Smart Discharges for Mom & Baby

NCT ID: NCT05730387

Last Updated: 2025-04-10

Basic Information

• Recruitment Status: COMPLETED
• Total Enrollment: 7182 participants
• Study Classification: OBSERVATIONAL
• Study Start Date: 2022-04-14
• Study Completion Date: 2024-04-30

Brief Summary

This study aims to build a predictive algorithm that identifies mother-newborn dyads most at risk of death or complications in the 6 weeks after birth. The investigators will conduct a multi-site cohort study with 7,000 dyads in Uganda and engage with local stakeholders (e.g., patients, healthcare workers, and health policy-makers) to develop an evidence-based bundle of interventions that address key practice gaps and the critical factors leading to death and complications in these dyads. In the investigator's epidemiological study of post-delivery post-discharge outcomes in 3,236 dyads in Uganda (2017-2020), results indicated that most newborn and maternal readmissions were due to infectious illness (i.e. sepsis, surgical site infections, malaria), and primarily occurred early in the post-discharge period. Thus, the focus of this study will be identifying interventions that target these common and early outcomes, for both mothers and newborns, using World Health Organization recommendations, patient and caregiver experiences, and stakeholder recommendations. If successful, results will inform the next steps of this project, which is the external validation of the model and clinical evaluation of a personalized approach to improving health outcomes and health-seeking behaviour for mothers and newborns.

Detailed Description

PURPOSE

Neonatal outcomes are highly correlated with the health of the mother, an example of this is shown repeatedly by poor rates of survival of infants after maternal death. Prediction of risk, based on the mother and infant as a pair, is a major gap in current research and yet vital to the survival of both the mom and the infant. Thus, maternal and child health outcomes can be improved by identifying both mothers and babies at increased risk of mortality or serious morbidity after hospital discharge and allocating scarce resources for targeted follow-up to those most vulnerable. This allows the investigators to not only improve health outcomes but benefits the health system with efficient use of resources.

JUSTIFICATION

Since 2011, the investigators have been working with partners in Uganda to develop, validate, and implement an innovative program for children under 5 years who have been discharged following hospitalization for suspected sepsis. In this research and implementation program, called Smart Discharges, healthcare workers use an individualized risk prediction score to identify children at high risk of death or complications after discharge from a hospital following treatment for suspected sepsis. They can then use this score to guide the intensity of a counselling and community-referral program. While all participants receive counselling, only those above a certain risk threshold receive down-referrals to community health facilities. The investigators have shown that this approach may reduce post-discharge child mortality after in-hospital treatment for suspected sepsis by as much as 30%. Now, the investigators are working to expand their innovative precision public health approach to improving post-discharge care for mother-newborn dyads.

Findings will inform the development an evidence-based bundle of care for both the mother and newborn. This package will ensure that low-risk mother-infant pairs receive less burdensome (yet pragmatic and feasible) postpartum care, while high risk pairs receive a more extensive bundle of interventions (such as education, nutrition, healthcare interaction and community support). The Smart Discharges for Mom & Baby package will include support targeting aspects of both clinical and emotional wellbeing. Additional extensions of this work will include validating the risk models in women who deliver at home or suffer a stillbirth to ensure that more women and babies can benefit from the proposed intervention.

HYPOTHESIS

Maternal and infant characteristics collected at the time of discharge following a facility delivery can predict the risk of maternal or neonatal death or need for re-admission within six weeks of birth.

OBJECTIVE

The primary objective is to inform the development of an integrated maternal and newborn risk-based post-discharge care program. Specifically, the study aims to (1) develop and internally validate clinical risk prediction models for identifying dyads at high-risk of death or hospital readmission in the 6-week post-delivery post-discharge period, and (2) identify gaps and opportunities during in-hospital, discharge, and post-discharge care to inform the future development of an evidence-and risk-based bundle of interventions to improve postnatal care (PNC) for dyads.

DESIGN

This is a mixed-methods study using both quantitative and qualitative techniques to explore and map the current postnatal discharge processes in Uganda using data from two distinct hospital settings.

• Phase I) The team will conduct an observational cohort study informed through direct observation of the mother and newborn dyad prior to facility discharge and after delivery and follow-up telephone interviews conducted at six-weeks post-discharge.
• Phase II) The team will conduct journey mapping with a subset of dyads enrolled in the observational cohort using direct observation and follow-up telephone interviews.
• Phase III) The team will conduct a process mapping exercise using focus group discussion methodology with select facility staff.
• Phase IV) The team will conduct focus group discussions with a subset of mothers enrolled in the observational cohort, as well as their family members.

STATISTICAL ANALYSIS

Quantitative analysis: The investigators will summarize all risk factors for mothers and newborns that do and do not experience poor outcomes and estimate univariate associations. For newborns, data will be reported by sex. Derivation of prediction models will be based on optimization of the area under the receiver operating curve (AUROC) and specificity across a variety of modeling and variable selection approaches (e.g., logistic regression, elastic net, support vector machines). Model performance will be based on appropriate re-sampling techniques for internal validation (e.g., cross-validation, bootstrapping). Focus will be on developing parsimonious predictive models (e.g., 5-10 predictor variables) with high sensitivity (>80%). AUROC, sensitivity, and specificity will be reported for each model, along with positive and negative predictive values. Site specific metrics will be compared to ensure consistency across settings, and re-calibration may be considered if individual site performance is lower than expected. Finally, the investigators will assess combined sensitivity and specificity when each individual model is applied to the dyad. Outside of prediction modelling, the sample size will allow the investigators to detect an odds ratio of at least 1.30 for a given risk factor with 80% power and 5% significance and relative precision of 25%. Statistical analysis of quantitative data from journey mapping observation surveys and patient interviews will be performed using R Statistical software to obtain descriptive statistics of the frequency and distribution of each variable.

Qualitative Analysis: the investigators will analyze data collected descriptively and report summary statistics. A diagram of the discharge process will be developed, identifying key areas for improvement during the peri-discharge and post-discharge process. Focus group discussion data will be analyzed using a framework method, which allows themes to be developed inductively from participants and deductively from existing literature. Through an iterative process, transcripts will be coded and analyzed for descriptive and interpretive themes using NVivo. Descriptive themes include barriers to care and post-discharge health-seeking behaviour, while interpretive themes focus on caregiver perspectives of maternal and neonatal death and the role of the health system. The investigators will generate frequencies to describe reported medical symptoms, health-seeking behaviour, and barriers to care, and summarize common themes. Member checking will be used to improve the validity of the results, creating a summary document of the main findings that will be reviewed by health workers who participated in the focus groups. Feedback from patients and families will be obtained over telephone with research nurses who will explain the main findings verbally.

Conditions

Maternal Sepsis; Neonatal Sepsis

Study Design

• Observational Model Type: COHORT
• Study Time Perspective: PROSPECTIVE

Study Groups

mother and newborn dyads

We will recruit 6700 mother and newborn dyads from the two participating hospitals. We will continue to follow-up with all patients enrolled in the study until 6 weeks (42 days) post delivery.

Observational only

Intervention Type: OTHER

This is a non-interventional study

Interventions

Observational only

This is a non-interventional study

Intervention Type: OTHER

Eligibility Criteria

Inclusion Criteria

• Women and adolescent girls aged 12 and above delivering a single or multiple babies at the study hospital during the active recruitment phase.

Exclusion Criteria

• Inability, for whatever reason, to provide informed consent.
• Language barrier
• Mother is from a refugee camp
• Mother has no access to phone or other means for follow-up
• Mother lives outside of hospital catchment area

• Minimum Eligible Age: 12 Years
• Eligible Sex: FEMALE
• Accepts Healthy Volunteers: No

Sponsors

University of British Columbia

OTHER

Sponsor Role: lead

Responsible Party

Matthew Wiens

Assistant Professor

Responsibility Role: PRINCIPAL_INVESTIGATOR

Principal Investigators

Matthew O Wiens, PharmD, PhD

Role: PRINCIPAL_INVESTIGATOR

University of British Columbia

Locations

BC Children's Hospital Research Institute

Vancouver, British Columbia, Canada

Countries

Canada

References

Wiens MO, Trawin J, Pillay Y, Nguyen V, Komugisha C, Kenya-Mugisha N, Namala A, Bebell LM, Ansermino JM, Kissoon N, Payne BA, Vidler M, Christoffersen-Deb A, Lavoie PM, Ngonzi J. Prognostic algorithms for post-discharge readmission and mortality among mother-infant dyads: an observational study protocol. Front Epidemiol. 2023 Nov 29;3:1233323. doi: 10.3389/fepid.2023.1233323. eCollection 2023.

Reference Type: DERIVED

PMID: 38455948 (View on PubMed)

Provided Documents

Document Type: Study Protocol and Statistical Analysis Plan

View Document

Document Type: Informed Consent Form

View Document

Other Identifiers

H21-03709

Identifier Type: -

Identifier Source: org_study_id