1.1 Study aim

This study aims to review the contextual issues associated with the access of screening services for Diabetic Retinopathy in sub-Saharan Africa.

2.1 Framework

The Population, Concept and Context (PCC) framework for conducting scoping reviews recommended by Joanna Briggs Institute (JBI) in 2013 was used to develop the review question [28]. Based on this method of evidence synthesis, we deduced the following and posed the review question below:

Population of interest: People living with Diabetes

Concept: Access of Diabetic Retinopathy screening services (DRSS).

Context: sub-Saharan Africa

2.2 Search strategy

To answer the review question, the Elton B. Stephens Company (EBSCO) host platform was accessed through the University of Limerick Glucksman Library website from 29^th May 2023 to 30^th June 2023. Electronic databases searched include MEDLINE with full text, PubMed, Embase, APA PsychINFO, Web of Science, CINHAL Complete, and Africa Journals Online. These databases were searched to ensure comprehensive coverage, capture discipline-specific studies, and avoid selection bias. There was no publication time limit on the searches because of the paucity of available studies on the topic after a preliminary search.

Searches were run for these terms in the title, abstract and full-text fields. To guide our search properly, we searched related terms for our key concept “Diabetic Retinopathy Screening Services” under the medical subject headings MeSH terms for MEDLINE, CINAHL subheadings for CINAHL complete using a combination of Boolean operators (AND, OR, NOT) and other search terms to narrow or broaden the search as required. Keywords in the studies we found were reviewed and references lists were scanned to expand, refine, and modify our search terms as required. Utmost care was taken to document our search strings after each session to keep track of all queries. Retrieved articles were exported to the reference manager, Endnote, for proper filing. Hand searches of grey literature were equally conducted.

2.3 Study selection

Articles found on the electronic research databases were imported to the artificial intelligence software application, Rayyan, where they were scrutinized for duplicates [29]. These duplicates were subsequently removed. Two authors (IN and AB) independently screened the title and abstracts of the remaining studies identified from the databases on Rayyan. Articles were assessed using a set of eligibility criteria (outlined below), after which conflicts were discussed. These conflicts were resolved by consensus and through arbitration of a review tiebreaker (NI) as required. The Preferred Reporting Items for Systematic review and Meta-Analysis (PRISMA) flow diagram was then created to illustrate the entire process of study selection. Reasons for exclusion of articles at the full-text stage were provided.

2.4 Data extraction

The full text of the included studies was read, and relevant data was extracted using Microsoft Excel®. Each article was assigned a number for easy identification. Information extracted include author, title, year of publication, study location (country and setting), study population and sample size, research methods used, emerging themes and 5As dimensions. Two authors (IN and NI) worked independently on data extraction, and iteratively identified emerging themes. Thematic synthesis was conducted in three overlapping steps involving line by line coding of the findings from the included studies, grouping these codes into descriptive themes, and generating analytical themes from these categories using the 5As framework [30]. The data extraction table was pilot tested to refine the categories, ensure clarity and promote consistency across the reviewers for more reliable data. A risk of bias or critical appraisal of the included studies was not conducted in this scoping review as it is considered optional [31].

2.5 Collating, summarizing and reporting the results

A narrative account of key findings from the included literature was summarized after collating and analysing the extracted data for emerging themes. To shed more light on some contextual factors related to the access of DR screening services in sub-Saharan Africa, we used the 5A’s conceptual framework of access to healthcare by Levesque et al (2013) to map and explain the emerging themes from the studies.

3.1 General characteristics of included studies

Nineteen studies published in different scientific journals between 2007 and 2022 were included in this review. All the studies were observational studies using cross-sectional study design to assess the factors associated with the access of DRSS. Three of the studies [#6, #11, #19] investigated DRSS access from the service provider’s point of view, while 15 articles [#1, #2, #3, #4, #5, #7, #8, #9, #10, #12, #13, #14, #15, #16, #18] assessed from the patients’ standpoint. Study #17 explored both patient and service provider perspectives. Of all 19 studies, 14 studies were quantitative studies relying on interviewer administered questionnaires, 2 studies [#11 and #17] were qualitative studies involving face-face-face semi-structured interviews while 3 articles [#5, #9 and #10] used mixed methods. 10 of the studies used systematic random sampling technique to recruit study participants, while 6 of the studies resorted to convenience sampling. Two studies [#17 and #19] utilized purposive sampling while study #6 relied on proportionate allocation. The sample sizes ranged from 27 to 424 participants. With regards to country and setting, the studies were conducted in various primary, district and tertiary healthcare facilities across sub-Saharan Africa (5 in Nigeria, 1 in Ghana, 2 in Ethiopia, 4 in Tanzania, 3 in Kenya, 3 in South Africa, 1 in Zimbabwe). See Table 1 and Fig 2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Geographical distribution of included studies and 5A dimensions as reported.

https://doi.org/10.1371/journal.pone.0315367.g002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Included studies, emerging themes and 5A dimensions/abilities.

https://doi.org/10.1371/journal.pone.0315367.t001

3.2 Thematic synthesis

Seven themes (health literacy, health-related factor, health beliefs, demographic factors, psychological factors, financial factors; residence, mobility, and social factors) emerged from the review in relation to the patients while five themes (service reach, service cost, screening competency, service accessibility and service coordination) emerged in relation to service providers.

Patient-related themes were categorized using the five abilities of the population in the 5As conceptual framework as follows:

Ability to perceive

• Health Literacy: This was one of the popular themes emerging from 63% of the studies [#2, #4, #5, #8, #9, #10, #13, #14, #15, #16, #17, #18]. These studies assessed the level of awareness and knowledge of Diabetes, Diabetic Retinopathy, and importance of periodic retinal screening among patients living with Diabetes. Majority of the studies showed moderate to high levels of awareness and knowledge of Diabetic Retinopathy, while in 4 studies [#8, #10, #16, #17], this was reportedly low.

• Health-related Factors: According to the studies, duration of Diabetes was one of the single largest determinants of DR screening and this was mentioned in 47% of the studies [#1, #4, #7, #9, #11, #13, #14, #15, #16]. It was found that a shorter duration of Diabetes was frequently associated with defaulting on screening appointments. The effect of the type of Diabetes, family history of Diabetes, and the presence of comorbidities and history of previous eye disease were equally assessed in the studies.
• Health Beliefs: The influence of health beliefs underlying the perception of ill health and the need to seek care were equally explored. Studies [#5, #12, #17, #18] highlighted some of the factors such as stigma, fear of surgery, fear of vision loss and denial that could impact the choice of a person living with Diabetes to seek care. For example, in study #17 by Wentzel and Mchiza, fear of vision loss was cited by one of the key informants as a reason for seeking care and this outweighed fear of the surgical procedure. Commenting on this, one of the key informants interviewed stated “The main thing, I think they (patients) are scared”. Some of the patients also admitted their fear of treatment, by making remarks like “I was scared. I didn’t know what to expect” (Patient 8). “… if I don’t do it then I’ll eventually go blind” (Patient 3); “… probably my fear that I’ll go blind if I don’t go” (Patient 7).

Ability to seek

• Demographic Factors: Seven studies [#4, #7, #8, #12, #13, #14, #16,] investigated the association between age and access of DRSS, with majority of the studies stating that age at onset was a significant predictor of DRSS access. While study #14 found strong evidence of association between having a dilated eye examination (ever) and increasing age, the effect size was quite small. Study #12, on the other hand, reported that age was not associated with the access of DRSS. Studies [#5, #16, #18] reported on the influence of gender on the access of Diabetic Retinopathy screening services with studies #5 and #18 finding that being male increased the likelihood of accessing DRSS. Four studies [#5, #7, #8, #9] examined the impact of the level of education and marital status on DR screening access while studies [#8, #9 and #14] looked at the effect of employment status and residence on DRSS access. Two studies [#2 and #5] quizzed respondents on the impact of cultural beliefs on screening access. While 53% of the respondents in study #2 reported that cultural and spiritual beliefs may be the reason why people did not access DRSS, it did not specify the specific cultural and spiritual beliefs that could have been responsible for this trend. Some study participants in study #5 cited the limited appreciation of the chronicity of Diabetes and a lack of understanding of the concept of being “sick” without symptoms as some cultural beliefs that led to poor DRSS access. Commenting on this, one respondent said, “Most Africans don’t have the habit of checking our health status until we get sick.”

Ability to reach

• Residence, Mobility and Social Factors: Three of the included studies [#5, #12, #17] stressed the role of family, friends, and other forms of social support on the access of DR screening. The need for an escort to screening appointments and its attendant implications were highlighted by both patients [#5, #12] and service providers [#17]. The impact of current state of health and mobility on the access of available DR services was explored in studies [#7, #10, #16, #17]. In study #10, the current status of eye health was investigated and 50.8% of people living with Type 2 Diabetes in the study reporting they had no previous dilated eye examination because they had no eye problems/visual symptoms. Similarly, being in a poor state of health as a result of Diabetes or another disease was highlighted as a major barrier to the access of DRSS. Three studies [#8, #9, #14] reported on the influence of urban vs rural residence on access to DRSS, with living in urban areas reported as having a strong association with access of DRSS.

Ability to pay

• Financial factors: Studies [#4, #5, #8, #13] explored the effect of different income levels and financial status on the access of DR screening services. Here, financial constraints and low income were reported as the reasons for the poor DRSS access. For example, in study #5, women were less likely to attend DR screening appointments because they did not have the financial independence to do so since in most families in the study, financial decisions were made by men. The availability of medical insurance coverage was equally highlighted as a predictor of access to DR screening facilities in studies [#5 and #7] while financial trade-offs and competing priorities were found to be significantly associated with the access of DRSS in studies [#12, #17, #18].

Ability to engage

• Psychological Factors: Five studies [#9, #10, #12, #17, #18] explored the association between self-efficacy, attitude towards DR, anxiety and the access of DR screening services. In studies #12 and #18, lack of motivation, forgetfulness and a negative attitude towards DR were reported to be significantly associated with poor access of DR screening services (DRSS).

Service provider related themes were classified using the 5As service dimensions as follows:

Approachability

• Service Reach: Three studies [#4, #8, #15] asked study participants about their source of information on Diabetes and diabetic eye disease, with majority of the study participants reporting healthcare providers as the main source of information. Other sources of information mentioned in the studies include friends and relatives, mass media and eyecare professionals. The lack of outreach services and mobile clinics was equally identified as a major obstacle to DR screening access in study #11.

Acceptability

• Screening Competency: Three studies [#6, #11 and #19] asked healthcare providers about their experience levels in conducting DR screenings and how knowledgeable they are about Diabetic Retinopathy, most of the participants reported having good knowledge of diabetic ocular complications, however, it was also determined that only a few of the study participants routinely conducted retinal examinations on diabetic patients.

Availability and accommodation

• Service Accessibility: Information on the accessibility of Diabetic Retinopathy screening service (DRSS) was solicited from service providers. Four studies [#3, #9, #10, #17] reported the ease of obtaining DR screening appointments as predictors of service utilisation while long waiting times was identified as a notable impediment in studies [#3, #17, #18]. Six studies [#2, #9, #10, #11, #12, #18] noted that travel distance to the screening service was a considerable predictor of DR screening access. Other factors indicated in the studies in relation to availability and accommodation include shortage of ophthalmic workforce [#9 and #19] and in study #17, the Covid19 pandemic was also mentioned as a remarkable cause of Diabetic Retinopathy screening service disruption during the lockdown period.

Affordability

• Service Costs: The direct cost of treatment was mentioned as a notable determinant of Diabetic Retinopathy screening services access in 5 studies [#5, #7, #9, #10, #12]. Indirect expenses such as transport and accommodation costs were equally highlighted in study #5 and #12 while the mode of health service payment was considered as a significant factor in study #7.

Appropriateness

• Service Coordination: Absence of functional screening equipment like ophthalmoscopes, fundus cameras, and diagnostic medications were highlighted as major hinderances to DR screening access in studies [#3, #6 and #11]. Task sharing between eye care professionals and the non-ophthalmic workforce were also identified as significant predictors of DR screening compliance in studies [#3, #9, #11, #17 and #19]. Some of the study participants in 37% of the studies reviewed [#1, #2, #4, #12, #14, #16, #17] reported there was poor access because their healthcare professional did not refer them to the Diabetic Retinopathy screening service (DRSS). Similarly, providing explanations on the outcome of retinal screening by healthcare professionals in studies [#3, #9, #17] and communication between services involved with the care of people living with diabetes were considered as factors affecting access of DRSS in studies #11 and #19 respectively.

The 5As framework was used to map these emerging themes as follows. See Table 2 and Fig 3.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Mapping DR screening services using the 5As framework of access to healthcare.

https://doi.org/10.1371/journal.pone.0315367.g003

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Thematic synthesis using the 5As framework of access to healthcare.

https://doi.org/10.1371/journal.pone.0315367.t002

4.1 Approachability/Ability to perceive

In terms of approachability of DR screening services in sub-Saharan Africa, most of the studies revealed screening was conducted mainly on an opportunistic basis since very few screening centres in the studies [#3 and #12] reported the existence of an electronic database for people living with Diabetes. This highlights remarkable infrastructural deficits in the health information systems needed for the invitation, recall and referral components of screening services [18]. Health care providers were cited as the main sources of information on regular eye checkup practices for people living with Diabetes in studies #4, #8 and #15 [35,39,46]. For service users, the ability to perceive is largely influenced by the level of knowledge and awareness possessed by an individual in relation to their health condition. Health beliefs like fear of blindness, stigma, and denial also play a vital role in the ability to perceive and seek care in these settings [41].Though regarded as a fundamental strategy used in improving health literacy, it was evident in a few of the studies that creating awareness of a disease condition alone does not necessarily guarantee attitudinal change that will lead to compliance. This was the case for both patients and service providers in studies #4 and #6 [35,37]. This finding tallies with a similar study in Indonesia which investigated how knowledge of DR improved screening practices among general practitioners [50]. Conversely, studies in Turkey, Hong Kong, and Saudi Arabia found significant associations between knowledge of DR and increased DR screening services access and treatment adherence [51–53]. This indicates that compliance to DR screening or treatment could be determined by a combination of factors with different variables playing a role in the process [54]. For instance, in study #2, some participants were non-compliant because they believed retinal screening would delay the process of acquiring new spectacles. This is because patients had to undergo retinal examinations first to ensure there were no abnormalities that would preclude prescribing of spectacles in that facility. This demonstrates how preconceived notions could impact compliance to screening [33]. Duration of Diabetes was frequently considered as a major predictor of Diabetic Retinopathy screening service (DRSS) access, with a shorter duration of Diabetes linked to poor compliance. This outcome was also resonated in a study carried out in Jeddah, Saudi Arabia[55] which found that duration of Diabetes and age were significant predictors of DR among people living with Diabetes.

4.2 Acceptability/Ability to seek

The rate of DR screening by eyecare practitioners and other healthcare providers in the reviewed studies was reportedly low as seen in studies #6, #11 and #19. Studies conducted in a high-income country like the United States of America also recorded a similar trend [56]. To ensure a screening service is acceptable, it is crucial to make sure the service is not only of high quality but culturally sensitive [57]. Demographic factors like age, gender, marital status, cultural beliefs and other social constructs have been shown to influence health seeking behaviour [58]. For instance, increasing age was noted as a major predictor of access to Diabetic Retinopathy screening services with a greater degree of non-compliance recorded in younger age groups[44,47]. This was also corroborated by a study in the United Kingdom which concluded that younger adults living with Diabetes had greater rates of screening non-attendance compared to older individuals [59]. In three of the studies [#5, #16, #18] that assessed the impact of gender on DR screening access, males had a higher rate of service utilisation in two studies, while study #16 found the contrary. The provision of gender sensitive interventions like the use of female health care workers to attend to diabetic women has been shown to improve screening compliance [60]. This review also revealed varying degrees of association between marital status and DR screening service access. In study #7, for instance, people who were divorced were more likely to utilise eye health services than their married counterparts [38]. Conversely, in a study of barriers to healthcare for women of different ethnic and racial backgrounds, it was shown that being single or divorced negatively impacted access to healthcare services [61].

4.3 Availability & Accommodation/Ability to reach

From the reviewed studies, we can deduce that where an individual resides and their proximity to healthcare facilities are important correlates of access to Diabetic Retinopathy screening services. People residing in urban or suburban settlements are more likely to gain quicker access to health information and tend to avail themselves of healthcare services more than their rural counterparts [62]. This also includes the availability of transportation, housing, and other forms of services to ensure optimal access to health services. The state of health and ambulation of a patient is also central to how they interact with the health system [63]. Some people might require the services of escorts to accompany them to their screening appointments and this could have far-reaching economic consequences for the parties involved, hence the presence of a robust social support system is considered a key determinant of screening adherence [36]. A study carried out in Egypt showed that lack of social support correlated positively with higher amounts of DR treatment drop-outs [64]. For service providers, availability entails the ease of booking screening appointments, waiting times and service convenience. The severe shortage of the eyecare workforce has been identified as a primary contributor to the inaccessibility of DR screening services [65]. Beyond recruiting additional human resources for eye health services, ensuring equitable distribution of staff across rural and urban areas is paramount [66]. Study #17 equally highlighted the service disruption brought about by unique circumstances such as the COVID-19 pandemic. This was true not just for DR screening services, but all other screening services regarded as non-essential services during the lockdown period [67].

4.4 Affordability/Ability to pay

Owing to limited resources and financial hardships that characterize sub–Saharan Africa, issues of direct and indirect cost of service, absence of medical insurance and competing priorities were mentioned as notable barriers to DRSS access. These prohibitive costs of care were equally highlighted as significant barriers to the timely access of DR screening services in a multi -country study to determine global perspectives on access to DR screening and treatment[68]. While the cost of DR screening and treatment is generally regarded as a major hinderance, it is noteworthy that some of the respondents in the reviewed studies did not consider DR screening a priority as seen in studies #10 and #12 respectively. This lack of prioritisation could influence willingness to pay for Diabetic Retinopathy screening services (DRSS) and this was investigated in a previous study in Taiwan [69]. In this Taiwanese study, it was found that level of income did not affect willingness to pay as much as level of education and degree of severity of Diabetic Retinopathy. Likewise, the availability of a robust medical health insurance scheme is considered an important predictor of DR screening and health service access [70]. Given the widespread poverty in sub-Saharan Africa, most people living with Diabetes ultimately rely on out-of-pocket expenses to keep up with their health needs leading to poor DRSS access as seen in study #5.

4.5 Appropriateness/Ability to engage

The introduction of innovative solutions and approaches like fundus photography, ocular coherence tomography, teleophthalmology and other novel means of improving screening efficiency has been hailed as major turning points in the trend of DR screening in resource-constrained settings [71]. Granted, these might go a long way to improve coverage in geographically inaccessible areas, however, the training of non-physician clinical officers and nurses on how to use these devices and interpret retinal photographs to accurately identify true cases of vision-threatening Diabetic Retinopathy (VTDR) is also needed [72]. These task-sharing strategies, when implemented at the primary care level could cushion the enormous workload on a terribly understaffed eye care sector region-wide [73]. Even though this is generally a capital-intensive venture with a huge potential to yield promising return on investments, the cost of maintenance and other overhead cost incurred over time are also worth-considering [74].

Finally, the absence of clear referral protocols to ophthalmologists by general practitioners or other qualified healthcare practitioners was highlighted as an impediment to the timely access of DRSS in studies [#1, #2, #4, #12, #14, #16, #17]. This could possibly emanate from insufficient knowledge on how to grade different stages of DR or lack of an existing referral pathway. Similarly, a systematic review involving a couple of low-and-middle-income countries revealed a national DR screening protocol is non-existent in most sub-Saharan African countries [2]. Furthermore, knowledge of the referral pathway alone does not suffice, the need for stringent follow-up care cannot be overemphasized [75]. Efforts to keep the channels of communication between the diabetic clinics and ophthalmology departments open is paramount [42]. There is need to develop integrated models of care for all facilities involved in the care and management of people living with Diabetes [76]. In addition, incorporating a system of constant DR screening reminders through mass media campaigns and personalized phone reminders could encourage people living with Diabetes to seek timely care [77]. Considering that people with vision-threatening Diabetic Retinopathy are often asymptomatic, committing to regular Diabetic Retinopathy screening using appropriate technology and adequately trained personnel could potentially ensure timely interventions culminating in better visual outcomes for people living with Diabetes [78,79].