Background

The post-hospitalization period is a precarious time for patients. Post-discharge nurse telephone call programs aiming to prevent unnecessary readmissions have had mixed results.

Objective

Describe a primary-care based program to identify and address problems arising after hospital discharge.

Design

A quality improvement program embedding registered nurses in a primary care practice to call patients within 72 h of hospital discharge and route problems within the practice for real-time resolution.

Participants

Adult patients with a primary care provider in the general internal medicine practice at the University of California San Francisco who were discharged home from the Medicine service.

Main measures

Patients reached directly by phone had a 'full-scripted encounter;' those reached only by voice-mail had a 'message-scripted encounter;' those not reached despite multiple attempts had a 'missed encounter.' Among patients with full-scripted encounters, we identified and cataloged problems arising after hospital discharge and measured the proportion of calls in which a problem was uncovered. For the different encounter types, we compared follow-up appointment attendance and 30-day readmission rates.

Key results

Of 790 eligible discharges, 486 had a full-scripted, 229 a message-scripted and 75 a missed encounter. Among the 486 full-scripted encounters, nurses uncovered at least one problem in 371 (76 %) discharges, 25 % of which (n = 94) included new symptoms, and 47 % (n = 173) included medication issues. Discharges with full-scripted and message-scripted encounters were associated with higher follow-up appointment attendance rates compared with those with missed encounters (60.1 %, 58.5 %, 38.5 % respectively p = 0.004). There was no significant difference in 30-day readmission rates (12.8 %, 14.8 %, 14.7 %; p = 0.72).

Conclusions

Our results suggest that centering a post-discharge phone call program within the primary care practice improves post-hospital care by identifying clinical and care-coordination problems early. With the new Medicare transitional care payment, such programs could become an important, self-sustaining part of the patient-centered medical home.

As behavioural science is increasingly adopted by organizations, there is a growing need to assess the robustness and transferability of empirical findings. Here, we investigate the transferability of insights from various sources of behavioural science knowledge to field settings. Across three pre-registered randomized controlled trials (RCTs, N = 314,824) involving a critical policy domain-COVID-19 booster uptake-we field tested text-based interventions that either increased vaccinations in prior field work (RCT1, NCT05586204), elevated vaccination intentions in an online study (RCT2, NCT05586178) or were favoured by scientists and non-experts (RCT3, NCT05586165). Despite repeated exposure to COVID-19 vaccination messaging in our population, reminders and psychological ownership language increased booster uptake, replicating prior findings. However, strategies deemed effective by prediction or intention surveys, such as encouraging the bundling of COVID-19 boosters and flu shots or addressing misconceptions, yielded no detectable benefits over simple reminders. These findings underscore the importance of testing interventions transferability to real-world settings.