Perception Among Diabetic Patients of Telescreening for Diabetic Retinopathy

Abstract

Purpose:

To evaluate the satisfaction of diabetic patients and its determinants of telescreening for diabetic retinopathy (DR) in Saudi Arabia.

Methods:

This cross-sectional survey was conducted in December 2018 in a diabetes center of Riyadh, Saudi Arabia. Ten questions were asked by the interviewer. A 5-point Likert scale was used to grade patient responses for each question. Rasch analysis was conducted to evaluate the response of the participants. The main outcome variable was the sum of the Logit values of the responses. The Rasch score was also compared among subgroups.

Results:

Two hundred (n = 200) diabetic patients were interviewed. The median Rasch score of client-perceived benefit of DR telescreening was +14.6 (25% quartile +3.09, minimum; −23.2, maximum; +35.7). A positive attitude of patients regarding DR telescreening was recorded in 159 (79.5%) participants. The perception of telescreening was better in younger diabetic patients than in older diabetic patients (Friedman p < 0.001). The perception was similar in both genders (Mann–Whitney p = 0.3). Diabetic patients from Riyadh and the southern region of Saudi Arabia had significantly more positive perception than those from north and eastern regions (Freedman p < 0.001).

Conclusion:

Diabetic patients have positive attitude toward tele-DR screening. Their cooperation is likely to be high if large scale tele-DR screening is implemented in the Kingdom.

Background

There are 425 million adults suffering from diabetes worldwide.¹ It is projected that the number of individuals with diabetes will increase to 642 million by 2040.² The genetic etiology of diabetes is complex and difficult to address clinically. Therefore, lifestyle-related factors are actively addressed for the prevention of diabetes and its complications. Obesity, unhealthy diet, sedentary lifestyle, smoking, and physical inactivity are the major modifiable risk factors for diabetes mellitus (DM).³

DM causes diabetic retinopathy (DR).^4,5 The prevalence of DR in type 1 and type 2 diabetes is 84.1% and 50.2%, respectively.⁴ In industrialized countries, complications of DR represent the leading cause of visual disabilities including blindness.^6,7 Ocular changes among diabetic patients include, DR, diabetic macular edema, cataract, glaucoma, optic neuropathy, ocular nerve palsies, and infections. The magnitude of DM and DR in the Middle East and North African (MENA) region (including Saudi Arabia) has reached epidemic levels.^8,9 Proactive and urgent initiatives are recommended for detecting DR cases in the early stages and prompt management of sight threatening DR is warranted.^10
–12

In the Kingdom of Saudi Arabia (KSA), one of the Eastern Mediterranean region member countries, there are nearly 1,032 ophthalmologists in government hospitals and 752 ophthalmologists in the private sector who serve a population of 3.9 million diabetic patients.^13,14 If all the type 2 diabetic patients are screened at least once yearly, these ophthalmologists will have to assess at least 10 diabetic patients daily on every working day over the course of a year. Therefore, task shifting needs to be adopted for DR screening in KSA. Nonophthalmologists and/or mid-level eye care professionals can be trained in screening for DR and they can reduce the burden of detection on the existing ophthalmic services.

The advent of technologies such as digital fundus cameras and smart phones allows retinal images to be captured and easily digitally transferred. Images can be transferred to a reading center by using the secure internet sites, file transfer protocol sites, teleophthalmology services, or cloud-based software. The reading center is manned by trained mid-level eye care professionals who interpret the retinal images under the direct supervision of medical retina specialists and provide feedback to the referring physician where the images were originally captured. The feedback involves informing the referring physician and diabetic patient on the status of DR and the actions required to address any ocular complications of diabetes.^15
–17

There are a number of barriers to the adoption of telescreening for DR, including the expense of digital fundus cameras and telelinkage facilities.^18,19 Other challenges include the lack of mydriatic eye drops at primary health centers (PHCs) due to the risk of precipitating acute angle closure. However, modern nonmydriatic cameras can acquire retinal images through an undilated pupil, providing high-quality retinal images.²⁰ To increase the uptake of telescreening for DR, there the perceptions of PHC staff and diabetic patients need to change. As many as 94% of diabetic patients from rural areas expressed their satisfaction of telescreening for DR as it reduced cost, decreased travel time, and was generally more convenient.²¹

To the best of our knowledge, there is no study evaluating the feedback of diabetic patients on telescreening for DR. In this study, we evaluate the satisfaction of diabetic patients and its determinants of telescreening for DR in KSA.

Methodology

In this cross-sectional study, old and new diabetic patients from the Salman Diabetes Center, Riyadh, Saudi Arabia, were invited for a 2-month period in 2018 for an interview. Those who agreed to participate in the survey were included. Those unable to reply the questionnaire due to severe illness or refusing to participate were excluded from the study.

The Institutional Research Board (IRB) approved this study (1816-R). Written informed consent was obtained from each patient who participated in the interview. Patient data were anonymized to maintain confidentiality. This study adhered to the tenets of the Declaration of Helsinki for research on humans.

The sample size was calculated using OpenEpi²² to represent 10,000 diabetic cases presenting to the Salman diabetes center during the study period. The perception of telescreening for DR was assumed to be adequately positive in 85% of diabetic patients.²³ To achieve 95% confidence interval (CI 95%) and 5% acceptable error margin, 194 ≡ 200 diabetic patients had to be randomly selected for an interview.

The data collection forms were adopted based on the earlier research by Valikodath et al.²⁴ The form was translated from English to Arabic by qualified translators. Reverse translation was performed to ensure that the meaning of the questions did not change.

The questionnaire included demographic information such as age, gender, and area of residence. The area of residence was defined based on the location of residence of the participant; central, eastern, western, northern, and southern regions of Saudi Arabia. Age was determined by the date of birth in the health records.

Ten questions were asked by the interviewer ( Appendix A1). If the participant opted not to respond to a question, the response was reported as “not applicable.” A 5-point Likert scale was used to grade patient responses for each question.²⁵

To ensure a high quality of the study methods and data collected, we conducted a standardization workshop. A pilot trail was conducted for data collection and study methods on 10 diabetic patients who were not involved in the study.

The data were tabulated using an Excel spreadsheet (Microsoft Corp., Redmond, WA). A frequency analysis was performed to check for inconsistencies and missing data. The data were then transferred to Statistical Package for Social Science software (SPSS 24; IBM Corp., Armonk, NY).

The main outcome variable was the sum of the Logit values of the responses of 10 questions for each participant. Rasch analysis was used to evaluate the response of the participants.²⁶ The Rasch score was compared among subgroups of males and females, area of residence, and age. A nonparametric method was used to validate the Rasch score in subgroups. The Rasch score was further divided into four grades as follows: >15 indicated excellent satisfaction, 0.1 to <15 indicated good satisfaction, a score up to −10 indicated poor satisfaction, and a score <−10 indicated very poor satisfaction. The percentage portions and 95% CI of grades of patient perception for DR telescreening were calculated. Qualitative data are presented as the frequency and the percentage proportions. Mean and standard deviations were calculated for normally distributed quantitative variable. If distribution was not normal, the median and interquartile range (IQR) were calculated.

Results

A total of 200 diabetic patients were interviewed for this study. Their mean age was 55 ± 15.9 years. Diabetic patients 40 years and older constituted 84% of the study sample. There were 128 (64%) males and 72 (36%) females. There were 44 (22%) residents from the Riyadh zone, 57 (28.5%) from the northern zone, 52 (26%) from the southern zone, 45 (22.5%) from the western province, and 2 (1%) residents from the eastern province. Apart from eastern province, most of the other areas of Saudi Arabia had approximately equal representation.

The median Rasch score of client-perceived benefit of DR telescreening was +14.6 (25% quartile +3.09, minimum; −23.2, maximum; +35.7). Forty-one (20.5%) diabetic patients had a negative Rasch score.

A positive response denoted a positive attitude of patients regarding DR telescreening. A positive response was recorded in 159 (79.5%) of the diabetic patients who participated in the study. Table 1 presents the grading of the patient response. More than half indicated excellent satisfaction and one-fourth had a positive outlook of DR telescreening. The patient feedback on the use of telescreening for DR was promising.

Table 1.

Grades of Diabetic Patient's Perception About Diabetic Retinopathy Telescreening (Rasch Score)

	N	%	95% CI
Excellent (>15 score)	110	55.0	48.1–61.9
Good (0.1 to 15 score)	49	24.5	18.5–30.5
Poor (<−10 to −0.1 score)	17	8.5	4.6–12.4
Very poor (>10.0 score)	22	11.0	6.7–15.3

CI, confidence interval.

Table 2 presents the outcomes from previous studies of the feedback of diabetic patients and their satisfaction with telescreening.^24,27,29

Table 2.

Diabetic Patient's Feedback and Satisfaction Regarding Telediabetic Retinopathy Screening

S NO.	AUTHOR	COUNTRY	YEAR	SAMPLE	REMARKS
1	Sreelatha and Ramesh²⁷	India	2016	6 articles	Review article shows positive impact and patient satisfaction.
2	Kurji et al.²³	Kenya	2013	26	88% satisfied with teleophthalmology services.
3	Valikodath et al.²⁴	United States	2017	97	69% believed that telemedicine would be more convenient than in-person eye examination.
4	Rotvold et al.²⁸	Norway	2003	42	High degree of satisfaction with the telemedicine examination.
5	Kumar S et al.²⁹	Australia	2003	118	98% expressed satisfaction with internet-based consultation
6	Paul et al.³⁰	South India	2006	348	99.8% satisfaction among rural patients.
	Present study	Saudi Arabia	2018	200	Excellent perception in 55% and good in 24.5% diabetic patients.

The correlation of the age of diabetic patients with the perception of telescreening using a nonparametric method was statistically significant (p < 0.001). Rasch analysis of the subgroups is presented in Table 3. As diabetic patients become older, their perception of telescreening for DR seems to be less positive.

Table 3.

Age Group of Diabetic Patients and Perception for Diabetic Retinopathy Telescreening

AGE GROUP	NO. OF PATIENTS	RASCH SCORE		VALIDATION
AGE GROUP	NO. OF PATIENTS	MEDIAN	INTERQUARTILE RANGE	VALIDATION
<40 years	32	27.8	15.0; 30.2	Friedman p < 0.001
41–50 years	34	19.0	9.8; 27.5
51–60 years	54	14.2	2.5; 26.5
>60 years	80	13.4	−5.7; 26.7

The Rasch score of 128 male diabetic patients indicated a median perception score of +12 (IQR 9.2; 18.0), and in 72 females, it was +16 (IQR 10.0; 18.0) (Mann–Whitney p = 0.3).

The Rasch score of patient perception of DR telescreening was compared among diabetic patients residing in five regions in KSA, Table 4. The diabetic patients from Riyadh and the southern region of Saudi Arabia had significantly higher Rasch scores than those in the north and eastern regions of KSA (Freedman p < 0.001).

Table 4.

Area of Residence and Patient's Perception for Diabetic Retinopathy TeleScreening

REGION	NO. OF PATIENTS	RASCH SCORE		VALIDATION
REGION	NO. OF PATIENTS	MEDIAN	INTERQUARTILE RANGE	VALIDATION
Central (Riyadh)	44	23.4	−5.1; 30.6	Friedman p < 0.001
North	57	14.2	2.9; 26.4
Eastern	2	8.5	−11.0; —
Western	45	17.6	6.7; 26.7
Southern	52	24.5	10.1; 27.1

Discussion

The outcomes of this study indicate that the majority of diabetic patients presenting to a diabetes center in Saudi Arabia had a positive perception of telescreening for DR. One in five diabetic patients had a negative perception. Diabetic patients who were younger and residents of the southern and central regions of Saudi Arabia were significantly more likely to have a positive perception. This perception was similar between genders.

The feedback of stakeholders such as diabetic patients and image readers is crucial for strengthening the telescreening system for DR. Saudi Arabia is in early stages of testing different models of telescreening for DR. Hence, the outcomes of this study are useful for ensuring that telescreening is more acceptable to diabetic patients and for establishing image reading centers. This is perhaps the first study in the KSA and in the Gulf countries that screened a large sample of patients with DR screening. Screening a large proportion DM cases for DR remains a major challenge.^9,31

The selection of diabetic patients from a busy governmental diabetes center was beneficial, since in this center, comprehensive diabetic care is provided to patients from different areas of the KSA. Thus, in this study, the perception of diabetic patients from all five regions was represented. In this study, a pretested questionnaire was adopted and translated to the local language.²⁴ Hence, the quality of patient perception is less likely to be affected by bias due to language difficulties. The images used to evaluate the quality in this study were captured in service scenario at PHCs and transferred to the reading center of a tertiary eye hospital. Therefore, application to a wider diabetic population within a national diabetes health framework is possible. The information transfer gap identified in this study should be addressed to further improve telescreening for DR.

We noted a positive outlook of telescreening for DR among diabetic patients. We compared patient perception in our study with the previous literature.^24,27
–29 Only one study enrolled a larger sample size than our study. Other studies were affected by nonresponse bias. The majority of studies focused on ophthalmic patients with diabetes. Valikodath et al.²⁴ published a study similar to ours. However, we used Rasch analysis for grading patient responses and the percentage proportion of DM patients with a good or better perception of telescreening for DR. Thus, we believe our study outcomes are comparable with previous studies and provide more solid evidence.

Diabetic patients residing in the central region of Saudi Arabia had a very positive attitude toward telescreening for DR. The presence of large cities such as Riyadh (population of 6.5 million) with a more educated population could result in a positive perception of the use of technology for screening cases of DR. Diabetic patients in the southern region of Saudi Arabia also expressed a positive perception of telescreening of DR. This could be due to longer distances they have to travel to see retinal specialists.³² There were very few cases from the eastern region likely due to the presence of a well-established diabetes and eye center in Dhahran.

In this study, age was negatively correlated with patient perception of telescreening for DR. Similarly, Gustke et al.³³ evaluated the barriers to using telemedicine for consultations and noted that older age negatively influenced uptake. Younger generations are more computer savvy and are more likely to accept telescreening initiatives.

This study found no differences between genders regarding the perceived utility of telescreening. This observation is similar to a Vietnamese study that reported that gender is not a significant barrier to accessing tertiary level eye care services.³⁴ In India, the gender gap in cataract surgery decreased if facilities were provided at primary eye care and camp levels instead of at urban eye hospitals.³⁵ With telescreening at the PHC level in the KSA, the gender barrier can be addressed.

There were some limitations to this study. The patients were recruited from one diabetes center. The diabetic patients were registered and managed at primary health centers in the KSA, and those not within the catchment area of the diabetes center were not represented in this study.³⁶ Hence, we recommend judicious interpretation of the outcomes of this study to the general pool of diabetic patients.

In view of a large number of diabetic patients and limited resources in areas within the Kingdom especially remote regions, telescreening for DR is the solution. Telescreening will help achieve annual DR screening targets and reduce the burden of early detection from retina specialists so that they can focus on managing advanced cases of DR. The perception of diabetic patients is positive for adopting these technologies.

What is Known in the Literature

Patient's cooperation is vital for the success of a national telescreening program.

Telescreening for DR is cost-effective and client friendly.

Tele-DR screening is widely accepted in industrialized countries as an alternative to screening in eye hospitals.

It has better acceptance if diabetic patients perceive it as useful to them.

What is Added in the Literature

The perception for telescreening for DR is high among diabetic patients of Saudi Arabia.

The perception for tele-DR screening varies with age and area of residence of the diabetic patient.

Health promotion needs to be focused more on subgroups noted with less positive perception for tele-DR screening in the national screening initiative.

Footnotes

Acknowledgments

This article is part of MPH thesis for K.A.-F. We thank the staff of Salman Diabetes Centre, Retina Clinic of King Khaled Eye Specialist Hospital (KKESH), and Research Department of KKESH for extending support to this research. We are also thankful to Mr. Ches Souru and Mr. Sejo Varghese for data entry of the survey forms.

Authors' Contributions

All four authors were involved in planning the research and writing the article. All authors with the exception of R.K. were involved in the field part of the study.

Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

Appendix

References

Key massages in International Diabetes Federation Diabetes Atlas-8th ed. 2017. Available at www.diabetesatlas.org/key-messages.html (last accessed March 19, 2018 ).

Silink

. A United Nations resolution on diabetes–The result of a joint effort. US Endocrine Disease, 2007; 1:12–14.

Mutie

, Giordano

, Franks

. Lifestyle precision medicine: The next generation in type 2 diabetes prevention?. BMC Med, 2017; 15:171.

Maric-Bilkan

. Sex differences in micro- and macro-vascular complications of diabetes mellitus. Clin Sci (Lond), 2017; 131:833–846.

Naskar

, Victor

, Nath

. Depression in diabetes mellitus-A comprehensive systematic review of literature from an Indian perspective. Asian J Psychiatr, 2017; 27:85–100.

Hernández

, Simó-Servat

, Bogdanov

, Simó

. Diabetic retinopathy: New therapeutic perspectives based on pathogenic mechanisms. J Endocrinol Invest, 2017; 40:925–935.

Wang

, Cheung

, Tapp

, Hamzah

, Tan

, Ting

, et al. Availability and variability in guidelines on diabetic retinopathy screening in Asian countries. Br J Ophthalmol, 2017; 101:1352–1360.

Alshaikh

, Filippidis

, Al-Omar

, Rawaf

, Majeed

, Salmasi

. The ticking time bomb in lifestyle-related diseases among women in the Gulf Cooperation Council countries; review of systematic reviews. BMC Public Health, 2017; 17:536.

Khandekar

. Screening and public health strategies for diabetic retinopathy in the Eastern Mediterranean region. Middle East Afr J Ophthalmol, 2012; 19:178–184.

10.

Ting

, Cheung

, Wong

. Diabetic retinopathy: global prevalence, major risk factors, screening practices and public health challenges: A review. Clin Exp Ophthalmol, 2016; 44:260–277.

11.

Sabanayagam

, Yip

, Ting

, Tan

, Wong

. Ten emerging trends in the epidemiology of diabetic retinopathy. Ophthalmic Epidemiol, 2016; 23:209–222.

12.

Das

, Aurora

, Chhablani

, Giridhar

, Kumar

, Raman

, et al. Evidence-based review of diabetic macular edema management: Consensus statement on Indian treatment guidelines. Indian J Ophthalmol, 2016; 64:14–25.

13.

Ministry of Health, Statistical book (table 1–3, page 30), (Table 1, page 29) Available at https://www.moh.gov.sa/en/Ministry/Statistics/book/Documents/Statistical-Yearbook-1437H.pdf (last accessed March 25, 2018 ).

14.

IDF Diabetes Atlas—8th ed. Available at www.diabetesatlas.org/ (last assessed March 25, 2018).

15.

Mohammadpour

, Heidari

, Mirghorbani

, Hashemi

. Smartphones, tele-ophthalmology, and VISION 2020. Int J Ophthalmol, 2017; 10:1909–1918.

16.

Caffery

, Taylor

, Gole

, Smith

. Models of care in tele-ophthalmology: A scoping review. J Telemed Telecare, 2019; 25:106–122.

17.

Kanjee

, Dookeran

. Tele-ophthalmology for diabetic retinopathy in Canada-meeting the needs of a growing epidemic. Can J Ophthalmol, 2017; 52(Suppl 1):S13–S14.

18.

Zimmer-Galler

, Kimura

, Gupta

. Diabetic retinopathy screening and the use of telemedicine. Curr Opin Ophthalmol, 2015; 26:167–172.

19.

Kanjee

, Dookeran

, Mathen

, Stockl

, Leicht

. Six-year prevalence and incidence of diabetic retinopathy and cost-effectiveness of tele-ophthalmology in Manitoba. Can J Ophthalmol, 2016; 51:467–470.

20.

Roser

, Kalscheuer

, Groener

, Lehnhoff

, Klein

, Auffarth

, et al. Diabetic retinopathy screening ratio is improved when using a digital, nonmydriatic fundus camera onsite in a diabetes outpatient clinic. J Diabetes Res, 2016; 2016:4101890.

21.

Rani

, Raman

, Manikandan

, Mahajan

. Patient satisfaction with tele-ophthalmology versus ophthalmologist-based screening in diabetic retinopathy. J Telemed Telecare, 2006; 12:159.

22.

Dean

, Sullivan

, Soe

. OpenEpi: Open Source Epidemiologic Statistics for Public Health, Version. www.OpenEpi.com, updated April 6, 2013. (last accessed March 19, 2018 ).

23.

Kurji

, Kiage

, Rudnisky

, Damji

. Improving diabetic retinopathy screening in Africa: Patient satisfaction with teleophthalmology versus ophthalmologist-based screening. Middle East Afr J Ophthalmol, 2013; 20:56–60.

24.

Valikodath

, Levenque

, Wan

, Lee

, Newman-Casey

, Hansen

, et al. Patient attitude towards telemedicine for diabetic retinopathy. Telemed J E Heath, 2017; 23:205–212.

25.

Pell

. Use and misuse of Likert scales. Med Educ, 2005; 39:970.

26.

Boone

, Staver

, Yale

. Rasch analysis in the human sciences. Dordrecht, Netherlands: Springer Science & Business Media, 2013.

27.

Sreelatha

, Ramesh

. Teleophthalmology: Improving patient outcomes?. Clin Ophthalmol, 2016; 10:285–295.

28.

Rotvold

, Knarvik

, Johansen

, Fossen

. Telemedicine screening for diabetic retinopathy: Staff and patient satisfaction. J Telemed Telecare, 2003; 9:109–113.

29.

Kumar

, Tay-Kearney

, Constable

, Yogesan

. Internet based ophthalmology service: Impact assessment. Br J Ophthalmol, 2005; 89:1382–1383.

30.

Paul

, Raman

, Rani

, Deshmukh

, Sharma

. Patient satisfaction levels during teleophthalmology consultation in rural South India. Telemed J E Health, 2006; 12:571–578.

31.

Scanlon

, Aldington

, Stratton

. Epidemiological issues in diabetic retinopathy. Middle East Afr J Ophthalmol, 2013; 20:293–300.

32.

Shrime

, Hamer

, Mukhopadhyay

, Kunz

, Claus

, Randall

, et al. Effect of removing the barrier of transportation costs on surgical utilisation in Guinea, Madagascar and the Republic of Congo. BMJ Glob Health, 2017; 2(Suppl 4):e000434.

33.

Gustke

, Balch

, West

, Rogers

. Patient satisfaction with telemedicine. Telemed J, 2000; 6:5–13.

34.

Van Huy

, Dung

, Thang

, Hanh

. Patient satisfaction with health care services at a national institute of ophthalmology. Int J Health Plann Manage, 2018; 33:e251–e262.

35.

Gupta

, Bansal

, Gupta

, Bhansali

. Sensitivity and specificity of nonmydriatic digital imaging in screening diabetic retinopathy in Indian eyes. Indian J Ophthalmol, 2014; 62:851–856.

36.

Riyadh, Saudi Arabia:. 1994. Quality Assurance in Primary Health Care Manual. MOH G.D. of Health Centers, WHO/EMRO. Available at https://applications.emro.who.int/dsaf/dsa231.pdf (last accessed June 5, 2020 ).