TOPIC: VISUAL IMPAIRMENT IN GHANA

Evaluation of study conducted to ascertain the epidemiology and clinical presentation of

Glaucoma in Ghana with a proposal for public health intervention.

INTRODUCTION

Glaucoma is a cluster of conditions that destroy the optic nerve which plays an essential role

in determining the vision of a person. In most cases, optic nerve damage is caused by

abnormally high pressure within the eyeball (American Academy of Ophthalmology, 2020).

According to the World Health Organization (WHO) (2004), glaucoma has emerged as the

second leading cause of blindness after cataract on a global scale. The portentous aspect of

this development is such that person of all age groups are affected and visual impairment or

blindness caused by glaucoma is irreversible. Again, persons with the common subtype of

glaucoma, such as open-angle glaucoma may present with no symptoms until being picked up

on routine clinical examination (Glaucoma Research Foundation, 2017). Ghana remains the

Sub-Sahara African country with the highest prevalence of glaucoma. At least 8.5 per cent of

the population above 40years and 7.7 per cent of persons above 30years carry the disease.

(Otabil, Tenkorang, Ankrah, and Otabil,2013). The country also maintains the second spot

for the highest prevalence of Glaucoma worldwide (Nelson-Ayifah, Mashige, 2020). In their

retrospective case series, Gyasi et al (2013) identified 24% of subjects reviewed were blind

due to glaucoma in at least one eye within the capital city of Ghana. Given the widespread

nature of glaucoma contributing to the growing incidence of visual impairment and or

blindness across the entire country, more studies are required to assist the development of

public health promotion intervention in addressing the issue.

Kyei et al (2021) seek to outline the epidemiology and clinical presentation of glaucoma in a

referral facility in Ghana to elicit lessons for public health intervention. This is to present

1
Word count: 1897
current and relevant data to shape the discourse about the menace of Glaucoma in Ghana.

Their cross-sectional study was a hospital based-retrospective review of clinical records of

glaucoma patients spanning ten years between 2010 and 2019. A sample population of

19,000 patient charts from one tertiary level eye centre archives were used for random

nonprobability-based sampling. With the inclusion criteria of visual acuity, intraocular

pressure, cup-to-disc ratios, and the average retinal nerve fibres thickness, 3.5 per cent

representing 660 records were matched. Even though some juvenile cases were identified, the

median age group were those of adult-onset. Socioeconomic inequalities are key determinants

to the diagnosis and treatment of glaucoma. Therefore, advocacy to intensify public health

promotion to shape policymaking in line with the identification and management of cases of

glaucoma was warranted (Kyei et al, 2021)

DISCUSSION

The following are the critical evaluation of the study conducted by Kyei et al (2020) as

enumerated above.

Choice of study design

The researchers opted for a cross-sectional study to review the epidemiology and clinical

management of glaucoma through a clinic-based retrospective review of patients’ charts.

Cross-sectional studies are essential in the establishment of the magnitude of disease and

associated risk in a defined population. Its usefulness is most essential in the study of chronic

diseases with high prevalence and with low incidence making the selection of a cohort

difficult (Zaccai, 2004). The study design is, therefore, fit for evaluating glaucoma, a known

chronic disease with high prevalence in Ghana.

Choice of the study population

2
Word count: 1897
The study population was selected among nineteen thousand patients charts among which

660 were diagnosed with glaucoma from a tertiary eye centre in Ghana. Notably, all the

selected study cases presented with glaucoma in both eyes. Based on the source of the study

population, selection bias is most likely to occur. In this study, the researchers selected an eye

centre that has known cases of glaucoma leaving little room for randomization (Institute for

Work and Health, 2014). This effected may be minimized by the selection of many patient

charts in excesses of 19,000. The inclusion criteria were based on cases with significant optic

nerve changes in the presence of increased intraocular pressure such that subclinical cases

were ignored. Since the study was done to review retrospective data, no attrition was

observed among study participants.

Exposure and outcome measures

The choice of data collection was specific for the study rather than being routine. Set-out

variables including socio-demographics of patients, visual acuity, increased intraocular

pressure, cup-to-disc ratios, cup volume, cup-to-disc area, among others were used to extract

information from patients’ chart. The validity of the data extracted may not be compromised

since well-established criteria for the case definition of glaucoma were used (Kroese and

Burton, 2003). Per established definition, enough information was collected from the charts

of study participants. The period used to follow each study participant could not be

elaborated by the researchers in the space where exposure and outcome were analyzed

concurrently. By employing standard tools of selection devoid of participant’s recollection,

no recall bias could occur in this study.

The study observes a 3.5% prevalence of glaucoma with a confidence interval of 95% among

all cases that had reported to the tertiary eye care centre. This twice as few as the prevalence

reported by Otabil et al (2013) among a similar group of the same Ghanaian population. The

study identifies most glaucoma patients related to significant religious groups in Ghana.

3
Word count: 1897
Primary open-angle glaucoma tops the causes of glaucoma as compared to normal-tension

glaucoma according to the study. This is in support of Ntim-Amponsah et al (2004) who

identified the same in their cross-sectional study of glaucoma prevalence in Ghana. The mean

age for glaucoma as established by the study was 47.3years as visual impairment is

associated with “productivity, quality of life, and family cohesion” (Kyei et al, 2021). A male

to female prevalence ratio of 3:1 for glaucoma was established by the study. This ratio is

higher relative to the finding from Budenz, Barton and Whiteside-de Vos (2013), where a

male to female ratio of 1.5:1 was elicited in a similar cross-sectional study. Socioeconomic

factors could play a significant role in the skew of the prevalence away from the female

group. Majority of study participants had significant loss of peripheral vision based on visual

field assessment indicator. The study relates to the prevalence of loss of peripheral vision as a

significant risk factor in road traffic accidents attributable to commercial drivers (Boadi et al,

2016, cited in Kyei et al, 2021). The management of established glaucoma is either by way of

medication or surgery. It was identified that a few cases of glaucoma get access to either

medication or surgical treatment. In most cases, surgical treatment is farfetched.

IMPLICATION

A cross-sectional study design is limited by cost and the need to gather a large amount of

historical data from all individuals surveyed. It was also difficult to establish a cause-and-

effect relationship in such a study since both exposure and outcome measured simultaneously

(Solem, 2015). In the modern era of digitization, a similar study could be conducted with far

less expensive resources and human capital where is readily available for analysis using

electronic medical records systems (Hoover, 2016).

4
Word count: 1897
Again, selection Bias is most a likely occurrence in cross-sectional studies. According to

Heckman J.J., (2005), selection bias occurs when the study population does not reflect a “true

representation of the target population to which conclusion is being extended to”. In this

study recruitment of participants were specific to patients who had eye care at established eye

centre not necessary a sampling from the general population. To correct this bias, large

sample size is required. The selection of 19,000 case chart may not be a true reflection of

adequate numbers required to offset the effect of this bias in the determination of the

outcome.

The study elicited a lower female to male prevalence ratio concerning the low socioeconomic

status of the female sex. Thus, due to their lower socioeconomic status, few females may be

capable of assessing a costly eye care service that could reveal glaucoma. Upon an

established direct relationship between female sex and prevalence of glaucoma, a

confounding variable of socioeconomic status emerges. Thomas (2020) defines a

confounding variable as “an unmeasured third variable that influences both the supposed

cause and the supposed effect.” The occurrence of this confounder could be eliminated by

restriction of the study group to a specific age range instead of a broad-based age group.

Bhandari (2020) defines external validity as the extent to which the findings of a study could

be generalized to other “situations, people, settings, and measures”. Even though cases

selected for the study came from a section of the population, it well established that majority

of glaucoma cases are asymptomatic and such patients would live in the general population

without reporting (Gyasi et al, 2014). Thus, sampling of data among subject who report at the

hospital via a non-probability convenience method may not be a true reflection of the

prevalence in the entire population. To correct the defect in external validity, the sample size

should include people with different characteristics (Bhandari, 2020)

5
Word count: 1897
CONCLUSION

A glaucoma is a group of conditions that result in the damage to the optic nerve which is a

major determinant of vision. The WHO identifies glaucoma as the second leading cause of

blindness globally as the spectrum of disease progression is generally irreversible. Ghana

tops the prevalence of glaucoma among all Sub-Saharan countries with prevalence more than

8.5 per cent among the population aged above 40years and 7.7 per cent among those above

30years. A cross-sectional study by Kyei et al (2021) seeks to outline the epidemiology and

clinical presentation of glaucoma in a referral facility in Ghana. The study aims to shape

public health discourse and policymaking in line with glaucoma. While adopting well case

definition for glaucoma, inclusion criteria were set out for 19,000 case charts selected of

patients among which 660 matched. The study established the prevalence of glaucoma to be

3.5 per cent. This finding contrasts with recent studies in Ghana that identified a higher

prevalence. The choice of a large case chart and expanding period of the clinic retrospective

study made little room selection bias. No attrition would be elicited since the study was a

review of retrospective data. With a specific choice of data collection, well-defined variables

were used to extract potential cases for analysis. No recall bias could be elicited since

participants were not directly engaged in the collection of data. The study also shows a

relatively higher male to female prevalence ratio of 3:1 as compared to similar studies that

identified a ratio of 1.5:1. A socioeconomic factor was noted to play a major role in the

determination of health-seeking behaviour, especially among the female sex.

By implication, the inclusion of electronic medical records systems could aid the collection of

data to reduce cost. Selection bias is a well-noted challenge to cross-sectional studies; thus,

the large sample size is required to offset its effect on the analysis. The study could have been

restricted to a specific age range to limit the occurrence of confounding variables among the

female sex. In effect, the study has a limited chance for external validity since the case

6
Word count: 1897
encompasses only subjects who opted for care at an established healthcare institution. It is

recommended that people with different characteristics are included at the case sampling

stage to enhance the study’s external validity.

7
Word count: 1897
REFERENCES

Bhandari P., (2020). Understanding external validity. Available at

https://www.scribbr.com/methodology/external-validity/ (Accessed on 21st January 2021)

Budenz D.L., Barton K, Whiteside-de Vos J, (2013). Prevalence of Glaucoma in an Urban

West African Population: The Tema Eye Survey. JAMA Ophthalmology.;131(5):651–658.

doi:10.1001/jamaophthalmol.2013.1686

Glaucoma Research Foundation (2017) Glaucoma Facts and Stats. Available at

https://www.glaucoma.org/glaucoma/glaucoma-facts-and-stats.php (Accessed on 20th

January 2021)

Gyasi, M. E., Francis, A. W., Chen, Y., Harrison, R. S., & Kodjo, A. R. (2014). Presentation

of glaucoma in the greater Accra metropolitan area of Ghana. Ghana medical journal, 48(3),

143–147. https://doi.org/10.4314/gmj.v48i3.4

Hoover R., (2016) Benefits of using an electronic health record, Nursing:46(7) p21-22

doi: 10.1097/01.NURSE.0000484036.85939.06

Institute for work and health, (2014). Selection bias. At Work. Available at

https://www.iwh.on.ca/what-researchers-mean-by/selection-bias (Accessed on 20th January

2021)

8
Word count: 1897
Kroese M, Burton H., (2003). Primary open angle glaucoma. The need for a consensus case

definition. Journal of Epidemiology & Community Health vol 57 pp752-754.

Kyei S., Obeng P.A., Michael A.K. and Assiamah F., (2021). Epidemiology and clinical

presentation of glaucoma in a referral facility in Ghana: Any lessons for public health

intervention? PLoS One, 16(1)

Nelson-Ayifah D, Mashige K.P., (2020). ‘Demographic and clinical characteristics of

patients with glaucoma in a tertiary eye facility in Ghana’. African Vision Eye Health.;79(1).

https://doi.org/10.4102/aveh.v79i1.521

Ntim-Amponsah, C. T., Amoaku, W. M., Ofosu-Amaah, S., Ewusi, R. K., Idirisuriya-Khair,

R., Nyatepe-Coo, E., & Adu-Darko, M. (2004). Prevalence of glaucoma in an African

population. Eye (London, England), 18(5), 491–497. https://doi.org/10.1038/sj.eye.6700674.

Otabil B.K, Tenkorang B.S, Ankrah L. M., Otabil E.A (2013). Prevalence of glaucoma in an

eye clinic in Ghana. Russian Open Medical Journal, 2(0310).

Solem R.C., (2015). Limitation of a cross-sectional study. American Journal of Orthodontics

and Dentofacial Orthopedics. ScienceDirect: 148(2) pp 205. DOI:

https://doi.org/10.1016/j.ajodo.2015.05.006

Thomas R., (2020). Understanding confounding variables. Available at

https://www.scribbr.com/methodology/confounding-variables/ (Accessed on 20th January

2021)

9
Word count: 1897
What is glaucoma? American Academy of Ophthalmology. Available at

https://www.aao.org/eye-health/diseases/what-is-glaucoma. (Accessed 18th January, 2021)

World Health Organization (2004). Glaucoma is the second leading cause of blindness

globally. Available at https://www.who.int/bulletin/volumes/82/11/feature1104/en/ (Accessed

on 20th January 2021)

Zaccai J.H., (2004). How to assess epidemiological studies. Post Graduate Medical Journal,

80(1) pp 140-147 doi: 10.1136/pgmj.2003.012633

10
Word count: 1897