Article Text

Download PDFPDF

SARS-COV-2 pandemic: establishing three risk levels for 19 Italian regions and two autonomous provinces
  1. Andrea Messori1,
  2. Manuela Angileri2,
  3. Marco Chiumente3,
  4. Laura Agnoletto4,
  5. Paolo Baldo5,
  6. Laura Bartoli1,
  7. Mario Cirino6,
  8. Vera Damuzzo7,
  9. Elisa Ferracane1,
  10. M Cecilia Giron8,
  11. Carolina Laudisio9,
  12. Daniele Mengato10,
  13. Angelo Claudio Palozzo11,
  14. Sabrina Trippoli12
  1. 1 HTA Unit, Regione Toscana, Firenze, Toscana, Italy
  2. 2 Hospital Pharmacy, Azienda Sanitaria di Firenze, Firenze, Toscana, Italy
  3. 3 Scientific Direction, SIFaCT, Milan, Italy
  4. 4 Hospital Pharmacy, Azienda ULSS 5 Polesana, Rovigo, Veneto, Italy
  5. 5 Hospital Pharmacy, Centro di Riferimento Oncologico, Aviano, Italy
  6. 6 Hospital Pharmacy, Azienda Sanitaria Universitaria Giuliano Isontina, Presidio Ospedale Maggiore, Trieste, Italy
  7. 7 Pharmaceutical Sciences, Universita degli Studi di Padova, Padova, Veneto, Italy
  8. 8 Department of Pharmaceutical Sciences, Universita degli Studi di Padova, Padova, Veneto, Italy
  9. 9 Pharmacy, Azienda Sanitaria Locale Salerno, Salerno, Campania, Italy
  10. 10 Hospital Pharmacy, Bolzano Hospital, Bolzano, Italy
  11. 11 Presidency, SIFaCT President, Italian Society for Clinical Pharmacy and Therapy, Milano, Milano, Italy
  12. 12 HTA Unit, Regional Health Service, Firenze, Italy
  1. Correspondence to Dr Andrea Messori, HTA Unit, Regione Toscana, 50129 Firenze, Toscana, Italy;{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

In Italy, there are 19 regions and two autonomous provinces, and each of these 21 jurisdictions has local responsibility for health care. As regards the COVID-19 pandemic, these 21 jurisdictions have been classified into three classes of risk (mild, moderate, and high) associated with three colours (green/yellow, orange, and red).

More than 20 indicators are routinely collected in these jurisdictions, and this information determines, every 2 weeks, the colour assigned and the consequent restrictions in everyday life. The replicability index (Rt according to the Italian notation) is the main parameter influencing this decision.

In November 2020, a number of health care professionals began a collaboration (based on a Facebook group named PIFO including 446 people) specifically aimed at discussing topics and controversies in the field of COVID-19.1 A total of 34 professionals participated in the experience described herein: 14 of these were invited to share their views through an ad hoc questionnaire.

Our purpose was to build a consensus on the following question: how can the 21 monitoring parameters established by national institutions be translated into the three risk classes? Reference 1 (Italian) reports in detail on the questions of our survey.

The criterion for agreeing the final consensus from individual responses was to select the most voted response for categorical variables, the median for numerical ones. The consensus statement was designed to, first, select a limited number of parameters (restricted to three) and to, consequently, determine the risk levels with the respective colours.

Three main parameters were identified in our survey: (1) replicability index; (2) incidence of new cases over 14 days (normalised to 100 000 people); and (3) percentage of intensive care unit (ICU) beds occupied by COVID-19 patients. These three parameters were prioritised in the above order.

In the decision model adopted by the group, the first parameter provisionally determines the colour assignment. Thereafter, the initial colour assignment can remain unchanged or, alternatively, can be worsened by one level if parameters (2) and/or (3) rise above a predetermined alarm threshold (table 1). This approach resembles the one currently adopted by the frameworks of ASCO and ESMO in evaluating innovative anti-cancer treatments.2–5 For each of the three parameters, our survey identified the cut-off values that separate individual risk levels. Rt was confirmed as the leading parameter to guide colour assignment, but the two co-primary parameters mentioned above were evaluated as well (table 1).

Table 1

Description of the algorithm that converts three parameters included in the national COVID monitoring framework into the risk levels associated with the three pre-determined colours (green/yellow, orange, and red)

In summary, our experience has shown that, in managing the COVID-19 epidemic, adopting a set of ranked risk levels according to a fully transparent approach is feasible. In our view, the decision process can better rely on three parameters rather than on one single parameter (Rt) (or, alternatively, 21 which are too many). In the perspective of further developments, one hypothesis could be to embed in the decision an outlook at 14 days (namely, the incidence projected at 14 days based on Rt) so that prospective projections rather than retrospective data determine the decision which is made every 2 weeks.

Ethics statements

Patient consent for publication


Part of this paper has been previously published as a preprint (see reference 1).



  • Twitter @mceciliagiron

  • Contributors All authors contributed equally to this paper.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; internally peer reviewed.