Psychiatric Disorders in Outpatients With Borderline Intellectual Functioning: Comparison With Both Outpatients From Regular Mental Health Care and Outpatients With Mild Intellectual Disabilities

Abstract

Objective:

In the Netherlands, patients with borderline intellectual functioning are eligible for specialized mental health care. This offers the unique possibility to examine the mix of psychiatric disorders in patients who, in other countries, are treated in regular outpatient mental health care clinics. Our study sought to examine the rates of all main Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision, Axis I psychiatric diagnoses in outpatients with borderline intellectual functioning of 2 specialized regional psychiatric outpatient departments and to compare these with rates of the same disorders in outpatients from regular mental health care (RMHC) and outpatients with mild intellectual disabilities (IDs).

Method:

Our study was a cross-sectional, anonymized medical chart review. All participants were patients from the Dutch regional mental health care provider Rivierduinen. Diagnoses of patients with borderline intellectual functioning (borderline intellectual functioning group; n = 235) were compared with diagnoses of patients from RMHC (RMHC group; n = 1026) and patients with mild ID (mild ID group; n = 152).

Results:

Compared with the RMHC group, psychotic and major depressive disorders were less common in the borderline intellectual functioning group, while posttraumatic stress disorder and V codes were more common. Compared with the mild ID group, psychotic disorders were significantly less common.

Conclusion:

Mental health problems in people with borderline intellectual functioning may not be well addressed in general psychiatry, or by standard psychiatry for patients with ID. Specific attention to this group in clinical practice and research may be warranted lest they fall between 2 stools.

In the International Classification of Diseases, 10th Revision, there is no specific code for it, but in the DSM-IV-TR, as well as in DSM-5, borderline intellectual functioning is a V code that can be used to signify problems associated with subaverage intellectual performance.^1–3 These patients function in between people with average or above average intelligence and people with ID. Adults with borderline intellectual functioning are believed to suffer from high rates of psychiatric disorders and rely mostly on outpatient treatment.⁴ In most countries they are not considered part of the ID population and are treated in the same clinics as patients without ID. However, it is unknown to what extent the mix of disorders with which they present themselves in mental health outpatient clinics differs from patients without ID. It is also unknown to what extent this mix of disorders differs from patients with lower IQs. Most studies in the general population, as well as in outpatients, did not include people with borderline intellectual functioning.^5–7 This may be important, as the prevalence of mental health disorders has been reported to vary by the severity of the ID.⁸

Unlike most other countries, in the Netherlands, patients with borderline intellectual functioning and comorbid psychiatric disorders are eligible for the same specialist psychiatric services as patients with ID. For instance, the Dutch regional mental health care provider Rivierduinen has over 10 years of experience with 2 outpatient mental health care centres specialized in psychiatry and ID, mainly for patients with borderline intellectual functioning and mild ID, apart from outpatient clinics for people without ID. Well established referral pathways and focus on patients with borderline intellectual functioning as a separate group made referral of patients with borderline intellectual functioning and psychiatric disorders to specialized mental health care the default procedure. Using data from these 2 specialized outpatient clinics and from a general outpatient clinic of Rivierduinen operating in the same region, we were able to compare psychiatric morbidity of patients with borderline intellectual functioning with outpatients from RMHC and with outpatients with mild ID.

Clinical Implications

• In patients with PTSD and V codes, one should extend the diagnostic process to borderline intellectual functioning as a co-existing problem.

• Caregivers should be aware of the possible hidden morbidity of depressive disorders in patients with borderline intellectual functioning or mild ID.

Limitations

• Results apply only to outpatients and cannot be generalized to inpatients with more severe illness.

• Our study did not account for comorbidity of Axis II disorders.

Methods

Participants

Our study was a cross-sectional anonymized medical chart review. All participants were patients from the Dutch regional mental health care provider Rivierduinen.

We compared anonymized data of patients from 2 regional secondary care adult outpatient departments, specialized in psychiatry and ID (Kristal Centre for Psychiatry and Intellectual Disability in the Leiden and Gouda locations; borderline intellectual functioning and mild ID groups) with anonymized data of patients from a regular secondary care outpatient department (RMHC group). The borderline intellectual functioning and mild ID groups came from the complete catchment area of Rivierduinen and the RMHC group came from one particular region within this area (Katwijk, Zuid Holland, the Netherlands). Both groups consisted of outpatients registered on January 1, 2011. All diagnoses were the DSM-IV-TR² Axis I diagnoses as recorded in the official registration system of the electronic patient file.

In the borderline intellectual functioning and mild ID groups, diagnoses were based on the integrative approach of Došen.^9–11 For people with ID, the integrative diagnosis considers the developmental perspective as the fourth dimension in addition to the biopsychosocial model. In daily clinical practice, this means that to consider al 4 dimensions, patients are always assessed by at least a certified and experienced psychiatrist, a certified and experienced mental health psychologist, and an experienced psychiatric community worker. DSM-IV-TR diagnoses were formulated using the DM-ID criteria.¹² The DM-ID offers adaptations of DSM-IV-TR diagnostic criteria and provides guidelines for making accurate psychiatric diagnoses in patients with various levels of IDs. In the RMHC group, diagnoses were formulated using the DSM-IV-TR.²

Among the 599 registered at the 2 specialized centres, diagnostic information was available for 576 patients (95.8%). Among these 576 patients, 511 (85.3%) were diagnosed with either borderline intellectual functioning or mild ID. A total of 65 patients (11.3%) were excluded from our study because the level of ID was not known at the time (2.3%) or because they had moderate (8.9%) or severe (0.2%) ID. For the RMHC group, diagnostic information was available for 1054 of the 1254 registered patients (84.1%). Among these patients, 14 (1.1%) were diagnosed with borderline intellectual functioning. These patients were excluded from our analyses.

Within our organization, patients with average or above average intellectual functioning with PDDs are referred to a special centre for autism spectrum disorders, while patients with a PDD and ID are referred to the Kristal Centre for Psychiatry and Intellectual Disability. This would lead to a possible referral bias when comparing the rates of autism because patients with PDDs were underrepresented in the RMHC group, compared with the borderline intellectual functioning and mild ID groups. Therefore, patients with PDDs were excluded from our analyses. One hundred and twenty-four (24.3%) people were excluded from the borderline intellectual functioning and mild ID groups, and another 14 (1.1%) were excluded from the RMHC group because of a diagnoses of PDD.

A total of 235 people were diagnosed with borderline intellectual functioning and 152 had mild ID. Thus the final groups consisted of 235 participants with borderline intellectual functioning (borderline intellectual functioning group), 1026 participants from RMHC (RMHC group), and 152 participants with mild ID (mild ID group).

Measures

Demographic Variables and Diagnostic Categories

The following variables were collected for each patient from the electronic patient file: age, sex, level of ID, and DSM-IV-TR Axis I diagnoses. All DSM-IV-TR Axis I diagnoses recorded in the official registration system of the electronic patient file were registered. For analyses, the DSM-IV-TR diagnoses were categorized as follows: psychotic disorders (subdivided into schizophrenia and psychotic disorder NOS), mood disorders (subdivided into MDD, dysthymic disorder, MDD NOS, and BD), anxiety disorders (subdivided into PTSD, panic disorder, GAD, social and specific phobia, and OCD), somatoform disorders (subdivided into somatization disorder, undifferentiated somatoform disorder, conversion disorder, pain disorder, and hypochondria), and V codes. People with ADHD and impulse control disorders NOS were categorized together as ADHD and impulse control disorders. Alcohol abuse and alcohol dependence were categorized together as alcohol abuse and dependence. Cannabis abuse and cannabis dependence were categorized together as cannabis abuse and dependence. The remaining diagnoses were not analyzed because rates were too low or they were absent. These included cognitive disorders, tic disorders, sexual disorders, other substance use disorders, eating disorders, and sleep disorders.

Intelligence

In the borderline intellectual functioning and mild ID groups, level of ID was based on IQ testing, using the WAIS-III.^13–15 Participants were divided into 2 groups: borderline intellectual functioning (TIQ of <85 and ≥70) and mild ID (TIQ of 50 to 55 and <70). There was no IQ testing in the RMHC group.

Statistical Analyses

Demographic and clinical variables were compared among the borderline intellectual functioning group and the RMHC and mild ID groups using ANOVA with post hoc Bonferroni correction for continuous variables (for example, age) and chi-square tests for dichotomous and categorical variables (for example, sex and diagnoses). First, all 3 groups were compared using a chi-square test. Second, when overall differences were found, a comparison was conducted comparing the borderline intellectual functioning group with both the RMHC and the mild ID groups. In additional analyses, outcomes were corrected for sex and age using binary logistic regression. All analyses were performed using SPSS Statistics version 16.0 (SPSS Inc, Chicago, IL). When chi-square conditions were not met, percentages were given but no statistical analyses were performed. A conservative level of significance was set at P ≤ 0.01.

Results

Demographic Characteristics

As shown in Table 1, the RMHC group consisted of 1026 participants. A total of 235 people were diagnosed with borderline intellectual functioning and 152 had a mild ID. Even though the percentage of females was highest in the borderline intellectual functioning group (66.8%), there was no significant difference among the 3 groups (χ² = 5.65, df = 2, P = 0.06). There was a significant difference in mean age among the 3 groups (P < 0.001). The mean age was lowest in the borderline intellectual functioning group (33.4 [SD 12.5]) and highest in the RMHC group (44.3 [SD 16.6], P < 0.001). There was a significant difference in mean age between the borderline intellectual functioning and the RMHC group (P < 0.001), as shown in Table 2. There was no difference in mean age between the borderline intellectual functioning and the mild ID group.

Table 1.

Demographic characteristics and statistical comparisons of the percentages of DSM-IV-TR Axis I diagnoses between the RMHC, borderline intellectual functioning, and mild ID groups

Characteristics	RMHC group n = 1026	Borderline intellectual functioning group n = 235	Mild ID group n = 152	P
Demographic characteristics
Sex, female, n (%)	603 (58.8)	157 (66.8)	96 (63.2)	0.06
Age, years, mean (SD)	44.3 (16.6)	33.4 (12.5)	37.2 (13.6)	<0.001
Age, years, range	15.8–95.4	16.3–78.2	16.8–70.6
DSM-IV-TR Axis I diagnoses, mean (SD)	1.46 (0.7)	1.49 (0.7)	1.38 (0.6)	0.3
DSM-IV-TR Axis I diagnoses	n (%)	n (%)	n (%)	P
Psychotic disorders	151 (14.7)	16 (6.8)	23 (15.1)	0.005
Schizophrenia	85 (8.3)	3 (1.2)	8 (5.3)	<0.001
Psychotic disorder NOS	30 (2.9)	10 (4.3)	13 (8.6)	0.003
Mood disorders	371 (36.2)	41 (17.4)	31 (20.4)	<0.001
MDD	217 (21.2)	21 (8.9)	18 (11.8)	<0.001
Dysthymic disorder	37 (3.6)	6 (2.6)	4 (2.6)	—
BD	69 (6.7)	4 (1.6)	5 (3.3)	—
Mood disorder NOS	6 (0.6)	7 (3.0)	4 (2.6)	—
Anxiety disorders	237 (23.1)	81 (34.5)	45 (29.6)	0.001
PTSD	107 (10.4)	46 (19.6)	30 (19.7)	<0.001
Panic disorder	57 (5.6)	10 (4.3)	4 (2.6)	0.26
GAD	14 (1.4)	3 (1.3)	1 (0.7)	—
Social phobia	16 (1.6)	5 (2.1)	2 (1.3)	—
Specific phobia	6 (0.6)	5 (2.1)	1 (0.7)	—
OCD	17 (1.7)	7 (3.0)	3 (2.0)	—
Anxiety disorder NOS	58 (5.7)	13 (5.5)	6 (3.9)	0.69
Somatoform disorder	26 (2.5)	3 (1.3)	5 (3.3)	0.39
Somatization disorder	2 (0.2)	0 (0.0)	0 (0.0)	—
Undifferentiated somatoform disorder	10 (1.0)	2 (0.9)	2 (1.3)	—
Conversion disorder	4 (0.4)	0 (0.0)	3 (2.0)	—
Pain disorder	4 (0.4)	1 (0.4)	0 (0.0)	—
Hypochondria	8 (0.7)	0 (0.0)	0 (0.0)	—
ADHD or impulse control disorder	135 (13.2)	31 (13.2)	10 (6.6)	0.67
Alcohol abuse or dependence	68 (6.6)	11 (4.7)	5 (3.3)	0.18
Cannabis abuse or dependence	22 (2.1)	12 (5.1)	2 (1.3)	0.02
V codes	97 (9.5)	53 (22.6)	20 (13.2)	<0.001

^—= conditions for chi-square test are not met

Table 2.

Demographic characteristics and statistical comparisons of the percentages of DSM-IV-TR Axis I diagnoses between the borderline intellectual functioning group, the RMHC, and the mild ID group

Characteristic	Borderline intellectual functioning group n = 235	RMHC group n = 1026	Pa	Mild ID group n = 152	Pb
Demographic characteristics
Sex, female, n (%)	157 (66.8)	603 (58.8)	0.02	96 (63.2)	0.46
Age, years, mean (SD)	33.4 (12.5)	44.3 (16.6)	<0.001	37.2 (13.6)	0.06
DSM-IV-TR Axis I diagnoses	n (%)	n (%)	P	n (%)	P
Psychotic disorders	16 (6.8)	151 (14.7)	0.001	23 (15.1)	0.008
Schizophrenia	3 (1.2)	85 (8.3)	—	8 (5.3)	—
Psychotic disorder NOS	10 (4.3)	30 (2.9)	0.29	13 (8.6)	0.08
Mood disorders	41 (17.4)	371 (36.2)	<0.001	31 (20.4)	0.47
MDD	21 (8.9)	217 (21.2)	<0.001	18 (11.8)	0.35
Anxiety disorders	81 (34.5)	237 (23.1)	<0.001	45 (29.6)	0.32
PTSD	46 (19.6)	107 (10.4)	<0.001	30 (19.7)	0.97
V codes	54 (22.6)	97 (9.5)	<0.001	20 (13.2)	0.02

^aP value of statistical comparisons between the borderline intellectual functioning and the RMHC groups

^bP value of statistical comparisons between the borderline intellectual functioning and the mild ID groups

^—= conditions for chi-square test are not met

Comparison of Diagnoses

The mean number of DSM-IV-TR Axis I diagnoses did not differ significantly among the 3 groups. Comparisons of the percentages of diagnostic categories among the 3 groups are presented in Table 1. In Table 2, percentages of different disorder types are presented between the borderline intellectual functioning group, the RMHC group, and the mild ID group.

Psychotic Disorders

A significant difference was found in the presence of psychotic disorders among the 3 groups (χ² = 10.7, df = 2, P = 0.005), with the lowest rates (6.8%) in the borderline intellectual functioning group (Table 1). The rate of schizophrenia was highest in the RMHC group (8.3%) and lowest in the borderline intellectual functioning group (1.2%) (χ² = 15.5, df = 2, P < 0.001). The rate of psychotic disorders NOS was highest in the mild ID group (8.6%) and lowest in the RMHC group (2.9%) (χ² = 11.8, df = 2, P < 0.003). Table 2 shows that the rate of psychotic disorders was significantly lower in the borderline intellectual functioning group, compared with both the RMHC (14.7%) and the mild ID (15.1%) groups (χ² = 7.1, df = 1, P = 0.008).

Mood Disorders

There was a significant difference among the 3 groups in the overall presence of mood disorders (χ² = 40.6, df = 2, P < 0.001) (Table 1). This was also true for MDD (χ² = 23.7, df = 2, P < 0.001). The highest rates of overall mood disorders (36.2%) and MDD (21.2%) were in the RMHC group. The rate of BD was also highest in the RMHC group (6.7%). The overall presence of mood disorders and the rate of MDD did not differ between the borderline intellectual functioning (17.4%) and mild ID groups (20.4%) (Table 2). However, most groups of mood disorders were too small to conduct statistical tests.

Anxiety Disorders (Including PTSD)

The rate of all anxiety disorders taken together differed significantly among the 3 groups (χ² = 14.3, df = 2, P = 0.001) (Table 1). This difference was mainly due to the higher rate of PTSD in the borderline intellectual functioning (19.6%) and mild ID groups (19.7%), compared with the RMHC (10.4%) group (χ² = 21.1, df = 2, P < 0.001). The rates of panic disorder, GAD, social phobia, specific phobia, OCD, and anxiety disorder NOS did not differ much among the groups. Most groups of specific anxiety disorders were too small to conduct statistical tests. There was no significant difference in the overall rate of anxiety disorders, or in the rate of PTSD between the borderline intellectual functioning and the mild ID groups (Table 2).

Somatoform Disorders

There was no significant difference among the 3 groups in the overall presence of somatoform disorders (Table 1). The different categories of somatoform disorders were too small to conduct statistical tests.

Substance Abuse and Dependence

There was no difference in the presence of alcohol and (or) cannabis abuse or dependence among the 3 groups (Table 1).

ADHD and Impulse-Control Disorders

There was no significant difference in the percentage of diagnosed ADHD and impulse-control disorders among the 3 groups (Table 1).

V Codes

There was a significant difference in the percentage of diagnosed V codes among the 3 groups (χ² = 31.2, df = 2, P < 0.001) (Table 1). The percentage of V codes was over twice as high in the borderline intellectual functioning group (22.5%), compared with the RMHC group (9.5%) (χ² = 31.3, df = 1, P < 0.001) (Table 2).

Discussion

In the Netherlands, patients with borderline intellectual functioning are eligible for specialized outpatient mental health care, offering the opportunity to examine psychiatric comorbidity in a group of people intellectually functioning in between people with and without an ID and often going unnoticed in most countries. In our study, the rates of DSMIV-TR Axis I psychiatric disorders were compared among patients with borderline intellectual functioning (borderline intellectual functioning group), outpatients from RMHC (RMHC group), and outpatients with mild ID (mild ID group). To our knowledge, there are no previous studies specifically focused on the rate of psychiatric disorders of patients with borderline intellectual functioning in outpatient mental health care. Most striking differences, compared with the RMHC group, were the high rate of PTSD and V codes in the borderline intellectual functioning group and the low rates of psychotic disorders and MDD. Also compared with the mild ID group, psychotic disorders were significantly less common in the borderline intellectual functioning group.

The rate of psychotic disorders was lower in the borderline intellectual functioning group than in both the RMHC and the mild ID groups. Considering the association found in earlier studies between lower IQ scores and an increased risk for schizophrenia,^16–18 this is a notable finding. It is unlikely that borderline intellectual functioning is associated with less chance of becoming psychotic. Based on our experience, we can say that patients with psychosis and borderline intellectual functioning are frequently referred from our outpatient department to teams specialized in the treatment of psychoses, such as the early detection and intervention team and functional assertive community treatment,¹⁹ more so than patients with psychosis and mild ID who do not seem able to profit from the above-mentioned approaches. More research is needed to explore this further.

The low rates of mood disorders in patients with borderline intellectual functioning, compared with the RMHC group, is also notable. It is contrary to what may be expected from literature. Hurley et al⁵ found high rates of mood disorders. We do not think it to be very probable that having borderline intellectual functioning predisposes to less mood disorders than having a more severe ID. This is in line with the absence of differences in our sample between borderline intellectual functioning and mild disorders, in this respect. We do not think that our therapists missed these diagnoses either. However, patients with depression may have been referred less often to our department as a lower mood is less easily recognized and labelled as a possible disorder by significant others.^20,21

Compared with the RMHC group, PTSD was almost twice as common in the borderline intellectual functioning group, and rates did not differ between patients with borderline intellectual functioning and patients with mild ID. We know that patients with ID are more likely to experience traumatic events.^22–24 They are also more vulnerable to the disruptive effects of trauma, and thus to PTSD.¹² Thus far, PTSD in patients with borderline intellectual functioning is underexposed in the literature. Our data warrant more attention to this subject.

Another result that merits discussion was the high rate of V codes in the borderline intellectual functioning group, compared with both the RMHC and the mild ID groups. In the DSM-IV-TR, V codes are used to indicate other conditions that may be a focus of clinical attention. V codes include, for instance, codes for relational problems, occupational problems, and phase of life problems. In general, V codes are used to capture clinically significant distress or problems functioning in daily life.² At the least, the high rates of V codes in the borderline intellectual functioning group suggest that these patients present with complex problems: ID, psychiatric disorders, and clinically significant distress or problems functioning in daily life as signified by these V codes. The alternative explanation may be that this specialized clinic is more likely to record these additional diagnoses. More research is needed to explore this further.

Our study has several strengths. First, our study examines a large sample of borderline intellectual functioning outpatients, which is a population not considered part of the ID population in most countries. They are thought to be especially vulnerable for developing psychiatric disorders and less likely to receive adequate treatment.⁴ Second, in the borderline intellectual functioning and mild ID patients, the level of ID was always recently established and based on the standardized WAIS-III, which ensured that the labels of borderline intellectual functioning and mild ID were carefully applied. Third, a broad range of DMS-IV-TR diagnoses was included. Fourth, these diagnoses were the diagnoses as recorded in the official registration system of the electronic patient file, applied after a careful diagnostic process. Fifth, the findings are based on large samples from a naturalistic outpatient setting, making them generalizable to the clinical field of interest.

The results should also be interpreted in light of some limitations. First, issues of referral most likely introduced some bias. The borderline intellectual functioning group consisted of significantly more females than the RMHC group. In addition, the mean age in the borderline intellectual functioning and mild ID groups was lower than in the RMHC group. Using binary logistic regression to correct for sex and age did not alter the outcomes (data not shown). Post hoc analyses showed that the difference in sex was accounted for by the high rate of female patients with PTSD, borderline intellectual functioning, and mild ID. When patients with PTSD were excluded, there was no longer a difference in the male-to-female ratio (data not shown). The mean age was lowest in patients with borderline intellectual functioning. This could mean that people with borderline intellectual functioning develop psychiatric symptoms at a younger age. It could also mean that older people with borderline intellectual functioning are less likely to be referred to specialized services. In both the borderline intellectual functioning and mild ID groups there were many patients diagnosed with an autism spectrum disorder because of a special referral policy. They were excluded from analysis, which means that part of the initial sample was not included. However, most patients diagnosed with an autism spectrum disorder did not report other psychiatric disorders and only made up a minor part of the rates of psychiatric disorders in the ID groups, meaning the extent of the introduced bias probably is small. A second limitation is that results apply only to outpatients and cannot be generalized to inpatients with more severe illness. Third, there was no IQ testing in the RMHC group. Fourth, demographic information was limited and information on treatment was not available. Fifth, we did not account for comorbidity of Axis II disorders. Future research could investigate the rates and comorbidity of Axis I and II disorders in patients with borderline intellectual functioning.

Conclusion

In conclusion, results indicate that people with borderline intellectual functioning are most commonly diagnosed with PTSD and V codes. Compared with patients from RMHC, they are younger, less likely to be diagnosed with psychotic and mood disorders, and more likely with anxiety disorders, more specifically, PTSD. Compared with their peers with mild ID referred to the same service, they are less likely to be diagnosed with psychotic disorders. Perhaps, the results remind us that, in many countries, this invisible group in the middle may not be well addressed by general psychiatry, or by ID psychiatry. Specific attention may be warranted in clinical practice, as well as in research, lest they fall between 2 stools.

Abbreviations

ADHD

attention-deficit hyperactivity disorder

BD

bipolar disorder

DM-ID

Diagnostic Manual–ID

DSM

Diagnostic and Statistical Manual of Mental Disorders

GAD

generalized anxiety disorder

ID

intellectual disability

IQ

intelligence quotient

MDD

major depressive disorder

NOS

not otherwise specified

OCD

obsessive–compulsive disorder

PDD

pervasive developmental disorder

PTSD

posttraumatic stress disorder

RMHC

regular mental health care

TIQ

total IQ

WAIS

Wechsler Adult Intelligence Scale