IIR 11-083
Effects of Mental Health Disorders on Pain Monitoring and Treatment in VA CLCs
Penny L. Brennan, PhD VA Palo Alto Health Care System, Palo Alto, CA Palo Alto, CA Funding Period: March 2013 - February 2016 Portfolio Assignment: Mental and Behavioral Health |
BACKGROUND/RATIONALE:
Recent VHA health care guidelines mandate prompt, effective pain treatment for residents of VA Community Living Centers (CLCs), a vulnerable population whose physical health and functioning may limit their ability to successfully initiate and sustain efforts to obtain effective treatment for their pain. Up to 60% of VA CLC residents have mental health disorders, including dementia, serious mental illness, depressive disorders, PTSD, and substance use disorders. Anecdotal clinical evidence and some research findings suggest that these mental health disorders may form an additional barrier to CLC residents obtaining adequate assessment and treatment for their pain. OBJECTIVE(S): This investigation determined short- and longer-term prospective relationships between CLC residents' mental health disorders and their subsequent pain treatments and pain treatment outcomes. It achieved these aims: (a) determined the short-term influence of older adults' mental health disorders on the pain, pain treatments, and pain treatment outcomes they experience as VA CLC residents; (b) determined whether a lack of specific MDS-assessed pain treatments, and lower intensity pain treatment, mediated between having a mental health disorder and poorer short-term pain treatment outcomes; and (c) ascertained how older adults' baseline mental health disorders affect the longer-term (9- month) course of their pain treatment outcomes, and the relationship of these 9-month pain treatment outcome trajectories to residents' 9-month physical, cognitive, and psychological functioning trajectories. METHODS: Minimum Dataset 3.0 (MDS 3.0) resident assessments comprised the data for this project. The MDS 3.0 resident assessment instrument is a tool used by VA CLC staff at residents' admission and discharge, and at quarterly intervals, to assess their mental and physical health conditions, pain, pain treatments, and functioning in the areas of activities of daily living, cognition, and psychosocial well-being. Our sample included VA CLC residents able to self-report pain, cognitive functioning, and depressive symptoms, who completed an MDS 3.0 resident assessment interview at baseline in July 2012. From these baseline assessments we extracted information about residents' demographic, diagnostic, pain, pain treatment, and functioning characteristics. Next, we obtained follow-up information in these domains from residents who were assessed with subsequent quarterly MDS 3.0 assessments three or more times over the next year. We analyzed data from these samples using descriptive statistics, logistic regression, latent growth trajectory modeling, parallel process modeling, and mediation analyses. All analyses adjusted statistically to account for variation in residents' demographic characteristics, mental health diagnoses, and, where analyses involved pain treatment, pain. FINDINGS/RESULTS: Aim 1: We found significant differences in pain and pain treatments of VA CLC residents with and without mental health disorder diagnoses. Residents with diagnosed dementia and serious mental illness (SMI; bipolar disorder, schizophrenia, and other psychoses) were significantly less likely than other residents to report having pain in the last 5 days; they also reported less frequent, less severe, and less debilitating pain (i.e., disruptive of sleep and daily activities). In comparison with other residents, those with dementia and SMI diagnoses were at heightened risk of obtaining no pain treatment, they were also less likely to obtain scheduled medication, as-needed medication, and non-pharmacological treatment for pain. In contrast, residents with diagnosed depressive disorders, post-traumatic stress disorder (PTSD), and substance use disorder (SUD) reported more frequent, severe, and debilitating pain than did other residents. They also were more likely to obtain scheduled pain medication and more intensive pain treatment (i.e., scheduled pain medication augmented with as-needed pain medication and/or non-pharmacological pain treatments). Three months after baseline assessment almost half (48%) of all residents had persisting pain; 12% had controlled pain, 10% had new pain, and about 30% were pain-free. Having a dementia or SMI diagnosis significantly reduced the likelihood of having persisting pain and doubled the likelihood of being pain-free from baseline to follow-up. Having a depressive disorder, PTSD, or SUD diagnosis at baseline predicted elevated risk of having persisting pain, and reduced the likelihood of being pain-free, at follow-up. Aim 2: We found no evidence for a mediating effect wherein mental health disorder diagnoses negatively impact residents' receipt of pain care, in turn leading to poorer pain treatment outcomes. In fact, residents with diagnosed depressive disorder, PTSD, or SUD at baseline were more likely to be provided scheduled pain medications, and to obtain higher intensity pain treatment, than were other residents; this was in turn associated with reduced risk of persisting pain over the 3-month follow-up interval. Aim 3: There was considerable variability in the 9-month pain treatment outcome trajectories of CLC residents. More than half (58%) of residents had low initial levels of pain frequency, severity, and interference, and these remained low for the next 9 months. However, 20% of residents experienced stable, high pain, 10% initially high and increasing pain, and 12% initially high but decreasing pain. Residents with dementia and SMI diagnoses were less likely to belong to the pain trajectory classes with higher initial and increasing pain symptoms whereas those with depressive disorder and PTSD diagnoses were at elevated risk of belonging to trajectory classes with higher initial and increasing pain frequency and severity. Having a PTSD diagnosis at baseline doubled the risk of having high initial and increasing pain interference. Having a baseline SUD diagnosis tripled the risk of being in the high, increasing pain trajectory category. However, it also increased the likelihood of belonging to the class of individuals with high but declining pain frequency and pain interference. Dementia and SMI diagnoses had no indirect negative effect, via pain symptoms, on residents' functioning at the 9-month follow-up. However, depressive disorder, PTSD, and SUD diagnoses at baseline were all indirectly associated, via elevated pain symptoms, with more depressive symptoms at the follow-up. IMPACT: These findings show that VA CLC residents with mental health disorders differ from other residents in their short- and long-term experience of pain, pain treatments, and pain treatment outcomes. They suggest a need for improved pain assessment and pain treatment methods especially for Veteran residents with dementia and SMI diagnoses. Further research is needed to better understand the reasons for elevated pain and special pain treatment needs of Veteran residents who have depressive disorder, PTSD, and SUD diagnoses. Information from this investigation can be used to inform educational and resident-care approaches aimed at improving the quality of pain assessment and pain treatment for Veteran CLC residents who have mental health disorders. External Links for this ProjectNIH ReporterGrant Number: I01HX000656-01A2Link: https://reporter.nih.gov/project-details/8398810 Dimensions for VADimensions for VA is a web-based tool available to VA staff that enables detailed searches of published research and research projects.Learn more about Dimensions for VA. VA staff not currently on the VA network can access Dimensions by registering for an account using their VA email address. Search Dimensions for this project PUBLICATIONS:Journal Articles
DRA:
Mental, Cognitive and Behavioral Disorders, Aging, Older Veterans' Health and Care
DRE: Treatment - Observational, Prognosis Keywords: none MeSH Terms: none |