Registry for Integrative Medicine Interventions Effectiveness

Rationale: According to the Agency for Healthcare Research and Quality (AHRQ), observational databases provide an opportunity for simultaneous exploration of the impact of integrative medicine on various medical conditions. As opposed to selecting a single condition for which patients may seek care at our sites, the project will allow us to collect prospective patient-reported outcomes on a number of frequently seen medical conditions. By clustering participants at the multiple IM clinical sites with the same medical condition, we will be able to compare the impact of various integrative therapies on patient-reported outcomes. Objectives: Primary Objective 1\. To evaluate the change in patient-reported outcomes (e.g. quality of life, mood and stress) over time Secondary Objectives 1. To evaluate whether patient-reported outcomes differ by baseline characteristics of the participants (e.g. demographics, clinical condition, intervention sought) 2. In a subset of clinics, to collect cost data from patient encounters to allow cost comparisons among the interventions and outcomes OUTCOME ASSESSMENTS Patient Reported Outcomes (PROs) The PRIMIER Registry will use the Research Electronic Data Capture (REDCap) as an online research management tool that can be used to collect PRO data. REDCap is a free, online research management tool that enables researchers to create study-specific Web sites for capturing participant data securely. Patient-Reported Outcomes (PROs) is the term used to denote health data that is provided by the patient through a system of reporting. A PRO is basically a patient's feedback on their feelings or functions as they are dealing with chronic diseases or conditions. PROs can also be measured when patients are undergoing treatment or are participating in a clinical trial. The Patient-Reported Outcome Measurement Information System (PROMIS®), funded by the National Institutes of Health, aims to provide clinicians and researchers access to efficient, valid, and responsive self-reported measures of health, including symptoms, function and well-being. PROMIS® instruments (short forms, computerized adaptive tests (CATs), profiles) are a central feature of the Assessment Center instrument library. PROMIS® is unique from other PRO systems in four ways: 1. PROMIS® data is comparable because there are common measurements across domains 2. PROMIS® measures are reliable and valid. They have been subjected to rounds of rigorous review and testing to ensure that the measurements are precise, consistent, and accurate. 3. PROMIS® is flexible: PRO tools or measures can be obtained by way of in-person interviews, written questionnaires or computer-assisted tools. 4. PROMIS® is inclusive in that all people regardless of literacy, language, physical function and life status can use its tools The PRIMIER Registry will use the PROMIS-29 as the core data PRO collection tool along with the Perceived Stress Scale-4 (PSS-4)and the Patient Activation Measure (PAM). Cost Indicators In a subset of sites, we will collect cost indicators of healthcare resource utilization, including prescription medications, over-the-counter medications, physician and ER visits, as well as measures of patient functionality such as work productivity. This analysis will use the treatment utilization data obtained from the patient visit questionnaires as these data allowed us to count the specific number of treatments each patient reported over the course of the study. Patients will be asked to report the number of each type of IM treatment session received at every data collection time point. In the event that a patient skipped an interim study visit, they will be asked to report cumulative counts of IM treatment sessions received since their previous visit. We recognize that patients may not readily recall this information, so to ensure accuracy of this measure, research coordinators (or other research staff) will review the patients' medical record to quantify the type and number of visits to the participating integrative medicine clinical sites during study participation. Statistical Design Primary Analysis The main analyses of the PRIMIER Registry will be descriptive and summary. Major Secondary Analyses Since the PRIMIER Registry will collect many of the same variables over a 2 year period, the secondary analyses will include examining the repeated variables and changes from baseline. General Statistical Methods We will summarize descriptive statistics using frequencies (percentages) for categorical variables. We will report means, standard deviations, medians, quartiles, minimums, and maximums for all continuous variables. We will use univariate analyses to detect outliers which will be subsequently submitted as queries to sites to ensure clean data. Questionnaires that can be aggregated into a score will be scored and the scores will be used for summarizing. We will perform repeated-measures analyses of variance to detect changes in PRO and cost indicators over time. We will summarize results with a point estimate of mean change in the outcome measures between baseline and any specified follow-up visit with 95% confidence intervals. We will report P values for overall model summary, indicating any difference between any two given points. If the P value is significant (\<0.05), we will perform follow-up linear regression analyses to verify that the trends are consistent in direction throughout all study visits, indicating constant improvement or decline from visit to visit. Exploratory Analyses We will look for variables and trends that are of interest to the collaborating investigators to explore. The appropriate analysis plan will be developed to take advantage of the wealth of data that will be collected in the PRIMIER Registry. At this time, we do not know what future analyses will be conducted. We do not know the medical conditions for which individual participants in this PRIMIER Registry will seek help at the IM clinic, nor will we know in advance which IM interventions patients will receive. With these limitations, we cannot predefine subgroups for statistical analyses at present. However, in the future we will define post-hoc groups by clinical condition (e.g., participants with cancer, chronic pain, headaches, etc.) or IM intervention received (e.g., acupuncture, meditation, massage, etc.) for further analyses.