J Community Health DOI 10.1007/s10900-015-0008-2
ORIGINAL PAPER
Determinants of Eye Care Utilization in Rural South-eastern Nigeria O. C. Arinze1 • B. I. Eze2 • N. N. Ude2 • S. N. Onwubiko2 • C. N. Ezisi1 C. M. Chuka-Okosa2
•
Ó Springer Science+Business Media New York 2015
Abstract To determine the barriers and incentives to eye care utilization (ECU) in Abagana, a rural south-eastern Nigerian community. The study was a population-based cross-sectional mixed method (quantitative and qualitative) survey of adult inhabitants of Abagana, in July–August, 2011. Data on respondents’ socio-demographics, barriers and incentives to ECU were collected, and analysed using descriptive and comparative statistics. A p \ 0.05 was considered statistically significant. Qualitative data were obtained from focus group discussions and in-depth interviews and analyzed using ATLAS.ti software. The 549 respondents (males 224) were aged 49.8 ± 15.9 SD years (range 18–93 years). Overall, orthodox eye care had ever been sought by 42.4 %, and 46.4 % of those who reported change in vision. Lack of awareness-31.8 %, cost-18.0 %, and fatalistic attitudes-15.9 % were the main ECU barriers. Possession of health insurance (OR 11.49; 95 % CI 4.21–31.34; p = 0.001), family history of eye disorder (OR 3.27, 95 % CI 2.03–5.26; p = 0.001), noticed change in vision (OR 11.30; 95 % CI 1.42–90.09; p = 0.022), current eye disease (OR 4.06; 95 % CI 2.29–7.19; p = 0.001) and systemic co-morbidity (OR 4.33; 95 % CI 2.67–7.02; p = 0.001) were the incentives to ECU. To enhance the low ECU in Abagana community, educational interventions on eye health maintenance and eye health seeking behaviours, and measures to reduce eye care cost are recommended. & B. I. Eze [email protected] 1
Department of Ophthalmology, Federal Teaching Hospital, Abakiliki, Ebonyi State, Nigeria
2
Department of Ophthalmology, University of Nigeria Teaching Hospital, Ituku-Ozalla 01139, Enugu, Nigeria
Keywords
Eye care utilization Barriers Incentives
Introduction Primary eye care (PEC) is the provision of essential, affordable, accessible, practical and sustainable eye healthcare to the general population [1]. PEC delivery utilizes the horizontal integration matrix model proposed by the World Health Organization (WHO) to incorporate PEC programs into the existing primary health care (PHC) structures. Eye care encompasses promotive, preventive, curative and rehabilitative services; its delivery could be institution- or community-based, or both [1]. The World Health Organization estimated the number of the world’s blind at 45 million and those visually impaired at 135 million [2]. In the Nigeria National Blindness Survey, avoidable causes, mainly cataract, accounted for 84 % of blindness [3]. However, in rural Nigeria, other prevalent eye diseases/disorders include glaucoma (21.1 %), allergic conjunctivitis (16.4 %), refractive errors (12.4 %), age-related macular degeneration (0.7 %) and corneal scarification (0.7 %) [4]. The survey further forecasted that, if the cataract surgical coverage remains unchanged, by the year 2020, the number of adults with operable cataract will increase by 43 % [3]. Of the global blindness burden, 90 % live in developing low and middle income countries (LMICs) where the resources and logistics for delivery of quality eye care are inadequate [2]. Consequently, regional [5, 6] and global [7] reports have observed that uptake of eye care services lags behind the established need for eye care. The factors responsible for initiation and sustenance of the gap between eye care need and uptake are multi-dimensional [8]. These include health system-dependent
123
J Community Health
factors like availability, and affordability of care; and patient-related factors like, gender, [9] awareness, attitudes to care, [10] direct [10, 11] and indirect [9] cost of care, and profile/severity of vision loss [9]. This needs-uptake gap accounts for non-detection of about half of the potentially blinding eye diseases, often with adverse consequences for mortality, morbidity and quality of life, among visually handicapped persons [11–13]. To clear the backlog, and treat incident cases of eye diseases, ‘VISION 2020: The Right to Sight’, the Global Initiative for the Elimination of Avoidable Blindness, has endorsed the need to reverse the increasing trend of global blindness through unrestricted access to eye care services [14]. Achieving the goals of this global initiative largely depends on the ability to identify and overcome barriers, and augment/expand the incentives, to utilization of eye care services [7]. Consequently, this survey, a mixed-methods quantitative and qualitative, cross-sectional population-based study, was conducted on the inhabitants of Abagana, a rural underserved south-eastern Nigerian community in July– August, 2011. It aimed to assess their utilization of, determine barriers to, and identify characteristics associated with utilization of available orthodox eye care services. The results, in addition to providing comparative loco-regional data, would assist eye care planners, and implementers, and health policy makers, in the study area and similar settings elsewhere, in strategizing for realization of the goals of ‘VISION 2020: The Right to Sight’.
-ophthalmologists, optometrists and nurses; and complemented by alternative/traditional medicine practitioners and patent medicine dealers. There are numerous public and privately-owned health care facilities. Of the public facilities, only Abagana Comprehensive Health Centre has a dedicated Eye Unit which provides medical, refraction and minor surgical eye care services to the inhabitants of the community and beyond. The centre is manned by visiting ophthalmologist and optometrist, and resident nurses and midwives. There is no privately-owned eye care facility in Abagana. This was a population-based qualitative and quantitative cross-sectional study conducted from July to August, 2011. Eligibility The participants comprised adult males and females, aged 18 years or older, who had resided continuously in Abagana for at least 6 months, who voluntarily consented to participation. Ethics Prior to commencement of study, ethics clearance consistent with the 1964 Helsinki Declaration, as revised in 2008, was obtained from the Medical and Health Research Ethics Committee (Institutional Review Board) of the University of Nigeria Teaching Hospital, Enugu. Study Design
Methods
Population-based quantitative and qualitative survey.
Background
Sample Size and Sampling
Anambra state, along with the states of Abia, Enugu, Ebonyi and Imo make up Nigeria’s south-east geo-political zone, one of Nigeria’s six geo-political divisions. The state is sub-divided into administrative sub-units or local government areas (LGAs). Each LGA is further sub-divided into political Wards; each ward is made up of villages. Abagana is one of the 6 rural communities that make up Njikoka LGA of Anambra state. It lies in the tropical rainforest climatic belt and has a landmass of approximately 9.2 km2. Based on the 2006 Nigerian National Census, the projected population of Abagana is 36,572. The inhabitants of Abagana are predominantly ethnic Ibos, whose main occupations are farming, trading, civil service and artisanship. In Nigeria, the eye care system comprises tertiary, secondary and primary levels of eye care. In Abagana, a rural community served by primary level eye care, eye care is provided mainly by orthodox eye care providers
The minimum sample size, n, was calculated using Fisher’s formula for sample size determination for a definite population i.e. n = [Z2a pq]/d2 when the population is more than 10,000. Where n = the minimum sample size; p = previously reported prevalence of 61 % in a related Nigerian report by Rabiu et al. [10] thus giving a prevalence (p) value of 0.61, Za = standard normal deviation of 1.96, at confidence level of 95 %, d = desired precision due to random sampling error of 5 % = 0.05, q = 1-p; [(1.96)2 9 0.61 9 0.39]/ 0.052 = 0.9139/0.0025 = 366. The calculated sample size of 366 was multiplied by 1.5, the design effect due to cluster sampling method, to give a modified sample size of 549. The sampling method was multi-stage cluster random sampling technique. The 10 villages that make up Abagana are already divided into four political Wards. The Wards served as clusters. The Wards and their component villages, and populations are: Ward 1: Adaegbe-Umudunu
123
J Community Health
and Uru-Okpala; population = 8071; Ward 2: EnuoraOraofia and Adaegbe-Oraofia; population = 8829; Ward 3: Enuora-Umudunu and Amagu; population = 7566; Ward 4 : Akpu, Amaenyi, Uru, and Adegbe; population = 12,106. From the proportion of the 549 participants, the modified sample size, to be recruited from each ward was calculated based on proportionate representation:—Ward 1 = 122; Ward 2 = 132; Ward 3 = 114; and Ward 4 = 181. Using a simple random sampling, the following villages were selected from their respective clusters (Wards). In each of the selected villages, all households were enumerated to get a sampling frame: Ward 1 = Uru-Okpara, 709 households; Ward 2 = Enuora-Oraofia, 728; Ward 3 = Enuora-Umudunu, 698; Ward 4 = Akpu, 463 and Adaegbe, 485. Using a systematic random sampling, k (sampling interval, skip) was calculated as k = N/n, where N = the population size of each ward and n = sample size i.e., the calculated proportion of the 549 participants to be recruited for each ward. Subsequently, starting from a randomly selected household and sampling direction, every sixth household in Uru-Okpara, Enuora-Oraofia and EnuoraUmudunu, and every fifth in Akpu and Adaegbe were selected until the required sample size of 549 households was obtained. From each of the selected households, one eligible adult was recruited by balloting. Any household in which nobody or eligible participant was around/available for recruitment or in which none gave voluntary informed consent was skipped without revisiting the household. A total of 12 focus group discussions (FGDs) comprising 6 all-male and 6 all-female groups, with 7 participants in each group, were conducted in the four Wards/clusters. In each ward, 3 FGDs, comprising at least 1 all-male and one all-female FGD, were conducted. The criteria for participation in FGD were age greater than 18 years, uninterrupted residence in the study community for the past 6 months and non-participation in the preceding questionnaire-based arm of the same survey. Each Ward/cluster comprised 2 villages, except Ward 4 which has 4 villages. For Wards 1, 2 and 3 which have 2 villages each, 1 village was randomly selected for male FGD and the remaining village for female FGD. Afterwards, the decision on whether to perform male or female FGD, to complete the three mandatory FGDs per ward, was reached by simple balloting (male vs. female). The village, within each ward, from where the participants in the supplementary i.e., third FGD were drawn, was similarly determined. In Ward 4, comprising 4 villages, the village for male FGD was randomly selected; this was followed by another random selection of another village, from the remaining 3 villages, for female FGD. The gender composition (male vs. female) and hosting village, for the third FGD, was each determined by simple balloting. The participants election procedure is illustrated in the flow chart below (Fig. 1).
Additionally, in-depth interviews (IDIs) were also conducted on the Igwe (Traditional ruler) of Abagana, President of Abagana Town Union, Chairman of Ward 4 (randomly selected from the four wards) and 3 health personnel working at the Comprehensive Health Center, Abagana.
Data Collection Procedures This was carried out by the researchers who are fluent in Igbo, the native language of the Abagana community, and English languages. Quantitative Data The study instrument, developed de-novo, was a pre-tested researcher-administered open- and closed-ended questionnaire with sections on participants’ socio-demographics (age, sex, education, marital status and occupation), clinical characteristics, pattern of eye care service utilization, barriers and incentives to eye care utilization. To ascertain its construct validity and psychometric reliability, the study questionnaire was pretested on a rural population, outside the study area, which share similar settings/characteristics as the study population. The final version of the questionnaire had internal consistency reliability [Kronbach alpha (a) coefficient of 0.8 and Lawshe’ s content validity index of 0.7]. The barriers to obtaining regular eye care services were explored among those who noticed a change in vision and did not seek orthodox eye care, and those who did not notice any change in vision but did not go for routine eye checks. When multiple reasons were given by the participants, the most important reason, as perceived by the subjects, was marked with a separate code. Qualitative Data The FGDs and IDIs were conducted by the researchers, using guides for FGDs and IDIs, while a trained research assistant took notes. The guides contained diagnostic questions on some issues raised in the questionnaire study. This was meant to provide in-depth knowledge of the dynamics in decision-making and response patterns with regards to barriers and incentives to eye care service utilization. The services of a sociologist were utilized in the development and administration of the FGD and IDI guides, as well as the analysis of the generated data. A total number of 12 FGDs were conducted with 7 persons in each group. In each ward, FGDs were held with representatives from 3 age groups (young adults aged 18–35, adults aged 36–59 and elderly people aged 60 or older). The moderator
123
J Community Health
conducted FGDs with sex-homogenous groups (i.e., each group comprising entirely of males or females), in culturally appropriate locations chosen by the participants. IDIs were conducted with health workers, and the opinion leaders of the community who are the custodian of the community’s norms, cultures and traditions. The language of administration of the survey was Igbo language, the native language of the host study community. Same sex FGDs and IDIs were conducted to overcome any potential sex-sensitive cultural inhibition to volunteering critical information relevant to the study. All the FGD and IDI sessions were recorded on tape.
Data Analysis Quantitative Analysis All data were cleaned and edited, double-entered into and analyzed using the statistical package for social sciences
Fig. 1 Flow chart illustrating the selection procedure for FGD participants
(SPSS) software, version 18.0 (SPSS Inc, Chicago, Illinois, USA). Data were subsequently categorized, and subjected to descriptive statistical evaluation to yield frequencies, percentages and proportions. Bivariate comparisons for statistical significance of observed betweenclass differences were performed using Chi square (v2) test for categorical variables and Student t test for continuous or metric variables. In all bivariate comparisons, a p \ 0.05 and the associated crude odds ratios and 95 % CI were considered statistically significant. In cases of multi co-linearity, all the significant (p \ 0.05) predictor variables were selected and fed into a multivariate logistic regression model to determine the independent effect of each predictor on the outcome of interest, eye care utilization. Adjusted (cofounder-corrected) odds ratios (OR) and their 95 % CIs, associated with p \ 0.05, were considered statistically significant. In the multivariate model, quantification of the effect size (effect size statistic) of each predictor variable was indicated by the associated odds ratio.
Total number of potential eligible participants that met the inclusion criteria = 143
Number that consented to parcipate= 120
Total number of randomly selected parcipants in the 4 wards, ensuring gender balance and proporonate distribuon by ward/village = 84
Ward 1 3 FGDs; 21 parcipants
123
Ward 2 3FGDs; 21 parcipants
Number that declined parcipaon= 23
Number not selected = 36
Ward 3 3 FGDs; 21 parcipants
Ward 4 3 FGDs; 21 parcipants
J Community Health
Qualitative Analysis Analysis of the qualitative data placed emphasis on the interpretation, description and recording/writing of what was actually said. The responses were translated into English language. In going through the transcription, phrases with contextual or special connotations were noted and pulled out as illustrative quotes for complementing the quantitative data. To do this, relevant themes were developed for the coding and sorting of the qualitative data. To facilitate the qualitative data processing, each of the transcripts was entered into a word processor and saved as a rich text format (RTF) file, which was formatted and coded using the ATLAS.ti. This was utilized in generating illustrative quotes from the qualitative data to support the quantitative data.
Results Participants’ Socio-Demographics There were 549 participants comprising 224 (40.8 %) males and 325 (59.2 %) females (male: female ratio = 1:1.5) who were aged 49.8 ± 15.9 SD years (range 18–93 years). The majorities of the participants were married—370 (67.4 %), possessed formal education—459 (83.6 %) and were frequently traders—138 (25.1 %). The participants’ demographic characteristics are shown in Fig. 2 while their socioeconomic profile is reported in Table 1. Clinical Profile Of the 549 participants, 233 (42.4 %) had ever utilized eye care services while 316 (57.6 %) had not. Five hundred (91.1 %) participants had ever noticed a change in their vision while 49 (8.9 %) had not. Of the 500 who noticed change in their vision, 232 (46.4 %) did, while 268 (53.6 %) did not, access orthodox eye care services as a result. Among the 49 participants who did not notice any
Table 1 Socio-economic profile of respondents Characteristics
Total (%) n = 549
Marital status Married Single Divorced/separated Widowed
370 (67.4) 88 (16.0) 2 (0.4) 89 (16.2)
Educational status Primary Secondary
192 (35.0) 173 (31.5)
Tertiary
94 (17.1)
None
90 (16.4)
Occupation Civil servant
130 (23.7)
Trading
138 (25.1)
Artisan
74 (13.5)
Farming
109 (19.9)
House wife
22 (4.0)
Others
76 (13.8)
change in their vision, only 1/49 (2.0 %) went for routine eye check (ocular health maintenance visits) by an orthodox eye care providers while 48 (98.0 %) did not. The main barriers to visual symptom-initiated eye care utilization were lack of awareness—85 (31.7 %) of available eye care services, cost—48 (17.9 %) and attitudinal barrier i.e., ‘I thought it was a minor thing’—42 (15.7 %). The participants’ leading barriers to routine eye check were selfassessment of one’s eyes being okay i.e., ‘my eyes are okay; no reason to go’-20 (41.7 %) and cost-8 (16.7 %)—Table 2. In the bivariate analysis, several of participants’ sociodemographic (age, sex, formal education, employment status and possession of health insurance) and clinical (noticed change in vision, current eye disease, family history of eye disease, non-ocular co-morbidity) were significantly (p \ 0.05) predictive of eye care service utilization, Table 3. However, in the multivariate logistic model, possession of health insurance, noticed change in vision, current eye disease, family history of eye disease and non-ocular co-morbidity were retained as significant independent predictors of eye care utilization, Table 4. Qualitative Results Focus Group Discussions
Fig. 2 Age-sex distribution of 549 participants
A total of 84 persons participated in the FGDs. On barriers to eye care services, 81 participants made comments that were grouped as lack of awareness (N = 24), cost (N = 21), attitudinal (N = 16), preference for alternative
123
J Community Health Table 2 Barriers to eye care services utilization among those who noticed a change in their vision but did not seek orthodox eye care and those who noticed no change in vision but did not seek routine eye check Barrier
Noticed change in their vision n (%) n = 268
Did not notice any change in their vision n (%) n = 48
Personal barriers Lack of awareness of eye care services
85 (31.7)
4 (8.3)
I thought it was a minor thing I prefer alternative eye care services
42 (15.7) 28 (10.5)
0 (0.0) 2 (4.2)
Fear of blindness resulting from treatment
28 (10.5)
5 (10.4)
My eyes are okay; no reason to go
11 (4.0)
20 (41.7)
7 (2.6)
0 (0.0)
48 (17.9)
8 (16.7)
I did not know that my eye problem can be treated Socio-economic barriers No money to pay for eye care services Poor vision is normal for old people like me
5 (1.9)
0 (0.0)
It is God’s will that my vision be bad
4 (1.5)
0 (0.0)
I am very busy I don’t have time to go
3 (1.1)
4 (8.3)
The health centre is too far from my house
3 (1.1)
4 (8.3)
I have nobody to escort me to the clinic
2 (0.8)
1 (2.1)
2 (0.8)
0 (0.0)
It is against our culture to have eye operation
eye cares services (N = 12) and fear (N = 8) barriers. As regards incentives, 68 participants made comments. Incentives mentioned included: having a health insurance policy (N = 26), noticing a change in vision (N = 22), and having an eye disease (N = 20). Few (N = 5) stated that having a first degree relative with an eye disease motivated them to see an eye doctor.
Discussion The participants comprised more females than males, often aged 40 years or older, who were frequently married, possessed formal education and had civil service and trading as their main occupations. The observed socio-demographic profile is similar to findings reported in Kaduna, Nigeria [11] and India [15]. These and our study were all conducted in rural communities where the majority of the men in the working age group are likely to be out, working in the cities, while the women are left to take care of the households, thus making them more available to participate in the study. The participants’ overall level of utilization of orthodox eye care services is low; utilization was more often symptom-initiated than asymptomatic routine eye checks. The findings in this present study are similar to findings in Calabar, Nigeria, [6] but contrast the report from Australia [17]. The Calabar, [6] Nigeria, report was on patients presenting to the hospital; as such, it likely addressed symptom-initiated hospital visits only, the dominant cause of consultation in the present report. The participants in the
123
Australian survey belong to a comparatively higher socioeconomic group, thus eliminating a critical health care access barrier, cost. These findings underscore the need to encourage timely presentation of symptomatic patients, and emphasize the constant necessity for routine eye checks, even in the absence of eye symptoms. Poor awareness of availability of eye care services, fatalistic attitudes, cost and preference for alternative eye care were the participants’ main barriers to accessing eye care services. This is similar to the findings in Calabar, [16] Nigeria, where lack of awareness constituted a major barrier to utilization of eye care services. Indeed, many of the participants from the FGDs said ‘we were not aware a hospital existed there, let alone the availability of eye care services.’ This could be explained by the location of the health centre at the border of Abagana with its neighboring community Enugwu-Ukwu. This location is quite distant from major residential areas of the community. Also, poor publicity, by the hospital, of its eye care services could be contributory. In contrast, a related Ghanaian study [10] reported a high level of awareness among its study participants. This could be explained by the observation from the Ghanaian study that majority of the cataract patients used for the study could access services within a short travelling time from their communities to outreach centers and as such had reasonably easier spatial access to eye care services. This, in addition to contributing directly to service underutilization, might encourage the villagers to patronize alternative eye care providers often with disastrous visual
J Community Health Table 3 Characteristics predictive of participants’ utilization of orthodox eye care services Characteristic
Total (%) n = 549
Utilisation of eye care services Yes n = 233
Odds ratio
95 % CI
p value
0.23–0.52
0.001*
No n = 316
Age (years) C40
375 (68.3)
188 (80.7)
187 (59.2)
0.35
\40 Gender
174 (31.7)
45 (19.3)
129 (40.8)
Reference
Female
325 (59.2)
152 (65.2)
173 (54.7)
0.65
0.46–0.91
0.017*
Male
224 (40.8)
81 (34.8)
143 (45.3) 1.1
0.69–1.74
0.761
1.99
1.31–3.33
0.01*
41.55
5.69–303.38
0.0001*
8.04
4.96–13.05
0.001*
5.4
3.63–8.05
0.001*
7.85
5.16–11.95
0.001*
9.73
4.49–21.04
0.001*
Possess formal education Yes
459 (83.6)
193 (82.8)
266 (84.2)
No
90 (16.4)
40 (17.2)
50 (15.8)
Employed
268 (48.8)
209 (89.7)
59 (18.7)
Unemployed
281 (51.2)
24 (10.3)
257 (81.3)
232 (99.6)
268 (84.8)
Employment status
Noticed a change in vision Yes
500 (91.1)
No
49 (8.9)
1 (0.4 %)
48 (15.2 %)
Has an eye disease currently Yes
378 (68.9)
No 171 (31.1) First degree relation has an eye problem
210 (90.1)
168 (53.2)
23 (9.9)
148 (46.8)
Yes
165 (30.1)
116 (49.8)
49 (15.5)
No
384 (69.9)
117 (50.2)
267 (84.5)
Has a non-ocular health problem Yes
164 (29.9)
124 (53.2)
40 (12.7)
No
385 (70.1)
109 (46.8)
276 (87.3)
Possess health insurance Yes
55 (10.0)
47 (20.2)
8 (2.55)
No
494 (90.0)
186 (79.8)
308 (97.5)
*Significant (p \ 0.05)
Table 4 Significant incentives to eye care service utilization on multivariate analysis
Incentive
Adjusted OR
95 % CI
p value
Noticed a change in my vision
11.30
1.42–90.09
0.022
Has an eye disease
4.06
2.29–7.19
0.001
1st degree relation has an eye problem
3.27
2.03–5.26
0.001
Has another non- ocular health problem Has a health insurance policy
consequences, and avoidable increase in the burden of eye diseases and blindness. To overcome the awareness barrier, the investigators recommend community-based eye health education utilizing culture- and settings-appropriate information dissemination channels. The education program should indicate the scope of available eye care services; and highlight the efficiency, efficacy, and cost-effectiveness advantage of orthodox versus alternative eye care. The prominent barrier role of cost in the present study is
4.33
2.67–7.020
0.001
11.49
4.21–31.34
0.001
consistent with the findings in developing [10, 18] and developed [19, 20] countries but differs from studies [21, 22] elsewhere where fear of surgery and treatment outcome were the major barriers. Indeed, many from the FGDs said ‘it is because we have no money and do not know where to go for our eye problem that we are forced to go to the traditional healers’. Many also said ‘‘we are unable to utilize the eye care services because they are not affordable’’. This has critical implications for presentation for
123
J Community Health
care and its timeliness, and by extension care outcomes, and might encourage recourse to hazard-prone cheaper alternatives to allopathic eye care services. This has a collateral adverse impact on the burden of blindness, which, if due to alternative medicine use, might be irreversible. The present data suggest the need for grass root economic empowerment of the study population and enhanced affordability of eye care services through service subsidy and provision of health insurance. The preference for alternative eye care services, self eye medication inclusive, by an appreciable proportion of the study participants corroborates the findings in Ghana, [23] Kaduna, Nigeria, [24] and Malawi [25]. This similarity could be explained by the relatively easy accessibility and affordability of these alternative eye care services to the community members. However, the findings of our study differ from the relatively strict preference for orthodox eye care reported by Dawodu et al. [26] in Benin, Nigeria. The difference between the present result and Dawodu et al’s. [26] could be attributed to the nature of the Benin study, which was an urban-based survey conducted on educated staff of state Ministries in Benin, who, considering their level of education were more likely to seek orthodox eye care services. In addition, information from the IDIs of the health workers suggests that inadequate infrastructure, equipment and skilled manpower results in weakness of the orthodox eye care system. This makes alternative eye care services appear more dependable to the community members. This finding underscores the need for strengthening the resource base of the eye care system; and expanding the scope of the earlier suggested eye health education to include the dangers associated with patronage of alternative eye care services. Alternatively, since they are communitybased and comparatively more accessible, these alternative eye care providers could be trained and incorporated into orthodox eye care system to provide supervised eye care. This approach has been shown to be feasible and effective in Malawi by Courtright [27] and Courtright et al. [28]. The barrier imposed by the fatalistic attitudes of the present survey participants towards eye diseases, similarly reported by Kovai et al. [29] and Bailey et al. [30] has enormous implications for care of initially asymptomatic, but potentially blinding, eye diseases like glaucoma. Therefore, preventive educational interventions on eye health should highlight this peculiar group of eye diseases and the need for their early diagnosis through regular eye health checks irrespective of presence of eye symptoms. Fear of unfavorable treatment outcomes constituted a barrier amongst participants who noticed change in vision but did not access eye care services. This has been previously reported in Tanzania [21] and India [22]. Especially in surgical intervention like cataract surgery, a single treatment failure needs about five well-advertised success
123
cases to neutralize its demotivating effects on the community [30]. Educating eye care consumers on the positive prognostic value of early presentation, other determinants of medical and surgical treatment outcomes and the comparatively higher risks of poor outcomes associated with unorthodox eye care interventions are suggested to dispel this fear. Towards this end, WHO further recommends better cataract monitoring and evaluation system against recommended guidelines and preferred practice patterns in order to improve visual outcome [31]. Possession of health insurance, but not age, gender, educational or employment status, was the only socio-demographic characteristic independently predictive of eye care utilization. The present result corroborates an Australian [32] report but differs from the findings of Schaumberg et al. [33] and Nirmalan [15] and associates. The positive predictive role of health insurance is consistent with the earlier identification of cost in our study as a major barrier to eye care utilization as observed in an American study [34]. This, provides suggestive evidence that the study population is not adequately empowered financially to bear the direct and indirect costs of orthodox eye care. The present data suggests further strengthening, and widening of the coverage of, health insurance policy in the study area, and under similar settings elsewhere. The curious lack of association between formal education and eye care utilization challenges the findings from previous studies [29, 35] and suggests gross deficiency of health education content of the formal educational training curriculum. As such, possession of formal education did not confer any health literacy advantage. However, the overriding influence of other barrier factors, especially cost barrier, could have masked the incentivizing effect of formal education. Consequently, curricular review, at various levels of education, with appropriate inputs from health educators, is instructive. Also contrary to previous reports, [35, 36] older age was not associated with significant increase in eye care utilization despite the established higher prevalence of agerelated eye diseases in this population sub-group. This observation could be attributed to age-related comprise of physical/financial independence and loss of autonomy on health-seeking behaviors/decisions. Many in this age group, with increasing age become progressively dependent on their immediate and extended family members and may not be able to seek orthodox eye care at will when needed, without assistance. To enhance the utilization of eye care services by this vulnerable age group, the authors suggest broadening of the scope of health insurance to include non-civil servants, and effective implementation of retirement and pension schemes to engender some degree of financial empowerment and autonomy, enough to carter for health care needs of people in this age group.
J Community Health
Institutions, specially equipped to take care of the elderly, should also be established. Eye care service utilization was significantly associated with noticed change in vision, current eye disease, family history of eye disease and concurrent non-ocular co-morbidity. Similar patterns of symptom-determined eye care service utilization have been found in India [28] and Ethiopia [9] and probably reflect the general trend in LMICs where limited financial resource and poor needs awareness restrict practice of routine/asymptomatic eye check. Again, this further reinforces the earlier introduced rationale for emphasizing this aspect-specific area of promotive and preventive interventions, during eye health education campaigns. Additionally, concurrent non-ocular co-morbidity and having a first degree relation with an eye problem were significantly associated with having seen an eye doctor. As suggested by Wang et al. [17], fear of developing similar eye conditions, as their first degree relatives, and the likelihood of the co-morbid systemic diseases conditions developing ocular complications could have caused these participants to seek orthodox eye care. Despite the advantage of its population-based mixedmethods design, the conclusions drawn from this study are limited by intrinsic bias inherent in participant-reported information. Also, the study setting was a rural area; so, the findings might not generalize to urban populations. However, the findings will potentially inform eye care practice, programmes and policy changes in similar settings in Nigeria, and other LMICs elsewhere.
Conclusions In Abagana, south-east Nigeria, the utilization of available eye care services is poor. The main barriers to eye care service utilization are poor awareness of the service availability, attitudinal barriers and cost. Possession of health insurance, family history of eye disease, noticed change in vision, current eye disease and non-ocular comorbidity are the incentives to service utilization. To increase eye care access, awareness creating measures, eye health education and enhanced service affordability are recommended. Towards these ends, eye care worker-led community-based public eye health education and tailored changes in rural eye care system, to overcome access barriers, are instructive. Future researches on patterns and determinants of eye care seeking behaviors among rural and underserved populations are warranted. Acknowledgments The authors wish to acknowledge the assistance of inhabitants of Abagana community during the study. Conflict of interest
None declared.
References 1. Eze, B. I., & Maduka-Okafor, F. C. (2009). An assessment of the eye care workforce in Enugu state, south-eastern Nigeria. Human Resources for Health, 7, 38. 2. Unite For Sight, Inc. Global Eye Health Statistics. http://www. uniteforsight.org/eye_stats.php. Date Accessed April 20, 2013. 3. The Nigeria national blindness and visual impairment survey 2005–2007. http://pbunion.org/Countriessurveyresults/Nigeria/ Nigeria_survey_Summary_report.pdf. Date Accessed April 7, 2013. 4. Adegbehingbe, B. O., & Majengbasan, T. O. (2007). Ocular health status of rural dwellers in south-western Nigeria. Australian Journal of Rural Health, 15(4), 269–272. 5. Whitworth, J. (1999). Determinants of attendance and patient satisfaction at eye clinics in south-western Uganda. Health Policy and Planning, 14(1), 77–81. 6. Lewallen, S., & Courtright, P. (2006). Increasing uptake of eye services by women. Journal of Community Eye Health, 19(60), 59–60. 7. Donoghue, M. (1999). People who don’t use eye services: Making the invisible visible. Journal of Community Eye Health, 12(31), 36–38. 8. Community Eye Health Online Course, Module 2: Patient barriers to eye care. http://www.uniteforsight.org/community-eyehealth-course/module2. Date Accessed February 30, 2013. 9. Melese, M., Alemayehu, W., Friedlander, E., & Courtright, P. (2004). Indirect costs associated with accessing eye care services as a barrier to service use in Ethiopia. Tropical Medicine and International Health, 9(3), 426–431. 10. Gyasi, M. E., Amoaku, W. M. K., & Asamany, D. K. (2007). Barriers to cataract surgical uptake in the Upper East Region of Ghana. Ghana Medical Journal, 41(4), 167–170. 11. Rabiu, M. M. (2001). Cataract blindness and barriers to uptake of cataract surgery in a rural community of Northern Nigeria. British Journal of Ophthalmology, 85(7), 776–780. 12. Zhang, X., Saaddine, J. B., Lee, P. P., Grabowski, D. C., Kanjilal, S., Duenas, M. R., et al. (2007). Eye care in the United States.Do we deliver to high-risk people who can benefit most from it? Archives of Ophthalmology, 125(3), 411–418. 13. Unite for Sight’s Evidence-based public health programs. http:// www.uniteforsight.org/start-a-chapter/best-practices. Date Accessed March 14, 2013. 14. Kyndt, M. (2001). Importance of affordable eye care. Journal of Community Eye Health, 14(37), 1–3. 15. Nirmalan, P. K., Katz, J., Robin, A. L., Krshnadas, R., Ramakrishnan, R., Thulasiraj, R. D., et al. (2004). Utilization of eye care services in rural south India: The Aravind comprehensive eye survey. British Journal of Ophthalmology, 88(10), 1237–1241. 16. Ekpenyong, B. N., & Ikpeme, B. M. (2009). Uptake of eye care services in university of Calabar Teaching Hospital, Cross River State, Nigeria. Journal of Nigerian Optometric Association, 15, 24–27. 17. Wang, J. J., Mitchell, P., & Smith, W. (1999). Use of eye care services by older Australians: The Blue Mountains Eye Study. Australia and New Zealand Journal of Ophthalmology, 27(5), 294–300. 18. Dandona, R., & Dandona, L. (2001). Review of findings of the Andhra Pradesh eye disease study: Policy implications for eyecare services. Indian Journal of Ophthalmology, 49(4), 215–234. 19. Bailey, R. N., Indian, R. W., Zhang, X., Geiss, L. S., Duenas, M. R., & Saaddine, J. B. (2006). Visual impairment and eye care among older adults-five states, 2005. Morbidity and Mortallity Weekly Reports, 55(49), 1321–1325.
123
J Community Health 20. Ellish, N. J., Royak-Schaler, R., Passmore, S. R., & Higginbotham, E. J. (2007). Knowledge attitudes and beliefs about dilated eye examinations among African Americans. Investigative Ophthalmology and Visual Science, 48(5), 1989–1994. 21. Geneau, R., Massae, P., Courtight, P., & Lewallen, S. (2008). Using qualitative methods to understand the determinants of patients’ willingness to pay for cataract surgery: A study in Tanzania. Social Science and Medicine, 66(3), 558–568. 22. Finger, R. P., Ali, M., Earnest, J., & Nirmalan, P. K. (2007). Cataract surgery in Andhra Pradesh state, India: An investigation into uptake following outreach screening camps. Ophthalmic Epidemiology, 14(6), 327–332. 23. Ntim-Amponsah, C. T., Amoaku, W. M. K., & Ofosu-Amaah, S. (2005). Alternate eye care services in a Ghanaian district. Ghana Medical Journal, 39(1), 19–23. 24. Ogwurike, S., & Pam, V. (2004). Means of transportation and its effects on eye care seeking behavior of patients in a rural setting. Annals of African Medicine, 3(2), 83–86. 25. Biska, T., Courtright, P., Geneau, R., Kasote, A., Chimombo, L., & Chirambo, M. (2009). Self-treatment of eye disease in Malawi. African Journal of Traditional Complementary and Alternative Medicine, 6(1), 23–29. 26. Dawodu, O. A., Okogie, O. H., & Ukponmwan, C. O. (2006). Pattern of utilization of eye care services in Benin City, Nigeria. Nigerian Postgraduate Medical Journal, 13(4), 323–325. 27. Courtright, P. (1995). Eye care knowledge and practices among Malawian traditional healers and the development of collaborative blindness prevention programs. Social Science and Medicine, 41(11), 1569–1575. 28. Courtright, P., Lewallen, S., & Kanjaloti, S. (1996). Changing patterns of corneal disease and associated vision loss at a rural
123
29.
30. 31.
32.
33.
34.
35.
36.
African hospital following a training program for traditional healers. British Journal of Ophthalmology, 80(8), 694–697. Kovai, V., Krishnaiah, S., Shamanna, B. R., Thomas, R., & Rao, G. N. (2007). Barriers to accessing eye care services among visually impaired populations in rural Andhra Pradesh, South India. Indian journal of ophthalmology, 55(5), 365–371. Ingrid, C. (2002). The patient’s view: How can we improve patient care? Journal of Community Eye Health, 15(41), 3–4. Informal consultation on analysis of blindness prevention outcomes. http://whqlibdoc.who.int/hq/1998/WHO_PBL_98.68. Date Accessed June 1, 2011. Keeffe, J. E., Weih, L. M., McCarty, C. A., & Taylor, H. R. (2002). Utilization of eye care services by urban and rural Australians. British Journal of Ophthalmology, 86(1), 24–27. Schaumberg, D. A., Christen, W. G., Glynn, R. J., & Buring, J. E. (2000). Demographic predictors of eye care utilization among women. Medical Care, 38(6), 638–646. Lee, D. J., Lam, B. L., Arora, B. S., Arheart, K. L., Mc-Collister, K. E., Zheng, D. D., et al. (2009). Reported eye care utilization and health insurance status among US adults. Archives of Ophthalmology, 127(3), 303–310. Robin, A. L., Nirmalan, P. K., Krishnadas, R., Ramkrishnan, R., Katz, J., Tielsch, J., et al. (2004). The utilization of eye care services by persons with glaucoma in rural south India. Transactions of American Ophthalmological Society, 102, 47–56. Orr, P., Barron, Y., Schein, O. D., Rubbin, G. S., & West, S. K. (1999). Eye care utilization by older Americans: The SEE project. Salisbury eye evaluation. Ophthalmology, 106(5), 904–909.
ORIGINAL PAPER
Determinants of Eye Care Utilization in Rural South-eastern Nigeria O. C. Arinze1 • B. I. Eze2 • N. N. Ude2 • S. N. Onwubiko2 • C. N. Ezisi1 C. M. Chuka-Okosa2
•
Ó Springer Science+Business Media New York 2015
Abstract To determine the barriers and incentives to eye care utilization (ECU) in Abagana, a rural south-eastern Nigerian community. The study was a population-based cross-sectional mixed method (quantitative and qualitative) survey of adult inhabitants of Abagana, in July–August, 2011. Data on respondents’ socio-demographics, barriers and incentives to ECU were collected, and analysed using descriptive and comparative statistics. A p \ 0.05 was considered statistically significant. Qualitative data were obtained from focus group discussions and in-depth interviews and analyzed using ATLAS.ti software. The 549 respondents (males 224) were aged 49.8 ± 15.9 SD years (range 18–93 years). Overall, orthodox eye care had ever been sought by 42.4 %, and 46.4 % of those who reported change in vision. Lack of awareness-31.8 %, cost-18.0 %, and fatalistic attitudes-15.9 % were the main ECU barriers. Possession of health insurance (OR 11.49; 95 % CI 4.21–31.34; p = 0.001), family history of eye disorder (OR 3.27, 95 % CI 2.03–5.26; p = 0.001), noticed change in vision (OR 11.30; 95 % CI 1.42–90.09; p = 0.022), current eye disease (OR 4.06; 95 % CI 2.29–7.19; p = 0.001) and systemic co-morbidity (OR 4.33; 95 % CI 2.67–7.02; p = 0.001) were the incentives to ECU. To enhance the low ECU in Abagana community, educational interventions on eye health maintenance and eye health seeking behaviours, and measures to reduce eye care cost are recommended. & B. I. Eze [email protected] 1
Department of Ophthalmology, Federal Teaching Hospital, Abakiliki, Ebonyi State, Nigeria
2
Department of Ophthalmology, University of Nigeria Teaching Hospital, Ituku-Ozalla 01139, Enugu, Nigeria
Keywords
Eye care utilization Barriers Incentives
Introduction Primary eye care (PEC) is the provision of essential, affordable, accessible, practical and sustainable eye healthcare to the general population [1]. PEC delivery utilizes the horizontal integration matrix model proposed by the World Health Organization (WHO) to incorporate PEC programs into the existing primary health care (PHC) structures. Eye care encompasses promotive, preventive, curative and rehabilitative services; its delivery could be institution- or community-based, or both [1]. The World Health Organization estimated the number of the world’s blind at 45 million and those visually impaired at 135 million [2]. In the Nigeria National Blindness Survey, avoidable causes, mainly cataract, accounted for 84 % of blindness [3]. However, in rural Nigeria, other prevalent eye diseases/disorders include glaucoma (21.1 %), allergic conjunctivitis (16.4 %), refractive errors (12.4 %), age-related macular degeneration (0.7 %) and corneal scarification (0.7 %) [4]. The survey further forecasted that, if the cataract surgical coverage remains unchanged, by the year 2020, the number of adults with operable cataract will increase by 43 % [3]. Of the global blindness burden, 90 % live in developing low and middle income countries (LMICs) where the resources and logistics for delivery of quality eye care are inadequate [2]. Consequently, regional [5, 6] and global [7] reports have observed that uptake of eye care services lags behind the established need for eye care. The factors responsible for initiation and sustenance of the gap between eye care need and uptake are multi-dimensional [8]. These include health system-dependent
123
J Community Health
factors like availability, and affordability of care; and patient-related factors like, gender, [9] awareness, attitudes to care, [10] direct [10, 11] and indirect [9] cost of care, and profile/severity of vision loss [9]. This needs-uptake gap accounts for non-detection of about half of the potentially blinding eye diseases, often with adverse consequences for mortality, morbidity and quality of life, among visually handicapped persons [11–13]. To clear the backlog, and treat incident cases of eye diseases, ‘VISION 2020: The Right to Sight’, the Global Initiative for the Elimination of Avoidable Blindness, has endorsed the need to reverse the increasing trend of global blindness through unrestricted access to eye care services [14]. Achieving the goals of this global initiative largely depends on the ability to identify and overcome barriers, and augment/expand the incentives, to utilization of eye care services [7]. Consequently, this survey, a mixed-methods quantitative and qualitative, cross-sectional population-based study, was conducted on the inhabitants of Abagana, a rural underserved south-eastern Nigerian community in July– August, 2011. It aimed to assess their utilization of, determine barriers to, and identify characteristics associated with utilization of available orthodox eye care services. The results, in addition to providing comparative loco-regional data, would assist eye care planners, and implementers, and health policy makers, in the study area and similar settings elsewhere, in strategizing for realization of the goals of ‘VISION 2020: The Right to Sight’.
-ophthalmologists, optometrists and nurses; and complemented by alternative/traditional medicine practitioners and patent medicine dealers. There are numerous public and privately-owned health care facilities. Of the public facilities, only Abagana Comprehensive Health Centre has a dedicated Eye Unit which provides medical, refraction and minor surgical eye care services to the inhabitants of the community and beyond. The centre is manned by visiting ophthalmologist and optometrist, and resident nurses and midwives. There is no privately-owned eye care facility in Abagana. This was a population-based qualitative and quantitative cross-sectional study conducted from July to August, 2011. Eligibility The participants comprised adult males and females, aged 18 years or older, who had resided continuously in Abagana for at least 6 months, who voluntarily consented to participation. Ethics Prior to commencement of study, ethics clearance consistent with the 1964 Helsinki Declaration, as revised in 2008, was obtained from the Medical and Health Research Ethics Committee (Institutional Review Board) of the University of Nigeria Teaching Hospital, Enugu. Study Design
Methods
Population-based quantitative and qualitative survey.
Background
Sample Size and Sampling
Anambra state, along with the states of Abia, Enugu, Ebonyi and Imo make up Nigeria’s south-east geo-political zone, one of Nigeria’s six geo-political divisions. The state is sub-divided into administrative sub-units or local government areas (LGAs). Each LGA is further sub-divided into political Wards; each ward is made up of villages. Abagana is one of the 6 rural communities that make up Njikoka LGA of Anambra state. It lies in the tropical rainforest climatic belt and has a landmass of approximately 9.2 km2. Based on the 2006 Nigerian National Census, the projected population of Abagana is 36,572. The inhabitants of Abagana are predominantly ethnic Ibos, whose main occupations are farming, trading, civil service and artisanship. In Nigeria, the eye care system comprises tertiary, secondary and primary levels of eye care. In Abagana, a rural community served by primary level eye care, eye care is provided mainly by orthodox eye care providers
The minimum sample size, n, was calculated using Fisher’s formula for sample size determination for a definite population i.e. n = [Z2a pq]/d2 when the population is more than 10,000. Where n = the minimum sample size; p = previously reported prevalence of 61 % in a related Nigerian report by Rabiu et al. [10] thus giving a prevalence (p) value of 0.61, Za = standard normal deviation of 1.96, at confidence level of 95 %, d = desired precision due to random sampling error of 5 % = 0.05, q = 1-p; [(1.96)2 9 0.61 9 0.39]/ 0.052 = 0.9139/0.0025 = 366. The calculated sample size of 366 was multiplied by 1.5, the design effect due to cluster sampling method, to give a modified sample size of 549. The sampling method was multi-stage cluster random sampling technique. The 10 villages that make up Abagana are already divided into four political Wards. The Wards served as clusters. The Wards and their component villages, and populations are: Ward 1: Adaegbe-Umudunu
123
J Community Health
and Uru-Okpala; population = 8071; Ward 2: EnuoraOraofia and Adaegbe-Oraofia; population = 8829; Ward 3: Enuora-Umudunu and Amagu; population = 7566; Ward 4 : Akpu, Amaenyi, Uru, and Adegbe; population = 12,106. From the proportion of the 549 participants, the modified sample size, to be recruited from each ward was calculated based on proportionate representation:—Ward 1 = 122; Ward 2 = 132; Ward 3 = 114; and Ward 4 = 181. Using a simple random sampling, the following villages were selected from their respective clusters (Wards). In each of the selected villages, all households were enumerated to get a sampling frame: Ward 1 = Uru-Okpara, 709 households; Ward 2 = Enuora-Oraofia, 728; Ward 3 = Enuora-Umudunu, 698; Ward 4 = Akpu, 463 and Adaegbe, 485. Using a systematic random sampling, k (sampling interval, skip) was calculated as k = N/n, where N = the population size of each ward and n = sample size i.e., the calculated proportion of the 549 participants to be recruited for each ward. Subsequently, starting from a randomly selected household and sampling direction, every sixth household in Uru-Okpara, Enuora-Oraofia and EnuoraUmudunu, and every fifth in Akpu and Adaegbe were selected until the required sample size of 549 households was obtained. From each of the selected households, one eligible adult was recruited by balloting. Any household in which nobody or eligible participant was around/available for recruitment or in which none gave voluntary informed consent was skipped without revisiting the household. A total of 12 focus group discussions (FGDs) comprising 6 all-male and 6 all-female groups, with 7 participants in each group, were conducted in the four Wards/clusters. In each ward, 3 FGDs, comprising at least 1 all-male and one all-female FGD, were conducted. The criteria for participation in FGD were age greater than 18 years, uninterrupted residence in the study community for the past 6 months and non-participation in the preceding questionnaire-based arm of the same survey. Each Ward/cluster comprised 2 villages, except Ward 4 which has 4 villages. For Wards 1, 2 and 3 which have 2 villages each, 1 village was randomly selected for male FGD and the remaining village for female FGD. Afterwards, the decision on whether to perform male or female FGD, to complete the three mandatory FGDs per ward, was reached by simple balloting (male vs. female). The village, within each ward, from where the participants in the supplementary i.e., third FGD were drawn, was similarly determined. In Ward 4, comprising 4 villages, the village for male FGD was randomly selected; this was followed by another random selection of another village, from the remaining 3 villages, for female FGD. The gender composition (male vs. female) and hosting village, for the third FGD, was each determined by simple balloting. The participants election procedure is illustrated in the flow chart below (Fig. 1).
Additionally, in-depth interviews (IDIs) were also conducted on the Igwe (Traditional ruler) of Abagana, President of Abagana Town Union, Chairman of Ward 4 (randomly selected from the four wards) and 3 health personnel working at the Comprehensive Health Center, Abagana.
Data Collection Procedures This was carried out by the researchers who are fluent in Igbo, the native language of the Abagana community, and English languages. Quantitative Data The study instrument, developed de-novo, was a pre-tested researcher-administered open- and closed-ended questionnaire with sections on participants’ socio-demographics (age, sex, education, marital status and occupation), clinical characteristics, pattern of eye care service utilization, barriers and incentives to eye care utilization. To ascertain its construct validity and psychometric reliability, the study questionnaire was pretested on a rural population, outside the study area, which share similar settings/characteristics as the study population. The final version of the questionnaire had internal consistency reliability [Kronbach alpha (a) coefficient of 0.8 and Lawshe’ s content validity index of 0.7]. The barriers to obtaining regular eye care services were explored among those who noticed a change in vision and did not seek orthodox eye care, and those who did not notice any change in vision but did not go for routine eye checks. When multiple reasons were given by the participants, the most important reason, as perceived by the subjects, was marked with a separate code. Qualitative Data The FGDs and IDIs were conducted by the researchers, using guides for FGDs and IDIs, while a trained research assistant took notes. The guides contained diagnostic questions on some issues raised in the questionnaire study. This was meant to provide in-depth knowledge of the dynamics in decision-making and response patterns with regards to barriers and incentives to eye care service utilization. The services of a sociologist were utilized in the development and administration of the FGD and IDI guides, as well as the analysis of the generated data. A total number of 12 FGDs were conducted with 7 persons in each group. In each ward, FGDs were held with representatives from 3 age groups (young adults aged 18–35, adults aged 36–59 and elderly people aged 60 or older). The moderator
123
J Community Health
conducted FGDs with sex-homogenous groups (i.e., each group comprising entirely of males or females), in culturally appropriate locations chosen by the participants. IDIs were conducted with health workers, and the opinion leaders of the community who are the custodian of the community’s norms, cultures and traditions. The language of administration of the survey was Igbo language, the native language of the host study community. Same sex FGDs and IDIs were conducted to overcome any potential sex-sensitive cultural inhibition to volunteering critical information relevant to the study. All the FGD and IDI sessions were recorded on tape.
Data Analysis Quantitative Analysis All data were cleaned and edited, double-entered into and analyzed using the statistical package for social sciences
Fig. 1 Flow chart illustrating the selection procedure for FGD participants
(SPSS) software, version 18.0 (SPSS Inc, Chicago, Illinois, USA). Data were subsequently categorized, and subjected to descriptive statistical evaluation to yield frequencies, percentages and proportions. Bivariate comparisons for statistical significance of observed betweenclass differences were performed using Chi square (v2) test for categorical variables and Student t test for continuous or metric variables. In all bivariate comparisons, a p \ 0.05 and the associated crude odds ratios and 95 % CI were considered statistically significant. In cases of multi co-linearity, all the significant (p \ 0.05) predictor variables were selected and fed into a multivariate logistic regression model to determine the independent effect of each predictor on the outcome of interest, eye care utilization. Adjusted (cofounder-corrected) odds ratios (OR) and their 95 % CIs, associated with p \ 0.05, were considered statistically significant. In the multivariate model, quantification of the effect size (effect size statistic) of each predictor variable was indicated by the associated odds ratio.
Total number of potential eligible participants that met the inclusion criteria = 143
Number that consented to parcipate= 120
Total number of randomly selected parcipants in the 4 wards, ensuring gender balance and proporonate distribuon by ward/village = 84
Ward 1 3 FGDs; 21 parcipants
123
Ward 2 3FGDs; 21 parcipants
Number that declined parcipaon= 23
Number not selected = 36
Ward 3 3 FGDs; 21 parcipants
Ward 4 3 FGDs; 21 parcipants
J Community Health
Qualitative Analysis Analysis of the qualitative data placed emphasis on the interpretation, description and recording/writing of what was actually said. The responses were translated into English language. In going through the transcription, phrases with contextual or special connotations were noted and pulled out as illustrative quotes for complementing the quantitative data. To do this, relevant themes were developed for the coding and sorting of the qualitative data. To facilitate the qualitative data processing, each of the transcripts was entered into a word processor and saved as a rich text format (RTF) file, which was formatted and coded using the ATLAS.ti. This was utilized in generating illustrative quotes from the qualitative data to support the quantitative data.
Results Participants’ Socio-Demographics There were 549 participants comprising 224 (40.8 %) males and 325 (59.2 %) females (male: female ratio = 1:1.5) who were aged 49.8 ± 15.9 SD years (range 18–93 years). The majorities of the participants were married—370 (67.4 %), possessed formal education—459 (83.6 %) and were frequently traders—138 (25.1 %). The participants’ demographic characteristics are shown in Fig. 2 while their socioeconomic profile is reported in Table 1. Clinical Profile Of the 549 participants, 233 (42.4 %) had ever utilized eye care services while 316 (57.6 %) had not. Five hundred (91.1 %) participants had ever noticed a change in their vision while 49 (8.9 %) had not. Of the 500 who noticed change in their vision, 232 (46.4 %) did, while 268 (53.6 %) did not, access orthodox eye care services as a result. Among the 49 participants who did not notice any
Table 1 Socio-economic profile of respondents Characteristics
Total (%) n = 549
Marital status Married Single Divorced/separated Widowed
370 (67.4) 88 (16.0) 2 (0.4) 89 (16.2)
Educational status Primary Secondary
192 (35.0) 173 (31.5)
Tertiary
94 (17.1)
None
90 (16.4)
Occupation Civil servant
130 (23.7)
Trading
138 (25.1)
Artisan
74 (13.5)
Farming
109 (19.9)
House wife
22 (4.0)
Others
76 (13.8)
change in their vision, only 1/49 (2.0 %) went for routine eye check (ocular health maintenance visits) by an orthodox eye care providers while 48 (98.0 %) did not. The main barriers to visual symptom-initiated eye care utilization were lack of awareness—85 (31.7 %) of available eye care services, cost—48 (17.9 %) and attitudinal barrier i.e., ‘I thought it was a minor thing’—42 (15.7 %). The participants’ leading barriers to routine eye check were selfassessment of one’s eyes being okay i.e., ‘my eyes are okay; no reason to go’-20 (41.7 %) and cost-8 (16.7 %)—Table 2. In the bivariate analysis, several of participants’ sociodemographic (age, sex, formal education, employment status and possession of health insurance) and clinical (noticed change in vision, current eye disease, family history of eye disease, non-ocular co-morbidity) were significantly (p \ 0.05) predictive of eye care service utilization, Table 3. However, in the multivariate logistic model, possession of health insurance, noticed change in vision, current eye disease, family history of eye disease and non-ocular co-morbidity were retained as significant independent predictors of eye care utilization, Table 4. Qualitative Results Focus Group Discussions
Fig. 2 Age-sex distribution of 549 participants
A total of 84 persons participated in the FGDs. On barriers to eye care services, 81 participants made comments that were grouped as lack of awareness (N = 24), cost (N = 21), attitudinal (N = 16), preference for alternative
123
J Community Health Table 2 Barriers to eye care services utilization among those who noticed a change in their vision but did not seek orthodox eye care and those who noticed no change in vision but did not seek routine eye check Barrier
Noticed change in their vision n (%) n = 268
Did not notice any change in their vision n (%) n = 48
Personal barriers Lack of awareness of eye care services
85 (31.7)
4 (8.3)
I thought it was a minor thing I prefer alternative eye care services
42 (15.7) 28 (10.5)
0 (0.0) 2 (4.2)
Fear of blindness resulting from treatment
28 (10.5)
5 (10.4)
My eyes are okay; no reason to go
11 (4.0)
20 (41.7)
7 (2.6)
0 (0.0)
48 (17.9)
8 (16.7)
I did not know that my eye problem can be treated Socio-economic barriers No money to pay for eye care services Poor vision is normal for old people like me
5 (1.9)
0 (0.0)
It is God’s will that my vision be bad
4 (1.5)
0 (0.0)
I am very busy I don’t have time to go
3 (1.1)
4 (8.3)
The health centre is too far from my house
3 (1.1)
4 (8.3)
I have nobody to escort me to the clinic
2 (0.8)
1 (2.1)
2 (0.8)
0 (0.0)
It is against our culture to have eye operation
eye cares services (N = 12) and fear (N = 8) barriers. As regards incentives, 68 participants made comments. Incentives mentioned included: having a health insurance policy (N = 26), noticing a change in vision (N = 22), and having an eye disease (N = 20). Few (N = 5) stated that having a first degree relative with an eye disease motivated them to see an eye doctor.
Discussion The participants comprised more females than males, often aged 40 years or older, who were frequently married, possessed formal education and had civil service and trading as their main occupations. The observed socio-demographic profile is similar to findings reported in Kaduna, Nigeria [11] and India [15]. These and our study were all conducted in rural communities where the majority of the men in the working age group are likely to be out, working in the cities, while the women are left to take care of the households, thus making them more available to participate in the study. The participants’ overall level of utilization of orthodox eye care services is low; utilization was more often symptom-initiated than asymptomatic routine eye checks. The findings in this present study are similar to findings in Calabar, Nigeria, [6] but contrast the report from Australia [17]. The Calabar, [6] Nigeria, report was on patients presenting to the hospital; as such, it likely addressed symptom-initiated hospital visits only, the dominant cause of consultation in the present report. The participants in the
123
Australian survey belong to a comparatively higher socioeconomic group, thus eliminating a critical health care access barrier, cost. These findings underscore the need to encourage timely presentation of symptomatic patients, and emphasize the constant necessity for routine eye checks, even in the absence of eye symptoms. Poor awareness of availability of eye care services, fatalistic attitudes, cost and preference for alternative eye care were the participants’ main barriers to accessing eye care services. This is similar to the findings in Calabar, [16] Nigeria, where lack of awareness constituted a major barrier to utilization of eye care services. Indeed, many of the participants from the FGDs said ‘we were not aware a hospital existed there, let alone the availability of eye care services.’ This could be explained by the location of the health centre at the border of Abagana with its neighboring community Enugwu-Ukwu. This location is quite distant from major residential areas of the community. Also, poor publicity, by the hospital, of its eye care services could be contributory. In contrast, a related Ghanaian study [10] reported a high level of awareness among its study participants. This could be explained by the observation from the Ghanaian study that majority of the cataract patients used for the study could access services within a short travelling time from their communities to outreach centers and as such had reasonably easier spatial access to eye care services. This, in addition to contributing directly to service underutilization, might encourage the villagers to patronize alternative eye care providers often with disastrous visual
J Community Health Table 3 Characteristics predictive of participants’ utilization of orthodox eye care services Characteristic
Total (%) n = 549
Utilisation of eye care services Yes n = 233
Odds ratio
95 % CI
p value
0.23–0.52
0.001*
No n = 316
Age (years) C40
375 (68.3)
188 (80.7)
187 (59.2)
0.35
\40 Gender
174 (31.7)
45 (19.3)
129 (40.8)
Reference
Female
325 (59.2)
152 (65.2)
173 (54.7)
0.65
0.46–0.91
0.017*
Male
224 (40.8)
81 (34.8)
143 (45.3) 1.1
0.69–1.74
0.761
1.99
1.31–3.33
0.01*
41.55
5.69–303.38
0.0001*
8.04
4.96–13.05
0.001*
5.4
3.63–8.05
0.001*
7.85
5.16–11.95
0.001*
9.73
4.49–21.04
0.001*
Possess formal education Yes
459 (83.6)
193 (82.8)
266 (84.2)
No
90 (16.4)
40 (17.2)
50 (15.8)
Employed
268 (48.8)
209 (89.7)
59 (18.7)
Unemployed
281 (51.2)
24 (10.3)
257 (81.3)
232 (99.6)
268 (84.8)
Employment status
Noticed a change in vision Yes
500 (91.1)
No
49 (8.9)
1 (0.4 %)
48 (15.2 %)
Has an eye disease currently Yes
378 (68.9)
No 171 (31.1) First degree relation has an eye problem
210 (90.1)
168 (53.2)
23 (9.9)
148 (46.8)
Yes
165 (30.1)
116 (49.8)
49 (15.5)
No
384 (69.9)
117 (50.2)
267 (84.5)
Has a non-ocular health problem Yes
164 (29.9)
124 (53.2)
40 (12.7)
No
385 (70.1)
109 (46.8)
276 (87.3)
Possess health insurance Yes
55 (10.0)
47 (20.2)
8 (2.55)
No
494 (90.0)
186 (79.8)
308 (97.5)
*Significant (p \ 0.05)
Table 4 Significant incentives to eye care service utilization on multivariate analysis
Incentive
Adjusted OR
95 % CI
p value
Noticed a change in my vision
11.30
1.42–90.09
0.022
Has an eye disease
4.06
2.29–7.19
0.001
1st degree relation has an eye problem
3.27
2.03–5.26
0.001
Has another non- ocular health problem Has a health insurance policy
consequences, and avoidable increase in the burden of eye diseases and blindness. To overcome the awareness barrier, the investigators recommend community-based eye health education utilizing culture- and settings-appropriate information dissemination channels. The education program should indicate the scope of available eye care services; and highlight the efficiency, efficacy, and cost-effectiveness advantage of orthodox versus alternative eye care. The prominent barrier role of cost in the present study is
4.33
2.67–7.020
0.001
11.49
4.21–31.34
0.001
consistent with the findings in developing [10, 18] and developed [19, 20] countries but differs from studies [21, 22] elsewhere where fear of surgery and treatment outcome were the major barriers. Indeed, many from the FGDs said ‘it is because we have no money and do not know where to go for our eye problem that we are forced to go to the traditional healers’. Many also said ‘‘we are unable to utilize the eye care services because they are not affordable’’. This has critical implications for presentation for
123
J Community Health
care and its timeliness, and by extension care outcomes, and might encourage recourse to hazard-prone cheaper alternatives to allopathic eye care services. This has a collateral adverse impact on the burden of blindness, which, if due to alternative medicine use, might be irreversible. The present data suggest the need for grass root economic empowerment of the study population and enhanced affordability of eye care services through service subsidy and provision of health insurance. The preference for alternative eye care services, self eye medication inclusive, by an appreciable proportion of the study participants corroborates the findings in Ghana, [23] Kaduna, Nigeria, [24] and Malawi [25]. This similarity could be explained by the relatively easy accessibility and affordability of these alternative eye care services to the community members. However, the findings of our study differ from the relatively strict preference for orthodox eye care reported by Dawodu et al. [26] in Benin, Nigeria. The difference between the present result and Dawodu et al’s. [26] could be attributed to the nature of the Benin study, which was an urban-based survey conducted on educated staff of state Ministries in Benin, who, considering their level of education were more likely to seek orthodox eye care services. In addition, information from the IDIs of the health workers suggests that inadequate infrastructure, equipment and skilled manpower results in weakness of the orthodox eye care system. This makes alternative eye care services appear more dependable to the community members. This finding underscores the need for strengthening the resource base of the eye care system; and expanding the scope of the earlier suggested eye health education to include the dangers associated with patronage of alternative eye care services. Alternatively, since they are communitybased and comparatively more accessible, these alternative eye care providers could be trained and incorporated into orthodox eye care system to provide supervised eye care. This approach has been shown to be feasible and effective in Malawi by Courtright [27] and Courtright et al. [28]. The barrier imposed by the fatalistic attitudes of the present survey participants towards eye diseases, similarly reported by Kovai et al. [29] and Bailey et al. [30] has enormous implications for care of initially asymptomatic, but potentially blinding, eye diseases like glaucoma. Therefore, preventive educational interventions on eye health should highlight this peculiar group of eye diseases and the need for their early diagnosis through regular eye health checks irrespective of presence of eye symptoms. Fear of unfavorable treatment outcomes constituted a barrier amongst participants who noticed change in vision but did not access eye care services. This has been previously reported in Tanzania [21] and India [22]. Especially in surgical intervention like cataract surgery, a single treatment failure needs about five well-advertised success
123
cases to neutralize its demotivating effects on the community [30]. Educating eye care consumers on the positive prognostic value of early presentation, other determinants of medical and surgical treatment outcomes and the comparatively higher risks of poor outcomes associated with unorthodox eye care interventions are suggested to dispel this fear. Towards this end, WHO further recommends better cataract monitoring and evaluation system against recommended guidelines and preferred practice patterns in order to improve visual outcome [31]. Possession of health insurance, but not age, gender, educational or employment status, was the only socio-demographic characteristic independently predictive of eye care utilization. The present result corroborates an Australian [32] report but differs from the findings of Schaumberg et al. [33] and Nirmalan [15] and associates. The positive predictive role of health insurance is consistent with the earlier identification of cost in our study as a major barrier to eye care utilization as observed in an American study [34]. This, provides suggestive evidence that the study population is not adequately empowered financially to bear the direct and indirect costs of orthodox eye care. The present data suggests further strengthening, and widening of the coverage of, health insurance policy in the study area, and under similar settings elsewhere. The curious lack of association between formal education and eye care utilization challenges the findings from previous studies [29, 35] and suggests gross deficiency of health education content of the formal educational training curriculum. As such, possession of formal education did not confer any health literacy advantage. However, the overriding influence of other barrier factors, especially cost barrier, could have masked the incentivizing effect of formal education. Consequently, curricular review, at various levels of education, with appropriate inputs from health educators, is instructive. Also contrary to previous reports, [35, 36] older age was not associated with significant increase in eye care utilization despite the established higher prevalence of agerelated eye diseases in this population sub-group. This observation could be attributed to age-related comprise of physical/financial independence and loss of autonomy on health-seeking behaviors/decisions. Many in this age group, with increasing age become progressively dependent on their immediate and extended family members and may not be able to seek orthodox eye care at will when needed, without assistance. To enhance the utilization of eye care services by this vulnerable age group, the authors suggest broadening of the scope of health insurance to include non-civil servants, and effective implementation of retirement and pension schemes to engender some degree of financial empowerment and autonomy, enough to carter for health care needs of people in this age group.
J Community Health
Institutions, specially equipped to take care of the elderly, should also be established. Eye care service utilization was significantly associated with noticed change in vision, current eye disease, family history of eye disease and concurrent non-ocular co-morbidity. Similar patterns of symptom-determined eye care service utilization have been found in India [28] and Ethiopia [9] and probably reflect the general trend in LMICs where limited financial resource and poor needs awareness restrict practice of routine/asymptomatic eye check. Again, this further reinforces the earlier introduced rationale for emphasizing this aspect-specific area of promotive and preventive interventions, during eye health education campaigns. Additionally, concurrent non-ocular co-morbidity and having a first degree relation with an eye problem were significantly associated with having seen an eye doctor. As suggested by Wang et al. [17], fear of developing similar eye conditions, as their first degree relatives, and the likelihood of the co-morbid systemic diseases conditions developing ocular complications could have caused these participants to seek orthodox eye care. Despite the advantage of its population-based mixedmethods design, the conclusions drawn from this study are limited by intrinsic bias inherent in participant-reported information. Also, the study setting was a rural area; so, the findings might not generalize to urban populations. However, the findings will potentially inform eye care practice, programmes and policy changes in similar settings in Nigeria, and other LMICs elsewhere.
Conclusions In Abagana, south-east Nigeria, the utilization of available eye care services is poor. The main barriers to eye care service utilization are poor awareness of the service availability, attitudinal barriers and cost. Possession of health insurance, family history of eye disease, noticed change in vision, current eye disease and non-ocular comorbidity are the incentives to service utilization. To increase eye care access, awareness creating measures, eye health education and enhanced service affordability are recommended. Towards these ends, eye care worker-led community-based public eye health education and tailored changes in rural eye care system, to overcome access barriers, are instructive. Future researches on patterns and determinants of eye care seeking behaviors among rural and underserved populations are warranted. Acknowledgments The authors wish to acknowledge the assistance of inhabitants of Abagana community during the study. Conflict of interest
None declared.
References 1. Eze, B. I., & Maduka-Okafor, F. C. (2009). An assessment of the eye care workforce in Enugu state, south-eastern Nigeria. Human Resources for Health, 7, 38. 2. Unite For Sight, Inc. Global Eye Health Statistics. http://www. uniteforsight.org/eye_stats.php. Date Accessed April 20, 2013. 3. The Nigeria national blindness and visual impairment survey 2005–2007. http://pbunion.org/Countriessurveyresults/Nigeria/ Nigeria_survey_Summary_report.pdf. Date Accessed April 7, 2013. 4. Adegbehingbe, B. O., & Majengbasan, T. O. (2007). Ocular health status of rural dwellers in south-western Nigeria. Australian Journal of Rural Health, 15(4), 269–272. 5. Whitworth, J. (1999). Determinants of attendance and patient satisfaction at eye clinics in south-western Uganda. Health Policy and Planning, 14(1), 77–81. 6. Lewallen, S., & Courtright, P. (2006). Increasing uptake of eye services by women. Journal of Community Eye Health, 19(60), 59–60. 7. Donoghue, M. (1999). People who don’t use eye services: Making the invisible visible. Journal of Community Eye Health, 12(31), 36–38. 8. Community Eye Health Online Course, Module 2: Patient barriers to eye care. http://www.uniteforsight.org/community-eyehealth-course/module2. Date Accessed February 30, 2013. 9. Melese, M., Alemayehu, W., Friedlander, E., & Courtright, P. (2004). Indirect costs associated with accessing eye care services as a barrier to service use in Ethiopia. Tropical Medicine and International Health, 9(3), 426–431. 10. Gyasi, M. E., Amoaku, W. M. K., & Asamany, D. K. (2007). Barriers to cataract surgical uptake in the Upper East Region of Ghana. Ghana Medical Journal, 41(4), 167–170. 11. Rabiu, M. M. (2001). Cataract blindness and barriers to uptake of cataract surgery in a rural community of Northern Nigeria. British Journal of Ophthalmology, 85(7), 776–780. 12. Zhang, X., Saaddine, J. B., Lee, P. P., Grabowski, D. C., Kanjilal, S., Duenas, M. R., et al. (2007). Eye care in the United States.Do we deliver to high-risk people who can benefit most from it? Archives of Ophthalmology, 125(3), 411–418. 13. Unite for Sight’s Evidence-based public health programs. http:// www.uniteforsight.org/start-a-chapter/best-practices. Date Accessed March 14, 2013. 14. Kyndt, M. (2001). Importance of affordable eye care. Journal of Community Eye Health, 14(37), 1–3. 15. Nirmalan, P. K., Katz, J., Robin, A. L., Krshnadas, R., Ramakrishnan, R., Thulasiraj, R. D., et al. (2004). Utilization of eye care services in rural south India: The Aravind comprehensive eye survey. British Journal of Ophthalmology, 88(10), 1237–1241. 16. Ekpenyong, B. N., & Ikpeme, B. M. (2009). Uptake of eye care services in university of Calabar Teaching Hospital, Cross River State, Nigeria. Journal of Nigerian Optometric Association, 15, 24–27. 17. Wang, J. J., Mitchell, P., & Smith, W. (1999). Use of eye care services by older Australians: The Blue Mountains Eye Study. Australia and New Zealand Journal of Ophthalmology, 27(5), 294–300. 18. Dandona, R., & Dandona, L. (2001). Review of findings of the Andhra Pradesh eye disease study: Policy implications for eyecare services. Indian Journal of Ophthalmology, 49(4), 215–234. 19. Bailey, R. N., Indian, R. W., Zhang, X., Geiss, L. S., Duenas, M. R., & Saaddine, J. B. (2006). Visual impairment and eye care among older adults-five states, 2005. Morbidity and Mortallity Weekly Reports, 55(49), 1321–1325.
123
J Community Health 20. Ellish, N. J., Royak-Schaler, R., Passmore, S. R., & Higginbotham, E. J. (2007). Knowledge attitudes and beliefs about dilated eye examinations among African Americans. Investigative Ophthalmology and Visual Science, 48(5), 1989–1994. 21. Geneau, R., Massae, P., Courtight, P., & Lewallen, S. (2008). Using qualitative methods to understand the determinants of patients’ willingness to pay for cataract surgery: A study in Tanzania. Social Science and Medicine, 66(3), 558–568. 22. Finger, R. P., Ali, M., Earnest, J., & Nirmalan, P. K. (2007). Cataract surgery in Andhra Pradesh state, India: An investigation into uptake following outreach screening camps. Ophthalmic Epidemiology, 14(6), 327–332. 23. Ntim-Amponsah, C. T., Amoaku, W. M. K., & Ofosu-Amaah, S. (2005). Alternate eye care services in a Ghanaian district. Ghana Medical Journal, 39(1), 19–23. 24. Ogwurike, S., & Pam, V. (2004). Means of transportation and its effects on eye care seeking behavior of patients in a rural setting. Annals of African Medicine, 3(2), 83–86. 25. Biska, T., Courtright, P., Geneau, R., Kasote, A., Chimombo, L., & Chirambo, M. (2009). Self-treatment of eye disease in Malawi. African Journal of Traditional Complementary and Alternative Medicine, 6(1), 23–29. 26. Dawodu, O. A., Okogie, O. H., & Ukponmwan, C. O. (2006). Pattern of utilization of eye care services in Benin City, Nigeria. Nigerian Postgraduate Medical Journal, 13(4), 323–325. 27. Courtright, P. (1995). Eye care knowledge and practices among Malawian traditional healers and the development of collaborative blindness prevention programs. Social Science and Medicine, 41(11), 1569–1575. 28. Courtright, P., Lewallen, S., & Kanjaloti, S. (1996). Changing patterns of corneal disease and associated vision loss at a rural
123
29.
30. 31.
32.
33.
34.
35.
36.
African hospital following a training program for traditional healers. British Journal of Ophthalmology, 80(8), 694–697. Kovai, V., Krishnaiah, S., Shamanna, B. R., Thomas, R., & Rao, G. N. (2007). Barriers to accessing eye care services among visually impaired populations in rural Andhra Pradesh, South India. Indian journal of ophthalmology, 55(5), 365–371. Ingrid, C. (2002). The patient’s view: How can we improve patient care? Journal of Community Eye Health, 15(41), 3–4. Informal consultation on analysis of blindness prevention outcomes. http://whqlibdoc.who.int/hq/1998/WHO_PBL_98.68. Date Accessed June 1, 2011. Keeffe, J. E., Weih, L. M., McCarty, C. A., & Taylor, H. R. (2002). Utilization of eye care services by urban and rural Australians. British Journal of Ophthalmology, 86(1), 24–27. Schaumberg, D. A., Christen, W. G., Glynn, R. J., & Buring, J. E. (2000). Demographic predictors of eye care utilization among women. Medical Care, 38(6), 638–646. Lee, D. J., Lam, B. L., Arora, B. S., Arheart, K. L., Mc-Collister, K. E., Zheng, D. D., et al. (2009). Reported eye care utilization and health insurance status among US adults. Archives of Ophthalmology, 127(3), 303–310. Robin, A. L., Nirmalan, P. K., Krishnadas, R., Ramkrishnan, R., Katz, J., Tielsch, J., et al. (2004). The utilization of eye care services by persons with glaucoma in rural south India. Transactions of American Ophthalmological Society, 102, 47–56. Orr, P., Barron, Y., Schein, O. D., Rubbin, G. S., & West, S. K. (1999). Eye care utilization by older Americans: The SEE project. Salisbury eye evaluation. Ophthalmology, 106(5), 904–909.
