum perspective. This process has been proposed for bipolar disorder, obsessive-compulsive disorder, and schizophrenia.69 Already, PDD NOS has been incorporated into an autism spectrum for DSM-5. The desire by many to record and use dimensional levels of psychopathology to increase diagnostic specificity is also strong.9,70 In this respect, Zimmerman et al71(p1917) wrote: “From a nosological perspective, the relatively high frequency of subthreshold diagnoses lends support to the dimensional rather than categorical approach toward classification.” Still another effort has been to weed out very vague NOS categories. As a result, the DSM-5 committee removed DSM-IV mood disorder NOS.
To increase the accuracy of the unspecified disorder category in the DSM-5, it seems possible to include a specifier as part of the diagnosis, preferably related to the degree of severity. The specifier could be modeled after the specifiers included within the diagnosis of MDD, one of which is rated as the degree of severity (from 1 to 4).
The diagnostic findings in this study are based on physician office visit data from a national probability sample of community medical treatment providers for youth. Although diagnoses were made by US medical providers using DSM-IV codes and were recorded as ICD-9 codes, the two were accurately matched. Psychiatric NOS prevalence data in adults were not included in this analysis; however, adult NOS findings are similar in many respects.72,73 Clinical diagnoses from visit data may not be as technically accurate as research diagnoses,74 although they provide useful information about medical diagnostic practice in the community. Details on severity and medication response associated with NOS diagnoses could not be assessed from the available data.
A driving force for relatively uncomplicated categorical psychiatric diagnoses is their practical time-saving value for clinicians of different professional backgrounds.50 Categorical diagnoses also simplify reimbursement efforts. Although the DSM-5 has added some dimensional diagnostic features, it still primarily depends on this categorical nomenclature and continues to be overinclusive.75 Other specified and unspecified diagnoses in the DSM-5 are generally coded the same as the DSM-IV NOS codes. Barring unforeseen marked scientific progress, which could fuel a large expansion in the use of dimensional DSM criteria, most of the unspecified increases in diagnosis described in this 1999-2010 DSM-IV data set are likely to continue into the near future.
Submitted for Publication: February 6, 2014; final revision received July 17, 2014; accepted July 21, 2014.
Corresponding Author: Daniel J. Safer, MD, 6310 Harford Rd, Baltimore, MD 21214 (dsafer@jhmi.edu).
Published Online: November 26, 2014. doi:10.1001/jamapsychiatry.2014.1746.
Author Contributions: Dr Safer had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Safer, Rajakannan, Zito.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Safer, Rajakannan, Zito.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Rajakannan, Burcu.
Study supervision: Safer, Zito.
Conflict of Interest Disclosures: None reported.
1.Rosen
DS. Eating disorders in children and young adolescents: etiology, classification, clinical features, and treatment.
Adolesc Med. 2003;14(1):49-59.
PubMedGoogle Scholar 3.Costello
EJ, Erkanli
A, Angold
A. Is there an epidemic of child or adolescent depression?
J Child Psychol Psychiatry. 2006;47(12):1263-1271.
PubMedGoogle Scholar 4.Eaton
WW, Kalaydjian
A, Scharfstein
DO, Mezuk
B, Ding
Y. Prevalence and incidence of depressive disorder: the Baltimore ECA follow-up, 1981-2004.
Acta Psychiatr Scand. 2007;116(3):182-188.
PubMedGoogle ScholarCrossref 5.Kessler
RC, Demler
O, Frank
RG,
et al. Prevalence and treatment of mental disorders, 1990 to 2003.
N Engl J Med. 2005;352(24):2515-2523.
PubMedGoogle ScholarCrossref 6.Murphy
JM, Horton
NJ, Laird
NM, Monson
RR, Sobol
AM, Leighton
AH. Anxiety and depression: a 40-year perspective on relationships regarding prevalence, distribution, and comorbidity.
Acta Psychiatr Scand. 2004;109(5):355-375.
PubMedGoogle ScholarCrossref 7.Polanczyk
GV, Willcutt
EG, Salum
GA, Kieling
C, Rohde
LA. ADHD prevalence estimates across three decades: an updated systematic review and meta-regression analysis.
Int J Epidemiol. 2014;43(2):434-442.
PubMedGoogle ScholarCrossref 8.Simpson
KR, Meadows
GN, Frances
AJ, Patten
SB. Is mental health in the Canadian population changing over time?
Can J Psychiatry. 2012;57(5):324-331.
PubMedGoogle Scholar 9.de Graaf
R, ten Have
M, van Gool
C, van Dorsselaer
S. Prevalence of mental disorders and trends from 1996 to 2009: results from the Netherlands Mental Health Survey and Incidence Study-2.
Soc Psychiatry Psychiatr Epidemiol. 2012;47(2):203-213.
PubMedGoogle ScholarCrossref 10.Spiers
N, Bebbington
P, McManus
S, Brugha
TS, Jenkins
R, Meltzer
H. Age and birth cohort differences in the prevalence of common mental disorder in England: National Psychiatric Morbidity Surveys 1993-2007.
Br J Psychiatry. 2011;198(6):479-484.
PubMedGoogle ScholarCrossref 11.Marcus
SC, Olfson
M. National trends in the treatment for depression from 1998 to 2007.
Arch Gen Psychiatry. 2010;67(12):1265-1273.
PubMedGoogle ScholarCrossref 12.Zito
JM, Safer
DJ, dosReis
S,
et al. Psychotropic practice patterns for youth: a 10-year perspective.
Arch Pediatr Adolesc Med. 2003;157(1):17-25.
PubMedGoogle ScholarCrossref 13.Moreno
C, Laje
G, Blanco
C, Jiang
H, Schmidt
AB, Olfson
M. National trends in the outpatient diagnosis and treatment of bipolar disorder in youth.
Arch Gen Psychiatry. 2007;64(9):1032-1039.
PubMedGoogle ScholarCrossref 14.Blader
JC, Carlson
GA. Increased rates of bipolar disorder diagnoses among US child, adolescent, and adult inpatients, 1996-2004.
Biol Psychiatry. 2007;62(2):107-114.
PubMedGoogle ScholarCrossref 15.Visser
SN, Danielson
ML, Bitsko
RH,
et al. Trends in the parent-report of health care provider–diagnosed and medicated attention-deficit/hyperactivity disorder: United States, 2003-2011.
J Am Acad Child Adolesc Psychiatry. 2014;53(1):34-46.e2. doi:10.1016/j.jaac.2013.09.001.
PubMedGoogle ScholarCrossref 16.Kessler
RC, Merikangas
KR, Berglund
P, Eaton
WW, Koretz
DS, Walters
EE. Mild disorders should not be eliminated from the
DSM-V.
Arch Gen Psychiatry. 2003;60(11):1117-1122.
PubMedGoogle ScholarCrossref 17.Raven
M, Parry
P. Psychotropic marketing practices and problems: implications for
DSM-5.
J Nerv Ment Dis. 2012;200(6):512-516.
PubMedGoogle ScholarCrossref 18.Dziegielewski
SF. DSM-IV-TR in Action.2nd ed. Hoboken, NJ: John Wiley; 2010:51.
19.First
MB. Clinical utility in the revision of the
Diagnostic and Statistical Manual of Mental Disorders (
DSM).
Prof Psychol Res Pract. 2010;41(6):465-473.
Google ScholarCrossref 20.Munson
CE. The Mental Health Diagnostic Desk Reference.2nd ed. New York, NY: Hawthorn Press; 2001:50.
21.Reichenberg
LW. DSM-5 Essentials: The Savvy Clinician’s Guide to Changes in Criteria. Hoboken, NJ: Wiley; 2014:4.
24.Axelson
DA, Birmaher
B, Strober
MA,
et al. Course of subthreshold bipolar disorder in youth: diagnostic progression from bipolar disorder not otherwise specified.
J Am Acad Child Adolesc Psychiatry. 2011;50(10):1001-1016.e3. doi:10.1016/j.jaac.2011.07.005.
PubMedGoogle ScholarCrossref 25.Bertha
EA, Balázs
J. Subthreshold depression in adolescence: a systematic review.
Eur Child Adolesc Psychiatry. 2013;22(10):589-603.
PubMedGoogle ScholarCrossref 26.Comer
JS, Gallo
KP, Korathu-Larson
P, Pincus
DB, Brown
TA. Specifying child anxiety disorders not otherwise specified in the
DSM-IV.
Depress Anxiety. 2012;29(12):1004-1013.
PubMedGoogle ScholarCrossref 27.Wesselhoeft
R, Sørensen
MJ, Heiervang
ER, Bilenberg
N. Subthreshold depression in children and adolescents: a systematic review.
J Affect Disord. 2013;151(1):7-22.
PubMedGoogle ScholarCrossref 28.Angold
A, Erkanli
A, Egger
HL, Costello
EJ. Stimulant treatment for children: a community perspective.
J Am Acad Child Adolesc Psychiatry. 2000;39(8):975-994.
PubMedGoogle ScholarCrossref 29.Balázs
J, Keresztény
A. Subthreshold attention deficit hyperactivity in children and adolescents: a systematic review.
Eur Child Adolesc Psychiatry. 2014;23(6):393-408.
PubMedGoogle ScholarCrossref 30.Döpfner
M, Breuer
D, Wille
N, Erhart
M, Ravens-Sieberer
U; BELLA Study Group. How often do children meet
ICD-10/DSM-IV criteria of attention deficit/hyperactivity disorder and hyperkinetic disorder? parent-based prevalence rates in a national sample: results of the BELLA study.
Eur Child Adolesc Psychiatry. 2008;17(1)(suppl 1):59-70.
PubMedGoogle ScholarCrossref 31.Whitely
M, Raven
M. The risk that
DSM-5 will result in a misallocation of scarce resources.
Curr Psychiatry Rev. 2012;8(4):281-286.
Google ScholarCrossref 32.Rosenberg
RE, Daniels
AM, Law
JK, Law
PA, Kaufmann
WE. Trends in autism spectrum disorder diagnoses: 1994-2007.
J Autism Dev Disord. 2009;39(8):1099-1111.
PubMedGoogle ScholarCrossref 33.Reiff
MI, Feldman
HM.
Diagnostic and Statistical Manual of Mental Disorders: the solution or the problem?
J Dev Behav Pediatr. 2014;35(1):68-70.
PubMedGoogle ScholarCrossref 35.Lauritsen
MB, Pedersen
CB, Mortensen
PB. The incidence and prevalence of pervasive developmental disorders: a Danish population-based study.
Psychol Med. 2004;34(7):1339-1346.
PubMedGoogle ScholarCrossref 36.Lecavalier
L, Gadow
KD, DeVincent
CJ, Houts
C, Edwards
MC. Deconstructing the PDD clinical phenotype: internal validity of the
DSM-IV.
J Child Psychol Psychiatry. 2009;50(10):1246-1254.
PubMedGoogle ScholarCrossref 37.Costello
EJ, Shugart
MA. Above and below the threshold: severity of psychiatric symptoms and functional impairment in a pediatric sample.
Pediatrics. 1992;90(3):359-368.
PubMedGoogle Scholar 39.Pincus
HA, Davis
WW, McQueen
LE. “Subthreshold” mental disorders: a review and synthesis of studies on minor depression and other “brand names.”
Br J Psychiatry. 1999;174(4):288-296.
PubMedGoogle ScholarCrossref 40.Olfson
M, Broadhead
WE, Weissman
MM,
et al. Subthreshold psychiatric symptoms in a primary care group practice.
Arch Gen Psychiatry. 1996;53(10):880-886.
PubMedGoogle ScholarCrossref 41.Frances
A. Essentials of Psychiatric Diagnosis.Rev ed. New York, NY: Guilford Press; 2013.
42.Maser
JD, Norman
SB, Zisook
S,
et al. Psychiatric nosology is ready for a paradigm shift in
DSM-V.
Clin Psychol Sci Pract. 2009;16(1):24-40. doi:10.1111/j.1468-2850.2009.01140.x.
Google ScholarCrossref 43.Shankman
SA, Lewinsohn
PM, Klein
DN, Small
JW, Seeley
JR, Altman
SE. Subthreshold conditions as precursors for full syndrome disorders: a 15-year longitudinal study of multiple diagnostic classes.
J Child Psychol Psychiatry. 2009;50(12):1485-1494.
PubMedGoogle ScholarCrossref 44.Rutter
M, Uher
R. Classification issues and challenges in child and adolescent psychopathology.
Int Rev Psychiatry. 2012;24(6):514-529.
PubMedGoogle ScholarCrossref 45.Paris
J. The Intelligent Clinician’s Guide to the DSM-5. New York, NY: Oxford University Press; 2013.
46.Ferris
TG, Saglam
D, Stafford
RS,
et al. Changes in the daily practice of primary care for children.
Arch Pediatr Adolesc Med. 1998;152(3):227-233.
PubMedGoogle ScholarCrossref 47.Goldberg
D, Simms
LJ, Gater
R, Krueger
PF. Integration of dimensional spectra for depression and anxiety into categorical diagnoses for general medical practice. In: Regier
DA, Narrow
WE, Kuhl
EA, Kupfer
DJ, eds. The Conceptual Evolution of DSM-5. Alexandria, VA: American Psychiatric Publishing Inc; 2011:19-31.
48.Jablensky
A, Kendell
RE. Criteria for assessing a classification in psychiatry. In: Maj
M, Gaebel
W, Lopez-Ibor
JJ, Sartorious
N, eds. Psychiatric Diagnosis and Classification. New York, NY: Wiley; 2002:1-25.
49.Jampala
VC, Sierles
FS, Taylor
MA. The use of
DSM-III in the United States: a case of not going by the book.
Compr Psychiatry. 1988;29(1):39-47.
PubMedGoogle ScholarCrossref 50.Westen
D, Heim
AK, Morrison
K, Patterson
M, Campbell
L. Simplifying diagnosis using a prototype matching approach. In: Beutler
LE, Malik
MC, eds. Rethinking the DSM. Washington, DC: American Psychological Association; 2002:224.
51.Zimmerman
M, Mattia
JI. Psychiatric diagnosis in clinical practice: is comorbidity being missed?
Compr Psychiatry. 1999;40(3):182-191.
PubMedGoogle ScholarCrossref 52.Zimmerman
M, Galione
J. Psychiatrists’ and nonpsychiatrist physicians’ reported use of the
DSM-IV criteria for major depressive disorder.
J Clin Psychiatry. 2010;71(3):235-238.
PubMedGoogle ScholarCrossref 53.Mojtabai
R. Clinician-identified depression in community settings: concordance with structured-interview diagnoses.
Psychother Psychosom. 2013;82(3):161-169.
PubMedGoogle ScholarCrossref 54.Lewinsohn
PM, Shankman
SA, Gau
JM, Klein
DN. The prevalence and co-morbidity of subthreshold psychiatric conditions.
Psychol Med. 2004;34(4):613-622.
PubMedGoogle ScholarCrossref 55.McClellan
J. Clinically relevant phenomenology: the nature of psychosis.
J Am Acad Child Adolesc Psychiatry. 2011;50(7):642-644.
PubMedGoogle ScholarCrossref 56.Valluri
S, Zito
JM, Safer
DJ, Zuckerman
IH, Mullins
CD, Korelitz
JJ. Impact of the 2004 Food and Drug Administration pediatric suicidality warning on antidepressant and psychotherapy treatment for new-onset depression.
Med Care. 2010;48(11):947-954.
PubMedGoogle ScholarCrossref 58.Dell’osso
L, Pini
S. What did we learn from research on comorbidity in psychiatry? advantages and limitations in the forthcoming
DSM-V era.
Clin Pract Epidemiol Ment Health. 2012;8(2):180-184.
PubMedGoogle ScholarCrossref 59.Cwikel
J, Zilber
N, Feinson
M, Lerner
Y. Prevalence and risk factors of threshold and sub-threshold psychiatric disorders in primary care.
Soc Psychiatry Psychiatr Epidemiol. 2008;43(3):184-191.
PubMedGoogle ScholarCrossref 60.Axelson
D, Birmaher
B, Strober
M,
et al. Phenomenology of children and adolescents with bipolar spectrum disorders.
Arch Gen Psychiatry. 2006;63(10):1139-1148.
PubMedGoogle ScholarCrossref 61.Birmaher
B, Axelson
D, Goldstein
B,
et al. Four-year longitudinal course of children and adolescents with bipolar spectrum disorders: the Course and Outcome of Bipolar Youth (COBY) study.
Am J Psychiatry. 2009;166(7):795-804.
PubMedGoogle ScholarCrossref 63.Hafeman
D, Axelson
D, Demeter
C,
et al. Phenomenology of bipolar disorder not otherwise specified in youth: a comparison of clinical characteristics across the spectrum of manic symptoms.
Bipolar Disord. 2013;15(3):240-252.
PubMedGoogle ScholarCrossref 64.Stringaris
A, Santosh
P, Leibenluft
E, Goodman
R. Youth meeting symptom and impairment criteria for mania-like episodes lasting less than four days: an epidemiological enquiry.
J Child Psychol Psychiatry. 2010;51(1):31-38.
PubMedGoogle ScholarCrossref 65.Menard
TV, Galanter
CA, Jensen
PS,
et al. Strategies for improved classification of pediatric bipolar biobank participants. Poster presented at: Annual Meeting of the American Academy of Child and Adolescent Psychiatry; October 2013; Orlando, FL.
66.Bakker
IM, Terluin
B, van Marwijk
HW, van Mechelen
W, Stalman
WA. Test-retest reliability of the PRIME-MD: limitations in diagnosing mental disorders in primary care.
Eur J Public Health. 2009;19(3):303-307.
PubMedGoogle ScholarCrossref 67.Galanter
CA, Pagar
DL, Oberg
PP, Wong
C, Davies
M, Jensen
PS. Symptoms leading to a bipolar diagnosis: a phone survey of child and adolescent psychiatrists.
J Child Adolesc Psychopharmacol. 2009;19(6):641-647.
PubMedGoogle ScholarCrossref 68.Galanter
CA, Hundt
SR, Goyal
P, Le
J, Fisher
PW. Variability among research diagnostic interview instruments in the application of
DSM-IV-TR criteria for pediatric bipolar disorder.
J Am Acad Child Adolesc Psychiatry. 2012;51(6):605-621.
PubMedGoogle ScholarCrossref 69.Nemeroff
CB, Weinberger
D, Rutter
M,
et al.
DSM-5: a collection of psychiatrist views on the changes, controversies, and future directions.
BMC Med. 2013;11(9):202. doi:10.1186/1741-7015-11-202.
PubMedGoogle ScholarCrossref 71.Zimmerman
M, Rothschild
L, Chelminski
I. The prevalence of
DSM-IV personality disorders in psychiatric outpatients.
Am J Psychiatry. 2005;162(10):1911-1918.
PubMedGoogle ScholarCrossref 72.Angst
J. Psychiatry NOS (not otherwise specified).
Salud Mental. 2009;32(1):1-2.
Google Scholar 73.Rucci
P, Gherardi
S, Tansella
M,
et al. Subthreshold psychiatric disorders in primary care: prevalence and associated characteristics.
J Affect Disord. 2003;76(1-3):171-181.
PubMedGoogle ScholarCrossref 74.Miller
PR, Dasher
R, Collins
R, Griffiths
P, Brown
F. Inpatient diagnostic assessments, I: accuracy of structured vs unstructured interviews.
Psychiatry Res. 2001;105(3):255-264.
PubMedGoogle ScholarCrossref 75.Polanczyk
GV. Dimensionality of childhood psychopathology and the challenge of integration into clinical practice.
Eur Child Adolesc Psychiatry. 2014;23(3):183-185.
PubMedGoogle ScholarCrossref