Pro-active monitoring and social interventions at community level mitigate the impact of coronavirus (COVID-19) epidemic on older adults' mortality in Italy: A retrospective cohort analysis.

BACKGROUND: The COVID-19 epidemic in Italy has severely affected people aged more than 80, especially socially isolated. Aim of this paper is to assess whether a social and health program reduced mortality associated to the epidemic.
METHODS: An observational retrospective cohort analysis of deaths recorded among >80 years in three Italian cities has been carried out to compare death rate of the general population and "Long Live the Elderly!" (LLE) program. Parametric and non-parametric tests have been performed to assess differences of means between the two populations. A multivariable analysis to assess the impact of covariates on weekly mortality has been carried out by setting up a linear mixed model.
RESULTS: The total number of services delivered to the LLE population (including phone calls and home visits) was 34,528, 1 every 20 day per person on average, one every 15 days during March and April. From January to April 2019, the same population received one service every 41 days on average, without differences between January-February and March-April. The January-April 2020 cumulative crude death rate was 34.8‰ (9,718 deaths out of 279,249 individuals; CI95%: 34.1-35.5) and 28.9‰ (166 deaths out of 5,727 individuals; CI95%:24.7-33.7) for the general population and the LLE sample respectively. The general population weekly death rate increased after the 11th calendar week that was not the case among the LLE program participants (p<0.001). The Standardized Mortality Ratio was 0.83; (CI95%: 0.71-0.97). Mortality adjusted for age, gender, COVID-19 weekly incidence and prevalence of people living in nursing homes was lower in the LLE program than in the general population (p<0.001).
CONCLUSIONS: LLE program is likely to limit mortality associated with COVID-19. Further studies are needed to establish whether it is due to the impact of social care that allows a better clients' adherence to the recommendations of physical distancing or to an improved surveillance of older adults that prevents negative outcomes associated with COVID-19.

1. Introduction

A disproportional rate of severe infection due to the COVID-19 epidemic was reported in Italian older adults, resulting in a very high age-specific mortality rate; in Italy, the population aged 80 years and older accounts for 7% of the total population, and more than 25% of the SARS-CoV-2 infections and about 60% of COVID-19-related deaths were reported in this population [1]. The experience of Italian older adults is like that of older adults in other European countries including Spain, the UK, France, Belgium, and Sweden where most deaths have been recorded in the population aged >80 years [2].

During the first months of 2020, the Italian provinces severely hit by the SARS-CoV-2 infection showed an increase in mortality of around 200% compared with that in the same period of 2019, and this figure mainly consisted of people aged >80 years [3]. Some reasons for the spread of infection in this population could be the high percentage of this population among the host of the Italian nursing homes that have been heavily affected by COVID-19, and the fragmentation of society that has led to many older adults living alone [4]. In fact, when one older person/an older couple lives alone, it could be very difficult for him/her/them to adhere to behavioural advice like the ones recommended during the COVID-19 epidemic, (e.g., stay at home) if there is no one to offer help. According to the Italian Institute of Statistics (ISTAT), more than 25% of Italian older adults claim that they cannot count on anyone for help in case of need [5]. Moreover, nursing home hosts and their carers showed many difficulties in practising social (or physical) distancing since many hosts need personal care several times per day. The lack of Individual protective devices, especially during the first phase of the epidemic spread, increased the risk of SARS-CoV-2 transmission.

Even if social fragmentation is associated with a high incidence of COVID-19 infection among Italian older adults, then social connectedness should be associated with a low mortality of COVID-19. In fact, higher mortality rate among older adults has been associated to smaller household size and higher Long-Term Care Facility (LTCF) bed rate [4]. The aim of this paper is to compare the mortality from COVID-19 in the general population and in a sample of older adults followed-up by the ‘Long Live the Elderly!’ (LLE) program, which is devoted to counteracting social isolation.

2. Materials and methods

This paper analyzed the trends of mortality from 1st January to 30th April 2020, and the potential determinants in the population aged above 80 years in three Italian cities: Rome, Genoa, and Novara. These three cities were chosen because they are the only ones where the LLE program has been running, and data about mortality in the general population by age group were updated at the end of April 2020, to include the first months in which the epidemic spread in Italy. Moreover, the three cities showed different levels of epidemic spread, high, medium, and low levels in Novara, Genoa, and Rome, respectively [3], allowing comparisons in different contexts.

2.1 Study design and setting

The study is a retrospective cohort analysis, based on an ecological approach, carried out in three Italian cities, Genoa, Rome, and Novara.

2.2 Data sources

A retrospective cohort analysis has been carried out on data from two sources: the LLE electronic record database and the ISTAT database on mortality, which is freely accessible online [6].

2.3 Inclusion criteria

Data about mortality and services provided to the LLE program participants, gathered on a routine basis during the study period have been included in the analysis.

2.4 Interventions and procedures

The LLE program started in Rome (Italy) in 2004 after the 2003 heat wave [7] to meet the care needs of the older population, particularly those who had limited relationships related to age transition. It is a community-based pro-active monitoring program based on periodical phone calls and ad hoc intervention (including home visits in case of need) to meet clients’ needs for care [8]. The program is implemented in agreement with the municipality, which participates especially by promoting awareness campaigns. The population included in the program has to meet two main criteria, ie living in an urban area and belonging to low socio-economic status household. All the people aged>80 living in the selected urban areas were contacted by letter and by phone. To be included they had to consent to the use of personal information for the program aims. The acceptance rate was higher than 90%, with minor differences between areas.

The frequency of phone calls is established according to the level of bio-psycho-social frailty (from one call every three months to one every two weeks). Bio-psycho-social frailty is assessed at the first contact by means of validated questionnaires and re-assessed once a year. The main objective of the program is to reach all those aged >80 years with a proactive approach and to offer them personalized and integrated social and health services according to their needs and wishes, as highlighted by the routine assessments. The aim of the LLE program is to support the client to have access to the most adequate package of services, either provided directly by the LLE program or by other available providers. Referral does not mean that the client is just addressed to a service; a continuous interaction between the provider and the LLE program is warranted. The program aims also to build a network of social relationships for the most isolated individuals to help them dealing with possible negative events, such as the death of the spouse, the development of a disease, or the worsening of physical functions. In these situations, tackling social isolation is often crucial in order to continue to live at home instead of being referred to a LTCF [8, 9].

In addition to routine activities, special interventions are activated during emergencies, such as heat waves or epidemics. All the program participants are called by phone once every two weeks (or more in case of need) to assess their physical conditions, the need for food or medicines, or to assist them in bureaucratic tasks or any other need. The aim of these actions is to allow the clients to keep the required physical distancing while avoiding social isolation. The emergency protocol started to be implemented since the first week of March until the last week of April. For this study, all the deaths that occurred from 1st January to 30th April 2020 have been recorded. In case of non-traceable individuals, news about them was gathered from their relatives or neighbours. A second wave of phone calls was performed in September to gather missing information.

2.5 Statistical analysis

Weekly death rates and cumulative standardized death rates were calculated. The Standardized Mortality Ratio (SMR) has been calculated according to the following formula: Number of Observed Deaths / Number of Expected Deaths. The observed deaths were the ones recorded among the LLE population, while the expected deaths were calculated applying the general population age-specific death rates to the LLE population (indirect standardization). To examine predictors of mortality, initial univariate analyses were run to identify characteristics associated with mortality at each time point. Possible determinants of mortality were assessed using Student’s t-tests and Mann-Whitney U-tests for continuous variables.

For the three cities and two groups (general population and LLE population) death rates for each one of the 18 time-points were calculated. The temporal trends of the obtained death rates were then modeled using multivariable generalized regression analyses based on Repeated Measures General Linear Models (RMGLM) in order to consider the temporal correlation of the weekly incidence. In fact, mortality among people aged more than 80 could be affected by the potential mortality displacement effect due to the pre-Covid-19 epidemic (ie, a cold wave). Due to the increased prevalence of frailty in this aged population, an increase of pre-Covid-19 epidemic mortality could affect the mortality rate in the following weeks. This is why the developed RMGLM models considered the weekly incidences as within-subjects factors, the groups (general vs. LLE population) as between-subjects factors, and the mentioned variables as covariates. Therefore, the models were adjusted in each group for the interaction of baseline characteristics (age, gender, and population size). Then the weekly incidence of Covid-19 and the LTCF bed rate by city has been included in the multivariable analysis as independent ordinal variables in the model and tested the significance of each covariate as well as of the interaction terms. Interaction terms with p < 0.10 were dropped from the model. A two-sided p<0.05 was considered as statistically significant, and 95% Confidence Intervals (CIs) were reported. Analyses were conducted using SPSS, version 25.

2.6 Ethics statement

Regarding the LLE database, a consent form for using anonymised aggregated data for analysis was signed by the older adults when they agree to participate in the LLE program. For the present analysis only data about mortality and provided services (gathered on a routine basis) have been used.

3. Results

In Table 1, the population is described by age and sex according to the intervention being evaluated. The differences in terms of age and gender between the two populations are not so relevant even if they must be taken into consideration in the analysis because of their impact on COVID-19 mortality [1].

Table 1

Population by age groups and sex.

		Age groups (years)			Males (%)	Females %	Individuals living in Nursing Homes (%)
		80–90 (%)	≥90 (%)	Total
Long Live the Elderly participants	Genova	357 (78.29)	99 (21.71)	456	34.6	65.4	6.9
	Novara	753 (83.85)	145(16.15)	898	33.7	66.3	1.6
	Roma	3,242 (80.25)	845 (20.67)	4,087	35.0	65.0	0.2
	Total	4,352 (79.99)	1,089 (20.01)	5,441	34.8	65.2	1.0*
Control groups	Genova	47,380 (80.18)	11,719 (19.82)	59,099	34.7	65.3	2.7
	Novara	6,915 (81.45)	1,575 (18.55)	8,490	35.8	64.2	4.1
	Roma	174,978 (82.45)	37,268 (17.55)	212,246	36.1	63.9	1.3
	Total	229,273 (81.94)	50,562 (18.06)	279,835	35.8	64.2	1.7*

The January-April 2020 cumulative crude death rate was 34.8‰ (9,718 deaths out of 279,249 individuals; CI95%:34.1–35.5) and 28.9‰ (166 deaths out of 5,727 individuals; CI95%:24.7–33.7) for the general population and the LLE sample, respectively. The LLE mean weekly mortality rate by month is lower than the general population one, except for February; however, the difference is statistically significant only in April (p = 0.006), when also the variance is similar (Levene’s test and Moses test not statistically significant). The weekly mortality increase observed during March and April among the general population sample is not confirmed among the older adults participating to the LLE program (Table 2).

Table 2

January-April average weekly mortality rate (per 1,000) by month, according to the participation to the LLE program.

		Mean	SD	t-test	U-Mann Withney test	Levene’s test	Moses test
January	Gen Pop	1.755	0.221	NS	NS	p<0.001	p>0.001
	LLE	1.599	1.353
February	Gen Pop	1.762	0.313	NS	NS	p = 0.023	NS
	LLE	2.089	1.232
March	Gen Pop	2.811	1.168	NS	NS	NS	NS
	LLE	1.973	1.534
April	Gen Pop	2.944	1.077	p = 0.006	p = 0.007	NS	NS
	LLE	1.670	1.261
January-April	Gen Pop	2.322	0.974	p = 0.025	p = 0.026	p = 0.042	p>0.001
	LLE	1.811	1.323

The death rate has been standardized (S1 Table) by age class (≤90 and >90), gender, and city to consider the differences in the three diverse epidemics. The Standardized Mortality Ratio (SMR) was 0.88 (CI95%: 0.78–1.03) for the whole period, and 0.78 (CI95%: 0.61–0.98) from March 3rd to April 26th, when the highest mortality has been recorded.

During the first four months of 2020, in the three cities chosen for this analysis, under the LLE program, 29,148 phone calls, 1,117 drug and food deliveries at home, and 2,379 home visits were made. The total number of services (also including received phone calls) was 34,528, 1 every 20 days per person on average, increased to 1 every 15 days during March and April. From January to April 2019 the same population received one service every 41 days on average, without differences between January-February and March-April.

Fig 1 shows the trend of the two populations’ weekly mortality rate: it is worth of note the increase of mortality from week 10 to week 17 in the general population compared with the lack of increase in the same weeks observed in the LLE sample. The multivariate linear mixed model, weighted for the size of the population, and adjusted for weekly mortality rate, percentage of individuals living in nursing homes, age, and percentage of males, confirms the impact of the “Long Live the Elderly!” intervention on the mortality (Table 3, p = 0.001). The inclusion of the Intensive Care Unit (ICU) regional bed rate―a proxy of the regional health system capacity to deal with the enormous increase of ICU demand due to the COVID-19 pandemic―did not modify the results of the model.

Fig 1

Average (Rome, Genoa, Novara) weekly mortality rate (rate per 1,000 inh. and CI95%) for the general population and LLE subsamples.

Table 3

Mixed linear model–fixed effect and covariances estimation*.

FIXED EFFECT	Estimation	SE	p	CI95%
				Lower Limit	Upper limit
Intercept	43.292	6.896	<0.001	29.697	56.887
Program (categorical, LLE = 0; GPop = 1)	0.390	0.157	0.013	0.082	0.699
COVARIANCE ESTIMATION		Wald-Z	p	CI95%
				Lower Limit	Upper Limit
Weekly incidence (Repeated Measures)		17.652	<0.001	1.075	1.342
Individuals living in nursing homes (%) + males (%) + age>90(%) + calendar week (interaction)		1.257	0.209	0.072	1.630

• “Program” variable is included in the model as a fixed effect variable while the “weekly mortality rate”, “percentage of individuals living in nursing homes”, “age” and “gender”, as random co-variables.

4. Discussion

The paper shows the impact of COVID-19 epidemic on the older adult population in three Italian cities with different incidence of COVID-19 related mortality, according to the “exposure” to a social intervention.

The SMR synthesizes the reduction of mortality rate among the LLE participants compared with the mortality rate of the general population. This reduction was more evident from March 3rd to April 26th, and in the cities of Novara and Genoa where the epidemic was more severe than in Rome. The triple standardization (gender, age, and city) takes into consideration the different stage of epidemic in the three cities. It is worthy of note that the mortality increased by 83%, 100% and 5.4% in Novara, Genoa, and Rome, respectively [10]. The LLE program showed a positive impact on mortality mainly in cities where the negative impact of COVID-19 on mortality was more prominent.

The linear mixed model analyzed the risk of mortality growth in the control population compared with the LLE one, confirming the impact of the LLE program on the mortality containment.

There are various explanations for this phenomenon: the first one is the limited number of individuals living in nursing homes in the LLE population, especially in the Rome sub-sample. Of note, avoiding admission to a nursing home is one of the main goals of the LLE program, so it should be considered a result of the program itself. The nursing home admission rate among the citizens who participated in the program in Rome during 2019 was 0.77 per 1,000 individuals, while that for the general population in Rome was 0.97 per 1,000 individuals, with a difference of about 20% (data not published). This is probably due to the indirect effect of increasing social connectedness, promoted by the LLE program, since living alone has been associated to a more frequent recourse to LTCF [11, 12] as well as the lack of social ties [13, 14].

However, the trend of epidemic is not leaded only by LTCFs bed rate that is not likely the only factor explaining the mortality increase. The role played by the LLE program could be also associated to a reduction in hospital admission that had been already shown [7, 15]. Of note, the hospital environment was a place of transmission of SARS-CoV-2 infection, especially at the beginning of the epidemic in the last weeks of February, before the implementation of effective preventive measures on large scale in the hospital settings. The potential reduction in hospital admissions could have played a protective role in the spread of the infection among the LLE program clients.

Social isolation is associated per se to an increase of mortality [1-18] as well as pre-frailty [19]. In general, social isolation and frailty are strongly associated [20], especially in men [21], while frailty showed to be associated to death in hospitalized COVID-19 patients [22]. Hence, a third protective factor could be the proactive monitoring and the following interventions. The LLE considers the COVID-19 epidemic an emergency due to the special attention needed by the older population. The emergency protocol was activated from the beginning of March to the end of April; it includes a phone call to every client twice a month (or more when needed) to check for any requirement for practical need (food, drugs, bill payments, etc.) to allow the clients to stay safe at home as much as possible. The services delivered during the study period (home visits, food/drug home delivering) increased if compared to the same period of 2019. This shows the efforts of the LLE program to meet the increased demand for care. It is possible that the mortality containment among the program participants during the first months of the COVID 19 crisis might be generated by an increase in social capital because of the program activity along the years, as well as by the protective effect of the intervention that facilitated the clients’ compliance to the general prescriptions to prevent SARS-CoV-2 contagion. For example, having someone who can control the adherence to the drug schedule and who drives the attention of the GP to a symptom might have avoided hospital admissions or access to emergency rooms, which are sometimes the starting point of dangerous vicious circles. The containment of mortality observed in the LLE population during the COVID-19 emergency is consistent with the one observed during the heat waves. The LLE emergency protocol is the same and limited the increase in mortality observed in the neighboring urban zones exposed to the same environmental stress by half [9]. The LLE program acts as a shock absorber, reducing the impact of emergencies such as the ones we analyzed. Of course, the containment of mortality in a population aged >80 years is very difficult to be observed under normal life conditions since in Italy the median age at death is 81 and 85 years for men and women, respectively. The protective action of the program was evident because of the rise of stress factors provoking an unexpected increase in mortality.

This paper presents several limitations. Firstly, the two population are not fully comparable in terms of age and gender distribution. The standardization procedure aims at overcoming the differences to allow the comparison. A limited percentage of the sample of LLE clients (<5%) was not reached by the researchers, but news about their health status have been obtained from neighbors: this might have led to a bias due to inaccurate information. However, also some information stemming from the municipality could be delayed as usually happen when dealing with official information on deaths (delay of notification). Other limitations are related to the lack of information about factors that might influence mortality, such as adherence to governmental public health measures (wearing masks, keeping distance, stay at home, etc.) that are well-known interventions to reduce mortality by decreasing transmission (Rt) [23]. Nevertheless, there are no reasons to think that these recommendations have been better accomplished by the LLE participants than the general population, except for the role played by the LLE program itself as a reinforcing factor. Therefore, we had to assume that these differences were not relevant to invalidate the results especially due to their diffusion that is likely to be similar in both populations. Further potential confounders that could bias the association between the LLE program and the containment of COVID-19 related mortality are listed in Table 4. For all these limitations, further analytical studies are warranted to confirm or question our results.

Table 4

Potential confounders.

1. Disparities among population level mitigation measures

2. Disparities among regional health systems capacities

3. Differences in terms of frequency of phone calls from friends or neighbours between the LLE and the general population

4. Differences in terms of frequency of visits from friends or neighbours between the LLE and the general population

5. Differences in terms of duration of the above visits/calls

6. Discrepancies among action performed by the visitors (bring food or medication, drive the participants to medical appointments and/or to pharmacies to collect medications, drive respondents to shops

7. Differences in illnesses among the two populations

8. Differences in drugs assumption among the two populations

9. Different causes of death among the two populations

10. Different impact of all these factors on nursing home residents included in the study

5. Conclusions

The analysis of the standardized death rates of people aged >80 years shows the increase in death rates observed in the general population, which is different from the one observed among the clients of the LLE program. It is likely that the intense activity of the LLE program, focused on supporting social connectedness, limited the negative impacts of the COVID-19 epidemic among older adults, especially when physical distancing is mandated. However, this conclusion should consider the limitations of the study, which are mainly related to the study design. Further studies are needed to highlight the relation between social care and mortality among people aged>80 during a pandemic crisis. In case a containment of mortality will be confirmed it will be crucial to establish whether it is due to the impact of social care that allows a better clients’ adherence to the recommendations of physical distancing or to an improved surveillance of older adults that prevents negative outcomes associated with COVID-19. Since the COVID-19 crisis will last, even if attenuated by vaccinations, the question on how to protect the older adults without increasing their social isolation―a well-known risk factor for mortality―is still open. However, the lesson learnt from the COVID-19 crisis is that each crisis calls for an increase in the attention and care provided to the frail population by community care services. The integration of health and social care at the community level is a crucial step for improving the quality of the intervention.

Death rate standardization (indirect procedure).

(XLSX)

Click here for additional data file.

11 Aug 2021

PONE-D-21-11121

Pro-active monitoring and social interventions at community level mitigate the impact of Coronavirus (COVID-19) epidemic on older adults’ mortality in Italy: a retrospective cohort analysis

PLOS ONE

Dear Dr. liotta,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Both reviewers feel that the study addresses an important question, and should eventually be published. However, both of them recommended that their concerns be addressed before this happens. In particular, Reviewer 1 is concerned about confounding factors that also affect COVID-19 mortality amongst seniors. These include vaccination of seniors, which reduces mortality, and the healthcare capacity being exceeded, which increases mortality. Please think about how these confounding factors can be eliminated from the data, before discussing the effectiveness of non-pharmaceutical intervention measures. Reviewer 2, on the other hand, asks for more details on the data collected.

Please submit your revised manuscript by Sep 25 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see:  http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at  https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols .

We look forward to receiving your revised manuscript.

Kind regards,

Siew Ann Cheong, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1.Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: No

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors aimed to assess if a social and health program reduced mortality associated to the COVID-19 epidemic in Italy, specifically to the older aged groups. The paper has important significance to the scientific readership in the quest to understand mitigation efforts that could reduce negative impacts to vulnerable groups. While the manuscript is well presented and written, I would like to deliver few comments that I feel the authors need to address prior to consideration into publication.

1. The intervention program (LLE) has been shown to have a positive effect to reduce mortality among old aged. Could this effect be actually confounded by national vaccination programs that have initially targeted vulnerable groups to reduce the impact on severe COVID-19 such as hospitalizations and mortalities? If this association was not tested, I suggest that the authors mention them in the limitations part or succinctly describe them briefly as potential public health implications. One suggestion is that, the infectiousness (Rt) overtime with population level mitigation measures could have, at the overall stage been reduced, due to government mitigation measures, and yet massive vaccinations targeting vulnerable groups to reduce complications or to achieve the somewhat “herd immunity” could have been likely reducing mortality rates. These interactions or confounders needs to be taken into consideration. The following literature are worthy to be cited:

• Ganasegeran, K.; Ch’ng, A.S.H.; Looi, I. What Is the Estimated COVID-19 Reproduction Number and the Proportion of the Population That Needs to Be Immunized to Achieve Herd Immunity in Malaysia? A Mathematical Epidemiology Synthesis. COVID 2021, 1, 13-19. https://doi.org/10.3390/covid1010003

• Kwok KO, McNeil EB, Tsoi MTF, Wei VWI, Wong SYS, Tang JWT. Will achieving herd immunity be a road to success to end the COVID-19 pandemic? J Infect. 2021 Jun 10:S0163-4453(21)00287-5. doi: 10.1016/j.jinf.2021.06.007.

2. The next potential confounder of deaths rates could be attributed to the fact that health systems are overwhelmed, lack of testing capacities and test sensitiveness. These should somewhat be mentioned in the introduction as potential factors to cause mortalities among the elderly, especially to those staying alone or within homes of the elderly that has the potential to yield “clusters.”

3. A crucial part that needs further elaboration here is that the description of “LLE.” The intervention was an adopted one aimed for a different purpose, “older population that faced lack of relationships associated with age transition.” It is unclear how authors could convince readers that the same program be adopted within a public health crisis such as during pandemics. One such issue I could postulate here is psychological repercussions and social isolation among the elderly that may affect some unprecedented issues, but how about the threat to the infection itself?

4. Can the authors subsection the methodology part as: study design and setting, study participants, inclusion and exclusion criteria, data sources, interventions and procedures, statistical analysis, ethics statement. This would give a more systematic reporting of the methodology part for clear readership.

5. What is CL95%? Do the authors point to confidence interval (CI)? I suggest to change to CI or define them at the beginning before the abbreviation being used.

6. Table 1 – Please include the proportion of females

7. Include the formula for SMR

8. Why choose a non-parametric approach to most of your analysis? Was your continuous data skewed?

9. The comparator was the general population, hence how could we know if LLE, specifically designed for the older aged will be effective, unless stratified by age in the comparator as well. Not very clear to me in the manuscript, but I appreciate if authors could explain.

10. Table 3 – your standard error value (SE) is quite huge. Please check.

Overall, this is a good study and worthy of publication.

Reviewer #2: This is a study of a serious problem, the absence of social and family support for elderly persons living alone. The severity of the psychosocial component was recognised by the creation of a new Ministry in England, the Ministry for Loneliness in Seniors.

Introduction, research plan and results:

In a retrospective cohort study like this the key problem is identifying which are the known confounders and then collecting data at low risk of bias to comprehensively examine these. Missing data are:

Number and content of contacts other than phone calls from the programme:

1. How were individuals selected or self-selected for the programme?

2. What were the attrition rates? What were the causes of attrition?

3. What other inputs did the participants receive?

4. How many phone calls from friends or neighbours?

5. How many phone visits from friends or neighbours?

6. How many phone calls from friends or neighbours?

7. How many phone visits from friends or neighbours?

8. Frequency and duration of the above visits/calls?

9. Did visitors bring food or medication?

10. Did visitors drive the participants to medical appointments and/or to pharmacies to collect medications?

11. Did visitors drive respondents to go shopping?

12. Did visitors perform needed house repairs? (this is a problem that sometimes causes residents to move

Illness and comorbidities:

1. What were their illnesses and comorbidities?

2. Did they have infleunza, pneumococcal or COVID-19 or other infections?

3. What medications did they take? Serious adverse effects of medications?

4. Cause of death?

The relevant items above need to be separately reported for those in nursing homes and living at home.

The proportion in nursing homes was 1.7% in the control and 1% in the intervention group (very low compared to other western countries). Did you control for this?

I mention all these considerations:

1. to ask if you have data bout them

2. To suggest that you can draw no causative conclusions from a retrospective cohort study with so many missing known confounders.

May I suggest you rewrite your study to provide as much data as you can for subsequent researchers and draw no causative conclusions. (can a phone call every three months be reasonably expected to have any effect?)

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Roger E. Thomas

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

22 Sep 2021

All the answers are included also in the rebuttal letter.

Review Comments to the Author

Reviewer #1: The authors aimed to assess if a social and health program reduced mortality associated to the COVID-19 epidemic in Italy, specifically to the older aged groups. The paper has important significance to the scientific readership in the quest to understand mitigation efforts that could reduce negative impacts to vulnerable groups. While the manuscript is well presented and written, I would like to deliver few comments that I feel the authors need to address prior to consideration into publication.

1. The intervention program (LLE) has been shown to have a positive effect to reduce mortality among old aged. Could this effect be actually confounded by national vaccination programs that have initially targeted vulnerable groups to reduce the impact on severe COVID-19 such as hospitalizations and mortalities? If this association was not tested, I suggest that the authors mention them in the limitations part or succinctly describe them briefly as potential public health implications. One suggestion is that, the infectiousness (Rt) overtime with population level mitigation measures could have, at the overall stage been reduced, due to government mitigation measures, and yet massive vaccinations targeting vulnerable groups to reduce complications or to achieve the somewhat “herd immunity” could have been likely reducing mortality rates. These interactions or confounders needs to be taken into consideration. The following literature are worthy to be cited:

• Ganasegeran, K.; Ch’ng, A.S.H.; Looi, I. What Is the Estimated COVID-19 Reproduction Number and the Proportion of the Population That Needs to Be Immunized to Achieve Herd Immunity in Malaysia? A Mathematical Epidemiology Synthesis. COVID 2021, 1, 13-19. https://doi.org/10.3390/covid1010003

• Kwok KO, McNeil EB, Tsoi MTF, Wei VWI, Wong SYS, Tang JWT. Will achieving herd immunity be a road to success to end the COVID-19 pandemic? J Infect. 2021 Jun 10:S0163-4453(21)00287-5. doi: 10.1016/j.jinf.2021.06.007.

In Italy, as well as in other EU countries, the vaccination program started on December 29, 2020, whereas data is gathered from January 1st to April 31st, 2020 (i.e. before the introduction of vaccinations). Therefore, vaccination cannot be considered a possible confounding factor. Moreover, since we are presenting a study based on an ecological approach considering the comparison between population exposed to the same mitigation interventions (wearing mask, keep distances, stay at home, etc..) it is unlikely that selection biases such to affect the results might have occurred. Nevertheless, as suggested by the reviewer, we reported the latter consideration as a study limitation

2. The next potential confounder of deaths rates could be attributed to the fact that health systems are overwhelmed, lack of testing capacities and test sensitiveness. These should somewhat be mentioned in the introduction as potential factors to cause mortalities among the elderly, especially to those staying alone or within homes of the elderly that has the potential to yield “clusters.”

This is another important point raised by the referee, and we are grateful for that. We actually presented data referring to three Italian administrative regions (Genoa–Liguria region, Novara-Piedmont, and Rome-Lazio). It is true there are differences in the capacity of the Regional Acute Health Care Systems among the Italian administrative regions especially in relation to different work loads. We tried to test the model introducing the regional Intensive Care Unit bed rate as a proxy of the capacity of the health systems to deal with an increase in the demand for acute intensive care like the one observed during the first phase of the epidemic in Italy. Result of the analysis did not substantially change, and the statistically significant result of the program was confirmed. So, we did not change the results but we presented the issue raised by the referee as one of the limitations of our study.

3. A crucial part that needs further elaboration here is that the description of “LLE.” The intervention was an adopted one aimed for a different purpose, “older population that faced lack of relationships associated with age transition.” It is unclear how authors could convince readers that the same program be adopted within a public health crisis such as during pandemics. One such issue I could postulate here is psychological repercussions and social isolation among the elderly that may affect some unprecedented issues, but how about the threat to the infection itself?

The reviewer is right. A more detailed description of the LLE intervention program is needed to appreciate the contribution of this program in this specific emergency. LLE was developed several years ago to deal with one of the most frequent crises faced by public health services in many Western countries for the last 20 years, ie the increase of mortality due to heat waves. During summer 2003 about 70,000 unexpected deaths have been recorded in Europe due to the heat waves. More than 90% of the deaths affected people aged more than 75, particularly when living alone. Something similar, even if not in the same size, happened during some of the following years. In the last five years before the COVID crisis (2015-2019) the LLE program showed the capacity of reducing mortality. Such a reduction was associated to social interventions aimed at supporting the older adults through a network of stable and enduring relationships.

The hypothesis on which the present study is based is that the negative consequences of the COVID crisis, included mortality, are to some extent due to the social isolation of many older adults. Isolation that was even worsened by the need to implement social distancing as a prevention measure against SARS-CoV-2 transmission. In other words, if you are alone, without significant relationships (as it happens to an increasing segment of older adults in Western countries) you are forced to leave your house for shopping or for getting drugs, exposing yourself to risk of transmission. If you have someone who can do these tasks for you, decreases such risky behaviour. Moreover, reduced contacts with your GP, as it happened during the crisis, can worsen your physical conditions, provoking hospitalizations and deaths even if not directly related to COVID-19. Of course, to increase its effectiveness it is much better if such a program is adopted as a standard of care all along the whole year than just during emergency, Nevertheless, many Italian municipalities launched similar programs to support older adults during the emergency because of the alarming number of older adults confined at home without any help.

All these considerations have been now synthesized and presented in the discussion. We believe that the remarks of the reviewer helped us to better describe and discuss the reasons of the impact of the intervention.

4. Can the authors subsection the methodology part as: study design and setting, study participants, inclusion and exclusion criteria, data sources, interventions and procedures, statistical analysis, ethics statement. This would give a more systematic reporting of the methodology part for clear readership.

The methods section has been rearranged according to the referee’s suggestions

5. What is CL95%? Do the authors point to confidence interval (CI)? I suggest to change to CI or define them at the beginning before the abbreviation being used.

Thank you for highlighting the typo. CL has been now corrected in CI

6. Table 1 – Please include the proportion of females

Proportion of females has been added to the table

7. Include the formula for SMR

Formula for SMR has been added to the Supplementary Table 1

8. Why choose a non-parametric approach to most of your analysis? Was your continuous data skewed?

The present study is an ecological one; several analysed data have not been collected by us but were retrieved from different published sources. Considering the whole Italian population of older adults, our sample was relatively small and skewed in the direction of cities (rather than countryside). In some cases, outliers were present. These outliers could not be legitimately removed as they were a genuine part of the distribution for the study area. For all these reasons we chose the more conservative nonparametric approach.

9. The comparator was the general population, hence how could we know if LLE, specifically designed for the older aged will be effective, unless stratified by age in the comparator as well. Not very clear to me in the manuscript, but I appreciate if authors could explain.

The comparator was always the general population aged > 80. The reviewer is right, this was not clear. We therefore added a clarification sentence in the paper.

10. Table 3 – your standard error value (SE) is quite huge. Please check.

Thank you, we double checked the SEs. They are correct, even if they are large due to the factors highlighted under bullet point 8.

Overall, this is a good study and worthy of publication.

Thank you very much for your very useful contribution

Reviewer #2: This is a study of a serious problem, the absence of social and family support for elderly persons living alone. The severity of the psychosocial component was recognised by the creation of a new Ministry in England, the Ministry for Loneliness in Seniors.

Introduction, research plan and results:

In a retrospective cohort study like this the key problem is identifying which are the known confounders and then collecting data at low risk of bias to comprehensively examine these. Missing data are:

Number and content of contacts other than phone calls from the programme:

In the manuscript we briefly presented the intervention delivered by the LLE program during the study period:

“During the first four months of 2020, in the three cities chosen for this analysis, under the LLE program, 29,148 phone calls, 1,117 drug and food deliveries at home, and 2,379 home visits were made. The total number of services (including also phone calls received) was 34,528, 1 every 20 days per person on average, increased to 1 every 15 days during March and April. From January to April 2019 the same population received one service every 41 days on average, without differences between January-February and March-April”.

However, the reviewer is right in underlining that this was not taken up in the discussion section. So we considered to do that in order to make more understandable the reasons why we associated the LLE program with a possible containment of mortality. The referee pointed out a lack of a clear link between the LLE program and mortality reduction. We acknowledge that due to the characteristics of such an ecological study, some information to prove the link is still lacking. So, we decided to attenuate our conclusion on these aspects.

1. How were individuals selected or self-selected for the programme?

The programme contacts all people aged>80 leaving in the areas selected for the intervention. The main selection criteria was living in a urban area with low socio-economic status. This is relevant because it is known that low socio-economic status is a risk factor for mortality. In this analysis a proxy of socio-economic status - the mean value of owned houses - was included; it did not show any statistically significant association with COVID 19 mortality. In any case, information about selection of urban areas and population has been now added to the methods section

2. What were the attrition rates? What were the causes of attrition?

The attrition rate during the four months of the study was less than 5%, even if information related to survival have been gathered also in this case, as reported in the limitation paragraph at the end of the paper. Main cause of attrition was moving to another location. The program acceptance rate is on average higher than 90%, and during the COVID crisis some clients who initially refused to be included into the program turned later to it because of the difficulties related to the COVID crisis.

Regarding some of the questions from 3 to 12, due to the characteristics of this study we cannot provide information about aspects not related to the LLE program operators or program volunteers. We have information about the consistency of the social network of relationship of the people included in the LLE program, as these pieces of information were gathered through the periodical assessment of frailty. But we cannot compare this information with the one of the general population (not included in the LLE program), as in this case these parameters were not available.

However, we have no evidence that allows us to state that the LLE program clients received a different input by friends/neighbours than the general population (apart from the program itself). We assume that on a similar base of social contacts the LLE program represents a booster with a relevant implication in the quality of life, resulting eventually also in a containment of mortality.

3. What other inputs did the participants receive?

4. How many phone calls from friends or neighbours?

5. How many phone visits from friends or neighbours?

6. How many phone calls from friends or neighbours?

7. How many phone visits from friends or neighbours?

8. Frequency and duration of the above visits/calls?

9. Did visitors bring food or medication?

10. Did visitors drive the participants to medical appointments and/or to pharmacies to collect medications?

11. Did visitors drive respondents to go shopping?

12. Did visitors perform needed house repairs? (this is a problem that sometimes causes residents to move

Illness and comorbidities:

1. What were their illnesses and comorbidities?

2. Did they have influenzas, pneumococcal or COVID-19 or other infections?

3. What medications did they take? Serious adverse effects of medications?

4. Cause of death?

Due to the study design this information is available for the LLE clients, but not for the general population; so we decided to exclude it from the analysis. However, we do not have reasons to think there were relevant differences between the LLE population and the general population about these issues.

The relevant items above need to be separately reported for those in nursing homes and living at home.

Unfortunately, due to the study design this information is not available

The proportion in nursing homes was 1.7% in the control and 1% in the intervention group (very low compared to other western countries). Did you control for this?

The percentage of people leaving in nursing homes has been included into the model as a co-factor

I mention all these considerations:

1. to ask if you have data bout them

2. To suggest that you can draw no causative conclusions from a retrospective cohort study with so many missing known confounders.

May I suggest you rewrite your study to provide as much data as you can for subsequent researchers and draw no causative conclusions. (can a phone call every three months be reasonably expected to have any effect?)

According to the suggestions of the reviewer, we tried now to include in the manuscript as much information as possible. However, as pointed out previously many pieces of information were not available. We agree with the reviewer that some conclusions must be softened, and we did it in this revised version.

Nevertheless, we would like to point out that the LLE program is not just based on a phone call every three months; it is rather a complex community care model which aims to increase the social capital of the population aged>80, representing the potential vector of a number of preventive interventions (like – just to mention the last example- the COVID-19 vaccination). Social isolation is a well-known risk factor for mortality, especially among the older adults population. It is possible that an intervention that counteracts social isolation can contain mortality in this population. Our data point in this direction, even if – of course – due to the many limitations of this study they cannot be considered conclusive.

Submitted filename: response to reviewers.docx

Click here for additional data file.

11 Oct 2021

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Siew Ann Cheong, Ph.D.

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Thank you for your revision. While I agree to almost all author responses and revisions, there are still two minor comments that I think authors need to address and provide appropriate conceptual evidence before the manuscript qualifies for publication.

The authors have mentioned in response to comment 1 that mitigation and containment strategies by the government could have somewhat influenced the number of cases and mortality. Authors responded in the last sentence “Nevertheless, as suggested by the reviewer, we reported the latter consideration as a study limitation.”

Authors added the following sentence in the limitations part “adherence to governmental public health measures (wearing masks, keeping distance, stay at home, etc.; diseases affecting the studied population or the causes of death).” These measures were known to affect the infectiousness (Rt) overtime, thereby determining cases escalation or decrease over time, yet may cause the mortality rates to increase or decrease overtime if health systems were overwhelmed or cases were undetected. As authors agreed to the suggestion and noted in the limitation, it needs to be justified with evidence as conceptualized in the reviewer’s comments. The following evidence needs to be cited to in accordance to authors acceptance of the proposed work and need to be corroborated based on the above argument:

Ganasegeran, K.; Ch’ng, A.S.H.; Looi, I. What Is the Estimated COVID-19 Reproduction Number and the Proportion of the Population That Needs to Be Immunized to Achieve Herd Immunity in Malaysia? A Mathematical Epidemiology Synthesis. COVID 2021, 1, 13-19. https://doi.org/10.3390/covid1010003

Second, the SMR formula needs to be incorporated under methods part as an operational definition.

Once these revisions have been made, I have no further objections for the paper to be published.

Reviewer #2: Thank you for your detailed replies to the reviewers. Supporting lonely isolated older people at risk is a key problem facing Western societies, Japan...

As has been stressed, you cannot make conclusions that the programme had an effect because of the large numbers of known and unknown confounders. I would like you to make a table listing all the confounders and items of data you were not able to collect listed by the two reviewers to illustrate to future researchers the data that they need to collect. This is not a criticism of your work, but important that as part of your scientific contribution your guide future researchers to identify methods to collect as much information on confounders as possible. Both reviewers listed large numbers of items of possible contacts that you were not able to assess

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Roger E. Thomas

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

2 Nov 2021

Dear Reviewers

please find below the answers to your suggestions that we included fully in the final manuscript. On behalf of the group of authors, I want to thank you for your contribution to improve the paper

with my best regards

Giuseppe Liotta

Reviewer #1: Thank you for your revision. While I agree to almost all author responses and revisions, there are still two minor comments that I think authors need to address and provide appropriate conceptual evidence before the manuscript qualifies for publication.

Thank you for your contribution to the improvement of the paper. Please find the answers to your comment by points

The authors have mentioned in response to comment 1 that mitigation and containment strategies by the government could have somewhat influenced the number of cases and mortality. Authors responded in the last sentence “Nevertheless, as suggested by the reviewer, we reported the latter consideration as a study limitation.”

Authors added the following sentence in the limitations part “adherence to governmental public health measures (wearing masks, keeping distance, stay at home, etc.; diseases affecting the studied population or the causes of death).” These measures were known to affect the infectiousness (Rt) overtime, thereby determining cases escalation or decrease over time, yet may cause the mortality rates to increase or decrease overtime if health systems were overwhelmed or cases were undetected. As authors agreed to the suggestion and noted in the limitation, it needs to be justified with evidence as conceptualized in the reviewer’s comments. The following evidence needs to be cited to in accordance to authors acceptance of the proposed work and need to be corroborated based on the above argument:

Ganasegeran, K.; Ch’ng, A.S.H.; Looi, I. What Is the Estimated COVID-19 Reproduction Number and the Proportion of the Population That Needs to Be Immunized to Achieve Herd Immunity in Malaysia? A Mathematical Epidemiology Synthesis. COVID 2021, 1, 13-19. https://doi.org/10.3390/covid1010003

Please see the discussion section, we have tried to report in the best possible way the considerations of the reviewer

Second, the SMR formula needs to be incorporated under methods part as an operational definition.

We added the SMR formula to the methods

Once these revisions have been made, I have no further objections for the paper to be published.

Reviewer #2:

Thank you for your detailed replies to the reviewers. Supporting lonely isolated older people at risk is a key problem facing Western societies, Japan...

As has been stressed, you cannot make conclusions that the programme had an effect because of the large numbers of known and unknown confounders. I would like you to make a table listing all the confounders and items of data you were not able to collect listed by the two reviewers to illustrate to future researchers the data that they need to collect. This is not a criticism of your work, but important that as part of your scientific contribution your guide future researchers to identify methods to collect as much information on confounders as possible. Both reviewers listed large numbers of items of possible contacts that you were not able to assess

We added a table with the potential confounding factors w cannot assess at the end of the discussion section. thank you very much for your contribution

Submitted filename: Answer to reviewers 31102021.docx

Click here for additional data file.

6 Dec 2021

Pro-active monitoring and social interventions at community level mitigate the impact of Coronavirus (COVID-19) epidemic on older adults’ mortality in Italy: a retrospective cohort analysis

PONE-D-21-11121R2

Dear Dr. liotta,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Siew Ann Cheong, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The latest version of the revised manuscript is acceptable. Authors have addressed my suggestions well. Thank you.

Reviewer #2: Thanks to the authors for their careful responses to all the reviewers' suggestions. The table of potential confounders is very helpful.

This is an interesting study and the authors are to be commended. The manuscript is carefully written and the conclusions are now appropriately guarded.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Roger E. Thomas

13 Dec 2021

PONE-D-21-11121R2

Pro-active monitoring and social interventions at community level mitigate the impact of Coronavirus (COVID-19) epidemic on older adults’ mortality in Italy: a retrospective cohort analysis

Dear Dr. liotta:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Siew Ann Cheong

Academic Editor

PLOS ONE