Maze Michael James, Williman Jonathan, Anstey Rebekah, Best Emma, Bhally Hasan, Bryce Aliya, Chang Catherina L, Chen Kevin, Dummer Jack, Epton Michael, Good William R, Goodson Jennifer, Grey Corina, Grimwade Kate, Hancox Robert J, Hassan Redzuan Zarool, Hills Thomas, Hotu Sandra, McArthur Colin, Morpeth Susan, Murdoch David R, Pease Fiona Elizabeth, Pylypchuk Romana, Raymond Nigel, Ritchie Stephen, Ryan Deborah, Selak Vanessa, Storer Malina, Walls Tony, Webb Rachel, Wong Conroy, Wright Karen
Department of Medicine, University of Otago, Christchurch, New Zealand.
Department of Population Health, University of Otago, Christchurch, New Zealand.
IJID Reg. 2024 Aug 13;12:100424. doi: 10.1016/j.ijregi.2024.100424. eCollection 2024 Sep.
COVID-19 severity prediction scores need further validation due to evolving COVID-19 illness. We evaluated existing COVID-19 risk prediction scores in Aotearoa New Zealand, including for Māori and Pacific peoples who have been inequitably affected by COVID-19.
We conducted a multicenter retrospective cohort study in adults hospitalized with COVID-19 from January to May 2022, including all Māori and Pacific patients, and every second non-Māori, non-Pacific (NMNP) patient to achieve equal analytic power by ethnic grouping. We assessed the accuracy of existing severity scores (4C Mortality, CURB-65, PRIEST, and VACO) to predict death in the hospital or within 28 days.
Of 2319 patients, 582 (25.1%) identified as Māori, 914 (39.4%) as Pacific, and 862 (37.2%) as NMNP. There were 146 (6.3%, 95% confidence interval 5.4-7.4%) deaths, with a predicted probability of death higher than observed mortality for VACO (10.4%), modified PRIEST (15.1%) and 4C mortality (15.5%) scores, but lower for CURB-65 (4.5%). C-statistics (95% CI) of severity scores were: 4C mortality: Māori 0.82 (0.75, 0.88), Pacific 0.87 (0.83, 0.90), NMNP 0.90 (0.86, 0.93); CURB-65: Māori 0.83 (0.69, 0.92), Pacific 0.87 (0.82, 0.91), NMNP 0.86 (0.80, 0.91); modified PRIEST: Māori 0.85 (0.79, 0.90), Pacific 0.81 (0.76, 0.86), NMNP 0.83 (0.78, 0.87); and VACO: Māori 0.79 (0.75, 0.83), Pacific 0.71 (0.58, 0.82), NMNP 0.78 (0.73, 0.83).
Following re-calibration, existing risk prediction scores accurately predicted mortality.
由于新冠病毒疾病不断演变，新冠病毒疾病严重程度预测评分需要进一步验证。我们评估了新西兰现有的新冠病毒疾病风险预测评分，包括对受新冠病毒疾病影响不平等的毛利人和太平洋岛民。
我们对2022年1月至5月因新冠病毒疾病住院的成年人进行了一项多中心回顾性队列研究，包括所有毛利人和太平洋岛民患者，以及每第二名非毛利、非太平洋岛民（NMNP）患者，以便按种族分组实现同等的分析能力。我们评估了现有严重程度评分（4C死亡率、CURB-65、PRIEST和VACO）预测住院期间或28天内死亡的准确性。
在2319名患者中，582名（25.1%）为毛利人，914名（39.4%）为太平洋岛民，862名（37.2%）为非毛利、非太平洋岛民。共有146例（6.3%，95%置信区间5.4 - 7.4%）死亡，VACO评分（10.4%）、改良PRIEST评分（15.1%）和4C死亡率评分（15.5%）的预测死亡概率高于观察到的死亡率，但CURB-65评分（4.5%）的预测死亡概率较低。严重程度评分的C统计量（95%CI）分别为：4C死亡率：毛利人0.82（0.75，0.88），太平洋岛民0.87（0.83，0.90），非毛利、非太平洋岛民0.90（0.86，0.93）；CURB-65：毛利人0.83（0.69，0.92），太平洋岛民0.87（0.82，0.91），非毛利、非太平洋岛民0.86（0.80，0.91）；改良PRIEST：毛利人0.85（0.79，0.90），太平洋岛民0.81（0.76，0.86），非毛利、非太平洋岛民0.83（0.78，0.87）；VACO：毛利人0.79（0.75，0.83），太平洋岛民0.71（0.58，0.82），非毛利、非太平洋岛民0.78（0.73，0.83）。
重新校准后，现有风险预测评分能准确预测死亡率。
