Stanford Clinics Physician Mindset Training

Overview

Mindsets play an important role in motivating and shaping health behavior and outcomes. For example, when patients have the mindset that a treatment will work, they are more likely to adhere to treatment medications and the treatment itself becomes more effective as a result of this mindset. Providers have an opportunity to shape important patient mindsets as part of clinical care, and these mindsets may influence patients' adherence to medication, screening and vaccination recommendations, and diet, exercise, and treatment recommendations that can help patients manage chronic illness. To help care teams capitalize on the potential of leveraging mindsets in medicine and improve patient health behavior and outcomes, we developed and implemented the Medicine Plus Mindset Training as part of Primary Care 2.0. Built on more than two decades of research, this training program (a) Informs Primary Care teams about the power of patient mindsets in shaping treatment outcomes (b) Provides care teams with a language and framework to identify which patient mindsets may be at play (i.e. patient mindsets about illness, treatment, their body, and the provider/care team) and (c) Equips care teams with skills and techniques to effectively shape patient mindsets to improve health outcomes. By motivating care teams to recognize patient mindsets that may be hindering health behavior change (such as "this illness is a catastrophe") or medication adherence (such as "this medication is going to cause side effects"), care teams become better equipped to help their patients adopt more useful mindsets (such as "this treatment will work," "this illness is manageable," "my body is capable," and "I am in good hands").

Healthcare providers at four Stanford clinic sites will be the investigators' main participants and the study will follow a wait-list control design. The investigators will track patient health outcomes. The intervention described in this study will only be for the healthcare team. The investigators will track both provider outcomes (using self-report survey measures) and patient outcomes (using health information already being collected by the clinic). Although the investigators will be tracking the outcomes of all physicians at the clinics, only select patient outcomes will be included as part of the study. The study will begin with the care team filling out baseline self-report surveys online and/or in person. Prior to delivering the intervention, the clinics will be split, and half of the clinic sites will be assigned to be wait-list control clinics, while the other half of sites will receive the mindset training. The intervention for the care team will be delivered in person by members of the research team. The care team will be told that the investigators are assessing the impact of a novel training program for providers.The care team will truthfully be informed that the training they are receiving is designed to enhance their interactions with patients. Questionnaires: The care team: Physicians will be asked to fill out an initial brief survey about their mindsets about connecting with patients, burnout, job satisfaction, and their efficacy using harnessing mindsets in clinical practice. Care team members will then be asked to fill out the same survey after receiving their training. Charts will be reviewed to assess patient health outcomes. Data from the electronic medical record will be used to assess patient health outcomes at all clinics. Overview of Expected Outcomes: This training was designed to improve care teams' ability to shape patient mindsets in clinical care, and therefore influence patient outcomes in the following ways: 1. Increase adherence to medication, screening, and vaccination recommendations. When care teams are better able to instill the mindset that treatment is likely to be beneficial and not harmful, patients will be more motivated to adhere to treatment recommendations. When care teams are better able to instill the mindset that patients are in good hands (e.g. the provider/care team is both warm/"gets them" and competent/"gets it"), patients will trust recommendations more and be more likely to follow through with them. 2. Reduce unnecessary antibiotic use. When care teams are better able to instill the mindsets that an illness is manageable and the patient's body is capable, patients will be more likely to believe care teams when they say antibiotics are unnecessary (e.g. in the case of colds or certain respiratory illnesses), thereby reducing overall antibiotic usage. 3. Improve health outcomes for patients with chronic illness. When care teams are better able to instill the mindsets that chronic illness is manageable, that patients' bodies are capable, that treatment is likely to be effective, and that they are in good hands, patients will be more likely to manage their chronic illness through both medication and lifestyle change, which will lead to improved health outcomes for these patients. Outcome Computation Plans for Requested Data: To assess the impact of the training, we will review patient-level primary care data from January 1, 2016-June 1, 2020. Using the data requested, we will compute outcomes for each of the three broad categories as follows. We will compare outcomes within each clinic before and after the training was implemented, and will also compare outcomes at the two initial intervention clinics to outcomes at the two initial-wait list clinics during the time period in which two of the clinics had received the training and two had not. 1. Increase Adherence Prescription fills, refills, \& discontinuations Refill requests will serve as a proxy measure for whether patients are taking their medications as prescribed. • Medications of interest: Prescriptions fills, refills, \& discontinuations for medications for chronic disease management medications such as: o Antidepressants • SSRIs \& SNIs * Statins * Hypertension medication • Beta blockers * Ace inhibitors * Calcium Channel Blockers * Thiazide diuretics * All common blood pressure medications * Diabetic medications • Variable(s) we will compute: * We will compute the percentage of patients who re-fill new prescriptions. * Denominator: Patients who have received a new prescription for one of the medications listed above. New prescriptions will be defined as any prescription not previously given in the past year (we will exclude patients for whom this is their first visit to this provider). Patients who are receiving a different prescription for the same condition will count as new prescriptions. * Numerator: Patients from the above group who requested a refill request for their eligible prescription within one year after the clinic visit. * If possible, we will also use pharmacy data to assess the number of patients who fill the new prescription initially. * If possible, we will also assess the number of patients whose medication is discontinued because the patient stopped taking the medication. Follow-up lab visit for patients with diabetes Patients with uncontrolled diabetes are recommended to have lab work every 3 months, and patients with controlled diabetes are recommended to have lab work every 6 months. Completion of such lab visits is an indication of adherence. • Variables we will compute: * We will compute the percentage of patients with uncontrolled diabetes (defined as A1c \> 8 or A1c \> 9) coming in for lab work follow ups within 5 months after clinic visit. * Denominator: Patients with uncontrolled diabetes * Numerator: Patients from the above group who come in for lab work within 5 months after clinic visit * We will compute the percentage of patients with controlled diabetes (defined as 6.5 \< A1c \< 8 or 6.5 \< A1c \< 9) coming in for lab work follow ups within 12 months after clinic visit. * Denominator: Patients with controlled diabetes * Numerator: Patients from the above group who come in for lab work within 12 months after clinic visit Referral adherence When providers refer patients for diagnostic screenings such as Colonoscopy, Fit or Mammograms, completion of these referrals is a sign of adherence. * Variables we will compute: o We will compute the percentage of patients following through with recommended screening referrals. * Denominator: Patients who are referred for Colonoscopy, Fit, or Mammogram. * Numerator: Patients from the above group who completed their screening within 6 months of order date Vaccine adherence Pneummococcol vaccinations are recommended to patients ages 65 and older. Completed pneummococcol vaccinations are an indication of patient adherence to provider recommendations. * Variables we will compute: o We will compute the percentage of patients 65 and older who received the pneumococcol vaccination. * Denominator: Patients in the clinic who are 65 and older * Numerator: Patients from the above group who completed the pneumococcol vaccination within 6 months of visit o We will compute the percentage of patients aged 65-67, who are newly eligible to receive the pneumococcol vaccination, who received the pneumococcol vaccination. * Denominator: Patients in the clinic who are between 65 and 67 years of age * Numerator: Patients from the above group who completed the pneumococcol vaccination within 6 months of visit 2. Reduce antibiotic use Reduction in antibiotic prescriptions While in most cases we hope to increase adherence to medication prescriptions, there are many cases in which antibiotics are unnecessary. For example, antibiotics are often unnecessarily prescribed for common colds, bronchitis, chest colds, flu, and sore throats. A 2016 CDC report found that an estimated 30% of antibiotics prescribed in outpatient settings are unnecessary. In order to combat antibiotic resistant bacteria, the CDC set a 2015 goal to reduce inappropriate antibiotic use in outpatient settings by 50% by 2020. Thus, reduction in overall antibiotic use is a desirable outcome. ● Variable(s) we will compute: * We will compute overall antibiotic prescriptions by looking at the number of patients prescribed antibiotics at clinic visits. * Denominator: Number of clinic encounters * Numerator: Number of antibiotics prescribed * We will compute antibiotic prescriptions for respiratory illness by looking at the number of patients with respiratory illnesses prescribed antibiotics at clinic visits. * Denominator: Number of encounters with respiratory illness diagnoses (including bronchitis and cough/congestion) * Numerator: Number of antibiotics prescribed at the above encounters 3. Improve health outcomes Reduction in patient BMI Reduction in BMI for patients who are overweight is an indication of improved health outcomes. ● Variable(s) we will compute: * Overall reduction in BMI for patients who are overweight ▪ First we will select patients with a BMI \> 25 ▪ Then we will compute change in BMI between visits for these patients. * Reduction in BMI for patients with diabetes * First we will select patients who have a diagnosis of diabetes and have a BMI \> 25 * Then we will compute change in BMI between visits for these patients. Reduction in patient blood pressure Reduction in blood pressure for patients with hypertension is an indication of improved health outcomes. * Variable(s) we will compute: * Overall reduction in systolic blood pressure for patients with hypertension ▪ First we will select patients with a diagnosis of hypertension ▪ Then we will compute change in blood pressure between visits for the above patients * Overall reduction in systolic blood pressure for patients with uncontrolled hypertension ▪ First we will select patients with blood pressure \> 140/90 ▪ Then we will compute change in blood pressure between visits for the above patients Reduction in patient A1c levels Improved A1c control for patients with diabetes is an indication of improved health outcomes. ● Variables we will compute: * Overall reduction in A1c for patients with diabetes ▪ First we will select patients with a diagnosis of diabetes ▪ Then we will compute change in A1c between visits for the above * Reduction in A1c for patients with uncontrolled diabetes ▪ First we will select patients with uncontrolled diabetes (A1c \> 8 or A1c \> 9) ▪ Then we will compute change in A1c between visits for the above patients