Advanced technologies have altered long-standing processes within customer-based industries, resulting in greater efficiency, productivity, and quality of service.1,2 Self-service technology (SST), a technological interface that allows business operators to produce a service without the involvement of a service employee, is a proven business model that generates favorable impact in industries such as airports, hotels, and retail.3,4 In settings with long lines and wait times, self-service kiosks have been shown to be a valuable tool in enhancing operational efficiencies and have been implemented across virtually all service industries,5 including health care.
Recent studies have evaluated the utilization of check-in kiosks within emergency departments (EDs), where overcrowding has been shown to have a significant detrimental impact on various clinical outcomes.6,7 In one clinical trial by Sinha et al., of 400 participants visiting a pediatric ED, patients randomized to the self-triage kiosk group had a shorter mean time to enter medical history data and lower mean inaccuracies in areas of medical, medication, and immunization histories compared to the group randomized to standard nurse triage.8 Another prospective trial by Coyle et. al. demonstrated that ambulatory patients who used self-check-in kiosks had shorter time to first identification, which was defined as the interval between ED arrival and identification in the hospital system.9
Long wait times in lower-acuity settings such as outpatient clinics have also been demonstrated to negatively affect quality of patient care. In one cross-sectional survey study by McMullen et al., the amount of time spent waiting in an ophthalmology clinic was the strongest driver of overall patient-reported satisfaction.10 With an expanding elderly patient population, increasing documentation and regulatory compliance requirements, and growing complexity and numbers of diagnostic tests and treatments, patient visit times may also increase in the future.11 There have been numerous studies aimed at decreasing total visit times, defined as time between check-in and check-out, using the “Lean Six Sigma” methodology.11–13 Lean Six Sigma is a combination of two manufacturing industry principles, Lean and Six Sigma, that have been applied to health care to eliminate inefficiencies, errors, and wasted resources in order to improve patient flow. One of the most significant bottlenecks identified in a tertiary clinic using the Lean Six Sigma method is the time from patient arrival to interaction with care provider,13 which has been demonstrated to improve following implementation of check-in kiosks.8,9
However, there is scarce literature on the use of check-in kiosks within real-life clinical settings, and little is known about the characteristics of patients who voluntarily choose to use a kiosk compared to those who do not. Within ophthalmology clinics specifically, impaired vision may be an additional barrier to kiosk usage. This study aims to address current knowledge gaps and to characterize routine use of check-in kiosks within a large, multidisciplinary ophthalmology outpatient center. Knowledge about patient factors that influence kiosk use will help develop future systematic changes that maximize usage, increase efficiency, and cut down on patient wait times.
Patients and Methods
This was a retrospective study performed at Cole Eye Institute, Cleveland, Ohio, after receiving approval from the Cleveland Clinic Institutional Review Board. Written informed consent was not required because of the retrospective nature of this study. All study-related procedures were performed in accordance with good clinical practice (International Conference on Harmonization of Technical Requirements of Pharmaceuticals for Human Use [ICH] E6), the Declaration of Helsinki, applicable FDA regulations, and the Health Insurance Portability and Accountability Act.
The dataset used in this study was created following a comprehensive electronic chart review of all ophthalmic appointments by patients 18 years of age or older at the Cole Eye Institute from August 1, 2019, to October 31, 2019 (Figure 1). No-shows and canceled appointments were excluded. Subsequent visits by the same patient within the 3-month study period were excluded, so that each unique visit was associated with one unique patient. Patients were then grouped into two cohorts according to whether they successfully utilized a check-in kiosk at that visit: “kiosk users” and “kiosk non-users.” Patients who attempted to use a kiosk to check-in but then were directed to the front desk to complete the check-in process were included in the “kiosk non-users” group.
Patient data acquisition.
The main outcome was the percentage of patients who used a kiosk to check-in. Secondary outcomes were to compare the demographics (specifically age, gender, race, insurance type, and median income) and best-corrected visual acuity (BCVA) of the better-seeing eye between kiosk users and non-users. Median annual household income corresponding to each patient's home zip code was obtained by inputting zip codes into American FactFinder, which provides population data from the U.S Census Bureau. BCVA was measured using Snellen charts at the time of the visit. Patients were then stratified by BCVA of the better seeing eye, with “good” vision defined as 20/40 or better, “intermediate” vision defined as 20/50 to 20/200, and “poor” vision defined as 20/200 or worse. For analysis, Snellen chart visual acuity (VA) was converted to approximated ETDRS letters.14
Categorical variables were described using frequencies and percentages, whereas continuous variables were described either as means with standard errors or as medians with interquartile range. Relationships between categorical variables were assessed using the Chi-Square test, while relationships between continuous variables were assessed using the t-test for means and Mann-Whitney test for medians. Analyses were performed using JMP Pro 14 software (SAS Institute).
Data from 31,612 appointments of patients older than 18 years of age within the months of August 2019 to October 2019 were obtained from the Cole Eye Institute database. Of those, 7,400 no-shows and canceled visits were excluded. Subsequent visits by the same patient within the study period, totaling 10,460, were also excluded. Thus, 13,745 patients and unique visits were included in this study (Figure 1).
The median age for the whole cohort was 65.9 years (interquartile range [IQR], 53.5–74.6). Of this cohort, 3,542 (26%) patients successfully used a check-in kiosk (Table 1). Kiosk users were significantly younger (median [IQR], 63.6 [49.4–72.6] vs. 66.6 years [55.0–75.4]; P < .0001) and had a greater proportion of women than kiosk non-users (2,149 [61%] vs. 5,780 [57%]; P < .0001). There were also significant differences in race; there was a higher proportion of white patients (2,640 [75%] vs. 6,799 [67%]; P < .0001) and lower proportion of Black patients (669 [19%] vs. 2,639 [26%]; P < .0001) in the kiosk user group compared to kiosk non-users. A greater proportion of kiosk users had commercial insurance (1,467 [42%] vs. 2,952 [30%]; P < .0001) and a smaller proportion had Medicare (1,761 [50%] vs. 5,836 [59%]; P < .0001) or Medicaid (282 [8%] vs. 930 [10%]; P < .0001). On average, kiosk users lived in areas with a greater median income than kiosk non-users (mean [± standard error [SE]); $58,421 [± 399] vs. $54,992 [± 236]; P < .0001).
Demographics of Kiosk Users and Non-Users
Best-Corrected Visual Acuity
There were significant differences in best-corrected VA (BCVA) between kiosk users and non-users (Figure 2). There was a greater proportion of kiosk users who had good (20/40 or better) vision (3,300 [93%] vs. 9,084 [89%]; P < .0001) and smaller proportion with intermediate (200 [6%] vs. 906 [9%]; P < .0001) and poor (20/200 or worse) vision (42 [1%] vs. 210 [2%]; P < .0001) compared to kiosk non-users (Figure 2a). On average, the kiosk user group had better BCVA (mean ETDRS [95% confidence interval (CI)], 80.5 [80.0–80.9] vs. 78.3 [78.0–78.6]; P < .0001). Following age stratification of the total cohort by quartiles (Qs) (Q1: < 54 years old, Q2: 54–66 years old, Q3: 67–75 years old, Q4: > 75 years old), this difference in mean BCVA persisted between groups in the first quartile (82.2 [81.4–83] vs. 79.8 [79.2–80.4]; P < .05), second quartile (81.3 [80.4–82.2] vs. 79.5 [79–80]; P < .05), and third quartile (81 [80.2–81.8] vs. 78.8 [78.3–79.3]; P < .05] of age (Figure 2b). There was no difference in mean BCVA in adults older than 75 years of age.
Visual acuity (VA) of the better-seeing eye in kiosk users and non-users. (A) Proportion of kiosk users and non-users with “good” (20/40 or better), “intermediate” (20/50 to 20/200), and “poor” vision (20/200 or worse). Kiosk users had a significantly greater proportion of patients with good vision and fewer proportions of patients with intermediate and poor vision (P < .0001). (B) Average best-corrected VA (BCVA) stratified into quartiles by age. Kiosk users had a significantly greater average BCVA than kiosk non-users in all age groups except adults older than 75 (*P < .05).
Reasons for Failure to Check in With Kiosk
In addition to the 3,542 patients who did successfully check-in using a kiosk, there were 308 (8% of all patients who attempted) patients who tried to use a kiosk but were redirected to the front desk to complete the check-in process (Table 2). The top reasons for this included the Medicare Secondary Payer questionnaire (MSPQ) not being complete (117 [38%]) or that it was too early (80 [26%]) or too late (28 [9%]) for the patient to check in for their appointment.
Kiosk Usage Rates
This study aims to report the rate of kiosk usage within a standard multispecialty ophthalmological practice and to characterize the patients who elected to use a check-in kiosk and those who did not. Of the total cohort of patients seen within a 3-month period, 26% of patients used a kiosk to check in. More than half of the patients included in this study were older than 65 years of age, which reflects the population seeing ophthalmological care. Age may be one reason for a low rate of kiosk usage, as there are many factors that impair a broad use of technology in older age, including psychosocial and ethical issues and fear of losing human interaction.15 Additionally, there may be cognitive and motor barriers that are experienced more commonly by older adults.16 This is further supported by our finding that on average, kiosk users were younger than kiosk non-users. A survey study of 104 participants by Lee et. al., which explored factors that motivated older consumers to adopt self-service technology (SST) within a retail setting, demonstrated that one significant predictor of use is prior experience with SSTs.17 Similarly, encouraging and assisting older patients to try using a kiosk for the first time may help them feel more comfortable and inclined to use one again in the future.
Kiosk users had a higher proportion of white patients and lower proportion of Black patients compared to kiosk non-users. Additionally, they had a higher proportion of patients with commercial insurance and lower proportion with Medicare — which could be partially attributed to a greater proportion of younger individuals in this group — and a lower proportion of patients with Medicaid. On average, they came from communities with a higher median income. These subtle yet significant differences suggest that socioeconomic background and race, which are often confounded, also influence kiosk usage. In one study by Ching et. al. that examined factors that influence technology use in 130 college students, family income was predictive of degree of personal technology use.18 Although this study was done in 2005 and degree of technology adoption has dramatically changed since then, there continues to be a digital divide based on socioeconomic status today. The Pew Research Center recently surveyed 1,502 adults living in all 50 U.S. states and found marked differences in technology use in adults depending on household income.19 More than four in 10 adults with household incomes below $30,000 a year did not own home broadband services (44%) or a traditional computer (46%), and a majority did not own tablets (64%). In contrast, each of these technologies were nearly ubiquitous among adults with households earning $100,000 or more a year. This disparity in technology use may contribute to differences in comfort level and decreased utilization of SSTs.
In addition to demographics, this study found significant differences in BCVA of the better seeing eye between kiosk users and non-users, even when stratifying by age. Although the majority of patients in both groups had good vision, kiosk users had a higher proportion of patients with good vision and lower proportion with poor vision. Kiosk users had a higher average BCVA than kiosk non-users in all age groups except adults who were 75 years of age or older, suggesting that in most patients, BCVA influences kiosk utilization independently of age. Possible changes to make kiosks easier to use by ophthalmic patients include making texts and icons larger, using vibrant colors to increase contrast, having auditory or vibratory cues, and having assistance readily available.20 Interestingly, there were some patients with 20/200 or worse BCVA in both eyes who used a kiosk to check in, which could be due to patients receiving assistance with the check-in process by companions or employees. This was a potential confounder that was not examined in this study.
The large majority of patients who attempted to use a kiosk were able to successfully check-in, indicating that kiosks were relatively easy to use. The primary reason for redirection to the front desk was due to an incomplete MSPQ. Incomplete or inaccurate completion of MSPQ results in significant problems for the revenue cycle. By promoting increased use of kiosks for check-in purposes, front desk personnel have more time to spend with patients on more involved tasks such as accurately filling out the MSPQ.
Strengths of this study include the use of a large sample size and real-life patient cohort seen at a multispecialty ophthalmological practice. The drawbacks of this study include lack of information on patient opinions toward kiosk use, whether patients had assistance with the check-in kiosk from a companion or employee, and the duration of kiosk use. Additionally, only patients who were redirected by the kiosk to the front desk were included as unsuccessful kiosk check-ins — those who used a kiosk and stopped on their own were not recorded, which could underestimate the true unsuccessful kiosk check-in rate. BCVA of the best seeing eye was used to reflect functional visual status, which may underestimate the degree of vision impairment, particularly when the worse eye is moderately to severely visually impaired.21 Although zip codes are not the most specific way to obtain median incomes, they still provided valuable information regarding patients' socioeconomic context.22 Lastly, the use of the first visit within a 3-month period does not capture changes in behavior in prior or subsequent visits.
In conclusion, this study found significant differences in demographic characteristics, including age, gender, and median income, and BCVA between kiosk users and non-users. The findings here can direct future changes to the check-in process, such as encouraging patients from populations less inclined to embrace SSTs to try using a check-in kiosk and to make kiosk displays easier to navigate by patients with poorer BCVA. Further research is needed to assess whether these changes would truly increase kiosk utilization and decrease patient waiting times, and also to gain perspective on patient attitudes towards kiosk utilization within an outpatient medical clinic.
- Johnson MW, Christensen CM, Kagermann H. Reinventing Your Business Model. https://www.hbs.edu/faculty/Pages/item.aspx?num=34830. December 2008. Accessed February 13, 2020.
- Brynjolfsson E, McAfee A. The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies. W. W. Norton & Company; 2014.
- Dabholkar PA, Michelle Bobbitt L, Lee E. Understanding consumer motivation and behavior related to self-scanning in retailing: implications for strategy and research on technology-based self-service. Int J Serv Ind Manage. 2003;14(1):59–95. doi:10.1108/09564230310465994 [CrossRef]
- Taufik N, Hanafiah MH. Airport passengers' adoption behaviour towards self-check-in Kiosk Services: the roles of perceived ease of use, perceived usefulness and need for human interaction. Heliyon. 2019;5(12):e02960. doi:10.1016/j.heliyon.2019.e02960 [CrossRef] PMID:31890945
- Shin H, Perdue RR. Self-Service Technology Research: A bibliometric co-citation visualization analysis. Int J Hospit Manag. 2019;80:101–112. doi:10.1016/j.ijhm.2019.01.012 [CrossRef]
- Bernstein SL, Aronsky D, Duseja R, et al. Society for Academic Emergency Medicine, Emergency Department Crowding Task Force. The effect of emergency department crowding on clinically oriented outcomes. Acad Emerg Med. 2009;16(1):1–10. doi:10.1111/j.1553-2712.2008.00295.x [CrossRef] PMID:19007346
- Bernstein SL, Asplin BR. Emergency department crowding: old problem, new solutions. Emerg Med Clin North Am. 2006;24(4):821–837. doi:10.1016/j.emc.2006.06.013 [CrossRef] PMID:16982341
- Sinha M, Khor K-N, Amresh A, Drachman D, Frechette A. The use of a kiosk-model bilingual self-triage system in the pediatric emergency department. Pediatr Emerg Care. 2014;30(1):63–68. doi:10.1097/PEC.0000000000000037 [CrossRef] PMID:24378865
- Coyle N, Kennedy A, Schull MJ, et al. The use of a self-check-in kiosk for early patient identification and queuing in the emergency department. CJEM. 2019;21(6):789–792. doi:10.1017/cem.2019.349 [CrossRef] PMID:31057137
- McMullen M, Netland PA. Wait time as a driver of overall patient satisfaction in an ophthalmology clinic. Clin Ophthalmol. 2013;7:1655–1660. doi:10.2147/OPTH.S49382 [CrossRef] PMID:23986630
- Ciulla TA, Tatikonda MV, ElMaraghi YA, et al. LEAN SIX SIGMA TECHNIQUES TO IMPROVE OPHTHALMOLOGY CLINIC EFFICIENCY. Retina. 2018;38(9):1688–1698. doi:10.1097/IAE.0000000000001761 [CrossRef] PMID:28723845
- Fischman D. Applying Lean Six Sigma methodologies to improve efficiency, timeliness of care, and quality of care in an internal medicine residency clinic. Qual Manag Health Care. 2010;19(3):201–210. doi:10.1097/QMH.0b013e3181eece6e [CrossRef] PMID:20588139
- Lin SY, Gavney D, Ishman SL, Cady-Reh J. Use of lean sigma principles in a tertiary care otolaryngology clinic to improve efficiency. Laryngoscope. 2013;123(11):2643–2648. doi:10.1002/lary.24110 [CrossRef] PMID:23529903
- Gregori NZ, Feuer W, Rosenfeld PJ. Novel method for analyzing snellen visual acuity measurements. Retina. 2010;30(7):1046–1050. doi:10.1097/IAE.0b013e3181d87e04 [CrossRef] PMID:20559157
- Pilotto A, Boi R, Petermans J. Technology in geriatrics. Age Ageing. 2018;47(6):771–774. doi:10.1093/ageing/afy026 [CrossRef] PMID:29546366
- Xie B. Older adults, computers, and the internet: future directions. Gerontechnology (Valkenswaard). 2003;2(4):289–305. doi:10.4017/gt.2003.02.04.002.00 [CrossRef]
- Lee H-J, Lyu J. Exploring factors which motivate older consumers' self-service technologies (SSTs) adoption. Int Rev Retail Distrib Consum Res. 2019;29(2):218–239. doi:10.1080/09593969.2019.1575261 [CrossRef]
- Ching CC, Basham JD, Jang E. The Legacy of the Digital Divide: Gender, Socioeconomic Status, and Early Exposure as Predictors of Full-Spectrum Technology Use Among Young Adults. Urban Education. 2016;40(4):394–411. doi:10.1177/0042085905276389 [CrossRef]
- Anderson M, Kumar M. Digital divide persists even as lower-income Americans make gains in tech adoption. Pew Res Cent. https://www.pewresearch.org/fact-tank/2019/05/07/digital-divide-persists-even-as-lower-income-americans-make-gains-in-tech-adoption/. May 7, 2019. Accessed February 21, 2020.
- Legge GE. Reading Digital with Low Vision. Visible Lang. 2016;50(2):102–125. PMID:29242668
- Finger RP, Fenwick E, Hirneiss CW, et al. Visual impairment as a function of visual acuity in both eyes and its impact on patient reported preferences. PLoS One. 2013;8(12):e81042. doi:10.1371/journal.pone.0081042 [CrossRef] PMID:24339893
- Berkowitz SA, Traore CY, Singer DE, Atlas SJ. Evaluating area-based socioeconomic status indicators for monitoring disparities within health care systems: results from a primary care network. Health Serv Res. 2015;50(2):398–417. doi:10.1111/1475-6773.12229 [CrossRef] PMID:25219917
Demographics of Kiosk Users and Non-Users
|Total||Kiosk Users||Kiosk Non-Users||P Value|
|n (% of Total n)||13,745||3,542 (26%)||10,203 (74%)|
|Age, Median (IQR)||65.9 (53.5–74.6)||63.6 (49.4–72.6)||66.6 (55–75.4)||< .0001a|
|Female, n (%)||7,929 (58%)||2,149 (61%)||5,780 (57%)||< .0001b|
|Race, n (%)||< .0001b|
| White||9439 (69%)||2,640 (75%)||6,799 (67%)||< .0001b|
| Black||3,308 (24%)||669 (19%)||2,639 (26%)||< .0001b|
| Other||998 (7%)||233 (6%)||765 (7%)||.0085b|
|Insurance, n (%)||< .0001b|
| Commercial||4,419 (32%)||1,467 (41%)||2,952 (29%)||< .0001b|
| Medicare||7,597 (55%)||1,761 (50%)||5,836 (57%)||< .0001b|
| Medicaid||1,212 (9%)||282 (8%)||930 (9%)||.0097b|
| Other||517 (4%)||32 (1%)||485 (5%)||< .0001b|
|Annual Median Income, Mean (±SE)||55,881 (±204)||58,421 (±399)||54,992 (±236)||< .0001c|
Kiosk Usage Rates
|Total Attempted, N (% Total)||3,850|
| Success||3,542 (92%)|
| Failure||308 (8%)|
|Reasons for Kiosk Check-In Failure, N (% Failure)|
| MSPQ not complete||117 (38%)|
| Too early to check in||80 (26%)|
| Too late to check in||28 (9%)|
| Primary insurance not verified||28 (9%)|
| Other||55 (18%)|