Abstract Background/Introduction Heart failure (HF) has high mortality and healthcare utilisation, and despite patients having similar palliative care needs to cancer patients(1), they are much less likely to have their PC needs assessed and managed (2-3). Expert consensus guidelines from cardiology and palliative care groups recommend early integration of a palliative approach in addition to usual care, but the complex and unpredictable trajectory of HF, lack of referral criteria and misunderstandings about PC are potential barriers that affect specialist PC referrals. Given the issues surrounding integration of PC in HF patients, we sought to understand the use of PC by patients with HF at our institution. Purpose To characterise the pattern and predictors of referrals of HF patients to the inpatient PC service in an Australian quaternary health service. Methods A single-centre retrospective cohort study was performed using electronic medical record data in consecutive patients admitted for HF at a quaternary health service in Australia, over a 12 month period from July 2019 to June 2020. All adults ≥ 18 years of age and admitted for HF were identified using the Diagnostic Related Group (DRG) admission code for ‘Heart Failure and Shock’. Results The 502 patients admitted for HF were elderly (median age 82 years, IQR 15) and 50% were male. Functional status was poor, with 29% unable to care for themselves and 56% requiring some level of assistance. General medicine was the major admitting unit with 67% (n=338 of 502) of patients, and the cardiology and advanced heart failure units admitted 12% (n=59) and 5% (n=26) of patients, respectively. Conversely, the main treating team in the ambulatory setting was cardiology in 51% (n=253). HF patients had high mortality rate at 32% during the median follow up time of 27 months (interquartile range 8 months). Seven percent (n=37 of 502) were referred to inpatient specialist PC, and often in the terminal phase. High dependency (defined as Australian-modified Karnofsky Performance Status 10-40) (62% vs 28%, p<0.01) and New York Heart Association class III-IV symptoms (86% vs 42%, p<0.01) were associated with referral, but one or more previous admission for HF in the last 12 months was not (35% vs. 28%, p=0.35). In all HF patients, symptoms were poorly assessed and discussion about prognosis occurred in 10% despite the high mortality. Referral to inpatient PC resulted in more prescribing of symptom medications on discharge (41% vs 2%, p<0.01) and referral to community PC (46% vs 1%, p<0.01). Conclusion Referral to inpatient specialist PC in hospitalised HF patients is low relative to the morbidity and mortality in these patients. In addition, treating teams do not routinely assess symptoms or discuss disease trajectory. More work is needed to ensure that HF patients’ PC needs are integrated into care as recommended by expert consensus.Table 1