Feelings of worthlessness during a single complicated major depressive episode predict postremission suicide attempt.

OBJECTIVE: To establish which symptoms of major depressive episode (MDE) predict postremission suicide attempts in complicated single-episode cases. METHOD: Using the nationally representative two-wave National Epidemiologic Survey on Alcohol and Related Conditions data set, we identified wave 1 lifetime single-episode MDE cases in which the episode remitted by the beginning of the wave 2 three-year follow-up period (N = 2791). The analytic sample was further limited to 'complicated' cases (N = 1872) known to have elevated suicide attempt rates, defined as having two or more of the following: suicidal ideation, marked role impairment, feeling worthless, psychomotor retardation, and prolonged (>6 months) duration. RESULTS: Logistic regression analyses showed that, after controlling for wave 1 suicide attempt which significantly predicted postremission suicide attempt (OR = 10.0), the additional complicated symptom 'feelings of worthlessness' during the wave 1 index episode significantly and very substantially predicted postremission suicide attempt (OR = 6.96). Neither wave 1 psychomotor retardation nor wave 1 suicidal ideation nor any of the other wave 1 depressive symptoms were significant predictors of wave 2 suicide attempt. CONCLUSION: Among depressive symptoms during an MDE, feelings of worthlessness is the only significant indicator of elevated risk of suicide attempt after the episode has remitted, beyond previous suicide attempts.

DSM-5, psychiatric epidemiology and the false positives problem.

The revision effort leading to the publication of the fifth edition of the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders (DSM-5) was flawed in process, goals and outcome. The revision process suffered from lack of an adequate public record of the rationale for changes, thus shortchanging future scholarship. The goals, such as dimensionalising diagnosis, incorporating biomarkers and separating impairment from diagnosis, were ill-considered and mostly abandoned. However, DSM-5's greatest problem, and the target of the most vigorous and sustained criticism, was its failure to take seriously the false positives problem. By expanding diagnosis beyond plausible boundaries in ways inconsistent with DSM-5's own definition of disorder, DSM-5 threatened the validity of psychiatric research, including especially psychiatric epidemiology. I present four examples: increasing the symptom options while decreasing the diagnostic threshold for substance use disorder, elimination of the bereavement exclusion from major depression, allowing verbal arguments as evidence of intermittent explosive disorder and expanding attention-deficit/hyperactivity disorder to adults before addressing its manifest false positives problems.

HIBAG--HLA genotype imputation with attribute bagging.

Genotyping of classical human leukocyte antigen (HLA) alleles is an essential tool in the analysis of diseases and adverse drug reactions with associations mapping to the major histocompatibility complex (MHC). However, deriving high-resolution HLA types subsequent to whole-genome single-nucleotide polymorphism (SNP) typing or sequencing is often cost prohibitive for large samples. An alternative approach takes advantage of the extended haplotype structure within the MHC to predict HLA alleles using dense SNP genotypes, such as those available from genome-wide SNP panels. Current methods for HLA imputation are difficult to apply or may require the user to have access to large training data sets with SNP and HLA types. We propose HIBAG, HLA Imputation using attribute BAGging, that makes predictions by averaging HLA-type posterior probabilities over an ensemble of classifiers built on bootstrap samples. We assess the performance of HIBAG using our study data (n=2668 subjects of European ancestry) as a training set and HLA data from the British 1958 birth cohort study (n≈1000 subjects) as independent validation samples. Prediction accuracies for HLA-A, B, C, DRB1 and DQB1 range from 92.2% to 98.1% using a set of SNP markers common to the Illumina 1M Duo, OmniQuad, OmniExpress, 660K and 550K platforms. HIBAG performed well compared with the other two leading methods, HLA*IMP and BEAGLE. This method is implemented in a freely available HIBAG R package that includes pre-fit classifiers for European, Asian, Hispanic and African ancestries, providing a readily available imputation approach without the need to have access to large training data sets.

Predictive validation of single-episode uncomplicated depression as a benign subtype of unipolar major depression.

OBJECTIVE: To evaluate the predictive validity of a proposed benign major depressive disorder (MDD) subtype, single-episode 'uncomplicated MDD', defined as MDD that remits within 6 months and lacks severe impairment, psychotic ideation, suicidal ideation, psychomotor retardation, and feeling worthless. METHOD: Using two-wave National Epidemiologic Survey on Alcohol and Related Conditions data, four groups differing in wave 1 lifetime MDD history (no history [n = 27 609]; single-episode uncomplicated [n = 418]; other single-episode [n = 1943]; multiple episode [n = 2473]) were evaluated for 3-year follow-up rates of major depressive episode (MDE), generalized anxiety disorder (GAD), and suicide attempt. RESULTS: Follow-up rates for no-MDD-history, single-episode uncomplicated MDD, other single-episode MDD, and multiple-episode MDD, respectively, were depressive episode 6.1%, 6.9%, 19.5%, 27.1%; GAD 2.7%, 4.3%, 7.8%, 11.2%; and suicide attempt 0.3%, 0.1%, 0.8%, 1.7%. For all validators, 3-year rates for single-episode uncomplicated cases were not significantly different from no-MDD-history rates, but significantly lower than both single- and multiple-episode other-MDD rates. Mild MDD, defined by having only five or six symptoms, did not yield similarly benign results; logistic regression showed 'uncomplicated' provides incremental validity over 'mild' in explaining validator rates. Validator differences were not explainable by treatment-rate differences. CONCLUSION: Single-episode uncomplicated MDD is a benign subtype lacking typical MDD negative sequelae. The planned DSM-5.1 revision should reinstitute an extended bereavement exclusion applied to all stressors.

Can the DSM's major depression bereavement exclusion be validly extended to other stressors? Evidence from the NCS.

OBJECTIVE: To evaluate whether the DSM's distinction between uncomplicated (normal) vs. complicated (disordered) bereavement-related depressive episodes can be validly extended to non-bereavement stressor-related depression (SRD). Previous findings supporting the uncomplicated/complicated SRD distinction's discriminant validity were criticized as tautological because of definitional biases (e.g., 'uncomplicated' requires brief duration, yet duration was a validator). We tested whether uncomplicated/complicated SRD validator differences are tautological or real. METHOD: Using National Comorbidity Survey data, we compared uncomplicated SRDs, complicated SRDs, and endogenous/psychotic MDD on levels of eight pathology validators. We identified definitional biases affecting six validators, and corrected them by deleting the biasing definitional components and recalculating validator levels. RESULTS: After correction of biases, uncomplicated SRDs had significantly lower pathology levels than both complicated SRDs and endogenous/psychotic MDD on seven of eight validators, disconfirming the tautology hypothesis. Regression analysis revealed that 'uncomplicated' cannot be equated with 'mild'. Extending the 'uncomplicated' durational threshold from 2 to 6 months yielded equal or stronger discriminant validity, suggesting the arbitrariness of the current durational criterion. CONCLUSION: Uncomplicated SRDs' lower pathology levels are because of real syndromal differences, not definitional tautologies. The uncomplicated/complicated distinction has discriminant validity when extended to non-bereavement SRDs as an indicator of normality vs. disorder.

Normal vs. disordered bereavement-related depression: are the differences real or tautological?

OBJECTIVE: To evaluate whether the DSM distinction between uncomplicated (normal) and complicated (disordered) bereavement-related depression (BRD) has discriminant validity on a range of pathology indicators. The DSM's major depression bereavement exclusion (BE) excludes BRDs from diagnosis when they are uncomplicated (defined by brief duration, non-severe impairment, and lack of certain pathosuggestive symptoms) but classifies all other ("complicated") BRDs as major depression. A previous report seemed to support the uncomplicated/complicated distinction's discriminant validity. However, those arguing for eliminating the BE from DSM-5 dismiss the findings as 'tautological,' attributing the validator differences to definitional biases (e.g. 'uncomplicated' requires 'no suicidal ideation,' yet 'lifetime suicide attempt' was a validator). This study empirically tests whether the uncomplicated/complicated differences are real or tautological. METHOD: Using National Comorbidity Survey data, confounds between definitional criteria for 'uncomplicated' and pathology validators were identified and corrected by deleting the biasing criteria and recalculating the corresponding validator's outcome. RESULTS: Six validators (interference with life, suicide attempt, melancholic depression, duration, hospitalization, and number of symptoms) were reanalyzed using unbiased definitions for 'uncomplicated.' All still yielded significantly lower pathology levels for uncomplicated BRDs, disconfirming the 'tautology' hypothesis. Regression analysis revealed that 'uncomplicated' offered incremental validity over severity alone in predicting pathology, so 'uncomplicated' cannot be equated with 'mild.' CONCLUSION: Uncomplicated BRDs' lower pathology validator levels are because of real syndromal differences, not definitional tautologies, supporting the BE's validity.

Relation between duration and severity in bereavement-related depression.

OBJECTIVE: Prolonged duration is commonly used as an indicator that bereavement-related depression (BRD) is pathological. DSM-IV replaced the traditional 1-year pathology cut-point by a 2-month cut-point. Yet, little evidence exists regarding the validity of these cut-points in indicating increased BRD severity. This study evaluated the validity of the 2-month and 1-year cut-points in differentiating less severe from more severe BRDs in a nationally representative U.S. sample. METHOD: National Comorbidity Survey respondents with BRD's (n = 152) lasting 0-8, 9-52 and >52 weeks were evaluated for depression severity using six severity indicators. Cut-point validity was established by discontinuities in severity levels between durations below and above the cut-point. RESULTS: Bereavement-related depressions of >52-week duration were significantly higher than 9- to 52-week BRDs on four of six severity indicators and on a cumulative overall severity measure of mean number of severity indicators per person, whereas ≤8-week and 9- to 52-week durational categories differed on one severity indicator and not on overall severity. Additional analyses using durations 0-12, 13-26, 27-52 and >52 weeks suggested that alternative <52-week cut-points also lack validity. CONCLUSION: The traditional 1-year cut-point validly identifies increasing BRD severity; DSM's 2-month cut-point does not. Duration does not indicate increasing BRD severity before 1 year. Research using the 2-month cut-point may yield misleading results.

The analyst's knowledge and authority: a critique of the "new view" in psychoanalysis.

A critical assessment is presented of positions recently taken by Mitchell and Renik, who are taken as representatives of a "new view" in psychoanalysis. One article by Mitchell and two by Renik are examined as paradigmatic of certain ways of construing the nature of mind, the analyst's knowledge and authority, and the analytic process that are unduly influenced by the postmodern turn in psychoanalysis. Although "new view" theorists have made valid criticisms of traditional psychoanalytic theory and practice, they wind up taking untenable positions. Specifically called into question are their views on the relation between language and interpretation, on the one hand, and the mental contents of the patient on the other. A disjunction is noted between their discussion of clinical material and their conceptual stance, and their idiosyncratic redefinitions of truth and objectivity are criticized. Finally, a "humble realism" is suggested as the most appropriate philosophical position for psychoanalysts to adopt.

The myth of DSM's invention of new categories of disorder: Houts's diagnostic discontinuity thesis disconfirmed.

Houts (2001) argues that increases in DSM diagnostic categories are due to the invention of new disorders that are discontinuous with old conceptions of disorder and would not have been previously diagnosed. He maintains that DSM category increases are not comparable in nature to ICD category increases, which are mainly refinements of recognized disorders. I survey categories of disorder introduced after DSM-II and assess whether they are discontinuous with old concepts and categories of disorder. Candidate categories are identified from: Houts and Follette (1998), Mentalism, mechanisms, and medical analogues: Reply to Wakefield. Journal of Consulting and Clinical Psychology; Kutchins and Kirk (1997) Making us crazy: DSM: The psychiatric bible and the creation of mental disorders. New York: Free Press; and my own list. The result is that virtually none of the candidate categories are invented, discontinuous categories. In almost every case, the newly labeled conditions were considered disorders at the time of DSM-II and would have been diagnosed under DSM-II categories. I also reexamine DSM-IV sleep disorder categories, which Houts claims are discontinuous with past diagnostic conceptions. The result is that all DSM-IV sleep disorders were recognized as disorders at the time of DSM-II, and most were recognized as mental disorders. I conclude that DSM category increases are comparable in nature to ICD category increases, and that the invention-of-disorder account cannot explain the vast majority of such increases.

Evolutionary history versus current causal role in the definition of disorder: reply to McNally.

The harmful dysfunction (HD) analysis (Wakefield, American Psychologist 47 (1992a) 373) asserts that "disorder" means "harmful dysfunction", where "harm" is a value concept anchored in social values and "dysfunction" is a factual concept referring to failure of a mechanism to perform a natural function. Additionally, the HD analysis claims that a mechanism's natural functions are its naturally selected effects. McNally (Behaviour Research and Therapy (2000) pp. 309-314) argues to the contrary that "dysfunction" is a value concept referring to negative failures of function, that "function" refers to current causal roles and not evolutionarily designed causal roles, and that "disorder" consequently means "harmful failure of a mechanism to perform a valued current causal role." I reply by showing that McNally's proposals lack the HD analysis's power to explain common judgments about function, dysfunction, and disorder. "Dysfunction" cannot be a negative value concept because many dysfunctions are positive or neutral; "function" cannot refer to current causal roles because many current causal roles are not functions and some functions are not current causal roles; and "disorder" cannot refer to harmful failures of current causal roles because that definition allows almost any negative condition whatever to be a disorder and thus fails to explain the distinctions we make between disorder and non-disorder.

Statistical issues in the analysis of disease mapping data.

In this paper we discuss a number of issues that are pertinent to the analysis of disease mapping data. As an illustrative example we consider the mapping of larynx cancer across electoral wards in the North West Thames region of the U.K. Bayesian hierarchical models are now frequently employed to carry out such mapping. In a typical situation, a three-stage hierarchical model is specified in which the data are modelled as a function of area-specific relative risks at stage one; the collection of relative risks across the study region are modelled at stage two; and at stage three prior distributions are assigned to parameters of the stage two distribution. Such models allow area-specific disease relative risks to be 'smoothed' towards global and/or local mean levels across the study region. However, these models contain many structural and functional assumptions at different levels of the hierarchy; we aim to discuss some of these assumptions and illustrate their sensitivity. When relative risks are the endpoint of interest, it is common practice to assume that, for each of the age-sex strata of a particular area, there is a common multiplier (the relative risk) acting upon each of the stratum-specific risks in that area; we will examine this proportionality assumption. We also consider the choices of models and priors at stages two and three of the hierarchy, the effect of outlying areas, and an assessment of the level of smoothing that is being carried out. For inference, we concentrate on the description of the spatial variability in relative risks and on the association between the relative risks of larynx cancer and an area-level measure of socio-economic status.

Point-source modelling using matched case-control data.

We describe an extension to matched case-control studies of the parametric modelling framework developed by Diggle (1990) and Diggle and Rowlingson (1994) to investigate raised risk around putative sources of environmental pollution. We use a conditional likelihood approach for the family of risk functions considered in Diggle and Rowlingson (1994). We show that the likelihood surface that results from these models may be highly irregular, and provide a Bayesian analysis in which we investigate the posterior distribution using Markov chain Monte Carlo. An analysis of one-one matched data that were collected to investigate the relationship between respiratory disease and distance to roads in East London is presented.

DSM-IV diagnostic criterion for clinical significance: does it help solve the false positives problem?

OBJECTIVE: A major change in DSM-IV is the inclusion in almost one-half of the diagnostic criteria sets of a clinical significance criterion, which requires that symptoms cause "clinically significant distress or impairment in social, occupational, or other important areas of functioning." In response to concerns that the DSM criteria are overly inclusive, the clinical significance criterion attempts to minimize false positive diagnoses in situations in which the symptom criteria do not necessarily indicate pathology. This article examines whether the clinical significance criterion achieves its purpose and considers its broader impact on diagnostic validity. METHOD: The effect of the clinical significance criterion on the diagnostic validity of DSM-IV criteria for a wide range of disorders was examined. RESULTS: For many diagnoses to which the clinical significance criterion was added, the symptom criteria are inherently associated with significant impairment, so the clinical significance criterion is redundant and therefore does not affect caseness. For some diagnoses, the clinical significance criterion is potentially helpful in eliminating false positives by elevating the level of required distress. However, there may be advantages to obtaining the same results by modifying some of the symptom criteria. Often the clinical significance criterion has led to the possibility of false negative diagnoses. CONCLUSIONS: In the process of revising DSM-IV, the generic use of the clinical significance criterion should be reconsidered. For each DSM diagnosis, it should be determined whether there is a need to raise the threshold of any of the existing symptom criteria or to add a criterion that excludes normal reactions to psychosocial stress.