Introduction

Malaysia signed the World Health Assembly resolution (WHA66.4: Towards Universal Eye Health: A Global Action Plan 2014–2019) during the 66^th WHA Geneva in May 2013. By signing the documents, the country pledged its commitment to implementing the items in the resolution and monitoring its progress [1,2]. In alignment with this commitment, a national-level initiative was undertaken to establish baseline data on key eye health indicators, aimed at guiding planning efforts, improving service delivery, and ensuring equitable access to eye care. Consequently, the National Eye Survey II (NES II) was conducted in 2014. Utilising the World Health Organization (WHO)-recommended Rapid Assessment of Avoidable Blindness (RAAB) methodology, the survey was implemented simultaneously across six administrative regions of Malaysia: Central, Eastern, Northern, Southern, Sabah and Sarawak.

NES II results revealed significant regional disparities in the prevalence of blindness. The Central region reported the lowest rate at 0.5%, followed by the Southern region at 0.9%. Higher prevalence was recorded in the Eastern (1.4%) and Northern (1.5%) regions, with the highest in Sabah (1.9%) and Sarawak (1.6%). In terms of cataract surgical outcomes, only 73.0% of patients in Sarawak and 73.2% in Sabah achieved good vision (6/12 or better) after surgery. The Eastern region had 70.6%, Southern 76.7%, Northern 79.1%, and the Central region recorded the best outcome at 81.9%. At the 3/60 threshold for effective Cataract Surgical Coverage (eCSC), Sabah recorded the lowest rate at 51.8%, followed by the Eastern region (57.4%), Sarawak (61.8%), Southern region (66.7%), Northern region (70.5%), and the Central region with the highest coverage at 81.1% [3].

Based on the discrepancies of key findings between the administrative regions during the NES II, the government launched Klinik Katarak-Kementerian Kesihatan Malaysia (Cataract Clinic Ministry of Health Malaysia, KK-KKM) as a community cataract pilot project in Sarawak (the Malaysian Borneo) and the Eastern Region of Peninsular Malaysia aiming to address the issue of high prevalence of blindness and cataract blindness [Fig 1]. The objective, concept and operational strategies of the program were subsequently audited and endorsed through the Case Study process of the WHO Western Pacific Innovation Challenge initiative in 2021/2022 [4].

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. The KK-KKM mobile unit en route by barge to the remote district of Kapit in Sarawak.

https://doi.org/10.1371/journal.pone.0319123.g001

In 2023, nearly a decade after the implementation of NES II, a follow-up survey was conducted in Sarawak and Eastern Region – areas where the KK-KKM mobile outreach service had been continuously deployed. The results demonstrated substantial improvements, with a notable decline in the prevalence of blindness: Sarawak reported a reduction to 0.6% and the Eastern region to 0.8% [5]. Visual acuity outcomes following cataract surgery also improved markedly, with 89.8% in Sarawak and 87.5% in the Eastern region achieving good visual outcomes (6/12 or better) [6]. Furthermore, effective cataract surgical coverage (eCSC) at the 3/60 threshold had increased to 83.8% in Sarawak and 74.4% in the Eastern region [7]. These positive trends indicate progress but also underscore the need for continued, targeted efforts to reduce regional disparities and ensure equitable access to quality eye care services nationwide.

Despite the establishment of service brands and the allocation of specific funds to the KK-KKM project, there remains a disparity in hospital and community-based data pertaining to cataract surgery (including patient profiles, surgeon practices, and key indicators for service performance and uptake) between the two regions, with the Eastern Region constantly underperforming [8–10].

We aim to analyse the association between the predictors and factors with the outcome (surgery done and surgery not done) in the community during the follow-up population survey after nine years of KK-KKM implementation in Sarawak and the Eastern Region. The findings from the study could potentially be used to formulate new policies or strengthen and optimise the implementation of the KK-KKM program, hence addressing the issue of service outcome discrepancy between the regions.

Materials and methods

This was a cross-sectional population survey using the Rapid Assessment of Avoidable Blindness (RAAB) technique to obtain data on eye health, specifically focusing on cataract surgery status. Additionally, structured interviews were conducted through a questionnaire to identify various factors that may influence surgical status, including knowledge, perceptions, attitudes, practices, as well as demographic and socioeconomic characteristics.

This study was conducted in two phases. Phase 1 involved the development and validation of a questionnaire. The factors and questionnaires were developed from an extensive literature search, expert brainstorming sessions, and evaluation of content and face validity. Content Validity Index (CVI) and Face Validity Index (FVI) were calculated at the initial phase of the questionnaire development. For the original 51-item questionnaire, low scale-level CVI values ranged from 0.50 to 0.71, and low item-level CVI values ranged from 0.17 to 0.67. These corresponded to eight items that were subsequently removed. After item removal, the mean item-level CVI was 0.98, and the average scale-level CVI was 0.93. Similarly, scale-level FVI values ranged from 0.55 to 0.78; and item-level FVI values from 0.30 to 0.70, leading to removal of seven items.

This process was followed by validation and reliability analysis using exploratory factor analysis (EFA), reliability tests, confirmatory factor analysis (CFA) and test-retest to produce a questionnaire named CatSurg-U Questionnaire (Cat = Cataract; Surg = Surgery; U = Uptake). It contained six factors and 22 questionnaire items [Table 1]. Cronbach’s alpha values of individual factors ranged from 0.50 to 0.72, and the total Cronbach’s alpha for the final model (22 items) was 0.72. CFA revealed a moderate absolute and reasonable parsimonious fit model. Additional details on the validity and reliability testing of the CatSurg-U questionnaire are included in a separate manuscript currently under submission.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Factors and questionnaire items (22 items) – for CatSurg-U data collection.

https://doi.org/10.1371/journal.pone.0319123.t001

Phase 1 was followed by Phase 2, a survey using the RAAB method to obtain population data on eye health and blindness [11,12]. The survey’s methodology, prevalence and cataract surgical outcome have been published [5,6]. Eligible subjects from the same sample were further interviewed using the validated CatSurg-U Questionnaire to identify factors influencing the cataract surgery uptake in both regions.

IOV CatSurg-U

During training, the assessment for IOV CatSurg-U was conducted after the IOV RAAB on subjects who were a mix of hospital staff and patients from the outpatient department. The subjects were different from the recruited subjects of the IOV RAAB. As per the questionnaire, 22 questions were asked by six interviewers representing each survey team. Twenty-two (22) subjects in Sarawak and 20 in the Eastern Region volunteered to be interviewed. The subjects included individuals with mixed normal and impaired vision, including cataracts and pseudophakia (had been operated on for cataracts). They were interviewed by rotation (each subject was interviewed six times using the same questionnaire, starting at different chronological numbers). The agreements were analysed using Fleiss Kappa [17]. For Sarawak, two items had a fair agreement, ten had a moderate agreement, eight had a substantial agreement, and two had an almost perfect agreement. For the Eastern Region, six items had a fair agreement, seven had a moderate agreement, seven had a substantial agreement, and two had an almost perfect agreement. For both regions, C8 and C9 had only fair agreement. All agreements were statistically significant [Table 2]. Similar to the intervention taken after IOV RAAB, post-mortem interviews and group retraining focusing on the disagreements were done before the team went to the field the next day for an actual survey. Each team member was re-explained regarding each question’s aim and trained on the best possible way to administer it to the subjects. The emphasis was also given only to allow the assigned person to conduct the interview throughout the survey.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Inter-interviewer Reliability Assessment of Six Interviewers of Sarawak and the. Eastern Region for agreement on CatSurg-U Questionnaire; Fleiss Kappa by Response Item (22 Questions).

https://doi.org/10.1371/journal.pone.0319123.t002

Statistical analysis

Statistical analysis was performed using IBM SPSS AMOS Statistical Package for Social Science (SPSS), version 26.0 (SPSS, Inc., Chicago, III., USA) for Windows. Inter-interviewer reliability assessment for the CatSurg-U Questionnaire was done using Fleiss Kappa. Binary Logistic Regression was done to identify the association of the predictor variables or factors (demographic, socioeconomic, knowledge, perception, attitude and practice factors) to the dependent outcome (surgery done vs surgery not done with the latter as the outcome event) in one or both eyes (unilateral vs bilateral) [Fig 2]. The analysis started with data exploration and cleaning, followed by simple logistic regression (SLogR) to screen for important independent variables. Variables with P-value < 0.25 and/or clinically important were selected for multiple logistic regression (MLogR) using forward and backward selection techniques to produce a preliminary main effect model. The modelling procedure was done using the Likelihood Ratio test based on Maximum Likelihood Estimates (MLEs). From SLogR to MLogR, variable selection was done by four major principles: (a) best fit, (b) parsimonious model, (c) biological plausibility, and (d) statistical significance. This step was followed by a multicollinearity & interaction check to produce a preliminary final model. The Hosmer–Lemeshow goodness-of-fit test, classification tables and area under the receiver operating characteristic (ROC) curve were examined to ensure the model’s fitness before the regression model was confirmed. The results were interpreted using crude and adjusted Odds Ratios, 95% confidence intervals and corresponding P-values [20,21].

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Conceptual framework.

https://doi.org/10.1371/journal.pone.0319123.g002

Results

Some of the results related to the CatSurg-U Questionnaire development, validation and the RAAB survey are still under review or have been published. This manuscript will only address the CatSurg questionnaire data collection results following the RAAB survey. A total of 1,119 subjects (Sarawak = 408, Eastern = 711) were recruited. The demographic profile of the eligible subjects for the CatSurg-U Questionnaire showed a significantly higher proportion of self-employed females, females with no formal educational background and females with income RM1500 or less per month in Sarawak. There was a significantly higher proportion of Malay females and a higher significant proportion of males with primary school qualifications in the Eastern Region. There was a significantly higher proportion of married males in both regions [Table 3].

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Demographic profile of subjects recruited for the CatSurg-U Questionnaire by gender (Sarawak n = 408, Eastern n = 711).

https://doi.org/10.1371/journal.pone.0319123.t003

There were two common significant factors identified which influenced the individuals with unilateral operable cataract to have “no surgery” in both regions (“perception to own sight” and “attitude towards treatment”). The other significant factor was “practice towards information” (specific to the Eastern Region) [Table 4].

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Associated Factors Influencing Individuals with UNILATERAL (one eye) Operable Cataract to have “No Surgery” by Simple and Multiple Logistic Regression Model (Sarawak n = 90, Eastern n = 212).

https://doi.org/10.1371/journal.pone.0319123.t004

In Sarawak:

1. 1). Individuals with better “perception to own sight” had 33.0% lower odds of having no surgery when they had an operable cataract in one eye [AOR: 0.67, 95% CI (0.53, 0.84) P  = 0.001, adjusted for “attitude towards treatment”].
2. 2). Individuals with better “attitude towards treatment” had 47.0% higher odds of having no surgery when they had an operable cataract in one eye [AOR:1.47, 95% CI (1.17, 1.85) P = 0.001, adjusted for “perception to own sight”].

In the Eastern Region:

1. 1). Individuals with better “perception to own sight” had 20.0% lower odds of having no surgery when they had an operable cataract in one eye [AOR: 0.80, 95% CI (0.69, 0.92) P  = 0.002, adjusted for “attitude towards treatment” and “practice towards information”].
2. 2). Individuals with better “attitude towards treatment” had 15.0% higher odds of having no surgery when they had an operable cataract in one eye [AOR: 1.15, 95% CI (1.03, 1.29) P = 0.016, adjusted for “perception to own sight” and “practice towards information”].
3. 3). Individuals with better “practice towards information” had 23.0% higher odds of having no surgery when they had an operable cataract in one eye [AOR: 1.23, 95% CI (1.01, 1.50) P  = 0.042, adjusted for “perception to own sight” and “attitude towards treatment”].

For bilateral operable cataract, the common significant factor that influenced the individuals to have “no surgery” in both regions was “knowledge on surgery”. The other significant factors were “perception to own sight”, ethnicity and level of education (specific to Sarawak) and “practice on surgery” (specific to the Eastern Region) [Table 5].

Download:

• PNG
  larger image
• TIFF
  original image

Table 5. Associated Factors Influencing Individuals with BILATERAL (both eyes) Operable Cataract to have “No Surgery” by Simple and Multiple Logistic Regression Model (Sarawak n = 243, Eastern n = 352).

https://doi.org/10.1371/journal.pone.0319123.t005

In Sarawak:

1. 1). Individuals with better “knowledge on surgery” had 65.0% lower odds of having no surgery when they had operable cataract in both eyes [AOR: 0.35, 95% CI (0.25, 0.50) P < 0.001, adjusted for ethnicity, level of education and “perception to own sight”].
2. 2). Individuals with better “perception to own sight” had 48.0% higher odds of having no surgery when they had operable cataract in both eyes [AOR: 1.48, 95% CI (1.15, 1.89) P = 0.002, adjusted for ethnicity, level of education and “knowledge of surgery”].
3. 3). Chinese individuals had 81.0% lower odds, compared to Malay individuals, of having no surgery when they had operable cataract in both eyes [AOR: 0.19, 95% CI (0.04, 0.88) P = 0.033, adjusted for level of education, “knowledge of surgery” and “perception to own sight”].
4. 4). Individuals who attended primary school had 5.5 times higher odds, compared to those who attended secondary school or above, of having no surgery when they had operable cataract in both eyes [AOR: 5.54, 95% CI (1.49, 20.69) P  = 0.011, adjusted for ethnicity, “knowledge of surgery” and “perception to own sight”].

In the Eastern Region:

1. 1). Individuals with better “knowledge on surgery” had 65.0% lower odds of having no surgery when they had operable cataract in both eyes [AOR: 0.35, 95% CI (0.26, 0.48) P < 0.001, adjusted for “practice of surgery”].
2. 2). Individuals with better “practice on surgery” had 28.0% lower odds of having no surgery when they had operable cataract in both eyes [AOR: 0.72, 95% CI (0.62, 0.84) P < 0.001, adjusted for “knowledge of surgery”].

Discussion

The KK-KKM project in both regions emphasises scheduled trips for screening and surgery and revisiting them after one month by optometrists to assess patients’ visual outcomes. The timetable for the mobile unit is distributed to all the Provincial Hospitals at the beginning of each calendar year. The fixed schedule allows people in remote areas to plan their finances and trips to come forward and seek eye treatment. Operating in proper operating rooms using standard cataract extraction techniques, adhering to the quality measurement of biometry and fixing the timetable for the service maximises access and ensures quality surgery for the people.

Like all other hospital facilities in the Malaysian Ministry of Health, data from cataract surgeries performed at the KK-KKM locations are entered into the National Eye Database, a web-based password-protected surveillance system collecting data on eye diseases and the clinical performance of the ophthalmology program in Malaysia. It consists of online systematic data entry according to predefined sets of preoperative, operative and outcome forms. Details on the Malaysian Cataract Surgery Registry and Cumulative Summation (CUSUM) techniques in cataract surgical performance monitoring have been published elsewhere [22,23]. The main reason behind these strategic measures was to ensure maximum access to quality cataract surgeries for every individual in the region, regardless of differences in socioeconomic and demographic profiles.

This study combined the RAAB survey findings (surgery status obtained from ocular examination by trained eyecare providers) with the CatSurg-U Questionnaire findings (factors and response/questionnaire items), demographic and socioeconomic data [Fig 2]. Peng et al. (2013), in their population survey to evaluate the utilisation of eye care services in a rural population in North China, used the frequency of visits to eye care providers as the outcome of multiple logistic regression in analysing the factors associated with the underuse of services [24]. Li et al. (2020), in a cross-sectional survey in rural Yueqing, Wenzhou, China, used a questionnaire on knowledge, attitudes and practices (KAP). Visual acuity (unilateral vs bilateral) was used as the outcome variable of multiple logistic regression to identify factors affecting the use of eye care services [25]. To the best of our knowledge, no available study in the literature yet evaluates the uptake of cataract surgery using an objective outcome measurement (surgery status) obtained from ocular examination.

In 1998, WHO defined a good visual outcome after cataract surgery as achieving visual acuity of ≥6/18, while an outcome of <6/60 was considered poor. It was recommended that at least 80% of patients should attain a good outcome without correction, and 90% with best-corrected visual acuity, with poor outcomes kept below 5%. In 2021, WHO updated these benchmarks, redefining a good surgical outcome as achieving presenting visual acuity of ≥6/12, and setting the eligibility threshold for cataract surgery at best-corrected visual acuity (BCVA) of <6/12 [26].

The following were the outcome variables collected during the fieldwork (RAAB survey) obtained from the ocular examination. They were evaluated and discussed with the research committee before the commencement of the survey:

1. I. Lens status

This variable was obtained from direct eye examination using a handheld direct ophthalmoscope. The findings included normal, opacity, aphakia, pseudophakia with no posterior capsular opacity, and pseudophakia with posterior capsular opacity or no view. Cataract surgery uptake was represented by either aphakia, pseudophakia (surgery done), or opacity (surgery not done).

• Distant Pinhole visual acuity (PinVA)

This variable was obtained by testing the subjects’ distant visual acuity with pinholes using the built-in chart in the RAAB7 apps. The findings were an option of the levels of PinVA (6/12, 6/18, 6/60, 3/60, 1/60, Perception of light or No perception of light). The inclusion criteria for the CatSurg Questionnaire interview was Pinhole VA worse than 6/18.

1. III. Cataract surgery

Four sub-variables represented this variable:

1. a). Age when the surgery was done (numerical)
2. b). Type of cataract surgery [categorical: options were no Intraocular Lens (IOL), IOL implantation or couching]
3. c). Cost of cataract surgery (categorical: options were totally free, partially free or fully paid)
4. d). Cause of poor vision after cataract surgery (categorical: options were ocular co-morbidity, operative complication, refractive error or long-term complications)

Unlike Variable II and III, Variable I had a clear binary definition between “surgery done” and “surgery not done”. For Variable II, subjects could have undergone surgery, but the visual acuity remained poor (e.g., surgery with intra-operative or post-operative complication); likewise, they could have lens opacity, but the visual acuity remained good. Variable III only captured subjects who had undergone cataract surgery (and excluded subjects who had cataracts but did not undergo surgery).

The results from the multiple regression analysis suggested that there were two common significant factors identified, which influenced the individuals with unilateral operable cataract to have “no surgery” in both regions (“perception to own sight” and “attitude towards treatment”). The other significant factor was “practice towards information” (specific to the Eastern Region). For bilateral operable cataract, the common significant factor that influenced the individuals to have “no surgery” in both regions was “knowledge on surgery”. The other significant factors were “perception to own sight,” ethnicity, and level of education (specific to Sarawak), as well as “practice on surgery” (specific to the Eastern Region).

In Sarawak, having better “perception to own sight” was associated with lower odds of “no surgery” in unilateral cataract. However, the factor was associated with higher odds if the person had a bilateral cataract. The seemingly contradictory role of the “perception to own sight” between unilateral and bilateral operable cataracts can be explained by differences in individuals’ visual awareness and adaptive behaviour. For individuals with unilateral cataract, the disparity in vision between the affected and unaffected eye is often noticeable and functionally limiting, which increases the likelihood of seeking surgical intervention. However, for those with bilateral operable cataracts, vision typically deteriorates gradually in both eyes, leading individuals to adapt to their reduced vision over time. This adaptation, combined with factors such as slow cataract progression, limited daily visual demands, cultural beliefs, and barriers related to access and cost, may result in a lower perceived need for surgery. As a result, the odds of not undergoing surgery are higher in this group despite similar or greater levels of visual impairment.

In Sarawak, where geographical and financial access could be an issue for the community, those with bilateral operable cataract should not be neglected and must be reached out to by expanding the coverage of the mobile unit.

In both regions, when a person had unilateral operable cataract, although a better “perception of own sight” was associated with lower odds of “no surgery”, the opposite was revealed for the “attitude towards treatment”. Even if the community had a positive attitude towards seeking cataract treatment, the odds of having “no surgery” were higher. They had good self-awareness about poor vision in one eye and possibly wanted to seek treatment, but access to treatment (geographical or financial) could be the limiting factor. Interpreting the “attitude towards treatment” and “practice towards information” together, within the scope of the questionnaire items (the better the attitude and practice, the higher the odds of “no surgery) in the Eastern Region pointed more to possible limited “supply” of service despite higher “demand”. A concerted effort to increase the coverage and capacity of services must be addressed to increase the uptake of cataract surgery in the population, especially in the Eastern Region.

Ethnicity (Chinese, compared to Malays) and level of education (primary school, compared to secondary school or above) contributed to the factors influencing individuals in Sarawak with bilateral operable cataract to have “no surgery”. These factors were not identified in the Eastern Region. Sarawak is a multicultural state. According to the 2020 Population and Housing Census, the Iban are the largest indigenous group in Sarawak, comprising approximately 30.0–35.0% of the population, Chinese 24.0–26.0%, Malays 20.0–25.0%, Bidayuh 6.0–7.0% and other indigenous groups, including the Orang Ulu, Melanau, Kayan, Kenyah, Kelabit, and Penan, who together make up around 10.0–15.0% of the population [27]. The Iban are mainly concentrated in rural and inland areas such as Kapit, Sri Aman, and Betong. The Bidayuh predominantly reside in the highland areas of Kuching, Bau, and Serian. Malays and Melanaus are more commonly found along the coastal regions, particularly around Kuching, Mukah, and Miri. The Chinese population is largely based in urban centres like Kuching, Sibu, Bintulu, and Miri. Smaller indigenous groups, including the Orang Ulu, are mostly located in the remote interior of northern Sarawak [27]. The lower odds among Chinese with bilateral operable cataract with “no surgery” perhaps was contributed by the ethnic composition in the state and their better access to treatment (geographically). Analysis of other specific ethnics was not significant. The individuals with primary school qualifications had 5.5 times higher odds of “no surgery “ than those with higher qualifications. These findings emphasised the importance of education in providing eye care in Sarawak and other regions.

A better “knowledge on surgery” if the individuals had bilateral operable cataract was associated with lower odds of “no surgery” in both regions. This indicated that if the community had substantial knowledge about cataract surgery, the odds of them coming forward to seek treatment would be higher. Similarly, “practice on surgery” was associated with lower odds of “no surgery” when the individuals had bilateral operable cataract, but only specific to the Eastern Region.

The strategies proposed to increase the uptake of cataract surgery in the community shall be targeted at the identified influencing factors associated with higher odds of “no surgery” in specific regions. In Sarawak (“attitude towards treatment” in unilateral operable cataract) and (level of education and “perception to own sight” in bilateral operable cataract). In the Eastern Region (“attitude towards treatment” and “practice towards information” in unilateral operable cataract)

Peng et al. (2013), in their evaluation of the use of eye care services in a rural population, revealed ocular comorbidities (glaucoma, age macular degeneration or refractive error) as the factors influencing the community to visit an eye care provider [24]. Li et al. (2020) stated “no need” and “schedule conflicts” as the main reasons for not seeking eye care. High education levels, older age, self-perceived vision condition and regular vision check behaviour were related to seeking eye care services [25].

The remedial measures for this study shall include expanding and upgrading cataract surgical services in both regions. This expansion/upgrade must include the hospital infrastructure, the mobile unit, and workforce capacity, especially the surgeons’ training. This expansion will allow active case detection, especially for individuals residing in remote areas with bilateral operable cataracts. Other remedial measures include strengthening community advocacy, education and engagement to empower and sustain the long-term project, especially in the Eastern Region.

Conclusion

The common factors influencing the cataract surgery uptake in Sarawak and Eastern Region are “perception to own sight”, “attitude towards treatment”, and “knowledge on surgery”. In Sarawak, ethnicity and education levels further influence the uptake. While in Eastern Region, “practice towards information” and “practice on surgery” factors were key barriers. Addressing both common and region-specific factors through targeted strategies, including the enhancement of mobile cataract surgery services, strengthening of public education and expansion of service capacity. Collaborative efforts and further research are needed to improve eye care delivery and ensure equitable access nationwide.

These efforts are vital to improving the uptake of cataract surgery and ensuring equitable access to eye care across the country. In particular, the Eastern Region would benefit from increased investment in infrastructure and service delivery. Collaborative engagement with relevant stakeholders is essential to raise public awareness and knowledge about cataract surgery, while further research among the eye care providers is required to uncover the barriers in the country’s delivery of cataract surgery services.

Supporting information

Acknowledgments

The authors would like to thank the Director General of the Ministry of Health Malaysia for his permission to publish this article. Additionally, the authors acknowledge the significant contributions of data collectors involved in data entry for both NES II and NES III.