To send content items to your account,
please confirm that you agree to abide by our usage policies.
If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account.
Find out more about sending content to .
To send content items to your Kindle, first ensure firstname.lastname@example.org
is added to your Approved Personal Document E-mail List under your Personal Document Settings
on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part
of your Kindle email address below.
Find out more about sending to your Kindle.
Note you can select to send to either the @free.kindle.com or @kindle.com variations.
‘@free.kindle.com’ emails are free but can only be sent to your device when it is connected to wi-fi.
‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.
To present evidence on the burden and outcomes of co-morbidities among severely malnourished (SAM) children admitted to outpatient therapeutic programme (OTP) facilities in the conflict setting of Borno, Nigeria.
Retrospective medical chart review.
Children aged 6–59 months with SAM enrolled in OTP between June and November 2016 whose medical records were analysed. Only pneumonia and diarrhoea were examined due to data limitations. Stata software was used for descriptive, multivariate and survival analyses.
Records of 396 children with median age of 15 months were identified and analysed from the date of enrolment to exit from OTP. Mean length of stay in OTP was 61d, with co-infected SAM children having shorter stay (P=0·006). Of the total, 148 (37·4 %) had at least one co-morbidity (pneumonia or diarrhoea), of which thirty-nine (26·4 %) had both. Cumulative rate of mortality during follow-up time was 9·5 (95 % CI 6·0, 15·1) per 10 000 child-days; SAM children with co-morbidities were ten times more likely to die than those without (hazard ratio=10·2; 95 % CI 3·4, 31·0). In multivariable analysis, co-morbidity (P=0·01), oedema (P=0·003), dehydration (P=0·02) and weight on admission (P=0·01) were associated with mortality. Both recovery and defaulter rates (57·8 and 36·1 %, respectively) did not meet SPHERE standards.
Children with SAM and co-morbidities are less likely to survive, presenting a significant barrier in improving child survival. The findings call for integrated OTP models that incorporate clinical algorithms and ensure prompt referral for SAM children with co-morbidity.
Eating disorders are commonly regarded as complex psychiatric conditions, although this perception might be a product of the high levels of physical and psychological co-morbidity that are present in many such cases, rather than being about complexity in the eating disorder per se. This paper will consider the reasons that eating disorders are seen as complex, and whether or not that perceived complexity should be seen as a genuine reason to deviate from existing evidence-based cognitive behavioural therapy (CBT) protocols. Case examples will be used to illustrate how complex presentations can require clinicians to work skilfully with relatively simple formulations to achieve the best outcome, rather than using unnecessarily complex formulations and treatments. The importance of clear supervision is also stressed, as it can play a role in clinicians’ perception of the need for a complex or simple formulation, it can support the clinician in developing a collaborative, focused and efficient formulation, and it can keep the clinician on track with an evidence-based CBT approach.
Although specific phobia is highly prevalent, associated with impairment, and an important risk factor for the development of other mental disorders, cross-national epidemiological data are scarce, especially from low- and middle-income countries. This paper presents epidemiological data from 22 low-, lower-middle-, upper-middle- and high-income countries.
Data came from 25 representative population-based surveys conducted in 22 countries (2001–2011) as part of the World Health Organization World Mental Health Surveys initiative (n = 124 902). The presence of specific phobia as defined by the Diagnostic and Statistical Manual of Mental Disorders, fourth edition was evaluated using the World Health Organization Composite International Diagnostic Interview.
The cross-national lifetime and 12-month prevalence rates of specific phobia were, respectively, 7.4% and 5.5%, being higher in females (9.8 and 7.7%) than in males (4.9% and 3.3%) and higher in high- and higher-middle-income countries than in low-/lower-middle-income countries. The median age of onset was young (8 years). Of the 12-month patients, 18.7% reported severe role impairment (13.3–21.9% across income groups) and 23.1% reported any treatment (9.6–30.1% across income groups). Lifetime co-morbidity was observed in 60.5% of those with lifetime specific phobia, with the onset of specific phobia preceding the other disorder in most cases (72.6%). Interestingly, rates of impairment, treatment use and co-morbidity increased with the number of fear subtypes.
Specific phobia is common and associated with impairment in a considerable percentage of cases. Importantly, specific phobia often precedes the onset of other mental disorders, making it a possible early-life indicator of psychopathology vulnerability.
Various sources indicate that mental disorders are the leading contributor to the burden of disease among youth. An important determinant of functioning is current mental health status. This study investigated whether psychiatric history has additional predictive power when predicting individual differences in functional outcomes.
We used data from the Dutch TRAILS study in which 1778 youths were followed from pre-adolescence into young adulthood (retention 80%). Of those, 1584 youths were successfully interviewed, at age 19, using the World Health Organization Composite International Diagnostic Interview (CIDI 3.0) to assess current and past CIDI-DSM-IV mental disorders. Four outcome domains were assessed at the same time: economic (e.g. academic achievement, social benefits, financial difficulties), social (early motherhood, interpersonal conflicts, antisocial behavior), psychological (e.g. suicidality, subjective well-being, loneliness), and health behavior (e.g. smoking, problematic alcohol, cannabis use).
Out of the 19 outcomes, 14 were predicted by both current and past disorders, three only by past disorders (receiving social benefits, psychiatric hospitalization, adolescent motherhood), and two only by current disorder (absenteeism, obesity). Which type of disorders was most important depended on the outcome. Adjusted for current disorder, past internalizing disorders predicted in particular psychological outcomes while externalizing disorders predicted in particular health behavior outcomes. Economic and social outcomes were predicted by a history of co-morbidity of internalizing and externalizing disorder. The risk of problematic cannabis use and alcohol consumption dropped with a history of internalizing disorder.
To understand current functioning, it is necessary to examine both current and past psychiatric status.
Obsessive–compulsive disorder (OCD) is often co-morbid with depression. Using the methods of network analysis, we computed two networks that disclose the potentially causal relationships among symptoms of these two disorders in 408 adult patients with primary OCD and co-morbid depression symptoms.
We examined the relationship between the symptoms constituting these syndromes by computing a (regularized) partial correlation network via the graphical LASSO procedure, and a directed acyclic graph (DAG) via a Bayesian hill-climbing algorithm.
The results suggest that the degree of interference and distress associated with obsessions, and the degree of interference associated with compulsions, are the chief drivers of co-morbidity. Moreover, activation of the depression cluster appears to occur solely through distress associated with obsessions activating sadness – a key symptom that ‘bridges’ the two syndromic clusters in the DAG.
Bayesian analysis can expand the repertoire of network analytic approaches to psychopathology. We discuss clinical implications and limitations of our findings.
Anxiety disorders are highly prevalent in people with bipolar disorder, but it is not clear how many have anxiety disorders even at times when they are free of major mood episodes. We aimed to establish what proportion of euthymic individuals with bipolar disorder meet diagnostic criteria for anxiety disorders.
We performed a random-effects meta-analysis of prevalence rates of current DSM-III- and DSM-IV-defined anxiety disorders (panic disorder, agoraphobia, social anxiety disorder, generalized anxiety disorder, specific phobia, obsessive–compulsive disorder, post-traumatic stress disorder, and anxiety disorder not otherwise specified) in euthymic adults with bipolar disorder in studies published by 31 December 2015.
Across 10 samples with 2120 individuals with bipolar disorder, 34.7% met diagnostic criteria for one or more anxiety disorders during euthymia [95% confidence interval (CI) 23.9–45.5%]. Direct comparison of 189 euthymic individuals with bipolar disorder and 17 109 population controls across three studies showed a 4.6-fold increase (risk ratio 4.60, 95% CI 2.37–8.92, p < 0.001) in prevalence of anxiety disorders in those with bipolar disorder.
These findings suggest that anxiety disorders are common in people with bipolar disorder even when their mood is adequately controlled. Euthymic people with bipolar disorder should be routinely assessed for anxiety disorders and anxiety-focused treatment should be initiated if indicated.
The presence of obsessive-compulsive symptoms (OCS) and obsessive-compulsive disorders (OCD) in schizophrenia is frequent, and a new clinical entity has been proposed for those who show the dual diagnosis: the schizo-obsessive disorder. This review scrutinizes the literature across the main academic databases, and provides an update on different aspects of schizo-obsessive spectrum disorders, which include schizophrenia, schizotypal personality disorder (SPD) with OCD, OCD with poor insight, schizophrenia with OCS, and schizophrenia with OCD (schizo-obsessive disorder). An epidemiological discussion on the discrepancies observed in the prevalence of OCS and OCD in schizophrenia across time is provided, followed by an overview of the main clinical and phenomenological features of the disorder in comparison to the primary conditions under a spectral perspective. An updated and comparative analysis of the main genetic, neurobiological, neurocognitive, and pharmacological treatment aspects for the schizo-obsessive spectrum is provided, and a discussion on endophenotypic markers is introduced in order to better understand its substrate. There is sufficient evidence in the literature to demonstrate the clinical relevance of the schizo-obsessive spectrum, although little is known about the neurobiology, genetics, and neurocognitive aspects of these groups. The pharmacological treatment of these patients is still challenging, and efforts to search for possible specific endophenotypic markers would open new avenues in the knowledge of schizo-obsessive spectrum.
Researchers have studied psychological disorders extensively from a common cause perspective, in which symptoms are treated as independent indicators of an underlying disease. In contrast, the causal systems perspective seeks to understand the importance of individual symptoms and symptom-to-symptom relationships. In the current study, we used network analysis to examine the relationships between and among depression and anxiety symptoms from the causal systems perspective.
We utilized data from a large psychiatric sample at admission and discharge from a partial hospital program (N = 1029, mean treatment duration = 8 days). We investigated features of the depression/anxiety network including topology, network centrality, stability of the network at admission and discharge, as well as change in the network over the course of treatment.
Individual symptoms of depression and anxiety were more related to other symptoms within each disorder than to symptoms between disorders. Sad mood and worry were among the most central symptoms in the network. The network structure was stable both at admission and between admission and discharge, although the overall strength of symptom relationships increased as symptom severity decreased over the course of treatment.
Examining depression and anxiety symptoms as dynamic systems may provide novel insights into the maintenance of these mental health problems.
It is unclear whether health service use influences the association between psychiatric and physical co-morbidity and suicide risk in older adults.
Controls were older adults (n = 2,494) participating in a longitudinal study on the health of the elderly carried out between 2004 and 2007, in Quebec. The cases were all suicide decedents (n = 493) between 2004 and 2007, confirmed by the Quebec Coroner's office. Multivariate analyses were carried out to test the association between suicide and the presence of psychiatric and physical illnesses controlling for health service use and socio-demographic factors by gender and age group. Interaction terms were also tested between suicide and co-morbidity on outpatient service use.
The presence of physical illnesses only, was associated with a reduced risk of suicide across all sex and age groups. The presence of a mental disorder only was associated with an increased risk of suicide overall and specifically in females and those aged 70 to 84 years of age. Suicide risk was lower in those with a psychiatric and physical co-morbidity and consulting mental health services.
Increased mental health follow-up in older adults with psychiatric illnesses is needed for the detection of suicidal behavior and reducing suicide risk in males. Further research should focus on the mitigating effect of the presence of physical illnesses on stigma and health service use and the presence of social support in the elderly.
Major depressive disorder (MDD) and generalized anxiety disorder (GAD) often co-occur with somatic symptomatology. Little is known about the contributions of individual symptoms to this association and more insight into their relationships could help to identify symptoms that are central in the processes behind the co-occurrence. This study explores associations between individual MDD/GAD symptoms and somatic symptoms by using the network approach.
MDD/GAD symptoms were assessed in 2704 participants (mean age 41.7 years, 66.1% female) from the Netherlands Study of Depression and Anxiety using the Inventory of Depressive Symptomatology. Somatic symptoms were assessed with the somatization scale of the Four-Dimensional Symptom Questionnaire. The technique eLasso was used to estimate the network of MDD/GAD and somatic symptoms.
The network structure showed numerous associations between MDD/GAD and somatic symptoms. In general, neurovegetative and cognitive/affective MDD/GAD symptoms showed a similar strength of connections to the somatic domain. However, associations varied substantially across individual symptoms. MDD/GAD symptoms with many and strong associations to the somatic domain included anxiety and fatigue, whereas hypersomnia and insomnia showed no connections to somatic symptoms. Among somatic symptoms, excessive perspiration and pressure/tight feeling in chest were associated with the MDD/GAD domain, while muscle pain and tingling in fingers showed only a few weak associations.
Individual symptoms show differential associations in the co-occurrence of MDD/GAD with somatic symptomatology. Strongly interconnected symptoms are important in furthering our understanding of the interaction between the symptom domains, and may be valuable targets for future research and treatment.
Mental health disorders commonly co-occur, even between conceptually distinct syndromes, such as internalizing and externalizing disorders. The current study investigated whether phenotypic, genetic, and environmental variance in negative emotionality and behavioral control account for the covariation between major depressive disorder (MDD) and alcohol use disorder (AUD).
A total of 3623 members of a national twin registry were administered structured diagnostic telephone interviews that included assessments of lifetime histories of MDD and AUD, and were mailed self-report personality questionnaires that assessed stress reactivity (SR) and behavioral control (CON). A series of biometric models were fitted to partition the proportion of covariance between MDD and AUD into SR and CON.
A statistically significant proportion of the correlation between MDD and AUD was due to variance specific to SR (men = 0.31, women = 0.27) and CON (men = 0.20, women = 0.19). Further, genetic factors explained a large proportion of this correlation (0.63), with unique environmental factors explaining the rest. SR explained a significant proportion of the genetic (0.33) and environmental (0.23) overlap between MDD and AUD. In contrast, variance specific to CON accounted for genetic overlap (0.32), but not environmental overlap (0.004). In total, SR and CON accounted for approximately 70% of the genetic and 20% of the environmental covariation between MDD and AUD.
This is the first study to demonstrate that negative emotionality and behavioral control confer risk for the co-occurrence of MDD and AUD via genetic factors. These findings are consistent with the aims of NIMH's RDoC proposal to elucidate how transdiagnostic risk factors drive psychopathology.
The number of beds in care homes (with and without nurses) in the United Kingdom is three times greater than the number of beds in National Health Service (NHS) hospitals. Care homes are predominantly owned by a range of commercial, not-for-profit or charitable providers and their residents have high levels of disability, frailty and co-morbidity. NHS support for care home residents is very variable, and it is unclear what models of clinical support work and are cost-effective.
To critically evaluate how the NHS works with care homes.
A review of surveys of NHS services provided to care homes that had been completed since 2008. It included published national surveys, local surveys commissioned by Primary Care organisations, studies from charities and academic centres, grey literature identified across the nine government regions, and information from care home, primary care and other research networks. Data extraction captured forms of NHS service provision for care homes in England in terms of frequency, location, focus and purpose.
Five surveys focused primarily on general practitioner services, and 10 on specialist services to care home. Working relationships between the NHS and care homes lack structure and purpose and have generally evolved locally. There are wide variations in provision of both generalist and specialist healthcare services to care homes. Larger care home chains may take a systematic approach to both organising access to NHS generalist and specialist services, and to supplementing gaps with in-house provision. Access to dental care for care home residents appears to be particularly deficient.
Historical differences in innovation and provision of NHS services, the complexities of collaborating across different sectors (private and public, health and social care, general and mental health), and variable levels of organisation of care homes, all lead to persistent and embedded inequity in the distribution of NHS resources to this population. Clinical commissioners seeking to improve the quality of care of care home residents need to consider how best to provide fair access to health care for older people living in a care home, and to establish a specification for service delivery to this vulnerable population.
Psychiatric co-morbidity is extensive in both psychiatric settings and the general population. Such co-morbidity challenges whether DSM-based mental disorders serve to effectively carve nature at its joints. In response, a substantial literature has emerged showing that a small number of broad dimensions – internalizing, externalizing and psychoticism – can account for much of the observed covariation among common mental disorders. However, the location of personality disorders within this emerging metastructure has only recently been studied, and no studies have yet examined where pathological personality traits fit within such a broad metastructural framework.
We conducted joint structural analyses of common mental disorders, personality disorders and pathological personality traits in a sample of 628 current or recent psychiatric out-patients.
Bridging across the psychopathology and personality trait literatures, the results provide evidence for a robust five-factor metastructure of psychopathology, including broad domains of symptoms and features related to internalizing, disinhibition, psychoticism, antagonism and detachment.
These results reveal evidence for a psychopathology metastructure that (a) parsimoniously accounts for much of the observed covariation among common mental disorders, personality disorders and related personality traits, and (b) provides an empirical basis for the organization and classification of mental disorder.
Rates of psychiatric disorders are highly prevalent among prison inmates, and recent evidence confirms over-representation of youths and adults with attention deficit hyperactivity disorder (ADHD). The risk for psychiatric co-morbidity may be greater among offenders with ADHD. We undertook a systematic review and meta-analysis of reported rates of co-existing psychiatric morbidity with ADHD in prison samples.
Studies published from 1980 to 2015 were identified using five bibliographic indexes, review articles and reference lists. Included studies had a defined ADHD group and provided additional prevalence on at least one of the following: conduct disorder, substance use disorder, mood disorder, anxiety disorder, or personality disorder. We performed meta-analytical estimates of the prevalence of each co-morbid disorder within ADHD, and estimated the risk for co-existing disorders among prisoners with ADHD by pooling odds ratios (OR) with 95% confidence intervals.
Eighteen studies with data for 1615 with ADHD and 3128 without ADHD were included. The risk (OR) of all psychiatric morbidity is increased among adult inmates with ADHD. Associations in youths with ADHD were restricted to mood disorder (OR 1.89, 95% confidence interval 1.09–3.28).
This study quantifies the extent of co-morbidity presented by offenders with ADHD, especially adults. The differences between risk estimates for youths and adults indicate an incremental effect in both frequency and severity for the development of further co-morbid pathology through adulthood. The findings have implications for clinical intervention and for criminal justice policy.
The first aim was to use confirmatory factor analysis (CFA) to test a hypothesis that two factors (internalizing and externalizing) account for lifetime co-morbid DSM-IV diagnoses among adults with bipolar I (BPI) disorder. The second aim was to use confirmatory latent class analysis (CLCA) to test the hypothesis that four clinical subtypes are detectible: pure BPI; BPI plus internalizing disorders only; BPI plus externalizing disorders only; and BPI plus internalizing and externalizing disorders.
A cohort of 699 multiplex BPI families was studied, ascertained and assessed (1998–2003) by the National Institute of Mental Health Genetics Initiative Bipolar Consortium: 1156 with BPI disorder (504 adult probands; 594 first-degree relatives; and 58 more distant relatives) and 563 first-degree relatives without BPI. Best-estimate consensus DSM-IV diagnoses were based on structured interviews, family history and medical records. MPLUS software was used for CFA and CLCA.
The two-factor CFA model fit the data very well, and could not be improved by adding or removing paths. The four-class CLCA model fit better than exploratory LCA models or post-hoc-modified CLCA models. The two factors and four classes were associated with distinctive clinical course and severity variables, adjusted for proband gender. Co-morbidity, especially more than one internalizing and/or externalizing disorder, was associated with a more severe and complicated course of illness. The four classes demonstrated significant familial aggregation, adjusted for gender and age of relatives.
The BPI two-factor and four-cluster hypotheses demonstrated substantial confirmatory support. These models may be useful for subtyping BPI disorders, predicting course of illness and refining the phenotype in genetic studies.
Malnutrition is common in patients admitted to hospital due to acute illness and contributes to negative patient outcomes. In Slovakia there is a lack of relevant data on malnutrition in hospitalized patients, particularly based on chronic co-morbidity and survival. The aim of the present study was to explore the prevalence of malnutrition in hospitalized chronic patients, its relationship to co-morbidity and its impact on 10-year survival.
Retrospective cohort study.
Nutritional status was estimated by Subjective Global Assessment (SGA), BMI and serum albumin level. Survival was assessed from the National Insurance Registry over a 10-year period. The association between nutritional status measured by SGA and 10-year survival controlling for age, gender, BMI and serum albumin was analysed using Cox regression.
Data were taken from the medical records of 202 consecutively admitted chronic patients.
Median age was 63·5 years; 55·4 % were males; median BMI was 25·9 kg/m2; median serum albumin level was 39·0 g/l. Based on SGA evaluation, 38·1 % did not have sufficient nutritional status (SGA classification B and C). Malnutrition was more common in patients who were older (P=0·023), with lower BMI (P<0·001), who had gastrointestinal (P=0·049) and oncologic co-morbidity (P=0·021) and lower albumin level (P=0·049). In-hospital mortality was 3 %, but during the following 10 years 52 % died. Cox regression analysis controlling for age, gender, BMI and serum albumin showed that SGA was an independent predictor of death (hazard ratio=1·55; 95 % CI 1·04, 2·32; P=0·031).
SGA is a simple screening tool that can be routinely used in hospitalized Slovak medical patients to predict the risk of death. Improving patient nutrition could thus reduce mortality.
Course and predictors of persistence of attention deficit hyperactivity disorder (ADHD) in adults are still largely unknown. Neurobiological and clinical differences between child and adult ADHD raise the need for follow-up studies of patients diagnosed during adulthood. This study investigates predictors of ADHD persistence and the possibility of full remission 7 years after baseline assessment.
A 7-year follow-up study of adults with ADHD (n = 344, mean age 34.1 years, 49.9% males) was conducted. Variables from different domains (social demographics, co-morbidities, temperament, medication status, ADHD measures) were explored with the aim of finding potential predictors of ADHD persistence.
Retention rate was 66% (n = 227). Approximately a third of the sample (n = 70, 30.2%) did not maintain ADHD criteria and 28 (12.4%) presented full remission (<4 symptoms), independently of changes in co-morbidity or cognitive demand profiles. Baseline predictors of diagnostic persistence were higher number of inattention symptoms [odds ratio (OR) 8.05, 95% confidence interval (CI) 2.54–25.45, p < 0.001], number of hyperactivity/impulsivity symptoms (OR 1.18, 95% CI 1.04–1.34, p = 0.01), oppositional defiant disorder (OR 3.12, 95% CI 1.20–8.11, p = 0.02), and social phobia (OR 3.59, 95% CI 1.12–11.47, p = 0.03).
Despite the stage of brain maturation in adults suggests stability, approximately one third of the sample did not keep full DSM-IV diagnosis at follow-up, regardless if at early, middle or older adulthood. Although full remission is less common than in childhood, it should be considered as a possible outcome among adults.
Severe depression may be a risk factor for diagnostic conversion into bipolar disorder (BD), and psychotic depression (PD) has been consistently associated with BD. The aims of the present study were to investigate the stability of the diagnosis of severe depression and the differences between PD and non-psychotic severe depression (non-PD), as well as to assess the effectiveness of electroconvulsive therapy (ECT).
Patients who were hospitalised for severe depression (diagnosed according to ICD-10) both with and without psychotic symptoms (n=89; mean age=55.6 years, SD=13.9) from 2001 to 2010 were retrospectively assessed.
By the 75th month of follow-up assessments, 11(12.4%) patients had developed BD. Among these 11 converters, nine had developed BD within 1 year after admission. Only sub-threshold hypomanic symptoms were significantly related to developing BD. The number of depressive episodes and history of physical diseases were significantly increased in non-PD compared with PD patients, whereas ECT was significantly increased in PD compared with non-PD patients. There was a significant association between length of stay at the hospital and the number of days between admission and ECT.
Sub-threshold hypomanic symptoms may represent a prodrome of BD or an indicator of an already manifest phenotype, especially in older patients, which suggests cautious use of antidepressants. In severe depression, non-PD may often occur secondary to physical diseases and patients may experience increased recurrences compared with PD patients, which may be a more ‘primary’ disorder and often requires ECT treatments. ECT is effective for severe depression regardless of the presence of any psychotic feature; the earlier ECT is introduced, the better the expected treatment outcome.
Co-morbidity among use of different substances can be explained by a shared underlying dimensional factor. What remains unknown is whether the relationship between substance use and various co-morbid mental disorders can be explained solely by the general factor or whether there remain unique contributions of specific substances.
Data were from the 2007 Australian National Survey of Mental Health and Wellbeing (NSMHWB). A unidimensional latent factor was constructed that represented general substance use. The shared and specific relationships between lifetime substance use indicators and internalizing disorders, suicidality and psychotic-like experiences (PLEs) were examined using Multiple Indicators Multiple Causes (MIMIC) models in the total sample. Additional analyses then examined the shared and specific relationships associated with substance dependence diagnoses as indicators of the latent trait focusing on a subsample of substance users.
General levels of latent substance use were significantly and positively related to internalizing disorders, suicidality and psychotic-like experiences. Similar results were found when examining general levels of latent substance dependence in a sample of substance users. There were several direct effects between specific substance use/dependence indicators and the mental health correlates that significantly improved the overall model fit but they were small in magnitude and had relatively little impact on the general relationship.
The majority of pairwise co-morbid relationships between substance use/dependence and mental health correlates can be explained through a general latent factor. Researchers should focus on investigating the commonalities across all substance use and dependence indicators when studying mental health co-morbidity.
Although variation in the long-term course of major depressive disorder (MDD) is not strongly predicted by existing symptom subtype distinctions, recent research suggests that prediction can be improved by using machine learning methods. However, it is not known whether these distinctions can be refined by added information about co-morbid conditions. The current report presents results on this question.
Data came from 8261 respondents with lifetime DSM-IV MDD in the World Health Organization (WHO) World Mental Health (WMH) Surveys. Outcomes included four retrospectively reported measures of persistence/severity of course (years in episode; years in chronic episodes; hospitalization for MDD; disability due to MDD). Machine learning methods (regression tree analysis; lasso, ridge and elastic net penalized regression) followed by k-means cluster analysis were used to augment previously detected subtypes with information about prior co-morbidity to predict these outcomes.
Predicted values were strongly correlated across outcomes. Cluster analysis of predicted values found three clusters with consistently high, intermediate or low values. The high-risk cluster (32.4% of cases) accounted for 56.6–72.9% of high persistence, high chronicity, hospitalization and disability. This high-risk cluster had both higher sensitivity and likelihood ratio positive (LR+; relative proportions of cases in the high-risk cluster versus other clusters having the adverse outcomes) than in a parallel analysis that excluded measures of co-morbidity as predictors.
Although the results using the retrospective data reported here suggest that useful MDD subtyping distinctions can be made with machine learning and clustering across multiple indicators of illness persistence/severity, replication with prospective data is needed to confirm this preliminary conclusion.