Missed medical appointments have repercussions for both individual-level and community-level health. The frequency of missed clinic visits, also known as clinic absenteeism, falls within the wide range of 12–80%, and varies with medical specialty, location, time of year, clinic setting, patient volumes, and clinic wait times.^1–8 Missed appointments not only impact individual patient health, but also affect the health of others who may have able to access care sooner during the unused time slot. In addition, the economic burden of absenteeism on the healthcare system is profound; it is estimated that $50 billion is lost in the United States (US) on missed clinic appointments annually.⁹ Unfilled appointment times are a substantial problem for maintaining the economic viability of outpatient clinics in communities with scarcity of healthcare resources.⁹

Rates of vision impairment and blindness are increasing, and yet rates of missed clinic visits to prevent and treat eye disease remain high.¹⁰ In 2020, 160.7 million people in the working age population, or 3.3% of the total workforce, were either blind or had severe visual impairment, leading to a $410.7 billion loss in societal productivity and purchasing power.¹¹ Beyond a person’s work-life, vision impairment impacts a person’s independence and ability to drive, read and perform activities of daily living.^11–14

Published rates for missed eye clinic visits fall between 15–30%,^15–19 and are even higher (up to 46%) among people with glaucoma, for whom there is a need to chronically monitor and adjust treatments to preserve vision.^20–22 Previous studies have found that demographic factors, including younger age, lack of insurance, and identifying as Black or Hispanic are risk factors for missing clinic visits.^16,22–26 In a five-year retrospective analysis of an eye clinic population in Detroit, MI, Khambati et al. found that older age, male sex, non-January appointment month, longer physician-patient history, Caucasian race, retina specialty, and clinic location were all significantly associated with appointment attendance.²² Additional qualitative and quantitative studies have noted poor health, poor communication between offices and patients, lack of transportation, large lead times before appointments, low health literacy, forgetfulness, fear of medical professionals, lack of support, and childcare-related issues as barriers to attending clinic visits. These barriers are not social risk factors that are routinely recorded on clinic intake forms.^{15,16,19,24,27–29}

The purpose of this study was to determine if neighborhood-level social risk factors, independent of individual demographic factors, were associated with missed clinic visits in a free glaucoma and eye disease screening program. Understanding whether and which neighborhood-level social risk factors are related to missed clinic visits could help inform policy on how to improve eye care utilization specifically and health care utilization more broadly.

A total of 1,431 participants scheduled their screening visits in the MI-SIGHT program, including 57% from the FQHC and 43% from the free clinic between July 28, 2020 and November 11, 2021. Participants were on average 54.1 (SD=14.5) years old at the time of scheduling their visit and 60.4% self-identified as female. Participants self-identified race and ethnicity were as follows: 51.4% identified as Black or African American, 36.6% identified as White, 8.9% identified as Hispanic or Latino, 5.1% identified as Asian, and 6.8% identified as Other race. Participants lived on average 7.7 (SD=8.1, range=0.1 to 83.5) miles from the clinic they were scheduled to attend. Both clinics were accessible through public transport. Among participants who attended the screening visits, 4.5% (49/1101) took public transport. The median home to bus stop distance for these participants was 0.1 miles and the median home to clinic distance was 3.2 miles (Table 2).

Participant characteristics were significantly different between the two clinics (Table 1). Participants scheduled at the FQHC were younger than those at the free clinic (53.3 vs. 55.2 years, p=0.012), a larger percentage were male (41.9% vs 36.6%, p=0.047), a larger percentage identified as Black or African American (64.2% vs 34.3%, p<0.000), and a smaller percentage identified as Hispanic or Latino (6.8% vs 11.9%, p=0.004). Participants at the FQHC lived closer to the bus stop (1.7 miles vs 6.1, p<0.0001) and traveled a shorter distance to the clinic than participants at the free clinic (5.5 miles vs 10.7, p<0.0001, Table 2).

Of the 1,431 participants, 1,423 (99.4%) had address information that could be mapped to a FIPS code for access to neighborhood-level measures (Table 2). Of these 1,423 participants, 0.2% to 12.7% of participants were missing some neighborhood-level measures because of suppressed data in ACS due to low population size. Participants lived in neighborhoods where the median HHI was $40,500 USD (IQR=$29,300-$63,200) and median renter cost burden was 31.6% of income (IQR=24.6%−43.1%). An average of 46.8% of renters (SD=23.8%) were cost burdened such that they paid ≥ 30% of their gross household income to rent their homes. In these neighborhoods, an average of 13.7% of households reported that they did not own a car (SD=13.4%). The average number of cars per household was 1.5 (SD=0.5). The state ADI rank was on average 6.8 (SD=3.2) and the national ADI was on average 76.1 (SD=26.5). Participants lived in neighborhoods where, on average, 5% (SD=2.7%) of households were considered energy burdened such that they spent approximately 5% of their gross household income on energy costs.³⁸ An average of 7.9% of households (SD=10.8%) lived in subsidized housing.

Of the first 1,431 participants recruited to participate in the ongoing MI-SIGHT program that provides free glaucoma and eye disease screening and helps participants access treatment and low-cost glasses, 23.1% of scheduled participants missed their appointments. The program purposefully was held in clinics embedded in neighborhoods with high levels of poverty because of the increased risks of eye disease and difficulties accessing high-quality eye care in this population. The median household income in our sample was $40,500 compared to the 2017 ACS national median income of $69,000.³³ In this study, we found that neighborhood-level measures of deprivation including ADI, median HHI, percentage of households with no vehicles, and energy cost burden were significantly associated with an increased probability of missing the scheduled screening appointment, even after controlling for individual-level characteristics including age, sex, race, and ethnicity.

Many previous studies have focused on individual-level characteristics that are associated with a higher probability of missing a clinic visit,^16,22 but here we specifically focused on the added effect of neighborhood-level characteristics on the probability of missing a clinic appointment.^41,45 In 2019, Chiam et al. conducted a study reviewing all new patient appointments scheduled in a suburban Pennsylvania clinic, and found that individuals living in zip codes with lower median household incomes had a higher likelihood of missed appointments (OR for <$35,667 USD vs. >$59,445 USD: 1.59, 95% CI 1.08, 2.34, p<0.001).¹⁶ Similarly, a retrospective chart review of 4,598 adult patients ≥18 years with a diagnosis of glaucoma, diabetic retinopathy (DR), or age-related macular degeneration (AMD) seen at Yale University identified a significant association of residing in a zip code with reduced median household income with a decreased probability of clinic attendance (OR 1.68, p < 0.0001).²⁶ In our current study, we found that other neighborhood-level metrics beyond HHI, including higher energy burden and ADI, were independently associated with an increased odds of missed clinic appointments. Like others, we found that, on average, lesser commute distance was related to higher rates of missed clinic visits.¹⁶ Possible explanations for this finding may include that those who lived farther away had higher disease acuity, fewer eye care providers, had greater access to transportation, or had higher incomes compared to those who lived closer to the clinics, which were situated in urban downtown areas. Of note, the strongest neighborhood-level predictor was percentage of households with no vehicles; participants in our sample resided in neighborhoods where an average of 13.7% (SD=13.4) of households did not own a car, compared to a national average of 9.0%.⁴⁶ Even still, only 4.5% of our participants used public transportation to attend their free eye screening visit and distance from home to bus stop or bus stop to clinic did not predict clinic attendance. Bus prices, convenient routes, frequency of buses, and safety of buses and stop locations all need to be considered beyond just availability of bus stops in assessing reasons for low public transport utilization.

The rate of missed clinic visits in our study, 23.0%, is higher than those reported by other studies investigating clinic absenteeism in ophthalmology.^15–19,22 Several studies have shown that people requiring glaucoma care have a higher likelihood of missing a clinic visit compared to people requiring care for other ophthalmologic conditions.^16,22 This may be due to the largely asymptomatic nature of early glaucoma where the focus is on prevention of future vision loss. Moreover, participants in the MI-SIGHT program predominantly came from socioeconomically deprived neighborhoods, which as we have shown here, affects the overall missed appointment rate. For comparison, Khambati et al. found that missed clinic visit rates were much lower in a satellite clinic that served a suburb with high socioeconomic status (Novi, Michigan, ADI = 1) compared to the missed clinic visit rate at the academic medical center’s main location in an urban area with high rates of poverty (Detroit, MI, ADI =4).²² In our cohort of participants presenting for screening appointments in Ypsilanti and Flint, MI, the overall mean ADI was 6.8 (SD=3.2), meaning that the neighborhoods in which MI-SIGHT participants lived were more deprived than 68% of neighborhoods in Michigan. Therefore, it is likely that the participants in this study come from more socioeconomically disadvantaged neighborhoods than those studied previously.

Despite the same appointment reminder strategies utilizing phone calls and text messages, there were significantly more missed clinic visits at the FQHC than at the free clinic (29.4% vs 14.6%, p<0.001), underscoring our findings that neighborhood factors matter. Participants at the FQHC came from neighborhoods with lower deprivation indices. For example, the average state ADI rank for FQHC participants was 8.6, meaning that the neighborhoods in which FQHC participants lived were more deprived than 86% of other neighborhoods in Michigan. In comparison, the mean ADI among free clinic participants was much better, at 4.5 (p<0.001). Using national ADI rank, FQHC participants lived in areas with an average ADI of 90.3 on a 100-point percentile scale, meaning that the neighborhoods in which FQHC participants lived were worse than 90% of other neighborhoods nationwide.

Our results contribute to the growing body of evidence that neighborhood characteristics affect health and health behaviors. Living in a neighborhood with a worse ADI has been associated with increased prevalence of adverse health and health care outcomes, including increased risk of hospital readmissions, poorer cognitive function in older adults, and worse survival with nonmetastatic cancers even after adjusting for patient income.^45,47,48 Social cognitive theory posits that neighborhood characteristics may influence health-related behaviors through exemplifying positive social norms such as seeing people walking and running outside.⁴⁹ Likewise, walkable, social, and safe neighborhoods provide opportunities for physical activity and increased health-related quality of life, as well as decreased levels of stress.⁴⁹ Another way to understand how physical neighborhoods impact missed clinic visits is through Maslow’s hierarchy of needs. Maslow’s hierarchy of needs is a motivational theory which centers around the fact that individuals are motivated to first achieve basic physiological needs that necessitate survival (such as food, shelter, security, and stability) before pursuing psychological needs (such as feelings of belonging or love, self-esteem, and self-actualization).⁵⁰ Preventive healthcare activities, like participating in screening for silent eye diseases, is a less pressing need than adequate food and shelter. People who live in neighborhoods with high deprivation measures such as high ADI, high rent burden, or high energy burden may have little reserve for pursuing higher needs of wellness such as preventive eye disease screening. Through this lens, it is not hard to imagine why residents of neighborhoods with fewer resources obtain less ophthalmic care and are at the highest risk for missing clinic appointments.^24,30,51,52

As part of the MI-SIGHT protocol, multiple phone calls were made to all participants who missed their clinic visits in order to reschedule appointments. While theoretically some individuals may have chosen to seek care elsewhere, we found this to rarely be the case. More often, based on documentation in our cancel logs, individuals missed appointments because they were ill, forgot their appointment due to other concerns, had transportation issues, or had difficulty taking time away from work. Understanding risk factors associated with missing clinic visits is crucial to developing strategies to address this issue. Strategies that have been previously utilized include reminder phone calls, automated emails/SMS/or phone calls or mailed reminder cards, and penalty fees.^1,53,54 These strategies, however, are likely to have only limited success as they place all the responsibility on the individual. Though forgetting to attend clinic appointments is a common occurrence, in neighborhoods with high levels of deprivation, the barriers to attending the clinic visit go well beyond forgetting, thereby making reminder messages a strategy unlikely to completely solve the problem. Additionally, penalty fees for missed visits, which we did not utilize, can deter people, especially those with more limited financial means, from seeking healthcare. Furthermore, one-way automated reminders do not facilitate re-scheduling appointments if participants are unable to make it to appointments due to last minute hurdles. In the MI-SIGHT program, we used a series of reminder methods, starting with a phone call two days prior to the appointment to give patients an opportunity to re-schedule the visit, and another phone call the day before the visit if the participant did not confirm the visit the day before. Only after the patient confirmed that they can attend the appointment do we send a text message reminder the day before the appointment. Using two-way communication – a phone call - allows participants the flexibility to reschedule the appointment if needed. The fact that we still had a 23% rate of missed appointments demonstrates that reminders are helpful, but address only one reason why people miss clinic visits for preventive healthcare.

We need to move beyond strategies to bolster clinic attendance that focus only on the individual and ignore upstream deterrents not under an individual’s control. Addressing the root of the issue through policy initiatives, such as tackling income inequality and its downstream effects and tackling unequal access to transportation, may help address clinic absenteeism. Specifically, providing more affordable, safe housing units, expanding clean outdoor green spaces within neighborhoods, ensuring access to healthy food, and improving public transportation could all aid people in coming to clinic visits for preventive eye care.^55–58 As part of a multi-method interventional strategy, DuMontier and colleagues identified and interviewed a cohort of patients who had missed six or more appointments in an 18-month period from May 2007 to June 2008 to better understand why they missed the appointments. Patients cited reasons ranging from managing family obligations to financial struggles to emotional and physical health concerns.⁵⁴ The authors then distributed a scripted discourse to be communicated by the receptionist to this cohort that made patients aware of their frequent missed appointments, described the effects on the clinic and the patient’s health, and negotiated a commitment from the patient to improve appointment adherence. They also implemented a double-booking strategy, whereby members of this cohort were first scheduled into a virtual visit on the same day as their clinical appointment to confirm their attendance. After these targeted strategies, the clinic’s rate of missed visits in the same cohort of patients fell from 33.3% to 17.7%. The authors posited that simply asking patients about their barriers to care in a respectful and empathetic manner improved health care providers’ understanding of their patients’ individual and collective issues, and subsequently decreased patient anxieties towards attending their appointments. This individual level intervention was highly successful, but still resulted in a 17.7% rate of missed clinic visits, which we hypothesize may be due to systemic barriers that were difficult for a person to overcome. To help address more upstream barriers, a community hospital-based family practice employed two social workers to connect patients to transportation and childcare services prior to their scheduled appointment and helped decrease their missed appointment rate from 25% to 7%.⁵⁹ More general area-based strategies have been limited by inconclusive results, difficulties with sustained funding, and general scalability.^55,60,61

Limitations of this study include its lack of generalizability as the analysis was limited to people who signed up for MI-SIGHT program appointments at two sites, an FQHC and a free clinic in southeastern Michigan. Further, we could not assess how individual-level economic indicators impacted participants because these variables were not collected for individuals who did not come to their MI-SIGHT program visit. An area of future study could involve following up with individuals who fail to attend a free clinic appointment to elicit individual level risk factors and specific barriers to seeking care. We do not have specific reasons for missed clinic visits for our cohort of patients. In addition, neighborhood-level data were extracted from national surveys based on the participants’ addresses. ACS data relies on self-reported outcome measures, which can lead to social desirability bias, and restricts data from geographic locations with smaller populations, potentially lowering generalizability. Lastly, the enrollment period for this study began after the lockdown period for the COVID pandemic ended, but was still during the height of the pandemic. While the effects of the pandemic on preventive care-seeking behavior would theoretically affect both those who came and did not come to their appointment equally as all had made the appointment, the impact of the pandemic on appointment attendance is not fully clear.

Missing clinic visits for screening for glaucoma and other eye diseases can lead to missed early detection of disease and downstream vision loss. In the MI-SIGHT program, a free eye disease detection and treatment program, missing clinic visits was associated with living in a neighborhood with higher levels of poverty, worse neighborhood level deprivation, and more people living in circumstances where it is difficult to meet basic needs including paying rent and having access to a car. In designing interventions to address missed clinic visits, most approaches include reminder systems with the underlying thought process being that if individuals prioritized their vision, they would come to the appointment. However, individual-level strategies are not enough as they fundamentally lay blame on the individual, rather than acknowledging the larger social barriers to healthcare utilization. Our findings indicate that broad, community-based interventions are needed, especially in communities with high rates of poverty, to address upstream social factors that impact people’s ability to attend clinic visits, including issues such as income inequality, access to transportation, childcare, eldercare, safe housing, employment and job flexibility.