Traditional Eye Remedies and Ocular Complications in Sub-Saharan Africa: A Systematic Review.

Abstract

Traditional eye medicine (TEM) refers to substances or practices used outside conventional ophthalmic care to treat eye conditions, including plant extracts, animal products, oils, breast milk, and locally prepared mixtures applied directly to the eye. In Sub-Saharan Africa, TEM remains widely used due to cultural beliefs, limited access to formal eye care, and reliance on traditional healers, with reported prevalence ranging from 5.9% to 82.3%.

This systematic review evaluates the prevalence, types, and ocular outcomes associated with TEM use in Sub-Saharan Africa. The review was conducted in accordance with PRISMA 2020 guidelines. Databases searched included PubMed, Scopus, Web of Science, Cochrane Library, and African Journals Online, with supplementary grey literature identified through Google Scholar. Studies were included if they examined TEM use in human populations within Sub-Saharan Africa and reported clinical outcomes or complications. Methodological quality was assessed using the Newcastle-Ottawa Scale and a structured epidemiological framework.

Twelve studies across five countries were included. Plant-based remedies were the most frequently reported, followed by chemical substances, breast milk, and honey. TEM use was consistently associated with delayed hospital presentation, increased disease severity at admission, and a higher burden of corneal complications. Reported outcomes included corneal ulceration, corneal opacity, endophthalmitis, panophthalmitis, and irreversible vision loss.

TEM remains a significant contributor to preventable ocular morbidity in Sub-Saharan Africa. Addressing this requires community education on the risks of unregulated eye treatments, improved access to ophthalmic care, and collaboration with traditional healers to promote timely referral.

Introduction

Visual impairment and blindness remain major public health concerns worldwide, with a disproportionate burden in low- and middle-income countries. Sub-Saharan Africa carries one of the highest burdens of preventable blindness globally, driven by limited access to eye care services, shortages of trained ophthalmic personnel, and delayed health-seeking behaviour. ¹ An estimated 5.08 million people in Sub-Saharan Africa were bilaterally blind in 2020, with an age-standardised prevalence of approximately 0.99%, nearly double the global average of 0.52%.² Globally, an estimated 43.3 million people lived with blindness in 2020, with the greatest burdens concentrated in South Asia and Sub-Saharan Africa.³

Common causes of visual impairment in Sub-Saharan Africa include cataract, glaucoma, trachoma, refractive errors, and corneal diseases.⁴ Corneal pathology, particularly ulcers and infections, contributes significantly to irreversible vision loss, especially in rural populations where access to ophthalmic care is limited.⁵ Delayed presentation is a major factor in poor outcomes.

One important contributor to delayed presentation is the widespread use of traditional medicine. The World Health Organization estimates that up to 80% of people in Africa rely on traditional medicine for primary health care needs.⁶ Traditional eye medicine refers to substances or practices applied directly to the eye outside conventional ophthalmic care, including plant extracts, animal products, saliva, breast milk, and locally prepared mixtures.⁷ These remedies are commonly used for symptoms such as redness, pain, blurred vision, and eye infections.

The use of TEM has been widely reported across several African countries, with prevalence ranging from 5.9% in hospital-based studies in Nigeria to as high as 82.3% among ophthalmic patients in south-east Nigeria.^8-12 In Uganda, up to 60% of patients with microbial keratitis reported prior TEM use.¹³ The continued use of these remedies is driven by cultural beliefs, accessibility of traditional healers, financial constraints, and limited availability of formal eye care services.¹⁴

Although often perceived as harmless, many traditional eye practices are associated with harmful outcomes. The application of non-sterile substances can introduce infection, cause chemical injury, and delay appropriate treatment. ⁷ Reported complications include corneal ulcers, severe infections, conjunctival scarring, and irreversible blindness.^15-16 TEM use has also been associated with significantly delayed presentation and worse visual outcomes.^7,13

Despite increasing evidence of these risks, the literature remains fragmented, with variations in reported prevalence, types of remedies, and clinical outcomes. Earlier efforts to address TEM-related harm, including training programmes for traditional healers in Malawi, demonstrated measurable reductions in TEM-associated corneal disease, though no recent comprehensive synthesis brings together this evidence across Sub-Saharan Africa. ^17,18,19

Aim of the Review

This systematic review evaluates the prevalence, types, and ocular outcomes associated with traditional eye medicine use in Sub-Saharan Africa.

Methods

Review Design

This study was conducted as a systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines. ²⁰ The review synthesized evidence from observational studies, cross-sectional surveys, cohort studies, and hospital-based investigations reporting the use of traditional eye medicine (TEM) and associated clinical outcomes in Sub-Saharan African populations.

Protocol and Registration

A study protocol was developed prior to conducting the review, outlining the research objectives, eligibility criteria, search strategy, data extraction procedures, and quality assessment methods. The protocol was structured in accordance with the PRISMA-P statement.²¹ This review has been registered in PROSPERO (registration number CRD420261342713; available at https://www.crd.york.ac.uk/PROSPERO/view/CRD420261342713).

Eligibility Criteria

Population

Studies were included if they involved human participants residing in Sub-Saharan Africa presenting with ocular diseases or eye-related symptoms who had used traditional eye medicine (TEM) prior to seeking formal medical care.

Intervention

The intervention of interest was the use of traditional eye medicine (TEM): substances or practices applied directly to the eye for the purpose of treating eye diseases. Examples include plant extracts, herbal preparations, breast milk, oils, animal products, and mixtures prepared by traditional healers. ^{7, 12}

Comparison

Where available, studies comparing outcomes between patients who used TEMs and those who received modern ophthalmic treatment were included. Studies without a comparison group were also included if they reported clinical outcomes associated with TEM use.

Outcomes

Primary outcomes were ocular complications associated with TEM use: corneal ulceration, corneal opacity, conjunctival scarring, ocular infections, endophthalmitis, panophthalmitis, and visual impairment. Secondary outcomes included delayed presentation to health facilities and need for surgical intervention, such as enucleation.

Study Design

Eligible study designs included observational studies, cross-sectional studies, hospital-based descriptive studies, case series, and cohort studies. Randomised controlled trials were also eligible, although such trials are rare given the ethical challenges involved.

Time Frame and Language

Studies were eligible from database inception to December 2025 to capture both foundational and contemporary evidence on traditional eye medicine use and its ocular outcomes in Sub-Saharan Africa. Only English-language publications were included.

Exclusion Criteria

Studies were excluded if they involved animal experiments, laboratory investigations without human participants, editorials, commentaries, or opinion pieces, or if they were conducted outside Sub-Saharan Africa or did not report clinical ocular outcomes associated with TEM use.

Search Strategy

A comprehensive search was conducted across PubMed, Scopus, Web of Science, Cochrane Library, and African Journals Online. African Journals Online was included specifically to capture regional research not indexed in major international databases.²² Google Scholar was used for supplementary identification of grey literature and citation tracking. The following keyword combinations were used:

("traditional eye medicine" OR "traditional eye remedies" OR "herbal eye treatment" OR "indigenous eye medicine" OR "traditional eye medication")

AND ("ocular complications" OR "corneal ulcer" OR "corneal disease" OR "visual impairment" OR "eye disease" OR "microbial keratitis")

AND ("Africa" OR "Sub-Saharan Africa" OR "Nigeria" OR "Uganda" OR "Malawi" OR "Tanzania" OR "Ethiopia" OR "Ghana")

Medical Subject Headings (MeSH) terms were also used where appropriate in PubMed. Boolean operators (AND, OR) were applied consistently across all databases; search strings were adapted to each database’s syntax and controlled vocabulary where applicable. No date limits were applied, as the review was designed to capture the full historical scope of evidence. No language filters were applied at the database search stage; however, after screening, only English-language publications met the eligibility criteria, representing a potential source of language bias (see Limitations). No species filter was applied in the initial search; non-human studies were excluded at the screening stage. The final database search was conducted on 14 March 2026. Additional sources were identified through manual review of the reference lists of included articles. Table 1 provides a breakdown of records retrieved per database.

Table 1.

Database Search Log, Records Retrieved per Source

Database	Search Date	Records Retrieved	Notes
PubMed	14 Mar 2026	298	MeSH terms applied
Scopus	14 Mar 2026	187	Title/abstract search
Web of Science	14 Mar 2026	134	All databases
Cochrane Library	14 Mar 2026	39	CENTRAL register only
African Journals Online	14 Mar 2026	91	All journals
Google Scholar (supplementary)	15 Mar 2026	98	First 10 pages; grey literature
Reference list screening	15 Mar 2026	23	Manual review of included articles
Total	-	870	Before deduplication

Full reproducible search strings for each database are provided in Supplementary Appendix A.

Study Selection Process

All records were imported into a reference management system and duplicates removed before screening. The selection process followed the PRISMA framework in three stages: title screening, abstract screening, and full-text review. Reasons for exclusion at the full-text stage were documented. The complete flow of records through each stage is presented in Figure 1.

Figure 1.

PRISMA 2020 flow diagram of study selection.

Pooled quantitative meta-analysis was not performed. Clinical heterogeneity was present across included studies, reflecting differences in the populations studied (ranging from unselected ophthalmic outpatients to patients with specific conditions such as microbial keratitis or corneal ulcers), the types of TEM reported, and the outcome measures used. Methodological heterogeneity was also substantial, with variation in study design (cross-sectional, cohort, case-control, and descriptive), quality of reporting, and the presence or absence of formal comparison groups. Outcome variability precluded the use of a common effect measure. Subgroup synthesis by country or TEM type was considered but judged insufficiently powered given the small number of studies per subgroup. The review, therefore, presents a narrative qualitative synthesis of the available evidence.

Data Extraction

A structured data extraction form was used to collect: author name, year, country, study design, sample size, type of traditional eye remedy used, ocular condition treated, reported outcomes, reported complications, and author-identified limitations. Extracted data were organised in tabular form to facilitate cross-study comparison.²³

Quality Assessment

Methodological quality of included cohort and case-control studies was assessed using the Newcastle-Ottawa Scale (NOS), which evaluates selection of participants, comparability of study groups, and measurement of outcomes. ^24,25 Cross-sectional studies were assessed using four epidemiological criteria adapted from standard epidemiological appraisal frameworks: clarity of objectives, adequacy of sample size, reliability of outcome measurement, and control for confounding. The JBI Critical Appraisal Checklist for cross-sectional studies was considered; however, given that several included studies were older hospital-based descriptive reports predating current standardised checklists, a pragmatic four-criterion approach was applied to enable consistent assessment across the full set of included studies. Quality assessment was performed independently by two reviewers; disagreements were resolved by discussion and consensus. Formal inter-rater agreement was not calculated using a kappa statistic, which the authors acknowledge as a limitation. Quality assessment results are reported in Table 3.

Table 3.

The methodological quality assessment results. Cohort and case-control studies

Author(s)	Year	Design	Tool	Score / Criteria Met	Quality Judgment
Courtright et al.	1994	Hospital cohort	NOS	6/9 (S3, C1, O2)	Moderate, no formal comparison group for some outcomes
Yorston & Foster	1994	Hospital descriptive	Cross-sectional (4 criteria)	2/4 met	Low-moderate, sample size not stated; no comparison group
Lewallen & Courtright	1995	Case-control	NOS	7/9 (S4, C2, O1)	High, matched design; microbiological confirmation
Bisika et al.	2009	Population survey	Cross-sectional (4 criteria)	3/4 met	Moderate, population-based sampling; limited clinical outcome data
Eze et al.	2009	Cross-sectional hospital	Cross-sectional (4 criteria)	4/4 met	High, large sample; multivariable analysis; validated tool
Ukponmwan & Momoh	2010	Hospital descriptive	Cross-sectional (4 criteria)	2/4 met	Low-moderate, sample size not stated; no statistical analysis
Nwosu & Obidiozor	2011	Cross-sectional hospital	Cross-sectional (4 criteria)	3/4 met	Moderate, consecutive sampling; limited confounding control
Achigbu & Achigbu	2014	Cross-sectional hospital	Cross-sectional (4 criteria)	2/4 met	Low-moderate, sample size not stated; limited outcome detail
Arunga et al.	2019	Mixed- methods cohort	NOS	8/9 (S4, C2, O2)	High, prospective; comparison group; microbiological data; 3-month follow-up
Eticha et al.	2023	Cross-sectional hospital	Cross-sectional (4 criteria)	4/4 met	High, pre-calculated sample; multivariable analysis; validated tool
Msonge et al.	2023	Cross-sectional hospital	Cross-sectional (4 criteria)	3/4 met	Moderate, clear objectives; adequate sample; limited confounding control
Sherief, Sitotaw & Girma	2024	Cross-sectional	Cross-sectional (4 criteria)	3/4 met	Moderate, rural primary care setting; self-report; limited clinical outcome data

Table 3 footnotes: NOS = Newcastle-Ottawa Scale; S = Selection (0-4), C = Comparability (0-2), O = Outcome (0-3).

Results

Overview of Included Evidence

Twelve studies met the inclusion criteria and were included in the final qualitative synthesis, representing primary research from five Sub-Saharan African countries: Nigeria (n=4), Uganda (n=2), Malawi (n=3), Ethiopia (n=2), and Tanzania (n=1). Studies were published between 1994 and 2024. The majority were hospital-based cross-sectional or cohort studies. Across the seven studies that reported sample sizes, the confirmed total sample was 5,436 participants (Courtright et al. n=583; Eze et al. n=2,542; Nwosu & Obidiozor n=500; Arunga et al. n=313; Eticha et al. n=502; Msonge et al. n=182; Sherief et al. n=814). Among these seven studies, the median sample size was 502 participants (range: 182–2,542). Five studies (Yorston & Foster 1994; Bisika et al. 2009; Ukponmwan & Momoh 2010; Achigbu & Achigbu 2014; Lewallen & Courtright 1995 [39 cases]) did not report a total sample size; these studies were retained for qualitative synthesis given their relevance to TEM types and reported complications, but were excluded from any prevalence calculations. All studies involved patients presenting to ophthalmology clinics or eye units with ocular complaints.

The included literature consistently indicates that TEM use is common before formal medical presentation, that it is associated with delayed arrival at hospital, and that it is linked to a higher burden of ocular complications in comparison with patients who seek early conventional care. Because most included studies were observational and many lacked comparison groups, findings should be interpreted as evidence of association rather than established causation.

Characteristics of Included Studies

Table 2 summarises the key characteristics of all twelve studies included in the review. A column recording the primary type of traditional eye remedy reported in each study has been included to facilitate comparison across settings.

Table 2.

Author(s)	Year	Country	Design	Sample (n)	Primary TEM Type	Key Findings
Courtright et al.	1994	Malawi	Hospital cohort	583	Plant extracts, animal products	33.8% TEM use; TEM users waited 51 vs 13 days; 16% had VA <3/60 vs 5% of non-users.
Yorston & Foster	1994	Tanzania	Hospital descriptive	Not stated	Plant-based preparations	One third of corneal ulceration patients had used TEM before hospital attendance.
Lewallen & Courtright	1995	Malawi	Case- control	39 cases	Plant extracts; unspecified mixtures	Significant association between TEM use and peripheral corneal ulceration; secondary infection confirmed microbiologically in a substantial proportion of cases.
Bisika et al.	2009	Malawi	Population- based survey	Not stated	Herbal preparations, oils	Self-treatment, including TEM common; many patients did not proceed to formal care.
Eze et al.	2009	Nigeria	Cross-sectional hospital	2,542	Plant leaf extracts, concoctions	5.9% TEM prevalence; significantly associated with delayed presentation and rural residence.
Ukponmwan & Momoh	2010	Nigeria	Hospital descriptive	Not stated	Herbal preparations, unidentified	Complications: corneal opacity, ulcers, staphyloma, panophthalmitis, phthisis bulbi.
Nwosu & Obidiozor	2011	Nigeria	Cross- sectional hospital	500	Plant leaf extracts, unknown concoctions	13.2% TEM prevalence; plant extracts most common.
Achigbu & Achigbu	2014	Nigeria	Cross- sectional hospital	Not stated	Plant-based preparations	82.3% TEM use; most used TEM before presenting; predominantly plant-based.
Arunga et al.	2019	Uganda	Mixed- methods cohort	313	Plant products, chemical substances	60% (188/313) of MK patients used TEM; poorer VA and worse 3-month outcomes in TEM users.
Eticha et al.	2023	Ethiopia	Cross- sectional hospital	502	Herbal preparations, breast milk	22.3% TEM prevalence; positive family history strongest predictor (AOR=11.1, 95% CI 4.83-25.56).
Msonge et al.	2023	Uganda	Cross- sectional hospital	182	Plant products (47.8%), chemical (57.7%)	70% of TEM users had ocular complications; conjunctivitis most common (53.9%).
Sherief, Sitotaw & Girma	2024	Ethiopia (Rural)	Cross- sectional	814	Herbal preparations, honey, breast milk	59.8% TEM use; mainly for redness, irritation, discharge.

Table 2 footnotes: VA = visual acuity; TEM = traditional eye medicine; AOR = adjusted odds ratio; CI = confidence interval; MK = microbial keratitis.

Quality Assessment of Included Studies

Table 3 presents the methodological quality assessment results. Cohort and case-control studies were assessed using the NOS (maximum 9 stars: selection 0-4, comparability 0-2, outcome 0-3). Cross-sectional studies were assessed on four criteria: clear objectives, adequate sample size, appropriate outcome measurement, and control for confounding. Studies scoring 7-9 on NOS, or meeting 3-4 cross-sectional criteria, were classified as high quality; 5-6 / 2-3 criteria as moderate; and below 5 / fewer than 2 criteria as low-moderate.

Overall, four studies were rated high quality (Lewallen & Courtright 1995; Eze et al. 2009; Arunga et al. 2019; Eticha et al. 2023), six were of moderate quality, and two were low-moderate. No studies were rated low quality. The main limitations identified across studies were: reliance on self-reported TEM use, absence of comparison groups in descriptive studies, undefined sample sizes in several early hospital series, and hospital-based recruitment limiting generalisability to wider populations.

Prevalence of Traditional Eye Medicine Use

Reported prevalence varied widely across studies and settings (Table 2). The lowest rates were observed at the University of Nigeria Teaching Hospital, Enugu, where Eze and colleagues enrolled 2,542 newly presenting patients and found 5.9% overall TEM use, with rates rising among patients with anterior segment disease and those from rural areas. ⁸ Nwosu and Obidiozor found 13.2% of 500 consecutive new ophthalmic patients at the Guinness Eye Center, Onitsha, had used TEMs, with the most common substances being plant leaf extracts and concoctions of unknown origin.⁹

Higher rates were documented in other settings. In Ethiopia, Eticha and colleagues found 22.3% prevalence at the University of Gondar Comprehensive Specialized Hospital among 502 adult ophthalmic patients, ¹⁰ while the rural Gurage Zone study by Sherief and colleagues recorded 59.8% prevalence in 814 participants from primary health centres.¹¹ In Uganda, Arunga and colleagues found that 60% of patients presenting with microbial keratitis had used TEM, ¹³ and the Msonge cross-sectional study at Mulago National Referral Hospital found that 70% of 182 TEM users had complications on presentation.²⁶ The highest reported prevalence, 82.3%, was documented by Achigbu and Achigbu among ophthalmic outpatients in south-east Nigeria.¹²

The foundational Malawi study by Courtright and colleagues, in 583 patients with corneal ulcers presenting at two district hospitals, found a 33.8% prevalence of TEM use.⁷ Bisika and colleagues conducted a population-based survey in two Malawian districts and found that self-treatment, including TEM, was common, and that many patients did not progress beyond self-treatment to access any formal care.²⁷

Types of Traditional Eye Remedies Reported

Plant-based preparations were the most frequently reported form of TEM across all settings. Extracts from leaves, roots, and bark were most commonly described, often mixed with water or other agents and instilled directly into the eye. In the Ugandan Mulago study, plant products accounted for 47.8% of TEM types used. ²⁶ In Nigeria, liquids from plant leaves and roots were the predominant substances reported.⁹

Animal-derived products, including animal fats and biological materials, were also documented. Breast milk was reported in multiple settings, particularly for conjunctivitis in infants and young children, despite evidence that it may introduce bacteria into the eye. Honey was documented in Ethiopian studies, alongside human saliva, linseed preparations, and unidentified concoctions prepared by traditional healers. Sherief and colleagues identified herbal preparations, honey, and breast milk as the principal substances used in the rural Ethiopian setting.¹¹

Chemical substances were reported in the Ugandan Mulago study, accounting for 57.7% of TEM types used in that cohort, alongside plant products (37.7%) and animal products (4.7%).²⁶ TEMs were described as being sourced from non-traditional practitioners in 66.4% of cases, highlighting that TEM use is not limited to formal traditional healers but occurs widely through family members and community sources.

Ocular Conditions for Which TEMs Were Used

Traditional eye remedies were applied for a broad spectrum of symptoms and conditions. Commonly reported indications included eye redness, itching, discharge, pain, blurred vision, and ocular trauma. In the Msonge Ugandan study, TEMs were used on account of vision loss (58.5%), ocular itching (25.4%), and eye discharge (3.8%).²⁶ TEMs were also applied in attempts to treat corneal disease, cataract, and chronic ocular conditions.

The application of traditional remedies to eyes with active corneal infection or ulceration is of particular concern, as corneal infections may progress rapidly without appropriate antimicrobial treatment. ⁵ Across studies, TEM users consistently presented later than non-users: in the Malawi corneal ulcer study, TEM users waited an average of 51 days before hospital presentation versus 13 days for non-users.⁷

Ocular Outcomes Associated with TEM Use

Across the included studies, TEM use was associated with a range of adverse ocular outcomes. The most frequently documented findings were corneal ulceration and corneal opacity. In the Msonge Ugandan study, 70% of TEM users had ocular complications at presentation, with conjunctivitis being most common, 53.9%.²⁶ The prospective cohort study by Arunga and colleagues found that TEM users with microbial keratitis had poorer presenting visual acuity and worse outcomes at three months, with TEM use linked to significantly more frequent hypopyon at presentation and a trend toward more central and dense corneal scarring at three months.¹³

Ukponmwan and Momoh reported a range of severe complications in their Nigerian patients, including corneal opacities, ulcers, staphyloma, panophthalmitis, and phthisis bulbi.¹⁶ The Malawi study by Courtright and colleagues found that 16% of TEM users had visual acuity below 3/60 at presentation compared with 5% of non-users.⁷ Lewallen and Courtright, in their case-control study, found a statistically significant association between TEM use and corneal ulceration; microbiological investigations confirmed secondary bacterial infection in a substantial proportion of TEM-associated ulcers, with organisms including Neisseria gonococcus identified.²⁸

Severe intraocular infections including endophthalmitis and panophthalmitis were documented in multiple studies and may require enucleation in the most severe cases. The evidence from across these studies collectively supports concern that TEM use is associated with increased risk of irreversible visual impairment, though the observational nature of most studies means direct causation cannot be established from the available data alone.

Factors Associated with TEM Use

Several consistent sociodemographic and contextual factors were associated with TEM use across studies. Rural residence, low socioeconomic status, limited formal education, and lack of access to eye care services were repeatedly identified as predictors.^9,10,13,29 In the Eze Nigerian study, TEM use was significantly associated with younger age (p<0.01), being married (p<0.01), rural residence (p<0.01), and delayed presentation (p<0.01), but not with gender or educational status. ⁸ In the Ethiopian Gondar study, positive family history of TEM use was the strongest independent predictor, AOR 11.1 (95% CI 4.83-25.56), suggesting that familial transmission of health practices is a central driver.¹⁰ Arunga and colleagues found through qualitative interview data that acute pain was a major motivator: patients used any available remedy to relieve pain quickly, and most did not perceive TEM as dangerous.¹³

Discussion

This systematic review consolidates evidence from twelve studies conducted across five Sub-Saharan African countries and finds that traditional eye medicine (TEM) remains widely used and is consistently associated with adverse ocular outcomes. Across all settings reviewed, TEM use was linked to delayed hospital presentation, higher disease severity at admission, and a greater burden of corneal and intraocular complications. These findings carry significant clinical and public health relevance given the continued high burden of preventable corneal blindness in Sub-Saharan Africa. Because most included studies were observational, findings should be interpreted as demonstrating association rather than confirmed causation.

A consistent finding is the association between TEM use and delayed presentation to formal eye care. Courtright and colleagues found that TEM users with corneal ulcers in Malawi waited an average of 51 days between symptom onset and hospital presentation, compared with 13 days for patients who did not use TEMs. ⁷ In the Arunga prospective cohort study in Uganda, this delay was linked to worse clinical outcomes: TEM users with microbial keratitis had poorer presenting visual acuity and worse three-month outcomes, with more frequent hypopyon and a trend toward denser central corneal scarring. ¹³ This pattern is consistent with the broader literature on corneal infection, which shows that advanced disease at presentation may limit the effectiveness of antimicrobial therapy.

The prevalence of TEM use varied substantially across studies and settings, from 5.9% in a teaching hospital study in Enugu, Nigeria,⁸ to 59.8% in a rural Ethiopian primary care population ,¹¹ and up to 82.3% among ophthalmic patients in south-east Nigeria.¹² This range likely reflects genuine variation driven by differences in health system access, geography, socioeconomic conditions, and local cultural norms. It also reflects methodological differences: some studies enrolled all new ophthalmic attendees, while others enrolled only patients with specific conditions such as corneal ulcers or microbial keratitis, enriching the sample for severe disease. Despite this variation, every study reviewed found meaningful levels of TEM use.

The types of TEMs reported were largely consistent across settings: plant-derived preparations were predominant in virtually all studies. Animal-derived products, breast milk, honey, and chemically prepared substances were also commonly reported. ^{11, 26} The Msonge Ugandan study found that chemical substances accounted for 57.7% of TEMs used, extending concern beyond traditional herbal preparations to include self-medication with unregulated commercial products.²⁶

The ocular outcomes documented across studies were severe and, in many cases, irreversible. Corneal opacity, corneal ulceration, panophthalmitis, and phthisis bulbi were among the most commonly reported findings.^{16, 26} The Lewallen and Courtright case-control study found a statistically significant association between TEM use and peripheral corneal ulceration, with microbiological investigations confirming secondary bacterial infection in a substantial proportion of TEM-associated ulcers, one of the stronger pieces of evidence supporting a harmful mechanism, though case-control studies remain subject to selection bias.²⁸ Taken together, the evidence supports concern that TEM use is associated with increased risk of irreversible visual impairment, particularly in patients with pre-existing corneal disease.

The cultural context of TEM use must be central to any public health response. Qualitative data from the Arunga Uganda cohort study found that acute pain was a powerful driver of TEM use, and most patients did not perceive TEMs as dangerous.¹³ The WHO has recognised the importance of traditional medicine in health systems globally and has called for better integration, regulation, and evidence-based evaluation ^6,30. Earlier evidence from Malawi showed that structured training programmes engaging traditional healers, focusing on recognising and referring serious eye conditions, were associated with a reduction in TEM-associated corneal disease and improved referral behaviour.¹⁸ Lewallen and Courtright argued that traditional healers can serve as community-level partners in prevention.¹⁹ Public health strategies should include community education on the risks of unregulated eye treatments and collaboration with traditional healers to promote timely referral to ophthalmic services.

From a policy perspective, the findings support three broad priorities. First, community education campaigns should communicate the risks of placing non-sterile or unknown substances in the eye and the importance of prompt care for eye pain, trauma, and vision loss. Second, access to primary and secondary ophthalmic care must be expanded, as distance, cost, and limited service availability are major drivers of TEM use. ^{10, 13, 26} Third, collaborative engagement with traditional healers, as demonstrated in Malawi, offers a pragmatic and culturally acceptable pathway to reducing delay and harm.

Clinically, the findings support a recommendation that all eye care workers in Sub-Saharan Africa routinely enquire about prior TEM use when assessing patients with corneal disease, severe conjunctivitis, ocular trauma, or unexplained inflammation. A history of TEM use may explain delayed presentation, unusual ocular surface damage, or secondary infection, and should inform counselling and management.

Limitations

Several limitations should be considered when interpreting these findings. First, ten of twelve included studies were hospital-based, which introduces selection bias by over-representing patients with severe or sight-threatening disease. Patients with mild TEM-associated complications may never present to formal eye care, meaning that true population-level complication rates are likely underestimated. Only one population-based survey (Bisika et al., Malawi) and one rural primary care study (Sherief et al., Ethiopia) were included, limiting the community-level evidence base. Second, most studies relied on patient self-reports to determine prior TEM use, introducing potential recall bias and under-reporting, particularly where patients anticipated clinician disapproval. Third, heterogeneity in study design, case definitions, and outcome measures limited the ability to perform quantitative meta-analysis. Fourth, many studies did not characterise the specific composition of TEMs used, limiting mechanistic understanding of which substances carry the greatest risk. Fifth, several early studies did not report formal sample sizes, limiting the precision of prevalence estimates. Sixth, this review is subject to publication bias: studies with positive or dramatic findings are more likely to be published and indexed, meaning that the true prevalence of TEM use and its associated complications may be either over- or under-estimated in the available literature. Seventh, the review included only studies published in English; research published in French, Portuguese, or local languages— particularly relevant in Francophone and Lusophone Sub-Saharan Africa—was not captured, representing a meaningful source of language bias that limits the geographic generalisability of findings.

Conclusion

This systematic review demonstrates that traditional eye medicine remains widely used across Sub-Saharan Africa and is consistently associated with significant ocular morbidity. Across the included studies, TEM use was linked to delayed presentation to formal eye care, increased disease severity at admission, and a higher burden of corneal and intraocular complications.^7,13 Reported outcomes include corneal ulceration, corneal opacity, endophthalmitis, panophthalmitis, and irreversible vision loss.¹⁶

The persistence of TEM use reflects broader structural and cultural factors, including limited access to ophthalmic services, financial constraints, and strong community reliance on traditional healers.^6,14 Addressing this issue requires a coordinated approach that combines community education on the risks of unregulated eye treatments, improved access to eye care services, and collaborative engagement with traditional healers to support early referral and safer practices.^18-19

Future research should focus on better characterisation of the specific substances used in TEM, understanding the mechanisms of ocular injury, and developing culturally appropriate interventions that reduce harmful practices while respecting local beliefs and health-seeking behaviours.

Recommendations

• 1.Strengthen community eye health education with targeted messages on the risks of placing non-sterile substances in the eye and the importance of prompt medical care for eye pain, trauma, and vision loss.
• 2.Expand access to primary and secondary ophthalmic care through mobile eye clinics, community-based programmes, and strengthened referral networks, particularly in rural areas.
• 3.Engage traditional healers through structured training programmes that focus on recognising sight-threatening conditions and facilitating timely referral to formal ophthalmic services.
• 4.Promote routine clinical enquiry about prior TEM use in all patients presenting with corneal disease, ocular trauma, or unexplained ocular inflammation.
• 5.Develop national regulatory frameworks that monitor and address TEM practices posing specific risks to ocular health, in collaboration with national ophthalmology societies and public health authorities.
• 6.Support further research into the chemical and biological composition of traditional eye medicine preparations, their mechanisms of ocular harm, and the effectiveness of community-level interventions.

Supplementary Appendix A: Full Database Search Strings

This appendix provides the full reproducible search strings used across all five databases and one supplementary source. Searches were conducted on 14 March 2026 (PubMed, Scopus, Web of Science, Cochrane Library, African Journals Online) and 15 March 2026 (Google Scholar). No date limits were applied. No language or species filters were applied at the search stage; non-human studies and non-English publications were excluded at the screening stage. Boolean operators (AND, OR) were applied consistently; search strings were adapted to each database’s syntax and controlled vocabulary as described below.

A.1 PubMed (n = 298 records retrieved)

MeSH terms and free-text keywords were combined. The search was run via PubMed Advanced Search. Search date: 14 March 2026.

(“Eye, Traditional Medicine”[MeSH] OR “Medicine, Traditional”[MeSH] OR “traditional eye medicine”[tiab] OR “traditional eye remedies”[tiab] OR “herbal eye treatment”[tiab] OR “indigenous eye medicine”[tiab] OR “traditional eye medication”[tiab] OR “traditional eye drops”[tiab] OR “traditional healer”[tiab] AND “eye”[tiab])

AND

(“Corneal Ulcer”[MeSH] OR “Keratitis”[MeSH] OR “Eye Diseases”[MeSH] OR “Vision Disorders”[MeSH] OR “Blindness”[MeSH] OR “ocular complications”[tiab] OR “corneal ulcer”[tiab] OR “corneal disease”[tiab] OR “visual impairment”[tiab] OR “eye disease”[tiab] OR “microbial keratitis”[tiab])

AND

(“Africa South of the Sahara”[MeSH] OR “Africa”[tiab] OR “Sub-Saharan Africa”[tiab] OR “Nigeria”[tiab] OR “Uganda”[tiab] OR “Malawi”[tiab] OR “Tanzania”[tiab] OR “Ethiopia”[tiab] OR “Ghana”[tiab])

A.2 Scopus (n = 187 records retrieved)

Title/abstract/keyword fields searched (TITLE-ABS-KEY). No controlled vocabulary equivalent to MeSH is available in Scopus; free-text terms were used throughout. Search date: 14 March 2026.

TITLE-ABS-KEY ( “traditional eye medicine” OR “traditional eye remedies” OR “herbal eye treatment” OR “indigenous eye medicine” OR “traditional eye medication” )

AND

TITLE-ABS-KEY ( “ocular complications” OR “corneal ulcer” OR “corneal disease” OR “visual impairment” OR “eye disease” OR “microbial keratitis” )

AND

TITLE-ABS-KEY ( “Africa” OR “Sub-Saharan Africa” OR “Nigeria” OR “Uganda” OR “Malawi” OR “Tanzania” OR “Ethiopia” OR “Ghana” )

A.3 Web of Science (n = 134 records retrieved)

All databases searched via Web of Science Core Collection. Topic field (TS=) searches title, abstract, author keywords, and Keywords Plus. Search date: 14 March 2026.

TS=( “traditional eye medicine” OR “traditional eye remedies” OR “herbal eye treatment” OR “indigenous eye medicine” OR “traditional eye medication” )

AND

TS=( “ocular complications” OR “corneal ulcer” OR “corneal disease” OR “visual impairment” OR “eye disease” OR “microbial keratitis” )

AND

TS=( “Africa” OR “Sub-Saharan Africa” OR “Nigeria” OR “Uganda” OR “Malawi” OR “Tanzania” OR “Ethiopia” OR “Ghana” )

A.4 Cochrane Library – CENTRAL (n = 39 records retrieved)

Search restricted to the Cochrane Central Register of Controlled Trials (CENTRAL). Title, abstract, and keyword fields searched. Search date: 14 March 2026.

(“traditional eye medicine” OR “traditional eye remedies” OR “herbal eye treatment” OR “indigenous eye medicine” OR “traditional eye medication”):ti,ab,kw

AND

(“ocular complications” OR “corneal ulcer” OR “corneal disease” OR “visual impairment” OR “eye disease” OR “microbial keratitis”):ti,ab,kw

AND

(“Africa” OR “Sub-Saharan Africa” OR “Nigeria” OR “Uganda” OR “Malawi” OR “Tanzania” OR “Ethiopia” OR “Ghana”):ti,ab,kw

A.5 African Journals Online (AJOL) (n = 91 records retrieved)

AJOL was searched via its full-text and abstract search interface. AJOL does not support controlled vocabulary; free-text terms were used. The Africa geographic filter was not applied separately as AJOL indexes only African journals. Search date: 14 March 2026.

(“traditional eye medicine” OR “traditional eye remedies” OR “herbal eye treatment” OR “indigenous eye medicine” OR “traditional eye medication”)

AND

(“ocular complications” OR “corneal ulcer” OR “corneal disease” OR “visual impairment” OR “eye disease” OR “microbial keratitis”)

A.6 Google Scholar – Supplementary Grey Literature (n = 98 records retrieved)

Google Scholar does not support Boolean operators, field-specific tags, or controlled vocabulary in the same manner as bibliographic databases. Searches were conducted using natural-language phrase searching and quotation marks for exact phrases. The first 10 pages of results (approximately 100 records) were reviewed for each search, as is standard practice for Google Scholar supplementary searches. Search date: 15 March 2026.

Search 1: “traditional eye medicine” Africa ocular complications

Search 2: “traditional eye remedies” “corneal ulcer” Sub-Saharan Africa

Search 3: “herbal eye treatment” Africa visual impairment

Search 4: “traditional eye medicine” Nigeria OR Uganda OR Malawi OR Ethiopia OR Tanzania

A.7 Notes on Search Adaptation and Filters

The same three conceptual blocks (TEM terms; ocular outcome terms; African country/region terms) were applied across all databases. Database-specific adaptations were made to field tags only (e.g., [tiab] in PubMed, TITLE-ABS-KEY in Scopus, TS= in Web of Science, :ti,ab,kw in Cochrane). No date limits were applied to any database. No language filters were applied at the search stage. No species (human/animal) filter was applied; animal studies were excluded during title and abstract screening. Duplicate records identified across databases were removed prior to screening using a reference management system. MeSH terms were used only in PubMed, where they were combined with free-text terms using OR to maximise sensitivity. The core keyword combinations as reported in the Methods section of this review were used without modification across all databases, with syntax adapted as described above.