Association Between Number of In-Person Health Care Visits and SARS-CoV-2 Infection in Obstetrical Patients

Abstract

This case-control study estimates the risk of contracting severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection among pregnant women making antenatal visits to 4 Boston, Massachusetts, area hospitals during April-June 2020 vs uninfected controls matched on gestational age.

A major concern that has emerged from the coronavirus disease 2019 pandemic is patient avoidance of necessary medical care.¹ Data regarding how in-person visits to medical facilities influence the risk of contracting severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection are limited. Obstetrical patients are a unique group who have required frequent in-person health care visits during the pandemic. The aim of this analysis was to examine whether the number of in-person health care visits was associated with the risk of SARS-CoV-2 infection.

Methods

Mass General Brigham institutional review board approval was obtained for this study and the need for informed consent waived. The study population included all patients delivering at 4 hospitals in the Boston, Massachusetts, area between April 19, 2020, and June 27, 2020, a period during which all obstetrical patients were tested for SARS-CoV-2 infection at the time of admission. All SARS-CoV-2 testing was performed on nasopharyngeal swabs using reverse transcriptase–polymerase chain reaction assays.

We performed a nested case-control study in which we used risk set sampling to match patients who tested positive for SARS-CoV-2 infection either during pregnancy or at the time of admission for labor and delivery with up to 5 control patients. The control matches were based on the gestational age of the cases and controls on the date the case tested positive for SARS-CoV-2 infection (±6 days), race/ethnicity (recorded in the patient’s medical record; Black vs Hispanic vs Asian or White), insurance type (Medicaid vs commercial), and SARS-CoV-2 infection rate in the patient’s zip code (divided in 20 groups by ventile).²

Based on electronic medical record data, we assessed the number of in-person visits for patients from March 10, 2020 (2 weeks prior to the closure of nonessential business in Massachusetts when community transmission was likely), to the date of the cases’ SARS-CoV-2 infection diagnosis. The association between the number of in-person visits and the odds of SARS-CoV-2 infection diagnosis was assessed using conditional logistic regression with adjustment for age, body mass index (BMI; calculated as weight in kilograms divided by height in meters squared), and essential worker occupation.³ We used multiple imputation to account for missing regression covariates (0.6% were missing BMI and 11.6% were missing essential worker occupation).

The odds ratios with corresponding standard errors were obtained from each of 10 imputed data sets and combined using the rules of Rubin⁴ to produce pooled estimates with 2-sided 95% CIs. We performed sensitivity analyses assessing the number of clinic visits after March 24, 2020 (the date of closure of nonessential businesses), excluding patients with a household member with known SARS-CoV-2 infection, patients testing positive for SARS-CoV-2 infection antenatally, and patients with incomplete covariate information. Precision around the measures of association is provided using 2-sided 95% CIs. Statistical analyses were performed using SAS software version 9.4 (SAS Institute Inc).

Results

The study population included 2968 deliveries; 5 patients were not tested for SARS-CoV-2 infection and were excluded. There were 111 patients (3.7% [95% CI, 3.1%-4.5%]) who tested positive for SARS-CoV-2 infection. Of these 111 patients, 45 tested positive for SARS-CoV-2 infection antenatally and 66 tested positive at the time of admission for labor and delivery.

We excluded patients residing outside Massachusetts (2.2%) and those missing data required for matching (0.8%). We then matched 93 cases with 372 control observations. The mean number of in-person visits was 3.1 (SD, 2.2; range, 0-10) for cases and 3.3 (SD, 2.3; range, 0-16) for controls. For the association between the number of in-person health care visits and SARS-CoV-2 infection, the odds ratio was 0.93 (95% CI, 0.80-1.08) per additional visit. Sensitivity analyses yielded similar results (Table).

Table. Association Between Each Additional In-Person Health Care Visit and Odds of SARS-CoV-2 Infection.

	Cases		Control observations		OR (95% CI)
	No.	Clinic visits, mean (SD)	No.	Clinic visits, mean (SD)
Primary analysisa
Unadjustedb	93	3.1 (2.2)	372	3.3 (2.3)	0.93 (0.80-1.07)
Adjustedc	93	3.1 (2.2)	372	3.3 (2.3)	0.93 (0.80-1.08)
Sensitivity analysesd
Assessing No. of clinic visits after March 24, 2020	90	2.5 (2.2)	357	2.6 (2.1)	0.91 (0.76-1.09)
Analyses excluding patients who
Had a household member with known SARS-CoV-2 infection	68	3.5 (2.2)	270	3.6 (2.3)	0.97 (0.82-1.14)
Tested positive for SARS-CoV-2 infection antenatally	53	4.2 (2.2)	201	4.3 (2.4)	0.97 (0.82-1.15)
Complete case analysis	82	3.1 (2.2)	318	3.2 (2.4)	0.91 (0.78-1.07)

Abbreviations: OR, odds ratio; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

^aAssessing the number of in-person visits for patients from March 10, 2020, which was 2 weeks prior to the closure of nonessential business in Massachusetts.

^bAfter matching on the gestational age of the cases and controls based on the date the case tested positive for SARS-CoV-2 infection (±6 days), race/ethnicity, insurance type (Medicaid vs commercial), and SARS-CoV-2 infection rate in the patient’s zip code.

^cAdjusting for age, body mass index, and essential worker occupation.

^dAll estimates were matched for the same covariates as in the primary analysis and were also adjusted for age, body mass index, and essential worker occupation.

Discussion

There was no meaningful association between the number of in-person health care visits and the rate of SARS-CoV-2 infection in this sample of obstetrical patients in the Boston area. Massachusetts had the third highest SARS-CoV-2 infection rate in the country during the spring 2020 surge, and the Boston area was particularly affected.

The findings from this obstetrical population who had frequent in-person visits to a health care setting and underwent universal testing for SARS-CoV-2 infection suggest in-person health care visits were not likely to be an important risk factor for infection and that necessary, in-person care can be safely performed. Limitations of this study include the restriction to obstetrical patients. Future studies are needed to determine whether these findings extend to other populations and health care settings.

Section Editor: Jody W. Zylke, MD, Deputy Editor.