Data dashboards are a common audit and feedback approach to support changes in processes and behavior. The rural Transitions Nurse Program (TNP) is a care coordination intervention for high-risk Veterans. An interactive dashboard was used to provide real-time performance metrics to sites. The feedback goal was to increase TNP Veteran discharges. One-year post implementation, discharge goals were not met. Control theory suggests that feedback using diverse methods can positively influence performance. Nudge emails can draw attention to performance metrics to improve awareness of current state. This study evaluated whether Veteran discharges and site communication increased when feedback occurred through a dashboard plus weekly nudge email versus dashboard alone.

This observational study reviewed discharge counts of urban, rural, and highly rural Veterans who were hospitalized and discharged from four VA hospitals participating in TNP. Transitions nurses and site champions implementing TNP at each site were surveyed.

Veteran discharge counts during the dashboard phase were compared to discharges during the dashboard plus weekly nudge email phase. The emails included run charts (figure) with site discharge averages and weekly discharge counts. The difference of means for weekly discharges between the two phases was calculated using Poisson distribution. After 3 months of nudge emails, project time was calculated and a survey assessing nurse and champion perceptions of the nudge emails was distributed.

Our sample included four VA medical centers. The average weekly discharges for all sites during the ~20-month dashboard phase was 4.23 Veterans. The weekly average during the 3-month dashboard plus nudge email phase was 4.21 Veterans. The difference in means was -0.03 (p=0.73). Adjusting for time trends had no further effect. Project time to create and communicate nudge emails was ~14 hours over 3 months. Four nurses responded to the survey. Two reported neutral and two reported positive perceptions of the nudge emails. No site champions responded to the survey.

Drawing attention to metrics, through nudge emails, maintained, but did not increase TNP Veteran discharges compared to dashboard feedback alone. These findings suggest the nudge effort has no effect on TNP.

The Division of Cardiology at the University of Colorado reviewed a “near-miss” patient safety event related to delayed performance of a STAT priority electrocardiogram (ECG) in a floor unit patient with high-risk chest pain. While the importance of prompt ECG completion (<10 mins) during pre-hospital and emergency room care is well-established,1 no similar measures exist for hospitalized patients. Meanwhile, delays in ECG acquisition contribute to delayed diagnosis and treatment of in-hospital STEMI.2. There are few published attempts to improve ECG completion times among hospitalized patients.

Analysis of 25,159 completed, time-stamped inpatient ECGs completed at UCH between 1/1/2018 and 10/30/2018 was completed. Descriptive statistics for ECG volume, ECG completion delays and total AHT staffing levels were calculated and correlations between ECG characteristics and delays were examined. Between February and April 2019, a trained qualitative researcher completed interviews and observations with UCH staff involved in ECG completion, focusing on work processes and the workplace environment. Based on these initial quantitative and qualitative analyses, a simulation model was developed to evaluate changes in 1) technician staffing models; 2) the proportion of STAT orders; and 3) nurse training to help perform ECGs.

ECGs were ordered with a STAT priority in 40% of cases and, among individual providers, use of the STAT priority varied from 7% to 95%. ECG completion was delayed (>15 minutes) for 35% of STAT ECGs, compared to less than 10% of non-stat ECGs. In qualitative interviews, technicians described a “cascade” effect to delays resulting from a compounding effect of a series of late ECGs and supported by the quantitative observation that delays are strongly correlated with STAT ECG volume. Technicians also described spikes in ECG ordering during hours in which staffing levels are low and for non-emergency reasons (such as discharge), a finding again supported by quantitative analysis. Results of discrete event simulation suggest: adding technician staffing hours during the day outperforms reducing the proportion of STAT ECGs; short shifts (4-8 hours) may be a cost effective way to add personnel; ECG training for Cardiology nurses could offload technicians and reduce delays; the negative effect of technician “attrition” – e.g. a technician calls in sick and is not replaced – is more powerful than the positive effect of any intervention.

In light of the above findings, UCH operations leadership trained charge nurses on Cardiology units to perform STAT ECGs and are discussing new ECG technician staffing models. The NavLab plans to create a dashboard for leadership to review ECG timeliness and the performance of ECG technicians.

MARGO S. HARRISON
Assistant Professor, University of Colorado

Postpartum contraception is important to prevent unintended, undesired, and closely-spaced pregnancies.1 If women receive comprehensive contraceptive education, around 11% will choose to use the contraceptive implant.3 Prior research from our community has found barriers to the use of long-acting reversible contraceptives (LARC) include lack of spousal approval, difficulty accessing methods, lack of knowledge, and fear of adverse effects.4 We hypothesized addressing access by bringing contraceptives to women’s homes would result in increased implant uptake.

In our study population of interest in the Southwest Trifinio, analysis of historical unpublished data suggests that about 88% of women in the region are using or are interested in using contraception by forty days postpartum. Of these users, 0.5% used condoms, 0.5% pills, 0.5% lactational amenorrhea, 1.5% natural family planning, almost 4% long-acting reversible contraceptives (around 3% using the implant), 21% sterilization, and 72% opted for injectable contraception.

Our study was a cluster-randomized parallel arm pragmatic trial to observe the association of home-based postpartum contraceptive provision, including the contraceptive implant, with implant utilization rates at 3 months post-enrollment.

In a region of rural Guatemala referred to as the Southwest Trifinio, twelve communities are served by a community-based antenatal and postnatal care program. The communities were combined into eight clusters based on 2017 birth rates and randomized to receive the home-based contraceptive delivery (condoms, pills, injection, implant) during the routine 40-day postpartum visit. All participants receive comprehensive contraceptive counseling beginning at the first antenatal visit, so control clusters received this as part of routine care; this education preceded the study intervention.

Of 208 women enrolled in the study, 108 were in four intervention and 100 in four control clusters. Three-month contraceptive initiation rates were 56.0% in the control clusters compared to 76.8% in the intervention clusters, p < 0.001. Women in control clusters overwhelmingly opted for the injectable contraceptive (94.6%) while women in intervention clusters chose both the injection (61.5%) and the implant (33.7%), p < 0.001. Implant use by 3 months, the primary outcome of the study, was significantly higher in the intervention arm (25.9%) compared to the control arm (2.0%), p < 0.001, OR 18.8 CI [4.3, 81.4].

Our study was designed to respond to previously identified barriers to contraceptive uptake, and it was successful. It increased overall use of contraception by 3 months, by shifting use away from short- in favor of long-acting methods, with high continuation and satisfaction rates and no adverse outcomes reported.