Confinement of the perimeter of basic health areas and incidence of COVID-19 in Madrid, Spain | BMC Public Health

This study assesses the impact of local perimeter lockdowns implemented during the second wave of COVID-19 in Madrid. This Autonomous Community has adopted a special model of public health measures based on perimeter closures by BHZ depending on the epidemiological situation over time. Our mathematical models showed no statistical difference in cumulative incidence for BHZs with and without perimeter closure.

In Spain, the universal wearing of a mask (indoor and outdoor) has been compulsory throughout the territory since May 19, 2020 [4]. On October 25, the second state of emergency began and additional public health measures were implemented, with a series of standards that varied according to the autonomous communities. These included curfews (set at 11:00 p.m. in Madrid for the period under review), limited seating capacity within restaurant premises, and gathering limitations (maximum of 6 people in outdoor restaurants and internal meetings, and 4 people in restaurants inside Madrid), and other general restrictions [15]. Prior to the state of emergency, perimeter confinements by BHZ were also activated, unlike the closures at the municipal level adopted in other autonomous communities (Galicia [16]Cantabria [17]among others).

We assessed whether BHZ perimeter confinements had a significant influence on the evolution of the epidemic curve by modeling them as explanatory covariates in several mathematical models. We found that the estimates provided by models that included perimeter confinements as an explanatory variable and those that did not were statistically very similar, indicating that perimeter confinements did not have a significant impact on the cumulative CI of 14 days.

Several factors limit the effectiveness of the BHZ locking system. For example, due to the high permeability between neighboring BHZs and the associated difficulty in assessing citizens’ compliance with the measure, it was not possible to determine whether the policy was implemented effectively. In addition, a low risk perception with respect to the COVID-19 pandemic was identified in the Spanish population during the period studied. [18]which could have led to a decline in compliance with the policy [19, 20].

While local mobility restrictions are effective in a theoretical modeling framework [21, 22]evidence suggests that an informed and coordinated approach is required for the effective implementation of such a response measure [23]. Being a rare policy, few studies on the effect of such selective confinements of units as small as BHZ are available. Fotan-Vela et al. [24] also analyze the case of BHZ closures in Madrid. Their analysis shows that the decline in the epidemic curve in Madrid began before the impact of perimeter closures could be reflected. Apart from Madrid, the only other context where a similar policy has been adopted is Chile, to our knowledge. Cuadrado et al. [25] and Li et al. [26] study local lockdowns active during the first wave of the COVID-19 pandemic in this country, obtaining respectively a reduction in the effective reproduction number (with a wide confidence interval, nevertheless), and a highly variable effectiveness of the policy ( depending on duration of intervention and ripple effect of neighboring areas).

Analysis limits

The average BHZ is an epidemiologically small unit, both in terms of population (22,750 inhabitants) and area (28 km2). For this reason, usual commonality methods for trend analysis are unlikely to reveal significant findings at the local BHZ level, lacking statistical significance. This is also the case for trend analysis on models that integrate information from all BHZs, due to the asynchronicity in the implementation of perimeter confinements between each of the BHZs. For the same reason, it is not expected that precise estimates will be obtained from models fitted to these data. We have therefore chosen to employ the current approach, sensitive to general trends in models that have been fitted differently, and focused on statistical evaluations rather than precise predictions. GAM models should capture a greater influence of the additional explanatory variables included than trend analysis models [12]and we incorporated higher significance by a cross-validation process without one out of 286 BHZ which involves choosing the best of 15 models at each step.

An additional confounding effect is due to the fact that perimeter confinements (and COVID-19 restrictions in general) were introduced after the fact. That is, restrictions are activated in response to the 14-day increase in IC, and thus there is a natural correlation between BHZs with a high IC and perimeter-confined BHZs. Again, an approach that does not focus on assessing the explicit and precise impact of these restrictions and rather on its statistical effect is therefore preferred, as misleading associations may otherwise be inferred.

Finally, the epidemiological threshold triggering closures changed during the study period. On September 21, weekly BHZ perimeter lockdowns were activated in BHZs where the 14-day cumulative incidence exceeded 1,000 cases per 100,000 population. This threshold was lowered to 750 cases on October 12, 500 cases on October 26 and 400 cases on November 23. [11]. We did not include the possible effect of this variation in our analysis, as we focused on the effect of actual perimeter lockdowns and not their dependence on the current epidemiological state.

Previous Kessler Topaz Meltzer & Checker, LLP: March 25, 2022
Next Sarah Raskin's systemic risk at the Federal Reserve