Prevalence and associated risk factors of Helicobacter pylori infection in East Africa: a systematic review and meta-analysis

Abstract

Background

Helicobacter pylori is the main cause of chronic gastritis, peptic ulcer, lymphoma, and gastric cancer in humans. The prevalence and factors associated with H. pylori infection are varied across countries. Thus, a comprehensive review has not been done on prevalence and associated factors in East Africa. Therefore, this systematic review and meta-analysis aimed to estimate the pooled prevalence and identify factors associated with H. pylori infection in East Africa.

Methods

Articles written in English language were retrieved from PubMed, Scopus, and Science Direct. Relevant articles were selected and screened using Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines. Publication bias was assessed qualitatively and qualitatively using funnel plot symmetry and Egger’s test, respectively. Heterogeneity was assessed using the I² measure. Data were analyzed using Stata software, version 14, and the “metan” command.

Results

A total of 231 articles were retrieved from nineteen countries in East Africa. Of these, 70 articles were eligible for the review. The pooled prevalence of H. pylori in East Africa was 50.98% (95% CI: 45.05–56.90). The prevalence of H. pylori infection ranged from 7.7 to 94.5% in East African countries. The highest pooled prevalence was from Sudan (61.3%, 95% CI: 52.6–69.9), and the lowest prevalence was reported from Uganda (40.7%, 95% CI: 33–48.3). Persons with no formal education (OR: 2.03; 95% CI: 1.22–2.83), lack of hand washing habit after toilet (OR: 2.24; 95% CI: 1.45–3.02), having a history of dyspepsia (OR: 2.25; 95% CI: 1.31–3.18), living in rural areas (OR = 1.80; 95% CI: 0.38–3.23), and having unclean water source (OR = 1.5; 95% CI:0.45–3.45) were all associated with higher risk for H. pylori infection.

Conclusion

More than half of the populations of East African countries were positive for H. pylori infection. Rural residence, source of water, and alcohol consumption were significantly associated with H. pylori infection. Therefore, healthcare workers could provide health education on the aforementioned risk factors, and the government and other stakeholders could improve the source of drinking water in East Africa.

Supplementary information

The online version contains supplementary material available at 10.1007/s42770-023-01190-0.

Introduction

Helicobacter pylori was discovered by Marshall and Warren at the beginning of the 1980s. The identification of H. pylori was considered as a turning point in understanding gastrointestinal microbial ecology and disease [1]. H. pylori is the main cause of chronic gastritis, peptic ulcer, mucosa-associated lymphoid tissue (MALT) lymphoma, and gastric cancer [2]. H. pylori is contagious, although the exact route of transmission is unknown [3]. Several factors such as age, socio-economic status, poor hygiene/deficient sanitation, having infected siblings, density/crowded living conditions, unsafe water sources, smoking, use of a non-steroidal anti-inflammatory drug, blood group O, high body mass index, and family history of gastric disease are associated with the acquisition and transmission of H. pylori [4–8].

About 50% of the world’s population is estimated to be infected with H. pylori, but the prevalence varies greatly among countries and among population groups within the same country. Helicobacter pylori is one of the most common infections in humans, affecting 30–40% of persons living in the developed and 80–90% of persons living in the developing world [1, 9]. The prevalence of H. pylori in Africa could reach as high as 80%, and the infection is acquired during childhood. A systematic review with meta-analysis that was carried out by Hooi et al. indicated that Africa had the highest rate of H. pylori infection with a prevalence of 70.1%, followed by South America and Western Asia with prevalence of 69.4% and 66.6%, respectively [10].

A high prevalence of H. pylori infection is expected in East Africa because of the low socio-economic status of the people living in this region. Most of the people have poor sanitary and health facilities and live in overcrowded conditions [11]. Gastritis was the most common disease associated with H. pylori infection in East Africa [12–14]. Several fragmented studies on the prevalence and associated risk factors of H. pylori from different countries in East Africa have been published. However, pooled data on the prevalence and associated risk factors of H. pylori is lacking in the region. This kind of information is vital for policymakers, regional administrators, and the scientific community to take appropriate measures. This study answered two basic questions: What is the overall pooled prevalence of H. pylori in East Africa? What factors are associated with H. pylori infections in this region? This study is a systematic review and meta-analysis of existing studies on H. pylori prevalence and associated risk factors in East Africa, and the data sources include peer-reviewed articles and gray literature. Therefore, this systematic review and meta-analysis aimed to determine the overall pooled prevalence and identify factors associated with H. pylori infections in East Africa.

Materials and methods

Eligibility criteria

Articles having reports about the prevalence of H. pylori infection and its associated risk factors in East African countries were included in this review. Original articles that were written in English language and published from 1985 to February 28, 2021, were included. All articles reporting H. pylori infection, regardless of the types of diagnostic techniques used for the study, were included in this review. The diagnostic techniques used in the eligible articles include stool antigen test, serology, rapid urease test, PCR, culture, histology, Loffler stain, immunohistochemistry, histopathology, and Giemsa stain. Studies that were conducted on animals were excluded from the analysis. The search was limited to East African countries.

Information sources and search strategies

Articles were retrieved from PubMed, SCOPUS, and Science Direct using Medical Subject Headings (MeSH) terms, keywords, and appropriate Boolean operators. An additional hand search was carried out in Google Scholar. The keywords included H. pylori and East Africa. The detailed search strategy is found in Supplementary Table S1.

Study selection

After importing identified articles into the EndNote reference manager, version 15, USA, the screening was carried out by titles and abstract. The full-text screening was conducted by ZM and MD independently; disagreements were solved in consultation with TH and EN. The selection and screening were carried out using Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines (Supplementary Table S2) [15].

Data collection process and data items

Data from the included articles were extracted using a pretested data extraction sheet. Data extraction from individual articles was carried out by two authors (ZM and MD) independently. Any discrepancy was resolved by consultation with TH and EN. The extracted data items included author name, year of publication, data collection period, study design, sample size, country, study population, diagnostic methods used, number of positive cases, estimated prevalence, and potential risk factors associated with H. pylori infection.

Quality assessment of individual studies

The Newcastle–Ottawa Scale (NOS) was used to assess the overall quality of papers included in this systematic review and meta-analysis (Supplementary Text S1). The quality of each study was declared using the three major assessment tools (methodological quality, comparability, and outcome and statistical analysis of the study). Two points were given to each criterion; publications with a total score of 5-6 points were considered to be high, 4 points to be moderate, and 0–3 points to be low quality [16].

Risk of publication bias across studies

The risks of publication bias across studies were assessed using funnel plot symmetry and Egger’s test. Funnel plots were drawn to assess the possibility of publication bias via visual inspection of the symmetry. Egger’s regression tests were used to assess publication bias quantitatively; Egger’s test p-value <0.05 suggests the presence of publication bias.

Summary measures and synthesis of results

The collected data were analyzed quantitatively using Stata software version 14, StataCorp LP, USA. There was a clear variation in the prevalence of H. pylori infection across studies included in this review. We used a random effect model at 95% confidence interval (CI) to estimate the pooled prevalence of H. pylori infection. The degree of heterogeneity was determined by I² statistic. The I² values below 25%, 25 to 50%, and above 50% were considered as low, moderate, and high heterogeneity across studies [17]. To sort out the causes of heterogeneity, subgroup analysis was conducted based on the study period, country, study design, sample size, and diagnostic methods used. In addition, a sensitivity test was used to determine the influence of each study on the pooled result. The summary measure was depicted as a forest plot. The box in the forest plot indicated the weight of articles from random effect analysis. The horizontal line indicated the 95% CI, and the solid vertical line is zero values to the x-axis. Furthermore, to pinpoint risk factors for H. pylori infection, a meta-analysis of the odds ratio was performed. All statistical analyses were carried out using the STATA statistical software package, version 14.0 (STATA Corp., College Station, Texas, USA), and the “metan” command. A statistically significant association was declared at p-value <0.05.

Results

Study selection

A total of 231 articles on the prevalence and associated risk factors of H. pylori infection in nineteen East African countries were retrieved. After removing 46 duplicate articles, the remaining 185 articles were screened by reading the title and their abstract. Thus, 12 and 55 articles were excluded after reading the title and abstracts, respectively. The remaining 118 full-text articles were assessed for eligibility based on the inclusion criteria. From the remaining 118 articles assessed, an additional 48 articles were excluded with specific exclusion criteria. As a result, only 70 articles that met the eligibility criteria were included in this systematic review and meta-analysis (Fig. 1).

Fig. 1.

PRISMA flow diagram of articles considered in this meta-analysis

Characteristics of original studies included in the meta-analysis

A total of 31,914 populations were participated in the 70 eligible articles. The studies were conducted between 1985 and 2021. The majority (55.8%) of the studies were from Ethiopia [14, 18–55], followed by Uganda. Most of the studies (68.6%) included in this review used cross-sectional study design [12–14, 18, 21, 24–34, 38–43, 45, 52, 55–78], fifty-six of the total eligible studies used a single diagnostic test to examine H. pylori infection (Table 1) [12–14, 18–20, 22–36, 38, 42–45, 47–49, 51, 52, 54–56, 58–61, 63–74, 77, 79–83].

Table 1.

Characteristics of the eligible articles included in this review

Authors	Country	Study design	Study subjects	Lab tests used	Sample size	Cases	Prevalence (%)	Quality
Abdallah et al. [56]	Sudan	Cross-sectional	Dyspeptic patients	Serology	225	148	65.8	4
Abebaw et al. [18]	Ethiopia	Cross-sectional	Dyspeptic patients	Serology	209	151	72.2	5
Ahmed et al. [79]	Sudan	Case control	Women with preeclampsia	Serology	186	132	70.96	4
Aitila et al. [57]	Uganda	Cross-sectional	Children (1 to 15 years)	Stool antigen, serology	304	74	24.3	3
Alebie [14]	Ethiopia	Cross-sectional	University students with gastritis	Serology	145	103	71	5
AliMohamed et al. [93]	Other countries	Case control	HIV seropositive patients with UGI	Rapid urease, histology	104	82	78.8	6
Alshareef et al. [58]	Sudan	Cross-sectional	Pregnant women	Serology	166	109	65.7	5
Amberbir et al. [20]	Ethiopia	Cohort	Children at age 3	Stool antigen	616	253	41.1	3
Amberbir et al. [19]	Ethiopia	Cohort	Children at age 5	Stool antigen	857	377	44	4
Ankarklev et al. [59]	Uganda	Cross-sectional	Asymptomatic children (0–12 years)	Stool antigen	427	189	44.3	5
Aoki et al. [60]	Tanzania	Cross-sectional	UGI symptoms	Serology	573	494	86.2	6
Asrat et al. [21]	Ethiopia	Cross-sectional	Dyspeptic patients	PCR, culture, rapid urease, histology, stool antigen, serology	300	273	91.0	6
Ayana et al. [12]	Tanzania	Cross-sectional	Adult dyspeptic patients	Rapid urease test	200	130	65	5
Ayele [22]	Ethiopia	Case control	Dyspeptic and non-dyspeptic patients	Stool antigen	168	13	7.7	4
Baingana et al. [61]	Uganda	Cross-sectional	Pregnant women	Stool antigen	447	2002	45.2	5
Bekele et al. [23]	Ethiopia	Retrospective	UGI symptoms	Stool antigen	1966	592	30.3	6
Belay et al. [24]	Ethiopia	Cross-sectional	Adult dyspeptic patients	Serology	208	89	42.8	5
Carrilho et al. [62]	Other countries	Cross-sectional	Dyspeptic patients	Histology, PCR	109	103	94.5	4
Desta et al. [25]	Ethiopia	Cross-sectional	Blood donors	Serology	150	133	89	3
Dilnessa [26]	Ethiopia	Cross-sectional	Dyspeptic and non-dyspeptic patients	Stool antigen	230	112	48.7
Gasim et al. [64]	Sudan	Cross-sectional	Pregnant women	Serology	186	132	71	4
Gupta et al. [80]	Uganda	Cohort	Children (1 to 10 years of age)	Serology	200	126	63	5
Hailu et al. [28]	Ethiopia	Cross-sectional	Upper GI symptoms	Stool antigen	349	177	50.7	4
Hailu et al. [27]	Ethiopia	Cross-sectional	Gastritis patients	Stool antigen	394	102	25.9	3
Henriksen et al. [94]	Ethiopia	Cohort	PUD and non PUD patients	Rapid urease, Loffler stain	290	234	80.7	4
Hestvik et al. [65]	Uganda	Cross-sectional	Healthy children (0–12 years of age)	Stool antigen	427	189	44.3	5
Hestvik et al. [66]	Uganda	Cross-sectional	HIV infected children	Stool antigen	236	53	22.5	6
Elsidding et al. [95]	Sudan	Cross-sectional	Dyspeptic patients	PCR	102	53	51.9	4
Jaka et al. [67]	Tanzania	Cross-sectional	Adult dyspeptic patients	Serology	202	79	39.1	3
Kahase [29]	Ethiopia	Cross-sectional	Adult dyspeptic patients	Stool antigen	405	208	51.4	4
Kakooza [68]	Uganda	Cross-sectional	Adults (18 and 40 years of age)	Serology	181	53	29.2	5
Kasew et al. [30]	Ethiopia	Cross-sectional	Dyspeptic patients	Stool antigen	354	133	37.6	4
Kebede et al. [31]	Ethiopia	Cross-sectional	TB and non TB patients	Stool antigen	108	20	18.5	3
Kenea et al. [32]	Ethiopia	Cross-sectional	Adult dyspeptic patients	Stool antigen	348	166	47.7	4
Kibru et al. [33]	Ethiopia	Cross-sectional	Dyspeptic patients	Stool antigen	401	210	52.4	5
Kitila et al. [34]	Ethiopia	Cross-sectional	Women of child-bearing age	Stool antigen	332	96	29	6
Lindkvist et al. [35]	Ethiopia	Case control	Children (2–4 years age)	Serology	242	116	47.9	5
Mathewos et al. [36]	Ethiopia	Retrospective	H. pylori suspects subject	Serology	1388	912	65.7	5
McLaughlin et al. [70]	Other countries	Cross-sectional	School children	Serology	87	53	60.9	6
Mbulaiteye et al. [69]	Tanzania	Cross-sectional	Participate in HIV serological survey	Serology	788	725	92	4
Moges et al. [38]	Ethiopia	Cross-sectional	Dyspeptic patients	Serology	215	184	85.6	5
Msekandiana et al. [71]	Tanzania	Cross-sectional	Children (6 months to 14 years)	Serology	200	23	11.5	6
Mubarak et al. [72]	Sudan	Cross-sectional	Pregnant women	Serology	179	125	69.8	5
Mungazi et al. [74]	Other countries	Cross-sectional	Asymptomatic patients	Serology	300	203	67.7	6
Mubiru et al. [73]	Uganda	Cross-sectional	Children with sickle cell anemia	Stool antigen	369	176	47.7	5
Mwangi et al. [75]	Other countries	Cross-sectional	Dyspeptic patients	Rapid urease test, culture, histopathology	487	199	40.8	5
Negash et al. [39]	Ethiopia	Cross-sectional	Dyspeptic patients	Stool antigen, serology	201	75	37.3	6
Ngaiza et al. [96]	Tanzania	Retrospective	Patients with gastric pathologies	Immunohistochemistry, histology	170	63	37.1	4
Noor et al. [76]	Sudan	Cross-sectional	Symptomatic and asymptomatic	Stool antigen, serology	100	53	53	4
Oling et al. [13]	Uganda	Cross-sectional	Adult dyspeptic patients	Histology	111	40	36	3
Salih et al. [77]	Sudan	Cross-sectional	Children	Serology	312	128	41	5
Schacher et al. [40]	Ethiopia	Cross-sectional	School children	Stool antigen, serology	434	285	65.7	5
Seid [41]	Ethiopia	Cross-sectional	Upper GI symptoms	Stool antigen, serology	342	104	30.4	6
Seid et al. [43]	Ethiopia	Cross-sectional	Upper GI symptoms	Stool antigen	318	99	31.1	4
Seid et al. [42]	Ethiopia	Cross-sectional	Upper GI symptoms	Serology	363	255	70.2	5
Seid et al. [44]	Ethiopia	Case control	Dyspeptic and non-dyspeptic HIV	Stool antigen	370	117	31.6	3
Shiferaw [45]	Ethiopia	Cross-sectional	Upper GI symptoms	Stool antigen	487	179	36.8	4
Taddesse et al. [46]	Ethiopia	Case control	Dyspeptic and non-dyspeptic patients	Stool antigen, serology	238	109	45.8	5
Tadesse et al. [48]	Ethiopia	Cross-sectional	Upper GI symptoms	Serology	408	340	83.3	4
Tadege et al. [47]	Ethiopia	Case control	Dyspeptic and non-dyspeptic patients	Serology	200	124	62	4
Taye et al. [49]	Ethiopia	Cohort	Children at age 6.5	Stool antigen	848	88	10.4	5
Tedros et al. [81]	Other countries	Retrospective	Dyspeptic patients	Serology	1844	386	20.9	5
Tedla [50]	Ethiopia	Cohort	Upper GI symptoms	Rapid urease, Loeffler, blue stain	444	324	73	6
Teferi et al. [51]	Ethiopia	Case control	Hyperemesis gravidrumwomen and without hyperemesis gravidrum	Stool antigen	150	37	24.7
Teka [52]	Ethiopia	Cross-sectional	HIV positive and negative patients	Serology	212	120	56.6	5
Tsega et al. [53]	Ethiopia	Case control	NUD and asymptomatic patients	Gram, Giemsa, gemineze stain	207	120	58	4
Wabinga [82]	Uganda	Retrospective	Upper GI symptoms	Endoscopy(Giemsa stain)	114	58	50.9	5
Walker et al. [83]	Other countries	Retrospective	Patients presenting for gastroscopy	Rapid urease	825	622	75.3	6
Workineh et al. [54]	Ethiopia	Retrospective	Dyspeptic patients	Serology	6566	2733	41.6	5
Yisak et al. [55]	Ethiopia	Cross-sectional	Pregnant women	Stool antigen	290	52	17.9	4

Quality of studies included in the meta-analysis

The quality scores of the studies included in this systematic review and meta-analysis ranged between three and six. The overall quality of the studies included was categorized as very good (Table 1).

Prevalence of H. pylori infections in East Africa

Out of the 31,914 study participants included in the eligible articles, 15,017 tested positive for H. pylori. The overall pooled prevalence of H. pylori infection in East Africa during the period under review was 51.04% (95% CI: 45.14–56.94; I² = 99.3%, p < 0.001) (Fig. 2). The prevalence of H. pylori infection among individual studies ranged from 7.7% in Ethiopia [22] to 94.5% in Mozambique [62]. A significant variation in the prevalence of H. pylori infection was observed within each country. The prevalence of H. pylori in Ethiopia ranged from 7.7% [22] to 91% [21], 22.5% [66] to 63% [80] in Uganda, 41% [77] to 71% [64] in Sudan, and 11.5% [71] to 92% [69] in Tanzania.

Fig. 2.

Forest plot of the pooled prevalence of Helicobacter pylori infection in East Africa

Subgroup analysis

As a result of high heterogeneity across studies observed in this systematic review and meta-analysis, we conducted a subgroup analysis based on various factors, including country of study, study design, types of diagnostic methods, sample size, and year of studies. The highest pooled prevalence was observed from Sudan (61.3%, 95% CI: 52.6–69.9), followed by Tanzania (55.2%, 95% CI: 29.1–81.3), while the lowest prevalence was observed from Uganda (40.7%, 95% CI: 33–48.3) (Table 2 and Fig. 3).

Table 2.

Pooled prevalence of H. pylori infections in East Africa across different study variables

Variables		Studies included (n)	Sample size	Cases	Prevalence (%)	95% CI	I2
Country	Sudan	8	1444	871	61.26	52.58–69.93	91.4
	Ethiopia	39	21,231	9669	48.90	41.89–55.91	99.2
	Uganda	10	2816	1260	40.66	33.04–48.27	94.2
	Tanzania	6	2133	1514	55.23	29.14–81.31	99.6
	Other East African countries	7	2931	1026	62.66	37.83–87.49	99.6
Study period	1985–2000	4	1183	794	65.08	51.36–78.80	96.2
	2001–2015	29	10,990	6513	60.52	49.85–71.19	99.5
	2016–2021	37	19,741	7710	41.96	37.08–46.84	97.9
Study design	Cross-sectional	49	13,909	7390	52.27	44.82–59.73	99.0
	Case control	9	1865	850	47.39	31.39–63.39	98.3
	Cohort	5	2639	1402	51.98	27.47–76.49	99.6
	Retrospective	7	12873	5366	45.94	32.14–59.73	99.6
Type of lab. tests used	Stool antigen	23	10,470	3651	34.64	28.71–40.58	97.8
	Serology	27	15,935	8176	60.79	50.71–70.88	99.5
	Stool antigen, serology	7	2046	889	42.89	31.11–54.67	96.7
	Several methodsa	13	3463	2297	64.32	53.93–74.71	98.0
No. of lab. tests used	Single	55	27,757	12,730	49.32	42.70–55.94	99.3
	Multiple	15	4157	2287	57.68	44.57–69.47	98.8
Sample size	<200	22	3304	1756	53.73	41.17–66.28	98.7
	201–350	23	6110	2962	49.08	39.01–59.15	98.6
	351–500	15	6030	3112	50.03	41.05–59.02	98.2
	>500	10	16,271	7182	50.73	33.58–67.89	99.8
Overall		70	31,914	15,017	51.04	45.14–56.94	99.3

^aRapid urease, histology, PCR, culture, Loffler stain, immunohistochemistry, methylene blue stain, Gram stain, Gemineze stain, Giemsa stain

Fig. 3.

Prevalence of H. pylori infection in countries of East Africa

Subgroup analysis based on the method of diagnosis showed that the highest pooled prevalence of H. pylori was obtained using a serology test (60.79%; 95% CI: 50.71–78.88), while the stool antigen test showed a lower prevalence rate (34.64%; 95% CI: 28.71–40.58) using stool antigen test. In addition, a lower prevalence of H. pylori was observed using single tests (49.32%; 42.7–55.9%) than using combined methods (57.68%; 95% CI: 44.57–69.47). We categorized the study period into three groups in order to see the trends of H. pylori infection in East Africa. The analysis showed a decreasing trend of H. pylori infection. Prevalence of H. pylori was 65.08% (95% CI: 51.36–78.80) during 1985–2000, 60.52% (95% CI: 49.85–71.19) during 2001–2015, and 41.96 % (95% CI: 37.08–46.84) in the period between 2016 to the beginning of 2021 (Table 2).

Risk of publication bias across studies included in the meta-analysis

The funnel plot symmetry suggested the presence of publication bias among studies included in this systematic review and meta-analysis (Fig. 4). Furthermore, Egger’s regression analysis test revealed that publication bias was indeed present among the studies (p <0.001).

Fig. 4.

Funnel plot symmetry showing publication bias across studies

Factors associated with H. pylori infections in East Africa

Helicobacter pylori infection was associated with different predisposing factors. The pooled odd ratio of potential risk factors such as socio-demographic factors, environmental factors, behavioral factors, and clinical factors was determined in this systematic review and meta-analysis.

Lack of formal education, rural residence, use of unprotected water source, and absence of hand washing habit after toilet showed 2-, 1.8-, 1.5-, and 2.2-fold higher risk of H. pylori infection than their counterparts, respectively. Similarly, participants who had alcohol drinking habits showed a 1.3-fold higher risk of acquiring H. pylori infection than participants who had no habit of drinking alcohol (Table 3).

Table 3.

Pooled results of potential risk factors associated with the prevalence of H. pylori infection in East Africa

Risk factors	No. of studies	Pooled odds ratio (OR)	95% CI	I2 in %	P-value	Supplementary figures
Educational status	5	2.03	1.22–2.83	0.0%	0.839	Fig. S1
Rural residency	5	1.80	0.38–3.23	66%	0.019	Fig. S2
Water sources	4	1.5	0.45–3.45	72.1%	0.013	Fig. S3
Hand washing habit after toilet	4	2.24	1.45–3.02	26.3%	0.254	Fig. S4
History of dyspepsia	8	2.25	1.31–3.18	24.8%	0.231	Fig. S5
Gender	6	1.43	1.13–1.73	34.2%	0.179	Fig. S6
Age group	5	1.51	1.25–1.78	24.2%	0.260	Fig. S7
Alcohol drinking	8	1.29	0.69–1.88	75.7%	< 0.001	Fig. S8

CI, confidence interval; OR, odds ratio; I², I² statistics

Whether the presence of gastrointestinal intestinal (GI) symptoms was associated with H. pylori infection was analyzed. Those participants that had GI symptoms such as dyspepsia were 2.25 times (OR: 2.25: 95% CI: 1.31–3.18) more likely to be infected with H. pylori infection than patients without this symptom (Table 3).

Discussion

Helicobacter pylori continues to be a major health problem globally, with high infection rates being reported from the developing world. The prevalence of H. pylori infection is higher in Africa, particularly in sub-Saharan Africa, as a result of high population density, unimproved drinking water sources, poor hygienic conditions, limited awareness, and ignorance of the infection in the region. The prevalence of H. pylori infection is even higher in East African countries, as reported by several fragmented studies. Summarized evidence is vital to designing appropriate intervention strategies by policymakers of the region. This study demonstrated that more than half (51%) of the East African population are positive for H. pylori infection. This result is in line with reports by the World Gastroenterology Organization [84], China [85], systematic review and meta-analysis in Ethiopia [37], and Asian countries [10]. In contrast to our results, a higher prevalence of H. pylori was reported in Latin America and the Caribbean [10] and Mongolia [86]. These reported differences might be attributed to methodological, socio-economical, behavioral, environmental, socio-demographic factors, and time trend of studies.

The trend of H. pylori infection in East Africa showed a decreasing trend over time from about 65% (between 1985 and 2000) to 41.96% (between 2016 and 2022). This decreasing trend might be ascribed to relative improvements in socio-economic status, personal and environmental hygiene and behavior factors, ways of prevention, and control of strategies for H. pylori infection in the region. A similar decreasing trend of H. pylori infection was reported in a systematic review and meta-analysis conducted by Hooi et al. in different regions, namely Europe, Northern America in Oceania [10], and Ethiopia [37].

The subgroup analysis showed that the highest (61.3%) and the lowest (40.7%) pooled prevalence rates of H. pylori infection were observed in Sudan and Uganda, respectively. These differences among countries might be attributable to socio-economic, socio-demographic, environmental, behavioral, and clinical factors and types of diagnostic strategies used in the studies.

The prevalence of H. pylori infection also differs based on the laboratory techniques used for the diagnosis. The highest prevalence of H. pylori was recorded when sensitive laboratory tests like rapid urease, histology, PCR, culture, and immunohistochemistry were used. On the contrary, the lowest prevalence was observed when a stool antigen test was applied. Our finding is in line with a study comparing the diagnostic capacity of PCR techniques with others for the detection of H. pylori infection [87]. The pooled prevalence of H. pylori was 57.68% when multiple diagnostic tests were used, whereas the prevalence was 49.32% when a single test was used. This clearly showed that the use of combined diagnostic methods would increase the detection of H. pylori.

Several risk factors for H. pylori have been reported by different studies conducted in East African countries. The results of these meta-analyses showed that living in rural areas was significantly associated with H. pylori infection. This might be attributed to the poor sanitary practice and lack of access to clean water as well as lack of awareness about H. pylori infection. This is in accordance with a study in Alaska [88] and China [85], but it is not in agreement with a study carried out in Egypt [89].

This study showed that study subjects who did not attend formal education were two times more likely to be infected with H. pylori than those subjects who attended formal education. It is believed that those who attended formal education might get better awareness about H. pylori and its mode of transmission, which would help them to apply preventive measures against the infection, unlike their counterparts.

The present study revealed that the source of drinking water is significantly associated with H. pylori infection. Study participants who drank water from unprotected water sources were more likely to be infected with H. pylori than those who used protected water sources. It is known that unprotected water may act both as a vehicle for the dissemination and source of H. pylori infection whenever the water sources are contaminated with human feces. This finding was consistent with studies conducted in Alaska [88] and Ethiopia [18, 24, 39], unlike a study reported from the five largest islands of Indonesia [90].

Among the behavioral factors, alcohol drinking was significantly associated with H. pylori infection. Even though the type, level, amount, and frequency of consumption were not described, those who consumed alcohol were about 1.3 times at higher odds of acquiring H. pylori infection compared to those who did not consume alcohol. This particular phenomenon might be attributed to the direct damaging effect of alcohol on the gastric mucosal layer, possibly predisposing consumers to H. pylori infection. It has also been implicated that alcohol could contribute to the adherence of the bacterium to its specific niche in facilitating infection within drinkers. This is in agreement with studies done in Ontario [91], China [92], Indonesia [90], and Ethiopia [38].

The results of this study showed that those participants with gastrointestinal symptoms were 3.8 times more likely to be infected with H. pylori than those who did not have gastrointestinal symptoms. This might be due to the effect of GI symptoms providing a growing medium, i.e., a change in gut microbiota, gastric ulceration, thinning of gastric walls, favoring a change in pH for the growth of bacteria, and establishing the infection. This is in line with studies reported in Tanzania [12] and Ethiopia [37, 47].

Limitations of the study

This study provides relevant information about the pooled prevalence of H. pylori infection and its associated risk factors in East Africa, but it also has its own limitations. First, the articles included in this meta-analysis were not derived from all countries of East Africa. Information about H. pylori infection was lacking from a few countries in East Africa. Second, this study included articles written only in the English language. Those articles written in French (especially Francophone countries) were not included. Third, most of the articles included in this meta-analysis were hospital based, and this may affect the generalizability of the findings of this meta-analysis.

Conclusion

This systematic review and meta-analysis revealed that more than half of the study subjects were positive for H. pylori in East Africa. The prevalence of H. pylori infection among countries of East Africa greatly varied. Rural residence, lack of formal education, use of unprotected water sources, lack of hand washing habit after the toilet, and history of dyspepsia were potential risk factors for H. pylori infection. Therefore, the governments and other stakeholders could provide safe drinking water sources and toilet facilities to the community. In addition, healthcare workers improve the awareness of the community about those risk factors described above. Moreover, populations in the region should implement recommendations given by healthcare workers to reduce the burden and impacts of H. pylori infection in East Africa.

Supplementary information

Below is the link to the electronic supplementary material.

Abbreviations

CI

confidence interval

GI

gastro intestine

HIV

human immune virus

MALT

mucosa-associated lymphoid tissue

MeSH

medical subject heading

NOS

Newcastle–Ottawa Scale

NUD

non-ulcer dyspepsia

PCR

polymerase chain reaction

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analysis

PUD

peptic ulcer disease

RUT

rapid urease test

UGI

upper gastro intestine

Author contribution

ZM and MD were involved in the design, data collection, processing, interpretation of the findings, and drafting of the manuscript. TH, EN, and DM were involved in the conception of the idea, drafting, reviewing, and editing of the manuscript.

Data availability

All data generated or analyzed during this study are included in this published article.

Code availability

Not applicable.

Declarations

Ethical approval

Not applicable.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declared no competing interests.