IntroductionBackground

Individuals with developmental disabilities (DD) such as autism and intellectual disability (ID) experience mental health crises more frequently than the general population [,]. A broad range of psychiatric and medical conditions can contribute to the agitation, aggression, and self-injury that often characterize these crises [-]. Rates of anxiety (20%-77%), depression (10%-20%), expansive mood and bipolar disorder (5%-11%), and psychosis (5%-10%) among individuals with autism exceed those in neurotypical individuals [-]. Elevated rates of psychiatric disorders have also been identified in individuals with ID, notably for unspecified psychosis (4.8%), schizophrenia (3.9%), and bipolar disorder (8%) [-]. A history of trauma or abuse should also be considered in individuals with DD presenting in crisis [].

When psychiatric and medical conditions are recognized as factors contributing to a person’s mental health crisis, clear long-term treatment targets emerge. Nevertheless, for those with DD, co-occurring medical and psychiatric conditions are often unrecognized, leaving them vulnerable to experiencing diagnostic overshadowing. Diagnostic overshadowing occurs when disruptive behaviors in individuals with DD are attributed to their disability without consideration of other potential medical or psychiatric conditions that could contribute to their behavioral presentation [].

Self-, parent-, and caregiver-report mental health questionnaires provide an efficient means of screening for common psychiatric conditions in the neurotypical population. However, for those with DD, self-report questionnaires may be impeded by communication deficits or a limited capacity to reflect on internal experiences. Parent- and caregiver-report questionnaires normed in typically developing children may also provide inadequate mental health screening for those with ID because they often include items that are inapplicable to children with minimal language ability, exclude severe conditions that disproportionately affect children with DD (eg, mania and psychosis), and overlook the individualized manner in which psychiatric symptoms manifest in this population [,-].

The American Psychiatric Association and the National Association for the Dually Diagnosed published the Diagnostic Manual–Intellectual Disability in 2007, and subsequently, in 2016, the second edition (Diagnostic Manual–Intellectual Disability–Second Edition; DM-ID-2) [,]. These texts adapt the Diagnostic and Statistical Manual of Mental Disorders criteria to reflect their presentation in individuals with ID. The Psychopathology Instrument for Mentally Retarded Adults and the Psychiatric Assessment Schedule for Adults with Developmental Disabilities (PAS-ADD) operationalize adapted diagnostic criteria into structured interviews to provide a framework through which to identify psychiatric conditions in this population [,]. These interviews are quite lengthy and require training to administer. Even as an abbreviated semistructured interview, the Mini PAS-ADD Clinical Interview takes approximately 45 minutes to complete []. Existing parent- and caregiver-report psychiatric screening tools for individuals with ID create a more efficient and practical means of collecting information [-]; yet, the checklist format of parent- and caregiver-report questionnaires limits depth and scope, both of which are necessary when evaluating crises in a population with complex medical and mental health needs. In addition, there is a great need for the inclusion of items that query symptoms of common medical conditions (eg, epilepsy, gastrointestinal disorders, and poor dentition) that manifest with agitation and aggression and occur more frequently in individuals with DD [,,].

Sources of Distress is a survey developed for parents and caregivers (hereinafter collectively referred to as caregivers) that uses a web-based branching logic format to screen for mental health and medical conditions among individuals with DD who present in crisis. This tool informs the care of individuals experiencing distress and is intended for use when the severity or persistence of disruptive behavior prompts the consideration of medication intervention. Screening information endorsed by caregivers is organized into relevant psychiatric and medical categories within a report. This report ( []) is developed for the caregiver and can subsequently facilitate their shared decision-making process with health care providers as specific underlying conditions are evaluated. Sources of Distress aims to minimize diagnostic overshadowing and optimize the ability of the caregiver and the provider to recognize the presence of psychiatric and medical conditions that merit targeted intervention. The web-based branching logic format is adaptive in nature—optimizing caregiver and health care provider convenience and efficiency and minimizing caregiver burden for survey completion [].

Objectives

This paper aims to (1) describe the initial development of Sources of Distress; (2) report on the findings from focus group evaluations and expert reviews and indicate how this feedback shaped the subsequent version of the survey; and (3) present the results from the evaluation of validity for Sources of Distress after its implementation in the clinical setting. The Methods and Results sections are divided into 3 subsections (apart from the Ethical Considerations section in Methods) corresponding to the development, initial evaluation, and clinical implementation phases of Sources of Distress.

MethodsEthical Considerations

The University of Utah Institutional Review Board approved focus group activities for Sources of Distress content validation (IRB_00111975). Focus group participants provided informed consent and received compensation for their time in the form of an Amazon gift card worth US $50. The University of Utah Institutional Review Board approved with a waiver of consent for the retrospective records review, data collection, and subsequent deidentified data analysis for individuals for whom Sources of Distress was completed as part of their clinical care (IRB_00170868).

Early Survey Development

Funding for the development of Sources of Distress was provided by the Autism Council of Utah based in Murray, Utah, United States []. The development team comprised a triple board physician (pediatrics, general psychiatry, and child and adolescent psychiatry), an educational psychologist, a medical student, and a business consultant grandparent of a child with autism and ID. In the initial development phase, Sources of Distress was built in Qualtrics (Qualtrics International Inc) using a branching logic format to approximate the history-taking component of a DD psychiatric evaluation. This evaluation queries psychiatric symptom clusters, physical complaints, and psychiatric medical history to support the development of a diagnostic impression for which treatment recommendations could be made.

Multiple expert opinion sources were reviewed to identify pertinent screening categories and corresponding items to include in Sources of Distress. The expert sources included published literature, the DM-ID-2, the Mini PAS-ADD Clinical Interview, and the screening interview for the Kiddie Schedule for Affective Disorders and Schizophrenia–Present and Lifetime (a semistructured psychiatric diagnostic interview for children and adolescents) [,,]. As Sources of Distress is intended for use in the context of distress, the presence of at least 1 manifestation of a behavioral or emotional crisis must be endorsed to initiate survey questions.

Initial Survey EvaluationFocus Group Evaluation

In 2018 and early 2019, focus group participants were recruited from (1) a university-based outpatient program that provides medical and psychiatric care for individuals with DD across the lifespan and (2) the Autism Council of Utah (a community stakeholder organization for individuals and families affected by autism). Six focus groups were conducted that consisted collectively of parents (6/17, 35%), professional caregivers (6/17, 35%), and adults with both DD and the ability to provide verbal feedback (5/17, 29%). Participants completed Sources of Distress before attending the focus group and reported on specific items, missing items, item wording, and attribution of items to corresponding conditions. Interviews and discussions were transcribed and analyzed following the framework analysis of Ritchie and Spencer []. Inductive reasoning and the constant comparative method put forth by Strauss and Corbin [] were used to compare statements by parents, professional caregivers, and individuals with disability within and across focus groups.

Expert Review Evaluation

Revisions were made to Sources of Distress based on focus group feedback. Experts reviewed the revised survey version, and additional changes were made. The experts included a pediatrician and 2 child psychiatrists, all with national recognition for their clinical and research work in DD.

Clinical ImplementationOverview

Sources of Distress was implemented in various clinical settings to augment the clinical history-taking process—outpatient (primary care, neurology, developmental pediatrics, and psychiatry), emergency department, psychiatric inpatient, and residential care. Caregivers were given a link to the survey when their health care provider identified the need for expert support in managing severe agitation and aggression. All caregivers (231/231, 100%) completed the survey outside of the clinical setting. An informal or formal consult followed survey completion for a subset of individuals. In August 2020, the survey was transitioned from the Qualtrics platform to the REDCap (Research Electronic Data Capture; Vanderbilt University) platform to automate the Sources of Distress report generation using the custom template engine []. This external REDCap module was developed and has been maintained by the Integrated Research Informatics Services of British Columbia Children’s Hospital Research Institute [].

Survey Data Collection

Sources of Distress responses were collected from its first use in a clinical setting from February 2019 through June 2022. The following information was obtained: respondent type, individual characteristics, caregiver-reported diagnoses, current medications, distress manifestations, psychiatric symptoms, and medical symptoms, conditions, or concerns. When multiple caregivers reported on the same individual, responses were used from the caregiver closest to where the individual lived (eg, parent for a child living at home and professional caregiver for an individual living in a residential setting). Psychotropic medications were organized within the following mutually exclusive categories: antipsychotics, antidepressants, non-antidepressant anxiolytics, anticonvulsants, lithium, alpha-2 agonists, stimulants, and atomoxetine.

Consults

A medical decision-making support consultation took place after survey completion as either an informal or a formal consult for a subset of individuals. This consult was conducted by a clinical team led by the triple board physician member of the survey’s development team. The consult team used DM-ID-2 criteria as the basis for establishing psychiatric diagnoses. At a minimum (as an informal consult), the consult involved a discussion between a DD clinical expert and the referring provider. This discussion resulted in a collective determination of working diagnoses and treatment plan. A formal consult included the additional components of medical records review, caregiver interview, and direct participant evaluation. Psychiatric diagnoses that were not reported in the survey but discussed by the provider or documented in the medical record were included among preexisting diagnoses.

Working diagnoses were abstracted from formal and informal consult documentation and served as the standard to define true case status.

Mood Disorder Classification

The presence of a mood disorder among preexisting and working diagnoses was classified into mutually exclusive categories such that there was no overlap among individuals across mood disorder categories to allow for direct comparisons across preexisting diagnoses, survey screening status results, and working diagnoses. The following mood disorder classification hierarchy was used from highest to lowest: (1) episodic expansive mood, hypomania, mania, and bipolar disorder, hereafter collectively referred to as bipolar disorder, (2) disruptive mood dysregulation disorder (DMDD) and unspecified mood disorder, and (3) unipolar depression. If an individual had a diagnosis of bipolar disorder, regardless of what other mood disorder diagnosis was reported or identified, their mood disorder classification would be bipolar disorder. An individual was only classified with unipolar depression if (1) they had a depression diagnosis and (2) they had no other mood disorder diagnosis.

Statistical Analyses and Evaluation of Validity

Descriptive statistics and chi-square tests were conducted in SPSS (version 28.0; IBM Corp) with an α of .05 selected to assess statistical significance. Differences between surveys with an accompanying consult and those without were measured. Positive predictive value (PPV), negative predictive value (NPV), and accuracy rates were calculated for (1) preexisting diagnoses and (2) survey screening results with working diagnoses used as the determinant of true case status. We calculated 95% CIs for the binomial distribution of accuracy rates.

ResultsEarly Survey Development

lists the modules and corresponding items initially selected as the categories, characteristics, and symptoms to be queried by Sources of Distress. The initial version of the survey included scoring algorithms to determine positive screen status for the following conditions: anxiety, unipolar depression, bipolar disorder, psychosis, and attention-deficit/hyperactivity disorder (ADHD).

Table 1. Description of Sources of Distress and additions in response to focus group feedback.ModuleOriginal itemsAdded in response to feedbackIntroduction and demographicsRespondent’s relationship to the individual who is affected
Distress symptoms
Language ability
Age
Known diagnoses
Current medications
For professional caregivers: how long have you known the affected individual?
Added “increased fixation on certain things” and “changes in behavior such as increased isolation, social withdrawal” to distress symptoms
Is there a difference in language ability at the physician’s office? If so, is there something the provider can do to improve the individual’s ability to speak for themselves?
Behavior patterns and triggersCircumstances of disruptive behavior (recognized triggers, patterns, motivation and reinforcement, and location)
Query perceived function to behavior surrounding distress
SleepTime of sleep onset and awakening
Middle-of-the-night interruptions
Naps
Activities interfering with sleep onset or returning to sleep
Intermittent periods of decreased need for sleep
Food seeking as an activity interfering with sleep
Sleep apnea diagnosis and symptoms
Discomfort precipitating sleep disturbance
AnxietyLeading to significant outbursts or discomfort: transitioning activities, getting stuck on certain topics or things, and minor changes in daily activities
Panic and nightmares
Sensory sensitivity that leads to discomfort
Repeated checking or rituals, which interferes with daily activities
DepressionLess energy than usual, increased crying spells, sadness, irritability, isolative, loss of interest in activities typically enjoyed, and excess sleep
Injures self on purpose; if yes: location of injury and whether self-injury is causing discomfort?
Whether self-injury is concerning to parent or caregiver
Whether self-injury could be perpetuated by attention seeking or avoidance
ManiaEstablish baseline energy
Query discrete periods out of the blue lasting ≥2 days of increased energy compared to baseline, laughing or vocalizing for no clear reason, particularly happy or giddy, risk taking, sexually acting out, increased impulsivity, and decreased need for sleep
PsychosisAppearing to be responding to internal auditory or visual stimuli
Yelling angrily in a room where no one else is present as if yelling at someone who is not there
ADHDaDifficulty following through on instructions, avoiding task demands, easily distractible, fidgety or restless, high activity when expected to remain in 1 place, constantly moving, blurting into other people’s conversations, and demanding attention or desired items
General medical problemsQuery history of headaches, seizures, injuries that can be causing discomfort, thyroid abnormalities, and tooth pain
Could any of these issues be contributing to distress?
Are there unusual ways of responding to physical discomfort?
Added joint pain; ear, nose, or throat pain; and seasonal allergies
TraumabHistory of trauma
Related to trauma: avoidance, flashbacks, and nightmares
Hypervigilance
Gastrointestinal concernsbBowel movement frequency
Query history of constipation, stool accidents, frequent stomachaches, food allergies, and acid reflux. Could any of these issues contribute to distress?
Subsequent additions: changes in appetite, nausea, and variable bowel movements
Menstrual concernsb (for female patients only)Query presence of mood changes during menses, endometriosis, polycystic ovary syndrome, significant menstrual pain, excess bleeding during or between cycles, and anxiety surrounding periods. Could any of these conditions be leading to distress?
Birth control: oral contraceptives, hormonal IUDd, nonhormonal IUD, and Depo-Provera (a contraceptive injection).
Dental concernsbWhen was the last dental visit?
Query presence of changes in eating patterns: texture preference, sensitivity to hot or cold food or drink preference for eating on 1 side of the mouth, and reduced oral intake
Grinding teeth

aADHD: attention-deficit/hyperactivity disorder.

bModule added in response to focus group feedback.

cN/A: not applicable.

dIUD: intrauterine device.

Initial Survey EvaluationFocus Group Feedback

During the focus groups, 3 main themes emerged in this analysis.

Theme A: respondents gave overall positive feedback regarding existing content and specific feedback regarding areas where there was room to expand content. describes the modules and items added in response to this feedback. Notably, a posttraumatic stress disorder (PTSD) module was added along with a PTSD scoring algorithm to determine positive screen status.Theme B: most of the respondents (15/17, 88%) agreed that the symptoms queried matched their understanding of the psychiatric and medical conditions to which they are attributed.Theme C: all participant groups reported positive acceptability of the branching logic format and time required to complete the measure.Expert Review

Overall, the expert review supported the Sources of Distress categories and respective items attributed to each condition. One expert recommended adding items that query gender and replacing sex as the basis for pronoun selection within the tool and its report. This expert also suggested that the report include screening results for each psychiatric condition. The former recommendations were implemented when Sources of Distress was transitioned to the REDCap platform. The latter recommendation was deferred until after screening algorithms are validated in a clinical setting.

Clinical ImplementationSample Characteristics

Surveys (N=264) were completed by parents or guardians (n=200, 75.8%), professional caregivers (n=43, 16.3%), and other caregivers (n=21, 8%) of 231 individuals (n=161, 69.7% men and boys; n=69, 29.9% women and girls; and n=1, 0.4% other; mean age 17.7, SD 10.3; range 2-65 years). Informal (n=62, 41.6%) and formal (n=87, 58.4%) consults were performed for 149 individuals collectively. presents sample characteristics, the manifestations of distress, and a comparison between individuals with a consult and those without.

Table 2. Sample characteristics and distress manifestations.CharacteristicsWith consulta (n=149), n (%)Without consult (n=82), n (%)Total (N=231), n (%)Chi-square (df)P valueGenderb2.3 (2).34
Man or boy102 (68.5)59 (72)161 (69.7)


Woman or girl47 (31.5)22 (25.5)69 (29.9)


Otherb0 (0)1 (1.2)1 (0.4)

Caregiverc7.8 (2).02
Parent or guardian108 (72.5)71 (86.6)179 (77.5)


Professional caregiver32 (21.5)6 (7.3)38 (16.5)


Other9 (6)5 (6.1)14 (6.1)

Age range (y)7.4 (2).03
<1346 (30.9)38 (46.3)84 (36.4)


13-2257 (38.3)30 (36.6)87 (37.7)


>2246 (30.9)14 (17.1)60 (26)

Language ability0.0 (2).99
Full verbal ability76 (51)42 (51.2)118 (51.1)


Limited use of words46 (30.9)25 (30.5)71 (30.7)


Nonverbal27 (18.1)15 (18.3)42 (18.2)

Manifestation of distress
Agitation130 (87.2)70 (85.4)200 (86.6)0.2 (1).69
Aggression97 (65.1)50 (61)147 (63.6)0.4 (1).53
Change in sleep86 (57.7)38 (46.3)124 (53.7)2.8 (1).10
Moodiness122 (81.9)65 (79.3)187 (81)0.2 (1).63
Increased fixation115 (77.2)57 (69.5)172 (74.5)1.6 (1).20
Change in eating patterns52 (34.9)19 (23.2)71 (30.7)3.4 (1).07
Change in personality99 (66.4)59 (72.0)158 (68.4)0.7 (1).39
Change in behavior96 (64.4)45 (54.9)141 (61)2.0 (1).15
Self-injurious behavior73 (49)39 (47.6)112 (48.5)0.0 (1).84Type of disability
Autism without IDd52 (34.9)47 (57.3)99 (42.9)10.9 (1)<.001
ID without autism15 (10.1)7 (8.5)22 (9.5)0.1 (1).71
ID and autism74 (49.7)18 (22.0)92 (39.8)17.0 (1)<.001
Genetic syndromee17 (11)17 (20.7)34 (14.7)3.7 (1).06

aIncludes informal and formal consults.

bOne participant reported other as gender: no participants reported non-binary as gender.

cWhen multiple caregivers completed Sources of Distress, the report from the caregiver with whom the participant spends the most time was used in this table.

dID: intellectual disability.

eGenetic syndrome includes some individuals who also populate the autism or ID categories.

Preexisting Psychiatric Diagnoses

The presence of at least 1 preexisting psychiatric diagnosis was reported in 65.4% (151/231) of the individuals. Individuals who received a consult compared to those without a consult were more likely to have a caregiver-reported history of psychotic disorder (14/149, 9.4% vs 1/82, 1%; P=.02; ).

Table 3. Medical conditions, preexisting psychiatric diagnoses, and psychiatric screening results.CharacteristicsWith consulta (n=149), n (%)Without consult (n=82), n (%)Total (N=231), n (%)Chi-square (df)P valueMedical conditionsb
Gastrointestinal concerns82 (55)37 (45.1)119 (51.5)2.0 (1).15
Dental concerns32 (21.5)25 (30.5)57 (24.7)2.3 (1).13
Menstrual concernsc16 (53.3)5 (27.8)21 (43.8)3.0 (1).13
General

Headache26 (17.4)8 (9.9)34 (14.7)2.5 (1).11

Ear, nose, and throat concerns17 (11.4)8 (9.9)25 (10.8)0.2 (1).70

Seasonal allergies34 (22.8)13 (15.9)47 (20.3)1.6 (1).21

Injury pain14 (9.4)9 (11)23 (10)0.2 (1).70

Thyroid abnormalities5 (3.4)6 (7.3)11 (4.8)1.8 (1).18

Joint pain9 (6)3 (3.7)12 (5.2)0.6 (1).44

Seizures29 (19.5)16 (19.5)45 (19.5)0.0 (1).99Seizure History40 (26.8)18 (22)58 (25.1)0.7 (1).41Sleep disturbance124 (83.2)68 (82.9)192 (83.1)0.0 (1).95Preexisting psychiatric diagnoses
Any psychiatric condition103 (69.1)48 (58.5)151 (65.4)2.6 (1).11
Depressiond20 (13.4)14 (17.1)34 (14.7)0.6 (1).46
Bipolar disorderd21 (14.1)9 (11)30 (13)0.5 (1).50
Unspecified mood disorder or DMDDd,e20 (13.4)7 (8.5)27 (11.7)1.2 (1).27
Anxietyf62 (41.6)31 (37.8)93 (40.3)0.3 (1).57
PTSDg11 (7.4)4 (4.9)15 (6.5)0.6 (1).46
Psychotic disorder14 (9.4)1 (1.2)15 (6.5)5.8 (1).02
ADHDh50 (33.6)26 (31.7)76 (32.9)0.1 (1).78Psychiatric screening status
Any psychiatric condition146 (98)80 (97.6)226 (97.8)0.1 (1).83
Unipolar depressiond61 (40.9)30 (36.6)91 (39.4)0.4 (1).52
Episodic expansive mood and bipolar disorderd60 (40.3)28 (34.1)88 (38.1)0.8 (1).36
Anxiety130 (87.2)71 (86.6)201 (87)0.0 (1).89
PTSD37 (24.8)15 (18.3)52 (22.5)1.3 (1).26
Psychosis52 (34.9)15 (18.3)67 (29)7.1 (1).008
ADHD102 (68.5)56 (68.3)158 (68.4)0.0 (1).98

aIncludes informal and formal consults.

bMedical conditions perceived by the caregiver as contributing to the current presentation of distress.

cAnalysis for menstrual concerns restricted to female patients aged >12.

dUnipolar depression, unspecified mood disorder and disruptive mood dysregulation disorder, and episodic expansive mood and bipolar disorder are mutually exclusive categories.

eDMDD: disruptive mood dysregulation disorder.

fPreexisting diagnosis of obsessive-compulsive disorder is included within the anxiety disorder category.

gPTSD: posttraumatic stress disorder.

hADHD: attention-deficit/hyperactivity disorder.

Caregiver-Reported Medical Conditions

describes medical conditions reported by caregivers. Caregivers of 73.2% (169/231) of the individuals identified at least 1 physical concern that they perceived as contributing to distress. The most common conditions were gastrointestinal concerns (119/231, 51.5%), menstrual concerns (21/48, 44% of female patients aged >12 y), seasonal allergies (47/231, 20.3%), and seizures (45/231, 19.5%).

Psychiatric Screening Results

lists the frequency of positive psychiatric screening results. All but 2% (5/231) of the individuals screened positive for a psychiatric condition, with a mean of 2.8 (SD 1.1; range 0-5) conditions per individual. Of those who were classified as having bipolar disorder, 89% (78/88) screened positive for a recent depressive episode. Positive screen status for psychiatric conditions were similar between those with a consult and those without, except in the case of psychosis (52/149, 34.9% vs 15/82, 18%; P=.008).

Psychotropic Medication Use

reports on the frequency of medication use. Most of the individuals (194/231, 84%) were taking psychotropic medication, and the majority were receiving antipsychotics (142/231, 61.5%) and antidepressants (129/231, 55.8%).

Table 4. Medication use reported in Sources of Distress.MedicationWith consult (n=149), n (%)Without consult (n=82), n (%)Total (N=231), n (%)Chi-square (df)P valueAny medication146 (98)68 (82.9)214 (92.6)17.6 (1)<.001Any psychotropic medication136 (91.3)58 (70.7)194 (84)16.6 (1)<.001
Antipsychotic102 (68.5)40 (48.8)142 (61.5)8.7 (1).003
Antidepressanta87 (58.4)42 (51.2)129 (55.8)1.1 (1).29
Anxiolyticb66 (44.3)23 (28.0)89 (38.5)5.9 (1).02
Anticonvulsantc45 (30.2)12 (14.6)57 (24.7)6.9 (1).009
Lithium10 (6.7)7 (8.5)17 (7.4)0.3 (1).61
Alpha-2 agonist72 (48.3)27 (32.9)100 (43.3)5.1 (1).02
Stimulant and atomoxetine30 (20.1)14 (17.1)44 (19)0.3 (1).57

aSelective serotonin reuptake inhibitors, duloxetine, tricyclics, mirtazapine, and trazodone were included exclusively within the antidepressant category.

bBenzodiazapines, buspirone, hydroxyzine, beta-blockers, and prazosin were included exclusively within the anxiolytic category.

cAnticonvulsant medication use in the absence of a reported seizure history.

Working Psychiatric Diagnoses

Of the 149 individuals who received a consult, 148 (99.3%) were diagnosed with at least 1 psychiatric condition with a mean of 2.7 (SD 1.0; range 0-5) diagnoses per individual. The conditions identified were anxiety (129/149, 86.6%), ADHD (84/149, 56.4%), bipolar disorder (67/149, 45%), unipolar depression (33/149, 22.1%), PTSD (35/149, 23.5%), and psychosis (31/149, 20.8%). Furthermore, 25 (16.8%) of the 149 individuals were diagnosed with either unspecified mood disorder or DMDD. Nearly all individuals identified with psychosis (29/31, 94%) had a co-occurring mood disorder diagnosis: bipolar disorder (22/31, 71%), unipolar depression (5/31, 16%), and unspecified mood disorder or DMDD (2/31, 6%).

Evaluation of Validity

Sources of Distress accuracy rates ranged from 76% (95% CI 69%-82%) for ADHD to 91% (95% CI 85%-95%) for PTSD and exceeded those of preexisting diagnoses, except in the case of psychosis, for which the accuracy rates were equivocal (82%, 95% CI 75%-87%; ). The survey demonstrated higher NPVs (81%-98%) than PPVs (51%-78%) for all conditions, with the exceptions of anxiety (53% and 92%, respectively) and episodic expansive mood bipolar disorder (85% and 90%, respectively). Low PPVs were notable for depression (51%) and psychosis (54%).

Table 5. Association between consult diagnoses after completing Sources of Distress with preexisting psychiatric diagnoses and Sources of Distress screening status (n=149).Working diagnosisPreexisting psychiatric diagnosisaSources of Distress screening status
Case negative, nCase positive, nPPVb (%)NPVc (%)Accuracy rated, % (95% CI)Screen negative, nScreen positive, nPPV (%)bNPV (%)cAccuracy rated, % (95% CI)Unipolar depressione,f508278 (71-84)

519879 (71-85)
Case negative10610


8630



Case positive2310


231


Episodic expansive mood and bipolar disordere,f766062 (55-70)

908587 (81-92)
Case negative775


766



Case positive5116


1354


DMDDg and unspecified mood disordere458882 (75-88)

N/AhN/AN/A
Case negative11311


N/AN/A



Case positive169


N/AN/A


Anxiety disorderi972052 (44-60)

925387 (81-92)
Case negative182


1010



Case positive6960


9120


Posttraumatic stress disorder1008384 (77-89)

789591 (85-95)
Case negative1140


1068



Case positive2411


629


Psychotic disorderj648482 (75-87)

549782 (75-87)
Case negative1135


9424



Case positive229


328


Attention-deficit/hyperactivity disorder845866 (59-74)

748176 (69-82)
Case negative578


3827



Case positive4242


975