Effect of an Electronic Health Record "Nudge" on Opioid Prescribing and Electronic Health Record Keystrokes in Ambulatory Care.

BACKGROUND: Multiple policy initiatives encourage more cautious prescribing of opioids in light of their risks. Electronic health record (EHR) redesign can influence prescriber choices, but some redesigns add to workload. OBJECTIVE: To estimate the effect of an EHR prescribing redesign on both opioid prescribing choices and keystrokes. DESIGN: Quality improvement quasi-experiment, analyzed as interrupted time series. PARTICIPANTS: Adult patients of an academic multispecialty practice and a federally qualified health center (FQHC) who received new prescriptions for short-acting opioids, and their providers. INTERVENTION: In the redesign, new prescriptions of short-acting opioids defaulted to the CDC-recommended minimum for opioid-naïve patients, with no alerts or hard stops, such that 9 keystrokes were required for a guideline-concordant prescription and 24 for a non-concordant prescription. MAIN MEASURES: Proportion of guideline-concordant prescriptions, defined as new prescriptions with a 3-day supply or less, calculated per 2-week period. Number of mouse clicks and keystrokes needed to place prescriptions. KEY RESULTS: Across the 2 sites, 22,113 patients received a new short-acting opioid prescription from 821 providers. Before the intervention, both settings showed secular trends toward smaller-quantity prescriptions. At the academic practice, the intervention was associated with an immediate increase in guideline-concordant prescriptions from an average of 12% to 31% of all prescriptions. At the FQHC, about 44% of prescriptions were concordant at the time of the intervention, which was not associated with an additional significant increase. However, total keystrokes needed to place the concordant prescriptions decreased 62.7% from 3552 in the 6 months before the intervention to 1323 in the 6 months afterwards. CONCLUSIONS: Autocompleting prescription forms with guideline-recommended values was associated with a large increase in guideline concordance in an organization where baseline concordance was low, but not in an organization where it was already high. The redesign markedly reduced the number of keystrokes needed to place orders, with important implications for EHR-related stress. TRIAL REGISTRATION: www.ClinicalTrials.gov protocol 1710018646.

Sharing Clinical Notes while Protecting Adolescent Confidentiality and Maintaining Parental Insight.

BACKGROUND: The 21st Century Cures Act mandates sharing electronic health records (EHRs) with patients. Health care providers must ensure confidential sharing of medical information with adolescents while maintaining parental insight into adolescent health. Given variability in state laws, provider opinions, EHR systems, and technological limitations, consensus on best practices to achieve adolescent clinical note sharing at scale is needed. OBJECTIVES: This study aimed to identify an effective intervention process to implement adolescent clinical note sharing, including ensuring adolescent portal account registration accuracy, across a large multihospital health care system comprising inpatient, emergency, and ambulatory settings. METHODS: A query was built to assess portal account registration accuracy. At a large multihospital health care system, 80.0% of 12- to 17-year-old patient portal accounts were classified as inaccurately registered (IR) under a parent or registration accuracy unknown (RAU). To increase accurately registered (AR) accounts, the following interventions were pursued: (1) distribution of standardized portal enrollment training; (2) patient outreach email campaign to reregister 29,599 portal accounts; (3) restriction of access to remaining IR and RAU accounts. Proxy portal configurations were also optimized. Subsequently, adolescent clinical note sharing was implemented. RESULTS: Distribution of standardized training materials decreased IR and increased AR accounts (p = 0.0492 and 0.0058, respectively). Our email campaign (response rate: 26.8%) was most effective in decreasing IR and RAU accounts and increasing AR accounts (p < 0.002 for all categories). Remaining IR and RAU accounts, 54.6% of adolescent portal accounts, were subsequently restricted. Postrestriction, IR accounts continued declining significantly (p = 0.0056). Proxy portal enhancements with interventions deployed increased proxy portal account adoption. CONCLUSION: A multistep intervention process can be utilized to effectively implement adolescent clinical note sharing at a large scale across care settings. Improvements to EHR technology, portal enrollment training, adolescent/proxy portal settings, detection, and automation in reenrollment of inaccurate portal accounts are needed to maintain integrity of adolescent portal access.

Promoting generic medication prescribing by order interface redesign: small change, large impact.

Prescription drugs are a significant component of the ever increasing health care costs. We describe the effects on generic medication prescribing behavior achieved through redesign of the order entry interface of our institutions ambulatory electronic health record. The redesign involved custom programming that automatically substituted brand medications with their generic equivalents and only allowed continuation with the brand medication if the clinician made an extra mouse click selecting "dispense as written". We conducted a before-after retrospective study around the time of the redesign and witnessed a net 36.9% percentage increase in the number of generic medications prescribed.

Early Adopters of Patient-Generated Health Data Upload in an Electronic Patient Portal.

BACKGROUND AND OBJECTIVE: Patient-generated health data (PGHD) may help providers monitor patient status between clinical visits. Our objective was to describe our medical center's early experience with an electronic flowsheet allowing patients to upload self-monitored blood glucose to their provider's electronic health record (EHR). METHODS: An academic multispecialty practice enabled the portal-linked PGHD tool in 2012. We conducted a retrospective observational study of adult ambulatory patients using this tool between 2012 and 2016, comparing clinical and demographic characteristics of data uploaders with those of a group of patients with diabetes diagnoses and patient portal accounts seen by the same health care providers. RESULTS: Over four years, 16 providers chose to use the tool, and 53 adult patients used it to upload three or more blood glucose values within any 9-month period. Of these patients, 23 were pregnant women and 30 were nonpregnant adults with diabetes. Uploaders had more encounters and portal log-ins than comparison patients but did not differ in socioeconomic status. Among the chronic disease patients, uploaders' mean hemoglobin A1c and body mass index (BMI) both dropped significantly in the months after upload. CONCLUSION: Despite the potential value of PGHD in health care, the rate of adoption of a tool allowing patients to upload PGHD to their provider's EHR has been slow. Among chronic disease patients, PGHD upload was associated with improvements in blood glucose control and BMI, but it is possible that the changes were because of increased motivation or intensive changes in medical management.

Effects of an e-Prescribing interface redesign on rates of generic drug prescribing: exploiting default options.

OBJECTIVE: Increasing the use of generic medications could help control medical costs. However, educational interventions have limited impact on prescriber behavior, and e-prescribing alerts are associated with high override rates and alert fatigue. Our objective was to evaluate the effect of a less intrusive intervention, a redesign of an e-prescribing interface that provides default options intended to "nudge" prescribers towards prescribing generic drugs. METHODS: This retrospective cohort study in an academic ambulatory multispecialty practice assessed the effects of customizing an e-prescribing interface to substitute generic equivalents for brand-name medications during order entry and allow a one-click override to order the brand-name medication. RESULTS: Among drugs with generic equivalents, the proportion of generic drugs prescribed more than doubled after the interface redesign, rising abruptly from 39.7% to 95.9% (a 56.2% increase; 95% confidence interval, 56.0-56.4%; P < .001). Before the redesign, generic drug prescribing rates varied by therapeutic class, with rates as low as 8.6% for genitourinary products and 15.7% for neuromuscular drugs. After the redesign, generic drug prescribing rates for all but four therapeutic classes were above 90%: endocrine drugs, neuromuscular drugs, nutritional products, and miscellaneous products. DISCUSSION: Changing the default option in an e-prescribing interface in an ambulatory care setting was followed by large and sustained increases in the proportion of generic drugs prescribed at the practice. CONCLUSIONS: Default options in health information technology exert a powerful effect on user behavior, an effect that can be leveraged to optimize decision making.

Association of patient recall, satisfaction, and adherence to content of an electronic health record (EHR)-generated after visit summary: a randomized clinical trial.

OBJECTIVE: Most electronic health record (EHR) systems have the capability of generating a printed after-visit summary (AVS), but there has been little research on optimal content. We conducted a qualitative study and a randomized trial to understand the effect of AVS content on patient recall and satisfaction. METHODS: Adult primary care patients (n = 272) with at least 1 chronic condition were randomly assigned to 4 AVS content conditions: minimum, intermediate, maximum, or standard AVS. Demographics and health literacy were measured at an index clinic visit. Recall and satisfaction were measured by telephone 2 days and 2 to 3 weeks after the clinic visit. RESULTS: Average age was 52 years; 75% of patients were female, 61% were Hispanic, and 21% were African American, and 64% had adequate health literacy. Average medication recall accuracy was 53% at 2 days and 52% at 3 weeks, with no significant difference among groups at either time. Satisfaction with AVS content was high and did not differ among groups. Recall of specific content categories was low and unrelated to group assignment. Health literacy was unrelated to recall and satisfaction. CONCLUSION: Primary care patients like to receive an AVS, but the amount of information included does not affect content recall or satisfaction with the information.

Brand Name Statin Prescribing in a Resident Ambulatory Practice: Implications for Teaching Cost-Conscious Medicine.

BACKGROUND: Several national initiatives aim to teach high-value care to residents. While there is a growing body of literature on cost impact of physicians' therapeutic decisions, few studies have assessed factors that influence residents' prescribing practices. OBJECTIVE: We studied factors associated with intensive health care utilization among internal medicine residents, using brand name statin prescribing as a proxy for higher-cost care. METHODS: We conducted a retrospective, cross-sectional analysis of statin prescriptions by residents at an urban academic internal medicine program, using electronic health record data between July 1, 2010, and June 30, 2011. RESULTS: For 319 encounters by 90 residents, patients were given a brand name statin in 50% of cases. When categorized into quintiles, the bottom quintile of residents prescribed brand name statins in 2% of encounters, while the top quintile prescribed brand name statins in 98% of encounters. After adjusting for potential confounders, including patient characteristics and supervising attending, being in the primary care track was associated with lower odds (odds ratio [OR], 0.38; P  =  .02; 95% confidence interval [CI], 0.16-0.86), and graduating from a medical school with an above-average hospital care intensity index was associated with higher odds of prescribing brand name statins (OR, 1.70; P  =  .049; 95% CI, 1.003-2.88). CONCLUSIONS: We found considerable variation in brand name statin prescribing by residents. Medical school attended and residency program type were associated with resident prescribing behavior. Future interventions should raise awareness of these patterns in an effort to teach high-value, cost-conscious care to all residents.

Promoting knowledge of statins in patients with low health literacy using an audio booklet.

BACKGROUND: Statins are generally well tolerated and effective at reducing a patient's risk of both primary and secondary cardiovascular events. Many patients who would benefit from statin therapy either do not adhere to or stop taking their statin medication within the first year. We developed an audio booklet targeted to low health literacy patients to teach them about the benefits and risks of statins to help the patients adhere to their statin therapy. METHODS: Through focus groups and an iterative design, an audio booklet was developed for both English-speaking and Spanish-speaking patients. We then compared the booklet with standard of care in 132 patients from our target patient population to measure its impact on knowledge and understanding of statins. RESULTS: The patients enjoyed the audio booklet and showed significant increases in knowledge after listening to it when compared with those who received the standard of care materials. CONCLUSION: The audio booklet shows promise as a tool that can be used effectively in clinical practice to teach patients about statin therapy.

Adherence to diabetes self-care behaviors in English- and Spanish-speaking Hispanic men.

We conducted a qualitative study to elicit attitudes, attributions, and self-efficacy related to diabetes self-care in both English- and Spanish-speaking Hispanic men. Transcripts from six focus groups (three in English and three in Spanish) were reviewed by the authors to extract principal and secondary themes. Participants could describe their medication and lifestyle regimens and were aware of whether they were adherent or nonadherent to physician recommendations. Lack of skills on how to incorporate diet and regular physical activity into daily living, lack of will power, and reluctance to change culturally rooted behaviors emerged as significant barriers to diabetes self-management. Medication adherence is for some men the principal diabetes self-care behavior. Nonadherence appeared to fit two profiles: 1) intentional, and 2) nonintentional. In both instances low self-efficacy emerged as a significant influence on attainment and maintenance of diabetes self-care goals. Participants also expressed a strong sense of fatalism regarding the course of their disease, and seemed to have little motivation to attempt long-term dietary control. Educational and counseling messages should stress that a diagnosis of diabetes is not a death sentence, and full functional capacity can be maintained with good control.

Receipt of recommended therapy by patients with advanced colorectal cancer.

OBJECTIVES: To evaluate utilization of surgery, chemotherapy, and radiation therapy among patients with stage II or III colon cancer and stage II/III rectal or rectosigmoid cancer, as recommended by current national guidelines. METHODS: This cross-sectional study at the Michael E. DeBakey Veterans Affairs Medical Center (Houston, TX) used 1999-2003 administrative data to identify patients with a diagnostic code for colorectal cancer. Medical charts were then abstracted to identify an incident cohort with stage II or III cancer. Outcome of interest was receipt of recommended therapy defined as surgery only (stage II colon) or surgery with adjuvant chemo- or radiotherapy (stage III colon or stage II/III rectal/rectosigmoid cancer). Potential determinants of receipt of recommended therapy were analyzed using logistic regression. RESULTS: Among 197 incident cases diagnosed or treated, mean age of patients was 66 yr (SD, 11 yr), 64% were Caucasian, and 98.5% were men. A gastroenterologist diagnosed 72.5% tumors including 62 stage II colon, 62 stage III colon, and 73 stage II/III rectal cancers. Referral to oncology occurred in 76% of stage II colon, 92% of stage III colon, and 99% of rectal cancers. 87% of stage II and 71% of stage III colon cancer patients received recommended therapy, compared to only 42.5% of rectal cancer patients. Predictors of receipt of recommended therapy among rectal cancers included being married (OR, 5.3; 95% CI: 1.6-17.1), presentation at tumor board (OR, 3.6; 95% CI: 1.2-11.2), or age<65 yr (OR, 3.5; 95% CI: 1.3-9.3). When patient's comorbidity and physician's decision-making process were considered in the assessment of the outcome, only presentation at tumor board remained predictive of receipt of recommended therapy. CONCLUSIONS: Most colon cancer patients at a major VA medical center receive recommended therapy. Among rectal cancer patients, those presented at tumor board are most likely to receive recommended therapy.