5 results
Development of a model to predict antidepressant treatment response for depression among Veterans
- Victor Puac-Polanco, Hannah N. Ziobrowski, Eric L. Ross, Howard Liu, Brett Turner, Ruifeng Cui, Lucinda B. Leung, Robert M. Bossarte, Corey Bryant, Jutta Joormann, Andrew A. Nierenberg, David W. Oslin, Wilfred R. Pigeon, Edward P. Post, Nur Hani Zainal, Alan M. Zaslavsky, Jose R. Zubizarreta, Alex Luedtke, Chris J. Kennedy, Andrea Cipriani, Toshiaki A. Furukawa, Ronald C. Kessler
-
- Journal:
- Psychological Medicine / Volume 53 / Issue 11 / August 2023
- Published online by Cambridge University Press:
- 15 July 2022, pp. 5001-5011
-
- Article
- Export citation
-
Background
Only a limited number of patients with major depressive disorder (MDD) respond to a first course of antidepressant medication (ADM). We investigated the feasibility of creating a baseline model to determine which of these would be among patients beginning ADM treatment in the US Veterans Health Administration (VHA).
MethodsA 2018–2020 national sample of n = 660 VHA patients receiving ADM treatment for MDD completed an extensive baseline self-report assessment near the beginning of treatment and a 3-month self-report follow-up assessment. Using baseline self-report data along with administrative and geospatial data, an ensemble machine learning method was used to develop a model for 3-month treatment response defined by the Quick Inventory of Depression Symptomatology Self-Report and a modified Sheehan Disability Scale. The model was developed in a 70% training sample and tested in the remaining 30% test sample.
ResultsIn total, 35.7% of patients responded to treatment. The prediction model had an area under the ROC curve (s.e.) of 0.66 (0.04) in the test sample. A strong gradient in probability (s.e.) of treatment response was found across three subsamples of the test sample using training sample thresholds for high [45.6% (5.5)], intermediate [34.5% (7.6)], and low [11.1% (4.9)] probabilities of response. Baseline symptom severity, comorbidity, treatment characteristics (expectations, history, and aspects of current treatment), and protective/resilience factors were the most important predictors.
ConclusionsAlthough these results are promising, parallel models to predict response to alternative treatments based on data collected before initiating treatment would be needed for such models to help guide treatment selection.
Development of a model to predict psychotherapy response for depression among Veterans
- Hannah N. Ziobrowski, Ruifeng Cui, Eric L. Ross, Howard Liu, Victor Puac-Polanco, Brett Turner, Lucinda B. Leung, Robert M. Bossarte, Corey Bryant, Wilfred R. Pigeon, David W. Oslin, Edward P. Post, Alan M. Zaslavsky, Jose R. Zubizarreta, Andrew A. Nierenberg, Alex Luedtke, Chris J. Kennedy, Ronald C. Kessler
-
- Journal:
- Psychological Medicine / Volume 53 / Issue 8 / June 2023
- Published online by Cambridge University Press:
- 11 February 2022, pp. 3591-3600
-
- Article
- Export citation
-
Background
Fewer than half of patients with major depressive disorder (MDD) respond to psychotherapy. Pre-emptively informing patients of their likelihood of responding could be useful as part of a patient-centered treatment decision-support plan.
MethodsThis prospective observational study examined a national sample of 807 patients beginning psychotherapy for MDD at the Veterans Health Administration. Patients completed a self-report survey at baseline and 3-months follow-up (data collected 2018–2020). We developed a machine learning (ML) model to predict psychotherapy response at 3 months using baseline survey, administrative, and geospatial variables in a 70% training sample. Model performance was then evaluated in the 30% test sample.
Results32.0% of patients responded to treatment after 3 months. The best ML model had an AUC (SE) of 0.652 (0.038) in the test sample. Among the one-third of patients ranked by the model as most likely to respond, 50.0% in the test sample responded to psychotherapy. In comparison, among the remaining two-thirds of patients, <25% responded to psychotherapy. The model selected 43 predictors, of which nearly all were self-report variables.
ConclusionsPatients with MDD could pre-emptively be informed of their likelihood of responding to psychotherapy using a prediction tool based on self-report data. This tool could meaningfully help patients and providers in shared decision-making, although parallel information about the likelihood of responding to alternative treatments would be needed to inform decision-making across multiple treatments.
B.5 Hospitalization in school aged children with cerebral palsy and population-based Controls: A Data Linkage Study
- O Fortin, P Ng, M Dorais, L Koclas, N Pigeon, M Shevell, M Oskoui
-
- Journal:
- Canadian Journal of Neurological Sciences / Volume 48 / Issue s3 / November 2021
- Published online by Cambridge University Press:
- 05 January 2022, p. S17
-
- Article
-
- You have access Access
- Export citation
-
Background: Predictors of hospitalization and reasons for admissions can inform healthcare planning and prevention. We sought to characterize the hospitalization pattern and risk factors for admission of children with cerebral palsy (CP). Methods: Data from the Registre de paralysie cérébrale du Québec and provincial administrative databases were linked. The CP cohort contained children born between 1999 and 2002. Data related to admissions were captured in 2012. Relative risks (RR) were calculated to identify factors increasing hospitalization risk. Peers without CP were matched from administrative databases in a 20:1 ratio. Chi-square tests and Student’s T-tests were used to compare cohorts. Results: 301 children with CP and 6040 peer controls were selected. Hospitalizations were increased in children with CP (raw mean difference (RMD) 5.0 95% CI 4.7 to 5.2), with significantly longer lengths of stay (RMD 2.8 95% CI 1.8 to 3.8) and number of diagnoses per hospitalization (RMD 1.6 95% CI 1.4 to 1.8). Increased risk of any hospitalization was observed in children with a more complex profile. Conclusions: Children with a more severe profile of CP and greater health care complexity face more frequent and longer hospital stays. Coordinated interdisciplinary care is needed in school-aged children with CP and medical complexity.
P.119 Emergency Department Use in Children with Cerebral Palsy: A Data Linkage Study
- O Fortin, P Ng, M Dorais, L Koclas, N Pigeon, M Shevell, M Oskoui
-
- Journal:
- Canadian Journal of Neurological Sciences / Volume 48 / Issue s3 / November 2021
- Published online by Cambridge University Press:
- 05 January 2022, p. S53
-
- Article
-
- You have access Access
- Export citation
-
Background: Improved understanding of factors predictive of emergency department (ED) visits in children with cerebral palsy (CP) can help optimize healthcare use. We sought to identify the pattern of ED consultations in these children. Methods: Data from the Registre de paralysie cérébrale du Québec and provincial administrative databases were linked. The CP cohort was comprised of children born between 1999 and 2002. Data pertaining to ED presentations between 1999 and 2012 were obtained. Relative risks were calculated to identify factors associated with increased ED visits. Peers without CP were selected from administrative databases and matched in a 20:1 ratio. Chi-square tests and Student’s T-tests were used to compare the two cohorts. Results: 301 children with CP and 6040 peer controls were selected. Ninety-two percent (92%) of the CP cohort had at least one ED visit, compared to 74% amongst controls. Children with CP had an increased risk of high ED use compared to peers (RR 1.40 95% CI 1.30-1.52). Factors predictive of high ED use were comorbid epilepsy, severe motor impairment and low socioeconomic status. Conclusions: Children with CP have a higher need for urgent health assessments than their peers, resulting in increased use of ED services. System factors and barriers should be investigated.
2196 Pre-treatment sleep disturbance as a risk factor for radiation therapy induced pain in 676 women with breast cancer
- Anita R. Peoples, Wilfred R. Pigeon, Dongmei Li, Joseph A. Roscoe, Sheila N. Garland, Michael L. Perlis, Vincent P. Vinciguerra, Thomas Anderson, Lisa S. Evans, James L. Wade III, Deborah J. Ossip, Gary R. Morrow, Julie R. Wolf
-
- Journal:
- Journal of Clinical and Translational Science / Volume 2 / Issue S1 / June 2018
- Published online by Cambridge University Press:
- 21 November 2018, pp. 45-46
-
- Article
-
- You have access Access
- Open access
- Export citation
-
OBJECTIVES/SPECIFIC AIMS: The purpose of the present secondary data analysis was to examine the effect of moderate-severe disturbed sleep before the start of radiation therapy (RT) on subsequent RT-induced pain. METHODS/STUDY POPULATION: Analyses were performed on 676 RT-naïve breast cancer patients (mean age 58, 100% female) scheduled to receive RT from a previously completed nationwide, multicenter, phase II randomized controlled trial examining the efficacy of oral curcumin on radiation dermatitis severity. The trial was conducted at 21 community oncology practices throughout the US affiliated with the University of Rochester Cancer Center NCI’s Community Oncology Research Program (URCC NCORP) Research Base. Sleep disturbance was assessed using a single item question from the modified MD Anderson Symptom Inventory (SI) on a 0–10 scale, with higher scores indicating greater sleep disturbance. Total subjective pain as well as the subdomains of pain (sensory, affective, and perceived) were assessed by the short-form McGill Pain Questionnaire. Pain at treatment site (pain-Tx) was also assessed using a single item question from the SI. These assessments were included for pre-RT (baseline) and post-RT. For the present analyses, patients were dichotomized into 2 groups: those who had moderate-severe disturbed sleep at baseline (score≥4 on the SI; n=101) Versus those who had mild or no disturbed sleep (control group; score=0–3 on the SI; n=575). RESULTS/ANTICIPATED RESULTS: Prior to the start of RT, breast cancer patients with moderate-severe disturbed sleep at baseline were younger, less likely to have had lumpectomy or partial mastectomy while more likely to have had total mastectomy and chemotherapy, more likely to be on sleep, anti-anxiety/depression, and prescription pain medications, and more likely to suffer from depression or anxiety disorder than the control group (all p’s≤0.02). Spearman rank correlations showed that changes in sleep disturbance from baseline to post-RT were significantly correlated with concurrent changes in total pain (r=0.38; p<0.001), sensory pain (r=0.35; p<0.001), affective pain (r=0.21; p<0.001), perceived pain intensity (r=0.37; p<0.001), and pain-Tx (r=0.35; p<0.001). In total, 92% of patients with moderate-severe disturbed sleep at baseline reported post-RT total pain compared with 79% of patients in the control group (p=0.006). Generalized linear estimating equations, after controlling for baseline pain and other covariates (baseline fatigue and distress, age, sleep medications, anti-anxiety/depression medications, prescription pain medications, and depression or anxiety disorder), showed that patients with moderate-severe disturbed sleep at baseline had significantly higher mean values of post-RT total pain (by 39%; p=0.033), post-RT sensory pain (by 41%; p=0.046), and post-RT affective pain (by 55%; p=0.035) than the control group. Perceived pain intensity (p=0.066) and pain-Tx (p=0.086) at post-RT were not significantly different between the 2 groups. DISCUSSION/SIGNIFICANCE OF IMPACT: These findings suggest that moderate-severe disturbed sleep prior to RT is an important predictor for worsening of pain at post-RT in breast cancer patients. There could be several plausible reasons for this. Sleep disturbance, such as sleep loss and sleep continuity disturbance, could result in impaired sleep related recovery and repair of tissue damage associated with cancer and its treatment; thus, resulting in the amplification of pain. Sleep disturbance may also reduce pain tolerance threshold through increased sensitization of the central nervous system. In addition, pain and sleep disturbance may share common neuroimmunological pathways. Sleep disturbance may modulate inflammation, which in turn may contribute to increased pain. Further research is needed to confirm these findings and whether interventions targeting sleep disturbance in early phase could be potential alternate approaches to reduce pain after RT.