Effects of a community-based health education intervention on eye health literacy of adults in Vietnam

ABSTRACT

Health education interventions are more commonly evaluated in hospitals or schools but rarely in the community. The purpose of this study was to assess the effects of a community-based eye health education intervention on the eye health literacy in an adult Vietnamese population. Four hundred households from two districts of Ba Ria-Vung Tau province were selected by multistage systematic random sampling. A pretested questionnaire was administered at baseline and re-administered, after an eye health education intervention in two of the four communes. The other two communes acted as controls. The intervention included eye health education through community presentation, brochures, posters and loudspeaker. Fisher’s exact test and logistic regression were used for statistical analysis. A total of 400 adults (mean age: 51.5 ± 14.5 years; range, 30–90 years) participated in the baseline survey and the repeat survey. Participants in the intervention group showed significantly greater awareness and knowledge of red eye preventive measures (Odds ratio range 2.1 to 4.1, p = 0.03 to 0.001) compared to control group. Participants in the intervention group were more than twice as likely to have heard of cataract (OR 2.3, p = 0.008), and more than three times as likely to be aware of cataract surgery (OR 3.1, p = 0.003) and know that the cataract blindness is reversible (OR 3.3, p = 0.002), than the control group. Multiple logistic regression showed that location, education, and eyeglasses ownership were the major factors associated with disease awareness. The eye health education intervention successfully increased awareness and knowledge of red eye and cataract in the intervention community.

KEYWORDS:

• Awareness
• eye health literacy
• promotion
• intervention
• Vietnam

Introduction

Vision impairment is a major avoidable public health problem worldwide. The prevalence of vision impairment and blindness in Vietnam is 11.4% and 1.8%, respectively (Ministry of Health 2015). Cataract and refractive error are the two main causes of vision impairment (Ministry of Health, Medical service administration, Vietnam 2015). Whilst virtually everyone over the age of 50 years requires some form of vision correction (Charman 2008), only about 39% of people over 50 years have ever had an eye examination in Vietnam (Vela, Samson, and Zunzunegui et al. 2012). Studies have reported that the low service utilisation is mainly linked to low health literacy, poor access to services and the associated costs (Dandona et al. 2001; Shrestha et al. 2014; Thapa, Berg, and Khanal et al. 2011; Huang et al. 2013; Livingston, McCarty, and Taylor 1998; Foreman et al. 2018; Marmamula et al. 2014). Studies have also documented an increase in health awareness and uptake of eye care services in communities after health promotion initiatives (Muller, Keeffe, and Taylor 2007; Baker and Murdoch 2008; Hobday, Ramke, and du Toit 2011; Banzi 2007; Forst, Lacey, and Chen et al. 2004).

Vietnam has a population of over 96 million, constituted of 54 different ethnic groups and spread over 6 distinct ecological zones, and has rapidly developed into a low-middle income country over the past 20 years. Vietnam reported significant decrease in cataract blindness between 2000 and 2015; however, vision impairment due to cataract has increased indicating significant cataract surgery backlog (Ministry of Health, Medical service administration, Vietnam 2015). Vision impairment due to uncorrected refractive errors, diabetic retinopathy, glaucoma and age-related macular degeneration have not reduced.

We conducted a baseline eye health literacy study in Vietnam that revealed a low uptake of services and this was attributed to low awareness of eye health conditions and treatment (Paudel, Kovai, and Naduvilath et al. 2016). This study indicated the need for health promotion interventions at the community level. Health promotion is an effective strategy for improving health awareness of people, changing their health-seeking behaviour and increasing uptake of eye care services (Hubley and Gilbert 2006; Dineen 1999). Brien Holden Vision Institute and Ba Ria Vung Tau Provincial Eye Hospital later planned eye health education campaigns in the province, but the effectiveness of community-level health campaigns was unknown.

In literature too, there is a paucity of studies that report the pre-test and post-test evaluation of eye health literacy in relation to community-based health education interventions (Baker and Murdoch 2008). Health intervention studies help determine what is most effective for people/patients with low health literacy and to help improve their knowledge and access to services (Gazmararian et al. 2003). We therefore developed a protocol to evaluate how effective the eye health education campaigns are in the community. The main aim of this study was to investigate the effects of a community-based eye health education intervention on eye health literacy in an adult Vietnamese population. The specific objectives were to: (i) assess the improvement in eye health literacy rate of adult population in response to an eye health education intervention, (ii) determine the factors associated with the improvement in disease awareness in the intervention communities, and (iii) determine the rates of utilisation of eye health services and eyeglasses after the intervention.

Materials and methods

We performed a prospective, cross-sectional interventional study with a baseline survey and repeat survey to evaluate the effects of a one-month long community-based eye health education intervention. We computed the sample sizes for (1) estimating baseline prevalence and (2) to determine difference before and after intervention. We arrived at the final sample as the maximum of the two sample estimates. The formula for baseline prevalence was: n = (1.962 pq/d2)x df/R, where p = prevalence, d = absolute precision and df = sampling design effect, R = response rate. The formula for pre-post differences was: n = (1.96 + 0.84)2 x (p1(1-p1)+p2 (1-p2))/(p1-p2)2, where p1 and p2 are pre and post prevalence.

Based on an assumed expectation that 50% of the target communities are aware of common ocular conditions, we determined that 400 subjects would be required to determine the true prevalence (awareness rate) with an absolute precision of 10%, accounting for 15% non-response rate and a cluster design effect of 2.0. This sample of 400 at pre- and post-intervention would have 80% power to detect a ± 10% change in awareness and knowledge at the 5% level of significance. The eligible participants in the study were one adult per household whose age should be 30 years and above, lived in the household at least 6 months every year and takes meal from the same kitchen. Visitors, guests, and rental tenants were excluded.

A baseline survey investigating knowledge, attitudes and practices (KAP) related to eye health was implemented in four communes (Phuc Buu Town, Hoa Binh Commune, Dat Do Town, Loc An Commune) that were randomly selected from two districts (Xyuen Moc and Dat Do) of Ba-Ria Vung Tau (BRVT) province. Following the baseline survey, an eye health education intervention was implemented in two randomly selected communes, namely Hoa Binh Commune and Dat Do Town. A repeat survey particularly related to awareness, knowledge and practices to eye health was conducted in all four communes six months after implementation of the baseline survey. The repeat survey applied similar sampling methods and sample size as described in the baseline survey (Paudel, Kovai, and Naduvilath et al. 2016).

We used multistage systematic random sampling to select the communes, villages, hamlets and households from the two districts (Figure 1). One urban and one rural commune was randomly selected from each district to interview the targeted sample population. From each commune, two villages were selected – one near to the commune health station and another far from the health station. In each selected village, two hamlets were chosen and then systematic random sampling was used to select households. One participant from each selected household was the final sampling unit. At the household level, eligible participants were stratified by age groups (30 to 49 years, and 50 years and above, as most vision impairing eye conditions are age-related) and gender. Final participants were selected to ensure equal distribution within each stratum. If a selected respondent refused to participate or was not available for the interview after a second visit to the original household, we visited the next household. The repeat KAP survey was conducted in the same villages selected in the baseline survey but not necessarily in the same households and the same subjects.

Figure 1. Multistage systematic random sampling undertaken in the study for the selection of households.

The questionnaire was adapted from previous studies (Dandona et al. 2001; Ormsby et al. 2012) (Supplimentary material) and was translated into Vietnamese with the help of professional translation service and in-country investigator. The questionnaire was revised to include questions relating to the most prevalent ocular conditions in the province. The first part of the questionnaire included questions relating to awareness and knowledge of red eye, cataract, refractive error, glaucoma, diabetic retinopathy and presbyopia. As described in previous publications (Dandona et al. 2001; Shrestha et al. 2014), ‘having heard of the condition’ was defined as ‘awareness,’ and having some understanding of the basic aetiology, symptoms of the condition and treatment options was defined as ‘knowledge’. The knowledge part of questionnaire included one open question about cataract, glaucoma, and diabetic eye disease. Having both awareness and knowledge of the eye conditions was defined as ‘health literacy’.

A pre-test (to check clarity, comprehension and content validity) was carried out by administering the questionnaire to 50 participants comprising equal numbers of subjects (n = 25) from a commune of each district, which were excluded in later study. Following the pilot, questionnaire items and responses were revised for clarity and description of specific terminology in the local language. The survey questionnaire was administered to 400 participants, by individual household visits. The questionnaire was self-administered by the participant during a face-to-face encounter with an interviewer in the household. Where needed, interviewers assisted participants with limited literacy by reading the questions and the response options.

Development of the eye health intervention

All communes received basic eye health information on the ‘World Sight Day’ (the second Thursday in October) through activities such as eye health practitioners’ street parades and free vision screening camps in the commune health stations, and also through local media news (radio, television and newspaper). However, the eye health education designed for this study was specifically carried out in one rural (Hoa Binh commune) and one urban (Dat Do Town) commune. The other two communes Phuc Buu Town and Loc An Commune acted as the control groups and did not receive our specific health education.

Health education materials were developed using the results of the baseline KAP survey, and provided information on common eye conditions such as red eye, refractive error, cataract, glaucoma, diabetic retinopathy and presbyopia, available services in the province, and recommendations for routine and review examinations. Materials provided the details of eye conditions and health information using suitable images and, simple and clear language. These materials were tested for face and content validity by a panel. The panel consisted of an in-country researcher, an ophthalmologist, ophthalmic nurses and commune health worker. The materials were also distributed to the patients attending BRVT Provincial Eye Centre to test their level of understanding and clarity of the language used. This testing resulted in some health information statements and terminology being revised for clarity and comprehension. All health education materials were delivered in Vietnamese. The eligible post-survey participants in the intervention communes were those who reported receiving one of four health education interventions that were delivered over a one month period (November 2013). These included:

1. Community presentation on eye health: Trained health workers in village health stations organised an educational community presentation on eye health inviting about 100 local residents aged 30 and above every weekend. Talks provided oral and visual information in Vietnamese, followed by responses to questions from the audience.

2. Loudspeaker broadcasting: Loudspeakers are often used in each village for broadcasting community notices. This facility was used once a day for a month to broadcast key health messages.

3. Poster display: Health message posters about red eye, refractive error, cataract and posterior segment disease were displayed in commune health stations, schools and market malls.

4. Brochure distribution: Brochures were distributed to households and also through health stations containing information on common eye diseases, symptoms and signs, recommendations for seeking eye examinations, and eye care services available in the province.

This project’s health education intervention approach was similar to those reported by other studies except for use of loudspeaker broadcasting (Baker and Murdoch 2008; Hobday, Ramke, and du Toit 2011).

Ethics

The study was conducted in accordance with the Declaration of Helsinki with written informed consent obtained from participants. The Vision Cooperative Research Centre and Institute for Eye Research Human Ethics Committee, Sydney, Australia provided the ethics approval (VIHEC No. 11/14). The Vietnam National Institute of Ophthalmology and BRVT Provincial Eye Centre in Vietnam provided the in-country permission to conduct this study in Vietnam.

Statistical analysis

Data were analysed using SPSS version 21 (Armonk, NY: IBM Corp.). The outcome for each question was coded in binary. The difference in proportion of the outcome between pre- and post-intervention was initially analysed using Fisher’s exact test. Outcomes that showed a significant change between pre- and post-intervention were subject to multiple logistic regression. The factors included pre vs. post, intervention vs. control commune and their interaction. The outcome variable for each logistic regression model was the awareness question recorded as binary (Yes/No). A significant interaction indicates that the change between pre- and post-intervention was different between intervention and control communes. Here the odds ratio of greater than one for an interaction would indicate that the difference between pre- and post- is greater in the test compared to the control communes. Demographic factors were compared between intervention vs. control communes. Demographic factors that were significant were added to the logistic regression as possible confounders to ensure that the efficacy of the intervention was estimated after accounting for possible effects of confounders. Further demographic factors were analysed to determine associations with awareness of the six conditions in the test community using multiple logistic regression. The strength of association between factors and outcomes was summarised as adjusted odds ratio (OR) and 95% confidence intervals (CI). The level of significance was set at 5%.

Results

Study participants

A total of 400 adults (mean age: 51.5 ± 14.5 years; range, 30–90 years) participated in the baseline survey and same number participated in the repeat survey. Equal numbers of participants by age group and gender status were included in the study. Participant’s representation from rural and urban areas was proportionate to rural urban population sizes (2:1). The proportion of people with secondary or higher level of education and higher income was greater in the post-survey. Higher level of education was significantly greater in control communes compared to intervention communes (48.4% vs 41.5%, p = 0.04) (Table 1). Due to the significant differences in education level between intervention and control communes, level of education was added as a confounding factor in the multivariate analysis. Of the four health education interventions, loudspeaker was the most common health intervention received (50%).

Table 1. Demographic characteristics of participants in the intervention and control groups.

	Intervention		Control
	Pre (N, %)	Post (N, %)	Pre (N, %)	Post (N, %)
Age (yrs) 30–49 50 above	100 (50) 100 (50)	100 (50) 100 (50)	100 (50) 100 (50)	100 (50) 100 (50)
Gender Male Female	100 (50) 100 (50)	100 (50) 100 (50)	100 (50) 100 (50)	100 (50) 100 (50)
Area Rural Urban	132 (66) 68 (34)	132 (66) 68 (34)	132 (66) 68 (34)	125 (62.5) 75 (37.5)
Education Primary or low Secondary or higher	139 (69.5) 61 (30.5)	95 (47.5) 105 (52.5)	119 (59.5) 81 (40.5)	86 (43.0) 114 (57.0)
Income Lower (<350,000VND) Middle (350–5,000 000 VND) High (>5,000 000 VND)	31 (15.5) 165 (82.5) 4 (2.0)	6 (3.0) 166 (83.0) 28 (14.0)	24 (12.0) 164 (82.0) 12 (6.0)	11 (5.5) 155 (77.5) 34 (17.0)

Awareness and knowledge of eye health

Overall, the post survey results showed that the awareness and knowledge of eye diseases significantly increased in the intervention group. On the other hand, the control group also showed a slight increase in awareness for red eye, refractive error, glaucoma and diabetic retinopathy. However, the interaction between pre- and post-test when compared between intervention and control groups, participants in the intervention group showed significantly greater knowledge of red eye preventive measures (use of clean water, avoid crowd places and use eye drops) (OR range 2.1 to 4.1, p = 0.03 to 0.001) compared to the control group (Table 2). Similarly, participants in the intervention group were more than twice as likely to have heard of cataract (OR 2.3, p = 0.008), and more than three times as likely to be aware of cataract surgery (OR 3.1, p = 0.003) and know that the cataract blindness is reversible (OR 3.3, p = 0.002), than the control group. Awareness of other eye conditions showed no significant interaction between the pre- and post-test when compared between intervention and control groups. Of those who were aware of eye diseases, more than half had previously heard about eye conditions from friends and family members. With regard to trusted and reliable information sources, about three-quarters of participants reported hospital/health centre staff and the mass media.

Table 2. Proportion, and odds (OR) of participants in the intervention and control groups reporting knowledge of eye diseases

	Intervention group (n=200)			Control group (n=200)			Intervention vs. Control*
	Pre-survey (N, %)	Post-survey (N, %)	p -value	Pre-survey (N, %)	Post-survey (N, %)	p- value	Odds ratio with 95% CI	p-value
Red eye Heard of Know about causes  Poor eye hygiene  Epidemic Infections Know preventive measures  Regular hand wash  Avoid crowded places  Use clean water  Avoid physical contact  Use eye drops	105 (52.5) 4 (2.0) 24 (12.0) 27 (13.5) 6 (3.0) 20 (10) 42 (21) 20 (10)	192 (96.0) 12 (6.0) 38 (19.0) 39 (19.5) 37 (18.5) 42 (21) 80 (40) 50 (25)	<0.001 0.07 0.07 0.13 <0.001 0.003 <0.001 <0.001	106 (53.0) 10 (5.0) 31 (15.5) 12 (6.0) 11 (5.5) 40 (20) 35 (17.5) 36 (18)	192 (96.0) 12 (6.0) 45 (22.5) 34 (17.0) 21 (10.5) 25 (12.5) 54 (27.0) 47 (23.5)	<0.001 0.80 0.09 <0.001 0.09 0.06 0.03 0.22	1.0 (0.3 to 2.9) 2.4 (0.6 to 10.2) 1.0 (0.5 to 2.2) 0.47 (0.2 to 1.1) 3.5 (1.1 to 11.4) 4.1 (1.9 to 9.1) 1.4 (0.7 to 2.7) 2.1 (1.0 to 4.5)	0.99 0.24 0.93 0.10 0.03 0.001 0.30 0.04
Cataract Heard of Know best treatment (surgery) Know blindness is reversible	43 (21.5) 22 (11) 22 (11.0)	93 (46.5) 43 (21.5) 57 (28.5)	<0.001 <0.05 <0.001	64 (32.0) 45 (22.5) 39 (19.5)	78 (39.0) 43 (21.5) 41 (20.5)	0.17 0.80 0.40	2.3 (1.3 to 4.2) 3.1 (1.5 to 6.4) 3.3 (1.6 to 7.0)	0.008 0.003 0.002
Refractive error Heard of Know best treatment (eyeglasses)	87 (43.5) 48 (24.0)	123 (61.5) 95 (47.5)	<0.001 <0.001	100 (50.0) 61 (30.5)	126 (63.0) 96 (48.0)	0.012 <0.001	1.2 (0.7 to 2.1) 1.3 (0.7 to 2.4)	0.60 0.38
Glaucoma Heard of Know best treatment (drop/ surgery) Know vision loss is permanent	14 (7.0) 9 (4.5) 8 (4.0)	43 (21.5) 28 (14) 17 (8.5)	<0.001 0.001 0.02	14 (7.0) 7 (3.5) 8 (4.0)	40 (20.0) 19 (9.5) 20 (10.0)	<0.001 0.015 0.006	1.1 (0.4 to 2.8) 1.2 (0.4 to 3.8) 0.8 (0.3 to 2.8)	0.87 0.80 0.77
Diabetic retinopathy Heard of Know best treatment (laser, surgery) Know vision loss is treatable	34 (17.0) 6 (3.0) 12 (6.0)	60 (30.0) 7 (3.5) 24 (12.0)	0.003 0.80 0.04	26 (13.0) 1 (0.5) 8 (4.0)	61 (30.5) 12 (6.0) 27 (13.5)	<0.001 0.001 <0.001	0.7 (0.3 to 1.4) 0.1 (0.01 to 0.9) 0.6 (0.2 to 1.7)	0.30 0.04 0.30
Presbyopia Heard of Know best treatment (eyeglasses)	96 (48.0) 58 (29)	102 (51.0) 79 (39.5)	0.62 0.03	102 (51.0) 56 (28)	106 (53.0) 73 (36.5)	0.75 0.07	1.0 (0.6 to 1.8) 1.0 (0.6 to 2.0)	0.99 0.93

*Adjusted for level of education; Bold values refer to statistical significance.

Multiple logistic regression was performed to determine the association of disease awareness with factors – locality, age, gender, education, family income, possession of eyeglasses and history of eye check, using the pre and post survey samples of test and control communities. Urban location, higher education and eyeglasses ownership were the major factors associated with disease awareness (Table 3). Participants from the urban areas and those who possessed eyeglasses were approximately two-fold more likely to be aware of cataract, refractive error, and glaucoma (p = 0.002 to <0.001). However, region (urban/rural) did not interact significantly with test/control community (p > 0.53), indicating that the greater awareness in the urban areas was not different between test and control communities. Participants with secondary education or higher were two-fold more likely to be aware of all eye conditions (p = 0.001 to <0.001) except glaucoma. Age, gender and family income were not associated with disease awareness, except for red eye, which showed association with income level.

Table 3. Factors associated with disease awareness among participants in the intervention and control groups.

	Heard of disease^* Odds ratio (95% CI); p-value
	Red Eye	Cataract	Refractive error	Glaucoma	Diabetic eye disease	Presbyopia
Locality Rural Urban	1 (Ref) 0.8 (0.6 to 1.2); 0.25	1 (Ref) 2.0 (1.4 to 2.8); <0.001	1 (Ref) 1.6 (1.2 to 2.2); 0.002	1 (Ref) 1.5 (0.9 to 2.3); 0.09	1 (Ref) 1.3 (0.9 to 1.8); 0.18	1 (Ref) 1.4 (1.0 to 1.9); 0.04
Age 30–49 years ≥50 years	1 (Ref) 0.7 (0.5 to 1.0); 0.05	1 (Ref) 1.1 (08 to 1.5); 0.59	1 (Ref) 0.9 (0.7 to 1.3); 0.72	1 (Ref) 1.4 (0.9 to 2.1); 0.16	1 (Ref) 1.3 (0.9 to 1.9); 0.16	1 (Ref) 1.0 (0.8 to 1.4); 0.82
Gender Female Male	1 (Ref) 0.8 (0.5 to 1.1); 0.10	1 (Ref) 1.1 (0.8 to 1.5); 0.57	1 (Ref) 1.0 (0.7 to 1.3); 0.81	1 (Ref) 0.8 (0.5 to 1.2); 0.20	1 (Ref) 1.1 (0.8 to 1.6); 0.51	1 (Ref) 0.9 (0.7 to 1.2); 0.52
Education School≤Primary School≥ Secondary	1 (Ref) 2.2 (1.5 to 3.2); <0.001	1 (Ref) 2.0 (1.4 to 2.8); <0.001	1 (Ref) 2.4 (1.8 to 3.3);<0.001	1 (Ref) 1.5 (0.9 to 2.3); 0.09	1 (Ref) 1.8 (1.3 to 2.6); 0.001	1 (Ref) 2.0 (1.5 to 2.7); <0.001
Family income Low income Middle income High income	1 (Ref) 2.7 (1.6 to 4.7);< 0.001 1.9 (1.1 to 3.3); 0.03	1 (Ref) 0.8 (0.4 to 1.4); 0.42 0.8 (0.5 to 1.5); 0.39	1 (Ref) 1.5 (0.8 to 2.5);0.17 1.3 (0.7 to 2.3);0.12	1 (Ref) 0.8 (0.4 to 1.7); 0.65 0.9 (0.4 to 1.9); 0.71	1 (Ref) 0.6 (0.3 to 1.2); 0.15 1.1 (0.6 to 2.1); 0.79	1 (Ref) 1.6 (0.9 to 2.8); 0.09 1.2 (0.7 to 2.2); 0.44
Possess eyeglasses No Yes	1 (Ref) 1.2 (0.8 to 1.7); 0.30	1 (Ref) 2.0 (1.5 to 2.8); <0.001	1 (Ref) 1.7 (1.2 to 2.3); 0.001	1 (Ref) 2.2 (1.4 to 3.4); <0.001	1 (Ref) 1.5 (1.1 to 2.1); 0.02	1 (Ref) 2.2 (1.6 to 2.9); <0.001
Eye check Never Yes	1 (Ref) 1.1 (0.8 to 1.5); 0.71	1 (Ref) 2.1 (1.5 to 2.9); <0.001	1 (Ref) 1.2 (0.9 to 1.6); 0.24	1 (Ref) 2.4 (1.5 to 3.8); <0.001	1 (Ref) 1.1 (0.8 to 1.6); 0.59	1 (Ref) 0.9 (0.7 to 1.2); 0.58

^aAdjusted for study intervention (test vs control); Bold values refer to statistical significance.

Use of eye care services

Over 80% of participants reported having had a health check-up in the past but previous eye examinations were reported by just over 45% of participants. The number of participants who reported having had an eye examination significantly increased in the intervention community after health promotion (45.5% to 55.5%; p < 0.05) (Table 4). Similarly, 6% more participants reported eyeglasses wear but this increase was not significant. There were no significant increases in reports of eye examinations or eyeglasses wear in the control community.

Table 4. Pre- and post-comparison of eye examination and treatment practices in the intervention and control groups.

	Intervention community		p-value	Control community		p-value
	Pre	Post		Pre	Post
Ever had health check  Yes  No	162 (81.0) 38 (19.0)	164 (82.0) 36 (18.0)	0.80	171 (85.5) 29 (14.5)	165 (82.5) 35 (17.5)	0.41
Ever had eye check  Yes  No	91 (45.5) 109 (54.5)	111 (55.5) 89 (44.5)	0.04	95 (47.5) 105 (52.5)	111 (55.5) 89 (44.5)	0.11
Wear eyeglasses  Yes  No	67 (33.5) 133 (66.5)	81 (40.5) 119 (59.5)	0.12	85 (42.5) 115 (57.5)	83 (41.5) 117 (58.5)	0.84
If do not wear eyeglasses, why?  I don’t have vision problem  I don’t like wearing eyeglasses  I have no money to purchase  I never have had eye examination	95 (71.4) 26 (19.5) 3 (2.3) 6 ((4.5)	64 (53.8) 46 (38.7) 1 (0.8) 4 (3.4)	NA	90 (78.3) 21 (18.3) 1 (0.9) 1 (0.9)	53 (45.3) 55 (47.0) 1 (0.9) 4 (3.4)	NA
Eyeglasses improved vision  Yes  No	61 (91.0) 6 (9.0)	70 (86.4) 11 (13.6)	0.38	70 (82.4) 15 (17.6)	67 (80.7) 16 (19.3)	0.78
Eyeglasses affordable  Yes  No	58 (86.6) 9 (13.4)	74 (91.4) 7 (8.6)	0.35	75 (88.2) 10 (11.8)	75 (90.4) 8 (9.6)	0.66

NA, Not applicable; Bold values refer to statistical significance.

Reasons for not wearing eyeglasses

In both the intervention and control communities, participants reported that the main reasons for not wearing eyeglasses were ‘I don’t have vision problem’ and ‘I don’t like wearing eyeglasses’. In both intervention and control communities, the majority of participants (>80%) who had worn eyeglasses stated that eyeglasses were affordable and improved their vision.

Discussion

This study is novel in reporting the impact of a community-level education campaign on eye health literacy and practices of an adult population. Evaluations have usually been performed in hospitals (Thapa et al. 2012; Noertjojo et al. 2006; Meyer et al. 2016; Gyawali and Sarkar 2014) or schools (Paudel, Yen, and Kovai et al. 2019; Kirag and Temel 2018; Hobday et al. 2015). This study reports awareness and knowledge of six common eye conditions and eye care practices in the communities before and after a month-long community-based health education intervention provided through four different media campaigns.

The results showed that the month-long health education intervention significantly increased the awareness and knowledge of cataract and red eye prevention measures in the intervention community compared to the control community. The awareness and knowledge increased more than two-fold for cataract and red eye. A study conducted in the United Kingdom showed that a public health education intervention increased the awareness of glaucoma by more than two-fold while awareness of cataract and macular degeneration showed no positive changes (Baker and Murdoch 2008). The reasons for this difference in the findings was that the eye health campaign materials in the United Kingdom were mainly focused on glaucoma awareness. Similar to other study findings (Muller, Keeffe, and Taylor 2007; Hobday, Ramke, and du Toit 2011), the present study showed a significant increase in the eye examination visits in the intervention group. As reported in previous studies (Dandona et al. 2001; Ormsby et al. 2012; Gyawali and Sarkar 2014), friends and family members were the main source of awareness of eye conditions. However, hospital/health centre staff and the mass media were regarded as trusted and reliable information sources both in pre- and post-survey. Evidence also suggests that health information shared by mass media is a powerful tool for public health campaigns as the information is likely to reach communities of all health literacy levels.² This indicates that mass media such as radio and television should be used when possible for health education campaigns to address health literacy equity issues within communities.

The reason for increased health literacy of red eye in our study could be its pervasive nature (60% of participants of all ages in our study reported to have experienced this in their life). It could also relate to the condition when people saw the campaign so they had greater recall of something that was already familiar. Previous reports have shown increased health literacy and service utilisation as attributed to health awareness of age-related health conditions and that campaigns are more successful (Sudore and Schillinger 2009). The increase in awareness means that the people in the intervention community are more likely to seek services and take care of their eyes. Although literature suggests increased knowledge doesn’t necessarily translate into changed behaviours (Kelly and Barker 2016), this study showed that visits for eye examinations to eye hospitals increased significantly in the intervention communities.

On the other hand, the interaction results showed that the proportion of increased awareness and knowledge were not substantial for other four eye conditions (glaucoma, diabetic retinopathy, refractive error and presbyopia) suggesting that a health education intervention conducted here is not effective at increasing the health literacy of these conditions or maybe the intervention itself was not adequate (Table 2). Similar to the findings of other studies (Dandona et al. 2001; Thapa, Berg, and Khanal et al. 2011; Ormsby et al. 2012), low level of awareness of glaucoma and diabetic retinopathy is reported in both the communities. This result is however different than that in the developed world where about 80% people are reported to be aware of glaucoma (Livingston, McCarty, and Taylor 1998) and above 90% were aware of diabetic retinopathy (Meyer et al. 2016). The increase of disease awareness in control communes could be attributed to education level that was higher compared to intervention communes.

Further, this study showed that the increased eye health awareness in the intervention community was significantly associated with the locality and education, suggesting people living in urban areas, and with secondary and higher education are more likely to have better health literacy and better service utilisation. This is consistent with other published studies (Dandona et al. 2001; Shrestha et al. 2014; Thapa, Berg, and Khanal et al. 2011; Huang et al. 2013; Sudore and Schillinger 2009; Ramke et al. 2012; Sathyamangalam, Paul, and George et al. 2009; Fotouhi, Hashemi, and Mohammad 2006). Awareness was also found to be significantly associated with past eye examinations and possession of eyeglasses. Those participants who possessed eyeglasses were twice as likely to have heard of cataract, refractive error, glaucoma and presbyopia. Similarly, those who had an eye examination were twice as likely to have heard of cataract and glaucoma. This indicates that getting people to an eye exam is more effective so as to improve the diseases awareness.

An eye examination for asymptomatic patients is recommended every five years for those under 40 years of age, every 2–4 years for those aged 40 to 54 years, every 1–3 years for those aged 55 to 64 years and every 1–2 years for 65 years and older (American Academy of Ophthalmology 2010). Although eye examinations and awareness of some eye conditions increased with intervention, the post-survey findings showed that about 45% participants reported never having had an eye health check-up. This proportion is higher than the percentage of no-visit for eye examination reported in other studies (i.e. about 35%) (Fotouhi, Hashemi, and Mohammad 2006; Marmamula, Pyda, and Khanna 2017). Interestingly, about 70% of respondents do not have any plans for an eye examination in the next year. This indicates that people are less likely to seek eye care services until they start experiencing eye problems.

The study has some limitations. First, although brochure distribution was planned to reach every house, our monitoring system could not ensure this. Second, as the participants were randomly chosen from the community, the participants in the post study had higher educational levels compared to baseline study. However, this confounding factor was adjusted in the multivariate analysis. Third, even though the results showed some positive changes in the knowledge of eye conditions, this may not predict long-term changes in behaviours and practices for service utilisation. A regular health education intervention program and a longitudinal study would be ideal to measure lasting change. Lastly, the study findings reflect the Vietnamese context and may not be applicable to or predict the health literacy situation in other countries.

In conclusion, this study showed that a health education intervention program conducted for a month with promotional materials, contributed to increased awareness and knowledge of red eye and cataract but limited changes for the other four conditions. If health promotion campaigns are well-monitored, organised with a wider population coverage for a reasonable amount of time and conducted periodically at a community level, awareness and the knowledge of eye diseases would be likely to improve affectedly. This study methods and results add to the knowledge of effective public health promotion and indicates that the carefully planned health education campaigns are important for improving health literacy and better usages of services, and therefore should be included when developing health service programs. The study findings may also help health organisations to introduce policy and plan for improved health literacy of the Vietnamese people.

Acknowledgments

We acknowledge the TNS Vietnam for the support in data collection and the Ba Ria Vung Tau Provincial Eye Hospital for the technical support during community based health intervention.

Supplementary material

Supplemental data for this article can be accessed here.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This study was conducted as a part of the Vietnam Australia Vision Support Program (VAVSP) financially supported by an Avoidable Blindness Initiative grant through the Department of Foreign Affairs and Trade, Australian Government.