Multivariate logistic regression for familial aggregation of two disorders. II. Analysis of studies of eating and mood disorders

[Affective disorders and eating disorders]

Epidemiologic studies show a frequent co-occurence of affective and eating disorders. The incidence of one disorder in patients suffering from the other disorder is well over the incidence in the general population. Several causes could explain this increased comorbidity. First, the iatrogenic origin is detailed. Indeed, psychotropic drugs, and particularly mood stabilizers, often lead to modification in eating behaviors, generally inducing weight gain. These drugs can increase desire for food, reduce baseline metabolism or decrease motor activity. Also, affective and eating disorders share several characteristics in semiology. These similarities can not only obscure the differential diagnosis but may also attest of conjoint pathophysiological bases in the two conditions. However, genetic and biological findings so far are too sparse to corroborate this last hypothesis. Nonetheless, it is noteworthy that comorbidity of affective and eating disorders worsens patients'prognosis and is associated with more severe forms of affective disorders characterized by an earlier age of onset in the disease, higher number of mood episodes and a higher suicidality. Lastly, psychotropic drugs used in affective disorders (lithium, antiepileptic mood stabilizers, atypical antipsychotics, antidepressants) are reviewed in order to weigh their efficacy in eating disorders. This could help establish the best therapeutic option when confronted to comorbidity.

Analysis of familial aggregation in total, against-the-rule, with-the-rule, and oblique astigmatism by conditional and marginal models in the Tehran eye study

Purpose:

The purpose was to determine the familial aggregation of the total, against-the-rule (ATR), with-the-rule (WTR), and oblique astigmatism by conditional and marginal models in the Tehran Eye Study.

Materials and Methods:

Total, ATR, WTR, and oblique astigmatism were studied in 3806 participants older than 5 years from August 2002 to December 2002 in the Tehran Eye Study. Astigmatism was defined as a cylinder worse than or equal to −0.5 D. WTR astigmatism was defined as 0 ± 19°, ATR astigmatism was defined as 90 ± 19°, and oblique when the axes were 20–70° and 110–160°. The familial aggregation was investigated with a conditional model (quadratic exponential) and marginal model (alternating logistic regression) after controlling for confounders.

Results:

Using the conditional model, the conditional familial aggregation odds ratios (OR) (95% confidence interval) for the total, WTR, ATRs, and oblique astigmatism were 1.49 (1.43–1.72), 1.91 (1.65–2.20), 2.00 (1.70–2.30), and 1.86 (1.37–2.54), respectively. In the marginal model, the marginal OR of the parent-offspring and sib-sib in the total astigmatism were 1.35 (1.13–1.63) and 1.54 (1.13–2.11), respectively; WTR 1.53 (1.06–2.20) and 1.94 (1.21–3.13) and; ATR 2.13 (1.01–4.50) and 2.23 (1.52-3.30). The model was statistically significant in sib-sib relationship only for oblique astigmatism with OR of 3.00 (1.25–7.20).

Conclusion:

The results indicate familial aggregation of astigmatism in the population in Tehran adjusted for age, gender, cataract, duration of education, and body mass index, so that the addition of a new family member affected with astigmatism, as well as having a sibling or parents with astigmatism, significantly increases the odds of exposure to the disease for all four phenotypes. This aggregation can be due to genetic and/or environmental factors. Dividing astigmatism into three phenotypes increased the odds ratios.

Analysis of familial aggregation studies with complex ascertainment schemes

Familial aggregation studies are a common first step in the identification of genetic determinants of disease. If aggregation is found, more refined genetic studies may be undertaken. Complex ascertainment schemes are frequently employed to ensure that the sample contains a sufficient number of families with multiple affected members, as required to detect aggregation. For example, an eligibility criterion for a family might be that both the mother and daughter have disease. Adjustments must be made for ascertainment to avoid bias. We propose adjusting for complex ascertainment schemes through a joint model for the outcomes of disease and ascertainment. This approach improves upon previous simplifying assumptions regarding the ascertainment process.

A structural approach to the familial coaggregation of disorders

The demonstration that 2 disorders coaggregate in families is often the first step in the exploration of genetic factors common to the 2 disorders. Previous methods of analyzing familial coaggregation have used either (1) a typical measure of association (eg, the odds ratio) between a disorder in an individual and another disorder in family members, or (2) a linear structural equation model (SEM). The association method accommodates case-control sampling of families, but may not assess the causal effect of interest because it is not based on an underlying causal model. The SEM method is based on a causal model, but cannot easily accommodate case-control sampling or direct effects of 1 disorder on the other within individuals. We develop a new method of analyzing coaggregation based on directed acyclic graphs. Because this method is a generalization of structural equation models and uses measures of association that accommodate case-control sampling and direct effects, it combines the strengths of both previous methods. In the absence of direct effects between disorders, our approach provides a valid estimate of the causal coaggregation effect. In the presence of direct effects, our approach provides an upper-bound estimate and (assuming additive linear effects of latent familial and nonfamilial factors) a lower-bound estimate of the causal coaggregation effect. For illustration, we applied our method to a family study of binge eating disorder and bipolar disorder.

A family history study of binge-eating disorder

Family studies of anorexia nervosa and bulimia nervosa have yielded important information about the etiologies of these eating disorders. By contrast, little is known about familial factors of etiologic importance for binge-eating disorder (BED). The purpose of the current family history study was to assess the prevalence of comorbid psychopathology in a non-treatment seeking female sample of 31 probands with BED, 32 control probands without BED, and their 283 first-degree relatives. In-person semistructured clinical interviews were conducted with the probands, who also served as informants for all of their first-degree relatives. Significantly higher lifetime rates of major depressive disorder, dysthymic disorder, and social phobia were found among women with BED compared with control women. Significantly higher lifetime rates of bipolar (I or II) disorder, any depressive disorder, nearly all anxiety disorders, anorexia nervosa, and BED were reported among the first-degree relatives of women with BED compared with the first-degree relatives of control women. Furthermore, female relatives of women with BED were reported to have higher rates of substance use disorders and dysthymic disorder compared with female relatives of control women without BED. Nearly all disorders that were elevated in relatives of women with BED followed a pattern of independent transmission from BED. The primary exception was substance use disorder among female relatives, whose transmission pattern was consistent with that of a shared etiology with BED. Thus, BED and substance use disorder may share a common mechanism of familial transmission among women.

Directed acyclic graphs, sufficient causes, and the properties of conditioning on a common effect

In this paper, the authors incorporate sufficient-component causes into the directed acyclic graph (DAG) causal framework in order to make apparent several properties of conditioning on a common effect. By incorporating sufficient causes on a graph, it is possible to detect conditional independencies within strata of the conditioning variable which are not evident on DAGs without the representation of sufficient causes. It is also possible to determine the sign of the conditional covariance of two causes when conditioning on their common effect if some knowledge of the sufficient-cause mechanisms for the common effect is available. The incorporation of sufficient causes within the DAG framework also allows for the representation of interactions on DAGs and for the unification of several different causal frameworks. For illustration, the results are applied to an example concerning the familial coaggregation of two disorders.

Speech and language skills of parents of children with speech sound disorders

PURPOSE

This study compared parents with histories of speech sound disorders (SSD) to parents without known histories on measures of speech sound production, phonological processing, language, reading, and spelling. Familial aggregation for speech and language disorders was also examined.

METHOD

The participants were 147 parents of children with SSD (58 fathers and 89 mothers) who were directly tested and interviewed for family history of disorders.

RESULTS

Thirty-six parents (18 mothers and 18 fathers) reported enrollment in speech therapy as children for SSD. Parents with a history of speech therapy obtained lower scores on the Multisyllabic Word Repetition, Nonword Repetition, and Tongue Twister tasks than parents without such histories. These parents also had poorer reading, spelling, and receptive language skills. Parents with histories of SSD and additional language impairments (LI) performed worse than parents with isolated SSD on all measures except Pig Latin and oral motor skills. Familial aggregation for SSD and LI was supported. In addition, the likelihood of SSD in a family member increased by a factor of 1.9 over rates of SSD found in individuals without additional family members with SSD, and the odds of LI increased by a factor of 4.1 over rates of LI found in individuals without additional family members with LI for each additional family member with SSD or LI, respectively.

CONCLUSIONS

The results documented both residual effects in adulthood of childhood SSD and familial aggregation for SSD. These residual difficulties do not appear to affect educational and occupational outcomes.

Further evidence of pleiotropy influencing speech and language: analysis of the DYX8 region

BACKGROUND/AIMS

Genetic studies have raised the possibility of common bases for cognitive linguistic disorders such as speech sound disorder (SSD), reading disorder (RD) and language impairment (LI). Thus, some of the same genes may jointly influence cognitive components within and between these three disorders. We examined the plausibility of this theory in a sample of families ascertained on the basis of a child with SSD.

METHODS

Using the method of generalized estimating equations to solve a bivariate family predictive model we obtained measures of comorbidity and familial aggregation of SSD and LI. We then used two methods of multipoint model-free linkage analysis to evaluate SSD and LI psychometric test measures over a region previously implicated in linkage studies of RD, DYX8 region, 1p34-p36.

RESULTS

Bivariate phenotypic analyses show evidence of comorbidity and within family aggregation and coaggregation of SSD and LI. In addition, two regions on chromosome 1 show suggestive evidence of linkage. The first region was previously reported in dyslexia studies. Our maximum linkage signal in this region measured articulation (p = 0.0009) in SSD sibling pairs. The second region is characterized by processes involved in language production, with the maximum linkage signal measuring listening comprehension (p = 0.0019) using all sibling pairs.

CONCLUSION

We conclude that the DYX8 region could bear genes controlling pleiotropic effects on SSD, LI and RD.

Psychological and physical correlates of disordered eating in male and female Turkish college students

The aim of this study was to examine the relationship between disordered eating and three psychological variables (obsessive and depressive symptoms, self-esteem) and body mass index (BMI) in a sample of male and female college students in Turkey. Maudsley Obsessive-Compulsive Inventory (MOCI), Beck Depression Inventory (BDI), Rosenberg Self Esteem Scale (RSES), Eating Attitude Test (EAT), and Bulimic Investigatory Test, Edinburgh (BITE) scales were administered to 408 college students (279 women; 129 men). By multiple regression analysis, the scales of MOCI, BDI, RSES and BMI were used as potential predictors of EAT and BITE scores for each gender. Obsessionality and BMI were the strongest predictors of bulimic and anorexic symptoms in women. In men, while the strongest predictors of bulimic symptoms were the depressive and obsessive symptoms, the best predictor of anorexic symptoms was obsession scores. There were significant differences and some similarities between male and female Turkish college students.

Comorbidity of bipolar and eating disorders: distinct or related disorders with shared dysregulations?

BACKGROUND

The co-occurrence of bipolar and eating disorders, though of major clinical and public health importance, remains relatively unexamined.

METHODS

In reviewing the literature on this comorbidity, we compared bulimia, anorexia nervosa, bulimia nervosa, binge eating disorders and bipolar disorders on phenomenology, course, family history, biology, and treatment response.

RESULTS

Epidemiological studies show an association between subthreshold bipolar disorder and eating disorders in adolescents, and between hypomania and eating disorders, especially binge eating behavior, in adults. Of the clinical studies, most show that patients with bipolar disorder have elevated rates of eating disorders, and vice versa. Finally, the phenomenology, course, comorbidity, family history, and pharmacologic treatment response of these disorders show considerable overlap on all of these parameters. In particular, on phenomenologic grounds--eating dysregulation, mood dysregulation, impulsivity and compulsivity, craving for activity and/or exercise--we find many parallels between bipolar and eating disorders. Overall, the similarities between these disorders were more apparent when examined in their spectrum rather than full-blown expressions.

LIMITATIONS

Despite an extensive literature on each of these disorders, studies examining their overlap across all these parameters are relatively sparse and insufficiently systematic.

CONCLUSIONS

Nonetheless, the reviewed literature leaves little doubt that bipolar and eating disorders--particularly bulimia nervosa and bipolar II disorder--are related. Although several antidepressants and mood stabilizers have shown promise for eating disorders, their clinical use when these disorders co-exist with bipolarity is still very much of an art. We trust that this review will stimulate more rigorous research in their shared putative underlying psychobiologic mechanisms which, in turn, could lead to more rational targeted treatments.

Familial aggregation and coaggregation of history of hypertension and stroke

We attempted to evaluate familial aggregation and coaggregation of history of hypertension and stroke. Past and family history of hypertension and stroke for 83 089 probands and their relatives were obtained from a data set for the Japan Collaborative Cohort Study for Evaluation of Cancer Risk sponsored by the Ministry of Education (JACC Study), which was initiated from 1988 to 1990. First, evaluation was performed for familial aggregation of each of two disorders using ordinal logistic regression of the generalized estimation equations (GEE) to account for dependence of observations within families. Secondly, in order to evaluate the familial congregation of the history of hypertension and stroke, a GEE-based multivariate probed predictive model was applied. After adjusting for the proband's age, level of obesity, smoking status, drinking status, habitation area, and the gender and type of the relatives, the estimated odds ratios for the intraindividual clustering and familial aggregation of the disease history showed statistically significant relationships. In addition, the history of the two disorders showed a significant relationship in terms of familial coaggregation independently of the aggregation of each disorder itself. Our results confirmed that hypertension and stroke coaggregate strongly within families through possible effects of genetic factors, which, alone or in conjunction with environmental factors, influence susceptibility to both hypertension and stroke.

Family study of fibromyalgia and affective spectrum disorder

BACKGROUND

Fibromyalgia is one member of a proposed group of psychiatric and medical disorders, collectively termed affective spectrum disorder (ASD), hypothesized to share possibly heritable pathophysiologic features. Two predictions of the ASD hypothesis were tested: ASD, taken as a single entity, aggregates in families; and fibromyalgia coaggregates with other forms of ASD in families.

METHODS

Probands with and without fibromyalgia, together with their first-degree relatives, were administered structured diagnostic interviews. Noninterviewed relatives were diagnosed according to information provided by interviewed relatives. Aggregation and coaggregation of disorders were analyzed with proband predictive logistic and linear regression models.

RESULTS

In 533 relatives of 78 probands with fibromyalgia and 272 relatives of 40 probands without fibromyalgia, the estimated odds ratio (OR) (95% confidence interval) for the familial aggregation of ASD was 1.8 (.97, 3.2), p = .065, and the increase in number of forms of ASD in a relative for each additional form of ASD in a proband was .076 (.027, .1240), p = .002. The OR for the coaggregation of fibromyalgia with other forms of ASD was 2.0 (1.2, 3.2), p = .004; this remained significant even after excluding all mood-disorder diagnoses: 1.8 (1.1, 3.0), p = .012.

CONCLUSIONS

These findings support familial aggregation of ASD collectively and familial coaggregation of fibromyalgia with other forms of ASD.

Power calculations for familial aggregation studies

Family studies are frequently undertaken as the first step in the search for genetic determinants of disease. Significant familial aggregation of disease is suggestive of a genetic etiology for the disease, and may lead to more focused genetic analyses. Many methods have been proposed in the literature for the analysis of family studies. One model that is appealing for its simplicity of computation and the conditional interpretation of its parameters is the quadratic exponential model (e.g., Zhao and Prentice [1990] Biometrika 77:642-648; Betensky and Whittemore [1996] Appl. Stat. 45:422-429; Hudson et al. [2001a] Am. J. Epidemiol. 153:500-514). However, a limiting factor in its application, as well as that of the other proposed methods, is that power and sample size calculations have not been derived. These calculations are essential for investigators who are designing family studies. Here we derive analytic approximations for power for testing for familial aggregation, for both randomly sampled and nonrandomly sampled families. We also present simulation studies of power for both single- and two-disease cases, both under random and nonrandom sampling.

Family study of fibromyalgia

OBJECTIVE

To assess for familial aggregation of fibromyalgia (FM) and measures of tenderness and pain, and for familial coaggregation of FM and major mood disorder (major depressive disorder or bipolar disorder).

METHODS

Probands meeting the American College of Rheumatology criteria for FM and control probands with rheumatoid arthritis (RA) and no lifetime diagnosis of FM were recruited from consecutive referrals to 2 community-based rheumatology practices. Probands were ages 40-55 years and had at least 1 first-degree relative age 18 years or older who was available for interview and examination. All probands and interviewed relatives underwent a dolorimeter tender point examination and a structured clinical interview. Interviewed relatives were asked about first-degree relatives who were not available for interview, using a structured family interview. Logistic and linear regression models, adjusting for the correlation of observation within families, were applied to study the aggregation and coaggregation effects.

RESULTS

Information was collected for 533 relatives of 78 probands with FM and 272 relatives of 40 probands with RA. FM aggregated strongly in families: the odds ratio (OR) measuring the odds of FM in a relative of a proband with FM versus the odds of FM in a relative of a proband with RA was 8.5 (95% confidence interval [95% CI] 2.8-26, P = 0.0002). The number of tender points was significantly higher, and the total myalgic score was significantly lower in the relatives of probands with FM compared with the relatives of probands with RA. FM coaggregated significantly with major mood disorder: the OR measuring the odds of major mood disorder in a relative of a proband with FM versus the odds of major mood disorder in a relative of a proband with RA was 1.8 (95% CI 1.1-2.9, P = 0.013).

CONCLUSION

FM and reduced pressure pain thresholds aggregate in families, and FM coaggregates with major mood disorder in families. These findings have important clinical and theoretical implications, including the possibility that genetic factors are involved in the etiology of FM and in pain sensitivity. In addition, mood disorders and FM may share some of these inherited factors.

Pharmacologic treatment of binge eating disorder

OBJECTIVE

To review the findings from pharmacologic trials of binge eating disorder (BED) and to provide guidelines for pharmacologic treatment.

METHODS

The literature was searched for studies of pharmacologic treatment of BED and related conditions, such as nonpurging bulimia nervosa.

RESULTS

Placebo-controlled studies of desipramine, fluvoxamine, fluoxetine, sertraline, citalopram, dexfenfluramine, sibutramine, and topiramate have demonstrated the efficacy of these agents in the treatment of BED. An open trial of venlafaxine has offered preliminary evidence for the efficacy of this medication. Guidelines for pharmacologic management of BED are provided.

CONCLUSIONS

The literature offers support for the use of agents from three categories of medication (antidepressants, appetite suppressants, and anticonvulsants) in the treatment of BED.

Regression methods for assessing familial aggregation of disease

This paper reviews methods for assessing familial aggregation of disease based on simple logistic regression models. Studies are based on a case-control sampling design, where the disease status of the first degree relatives of both cases and controls are obtained. Both 'proband predictive' and 'family predictive' models are discussed, and an example is given using a case-control sample from a lung cancer study in non-smokers. The methods are extended to characterize co-aggregation of two disorders, that is, presence of one disorder in the proband increases the risk of a second disorder in the relative. An example involving eating disorders and depression is given.

Multivariate logistic regression for familial aggregation of two disorders. I. Development of models and methods

The question of whether two disorders cluster together, or coaggregate, within families often arises. This paper considers how to analyze familial aggregation of two disorders and presents two multivariate logistic regression methods that model both disorder outcomes simultaneously. The first, a proband predictive model, predicts a relative's outcomes (the presence or absence of each of the two disorders) by using the proband's disorder status. The second, a family predictive model derived from the quadratic exponential model, predicts a family member's outcomes by using all of the remaining family members' disorder statuses. The models are more realistic, flexible, and powerful than univariate models. Methods for estimation and testing account for the correlation of outcomes among family members and can be implemented by using commercial software.