Cardiovascular

Heart Failure Care and Post-Hospital Outcomes in Northwestern Tanzania.

TL;DR

Mortality following HF admission in Northwestern Tanzania remains high, and factors associated with post-hospital mortality may include poor linkage to post-discharge care and low sustained care engagement.

Key Findings

Mortality was very high both during the index HF hospitalization and after discharge among adults admitted to Bugando Medical Centre in 2023.

  • Among 177 patients admitted with HF, 23 (13%) died during the index hospitalization.
  • Of the 154 patients alive at discharge, 63 (40.9%) died within six months.
  • 86 (56%) of discharged patients died within a year.
  • Vital status was ascertained by telephone, with follow-up through August 31, 2025.

There were substantial gaps in the quality of inpatient and post-discharge HF care among patients discharged alive.

  • 79% received indicated testing.
  • 50% were on guideline-directed medical therapy (GDMT) at discharge.
  • 68% attended the first post-discharge appointment.
  • 53% received ongoing care three months after discharge, and 39% were on GDMT at three months.

Attending the first post-discharge visit was independently associated with improved survival.

  • Adjusted hazard ratio 0.32 (95% CI: 0.17, 0.63).
  • Estimated using Cox proportional hazards models of gaps in quality measures and mortality.
  • Retrospective chart review design, so the association is not proof of causation.

Remaining in care three months after discharge was independently associated with improved survival.

  • Adjusted hazard ratio 0.36 (95% CI: 0.16, 0.81).
  • Only 53% of patients alive at discharge received ongoing care three months post-discharge.

GDMT use was low across outpatient visits, with frequent medication changes and de-escalation over time.

  • GDMT use across outpatient visits was 47%.
  • GDMT use fell from 50% at discharge to 39% at three months post-discharge.
  • Medication changes and de-escalation were frequent.

The authors conclude that implementation research is needed to improve post-discharge HF outcomes in East Africa, focusing on pre-discharge GDMT optimization and early linkage to care.

  • Poor linkage to post-discharge care and low sustained care engagement may contribute to post-hospital mortality.
  • The study took place at a large academic regional referral hospital in Mwanza serving a catchment of about 22 million people.
  • Study was a retrospective chart review of all adult HF admissions in 2023 (n=177).

What This Means

This study looked at the records of 177 adults admitted with heart failure to a large referral hospital in Mwanza, Tanzania, in 2023, and followed them by chart review and phone calls until August 2025. Outcomes were poor: 13% died in the hospital, and of those who went home, about 41% died within six months and 56% within a year. Many patients did not receive recommended care after leaving the hospital. Only half were on the recommended heart failure medications at discharge, about two-thirds attended their first follow-up visit, just over half were still in care three months later, and fewer than 4 in 10 were on the recommended medications at that point. Patients who attended the first follow-up visit or who remained in care at three months had substantially lower risk of dying, even after accounting for other factors. This research suggests that getting patients onto appropriate medications before discharge and connecting them to follow-up care soon after may be important targets for improving survival. Because the study was a retrospective review at a single hospital, it shows associations rather than proving that follow-up care causes better survival. The authors call for further research on how to improve care engagement and medication use after heart failure hospitalization in East Africa.

Have a question about this study?

Citation

Kalokola F, Wajanga B, Deodatus M, Mujuni E, Mayanja P, Majule D, et al.. (2026). Heart Failure Care and Post-Hospital Outcomes in Northwestern Tanzania.. Global heart. https://doi.org/10.5334/gh.1593