Medical and Resource Facilitation Intervention After Traumatic Brain Injury

Summary

Participation Requirements

Description

This Integrated Medical and Resource Facilitation Intervention (MRFI) trial will address: 1) The ineffective connection of individuals with traumatic brain injury (TBI) and their families to specialized medical and community resources after hospital-based care; 2) Limited access to TBI experts, and;...

This Integrated Medical and Resource Facilitation Intervention (MRFI) trial will address: 1) The ineffective connection of individuals with traumatic brain injury (TBI) and their families to specialized medical and community resources after hospital-based care; 2) Limited access to TBI experts, and; 3) A lack of knowledge by primary care providers (PCP) about the complex needs of individuals with TBI. This clinical trial will test a theory-driven complex behavioral intervention that integrates the medical and TBI expertise of Mayo's Brain Rehabilitation Clinic (BRC) with Minnesota Brain Injury Alliance's (MN BIA) highly developed Resource Facilitation (RF) program. The target populations for the proposed study are: 1) Individuals with TBI eligible for MN BIA RF services; 2) Their families or caregivers; and 3) Their PCPs. This research uses a community-based pragmatic clinical trial (PCT) to test a complex behavioral intervention. A PCT: 1) Compares multiple clinically relevant interventions rather than comparing an intervention of interest against a placebo or control condition; 2) Includes a heterogeneous population of participants in multiple experimental settings as opposed to a narrowly controlled homogeneous sample in practice settings of a single kind; and 3) Measures a broad range of health care outcomes. The active elements in the proposed trial define a 'complex behavioral intervention'. The intervention's integrated MRFI model of care consists of: multiple components within the experimental and control interventions that interact; multiple and complex behaviors associated with those delivering and receiving the intervention; multiple target populations; multiple outcome measurements; and, adapting the intervention based on need and context. At trial's end, the investigators will estimate the difference in the use of health care resources between individuals with TBI who receive the intervention versus those who receive usual care, by completing a cost-effectiveness analysis. True success of a clinical intervention is achieved when its benefits are realized at a similar or diminished cost compared to an alternative treatment. Specific Aims Specific Aim 1: To assess the effectiveness of a medical/RF intervention provided remotely in improving participation outcomes of individuals with TBI when compared to a similar group receiving usual care. Hypothesis 1: Participants receiving the trial intervention will show greater improvement in participation outcomes when compared to those receiving usual care. Specific Aim 2: To test for the differences in caregiver burden and quality of life in the families of individuals with TBI involved in each arm of this trial. Hypothesis 2: The family members of individuals with TBI receiving the trial intervention will report lower caregiver burden and higher quality of life when compared to family members of individuals with TBI receiving usual care. Specific Aim 3: To measure differences in efficacy and mastery among PCPs in caring for individuals with TBI. Hypothesis 3: PCPs for individuals with TBI receiving the trial intervention will report higher self-efficacy and mastery, when compared to providers for individuals with TBI receiving usual care. Specific Aim 4: To compare the cost-effectiveness of the integrated service intervention arm of the trial with the usual care arm. Hypothesis 4: Cost-effectiveness will be the same or less overall for individuals with TBI receiving the trial intervention when adjusted for covariates compared to those individuals receiving usual care. Methods Sample Minnesota has statutory authority to identify individuals discharged from hospitals with TBI for epidemiological surveillance and to connect them with the MN BIA to offer RF. During subsequent RF intake, the recruitment and consent process will commence with RF staff introducing the study. A list of interested individuals will be securely submitted to the Mayo Research Coordinator; these individuals will then be contacted for consent and randomized. If the individual with TBI is unable to consent based on a cognitive screen, their legal representative will be approached. Participants will not be required to have a family member or PCP enrolled to be eligible for the study, but will be strongly encouraged to involve both. A consented participant's family and PCP will be contacted for consent as research participants. They will be assigned to the same group as the individual with TBI. Randomization to ensure representation of known prognostic factors where age (?65 and >65), gender, and residence (rural or urban) will be used as stratification factors, with treatment assignment balanced in blocks of 4. Because a simple random scheme may limit enrollment of rural participants the investigators will over-sample in a ratio of 2:1 rural/urban. Sample Size and Power Calculations For Specific Aim 1, a sample size of 500 individuals is considered desirable based on power calculations as described. By achieving this sample size a minimum difference of at least 0.25 standard deviations can be detected between groups for each continuous outcome of interest, with 80% power based on a two-sample t-test. In 2014, 3,868 people 15 years old or older were hospitalized with TBI in MN and 1,108 people with TBI received RF in MN in 2016. The investigators plan a 24-month enrollment period, so 2,216 individuals with TBI will be potentially eligible. The minimum detectable difference between the groups based on a two sample t-test will be 0.25 standard deviations. The sample size of 250 per group will have 80% power to detect a correlation of 0.18 or larger between any two continuous outcomes within the group. For an assessment of associations between group and other categorical variables, there would exist 80% power to detect a difference of 11% or larger for the presence of a categorical outcome (15% among participant with usual care versus 25% among participants receiving the study intervention) based on a Chi-square test of two proportions. Intervention All intervention components will be delivered remotely: there will be no face-to-face interaction with the research participants. The complex intervention being tested is comprised of the clinical, educational, and supportive services of the Mayo BRC integrated with the MN BIA RF program. The specific ICT modes used to interact with intervention participants, and the specific clinical services that are provided by these modes, will be determined by clinical need, individual preference, and technological capacity. Individuals with TBI, their family members and PCPs assigned to the usual care group will receive care and provide services as usual in their communities. Individuals with TBI assigned to the usual care group will receive RF as routinely provided by MN BIA. For individuals with TBI, their families and PCPs assigned to the intervention group, their health, social, community, education, support, resource, and care coordination needs will be assessed. Individuals with TBI assigned to the intervention group will receive RF as integrated into the intervention. Every intervention group participant in all target populations will receive a unique combination of services. It is expected that the composition of each participant's integrated team will vary based on individual needs, preferences, and technological capacity. Ideally, all teams will include a PCP. If a PCP is not already established, the Advanced Practice Registered Nurse will help facilitate this. The MRFI intervention expands on a previously established web-based platform and other ICT, incorporating synchronous direct clinical care (telemedicine). Other synchronous and asynchronous intervention will be delivered as needs indicate. Participants will use their own devices and internet access for any and all interactions with Mayo Clinic providers. Data Analysis Data for continuous variables will be summarized using descriptive statistics. Frequencies and percentages will be calculated for categorical variables. Baseline characteristics and demographic variables will be compared using two sample t-test or Wilcoxon rank sum test for continuous variables and using Chi-square test or Fisher's exact test for categorical variables of interest. Associations between continuous variables within each group will be estimated using Pearson or Spearman correlation coefficient. Comparison of correlations across the groups of interests will be performed using the Fisher's r-to-z transformation, which calculates a value of z that can be applied to assess the significance of the difference between two correlation coefficients. All the statistical tests will be 2-sided with an alpha level of 0.05. Statistical analysis for Specific Aim 1 For each individual outcome measure the investigators will first compute percent change from baseline to last follow-up measurement to account for baseline measurements. Further analysis will be performed using these percent changes. However, analyzing effectiveness using each individual measure will not provide a single p-value to measure the intervention's overall effect on participation outcomes. Therefore, the investigators will perform the analysis using a composite endpoint that will allow us to assess the global impact of the intervention using single comparisons performed as two sample t-tests or a Wilcoxon rank sum test. All analyses will be performed on an intent-to-treat basis. If there are any baseline covariate imbalances between the groups, the investigators will perform linear regression analyses, where summed ranks for each individual will be an outcome of interest and the primary predictor variable will be group, adjusting for those covariates that had imbalances among the groups at baseline. Statistical analysis for Specific Aims 2 and 3 The primary goal of these Aims is to measure caregiver burden and family needs, and efficacy and clinical mastery in PCPs. Percent change from baseline will be reported as mean and standard deviation or median as appropriate. Comparisons of percents between the groups assigned to intervention versus usual care will be performed using two sample t-tests or Wilcoxon rank sum test as appropriate. If there is imbalance in any baseline covariates of interest, analysis will be performed using linear regression adjusting for those covariates of interest. Statistical analysis for Specific Aim 4 In order to assess the cost-effectiveness of the proposed intervention, the trial patients will be linked to the Minnesota All Payers Claims Database (MN-APCD) developed and maintained by MDH. Both healthcare utilization and cost outcomes will be compared between the participants in the intervention and usual care arms. Adjustments will be made to account for different payment levels between governmental and commercial payers. Descriptive statistics of these outcomes including mean, median, and standard deviation will be provided. If any difference between the intervention and usual care groups in terms of their baseline characteristics is observed, multivariate adjustments will be conducted; due to expected skewness of underlying distribution, utilization outcomes will be analyzed using count regression (e.g., negative binomial regression) while cost will be analyzed using generalized linear modeling with gamma distribution for cost, and logarithmic link. In order to meet the confidentiality terms of the data submitters to MN-APCD, trial participants' records will be matched by applying the same MN algorithms to generate the hashed identifiers used to link claims for the same person through the following set of identifiers: name (first, middle, and last name), 5-digit zip code, age, gender, and date of birth. To carry out this linkage and avoid the possibility of participant re-identification, the creation of the analytic dataset from MN-APCD and subsequent analyses will be conducted by the MDH staff. The match rate of the trial participants to the MN-APCD will be reported.

Tracking Information