In 2021, the HIV prevalence among South African adults was 18% and more than 2 million people had uncontrolled HIV and, therefore, had increased risk of poor outcomes with SARS-CoV-2 infection. We investigated trends in COVID-19 admissions and factors associated with in-hospital COVID-19 mortality among people living with HIV and people without HIV. In this analysis of national surveillance data, we linked and analysed data collected between March 5, 2020, and May 28, 2022, from the DATCOV South African national COVID-19 hospital surveillance system, the SARS-CoV-2 case line list, and the Electronic Vaccination Data System. All analyses included patients hospitalised with SARS-CoV-2 with known in-hospital outcomes (ie, who were discharged alive or had died) at the time of data extraction. We used descriptive statistics for admissions and mortality trends. Using post-imputation random-effect multivariable logistic regression models, we compared characteristics and the case fatality ratio of people with HIV and people without HIV. Using modified Poisson regression models, we compared factors associated with mortality among all people with COVID-19 admitted to hospital and factors associated with mortality among people with HIV. Among 397 082 people with COVID-19 admitted to hospital, 301 407 (75·9%) were discharged alive, 89 565 (22·6%) died, and 6110 (1·5%) had no recorded outcome. 270 737 (68·2%) people with COVID-19 had documented HIV status (22 858 with HIV and 247 879 without). Comparing characteristics of people without HIV and people with HIV in each COVID-19 wave, people with HIV had increased odds of mortality in the D614G (adjusted odds ratio 1·19, 95% CI 1·09–1·29), beta (1·08, 1·01–1·16), delta (1·10, 1·03–1·18), omicron BA.1 and BA.2 (1·71, 1·54–1·90), and omicron BA.4 and BA.5 (1·81, 1·41–2·33) waves. Among all COVID-19 admissions, mortality was lower among people with previous SARS-CoV-2 infection (adjusted incident rate ratio 0·32, 95% CI 0·29–0·34) and with partial (0·93, 0·90–0·96), full (0·70, 0·67–0·73), or boosted (0·50, 0·41–0·62) COVID-19 vaccination. Compared with people without HIV who were unvaccinated, people without HIV who were vaccinated had lower risk of mortality (0·68, 0·65–0·71) but people with HIV who were vaccinated did not have any difference in mortality risk (1·08, 0·96–1·23). In-hospital mortality was higher for people with HIV with CD4 counts less than 200 cells per μL, irrespective of viral load and vaccination status. HIV and immunosuppression might be important risk factors for mortality as COVID-19 becomes endemic. South African National Institute for Communicable Diseases, the South African National Government, and the United States Agency for International Development.

中文翻译：

---

南非艾滋病毒感染者中 COVID-19 入院和死亡趋势：国家监测数据分析

2021 年，南非成年人的艾滋病毒感染率为 18%，超过 200 万人的艾滋病毒感染率不受控制，因此感染 SARS-CoV-2 后出现不良后果的风险增加。我们调查了 HIV 感染者和未感染者中 COVID-19 入院趋势以及与院内 COVID-19 死亡率相关的因素。在对国家监测数据的分析中，我们链接并分析了 2020 年 3 月 5 日至 2022 年 5 月 28 日期间从 DATCOV 南非国家 COVID-19 医院监测系统、SARS-CoV-2 病例系列列表和电子疫苗接种数据系统。所有分析均包括因 SARS-CoV-2 住院且在数据提取时已知院内结局（即活着出院或已死亡）的患者。我们使用入院率和死亡率趋势的描述性统计数据。使用插补后随机效应多变量逻辑回归模型，我们比较了艾滋病毒感染者和非艾滋病毒感染者的特征和病死率。使用改良的泊松回归模型，我们比较了与所有住院的 COVID-19 患者的死亡率相关的因素以及与 HIV 感染者的死亡率相关的因素。在 397 082 名住院的 COVID-19 患者中，301 407 人 (75·9%) 存活出院，89 565 人 (22·6%) 死亡，6110 人 (1·5%) 没有记录结果。 270 737 (68·2%) 名 COVID-19 感染者记录了 HIV 状况（22 858 名感染 HIV 的人，247 879 名未感染 HIV 的人）。比较每一波 COVID-19 中未感染 HIV 的人和 HIV 感染者的特征，D614G 中 HIV 感染者的死亡率增加（调整后的比值比 1·19，95% CI 1·09–1·29），β（ 1·08、1·01–1·16)、Delta (1·10、1·03–1·18)、omicron BA.1 和 BA.2 (1·71、1·54–1·90)、以及 omicron BA.4 和 BA.5 (1·81, 1·41–2·33) 波。在所有入院的 COVID-19 患者中，既往感染过 SARS-CoV-2 的患者（调整后的发病率比为 0·32，95% CI 0·29–0·34）和部分感染过的患者（0·93，0·95% CI 0·29–0·34）死亡率较低。 90–0·96)、全面 (0·70、0·67–0·73) 或加强 (0·50、0·41–0·62) COVID-19 疫苗接种。与未接种疫苗的未感染艾滋病毒的人相比，未接种疫苗的未感染艾滋病毒的人的死亡风险较低（0·68、0·65–0·71），但接种疫苗的艾滋病毒感染者的死亡风险没有任何差异（1 ·08, 0·96–1·23)。无论病毒载量和疫苗接种状况如何，CD4 计数低于 200 个细胞/μL 的 HIV 感染者的院内死亡率较高。随着 COVID-19 流行，HIV 和免疫抑制可能是导致死亡的重要危险因素。南非国家传染病研究所、南非国家政府和美国国际开发署。