Article Text

Download PDFPDF

Temporal trends and disparities in gastroenterology care use before, during, and after COVID-19 lockdown
Free
  1. Harsha Sanaka1,
  2. Rajat Garg1,
  3. Vidhi Patel1,
  4. John McMichael1,2,
  5. Carole Macaron1
  1. 1Department of Gastroenterology and Hepatology, Cleveland Clinic, Cleveland, Ohio, USA
  2. 2Department of General Surgery, Cleveland Clinic, Cleveland, Ohio, USA
  1. Correspondence to Dr Carole Macaron, Department of Gastroenterology and Hepatology, Cleveland Clinic, Cleveland, USA; macaroc{at}ccf.org

Abstract

Socioeconomic disparities adversely affected healthcare use during COVID-19 lockdown. However, trends in these disparities post lockdown are unknown. Therefore, our aim was to study temporal trends and factors associated with gastroenterology healthcare access and disparities during and after COVID-19 lockdown. This cohort study consisted of patients receiving outpatient care in the Cleveland Clinic gastroenterology department between March 2020 and June 2020 and corresponding time periods in 2019 and 2021. Patient demographics and socioeconomic factors were extracted and analyzed. There were 47,031 patients (mean age 56.3±17.6 years, 61.9% female and 76.4% white) included. Patients ≥65 years sought healthcare less frequently during and after the lockdown (40.1% vs 34.8% vs 35.2% in 2019, 2020, and 2021 respectively). Missed visits (4.2% vs 10% vs 10.4%), tobacco (11.4% vs 15.9% vs 16.1%), alcohol (38.6% vs 45.5% vs 50.9%), and illicit drug use (3.5% vs 5.8% vs 10.7%) have steadily increased during and after the lockdown compared with prepandemic levels. Factors associated with reduced telehealth use were black race (OR 0.89, 95% CI 0.81 to 0.99), Hispanic race (OR 0.63, 95% CI 0.51 to 0.77)), Medicaid/other public insurance (OR 0.87, 95% CI 0.79 to 0.95)), unemployed status (OR 0.85, 95% CI 0.79 to 0.92)), and non-English/Spanish speakers (OR 0.66, 95% CI 0.46 to 0.94)). In conclusion, socioeconomic and ethnic disparities persist in healthcare use even a year after the onset of the COVID-19 pandemic. There is an alarming increase in missed visits and substance abuse. Therefore, efforts should be targeted on improving healthcare access for these aforementioned vulnerable groups.

  • COVID-19
  • public policy
  • socioeconomic factors

Data availability statement

Data are available upon reasonable request. Deidentified data can be shared with the permission of institution upon request.

This article is made freely available for personal use in accordance with BMJ’s website terms and conditions for the duration of the covid-19 pandemic or until otherwise determined by BMJ. You may use, download and print the article for any lawful, non-commercial purpose (including text and data mining) provided that all copyright notices and trade marks are retained.

https://bmj.com/coronavirus/usage

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

WHAT IS ALREADY KNOWN ON THIS TOPIC

  • COVID-19 pandemic has adversely influenced existing socioeconomic and ethnic disparities in healthcare use. However, the long-term effects are not known.

WHAT THIS STUDY ADDS

  • Older patients sought gastroenterology healthcare less frequently during and after the pandemic lockdown. Substance abuse and missed visit rates continue to increase. Blacks, Hispanics, unemployed patients and patients with public insurance or Medicaid were less likely to use video visits and more likely to miss scheduled appointments.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

  • This study identified vulnerable populations with decreased gastroenterology care use due to the COVID-19 pandemic. Future efforts are needed to target and improve access for these vulnerable groups.

Introduction

Demographic and socioeconomic inequities play a critical role in the access and use of healthcare services. Although these disparities have been long-standing, they first gained official recognition with the landmark Heckler Report in 1985, which reported that 60,000 deaths occurred each year in the USA due to health disparities and also provided recommendations to reduce such health disparities.1 Since then, there have been several advances in medical technology and community-level interventions that have improved health and reduced disparities. However, the availability of newer and expanded healthcare services has not translated to equitable distribution to all Americans, and disparities have persisted despite garnering increased attention. For instance, a 2019 report found that black people have a life expectancy 4 years less than white people.2

In the initial stages of the COVID-19 pandemic, hospitals opted to replace traditional in-person clinic visits with telehealth visits and continued to offer the latter as an increasingly viable alternative to the former even after the lockdown. This shift to telehealth had many potential benefits: not only would telehealth mitigate the spread of COVID-19 by removing interpersonal contact, but also telehealth is positively associated with improved outcomes, ease of use, cost savings, and improved communication among other benefits.3 Unfortunately, the long-lasting barriers and disparities in the American healthcare system have resulted in inequities in using the telehealth at the start of the pandemic.4–7 However, while these studies have looked at healthcare access disparities during the first surge of COVID-19, they have not examined whether these trends in healthcare access have persisted in 2021, after the country had adequate time to become accustomed to the pandemic landscape. Consequently, the true gravity of these healthcare access disparities has not been accurately measured. Therefore, we aimed to identify the trends in demographic and socioeconomic factors associated with healthcare access before, during, and a year after the onset of the COVID-19 pandemic and factors predictive of telehealth use and missed visits in our gastroenterology department.

Materials and methods

The study cohort consisted of patients scheduled for clinic visits in the Cleveland Clinic Department of Gastroenterology in three distinct time periods: March 9–June 18, 2020, and corresponding time periods in 2019 and 2021. In the early stages of COVID-19 pandemic, a state of emergency was declared on March 9, 2020, and Ohio went into lockdown. A gradual reopening of the state was started on May 1, 2020, with most of the lockdown restrictions lifted by June 18, 2020. The Cleveland Clinic shifted to telehealth, and its outpatient visits became predominantly remote during the lockdown period.8 Interstate travel was discouraged, limiting in-person visits from out-of-state patients; hence, only Ohio residents were included in this study. Patient groups from corresponding time periods in 2019 and 2021 were included to study the temporal trends in socioeconomic and ethnic disparities. The following data were extracted: age, sex, race (black, white, Hispanic, and other), area of residence, employment status, marital status, primary language, type of insurance (Medicare, Medicaid or other public insurance, commercial, or uninsured), and substance (tobacco, alcohol, or illicit drugs) abuse. All visits, whether completed or missed (no-show/cancellation), and all types of visits—in-person, video, or telephone—were considered for study purposes.

Video visits were defined as those provided using an electronic-based communication network with audiovisual input. Telephone visits consisted of interactions that used audio input only. Patient visits were considered missed visits if the patient did not complete their scheduled appointment due to cancellation or no-show. If a patient had multiple visits during the study period, only their first visit was considered in order to prevent duplicates and minimize bias. During the lockdown, in-person visits were offered only at the provider’s discretion. All patients were offered video visits first, and if a patient was unable to schedule a video visit or declined, then a telephone visit was offered. Video visits were conducted through different audiovisual platforms including Epic (Epic Systems Corporation, Verona, Wisconsin, USA) video platform, Amwell (Boston, Massachusetts, USA), or FaceTime (Apple, Cupertino, California, USA). During the 2021 study period, video, telephone and in-person visits were offered at patient and physician discretion. Primary language was determined based on the need for a translator. A proxy median household income for the patients’ residential addresses was obtained from US Census 2010 estimates, and patients were grouped into quartiles.9

Statistical analysis

Data were presented as mean±SD for continuous variables and n (%) for categorical factors. Independent sample t-test or the non-parametric Kruskal-Wallis test was used to compare continuous variables, and Pearson’s χ2 test was used to compare categorical variables. Univariate analyses was performed to assess the characteristics of patients seen in 2019, 2020, and 2021 and to compare demographics and socioeconomic factors associated with (1) video visit versus other visit types, (2) telephone visits versus other visit types, (3) in-person visits versus other visit types, and (4) completed visits versus missed visits. To identify predictors of each type of visit, multivariate logistic regression modeling was performed to calculate ORs. Input variables were included if they met the cut-off alpha level of ≤0.05 on univariate testing. All statistical analyses were performed using the SPSS software V.24. A p value of <0.05 was considered statistically significant.

Results

During the 2020 COVID-19 lockdown period, 20,059 visits were scheduled, of which 6720 patients had multiple visits. During the corresponding 2019 and 2021 time periods, 20,727 and 26,010 visits were scheduled, respectively. For study purposes, only the patient’s initial visit was considered, leading to a total of 17,335 visits in 2019, 13,339 visits in 2020, and 16,357 visits in 2021. Overall, there were a total of 66,796 scheduled visits during the three study periods, of which 19,764 patients had multiple visits, yielding 47,032 visits. The study cohort (women 29,116 (61.9%), mean age 56.3±17.6 years) predominantly had private insurance (61.7%) with a racial distribution as follows: white (76.4%), black (14.5%), Hispanic (3.8%), and other (5.3%). The cohort had an almost equal distribution of employed (39%) and unemployed (38.8%) individuals as well as married (49.6%) and single (49%) individuals. The vast majority spoke English as their primary language (97.9%) with 2.1% requiring translator services for the visits.

Temporal trends

During the pandemic lockdown, there was a decline in total visits scheduled with a gradual improvement thereafter. The majority of visits in 2020 were telephone visits (50%) and video visits (22.7%) with a slow return to in-person visits (76.6%) in 2021 (figure 1). The COVID-19 pandemic led to different trends in various groups seen in our department. Older patients (age ≥65 years, those with Medicare insurance, and retired) sought healthcare less frequently during and after the lockdown (table 1). Women used healthcare more often than men before, during, and after the lockdown (61.4% vs 60.8% vs 63.4%, p<0.001). The proportion of non-white patients (22.2% vs 24.4% vs 24.4%, p<0.001), patients with Medicaid and other public insurance (14.3% vs 18% vs 16.5%, p<0.001), and single individuals continued to increase during and after the lockdown compared with prepandemic levels. On the other hand, patients with private insurance sought care less frequently during the lockdown but more frequently a year afterwards (61.6% vs 59.8% vs 63.4%, p<0.001). Patients in the lowest quartile of income sought care more frequently during the lockdown in contrast to those in the higher quartiles; however, for the highest quartile, visits decreased during the lockdown but have resumed to higher than prepandemic levels (table 1). The number of missed visits and tobacco, alcohol, and illicit drug use have steadily increased during and continue to increase after the pandemic lockdown (table 1).

Figure 1

Visit distribution during the three study periods.

Table 1

Temporal trends in patient characteristics and patient visits

Predictors of video visits

During the three study periods, 5923 video visits were scheduled and 5352 visits were completed. In 808 visits, the specific type of video platform used was not mentioned. Of the remaining 5115 visits, there were 3848 visits via Zoom, 185 via FaceTime, and 1082 via Amwell scheduled. There were no statistically significant differences between the completed and missed visit rates between the Zoom, FaceTime, and Amwell platforms (completed visits: 3511, 163, and 980, respectively, and missed visits: 337, 22, 102, respectively; p value=0.30).

Compared with patients who had other visits (n=37,911), video visit patients (n=5352) were more likely to be young (51.8 vs 57.3), white (79.7% vs 77%), English-speaking (98.6% vs 97.8%), unemployed (47.1% vs 39.4%), have higher income, and have private insurance (68.5% vs 61.6%) (p<0.05 for all) (table 2). There were no significant differences in gender, marital status, or smoking between the two groups.

Table 2

Patient characteristics associated with video visits

On multivariate analysis, for every 1 year increase in age, there was a 2% lower chance of video visits (OR 0.98, 95% CI 0.98 to 0.98, p<0.001). Compared with white subjects, black subjects were 10% (OR 0.9, 95% CI 0.82 to 0.99, p=0.048) and Hispanics 36% less likely to use video visits (OR 0.64, 95% CI 0.53 to 0.78, p<0.001). Patients with Medicaid or other public insurance were 14% less likely to use video visits compared with those with private insurance (OR 0.86, 95% CI 0.78 to 0.94, p=0.002). Patients requiring translators for ‘other languages’ were 30% less likely to use video visits compared with English speakers (OR 0.7, 95% CI 0.5 to 0.99, p=0.049). Alcohol use and illicit drug use were associated with increased usage of video visits (table 2).

Predictors of telephone visits

A total of 6921 patients completed a telephone visit and 36,324 patients had other visits. On comparing patients who had telephone visits to others, the former was more likely to be young (56.1 vs 56.6 years), black (15.2% vs 13.6%), unemployed (43.1% vs 38%), single (51% vs 48.6%), smoke (15.9% vs (13.4%), have Medicare (21.4% vs 21.2%), Medicaid or public insurance (18.2% vs 14.6%), have lower median household income (lowest quartile 29.2% vs 22.85%), and need a Spanish translator (1.3% vs 0.9%) (p<0.05 for all values) (table 3). On multivariate analysis, Hispanics were 21% (OR 0.79, 95% CI 0.68 to 0.92, p=0.002) and others were 14% (OR 0.86, 95% CI 0.76 to 0.98, p=0.018) less likely than white subjects to use telephone visits. Those with Medicaid or other public insurance were 13% more likely to use telephone visits than those with private insurance (OR1.13, 95% CI 1.04 to 1.22, p=0.003). Patients in the second, third, and highest quartiles of household income were 21%, 28%, and 24% less likely to use telephone visits than the lowest quartile (OR 0.79, 95% CI 0.73 to 0.86; OR 0.72, 95% CI 0.66 to 0.78; OR 0.76, 95% CI 0.71 to 0.82, respectively; p<0.001). Unemployed status, smoking, and Spanish-speaking status were also significant predictors of telephone visits (table 3).

Table 3

Patient characteristics associated with telephone visits

Predictors of in-person visits

On comparing patients who had in-person visits (n=30,987) to other visits (n=12,276), those who used in-person visits were more likely to be older (57.5 vs 54.3), retired (21.4% vs 17%), have Medicare insurance (22.1% vs 18.9%), higher median household income, married (50.2% vs 48.1%) and less likely to report substance abuse (table 4). On multivariate analysis, black and Hispanic subjects were 8% (OR 1.08, 95% CI 1.008 to 1.15, p=0.029) and 43% (OR 1.43, CI 1.26 to 1.63, p<0.001) more likely than white subjects to use in-person visits. Compared with English-speaking patients, patients who needed a translator for other languages were 26% more likely to use in-person visits (OR 1.26, 95% CI 1.005 to 1.58, p=0.045). Patients in the second, third, and highest quartiles of household income were 19%, 13%, and 14% more likely to use in-person visits than the lowest quartile, respectively. Older age was also a significant predictor of in-person visits (table 4). Tobacco and alcohol use and single and Spanish-speaking status were less frequently associated with in-person visits (table 4).

Table 4

Patient characteristics associated with in-person visits

Predictors of missed visits

Of the 47,032 total visits, 3768 were missed. Those who missed visits were more likely to be young (52.9 vs 56.6), belong to a non-white race (34.3% vs 32.7%), single (57.1% vs 49%), unemployed (52% vs 38.8%), in the lower two quartiles of household income (lowest to highest: missed: 34.6%/29%/17.6%/18.8% vs completed: 23.9%/24.7%/25.8%/25.6%), have Medicaid and other public insurance (26.8% vs 15.2%), report substance abuse (tobacco: 20.4% vs 13.8%; illicit drug: 9.7% vs 6.4%), and need a translator than those who completed visits (3.4% vs 2.1%, p<0.05 for all) (table 5). On multivariate analysis, black subjects, Hispanic subjects, and those of other races were 42%, 34%, and 44% more likely than white subjects to miss visits, respectively. Those with Medicaid insurance were 35% more likely to miss visits than those with private insurance (OR 1.35, 95% CI 1.23 to 1.48, p<0.001). Patients in the third, and highest quartiles of median household income were 36%, and 29% less likely to miss visits than the lowest quartile, respectively. Lastly, unemployed status, single marital status, smoking and illicit drug use, and need of a translator for other languages were significant predictors of missed visits (table 5).

Table 5

Patient characteristics associated with missed visits

Discussion

In this large retrospective cohort study of over 47,000 patients seen in the department of gastroenterology at a tertiary care center, we observed that COVID-19 has exacerbated the pre-existing inequities in healthcare access and use. Older patients (age ≥65 years, those with Medicare insurance, and retired individuals) sought healthcare less frequently during and after the pandemic lockdown. Substance abuse and missed visit rates have steadily increased in the pandemic. On further analysis, we saw new patterns: although patients belonging to vulnerable groups (ethnic and socioeconomic minorities) were more likely to have scheduled visits during the pandemic lockdown, they missed appointments more often and were also less likely to use video visits. Conversely, groups traditionally considered to be socioeconomically advantaged (white, employed, patients with private insurance and higher household incomes) sought less care during the lockdown but resumed visits post lockdown and were more likely have completed visits and used video visits.

Our analysis found concerning patterns in elderly (age 65+) patients’ and the overlapping group of patients with Medicare’s reduced healthcare access since the COVID-19 onset, which is persisting even after resumption of in-person visits. There are potential explanations for this phenomenon. One issue is delaying or avoidance of medical care due to concerns about both the risk of COVID-19 infection and its severe nature in the elderly.10 Another is that elderly patients may be at a disadvantage in a transitioned-healthcare system that has focused on increased usage of telehealth. Hospitals have used telehealth at exponentially higher rates as a result of the pandemic,11 and elderly patients have responded poorly to this shift. In fact, a similar study by Darrat et al also found increasing age to be inversely proportional to the likelihood of completing a video visit.4 The pandemic may have exacerbated the societal issue of the digital divide, in which elderly individuals are left behind in a world that becomes increasingly dependent on technology. Prior research has shown this same issue, in that older patients take longer amounts of time to familiarize themselves with technology in healthcare settings.12 Regardless of telehealth use our data demonstrates an even more pressing issue: despite the elderly’s overall lack of healthcare access during the pandemic, increasing age was associated with a higher likelihood of an in-person visit, which can potentially increase the risk of exposure to COVID-19. These trends require attention when coupled with the fact that older age is associated with increased hospitalization rate due to COVID-19.13 Simply put, the healthcare system must make more concerted efforts to integrate elderly patients into telehealth.

This study found an alarming upward trend in substance abuse since the onset of the pandemic. Increased alcohol usage has been a well-documented by-product of COVID-19.13 A 2021 systematic review by Roberts et al found an increase in usage of cannabis, opioids, and stimulants during the pandemic and a strong correlation to mental health issues.14 A large European survey also found an overall increase in tobacco and cannabis use during the pandemic.15 The major concern is that these substance abuse trends persisted and worsened well over a year since the onset of the pandemic. With the rise of more COVID-19 variants, such as Delta and Omicron, Americans may continue to turn to substance abuse to cope with this extended period of pandemic-related mental health damage.

Our findings suggest that non-white patients, patients with Medicaid and public insurance, unemployed patients, and those of the lowest quartile of household income are a vulnerable group in this pandemic landscape, with decreased telehealth use and increased missed visit rates. There are several studies in the current literature which have also found decreased telehealth use rates among non-white subjects, Medicaid or Medicare insurance, and single status individuals.5–7 16 This decrease in telehealth use can also be attributed to the aforementioned digital divide. The so-called ‘racial digital divide’ is multilayered and goes beyond owning a telephone and having internet service, for instance. A 2016 Pew Research Center study showed that a much greater percentage of black and Hispanic subjects would like training to become comfortable with usage of technology compared with white subjects.17 Our finding that non-white subjects were more likely to have missed visits is not new; for instance, Shuja et al showed that African–Americans patients had higher no-show rates than their white counterparts.18 In a prepandemic systematic review, characteristics associated with no-show visits were younger age, lower socioeconomic status, lack of private insurance, and high lead time to an appointment.19 Actually, telehealth may actually mitigate some healthcare disparities by promoting equal access to care for those with transportation or social support barriers. Providing technology training resources to benefit the underserved populations could potentially be a solution.

One final overarching theme is that the proportion of missed visits is steadily rising. The current literature shows conflicting trends; for instance, studies have found that COVID-19 had no impact on the rate of no-shows20 or that COVID-19 no-show rates were lower than those pre-COVID-19,21 while other research has shown significant increases in no-show rates during COVID-19.22 On top of that, many prior clinical studies have shown that the variables associated with no-shows can go much farther beyond the factors that our analysis looked at. Some of these more complex variables include environmental factors, such as weather and commute distance for a patient; the time between a visit and scheduling date; and a patient’s prior history of no-show.23 24 Although our analysis did not identify which type of visits were associated with no-shows and cancellations, a study by Alkilani et al found that telehealth visits during the pandemic were associated with significantly fewer no-show and cancellation rates than in-person visits.25 Thus, increased effort towards breaking down the aforementioned digital divide could hold several benefits: the resulting increased usage of telehealth could result in mitigating the spread of COVID-19 and also increasing the percentage of completed visits.

Though this study is the first to report on long-term disparities in gastroenterology care during the COVID-19 era, it has some limitations. Several factors which may impact access to healthcare, such as distance from patient residence to clinic, access to high speed internet, presence of disabilities and transportation needs, reason for the visit, appointment wait time, adequacy of insurance, and environmental factors were not assessed in this study. Nevertheless, this study was able to identify the healthcare access disparities that COVID-19 may have instigated or exacerbated and also whether these disparities have continued after the lockdown has been lifted. On top of that, rather than solely looking at telehealth use, this study looked at all different types of visits: in-person, video and telephone visits, and missed visits. Considering that telehealth will likely play an increasing role in American healthcare for many years to come, we were able to see which demographic groups will need increased assistance in using the telehealth and also identify which patients are having missed visits in the pandemic landscape. Lastly, despite being a single center study, the study population included a large, diverse sample size drawn from a tertiary care hospital as well as several community centers for the state of Ohio and hence is representative of the Midwestern population. However, these results may not be generalizable to practices that serve populations with different demographic and socioeconomic conditions.

In conclusion, the current study identified several alarming trends that are persisting amidst the pandemic: the elderly are seeking less care; socioeconomic and ethnic minorities are facing barriers in telehealth use and have higher missed visit rates, and substance abuse is peaking among other trends. There are several potential solutions to each of these problems; however, recognition is the first step. Future research efforts could involve performing a wider-scale study that is generalizable to the whole country. As of now, healthcare professionals may use this study’s findings to determine which underserved populations would need increased medical outreach and resources in the pandemic landscape.

Data availability statement

Data are available upon reasonable request. Deidentified data can be shared with the permission of institution upon request.

Ethics statements

Patient consent for publication

Ethics approval

This study was approved by the Cleveland Clinic institutional review board (IRB # 21–870).

References

Footnotes

  • Contributors HS: study concept and design, analysis and interpretation of data, drafting of the manuscript, and critical revision of the manuscript for important intellectual content. RG: statistical analysis and critical revision of the manuscript for important intellectual content. VP and JM: acquisition of data and critical revision of the manuscript for important intellectual content. CM: Study concept and design, analysis and interpretation of data, critical revision of the manuscript for important intellectual content, and study supervision. CM is the guarantor of the article.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.