In-Person and Virtual Clinic Visit Frequency to Rheumatologists for Rheumatoid Arthritis at an Academic Medical Center Before, During, and After COVID Lockdown

Abstract

Introduction

This study aimed to describe outpatient visit volume in a subspecialty clinic before, during, and after COVID lockdown.

Methods

We assessed monthly in-person and virtual visit volume (telephone-only or video) of 257 patients with rheumatoid arthritis (RA) at one academic center before, during, and post COVID lockdown, November 2018 to September 2021. The primary outcome was monthly visit volume to a rheumatologist. Visit volume, visit type (in-person vs. virtual), and annual visit frequency per patient were assessed. Piecewise Poisson regression models were constructed to examine visit volume trends. Predictors of patient’s visit volume before and after the lockdown were examined using multivariable linear regression.

Results

Median patient age was 58 years; 84% were female; 82% used any disease-modifying anti-rheumatic drug (DMARD), and 62% used a targeted or biologic DMARD. Visit volume was stable 18 months prior to the COVID pandemic [slope 1.00 (95% confidence interval (CI) 0.99–1.01)] and increased at a rate of 2% per month post-lockdown [1.02 (95% CI 1.01–1.03)]. In-person visit volume was greatly reduced during the lockdown, with 61% virtual (51% video, 10% telephone). In the 18 months after lockdown, visit volume rebounded to pre-pandemic levels and continued to increase, with 11% virtual. Older age, serologic status, use of combination DMARDs, and non-steroidal anti-inflammatory drug (NSAID) use predicted greater visit volume during the pre-lockdown period. No variables predicted visit volume post-lockdown.

Conclusion

While COVID caused a huge disruption in rheumatology practice, visit volume for RA rebounded in one American academic center, with an increasing slope in visit volume after lockdown.

Key Summary Points

Why carry out this study?

Visit frequency in rheumatology is largely without a substantial evidence basis and was called into question during COVID.

Hypothesis: Visit frequency would differ by patient characteristics.

What was learned from the study?

Visit volume for rheumatoid arthritis increased at a rate of 2% per month after the COVID lockdown.

In-person visit volume was greatly reduced during the lockdown.

After lockdown, in-person and virtual visit volume rebounded to pre-pandemic levels and continued to increase with a substantial portion being virtual.

Introduction

The COVID-19 pandemic caused major disruptions to care delivery. Many physicians’ offices were closed for several months, and virtual care dominated. Visit volumes initially plummeted [1]. These changes have been well documented in primary care settings [2], but less is known about how the pandemic affected subspecialty care. The impact of the pandemic on visit patterns may have long-lasting effects on access and utilization of care, as is suggested by an analysis of visit data in California showing a sustained increase in virtual care [3].

To understand how the pandemic affected and may continue to influence visit patterns in subspecialty care, we conducted a retrospective analysis of the electronic health records of a cohort of patients being treated for rheumatoid arthritis (RA) before, during, and after the peak of the COVID-19 pandemic at a single large academic rheumatology practice. Our goals were, first, to describe outpatient rheumatology visit frequency for RA during this time (2019 through 2021) for in-person, virtual, and total visits; and second, to examine how visit frequency differed for different patient characteristics during the same period.

Methods

Study Design and Population

This was a longitudinal cohort study of a convenience sample of 257 patients with RA treated by a group of academic rheumatologists who are primarily engaged in clinical care (not research) at one medical center in Boston MA. The study period of interest encompassed the 18 months prior to COVID pandemic (November 2018–February 2020), the 3 months of COVID lockdown (rheumatology clinic closed except emergencies, March–May 2020), and the 18 months after COVID lockdown (June 2020–September 2021).

All patients were seen by one of 11 rheumatologists participating in a separate longitudinal study of patient symptoms. Patients were included if they had been diagnosed with RA according to their rheumatologist and had at least one visit within 18 months prior to plus at least one visit within 18 months after the lockdown. Patients were excluded if they were followed for less than 1 year by the same rheumatology provider to reduce the number of one-time consultations.

This study was approved by the Brigham and Women’s Hospital (BWH) Institutional Review Board (protocol number 2021P000790).

Outcome Definitions

The primary outcome was the volume of in-person plus virtual visits (telephone-only or video) for RA to the patient’s primary rheumatologist. Visits were identified from the hospital billing system using specific common procedural terminology (CPT) codes entered by the specified rheumatologists during the date ranges of interest. We calculated monthly RA visit volume during the three periods of interest: pre-lockdown, lockdown, and post-lockdown. The volume of total visits and types of visits were compared across the periods of interest (see below).

The secondary outcome was annualized visit frequency per patient. In addition, the median annualized visit frequency was estimated for each rheumatologist.

Other Variables

We extracted baseline variables of interest from the medical record, using information from the 12 months prior to the start of the study period (November 2017–November 2018). These included sociodemographic (age and sex), RA characteristics (serologic status, use of disease-modifying anti-rheumatic drugs (DMARDs), use of corticosteroids, use of opioids, use of non-steroidal anti-inflammatory drugs, NSAIDs), and the specific rheumatologist seen by the patient. These variables have all been related to visit frequency in prior studies [4].

Statistical Analysis

We first derived descriptive statistics of the 257 included patient characteristics, presented as median (interquartile range, IQR) for continuous variables and frequency (%) for categorical variables. Summary statistics of the total number of outpatient rheumatology visits for the three periods were calculated, overall and by type (in-person vs. virtual). We considered visits in two ways. First, an overall visit volume by month for all 257 patients included. Second, we calculated the annualized visit frequency for each patient.

The total visit volume was plotted by month, and locally estimated scatterplot smoothing (LOESS) curves were fit to allow visual inspection of any potential trends. We used piecewise Poisson regression with knots at month 18 and 22 to estimate the trends in the count of monthly visits during the three time periods [pre-lockdown (months 0–18), lockdown (months 19–21), post-lockdown (months 22–39)]. The slope of each of the spline curves can be described as a change in the rate of visit volume per month. The 95% confidence intervals in monthly slope were estimated. No statistical testing between the slopes was pursued as there was no a priori hypothesis regarding the slopes.

We subsequently investigated the contributions of patient characteristics on visit volume by constructing linear regression models with the number of rheumatologist visits as the dependent variable. Models were constructed separately for the 18 months pre-lockdown and the 18 months post-lockdown. We also combined these periods in a secondary analysis. The models used linear mixed-effects regression models to allow for inclusion of a random effect for rheumatologist. All variables were tested in univariate models for the two separate time periods; non-significant variables were removed. Patient’s age, sex, and race were forced into all models. Additional patient characteristics considered for inclusion were comorbidity index, medication use (NSAIDs, opioids, glucocorticoids, and DMARDs), serologic status, and C-reactive protein level. Variables with p < 0.1 on univariate regression were advanced to the multivariable models. To explore the effect of clinician on visit volume, the median number of visits per patient to each rheumatologist was examined separately. All statistical analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC, USA) and R version 4.2.2 (R Core Team 2022). P values were two-sided, and statistical significance was set at p < 0.05.

Results

As seen in Table 1, a total of 257 patients were included in our study with a median age of 58.2 years. A majority of patients were white (87.5%) and female (84.0%) with a median Charlson-Deyo comorbidity index of 1 (IQR 1, 1). In the 12 months prior to the start of the study period, over half of the study population (68.0%) were seropositive for either rheumatoid factor or anti-CCP antibody with a median high-sensitivity C-reactive protein (hsCRP) level of 2.0 mg/L (IQR 0.7, 5.0). Out of all included patients, 14.8% were receiving an opioid, a quarter were receiving NSAIDs, and a third were taking corticosteroids. Rates of use of conventional synthetic disease-modifying anti-rheumatic drugs (csDMARDS) were 57.2% and targeted/biologic DMARDS (t/bDMARDS) 67.5%, with 37.0% patients receiving both.

Table 1. Baseline characteristics of 257 patients with rheumatoid arthritis followed in the study cohort

Patient characteristics	N (%) or median (IQR)
Age, years	58.2 (48.7, 65.8)
Female sex	216 (84%)
Charlson-Deyo comoribidity index	1.0 (1.0, 1.0)
Race
Black	11 (4.3%)
Other	21 (8.2%)
White	225 (87.5%)
Seropositive (n = 128)	87 (68%)
hsCRP, mg/L	2.0 (0.7, 5.0)
DAMRD use
csDMARD	147 (57.2%)
t/bDMARD	158 (61.5%)
Both	95 (37%)
Neither	47 (18.3%)
Non-steroidal anti-inflammatory drug use	61 (23.7%)
Glucorticoid (oral) use	84 (32.7%)
Opioid use	38 (14.8%)
Identify of the provider
A	24 (9.3%)
B	15 (5.8%)
C	9 (3.5%)
D	8 (3.1%)
E	41 (16.0%)
F	35 (13.6%)
G	4 (1.6%)
H	5 (1.9%)
I	16 (6.2%)
J	89 (34.6%)
K	11 (4.3%)

Seropositive refers to either rheumatoid factor or anti-CCP antibody positive

Abbreviations: IQR interquartile range, DMARD disease-modifying anti-rheumatic drug, csDMARD conventional synthetic DMARD, t/b DMARD targeted/biologic DMARD

Figure 1 illustrates the monthly visit volume during the study period. The median visit volume pre-lockdown was 61 (IQR 51, 58), during lockdown 58 (51, 65), and went up post-lockdown 68 (IQR 50, 73). During the first 3 months post lockdown (June–August 2020), virtual visits comprised 61% of all visits (51% videoconferencing, 10% telephone only). By the 12th month post-lockdown (April 2021), virtual visits comprised 11% of all visits; all visits were via videoconferencing. The annualized patient visit frequency in the 18 months prior to COVID lockdown was 2.7 (IQR 2.0, 3.3); after lockdown it was also 2.7 (IQR 2.0, 3.3). Since lockdown was only 3 months, we did not calculate an annualized visit frequency.

[See PDF for image]

Fig. 1

Longitudinal pattern of monthly visit volume across 257 patients with rheumatoid arthritis

We visualized the association between visit volume and time using LOESS curve (Supplemental Figure), which suggested increasing visit volume by time. The smoothed curve in Fig. 2 is based on linear splines using piecewise Poisson regression. It demonstrates that during the pre-lockdown period (months 0–18) the slope was stable at 1.00 (95% CI 0.99–1.0; p = 0.81). During the post-lockdown period (months 22–39), the slope was increasing at 2% per month (1.02, 95% CI 1.01–1.03; p = 0.037).

[See PDF for image]

Fig. 2

The smoothed curve is based on linear splines using piecewise regression. The curves demonstrate the predicted curve based on the actual monthly visit volume (shown in blue circles). The x-axis is the month: the left side (months 0–18) the pre-lockdown period, center section (months 19–21) the lockdown period, and the right side (months 22–39) the post-lockdown period. The y-axis represents the number of rheumatoid arthritis visits per month for the total sample of patients. The pre-lockdown slope was 1.00 (95% CI 0.99–1.01) and the post-lockdown slope was 1.02 (95% CI 1.01–1.03)

Multivariable linear mixed effects models were constructed to investigate the relationship between patient visit volume and patient characteristics. Models accounted for the clustering of patients within rheumatologists. The models for the 18-month pre- and post-lockdown periods, and combined, gave somewhat different results (see Table 2): patient predictor for more frequent visits pre-lockdown were older age, seropositive status, use of combination DMARDs, and NSAID use; whereas post-lockdown, no variables were significant predictors of visit frequency. None of the other patient characteristics tested (e.g., sex, race, or oral corticosteroid use) were significantly associated with visit volume (see Supplemental Table for unadjusted regression coefficients).

Table 2. Multivariable linear regression for 257 patients with RA seen over 36 months, total number of rheumatology visits as the outcome

Patient characteristics	Beta coefficient (95% CI)
	Pre-lockdown 18 months	Post-lockdown 18 months	Combined 36 months
Age, per year increase	0.02 (0.00, 0.04)	0.01 (−  0.01, 0.03)	0.03 (0.00, 0.06)
Sex
Female	Reference	Reference	Reference
Male	− 0.05 (− 0.68, 0.58)	− 0.49 (− 1.1, 0.12)	− 0.54 (− 1.6, 0.51)
Race
Black	0.04 (− 1.1, 1.2)	− 0.23 (− 1.3, 0.87)	− 0.29 (− 2.1, 1.6)
Other	0.04 (− 0.38, 1.3)	− 0.5 (− 1.3, 0.32)	0.07 (− 1.3, 1.5)
White	Reference	Reference	Reference
Seropositive (n = 128)	0.80 (0.11, 1.5)	NA	0.56 (− 0.59, 1.7)
DAMRD use, none	Reference	Reference	Reference
csDMARD	0.24 (− 0.52, 1.0)	0.24 (− 0.52, 1.0)	− 0.71 (− 2.0, 0.55)
t/bDMARD	0.12 (− 0.61, 0.85)	0.12 (− 0.61, 0.85)	− 0.52 (− 1.7, 0.68)
Both	0.40 (0.05, 1.1)	0.40 (0.05, 1.1)	0.02 (− 1.1, 1.2)
Non-steroidal anti-inflammatory drug use	0.59 (0.05, 1.1)	NA	0.79 (− 0.10, 1.7)
Glucorticoid (oral) use	0.28 (− 0.23, 0.80)	0.28 (− 0.23, 0.80)	0.70 (− 0.15, 1.6)

The Beta coefficient can be interpreted as the increased (or decreased) number of visits associated with a give patient characteristic

Seropositive refers to either rheumatoid factor or anti-CCP antibody positive. NA, not applicable since the variable was not advanced to multivariable models

Abbreviations: IQR interquartile range, DMARD disease-modifying anti-rheumatic drug, csDMARD conventional synthetic DMARD, t/b DMARD targeted/biologic DMARD

Figure 3 shows the median number of monthly visits to rheumatologist for the same 257 patients in the study cohort during the pre-lockdown (Fig. 3a) and post-lockdown (Fig. 3b) for each of the 11 providers. The two graphs show similar results with slightly more variability post-lockdown.

[See PDF for image]

Fig. 3

Median number of per patient visits and interquartile range for each rheumatologist. a 18 months before lockdown. b 18 months after lockdown

Discussion

We studied the clinical visit frequency and visit volume for patients with RA at a large academic medical center, with a study period spanning the 18 months before, 3 months during, and 18 months after COVID lockdown. We focused on their visits to 11 different rheumatologists, examining trends and predictors of visit volume. Visit volume appeared stable in the 18 months prior to the COVID pandemic with a median annual visit frequency per patient of 2.7. While in-person visit volume was greatly reduced during the 3-month lockdown with a switch to virtual care, visit volume rebounded to pre-pandemic levels with an increasing trend post-lockdown and a sustained use of virtual visits.

In contrast to the trends in outpatient visit frequency from a 2020 US study using national databases, our outpatient rheumatology clinic did not experience a sharp decline in in-person visit in mid-February 2020 [5]. In fact, our in-person visit volume remained constant for the 18 months preceding COVID lockdown, until March 2020 when our outpatient clinic shutdown (Fig. 1). The subspecialty clinic being studied experienced a rapid replacement of in-person visits by telephone and videoconferencing visits; both contributed to an increasing trend in visit volume after the lockdown. The national average for all outpatient visit volume during the same time period (February 2020–May 2020) was only at 60% of the pre-pandemic levels [5]. The subspecialty rheumatology clinic’s reduction in total visit volume during the COVID lockdown was small and below the average in the US Northeast, which reported an average reduction between 31% and 73% [5]. In contrast to all-specialty studies done in the New England region from the same time period, which reported a seemingly stable percentage of virtual visits from July 2020 to April 2021 (29.6%), the clinic being studied showed a diminishing trend of virtual visits since the lockdown ended in May 2020 [6]. Many factors may contribute to this difference, including state insurance regulations, clinic-specific practices in masking, and ease of virtual visits.

Further comparisons can be drawn between the results of this study with other rheumatology-specific studies, such as those reported for the province Ontario, Canada from March 2019 (during lockdown) to October 2021 [7]. Consistent with our study, results from the Ontario study demonstrated stable in-person visit volumes until the COVID lockdown occurred in March 2020. In contrast to our findings, the Ontario study observed a longer lag to reach their pre-lockdown visit volumes: they reached pre-lockdown visit volumes by September 2020 and the clinic we studied reached them by summer of 2020 (Fig. 1). While the exact reasons are unclear, it may be that clinic infrastructure and financial incentives allowed for a quicker rebound.

While this set of analyses was not undertaken with a specific hypothesis regarding trends in visit volume pre- and post-lockdown, the slopes suggest visit volume differences (Fig. 2). The pre-lockdown period slope was “flat”, and post-lockdown the slope was upward. It is possible that visit volumes became more stable later in 2021 and into 2022. However, it is possible that the ability to use virtual visits plus the pandemic stimulated an increase in visit volume; this has been suggested by other research [8]. This will need to be tested with more recent data.

The current study has important strengths. First, it represents a detailed study of visit frequency across the COVID pandemic, using a subspecialty practice in the USA. Second, we identified patient factors associated with visit volume, considering the clustering of patient factors within clinicians’ practices; there has not been a similar study conducted in rheumatology since 1997 [9]. Limitations of this study are its relatively small size and the focus on one academic medical center, which may not be representative. We included 11 rheumatologists, but the number of patients was uneven across these clinicians. The pre- and post-COVID periods were chosen to be 18 months, which was felt to be appropriately long. However, the choice of 18 months is somewhat arbitrary.

Conclusions

We found outpatient visit volumes in one subspecialty practice quickly resumed and appeared to be trending upward as virtual visits and in-person visits are both used post-lockdown. The COVID-19 pandemic has had profound impact on all facets of outpatient healthcare delivery which will reverberate for years. Virtual visits became the standard of care during the COVID pandemic lockdown, a format that many patients grew to appreciate. The convenience of virtual care was made possible by reducing patients’ logistical and financial burden, thereby improving healthcare access. Improved access, in turn, allows more timely treatment responses to changes in disease activity. The virtual care format was similarly appreciated by clinicians, many of whom scheduled entire sessions as virtual, allowing them to not travel from home to their office. A better understanding of how to improve the efficiency and outcomes of outpatient care becomes more important as virtual care becomes easier.

Acknowledgements

We thank the participants of the study.

Author Contributions

Yuxuan Jiang: study conceptualization, data interpretation, wrote first draft; Robert Rudin: study conceptualization, revised first draft; Leah Santacroce: data analysis, revised first draft; Jamie Collins: data analysis, revised first draft; Jackie Stratton: revised first draft; Hallie Altwies: revised first draft; Daniel Solomon: obtained funding, study conceptualization, revised first draft.

Funding

Rheumatology Research Foundation funded the drafting of the manuscript. The journal’s Rapid Service Fee was funded by the authors.

Data Availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Conflict of Interest

Daniel Solomon has received salary support in the last year from research contracts to Brigham and Women’s Hospital with CorEvitas, Janssen, Novartis, and Moderna. He also receives royalties for work on NSAIDs and coxibs from UpToDate. Jamie Collins serves as a statistical consultant for Boston Imaging Core Labs. No other authors (Yuxuan Jiang, Leah Santacroce, Jackie Stratton, and Hallie Altwies) have any conflicts to declare. Robert Rudin participated in this work outside of his employment at the RAND Corporation.

Ethical Approval

This study was approved by the Brigham and Women’s Hospital (BWH) Institutional Review Board (Protocol Number 2021P000790).

References

1. Baum, A; Kaboli, PJ; Schwartz, MD. Reduced in-person and increased telehealth outpatient visits during the COVID-19 pandemic. Ann Intern Med; 2020; [DOI: https://dx.doi.org/10.7326/M20-3026] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32776780][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7429994]

2. Tu, K; Sarkadi Kristiansson, R; Gronsbell, J et al. Changes in primary care visits arising from the COVID-19 pandemic: an international comparative study by the International Consortium of Primary Care Big Data Researchers (INTRePID). BMJ Open; 2022; 12, 5 [DOI: https://dx.doi.org/10.1136/bmjopen-2021-059130] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35534063]e059130.

3. Whaley, CM; Ito, Y; Kolstad, JT; Cowling, DW; Handel, B. The health plan environment in California contributed to differential use of telehealth during the COVID-19 pandemic. Health Aff; 2022; 41, 12 pp. 1812-1820. [DOI: https://dx.doi.org/10.1377/hlthaff.2022.00464]

4. Jiang, Y; Rudin, RS; Solomon, DH. Clinical visit frequencies in rheumatology: a systematic literature review. Arthritis Care Res; 2023; [DOI: https://dx.doi.org/10.1002/acr.25106]

5. Patel, SY; Mehrotra, A; Huskamp, HA; Uscher-Pines, L; Ganguli, I; Barnett, ML. Trends in outpatient care delivery and telemedicine during the COVID-19 pandemic in the US. JAMA Intern Med; 2021; 181, 3 pp. 388-391. [DOI: https://dx.doi.org/10.1001/jamainternmed.2020.5928] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33196765]

6. Zachrison, KS; Yan, Z; Schwamm, LH. Changes in virtual and in-person health care utilization in a large health system during the COVID-19 pandemic. JAMA Netw Open; 2021; 4, 10 [DOI: https://dx.doi.org/10.1001/jamanetworkopen.2021.29973] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34705016][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8552053]e2129973.

7. Kwok, TSH; Kuriya, B; King, LK et al. Changes in service delivery and access to rheumatologists before and during the COVID-19 pandemic in a Canadian universal healthcare setting. J Rheumatol; 2022; [DOI: https://dx.doi.org/10.3899/jrheum.220658] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36521914]

8. Ashwood, JS; Mehrotra, A; Cowling, D; Uscher-Pines, L. Direct-to-consumer telehealth may increase access to care but does not decrease spending. Health Aff (Millwood); 2017; 36, 3 pp. 485-491. [DOI: https://dx.doi.org/10.1377/hlthaff.2016.1130] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28264950]

9. Ward, MM. Rheumatology visit frequency and changes in functional disability and pain in patients with rheumatoid arthritis. J Rheumatol; 1997; 24, 1 pp. 35-42.1:STN:280:DyaK2s7lsFaqtg%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/9002008]