Vision impairment, resulting in vision difficulties, is a leading cause of disability, and hence one of the key barriers for people to access education and employment, which may force them into poverty.
The objective of this study was to determine the prevalence of self-reported vision difficulties as an indicator of vision impairment in economically disadvantaged regions in South Africa, and to examine the relationship between self-reported vision difficulties and socio-economic markers of poverty, namely, income, education and health service needs.
A cross-sectional study was conducted in economically disadvantaged districts to collect data from households on poverty and health, including vision difficulty. As visual acuity measurements were not conducted, the researchers used the term vision difficulty as an indicator of vision impairment. Data were collected from 27 districts (74 901 respondents). Logistic regression analysis and chi-square tests were used to determine bivariate relationships between variables and self-reported vision difficulty. Kernel density estimators were used for age, categorised by self-reported and not reported vision difficulty.
Prevalence of self-reported vision difficulty was 11.2% (95% CI, 8.7% – 13.7%). More women (12.7%) compared to men (9.5%) self-reported vision difficulty (
The evidence from this study suggests associations between socio-economic factors and vision difficulties that have a two-fold relationship (some factors such as education, and access to eye health services are associated with vision difficulty whilst vision difficulty may trap people in their current poverty or deepen their poverty status). The results are thus indicative of the need for further research in South Africa.
Vision impairment, causing vision difficulties, is the leading cause of disability, and hence one of the key barriers for people in South Africa to access education and the labour market, which may force them into poverty. The War on Poverty Campaign (WOPC), an initiative of the South African Government to provide services and support for the ‘poorest of the poor’ families in the country, has developed a large database of households that were identified as being economically disadvantaged. The WOPC was commissioned by the presidency as part of the government's Apex priorities, with the deputy president as the lead person. The campaign seeks to raise the profile of current poverty eradication strategies and reach out to more people in South Africa. In addition, the campaign attempts to ensure that civil society, business, non-government organisations, and community-based organisations join in the anti-poverty effort (Santiaguel
In line with the objectives of the WOPC, a strategic partnership was formed between three eye health institutions namely, Orbis, the Brien Holden Vision Institute, and the African Vision Research Institute (AVRI) and the Department of Social Development to investigate poverty and eye health in South Africa. As part of the WOPC, this research collaboration brought together specialists – the researchers and authors – in the field of eye health to investigate the prevalence of eye diseases in South Africa. This three-pronged collaboration also aims to identify, implement, and monitor eye health services that are desperately needed to improve sight. This article, however, only presents data that were collected from the WOPC initiative, which provided the researchers with a select sample of poor households that had been identified with self-reported vision difficulties. Other studies, as part of this collaboration, are currently being conducted. In previous studies a general population would have been sampled, prevalence of blindness and vision impairment determined by means of visual acuity measurements, and then correlated with employment, education, quality of life and other socio-economic indicators. In this study the researchers have, however, direct access to households that the WOPC identified as being in poverty. Data collected from the WOPC include information on health and vision difficulties. Thus, access to the WOPC database provided the researchers with an opportunity to determine the prevalence of self-reported vision difficulties in economically disadvantaged regions of South Africa.
The purpose of this article was, therefore, to explore and analyse demographic (age, gender, education), socio-economic (housing, employment, social and documentation) needs, self-reported vision difficulty data as an indicator of vision impairment, and eye health service needs that were extracted from the WOPC database. The article examines the relationship between self-reported vision difficulty and indicators such as income, education and eye health service needs which are often used as markers for poverty in resource limited communities.
In this study we sought subjective (self-reported) responses. As visual acuity measurements were not conducted, the researchers used the term vision difficulty as an indicator of vision impairment which is usually classified by the World Health Organisation (WHO n.d.) according to a measured visual acuity.
South Africa has a population of 51.8 million people (Statistics South Africa [StatsSA]
High rates of unemployment, poverty and inequality in South Africa can be traced to colonial exploitation and the apartheid system as well as policies that denied African people access to opportunities, including access to land, to run businesses, to own certain assets, to quality education and to live in areas that were well established and located (Office of the Presidency
The
Poverty is much higher in rural areas, particularly in the former bantustans; however with the inward migration of people in search of work, deep poverty is also found in cities (Office of the President 2011:09). More than two-thirds (68.8%) of rural dwellers were estimated to be living in poverty in 2011, as compared with less than a third (30.9%) of residents in urban areas (StatsSA
Poverty is closely related to poor education and lack of employment. Almost 25% of the population is unemployed (StatsSA
According to May, the nature of poverty in South Africa is changing as is evident in the rural-urban migration trend, whereby the urban population increased by 9.5 million, thereby increasing the number of urban poor by 4.7 million, whilst the number of rural poor declined by 770 000 (May 2010:02). Despite this, livelihoods in poor rural communities are characterised by asset poverty, lack of access to the resources for food production and high levels of monetisation and integration into the broader economy (Du Toit
According to the South African National Council for the Blind (SANCB
There are 400 000 people in South Africa who are blind, that is, unable to see at three metres what a normal sighted person sees at 60 metres or vision worse than this (Mercy Vision
The South African Government has restructured the health system to improve equity and access to primary health care after the first democratic elections in 1994. However, major inequities remain despite achieving democracy and new leadership. Across South Africa's nine provinces, there are significant variations in health status and access to health services (Lawn & Kinney
The overwhelming majority of South Africa's poor live in rural areas without access to quality eye care services. Current estimates indicate that there are 3408 optometrists registered with the Health Professions Council of South Africa (Health Professionals Council of South Africa [HPCSA]
Furthermore, ophthalmology and optometry services are expensive and, until just a few years ago, were almost exclusively accessible only to those living in urban areas through the private sector (Rawlings
Theoretical evidence that poverty and poor eye health in developing countries is closely related has been succinctly captured in a literature review documented by Jaggernath
In considering the literature consulted on poverty, vision impairment, blindness and eye health services in South Africa, the aim of this study was to determine the association between self-reported vision difficulty, a proxy indicator of vision impairment, and socio-economic status in economically disadvantaged districts in South Africa. Specific objectives include determining the prevalence of self-reported vision difficulty in the economically disadvantaged regions of South Africa and examining the relationship between self-reported vision difficulty and indicators such as income, education and health service needs which are often used as markers for poverty in resource limited communities.
A cross-sectional study was conducted by the Department of Social Development in South African. The study was conducted in way of a census by eliciting data on poverty and health, including vision difficulties from all respondents within selected economically disadvantaged and not economically challenged wards.
Wards were allocated a poverty index and wards that were most economically challenged were identified through a prioritisation process using key indicators such as income, education level, and employment and mapping households in those wards to confirm their status as households in poverty. Twenty seven districts from all nine provinces in South Africa were found to be most economically challenged and 18 016 poverty-struck households that were randomly selected from these districts were profiled by the WOPC. Data were obtained from 74 901 respondents.
The WOPC collected household data by way of a census, using survey questionnaires. The preferred method of data collection was mobile phone questionnaires, because they were directly linked to the government's database and responses from participants are automatically stored on the database. Face-to-face interviews need to be manually entered into the government database, thus, when respondents had no access to mobile phones, face-to-face interviews were conducted. Data were obtained from 51.8% of households through mobile phone questionnaires and 48.2% of households by conducting face-to-face interviews. Questionnaires included sections for demographic data, socio-economic conditions and health and disability. Questions on health and disability included vision difficulty. The data were collected at the household level by trained enumerators. Information was provided by a household representative, either by the household head or any other member of the household who was over the age of 18 years. Following data collection, the questionnaire responses were captured by trained data entry technicians into an excel database. This database provided an opportunity to extract, analyse and disseminate information on poverty and self-reported vision difficulty in the economically disadvantaged districts in South Africa and investigate the prevalence of self-reported vision difficulty as well the relationship between vision difficulty and poverty, using indicators for poverty, such as income, education, and health service needs in resource limited communities. These indicators were selected as they formed questions in the WOPC questionnaire that related to the researchers’ objectives of correlating poverty markers with self-reported vision difficulties.
Logistic regression analysis was conducted to determine the association between poverty indicators and vision difficulty. Chi-square tests were employed to determine bivariate relationships between each variable and the reported vision difficulty.
Operational definitions of variables.
Data variable | Data explanation | Data type | Categories |
---|---|---|---|
Visual difficulty | Dependent: Difficulty in seeing even when using glasses | Categorical | 0=No; 1=Yes; 99=Do not know* |
Independent variables | - | - | |
Age | Age at last birthday | Continuous | - |
Gender | Sex of the respondent | Categorical | 1=Male; 2=Female |
Education | Highest education level successfully completed | Categorical | 0=no formal school; 1=primary school; |
Expenditure | Total household expenditure in the last month | Categorical | 1=0; 2=1-199; 3=200-399; 4=400-799; |
Heath care needs | Independent | Categorical | 0=No; 1=Yes |
Education needs | Independent | Categorical | 0=No; 1=Yes |
Documentation needs | Independent | Categorical | 0=No; 1=Yes |
Employment | Employment status | Categorical | 0=No; 1=Yes |
Province | Province of residence in South Africa | Categorical | 0=No; 1=Yes |
Household size | Total number of household members | Continuous | 0=No; 1=Yes |
Each option for occupation | - | Categorical | 0=No; 1=Yes |
Each option for business activities | - | Categorical | 0=No; 1=Yes |
*, a, do not know; b, refused or c, not stated were removed from the analysis but are presented in Tables.
The question on income in the WOPC questionnaire was scarcely reported; therefore, a limitation to the analysis was that respondents’ expenditure was utilised as a proxy to income. Importantly, there are three categories that do not state the exact expenditure: do not know, refused, and not stated. Non-responses were removed from the analysis.
Age distributions were compared between respondents who self-reported having vision difficulty even when wearing spectacles and those who did not have any vision difficulty. This was conducted by plotting kernel density estimators for age, categorised by self-reported and not reported vision difficulty.
Vision difficulty was examined in relation to the education and health care needs. Education needs include school feeding, fees and uniform, scholar transport, career guidance, access to bursaries, special educational needs, vocational skills development, further education and training (FET) and textbooks. Health service needs included the need for a road to health card (RTC), treatment or medication requirement, medical check-up, rehabilitation services, assistive devices, nutrition programmes, voluntary counselling and testing (VCT), immunisation and anthropometric measurements. In addition, documentation needs were combined to include identity document (ID), birth, marriage and death certificate, passport and residence permit.
Working was defined in the study as being formally employed and earning a salary. This question was administered to respondents who are above the age of 15 years and respondents who are not currently attending schools. In consideration to the possibility that some respondents may have multiple skills, respondents were given a multiple responses option for confirming their different skills. All tests were conducted at the 5% level of significance, taking into consideration the sample design.
The sample comprised of 55% women and 45% men. Respondents less than 15 years old comprised 32% of the sample, 32% were between the ages 15 and 34 years and 30% were aged 35 years and older. The ages were not stated by 6% of the respondents.
The prevalence of self-reported vision difficulty was 11.2% (95% CI: 8.7% – 13.7%) with 32.7% of these respondents wearing spectacles. Of the total sample 5.1% (95% CI: 3.78% – 6.36%) wore spectacles. In addition, 9.8% (95% CI: 7.29% – 12.30%) self-reported that their eye condition was permanent. A significant percentage of women (12.7%) compared to men 9.5% self-reported vision difficulty (
The highest level of education was not reported by 3% of the respondents. Educational level categories included no formal schooling, primary schooling, secondary school and tertiary education. There were 14.7% of respondents who did not state an education category; they, however, self-reported vision difficulty and were removed from this particular analysis. There is a statistically significant relationship between highest level of education and self-reporting of vision difficulty (
Self-reported vision difficulty was higher for respondents who are dependent on others (14.2%) and respondents who spend R10 000 or more per month (15.9%). Respondents who were employed have a significantly higher prevalence of self-reported vision impairment than those who are not employed (
Employment and vision status.
Expenditure was reported at household level and vision difficulty was self-reported at an individual level. However, it can be observed that self-reporting of vision difficulty for both the unemployed and the employed decreases as the household expenditure increases, save for household expenditure greater than R10 000 (
Self-reporting of vision difficulty by expenditure and employment status.
Self-reported vision difficulty was significantly higher for the Free State, North West, Western Cape and Gauteng provinces with 17.8%, 14.2%, 13.3% and 10.8% respectively (
Self-reported vision difficulty in the nine provinces of South Africa.
The highest number of occupants in a single household was 23 (
Self-reported vision difficulty by household size.
The following occupations were significantly related to self-reported vision difficulty: painting (
Odds ratios for skills.
Skill | Oddsratio (95%CI) | |
---|---|---|
Painting | 1.2 (1.06–1.43) | (p = 0.01) |
Bricklaying | 1.3 (1.12–1.42) | (p < 0.01) |
Waiter | 0.8 (0.68–0.99) | (p = 0.041) |
Security | 0.9 (0.64–1.24) | (p = 0.493) |
HCBC | 1.4 (1.13–1.64) | (p < 0.01) |
Welding | 1.4 (1.25–1.66) | (p < 0.01) |
Carpentry | 1.2 (1.02–1.46) | (p = 0.029) |
Electrical | 0.7 (0.54–0.85) | (p < 0.01) |
Plumbing | 1.0 (0.76–1.33) | (p = 0.978) |
Childcare | 1.4 (1.15–1.62) | (p < 0.01) |
Plastering | 1.1 (0.83–1.34) | (p = 0.660) |
Farming | 1.7 (1.45–2.03) | (p < 0.01) |
Sewing | 2.1 (1.81–2.40) | (p < 0.01) |
Bookkeeping | 0.9 (0.61–1.35) | (p = 0.615) |
More respondents aged below 40 years reported that they have no vision difficulty in comparison to those who are above 40 (
Comparisons between ages for those with self-reported vision difficulties and those without vision difficulties.
The distributions of the two functions using the Kolmogorov-Smirnov test for equality between distribution functions (
In relation to business activities some households reported multiple responses for the types of business activities they were involved in; meaning that they are involved in several business activities. In addition, when a business activity is reported for a household, every individual in that household is reported to be engaged in that business activity regardless of who is taking part in the type of business. Considering this, removing minors from the analysis would be speculative. This is because most of these businesses are operated from home and minors are exposed to similar environments as adults. For example, a 10-year old child may assist in selling goods or services in their family spaza [tuck shop] after school. About 7% of the respondents reported that they were involved in the selling of goods and services on the streets. Other business activities mentioned involved hair salons, welding, panel beating, vehicle repairs, wedding planners, property owners and carpenters. The self-reported vision difficulty for the different categories ranged from a high of 20% (30 out of 150) for those in the wood or fuel business and a low of 8% for those transporting goods.
Health and education needs were highest amongst respondents, 20.7% and 19.6% respectively, whilst social and documentation needs were 9.7% and 9.3%, respectively. The results show that 21.8% of the respondents with health needs self-reported that they have vision difficulty, with 5.8% of the respondents with educational needs self-reporting vision difficulty.
There is a significant relationship between self-reporting of vision difficulty and education needs (
Ethical clearance for presenting data extracted from the WOPC database was received from the University of KwaZulu-Natal Humanities and Social Sciences Research Ethics Committee and the South African Department of Social Development.
The primary research that informed this article was conducted by the government (Department of Social Development) in areas defined as economically disadvantaged in South Africa. Although the survey instrument used collected household members names and addresses, the data that were retrieved from the Department of Social Development were anonymous. Thus the responses received were not linked to any particular individual or households.
Secondary data were employed; therefore, particular information pertaining to questionnaire development and data collection process was carefully examined so as to inform the analysis methods and the limitations of the research results.
This research employed a quantitative approach; therefore the meaning of reported numbers will remain consistent over time. The questionnaire used was a revised version of a preceding similar study after removing ambiguous sections. In addition, the sampling entailed the profiling of wards in South Africa, followed by conducting censuses in the selected wards which increased the sample size making the estimates accurate for the population.
Validity determines if the research truly measures what it is intended to measure (Joppe
The prevalence of self-reported vision impairment or blindness found in this study (11.2%) is lower than the prevalence found in other prevalence studies (Bucher & Ijesselmuiden
Household size was a significant factor in the self-reported vision impairment. This could be the result of lesser resources or that limited resources have to be prioritised and vision is, thus, not high on the agenda when such choices have to be made. However, there is no definitive data to conclude this and further investigation in this regard will be useful.
When resources are limited, families prioritise certain expenses or services with women and girls being considered less of a priority (Courtright & Lewallen
The categories with the least expenditure and the most expenditure have the highest prevalence of vision impairment. If we take expenditure as a proxy for income and level of poverty this may be explained by the fact that those with the least income are less likely to access eye care services. Conversely those with the highest level of income are more likely to have jobs with a higher demand for vision tasks such as administration jobs and are more affected by the lack of vision correction or by vision impairment because of other causes.
Employed individuals and those with a higher family income have a higher prevalence of self-reported vision impairment. This could be a consequence of greater vision demands in work situations, which impacts subjectively on an individual as opposed to someone who does not have a great demand for vision. This would be particularly relevant in situations where near vision is demanded, such as for reading or sorting out products.
Two of the four provinces (Western Province and North West Province) which self-reported the highest level of vision impairment do not have optometrists employed in the public sector. These provinces are better resourced than some of the other provinces but for the poor very little affordable options exist when they cannot access public sector optometrists and affordable spectacles. Given that refractive error was found in an epidemiological study to be the major cause of vision impairment in South Africa (Naidoo
The higher self-reported vision difficulty in smaller families could be related to the fact that smaller households are composed of pensioners whose children live elsewhere and, as such, the relative influence of age-related conditions such as cataracts and presbyopia on vision impairment is magnified. The study showed that there was a steady increase in self-reported vision impairment from age 40 years and onwards, a peak prevalence at around 60 years and a subsequent decline post age 60 years, which correlates with the onset of presbyopia (ages 35–40 years) and cataracts (ages 55–60 years). South Africa enjoys one of the highest cataract surgery rates in sub-Saharan Africa; despite this a significant backlog of services exists. The current cataract surgical services may therefore explain the post age 60 years decline.
Those with health needs self-reported the highest percentage of vision impairment compared to social, documentation and education needs. The lack of health services in general could be indicative of the lack of access to eye care services for affected individuals and communities. This could also be influenced by other co-morbidities such as HIV, diabetes and hypertension leading to a prioritisation of health care needs amongst respondents. Some of these conditions also impact vision.
This study has limitations as the questions regarding vision difficulty were limited in the WOPC questionnaire. Further in-depth interviews and questions pertaining to poverty and vision difficulties will provide more information on the link between socio-economic status and vision problems. Unfortunately this was not possible at this stage, as the investigators analysed secondary data that were extracted from the WOPC questionnaire. A further limitation to the study was that vision difficulty in young children can be problematic to report as they are unable to effectively complain about an eye condition. In addition, it is difficult for parents to identify and report when young children have vision problems or diseases as many poor households do not access adequate health services to be able to detect vision problems early, unless they are severe or have been detected at birth.
The results of this study serve as an indicator of people's perception of their vision status and difficulties experienced. The evidence from this study suggests associations between socio-economic factors and vision difficulties that have a two-fold relationship: some factors, such as access to eye health services may lead to vision difficulty whilst vision difficulty may trap people in their current poverty or deepen their poverty status. However, the responses of the subjects do not indicate a clear relationship between income and employment and vision difficulty in KwaZulu-Natal as income was scarcely reported by the subjects enumerated and there is a higher prevalence of vision difficulties amongst the employed individuals in comparison with those who indicated that they were unemployed. Despite this, it should be noted that many individuals were found to be employed in areas of work that are likely to require good vision and perform tasks that may have increased vision demands, which may present vision difficulties such as painting, bricklaying welding, plastering plumbing. However given the relationship between vision demands and a greater awareness of vision impairment in work situations, it is necessary to investigate further the unemployed group via clinical evaluations as well as comprehensive qualitative studies to determine if a vision difficulty restricts their capacity to gain employment. There are also other dimensions to poverty such as social well-being, health and living conditions, and others, which impact on vision but have not been investigated in this study in relation to vision difficulty. The current data are therefore insufficient to examine this relationship. The results should therefore be considered as indicative of further areas for research.
The authors have earlier presented specific research questions that need further investigation on poverty and eye health (Jaggernath
Research areas emanating from the study.
Research area | Research interests |
---|---|
Income | Expenditure was utilised as a proxy to income in this study because the question pertaining to income in the WOPC questionnaire was scarcely reported. This could likely be because of the respondents not wanting to disclose their actual income to government as they may feel that they will be denied government grants if they state an income that is too high, or that they may be feeling a sense of inferiority and were afraid of the enumerator's perception of their response and thus avoided the question. However expenditure does not provide the actual earnings of the individuals in the households that were surveyed and would need to be cross-tabulated against household's income. Thus, a national study that does not place focus on poverty only, such as an investigation into poverty and eye health study, can help to elicit responses on income (in categories), without the fear of respondents being afraid to declare their earnings. |
Inequality | Considering the high rates of unemployment, poverty and inequality in South Africa, especially within race and specifically on the African population, it is imperative that further studies be conducted to investigate the rate of inequality with regard to income, in the African population and within the different historical racial categories.Furthermore, despite advances in areas such as electrification and access to education that have increased equality of opportunities, more research is needed to determine what other opportunities (social, economic and health) have increased in terms of equality. Specifically, studies that investigate inequality in access to opportunities that improve eye health should be carried out. |
Migration patterns | Leibbrandt |
Eye health services | The study suggested that those with health needs have a higher percentage of vision impairment compared to other needs such as social, education and documentations. Lack of health services in general, could be indicative of the lack of access to eye care services for affected individuals and communities. An investigation needs to be conducted on the barriers that poor people face with regards to access to eye health services. |
Gender and access to eye health services | |
Household priorities and eye health | In this study, household size was a significant factor in the self-reporting of vision impairment. An assumption of the authors are that larger houses mean that there are lesser resources or that limited resources for poor people need to be prioritised before eye health; thus vision problems are less of a priority and remain untreated. However, further investigation on the association between household size, household priorities and eye health need to be carried out to verify this assumption. |
Vision difficulties in children |
The authors would like to express gratitude to the South African Government's Department of Social Development for access to the War on Poverty Campaign (WOPC) data. We also thank the Vision Cooperative Research Centre (VCRC) for supporting the contribution of the African Vision Research Institute to the Poverty and Eye Health project.
The authors declare that they have no financial or personal relationships which may have inappropriately influenced them in writing this article.
K.S.N. (Brien Holden Vision Institute, African Vision Research Institute) made a substantial contribution to the research design, was responsible for conceptualisation of the research and coordination of the research manuscript. J.J. (University of KwaZulu-Natal, Vision Cooperative Research Centre), P.R. (Orbis) and F.C. (Brien Holden Vision Institute, African Vision Research Institute) were responsible for research design and analysis of the data and the write-up of the research findings. T.Z. (Department of Social Development) was responsible for data collection performed by that department and made intellectual contribution in the manuscript. L.Ø. (Orbis) was responsible for the conceptualisation of the research, and made a substantial intellectual contribution to the manuscript.