Systematic Review on Barriers and Facilitators for Access to Diabetic Retinopathy Screening Services

Objectives The aim of this systematic review is to identify the barriers/enablers for the people with diabetes (PwDM) in accessing DRS services (DRSS) and challenges/facilitators for the providers. Background Diabetic retinopathy (DR) can lead to visual impairment and blindness if not detected and treated in time. Achievement of an acceptable level of screening coverage is a challenge in any setting. Both patient-related and provider-related factors affect provision of DR screening (DRS) and uptake of services. Methods We searched MEDLINE, Embase, CENTRAL in the Cochrane Library from the databases start date to September 2016. We included the studies reported on barriers and enablers to access DRS by PwDM and studies which have assessed barriers or facilitators experienced by the providers in provision of DRSS. We identified and classified the studies that used quantitative or qualitative methods for data collection and analysis in reporting themes of barriers and enablers. Main Results We included 63 studies primarily describing the barriers and enablers. The findings of these studies were based on PwDM from different socio-economic backgrounds and different levels of income settings. Most of the studies were from high income settings (48/63, 76.2%) and cross sectional in design (49/63, 77.8%). From the perspectives of users, lack of knowledge, attitude, awareness and motivation were identified as major barriers to access DRSS. The enablers to access DRSS were fear of blindness, proximity of screening facility, experiences of vision loss and being concerned of eye complications. Providers often mentioned that lack of awareness and knowledge among the PwDM was the main barrier to access. In their perspective lack of skilled human resources, training programs and infrastructure of retinal imaging and cost of services were the main obstacles in provision of screening services. Conclusion Knowing the barriers to access DRS is a pre-requisite in development of a successful screening program. The awareness, knowledge and attitude of the consumers, availability of skilled human resources and infrastructure emerged as the major barriers to access to DRS in any income setting.


143
There are no published reviews related to this topic. One protocol was available however it is yet to be 144 published. (26) Most of the individual studies had provided the evidence of barriers to access DRSS 8 145 according to the typology of barriers. The processes related to DRS uptake can be considered in three 146 levels i.e., service user, service provider and eyecare system. Therefore, in this review we categorised 147 the reported themes or variables under above categories. This review was specifically assessed the 148 barriers to access DRSS at established health care facilities and challenges / barriers faced by the 149 providers in those institutions. In broad definitions, barriers to access to DRS is not only limited to the 150 access issues at the point of delivery, but it also involves all the steps which take place starting from 151 perceptions of a PwDM at one end to the whole eye care system at the other which are inter-related 152 and connected to each other.

155
The overall aim of the review was to explore barriers to access DRS. The review has the following 156 specific objectives.

157
-To assess the barriers and enablers to uptake of DRSS by PwDM.

158
-To assess the challenges faced by the services providers in provision of DRSS and to identify 159 the facilitators for development of a DRSP.

160
The secondary objectives of this review were; 161 -To assess the socio-economic factors that could affect DRS uptake.

162
-To assess the barriers or enablers to develop DRSP in a health care system.

169
We included studies that focused on assessing barriers / challenges and enablers / incentives to access 170 DRS. In addition, we found studies that described factors affecting the uptake of DRSS. Following 171 criteria were used for assessment of eligibility of the studies. (There is no protocol registration for this 172 review and PRISMA checklist was included as S1

234
In the synthesis of evidence "informants" were authors of the individual studies rather than the 235 participants. The authors' interpretations were presented as narrative themes supported by numerical 236 values of statistical significance levels wherever available. While authors' interpretations were primarily collected from results section of each paper, author 238 interpretations were sometimes also found in the discussions sections and these were also extracted 239 when relevant and well supported by data. 240

241
We carried out the risk of bias and quality assessment according to the guidelines of critical appraisal 242 of skills program (CASP) tools for case-control, qualitative, cohort and randomised controlled study 243 designs (28) and National Institute of Health, United States quality assessment tool for observational 244 cohort and cross sectional study designs (NIH-QAT) for cross sectional study designs.(29) Two 245 reviewers (SK and PN) independently applied set of quality criteria to each included study. We 246 appraised how well the individual studies conducted which contributed to narrative synthesis using 247 the above tools. Emphasis was given more over the applicability of the study according to the 248 inclusion criteria. It has been noted that applicability to review question was the main concern in the 249 synthesis rather than the overall level of quality of a study (S3 Table).

252
When several studies with varied methodological limitations contributed to a finding, we made an 253 overall judgement about the distribution of strengths and weakness of the study rather than for 254 individual components in the tools.

256
We assessed the coherence of each review finding by looking at extent to which we could identify a 257 clear pattern across the data contributed by each of the individual studies. This was supported by 258 when clarity of the themes was consistent across different contexts and the variations were explained 259 by the study authors according to the data collected, when supported by numerical data (odds ratios).

260
This was further strengthened when findings were drawn from different settings 13 261

262
Most of the eligible studies were observational and descriptive in nature hence narrative reporting 263 approach was used in this review. We analysed and synthesised the descriptive and qualitative data 264 narratively supported by other associated variables with levels of statistical significance. We The methodological quality assessments of included studies were presented in S3 Tables 1 to 5  confounders were not adjusted in ten studies (10/50, 20%).

288
Acceptable method of recruitment of the cohort was not followed in all three included cohort studies.

289
In included randomised controlled study designs, applicability of the results to the PwDM was not 290 observed in two studies (2/3, 66%). In qualitative study designs, most of the quality assessment 291 criteria were met except, relationship between researcher and the participants were not adequately 292 considered in two studies (2/5, 40%). The following main themes were derived from descriptive and qualitative studies (S5 Table 1 to 4).

341
The most prominent barrier to access DRS among the consumers in LIC were lack of knowledge on 342 DM eye complications, lack of awareness about importance of eye examination and lack of 343 knowledge about availability of eye clinics. Among providers, main challenges were lack of skilled 344 human resources and lack of access to DR imaging and treatment infrastructure. Further non-345 existence of a referral system and lack of multi-disciplinary care approach were barriers to provision 346 of DRSS. In LIC lack of a national policy and competing disease priority environments were the main 347 obstacles in the system (S5 Table 1).

349
Consumers' barriers related to knowledge and awareness could be observed in the LMIC as well.  Table 2).

359
The lack of awareness and knowledge on DR emerged as the main barrier among the PwDM in 360 UMIC. Poor physician-patient communication was also a barrier in these countries. In provider 361 perspectives scarce human resources, lack of training, high number of PwDM were the main 362 challenges faced. In the system analysis limitations in prevention and health promotion, civil unrest, 363 disparity in urban and rural services, lack of transportation and problems in insurance schemes were 364 the main barriers to accessing DRSS (S5 Table 3 4). Enablers for providers were educating the users on regular eye examination and 647 providing better access for PwDM (S5 Fig 5). Limitations

649
The included studies reflected the barriers in a cross section of a time. All the studies used diagnosed 650 PwDM at institutional level as their study samples. There were no studies that used long term 651 sociological and ethnographic approaches to study barriers to access long term in their natural 652 environment.

653
Many of the barriers or enablers identified in this review was peculiar to modality of screening in the 654 local context. We used reductionistic approach in this narrative synthesis without further synthesis of 655 new themes. Another aspect is the barriers or enablers were assessed in different health systems which 656 has different socio cultural and economic back grounds. Therefore, we could not assess the 657 interactions in between each theme we derived. Though we simplified and de-contextualised the 658 barriers themes, generalizability may depend on the context.

659
One of the limitations of this review is lack of eligible randomised controlled trials on this review 660 question and primary outcomes were described as explained by the authors. Considering the paucity 661 of systematic reviews under this topic, it is difficult to compare and comment our findings.

663
Implications and public health significance of the findings 664 Diabetic retinopathy screening program implementation involves a high capital expenditure. There 665 will be a high level of financial risk when implementing a program for the first time. By knowing the 666 potential barriers, the risks can be minimised, and access can be improved by implementing 667 interventions to overcome potential barriers.

668
The outcomes of current review will be useful to identify the modifiable barriers which could be 669 further explored in a local context before implementing costly interventions. Identification of user and 670 provider perspectives together will enable to identify and cater needs of demand side as well as supply