1 Introduction

Myopia is a common form of refractive error that has gained considerable public health attention across the globe in recent years. Pivotal to concerns about myopia is its attributable negative impact on vision and eye health [1], coupled with its increasing global prevalence [2].

Despite the lack of a clear pathophysiological mechanism for the onset and progression of myopia, environmental and genetic factors have been implicated by several studies [2, 3]. While the role of genetics is thought to be involved in the onset and progression of myopia,3 environmental factors such as the level of outdoor activities [4], level of education and greater amount of near work [5], reading distance [6], stature [7], urbanisation and socioeconomic status [8], low birth weight [9], lack of breastfeeding [10], and parental education [11] have been implicated in myopia. Furthermore, associations between myopia and ethnicity, age and sex have also been described [12].

Aside the obvious debilitating effect of myopia on the quality of vision, myopia may also present with axial elongation [13], changes in central corneal thickness [14], deeper anterior chamber [15], decreased choroidal circulation and decreased scleral thickness [16], and thinner retina and retinal detachment [17], which may lead to permanent visual impairment.

According to global estimates, myopia affects approximately 22.9% of the global population [18], while projections are that myopia and high myopia could affect approximately 49.8 and 9.8%, respectively, of the world population by 2050 [18]. According to reports, Eastern Asia is particularly noted for a relatively high prevalence of myopia: over 90% in South Korea, up to 84% in Taiwan and between 36.7 and 53.9% among children in China 69.5% in Singaporean children [19]; between 36.7 and 53.9% among children in China [20, 21]. In Europe and North America, the prevalence of myopia ranges from 6.2 to 26.2%, respectively [22]. Despite global reports about the high prevalence of myopia in different geographical regions, there is inadequate information on the regional prevalence of myopia in Africa. Recently, the prevalence of myopia in Africa was estimated to be between 4.7 and 6.2% among children [23, 24].

In Ghana, there are inadequate studies regarding the prevalence of myopia, which pegs the figures between 1.7 and 7% among school children [25, 26]. These prevalence studies are mostly confined to specific age groups and do not cut across the different stages of life to give a holistic situational report of myopia for planning and resource allocation. Due to the alarming spate of myopia in other geographical regions of the world especially after the COVID 19 pandemic, it is imperative to retrospectively examine the epidemiological profile of myopia in Ghana before the pandemic to serve as a reference for comparative analyses for myopia during and post-COVID era in future studies in the country. Ultimately, this will enhance clinical practice and inform future health policies. Therefore, this study aims to explore the epidemiology and demographic risk factors for myopia in Ghana using large-scale retrospective data gathered from a tertiary eye referral hospital that approximately represents the demographic distribution of Ghana.

2 Methods

2.1 Study setting and participants

This was a 5 year hospital-based retrospective cross-sectional study conducted at Dr. Agarwal Eye Hospital in Accra, Ghana, from 2015 to 2019. This was selected to exclude the peak period of COVID-19 that affected attendance at this eye care facility. It is one of the major eye hospitals in Ghana that receives referral cases nationwide, with an annual average patient inflow of over 4000. Their services include medical and surgical eye services and refractive services, including refraction and optical correction with contact lenses and spectacle lenses, except laser refractive surgery. Therefore, all the post-laser refractive patients had already had their refractive surgeries elsewhere as captured in their ocular histories. The facility is located in Accra, in the Greater Accra Region of Ghana, which is the region with the highest population [27]. Although Accra serves as the home city of the Ga-Adangbe ethnic group, it is also highly cosmopolitan, with sizeable proportions of all the other ethnic groups in Ghana residing in the city. Thus, its population diversity makes it roughly representative of the Ghanaian population. Ghana had ten administrative regions for the period under consideration, namely, Greater Accra, Volta, Western, Central, Upper East, Upper West, Northern, Brong Ahafo, Eastern and Ashanti. These regions were originally inhabited by one or more of the five main ethnic groups, including Akan, Ewe, Ga-Adangbe, Mole-Dagbani, and Guans, with Akan being the largest ethnic group in Ghana.

2.2 Data collection procedure

Myopia cases were extracted from the records of patients who visited the facility from 2015 to 2019. Data from noncycloplegic objective refraction as obtained by an autorefractometer (Topcon RM-8800, Tokyo, Japan) during the first hospital attendance were recorded for each eye as spherical equivalent powers (SE) by obtaining the algebraic sum of half the cylinder power and the sphere power [(SE: sphere + 0.5(cylinder)] [25]. Myopia was defined as spherical equivalent power ≤ − 0.50 D, and high myopia was defined as a spherical equivalent power ≤ − 5.00 D in accordance with the most recent definitions [18]. Patients were classified as myopic if at least one eye was myopic by a spherical equivalent power of ≤ − 0.50 dioptres, and in the case of bilateral myopia, the maximum myopic spherical equivalent power of either eye was used to classify the patient as either having myopia or high myopia. Visual impairment for each eye before and after refraction was classified based on the entrance visual acuity (VA) into mild (VA between 6/12 and 6/18), moderate (VA worse than 6/18–6/60), severe (VA worse than 6/60–3/60), and blindness (VA worse than 3/60) [28]. Demographic data, including age, sex, ethnicity, area of residence, and occupation, were extracted. The International Standard Classification of Occupations as endorsed by the International Labour Organisation in 2008 (ISCO-08) was used to broadly classify similar occupations into broad categories [29]. However, an exception to the ISCO-08 was students and the unemployed, who were considered separate categories. We further grouped the residential area data into urban, suburban and rural settlements [30]. Data from patient folders with incomplete records, those who had their refraction elsewhere, and those outside the study period were excluded.

2.3 Statistical analysis

The data extracted were screened for missing data, processed and analysed using IBM SPSS Statistics version 25 (IBM, Armonk, NY USA). Statistical associations were determined using Chi-square and independent t tests. Multivariable regression analysis was performed to examine the demographic factors associated with myopia. A p value < 0.05 was considered statistically significant.

2.4 Ethical consideration

The study protocol was approved by the Institutional Review Board of the University of Cape Coast with clearance identification number UCCIRB/CHAS/2022/40 prior to obtaining permission from the management of Dr Agarwal’s Eye Hospital. Inform consent is waived by the Institutional Review Board of the University of Cape Coast since it involved a retrospective review of patient case notes only. However, the study ensured strict anonymity and confidentiality of the medical records of patients and adhered to the tenets of the Declaration of Helsinki.

3 Results

A total of 21016 patients visited the facility within the 5 year study period, of whom 15807 patients accessed refractive services and had complete data for analysis. Therefore, the accessibility rate for refractive services at the facility for the 5 year period was 75.2%. Out of these patients, 50.9% were males, and 49.1% were females. The overall mean ± SD age was 42.81 ± 18.99 years (range: 2–98 years), while the mean ages for the myopic and non-myopic groups were 40.5 ± 20.94 years and 44.7 ± 17.06 years, respectively, and the difference was significant (p < 0.001). The mean spherical equivalents were − 1.54 ± 1.31 D (range: − 0.50 D to − 4.63 D) and − 6.45 ± 3.91 D (range: − 5.25 D to − 25.00 D) for myopia and high myopia, respectively.

Table 1 presents the demographic distribution of the patients and their associations with myopia. Overall, myopia was present in 7021 patients, representing a prevalence of 44.4%. On further sub-classification, the prevalence of myopia (SE > − 5.00 D) and high myopia (SE ≤ − 5.00 D) were 37.2% and 7.2%, respectively. Myopia was associated with age group (p < 0.001), ethnicity (p < 0.001), and occupation (p < 0.001) but not sex (p = 0.291). From Table 1, the prevalence of myopia was highest among the 30–39 age group (18.1%) but lowest in the 0–9 age group (3.9%) (p < 0.001), which was higher among the Akans (45%) than among the rest of the ethnic groups (p < 0.001). Moreover, myopia was also more prevalent among urban dwellers (82.5%) than among suburban and rural dwellers (p < 0.001). Regarding occupations, myopia was highest among students (25.8%) but lowest among elementary occupations (p < 0.001).

Table 1 Demographic characteristics of the study participants
Full size table

The distribution of myopia according to the various age groups is presented in Table 2. It is shown that the proportions of both myopia and high myopia were still higher in the 30–39 age group followed by the 20–29 age group but least in the 0–9 age group. In terms of national distribution, 72.4% of the cases of myopia were residents of the Greater Accra region (Fig. 1).

Table 2 Distribution of myopia and high myopia according to various age groups
Full size table
Fig.1
figure 1

Geographic distribution of the myopic and non-myopic patients among the hospital-based sample. G. Accra Greater Accra region, ER Eastern region, WR Western region, CR Central region, UER Upper East region, BAR Brong Ahafo region, NR Northern region, AR Ashanti region, UWR Upper West region, VR Volta region

Full size image

Table 3 summarises the relative proportions of eyes, right and left, that had various levels of visual impairment from mild to blindness before and after refractive intervention. Visual impairment levels were classified based on the criteria suggested by the World Health Organization (mild—VA between 6/12 and 6/18, moderate—VA worse than 6/18–6/60, severe—VA worse than 6/60–3/60, and blindness—VA worse than 3/60) [28]. Before analysis, 3 prosthetic right eyes and 2 prosthetic left eyes were excluded. Following refractive intervention for the right eyes, there was an exceptionally high improvement rate for eyes with moderate visual impairment (an improvement in the right eyes of 1417 (98.6%) patients initially with VAs worse than 6/18–6/60). On the other hand, for the left eyes, those with severe visual impairment saw the highest improvement rate (an improvement in the left eyes of 272 (81.4%) patients initially with VAs of worse than 6/60–3/60) (Table 3). Spectacles were the major mode of refractive correction, accounting for 91.1% of the correction, while contact lens (0.5%), LASER refractive surgery (1.1%), both contact lens and spectacles (0.3%), and LASER combined with spectacles (0.3%) were very less frequent management options.

Table 3 Comparative proportions of visual impairment before and after refractive correction
Full size table

Demographic risk factors for myopia were analysed based on the associated factors in Table 1 by estimating age-adjusted odds ratios for all the risk factors (Table 4). Based on the results, occupation classifications, including professionals (OR, 2.93; p < 0.001), technicians and associated professionals (OR, 6.71; p < 0.001), service and shop and market sales workers (OR, 11.74; p < 0.001), agricultural and fishery workers (OR, 1.88; p < 0.001), craft workers (OR, 3.41; P < 0.001), armed forces (OR, 10.26; P < 0.001), students (OR, 3.49; P < 0.001), and unemployed workers (OR, 6.64; P < 0.001), were all significant risk factors for myopia. Additionally, the Ewe (OR, 1.48; p = 0.001) and Ga-Adangbe (OR, 1.28; p = 0.012) ethnic groups were more likely to be myopic than the other ethnic groups. (Table 4).

Table 4 Multivariable analysis of the demographic risk factors for myopia
Full size table

4 Discussion

The global trend of increasing prevalence of myopia [2] has necessitated several population and hospital-based studies in different countries to inform health policies. In the present hospital-based study, the overall prevalence of myopia was 44.4%, of which the Greater Accra region had the greatest proportion of cases. The location of the study setting skewed the distribution of myopia to residents of Accra due to the advantage of greater accessibility. This current estimate of myopia in Ghana is lower than that of a previous hospital-based study, which reported 54.1% [31]. Methodological limitations could account for these inconsistencies in the prevalence figures. For instance, our study facility was a private eye hospital and may not meet the financial capabilities of some section of the population, as opposed to the previous study [31], which included public facilities. Regardless, the larger sample size of this study still offers a comparatively stronger predictive power.

Aside the overall myopia, the WHO also recognises myopia and high myopia as distinct types of myopic refractive error due to the increased risk of sight-threatening retinal damage and other complications associated with the latter [18]. Myopia currently refers to the range of low to moderate degrees of myopia [18], which accounts for the majority of clinically reported cases. In the current study, the prevalence of myopia and high myopia were 37.2 and 7.2%, respectively, in contrast to a previous study in Ghana that pegged myopia and high myopia at 49.4 and 4.75%, respectively [3]. An increasing amount of near-related work via digital devices and computers in recent decades poses a significant risk for myopic development and progression [5]. A possible shift of moderate myopia into high myopia over the course of time may explain the variations in estimates. Notwithstanding, longitudinal studies are recommended to unravel the time dynamics of myopia in Ghana.

Overall, the prevalence of myopia and high myopia were lowest in the early years of life (0–9 years) but increased from 10 to 19 years, before peaking in prevalence among the 30–39 age group. Subsequently, a decline in prevalence was observed for only myopia, despite some fluctuations, from 40 years to over 70 years. However, in high myopia, a reverse trend was noticed with increasing prevalence from 60 to 69 years onwards. Previous epidemiological studies reported similar findings in Ghana [31] and elsewhere [32, 33]. In studies conducted in Ghana and Tunisia where non–cycloplegic refraction was used to determine refractive error refraction in children aged 5–18 years, the reported prevalence ranged from 1.7 [25] to 22.6% [34]. Owing to Africa’s increasing urbanization [35], more children and young adults in Africa are increasingly engaged in indoor and near work activities compared to earlier generations [36]. While children spend long hours doing school work or gaming with mobile devices [37], young adults increasingly engage in sedentary computer-related work activities [38], thus explaining the increase in myopia between 10 and 39 years. Moreover, emerging physiological evidence suggests that lens biometric changes, with a resultant decrease in the refractive index of the crystalline lens with ageing, may eliminate low degrees of myopia after age 40 years [39,40,41].

Reports of studies regarding sex-related differences in the prevalence of myopia have been contradictory. While some studies reported significant associations [42, 43], others found otherwise [44] or differences only in a specific age group [45]. In Ghana, Koomson et al. [31] reported a higher prevalence in males than in females, in contrast to other studies that found otherwise [42, 43]. We found no sex-dependent differences in the prevalence of myopia.

To the best of our knowledge, the present study is the first to examine the ethnicity-related prevalence of myopia in Ghana. From our study, myopia was significantly more prevalent among the Akan ethnic group, followed by the Ga-Adangbe ethnic group compared to the rest. This observation may be accounted for by the population of Akans and the study setting. Akans are the largest ethnic group in Ghana and are predominantly located in the middle and southern parts of the country. Thus, by virtue of their numbers, Akans would naturally have a higher hospital attendance rate for facilities located in southern Ghana compared to other ethnic groups. Nevertheless, ethnic differences in myopia prevalence have been reported by other studies in Asia [46]. Myopia was also more prevalent among urban settlers, followed by suburban dwellers, but it was least prevalent among rural dwellers. This difference could be attributed to the urban location of the study setting and its relatively reduced accessibility to suburban, and especially rural, settlers. Large national myopia studies using multi-stage random sampling should be conducted for further investigations of ethnic- and residential-dependent variations in myopia.

Available evidence suggests that myopia onset and progression are influenced by the amount of near-related and outdoor activities [4, 5]. Occupations may differ in terms of the amount near the visual task involved. In this study, students had a higher prevalence of myopia than others, while executives, senior officials and managers had the lowest prevalence. Unlike students who must read large volumes of text books to make academic progress, senior officials and executive workers may play more of a supervisory role at work. Additionally, these groups of workers are usually aged 40 and above, unlike students. Therefore, another reason for the observed difference could be the effect of age on the prevalence of myopia [41].

This study evaluated the effect of demographic risk factors on myopia. From the results, students and workers categorised under armed forces, craft-related work, agriculture, services, technicians, professionals, and the unemployed had higher odds of increased myopia prevalence. Occupational risk for myopia could be explained by the level of near and outdoor visual tasks involved [4, 5]. Although the unemployed group was not technically involved with any work, their higher risk for myopia could be due to their previous activities, such as schooling or a greater leisure time, which increased their chances of watching TV or playing games. Additionally, the Ewe and Ga-Adangbe ethnic groups of Ghana were also at risk for increased myopia—the first of such a study in Ghana. However, due to the limitation imposed by the study setting, further population-based studies are required to further examine ethnic risk factors for myopia in Ghana on a national scale.

On the other hand, urban and suburban settlements as well as ages 40–49 years, 50–59 years, and 60–69 years were all protective factors for myopia, similar to other reports on age [41] and urbanisation (the prevalence of myopia was significantly lower in those aged 40–69 years). Despite the significantly higher prevalence of myopia among urban dwellers in this study, urbanisation was not a risk factor for myopia, which corroborates another study [47]. Regarding age, other studies also reported that age 40 years and above may be protective against myopia [41].

In recent times, myopia control has garnered much popularity due to convincing evidence about the increasing prevalence of myopia [18], which is associated with an increased risk for permanent visual impairment [1]. These factors have triggered studies into the slowing of myopia progression via therapies including overnight orthokeratology, atropine, myopia control spectacles, dual defocus contact lenses, and multifocal contact lenses [48]. Despite the various myopia control options, this study shows that conventional spectacles remain the mainstay management option for myopia in Ghana. This is consistent with the recent report of a survey on eye care practitioners across Africa in which clinicians were less abreast of the current compelling evidence on myopia control [49]. Therefore, a roll-out of continuing professional development courses on the recent advances in myopia control strategies for optometrists in Ghana is highly recommended.

Furthermore, this study noted that refraction was very effective at reducing the burden of myopia-related visual impairment burden. For instance, between 30 and 98.6% of both the right and left eyes that were initially classified as having moderate to blind visual impairment improved after refraction. Given that an overwhelming majority of the patients received spectacle correction and had a significant reduction in visual impairment, then by extending appropriate refractive services, the burden of myopic visual impairment in underdeveloped countries can be reduced drastically [25]. Spectacles are relatively more affordable than other means of myopia management. Therefore, programs aimed at reducing the global burden of reversible visual impairment attributable to myopia in underdeveloped countries must focus on training more optometrists.

4.1 Strengths and limitations

The large size and diversity of the sample study sample is a considerable strength of this study, which makes the findings generalisable to the epidemiological profile of myopia in a hospital population and, to some extent, the general population. Nonetheless, this study has some limitations. First, as a hospital-based study, caution must be exercised in an attempt to extrapolate the findings to the general population due to the risk of overestimation. Also, due to the study’s retrospective design, it was prone to selection bias. Moreover, the urban nature of the study setting skewed its accessibility to mostly urban and suburban dwellers relative to rural inhabitants. A more uniform distribution in the sample would have improved the representativeness of the findings to the Ghanaian population. Notwithstanding the inherent challenges, the large sample size of this study and the inclusion of all the major ethnic groups in Ghana provided a somewhat reasonable estimation.

5 Conclusions

Ewe and Ga-Adangbe ethnic groups, and being a student are demographic factors associated with an increased risk of myopia within a clinical sample in Ghana.