Prevalence and dynamics of clinically significant bacterial contaminants in herbal medicines sold in East Africa from 2000 to 2020: a systematic review and meta-analysis

Abstract

Background: Infectious diseases remain a leading cause of mortality and morbidity around the world, and those caused by bacteria are common in the East African region. In this region, trade and consumption of herbal medicine has been expanding in the recent decades. Herbal medicines may be contaminated with pathogenic bacteria; however, there is limited information due to fragmented studies in East Africa. In this meta-analysis, we critically analyzed original research related to the incidence of pathogenic bacterial contaminants of HM in the East African region since 2000. The aim was to create a comprehensive understanding of the extent and dynamics of bacterial contamination in HM, to guide future research and concerted public health protection in the region.

Methodology: The study was conducted according to the standards of the Preferred Reporting Items for Systematic Reviews and Meta-analyses. We searched and evaluated published articles from eleven electronic databases (Google Scholar, PubMed, HerbMed, MEDLINE, Science Direct, Scifinder Scholar, Cochrane Library, International Pharmaceutical Abstracts, EMBASE, Biological Abstracts and Commonwealth Agricultural Bureau Abstracts). Prevalences of different bacterial species, Cochran's Q test, and the I² statistic for heterogeneity were evaluated using a software called MedCalcs. Random and fixed effects models were used to determine the pooled prevalence of clinically significant bacteria from studies which were included in this meta-analysis. The potential sources of heterogeneity were examined through sensitivity analysis, sub-group analysis, and meta-regression at 95% level of significance.

Results: Fourteen studies met our inclusion criteria. Overall, the studies were highly heterogeneous (I² = 98.48%) and there was no evidence of publication bias. Escherichia coli was the most prevalent contaminant. Salmonella spp. and Shigella spp. were the most frequently reported primary pathogens with pooled prevalence of 10.4% and 6.3%, respectively. Our findings are in tandem with recent systematic reviews conducted in Europe and Asia, but are in discrepancy with the reviews recently conducted in southern Africa.

Conclusion and recommendations: The East African herbal medicine industry poses considerable health risks to communities through dissemination of clinically significant bacteria. Presence of enteric bacterial contaminants indicates possible fecal pollution of herbal medicine region-wide. Adequate research pertaining to microbial safety of herbal medicine in the East African countries remains highly desired. The latter will enable establishment of strong, region-wide herbal safety mechanisms in order to support comprehensive public health protection in East Africa.

Keywords: Bacterial contamination; East Africa; Herbal medicine; Systematic review.

Fig. 1

Flow chart for study eligibility screening of the research articles related to bacterial contamination of herbal medicines in East Africa, following PRISMA criterion

Fig. 2

Pooled prevalence estimates of bacterial contamination in herbal medicines sold in East African countries from the year 2000 to 2020, using random effects model

Fig. 3

Bias assessment plot of studies that reported bacterial contamination in herbal medicines sold in East African countries from the year 2000 to 2020

Fig. 4

Distribution of potentially harmful bacterial contaminants in herbal medicines in East Africa from the year 2000 to 2020

Fig. 5

Pooled prevalence estimates of bacterial contamination in herbal medicines sold in Kenya from the year 2000 to 2020, using random effects model

Fig. 6

Pooled prevalence estimates of bacterial contamination in herbal medicines sold in Tanzania from the year 2000 to 2020, using random effects model

Fig. 7

Bias assessment plot of studies that reported bacterial contamination in herbal medicines sold in Kenya from the year 2000 to 2020

Fig. 8

Bias assessment plot of studies that reported bacterial contamination in herbal medicines sold in Tanzania from the year 2000 to 2020

Fig. 9

Bias assessment plot of studies that reported bacterial contamination in herbal medicines sold in Uganda from the year 2000 to 2020

Fig. 10

Pooled prevalence estimates of bacterial contamination in herbal medicines sold in Uganda from the year 2000 to 2020, using random effects model

Fig. 11

Meta-regression analysis by sample size and contaminated samples (a) and by year of publication and bacteriologically contaminated HM samples (b), sold in E. Africa from 2000 to 2020

Fig. 12

Forest plot showing sensitivity analysis of the pooled prevalence of bacterial contamination in herbal medicines sold in Tanzania from the year 2000 to 2020