The burden of hospital-acquired infections (HAI) in sub-Saharan Africa: a systematic review and meta-analysis

Background

Hospital-acquired infections (HAI) are a leading cause of morbidity and mortality globally. These infections are diverse, but the majority are lower respiratory tract infection (LRTI), surgical site infection (SSI), bloodstream infection (BSI), and urinary tract infection (UTI). For most sub-Saharan African countries, studies revealing the burden and impact of HAI are scarce, and few systematic reviews and meta-analysis have been attempted. We sought to fill this gap by reporting recent trends in HAI in sub-Saharan Africa (SSA) with attention to key patient populations, geographic variation, and associated mortality.

Methods

Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, we conducted a literature search of six electronic databases (Web of Science, Pubmed, APA PsycInfo, CINAHL, Embase, and the Cochrane Library) to identify studies assessing the prevalence of HAI in SSA countries. Studies published between 01 January 2014 and 31 December 2023 were included. We applied no language or publication restrictions. Record screening and data extractions were independently conducted by teams of two or more reviewers. Using the R software (version 4.3.1) meta and metafor packages, we calculated the pooled prevalence estimates from random-effect meta-analysis, and further explored sources of heterogeneity through subgroup analyses and meta-regression. This study is registered with PROSPERO, CRD42023433271.

Findings

Forty-one relevant studies were identified for analysis, consisting of 15 from West Africa (n = 2107), 12 from Southern Africa (n = 2963), 11 from East Africa (n = 2142), and 3 from Central Africa (n = 124). A total of 59.4% of the patient population were associated with paediatric admissions. The pooled prevalence of HAI was estimated at 12.9% (95% CI: 8.9-17.4; n = 7336; number of included estimates [k] = 41, p < 0.001). By subregions, the pooled current prevalence of HAI in the West Africa, Southern Africa, East Africa and Central Africa were estimated at 15.5% (95% CI: 8.3-24.4; n = 2107; k = 15), 6.5% (95% CI: 3.3-10.7; n = 2963; k = 12), 19.7% (95% CI: 10.8-30.5; n = 2142; k = 11) and 10.3% (95% CI: 1.1-27.0; n = 124; k = 3) of the patient populations respectively. We estimated mortality resulting from HAI in SSA at 22.2% (95% CI: 14.2-31.4; n = 1118; k = 9).

Interpretation

Our estimates reveal a high burden of HAI in SSA with significant heterogeneity between regions. Variations in HAI distribution highlight the need for infection prevention and surveillance strategies specifically tailored to enhance prevention and management with special focus on West and East Africa, as part of the broader global control effort.

Funding

No funding was received for this study.

Prevalence of locoregional recurrence and survival post-treatment of head and neck cancers in Africa: a systematic review and meta-analysis

Background

Recurrent cancers of the head and neck are associated with poor survival outcome. Yet, their burden in Africa is not reliably known. We therefore aimed to estimate the prevalence of recurrence and the 5-year overall survival among patients treated for head and neck cancers (HNC) in Africa.

Method

In this systematic review and meta-analysis, we searched four electronic databases (Pubmed, CINAHL, MEDLINE, and Web of Science) and the grey literature for studies reporting the prevalence of HNC recurrence and 5-year overall survival post treatment, published between January 1, 2002, and December 31, 2022. We contacted corresponding authors of relevant studies. Searches were extended to reference lists of review articles and other relevant sources for potentially eligible studies. Each record was assessed for inclusion or exclusion by two independent reviewers. Records with individual-level data on recurrence and survival conducted in Africa were included while exclusion was based on the study design and availability of relevant data. Data were independently extracted by three reviewers from eligible studies, and summary estimates were sought. Our primary outcomes were recurrence and 5-year overall survival of patients who have been treated for HNC, and our secondary outcomes included risk factors, tumor site, squamous cell histology, clinical stage of tumor, and treatment options received. Only records selected for primary outcomes were assessed for secondary outcome data extraction. Random-effects meta-analysis was conducted for each outcome. Meta-regression models were used in addressing sample heterogeneity among the studies. Protocol for this study was registered with PROSPERO, CRD42022372307.

Findings

This systematic review and meta-analysis returned 3998 records, yielding 28 included studies after exclusion. Eighteen studies reported on the prevalence of HNC recurrence while 24 articles reported on the 5-year overall survival. Of the pooled total study population, 7199 (70.5%) of 10,218 patients were males while 2603 (25.5%) were females. We found that the prevalence of HNC recurrence was 15.4% (I² = 96.2%; 95% CI: 9.5-22.3; n = 3214; k = 18), and the 5-year overall survival was 54.4% (I² = 99.5%; 95% CI: 40.1-68.4; n = 9798; k = 24). We also found that the prevalence of smoking and alcohol consumption as risk factors for HNC were 42.6% (I² = 98.8%; 95% CI: 25.2-61.0; n = 4374; k = 15) and 35.8% (I² = 98.9%; 95% CI: 21.7-51.4; n = 4110; k = 11) respectively. The pooled current prevalence for advanced HNC (clinical stages III-IV) was 80.0% (I² = 99.2%; 95% CI: 68.6-89.5; n = 7624; k = 18) compared to 12.2% (I² = 96.4%; 95% CI: 6.2-19.8; n = 7624; k = 18) in early disease (clinical stages I-II).

Interpretation

The results showed significantly high prevalence of cancer recurrence, poor 5-year overall survival and very high prevalence of advanced cancers at time of diagnosis. This study provides robust evidence for strategies towards prompt diagnosis and appropriate management of HNC to improve patients' outcome in the African continent.

Funding

This study was not supported by any funding.