The emergency visits for back and neck pain during lockdowns


 Purpose

To assess the rate of visits to the emergency department of our medical center concerning low back or neck pain as a factor of COVID-19 confinement.
Methods

The study period was a 30-week interval during the COVID-19 pandemic contrasted by a similar stretch in the year preceding the epidemic. Visits to the emergency department prompted by low back or neck pain were recorded prior to and during lockdowns of the pandemic. The significance of the confinements for the development of pain syndromes was evaluated.
Results

A total of 1530 patients with newly diagnosed back or neck pain were enrolled. Most patients visited our emergency department for low back pain, commonly those older than 60 years. No significant gender variance was disclosed, although most visits of females were for low back pain. Low back pain presentations were curbed following confinement, but the rate of stays for neck pain swelled by more than 10%. Despite back pain predominance, visits for neck pain persisted. Before COVID-19, the average weekly number of emergency department visits was 38.5. This was followed by sharp drops during the COVID-19 lockdown (mean difference=-22.2, 95% CI=-28.7, -15.7, p<0.001) (not significant).
Conclusions

COVID-19 lockdowns have a significant impact on emergency department presentations due to back and neck pain. A higher rate of presentation for back pain compared to neck pain is probably related to COVID-19, without being affected directly by SARS-CoV-2: confinement-induced immobility might instigate musculoskeletal sequelae, which may be attributed to stress or other psychosocial afflictions.


Introduction
Low back pain (LBP) and neck pain (NP) are common complaints that tend to recur with aging. The etiology varies from nonspeci c causes to speci c disease entities, such as herniated discs, infectious diseases or tumors [1,2]. The mode of expression is therefore diversi ed. In addition to comparing LBP and NP, we intend to assess these complaints as they develop in different categories of patients [3][4][5].
Remarkably, we were able to scrutinize the behavior of two conditions that, a priori, should exhibit two cohorts with LBP and NP patients who display a similar clinical picture and a parallel clinical outcome [6].
Both subsets were composed of adult patients with the abovementioned syndromes. Both cohorts were free from associated COVID-19 since a prerequisite for ED attendance was the lack of any respiratory or febrile complaint. Only patients who were negative for the virus were treated at the ED. Patients with a positive SARS-CoV-2 RT-PCR test were steered to the COVID-19-ED. The association between COVID-19 con nement and musculoskeletal conditions was highlighted [3,5].

Materials And Methods
It is hypothesized that, as a consequence of the con nements implemented by the authorities to ght the spread of SARS-CoV-2, general immobilization will develop. A secondary effect may evolve in the form of a marked limitation of presentations to the ED, among others of LBP and NP.

Study design and population
An acceptable terminology should rst be established. Lockdown is employed interchangeably for con nement; ED visit (ED stay, ED sojourn) for ED presentation; low back pain for low back pain/sciatica; neck pain for neck pain/cervicalgia. The simplest terms are to be adopted whenever possible.
Back and neck pain were de ned by the International Statistical Classi cation of Diseases and Related Health Problems, Tenth Revision (ICD-10) code (Fig. 1). All patients included in our study were SARS-CoV-2 RT-PCR negative and were cared for in the general ED. Each patient's demographics, length of stay in the ED and diagnoses were retrieved from the hospital electronic database. No individual identi ers (e.g., name and address) were disclosed. Patients were excluded from the analysis if they were readmitted or if they were younger than 18 years. The study was approved by the Institutional Review Board prior to data collection.

Exposure and Outcome
The study period was categorized into two groups, namely, 30 weeks before the COVID-19 pandemic lockdowns began and 30 weeks after that. The primary outcome was the mean weekly number of patients with newly identi ed back or neck pain. A relationship between spinal syndromes and reduced physical activity was displayed.

Statistical analysis
The number of patients n (%) and mean (standard deviation, SD) were used to describe continuous and categorical variables, respectively. Analysis of variance (ANOVA) or the chi-squared test was used to examine differences in demographics between two groups (before and after lockdown). We applied an interrupted time-series analysis using segmented linear regression to explore the impact of COVID-19 lockdowns on the trends of ED presentations occasioned by LBP and NP [7]. In the interrupted time series analysis, β 0 estimated the start level of the weekly ED presentations at 1 year before lockdown (intercept); β 1 estimated the before-lockdown slope, which quanti ed the trend for the weekly ED presentations before lockdown; β 2 estimated the change in level at the 1st lockdown point, which can arguably be attributed to lockdown; and β 3 estimated the change in slope from before to after lockdown, which quanti ed the difference between the prior and ensuing lockdown slopes. All analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). Two-sided p <0.05 was considered statistically signi cant.

Results
A total of 1898 emergency department visits for neck and back pain were recorded in the hospital database between March 2019 and November 2020. Of these, 368 patients were excluded since they did not belong in the study period or due to different diagnoses or missing data. Figure 2 describes the ow diagram de ning the eligible patients. Table 1. Represents the patients' weekly presentation to the ED in the period preceding the pandemic, as well as during the lockdowns. As expected, the variance between the mean number of sojourns in both periods is signi cantly different, with a median number of 36.5 visits a week during the year preceding the pandemic, compared to 16 visits along its stretch.
As shown in Table 2, there was a signi cant difference between the weekly visits at the beginning of lockdowns (mean difference=-22.2, 95% CI=-28.7, -15.7, p<0.001). However, there was no signi cant difference in the weekly visit trend between the period preceding the pandemic and that encompassing the lockdowns (slope different = 0.19, 95% CI=-0.08, 0.45, p=0.17). Table 3 highlights the demographics and diagnoses attributed in the ED, as they relate to the status of the lockdowns. More males were identi ed. The mean age was higher before the pandemic. The mean length of stay in the ED was longer before the lockdowns. Although a sustained predominance of LBP is evident at all times, neck pain exhibits discordance: while the number of visits is reduced, from "before lockdowns", the rate of visits with neck pain increases from 24.4% to 34.1%.
In the rst quarter of the pandemic, the number of presentations dropped signi cantly to 25 a week, in contrast to 31 in the previous year. However, in the third quartile, which occurred after this country had left behind it the rst episode of con nement, the mean number of visits to the ED expanded to a weekly mean of 50, similar to that of the parallel quartile, in the year preceding the epidemic.   Immobility and lack of physical activity is proposed as a potential mechanism by which back and neck pain syndromes might be established following COVID-19 lockdowns [3,5].

Discussion
The present study intended to scrutinize whether the COVID-19 pandemic changed the pattern of attendance of patients with low back or neck pain to the ED. Nonspeci c pain in the neck or the low back is considered one of the most frequent painful syndromes, a majority of which are self-limited. Various risk factors have been identi ed regarding these conditions, such as age, obesity, inactivity (or in contrast, hard physical work) and smoking [7]. Our basic assumption is that since the COVID-19 pandemic did not cause an essential change in most of the above factors, a signi cant difference in the rate of visits to the ED related to these syndromes will not be observed.
A marked reduction in ED visits during the lockdowns might suggest a possible abuse of the facility prior to the pandemic. A severe drop in the frequency of cardiologic and neurologic visits to the ED is especially vexing [8]. A report from Finland supports a marked decrease in ED consults regarding back and limb pain but not concerning acute myocardial infarction during the early con nements of the COVID-19 pandemic [9]. A signi cant limitation in the number of presentations during lockdowns in the ED was interpreted as a lack of willingness to invest the necessary effort to acquire a medical facility if and when required [10]. An evaluation of the number and rate of ED visits before the start of the COVID-19 pandemic was rare. One study, however [11], did rise to the challenge. Here, the number of ED visits after the start of con nement displayed a marked reduction. In the present study, the majority of visits to the ED during the pandemic were due to LBP, mostly in patients older than 60 years of age. In contrast, most patients who visited the ED at the beginning of the pandemic were between 40 and 60 years of age. From a gender perspective, most women who visited the ED had LBP.
Consideration given to the number of weekly visits, as they occur before the epidemic or during its course (Table 1. ), one observes a variance between the two eras, and that, to a signi cant degree! During COVID-19, patients suffering from back-or neck pain visited the ED in reduced numbers. It cannot be excluded that the above drop in presentation of the ED may represent a reduced morbidity from these disorders. However, another option is that, due to the lockdowns, individuals did indulge to a lesser degree in all forms of physical activity, including that practiced in closed spaces. Therefore, limited episodes of low back and neck pain are expected, which requires a less frequent rate of ED attendance. An alternative explanation may be that, in a normalized period, the emergency facilities are freely attainable, and it is possible that a proportion of the visitors of the ED do not precisely need the exam or the treatment, and such abuse might be practiced at times. Support for this thesis is displayed in the high variance between the low mean number of visits in the rst quartile of the pandemic and the near-normal span that follows. Unfortunately, the above thesis is not anchored onto hard data. One is therefore invited to evaluate the proposed ideas and to suggest some of his own, since speculations are not excluded hereby.
In contrast with the LBP patients, there were fewer complaints of NP. While there was a contraction in the absolute number, they represented a higher rate of ED visits (Tables 3, 4). The role that gender might have played in this investigation is probably limited. Low back pain is connected mainly with the precon nement days, with age groups 40 to 60. The time spent in the ED by these patients is usually more than two hours (Table 4). Figure 3 underlines the breakdown occurring during the transition from prelockdown into the rst con nement. A marked drop in the number of visits to the ED may suggest that the effect described in relation to Table 1 might not be an artefact. During the rst lockdown, many people had avoided hospitals, even when in serious need, out of fear of contamination. It is probable that during the later con nements, the segregation between COVID-19 ED and general medical ED was tighter.
Two values in Table 2 are worth a discussion. One is the intercept, describing the rate of visits to the ED one year before lockdown. The number of ED visits per week is high, and the trend is positive. The second is the degree of change after the lockdown. The rate of visits per week is moderate, but the trend is negative. It appears that together with musculoskeletal (MSK) symptoms, psychological and psychosocial complications become evident when the spine is predominantly affected. Thus, the inactivity and immobility brought about by con nement may be at the origin of the symptoms. It has been stated previously that the predominant occurrence of LBP is overwhelming and is independent of whether con nement is taking place. The lack of consequent genetic or environmental factors is suggested. The lockdown re ects immobility and a limitation regarding work activity [13,14].
Overall, we con rm that more women than men visit the ED due to LBP. The causes have not been elucidated. One possible factor includes the hardship in sustaining quarantine, though it might be shared with lonely men. Visiting the ED offers the potential to socialize with other human beings. Hormonal factors should be taken into consideration, perhaps via the intermediary of calcium-modulating hormones. Last, we note that different emotional stressogenic causes may contribute to MSK pain of inde nite quality, which in some cases is di cult to differentiate from a standard initiation of LBP. One should look at the COVID-19 pandemic's lockdowns as potential stimulators of such episodes. Limitations: The duration of 30 weeks suggested for the contrasting period associated with the compiled lockdown intervals is not precisely estimated. The reason for this miscalculation originates both in a variable number of con nements and in the changing nature of the lockdowns. While the rst two sequences were strictly observed, this was not the case for the last two con nements. Conclusions: Evidence of a higher rate of LBP, as well as of NP, to a lesser level, which may be associated indirectly with the COVID-19 pandemic, is demonstrated. The association is generated by SARS-CoV-2 but rather as a result of the use of lockdowns as tools in pandemic management.  Table 3. The weekly number of ED visits before and during con nements.   Notes, Chi-square tests or T tests were used to test group differences. Table 5. Multivariable logistic regression relating the visits of 1090 patients diagnosed with LBP in the ED.

Figure 1
A list of the diagnoses that were used to identify patients in the study group