Changes in Test Volumes During Coronavirus Disease 2019 (COVID-19): A Laboratory Stewardship Opportunity

The first case of COVID-19 in Texas was reported on March 4, 2020.⁸  Texas Children's Health System began testing for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA on March 10, and through the end of August 2020 performed 37 513 SARS-CoV-2 RNA tests. On March 19, Texas restricted activity, limited gatherings, and closed schools, gyms, restaurants, and bars.^6,8  From March 22 through April 22, the governor of Texas issued an executive order to postpone all elective surgery. Phased reopenings began in May. During March to June 2020, the most restrictive stay-at-home period, hospital admissions and outpatient visits decreased by 17% (from 72 544 to 59 854) and 41% (from 506 879 to 298 065), respectively, compared with the corresponding period in 2019 (solid and dotted black lines in Figures 1 through 3). Telehealth visits increased nearly 275-fold, from 537 visits in January through August 2019 to 147 259 visits in January through August 2020, which helped mitigate the impact of reduced ambulatory visits. Despite this huge increase, telehealth visits (147 259) never exceeded more than 10% of the total outpatient visits (1 552 633).

Laboratory test volumes displayed 3 distinct patterns during the pandemic: (1) a decrease coinciding with state closing, followed by near-complete or complete recovery, (2) no change, and (3) a persistent decrease. The first category of tests was impacted by the pandemic but mostly recovered, their volumes closely mirroring the health system's inpatient census. With cancelled elective surgery, surgical pathology total case counts and inpatient complete blood count test counts decreased by 28% (from 9672 to 6941) and 34% (from 106 971 to 70 854), respectively, from 2019 to 2020 during the same March to August period. Both approached close to prepandemic volumes 4 months after the phase 1 reopening (Figure 1, A and B). Estradiol, commonly used when managing women seeking fertility treatment, displayed an initial precipitous fall followed by a dramatic return to prepandemic levels once restrictions eased (Figure 2, A). Peripartum testing, for example group B Streptococcus testing of pregnant women and neonatal bilirubin (in patients ≤30 days old), remained unaffected by the pandemic, reflecting the pandemic-invariant birth counts during this period (Figure 2, B and C). The last category of tests was those used for diagnosis or monitoring of chronic illness or for screening for disease at well-patient visits, for example HbA_1c, blood lead, Papanicolaou smears, and whole-exome sequencing for childhood genetic conditions. This category recovered only partially. Hemoglobin A_1c volumes declined 80% (2232 to 452) for April compared with the same period in 2019. The normally seen seasonal trend of HbA_1c, with test volumes increasing prior to the start of school in the fall, was also disrupted. By August's end, despite the easing of mobility restrictions, there was a sustained drop in tests of approximately 16% by the end of August, from 117 in 2019 to 98 in 2020 (Figure 3, A, 28-day moving averages). Blood lead volumes dropped 39% (from 2158 to 1314) for April and ended August 2020 23% lower (from 43 to 33), compared with 2019 (Figure 3, B, 28-day moving averages). Papanicolaou smears dropped by 77% (from 719 to 167) in April and displayed a sustained decrease of more than 29% (from >21 to <15 in August) until the end of August (Figure 3, C, 28-day moving averages). Large decreases (45%) in whole-exome sequencing volumes (from 127 to 70) persisted into August 2020 (Figure 4).

Coronavirus disease 2019 has impacted health in many ways. The Blue Cross Blue Shield Association reported that routine childhood vaccinations decreased by 26% in 2020 compared with 2019.⁹  We examined COVID-19′s impact, from a laboratory diagnostics standpoint, on the health of children and women in Texas. Most laboratory tests initially declined as communities locked down and elective surgeries were canceled. Some tests, like estradiol, rebounded rapidly. The rapid fall and rise in demand of this test may reflect the discretionary nature of fertility treatments and the desire to resume them as soon as possible. The return of surgical pathology cases reflects the resumption of elective surgeries. Other tests recovered slowly, reflecting partial continuing deferral.

Tests that decreased in volume but did not completely recover are of special concern. Outpatient HbA_1c declined but only partially recovered, suggesting suboptimal pediatric diabetes monitoring. The summer rebound reflects the expected annual increase in testing before school reopenings, but this increase still does not reach the volumes seen in 2019. Newborn lead screening for at-risk children identifies environmental exposure, facilitating interventions that reduce developmental delay. Texas recommends lead testing for Medicaid patients in targeted areas, with parental request or with clinical suspicion.¹⁰  Our data suggest many children have missed blood lead screening. Because of the importance of recognizing and mitigating lead exposure early, pediatricians must quickly identify patients and arrange for screening. Decreases in whole-exome sequencing testing suggest patients with complex genetic conditions may remain undiagnosed longer, possibly leaving only suboptimal and more expensive symptomatic treatment options.¹¹ 

We do not believe that downtrends in test volumes are attributable to correction of test overuse because there were no initiatives to reduce test misuse within the data collection period. Also, 2 studies^12,13  showed that test volume variations are not associated with procedure appropriateness. Testing downtrends were also unlikely due to population loss. Allied Van Lines reported the percentage of moves into Houston (57.8%) exceeded moves out (42.2%) during 2020.¹⁴ 

Routine Papanicolaou smears declined, consistent with women's limited access to routine primary care services during the second quarter of 2020. There was a partial recovery by summer, reflecting in part a slow return of patients for regularly scheduled ambulatory visits. Current US Preventive Services Task Force guidelines recommend Papanicolaou testing at 3-year intervals or high-risk human papillomavirus testing, with or without Papanicolaou testing, at 5-year intervals.¹⁵  In July 2020, the American Cancer Society issued updated recommendations favoring primary high-risk human papillomavirus screening.¹⁶  However, based on the timing of this study and ongoing professional dialogue about the relative clinical utility of high-risk human papillomavirus testing alone versus cotesting,^17,18  new 2020 guidelines would not have influenced practice during this study period. Given the relatively long intervals specified for repeat testing, delays of several months are unlikely to significantly impact the population-based risk for cervical cancer. However, for an individual patient, a delay could be associated with adverse clinical outcomes. Furthermore, as the pandemic continues, it is important that health care providers ensure that those who have missed a scheduled cervical cancer screening obtain that service in a timely manner.

Laboratory medicine and the professionals engaged in this practice have access to data that can provide insight into clinical practice and can identify potential gaps in care. Laboratory stewardship programs are growing nationally¹⁹  and provide a framework to use data and analytic approaches to achieve effective test utilization.²⁰  Effective test utilization strives to curtail overuse, underuse, and misuse of laboratory services.¹⁹  In the context of the pandemic, supported by the data presented above, the biggest patient threat that likely exists is underuse, particularly of those tests that are essential for screening and monitoring patients for chronic diseases. Data also suggest that underuse is prevalent in traditionally underserved communities and among people of color.²¹  An effective laboratory stewardship initiative, working within its health system, offers the opportunity to identify and close these gaps. When done effectively, a laboratory stewardship program, particularly focused around care during and after the pandemic, will position laboratory professionals as health care leaders with a commitment to assure appropriate, equitable, and efficient care.

The authors thank Erin Monteverdi, BA, for project coordination, and Bill Robinson, BS, and Ryan Kennedy, BA, MBA, of hc1 for their early input in conceptualizing the data analysis.