Fanyin He

Optimism, response to treatment of depression, and rehospitalization after coronary artery bypass graft surgery.

Objective Optimism has been associated with a lower risk of rehospitalization after coronary artery bypass graft (CABG) surgery, but little is known about how optimism affects treatment of depression in post-CABG patients. Methods Using data from a collaborative care intervention trial for post-CABG depression, we conducted exploratory post hoc analyses of 284 depressed post-CABG patients (2-week posthospitalization score in the 9-item Patient Health Questionnaire ≥10) and 146 controls without depression who completed the Life Orientation Test – Revised (full scale and subscale) to assess dispositional optimism. We classified patients as optimists and pessimists based on the sample-specific Life Orientation Test – Revised distributions in each cohort (full sample, depressed, nondepressed). For 8 months, we assessed health-related quality of life (using the 36-item Short-Form Health Survey) and mood symptoms (using the Hamilton Rating Scale for Depression [HRS-D]) and adjudicated all-cause rehospitalization. We defined treatment response as a 50% or higher decline in HRS-D score from baseline. Results Compared with pessimists, optimists had lower baseline mean HRS-D scores (8 versus 15, p = .001). Among depressed patients, optimists were more likely to respond to treatment at 8 months (58% versus 27%, odds ratio = 3.02, 95% confidence interval = 1.28–7.13, p = .01), a finding that was not sustained in the intervention group. The optimism subscale, but not the pessimism subscale, predicted treatment response. By 8 months, optimists were less likely to be rehospitalized (odds ratio = 0.54, 95% confidence interval = 0.32–0.93, p = .03). Conclusions Among depressed post-CABG patients, optimists responded to depression treatment at higher rates. Independent of depression, optimists were less likely to be rehospitalized by 8 months after CABG. Further research should explore the impact of optimism on these and other important long-term post-CABG outcomes.

The impact of pain and depression on recovery after coronary artery bypass grafting.

Objective: To describe the relationship between pain and depression on recovery after coronary artery bypass grafting (CABG). Methods: A secondary data analysis on 453 depressed and nondepressed post-CABG patients enrolled in a randomized, controlled, effectiveness trial of telephone-delivered collaborative care for depression. Outcome measures were collected from March 2004 to September 2007 and included pain, physical function, and mood symptoms. Results: Depressed patients (baseline Patient Health Questionnaire-9 score ≥10) versus those without depression reported significantly worse pain scores on the 36-Item Short Form Health Survey Bodily Pain Scale at baseline and up to 12 months post-CABG, p < .05. Among patients with depression, those who received collaborative care reported significantly better pain scores at each time point between 2 and 12 months post-CABG versus depressed patients randomized to the usual care control group, p < .05. Regardless of intervention status, depressed participants with at least moderate pain at baseline reported significantly lower functional status (measured by the Duke Activity Status Index) at 8 and 12 months versus depressed patients with none or mild pain, p < .05. Depressed patients with at least moderate pain at baseline were also significantly less likely to show improvement of depressive symptoms throughout the course of follow-up versus depressed patients with little or no pain, p < .05. These findings controlled for age, gender, education, race, comorbid conditions, and baseline pain diagnosis. Conclusions: Depression and pain seem to influence functional recovery post-CABG. The relationship between these two conditions and 12-month outcomes should be considered by clinicians when planning treatment. HTN = hypertension; CVA = cerebral vascular accident; COPD = chronic obstructive pulmonary disease; CHF = chronic heart failure; PHQ = Patient Health Questionnaire; HRS-D = Hamilton Rating Scale-Depression; DASI = Duke Activity Status Index; NSAIDs = nonsteroidal antiinflammatory drug.

A randomised controlled trial of adjunctive yoga and adjunctive physical exercise training for cognitive dysfunction in schizophrenia.

Background Yoga and physical exercise have been used as adjunctive intervention for cognitive dysfunction in schizophrenia (SZ), but controlled comparisons are lacking. Aims A single-blind randomised controlled trial was designed to evaluate whether yoga training or physical exercise training enhance cognitive functions in SZ, based on a prior pilot study. Methods Consenting, clinically stable, adult outpatients with SZ (n=286) completed baseline assessments and were randomised to treatment as usual (TAU), supervised yoga training with TAU (YT) or supervised physical exercise training with TAU (PE). Based on the pilot study, the primary outcome measure was speed index for the cognitive domain of ‘attention’ in the Penn computerised neurocognitive battery. Using mixed models and contrasts, cognitive functions at baseline, 21 days (end of training), 3 and 6 months post-training were evaluated with intention-to-treat paradigm. Results Speed index of attention domain in the YT group showed greater improvement than PE at 6 months follow-up (p<0.036, effect size 0.51). In the PE group, ‘accuracy index of attention domain showed greater improvement than TAU alone at 6-month follow-up (p<0.025, effect size 0.61). For several other cognitive domains, significant improvements were observed with YT or PE compared with TAU alone (p<0.05, effect sizes 0.30–1.97). Conclusions Both YT and PE improved attention and additional cognitive domains well past the training period, supporting our prior reported beneficial effect of YT on speed index of attention domain. As adjuncts, YT or PE can benefit individuals with SZ.

Non-parametric MANOVA approaches for non-normal multivariate outcomes with missing values

ABSTRACT Between-group comparisons often entail many correlated response variables. The multivariate linear model, with its assumption of multivariate normality, is the accepted standard tool for these tests. When this assumption is violated, the non-parametric multivariate Kruskal–Wallis (MKW) test is frequently used. However, this test requires complete cases with no missing values in response variables. Deletion of cases with missing values likely leads to inefficient statistical inference. Here we extend the MKW test to retain information from partially observed cases. Results of simulated studies and analysis of real data show that the proposed method provides adequate coverage and superior power to complete case analyses.

Benefits of collaborative care for post-CABG depression are not related to adjustments in antidepressant pharmacotherapy.

OBJECTIVE To determine whether the use and adjustment of antidepressant pharmacotherapy accounted for the beneficial effects of collaborative care treatment on the improvement of mood symptoms and health-related quality of life (HRQoL) after coronary artery bypass graft (CABG) surgery. METHODS In a post-hoc analysis of data from the Bypassing the Blues (BtB) trial we tested the impact of antidepressant medication on changes in depression and HRQoL from the early postoperative period to 8-month follow-up. Two hundred fifty-nine depressed post-CABG patients scoring ≥10 on the Patient Health Questionnaire-9 were classified in four groups according to whether or not they received antidepressants at baseline and 8-months following randomization. RESULTS Patients using antidepressant pharmacotherapy at baseline and follow-up tended to be younger and female (p≤0.01), but were similar in various clinical characteristics. Just 24% (63/259) of patients were on an antidepressant at baseline which increased to 36% at follow-up (94/259). Compared to other groups, patients on antidepressants at both baseline and follow-up assessment showed the smallest improvement in mood symptoms and HRQoL. While multivariate analyses confirmed that randomization to collaborative care was associated with greater improvement in mood symptoms (odds ratio [OR]=3.1; 95%-confidence interval [CI]=1.8-5.4, p<0.0001) and mental HRQoL (OR=3.6, CI=1.4-9.3, p=0.01), use of antidepressant medication had no differential impact on either measure (p=0.06 and p=0.92, respectively). CONCLUSION The beneficial effects of collaborative care for post-CABG depression were not generated by adjustments in antidepressant medication. TRIAL REGISTRATION Clinicaltrials.gov Identifier: NCT00091962. (http://clinicaltrials.gov/ct2/show/NCT00091962?term=rollman+cabg&rank=1).

Joint analysis of cognitive and circadian variation in Schizophrenia and Bipolar I Disorder

BACKGROUND Impairment in cognitive variables and alterations in circadian function have been documented among patients with schizophrenia (SZ) and bipolar I disorder (BP1), but it is not known whether joint analysis of these variables can define clinically relevant sub-groups in either disorder. OBJECTIVES To evaluate the pattern and relationship of cognitive and circadian function in SZ and BP1 patients with respect to diagnosis and indices of clinical severity. METHODS Among patients with SZ and BP1, cognitive function was evaluated using the Penn Computerized Neurocognitive Battery and circadian function was assessed using the Composite Scale of Morningness/ Eveningness (CSM). Clinical severity was estimated using the Global Assessment of Function (GAF) scale, and age at onset of illness (AAO). The patients were compared with community based non-psychotic control individuals and non-psychotic first degree relatives of the SZ patients. The cluster distributions of cognitive function, circadian function and clinical severity were investigated and identified clusters compared across diagnostic groups. RESULTS Across participants, the cognitive domains could be separated into two clusters. Cluster 1 included the majority of control individuals and non-psychotic relatives, while SZ patients predominated in Cluster 2. BP1 patients were distributed across both clusters. The clusters could be differentiated by GAF scores, but not AAO. CSM scores were not significantly correlated with individual cognitive domains or with the clusters. CONCLUSIONS Clusters based on levels of cognitive function can discriminate SZ patients from control individuals, but not BP1 patients. CSM scores do not contribute to such discrimination.