Unintentional injuries and potential determinants of falls in young children: Results from the Piccolipiù Italian birth cohort.

OBJECTIVES: Unintentional injuries such as falls, are particularly frequent in early childhood. To date, epidemiological studies in this field have been carried out using routine data sources or registries and many studies were observational studies with a cross-sectional design. The aims of the study are to describe unintentional injuries in the first two years of life in the Piccolipiù birth cohort, and to investigate the association between mother and children characteristics and the First Event of Raised surface Fall (FERF).
METHODS: This longitudinal observational study included 3038 children from an Italian birth cohort. Data on socio-demographic factors, socio-economic indicators, maternal health and lifestyle characteristics and child's sleeping behavior, obtained from questionnaires completed at birth, 12 and 24 months of age, were considered in the analyses as potential risk factors of FERF. Time of occurrence of FERF was analyzed using the Kaplan-Meier method. The multivariable analysis for time to event was carried out using a Cox proportional hazards model.
RESULTS: Falls from raised surfaces are the leading cause of unintentional injuries in the cohort with 610 (21.1%) and 577 (20.0%) cases among children during the first and second year of life, respectively. An increased risk of FERF was associated with several risk factors: maternal psychological distress (HR 1.41, 95%CI 1.10-1.81), maternal alcohol intake (HR 1.26, 95%CI 1.10-1.45), and child's sleeping problems (HR 1.28, 95%CI 1.09-1.51). Children with older aged mothers (HR 0.98, 95%CI 0.96-0.99) and living in northern Italy (HR 0.64, 95%CI 0.55-0.75) had a lower risk of FERF.
CONCLUSION: The results of the study suggest that a higher risk of FERF is associated with socio-demographic factors, maternal characteristics and child sleeping behavior that could hinder parent empowerment.

Introduction

Unintentional injuries are a leading cause of death among children, an important health threat and a public health issue [1]. The pattern and etiology of injuries and their outcome vary substantially within populations and across countries but worldwide, approximately 950,000 children die every year due to unintentional injury [2]. Previous studies investigating the circumstances leading to child mortality following injury have found that most injuries could be prevented [3-5].

Unintentional injuries due to falls, near-drowning and burns are particularly frequent in early childhood, even in high-income countries [6]. A population-based cohort study conducted in Japan estimated that more than 60% of children aged 1.5 years were affected by unintentional injuries in their first year of life [6]. Individual and family-related risk factors associated to child injury from previous studies include: gender (being a male), having a large number of siblings (3 or more children), having mothers with psychological, and/or behavioral problems and having a young mother (aged < = 22 years) [7-13].

So far, epidemiological studies on unintentional injuries in children have been mainly carried out using routine data sources and registries [14-17], which included only severe injuries reported to secondary health care service, or have often used the cross-sectional design [8, 12, 18], which is generally considered of low-quality. Moreover, the contribution from birth and child cohorts in this field was focused on school-children and adolescents [19-22]. European studies including preschoolers have been mainly conducted in Northern countries [23-25], and the evidence on Mediterranean countries is limited.

Piccolipiù is a prospective cohort of newborns enrolled in Italy. Since fetal and infant life are periods of rapid development, characterized by high susceptibility to exposures, this prospective cohort was set up to investigate the effects of environmental exposures, parental conditions and social factors acting during pre-natal and early post-natal life on infant and child health and development [26].

The aims of the present study are to describe unintentional injuries in early life, with a particular focus on those due to falls from a raised surface, and to identify potential risk factors associated with the First Event of Raised surface Fall (FERF) during the first two years of life in the Piccolipiù cohort.

Methods

Study population

Piccolipiù is the name of a birth cohort of 3,358 children enrolled in six maternity wards located in five Italian cities (Florence, Rome, Trieste, Turin, and Viareggio) between October 2011 and March 2015 [26]. Women were contacted during pregnancy or at delivery and a written informed consent form for participation was signed by both parents at enrollment. Enrolled mothers were asked to complete a baseline questionnaire, with questions on demographics, environmental exposures, and mother’s health. Additional information was obtained either from medical records or directly from the mother within the 48 hours after delivery. Parents were then contacted at 6, 12, 24 months, 4 and 6 years after delivery and asked to fill in self-administrated questionnaires which included information on demographics, environmental exposures, lifestyle, and mother’s and child health. The Piccolipiù study was approved by the Ethics Committee of the Local Health Unit Roma E, national coordinator of the project (Prot. CE/82 09/06/2011), and of each local center [27].

For the purpose of this study, only data obtained at birth (n = 3,038), at 12 months (n = 2,897) and 24 months (n = 2,751) of age were considered. The response rates were 99.4%, 87.2% and 83.8%, respectively.

Study outcomes

Data on unintentional injuries were provided in both 12 and 24 month’s questionnaires, where parents reported if their children had experienced at least one of the following injuries in the previous twelve months (multiple answers were allowed): falls from a raised surface (e.g. bed, furniture, table, chair), burns/scalds, poisoning, foreign body ingestion, road traffic injuries and other injuries. Moreover, details about circumstances of each fall from raised surface were collected; in particular where and when the fall happened, who was with the child, type/site of injury (multiple choice answer) and the treatment required.

Potential risk factors

A series of potential risk factors were considered in the analyses, namely: socio-demographic factors, socio-economic indicators, maternal health and lifestyle characteristics and child’s sleeping behavior.

The enrollment area (center/north of Italy), child gender, maternal age when the child was born (<30, 30–34, > = 35), maternal education level, paternal occupation (employed, not employed), maternal smoking (no/yes) during pregnancy and alcohol consumption (no/yes) during pregnancy were obtained from the baseline questionnaire. Maternal educational level was classified in three categories: low (primary school), medium (secondary school) and high (university degree).

All others variables, including maternal occupation (employed, not employed), siblings (no/yes), day care attendance (no/ yes), maternal psychological distress (high distress, moderate distress, feel good), and variables concerning child sleeping patterns, were assessed at 12 months.

Moreover, the Equivalised Household Income Indicator (EHII), was used as indicator of the total disposable monthly household income at birth, standardized by household size and composition in our cohort [28]. Piccolipiù cohort data (maternal age, cohabitation status, country of birth, educational level, occupational status and occupational code; paternal/partner age, country of birth, educational level, occupational status and occupational code; and household size and tenure status) and external data from the Italian 2011 “European Union Statistics on Income and Living Conditions” (EU- SILC) survey [29, 30] were considered. The EHII was reported as a three-level categorical variable by using the 25th (933 Euro) and 75th (1,810 Euro) percentiles of total monthly disposable household income in the Italy-EUSILC survey as cut-off thresholds.

To measure maternal psychological distress, the Italian version of the 12-items General Health Questionnaire (GHQ-12) was administrated at the same time as the 12 months questionnaire. This is a self-administered questionnaire aimed at detecting current levels of general (not psychotic) psychiatric morbidity, mainly in the anxiety/depression spectrum over the past two weeks. It has been used extensively in many community and hospital settings in different countries, including Italy and showed high validity and reproducibility [31, 32]. Mothers were asked to rate the degree to which they had experienced several symptoms and/or mood states. Answers are reported according to a four-level Likert-type scale (from ‘‘not at all” to ‘‘much more than usual”). The GHQ-12 was reported on a two-level scale by collapsing the four Likert categories into two (coded 0–1) [31]. The total score was computed by summing up the single item scores. Thus, the theoretical range was 0–12, with higher values indicating more severe distress. We considered scores between 2 and 4 to identify moderate psychological distress, and > = 5 to identify severe distress. These cut-offs have previously been adopted in the literature to screen mental problems [31, 33–35].

Children’s sleeping behavior was assessed considering data provided in the questionnaire: (i) time needed to fall asleep (a categorical variable was created: < = 30 minutes, > 30 minutes); (ii) where the child sleeps (a categorical variable was created: in parents’ bed (co-sleeping), in a room with others, in a room alone; (iii) the use of a comfort object while sleeping such as a pacifier, a cuddly toy, sucking the thumb (no/yes); (iv) parental perception of child’s sleeping behavior. The latter, was recoded into a dichotomous variable (no/yes), in which “yes” comprised both “somewhat of a problem” and”quite a problem” responses.

Statistical analysis

Firstly, unintentional injuries from birth to 12 months and from 13 to 24 months were described as absolute and relative frequencies.

Time to occurrence of FERF was defined as the time from birth to the first raised surface fall event. In case of multiple falls, the first one was considered. Children without a documented fall from raised surface at the end of the study were censored at the date of the last available questionnaire. For example, for a child who didn’t have any fall event and had the 24 months’ questionnaire filled in (the end of study period or follow-up), the time to event considered was 24. The time of occurrence of FERF was analyzed using the Kaplan-Meier method.

The Log-rank test was used to compare time of occurrence of FERF and all predictive factors.

All variables described in the previous section (“Potential risk factors”) were included in the univariable analysis based on Log-rank test (S1 Table) and those significant (p-value <0.05), at this first stage, were analyzed with the Chi-Square test to evaluate their independence. Risk factors significant in the univariate analysis based on Log-rank test and not correlated among each other were included in the multivariable analysis. Time to event analysis was performed using Cox proportional hazard models, without testing interaction between covariates. We ran Cox proportional hazard models considering mother‐child pairs with complete information of outcome and potential risk factors (N = 2,886). We evaluated multicollinearity in the multivariable model using the variance inflation factor (VIF), considering the presence of collinearity when VIF was higher than 10. The proportional hazards assumption was assessed through scaled Schoenfeld residuals and no relevant departure was detected. Since the missing rate was lower than 10%, we handled missing as missing at random (the pattern of missingness is not related to other variables in the dataset). All analyses were conducted using STATA 12 software (StataCorp).

Results

In Table 1 unintentional injuries in children occurring in the first and second year of life are described by leading cause.

Table 1

Leading cause of unintentional injuries during the first and second year of life.

	0–12 months (children injured = 746)		13–24 months (children injured = 994)
	N	%	N	%
Fall from a raised surface	610	76.9	577	50.3
Burn/scald	39	4.9	81	7.1
Poisoning	3	0.4	13	1.1
Foreign body ingestion	24	3.0	25	2.2
Road accident	10	1.3	15	1.3
Other injury	107	13.5	435	38.0
Total	793	100	1146	100

During the first year of life, 746 out of 2,896 children (25.8%) who filled in the questionnaire at 12 months had at least one unintentional injury, for a total number of 793 injuries. During the second year of life, 994 out of 2,751 children (36.1%) who filled in the questionnaire at 24 months had at least one unintentional injury, for a total number of 1,146 injuries. Falls from raised surface were the most frequent leading cause of unintentional injury, both in the first and in the second year of life (respectively 76.9% and 50.3%).

Table 2 summarizes details of falls from a raised surface in the cohort.

Table 2

Description of every raised surface fall reported during the first and the second year of life.

	0–12 months (raised surface falls reported = 766)			13–24 months (raised surface falls reported = 706)
	N	%		N	%
Age at fall			Age at fall
1–4 months	56	7.4	13–16 months	150	21.6
5–8 months	282	37.2	17–20 months	262	37.8
9–12 months	420	55.4	21–24 months	282	40.6
Place where fall happened			Place where fall happened
Bedroom	479	62.8	Bedroom	268	38.1
Living room	147	19.3	Living room	218	31.0
Nursery	6	0.8	Nursery	27	3.8
Playground/garden	31	4.1	Playground/garden	75	10.7
Other	100	13.1	Other	115	16.4
From where the child fell			From where the child fell
Table/chair/high chair	89	11.6	Table/chair/high chair	194	27.5
Bed	437	57.0	Bed	246	34.8
Changing table	46	6.0	Changing table	16	2.3
Arms	15	2.0	Arms	20	2.8
Other (e.g. stroller, stairs, etc)	179	23.4	Other (e.g. stroller, stairs, etc)	230	32.6
Who was with the child			Who was with the child
Parents	690	92.0	Parents	594	84.7
Granparents/relatives	42	5.6	Granparents	58	8.3
Siblings	8	1.1	Siblings	13	1.9
Baby sitter	10	1.3	Baby sitter	17	2.4
Other	0	0.0	Other	19	2.7
Injury type/site of injury *			Injury type/site of injury *
Bruise	419	59.8	Bruise	319	40.9
Bleeding	63	9.0	Bleeding	73	9.4
Wound	18	2.6	Wound	61	7.8
Fracture	8	1.1	Fracture	14	1.8
Blow	48	6.8	Blow	54	6.9
Head injury	14	2.0	Head injury	15	1.9
Other	131	18.7	Nothing	219	28.1
			Other	25	3.2
Treatment			Treatment
Medical examination	45	5.9	Medical examination	18	2.6
Emergency Hospital Department	152	20.1	Emergency Hospital Department	153	22.0
Hospitalization	16	2.1	Hospitalization	1	0.1
Nothing	540	71.3	Nothing	517	74.2
Other	4	0.5	Other	8	1.1

Total number may vary across variables due to missing values

question with multiple choice

During the first year of life, most falls occurred after eight months (55.4%), under parental supervision (92%) and mainly from the bed (57%). During the second year of life, falls from a raised surface occurred more frequently after twenty months (40.6%) again under parents’ supervision (84.7%) and mainly from the bed (34.8%). Interestingly, the percentage of children fallen in the nursery was very low (0.8% and 3.8% in the first and in the second year respectively). Moreover, the percentage of children taken to an Emergency Department or hospitalized as a result of the fall was similar in the first two years (around 22%).

Potential risk factors in the 2,886 mother-child pairs considered in the present study are reported in Table 3.

Table 3

Potential risk factors of the 2,886 mother‐child pairs.

	N	%
Enrollment area
Central Italy	1929	66.8
Northern Italy	957	33.2
Child gender
Male	1471	51.0
Female	1415	49.0
Maternal age at delivery
<30	556	19.3
30–34	992	34.4
> = 35	1,338	46.3
Maternal education
Primary school	301	10.4
Secondary school	1,242	43.1
University degree or above	1,343	46.5
Maternal employment
No	739	25.6
Yes	2,091	72.5
Not respondant	56	1.9
Paternal employment
No	84	2.9
Yes	2,784	96.5
Non-responders	18	0.6
EHII *
Low income	126	4.4
Medium income	1,607	55.7
High income	965	33.4
Non-responders	188	6.5
Number of siblings
No	1,707	59.2
Yes	1,175	40.7
Non-responders	4	0.1
Day care attendance
No	2,000	69.3
Yes	708	24.5
Non-responders	178	6.2
Maternal distress
Feel good	2,146	74
Moderate distress	335	13,0
High distress	106	4.1
Non-responders	299	10.4
Maternal smoking during pregnancy
No	2,261	78.3
Yes	623	21.6
Non-responders	2	0.1
Maternal alcohol intake during pregnancy
No	1,161	40.2
Yes	1,667	57.8
Non-responders	58	2.0
Time needed to fall asleep
< = 30 min	2,576	89.3
>30 min	221	7.7
Non-responders	89	3.0
Sleeping problems (reported by parents)
No	2,341	81.1
Yes	480	16.6
Non-responders	65	2.3
Where the child sleeps
Own bed	636	22
Bed in a room with orthers	1,554	53.9
Parents bed (Cosleeping)	584	20.2
Non-responders	112	3.9
The child uses a comfort object for sleeping
No	1,005	34.8
Yes	1,820	63.1
Non-responders	61	2.1

Non-responders indicates an unanswered question

* low income = <933 euro, medium income = 934–1809, high income = > = 1810 euro

Overall, during the first two years of life, 996 out of 2886 children (35.9%) reported a FERF.

Fig 1 displays the Kaplan-Meier survival function for the probability of FERF overall. At 12 months of age 79% of children had never experienced a fall (95% CI 77.5%-80.4%), while at 24 months of age the percentage declined to 64% (95% CI 62%-65.7%).

Fig 1

Kaplan Meier survival function for the probability of first fall from a raised surface overall during the first two years of life.

FERF = First Event of Raised surface Fall.

The results of the adjusted Cox model, including the predictive variables significant in the previous univariable analysis based on Log-rank test (S1 Table), are shown in Table 4. No collinearity was observed between predictors included in our model (VIF = 1.46).

Table 4

Estimated hazard ratios (HRs) for FERF from the multivariable Cox model.

		FERF 0–24 months
		HR	CI 95%
Enrollment site
	Central Italy	1,00
	Northern Italy	0.66	0.57–0.77
Maternal age
	<30	1,00
	30–34	0.80	0.66–0.96
	> = 35	0.71	0.60–0.85
Maternal distress
	Feel good	1,00
	Moderate distress	1.41	1.18–1.69
	High distress	1.50	1.12–2.01
Maternal alcohol intake
	No	1,00
	Yes	1.23	1.07–1.41
Child sleeping problems
	No	1,00
	Yes	1.33	1.13–1.56

A lower risk of FERF was observed in children living in Northern Italy (HR 0.66, 95% CI 0.57–0.77), compared to central Italy. With regards to maternal age, a lower risk was found among children with mothers aged 30–34 years (HR 0.80, 95% CI 0.66–0.96) and >35 years (HR 0.71, 95% CI 0.60–0.85) compared to those with younger mothers (<30 years old). Moreover, an increased risk of FERF was observed in children with mothers with a moderate (HR 1.41, 95% CI 1.18–1.69) and high (HR 1.50, 95% CI 1.12–2.01) psychological distress. Maternal alcohol intake during pregnancy was also found to be associated with an increasing risk of FERF among children (HR 1.23, 95% CI 1.07–1.41). Furthermore, an increased risk of FERF was found among children having sleeping problems (HR 1.33, 95% CI 1.13–1.56).

Discussion

Our study identified maternal age, maternal distress, maternal alcohol consumption, child sleeping problems and enrollment area as predictive factors of FERF.

Children whose mothers were relatively older (> = 30 years old), had a lower risk of FERF. This finding is coherent with the evidence in the literature, in particular a study conducted in the UK reported that younger maternal age is an important risk factor for unintentional injuries among preschoolers [23]. As Mitton and colleagues hypothesized, younger mothers may be less aware of the risks that children encounter and could be more prone to injuries growing up [36].

Regarding maternal psychological distress, our study found that children with mothers with the highest GHQ-12 score are at higher risk of FERF. Factors affecting the association between maternal psychological health or depression and unintentional injury in children of pediatric age (0–3 years) have already been investigated in two longitudinal studies conducted in the UK and US [12, 25]. One possible explanation linking maternal depression and children’s risk of injury is that chronically depressed mothers do not appropriately safeguard the physical environments children engage in [37, 38]. Furthermore, Phelan and colleagues reported that child supervision behavior on children aged 0–36 months differed between mothers suffering from depression and those who do not [39]. In contrast to these findings, a study conducted in UK in a deprived setting, found that maternal depressive symptoms, stress and a lack of social support do not influence the adoption of safety practices (such as smoke alarms, fireguards, safe storage of sharp objects/medicines, stair gates and window locks) [40].

Maternal alcohol consumption was associated with an increased risk of their child having a FERF occurring in the first two years of life in our study. This finding is coherent with previous studies where parental alcohol consumption has been found to increase the risk of unintentional injuries [18, 41].

Children with sleeping problems were found at higher risk of FERF compared to children with no sleeping problems. Sleeping difficulties in children might be correlated to sleep deprivation in parents. Considering that parental supervision plays a critical role in maintaining child safety, the consequence of parental sleep deprivation might be indicative of a less attentive or effective supervision of their children [42, 43].

The geographical differences in the risk of FERF observed in our study is somewhat difficult to explain with the data at hand and requires further analysis to evaluate if safety behaviors and risk perceptions differ by regional, social and cultural setting in Italy.

No association was found between FERF and factors such as maternal education, parental employment (a proxy of availability for supervision of the child at home), having siblings, EHII, maternal smoking during pregnancy (which could indicate a lower attention to the needs and health of the child), nursery attendance, thus we were not able to give strength to the role of “moderation” of welfare factors in injuries, as previously shown by other authors [17].

Although it is well known that boys experience injuries and falls more often than girls, this study did not show any difference by gender [9–11, 44]. Although the mechanisms through which this disparity may arise are not entirely clear, it has been suggested that boys are generally more active than girls thus are more likely to incur in injuries at an earlier age [44].

Furthermore, it is important to recall that our analysis included FERF occurring during the first year of life in which child-related factors may be less important compared to mother-related factors, potentially having a role in explaining the null association found between gender and FERF.

Several limitations of this study should be mentioned. First of all, since all the information about unintentional injuries were collected from self-reported questionnaires, parents can intentionally or unintentionally underreport injuries that children have experienced; a questionnaire administrated by a properly trained interviewer, might have reduced the risk of bias and ensure a higher quality of data, reducing definitely the amount of missing data. A social desirability bias may exist so that fall injuries can be underreported; moreover, a recall bias due to a relatively long time from injury to questionnaires may occur [45, 46]. Second, collected data included maternal and child factors, but not environmental factors (such as baby gates, window guards, restraining strap use, etc.) or other factors related to the kinetic energy on impact (such as fall height and cushioning capacity of the surface of impact).

Finally, since this study considered only the first event of fall from raised surface, future studies could be focused to assess multiple and repeated falls and evaluated if risk factors are confirmed or differ.

Conclusion

Despite the limitations mentioned above, the present study represents a crucial investigation regarding the predictive factors of FERF in children, simultaneously explored in early life, and adds evidence in this field of research, where the role of birth cohorts is limited.

Results of this study suggest that unintentional injuries in early life can be addressed by interventions and policies that target supervision of the child, especially during pregnancy and toddlers’ early life, when parental role is critical to prevent childhood injury.

Further investigations will be essential to strengthen these findings, by means of which policy makers and health professionals could design prevention strategies to empower parents and significantly reduce unintentional injuries in early life.

Results of the Log-rank test.

(PDF)

Click here for additional data file.

2 May 2022

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,

Angela Lupattelli, PhD

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. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide

Additional Editor Comments:

Dear authors, please describe more thoroughly how the multivariable model was built after selection of variables in the univariate analysis. This procedure is not fully described and it is difficult to reproduce by others as it stands now. It is unclear what the criteria for retaining and removing variables from the multivariable model were; similarly, no test for predefined interaction terms (if any) are mentioned. Please also reconsider the use of a p-value 0f 0.05 for variable selection in the univariate analysis; by using this threshold of significance, you may lose important information at the stage of variable selection. Please refer to the Hosmer, Applied Logistic Regression, as example foe how predictor model building should be performed. It is also unclear how multicollineraity was evaluated; it seems that was examined between covariates, before model building - this step should be clearly described.

Missing values on covariate: please indicate the overall extent of missing values on covariates in the study, and indicate how they were handled in the analysis. It seems that a listwise deletion approach was used. If this is the case, please acknowledge the limitation of having such an approach on your results as compared to more robust methods such as multiple imputation (depending on the assumption as to why data are missing).

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: Partly

Reviewer #2: Yes

**********

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

Reviewer #1: Yes

Reviewer #2: Yes

**********

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: Yes

**********

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: It is valuable to build high-quality cohorts to support examining important research questions for injury prevention. Nevertheless, however, this research did not justify several critial issue this their injury cohort.

1. It seems that the authors did not adequately justify the rationale of implementing the Piccolipiù Italian birth cohort. Specifically, it is unclear the innovations of this cohort compared to the published cohorts related to injury prevention. Consequently, the research questions to be addressed by the Piccolipiù Italian birth cohort are unclear.

2. The raised surface fall (FERF) was not clearly defined in the manuscript.

3. The selection of potential risk factors and their operational definitions was not clearly described. This is particularly important for addressing high-quality research questions.

4. In addition, it is hard to understand the "raised surface falls reported" among children within 6 months. Because we know such young children cannot independently walk. Please explain the reason for it.

Reviewer #2: This paper describes unintentional injuries and risk factors of falls in young children from a Piccolipiù Italian birth cohort. This paper addresses an important topic in a niche population. However, it would benefit from restructuring, strengthening the introduction, distinguishing the results and the discussion, and carefully reading through for any grammatical and clarity concerns.

Abstract

• Include any hypothesized findings

Introduction

• The introduction starts off strong but weakens near the end. The structure somewhat falls apart and some of the implications of the study are lost.

• Lines 58-60 on page 3 are a little confusing and could be reworded.

• Why are we specifically describing injuries in this cohort and in Italy? I think the introduction would benefit for more context on why this study is relevant and how it can be helpful to this population.

Methods

• I’m a little confused about the meaning of “Piccolipiù.” It’s described as a “birth cohort” but what does that mean in this context? At first I though it was a city in Italy but that does not seem accurate either. Further clarity on this is essential to really understand the population of this paper.

• Do you have any examples that can be provided regarding the details about the raised surface falls?

• Do you have any psychometrics to report on the EHII?

• On page 5, the line about children’s sleeping behavior (lines 121-123) is confusing. And why did it get dichotomized?

• Lines 128-134 feel out of place in the methods section. I recommend moving them to the Statistical Analysis section.

Statistical Analysis

• Lines 137-142 should be placed in the Methods section.

• I don’t understand what is meant by the following: “Children without documented falls at the end of the study were censored at the date of the last available questionnaire.” What does “censored” mean in this context?

Results

• Page 7, lines 152-153 need to be revised as they are currently confusing: “Unintentional injuries occurred to children of the Piccolipiù cohort in the first and in the second year of life are presented in Table 1.”

• For Table 2, why were some categories multiple choice and not others?

• This section overall needs improved set up and structure. It is difficult to follow the separate ideas because of the current presentation.

Discussion

• Why might younger mothers be at higher risk? There is no interpretation of this finding.

• Any thoughts on the contrast in findings listed on page 12 lines 118-221? Also, is that supposed to be a separate paragraph? It seems randomly placed.

• I am not sure what is meant for lines 231-232 (page 13), can you elaborate further?

• The limitations section can benefit from adding what can be done differently next time and/or why the limitation existed to begin with.

• Consider adding some of the reported findings currently in the discussion to the results section. Right now the discussion feels too much like the results.

Conclusion

• Could benefit from a comment on any intervention being done to address these concerns.

**********

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: No

[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.

8 Jul 2022

Editor Comments:

Dear authors, please describe more thoroughly how the multivariable model was built after selection of variables in the univariate analysis. This procedure is not fully described and it is difficult to reproduce by others as it stands now. It is unclear what the criteria for retaining and removing variables from the multivariable model were; similarly, no test for predefined interaction terms (if any) are mentioned. Please also reconsider the use of a p-value 0f 0.05 for variable selection in the univariate analysis; by using this threshold of significance, you may lose important information at the stage of variable selection. Please refer to the Hosmer, Applied Logistic Regression, as example foe how predictor model building should be performed. It is also unclear how multicollineraity was evaluated; it seems that was examined between covariates, before model building - this step should be clearly described.

RE: Dear editor, we want to thank the editor and the reviewers for the useful suggestions that we believe significantly increased the quality of the Manuscript. Regarding the issue of selection of the predictor variables considered for the multivariable model we preferred to test each potential predictor at a time, before to include it into the multivariable model. In the first step we aimed to select the predictors with a stronger association with our outcome using a log-rank test with a p-value <0.05 as inclusion criteria, since we are in a Cox model setting. About the choice of the threshold of significance, we preferred to follow a conservative approach above all, because of the big number of predictors that we had (as reported in ‘Potential risk factors’ paragraph) in order to build a Cox regression model that included the minimum set of predictors thus avoiding over adjustment and multicollinearity. Moreover, since there is not a gold standard for the cut-off to identify predictor as “statistically significant” associated to the outcome, we think this cut-off (p<0.05) was able to discriminate among many competing hypotheses, selecting those more compatible with the data. (Greenland et al. 2016, doi: 10.1007/s10654-016-0149-3; Miller et al. 2019, https://doi.org/10.1371/journal.pone.0208631).

In the second step we tested for independence among predictors, resulted significant in the first step, using a Chi-Square test since all predictors were qualitative. In the third step we run the explicative multivariable Cox proportional hazards model including all predictors significantly associated to the outcome and independent among each other. In the final model we did not use a stepwise method for retaining or removing variables and we did not test interaction terms between covariates.

To check the presence of multicollinearity in the final model we used the variance inflation factor (VIF), considering the presence of collinearity when VIF was higher than 10.

We modified accordingly the Method section in the lines 144-157.

Missing values on covariate: please indicate the overall extent of missing values on covariates in the study, and indicate how they were handled in the analysis. It seems that a listwise deletion approach was used. If this is the case, please acknowledge the limitation of having such an approach on your results as compared to more robust methods such as multiple imputation (depending on the assumption as to why data are missing).

RE: As for missing values of covariates we reported the “non- responders” in the Table 3. Since the percentages of the missing values in the covariates were between 0.1% and 6.5% (with the only exception of a complex index of maternal distress with a missing value of 10%), we used a complete-case approach.

We added accordingly a sentence on this aspect at page 7 line 155-157.

Reviewers' comments:

Reviewer #1: It is valuable to build high-quality cohorts to support examining important research questions for injury prevention. Nevertheless, however, this research did not justify several critial issue this their injury cohort.

1. It seems that the authors did not adequately justify the rationale of implementing the Piccolipiù Italian birth cohort. Specifically, it is unclear the innovations of this cohort compared to the published cohorts related to injury prevention. Consequently, the

research questions to be addressed by the Piccolipiù Italian birth cohort are unclear.

RE: We thank the reviewer for this observation. We better clarified that epidemiological injury research has mostly used routine data sources and registries. Moreover, some work has used birth or child cohorts which, however, were mainly performed in Northern European countries (lines 60-68).

2. The raised surface fall (FERF) was not clearly defined in the manuscript.

RE: FERF is an acronym for “First Event of Raised Surface Fall” during the first two years of life; we have now written the acronym in capital letters (line 71).

3. The selection of potential risk factors and their operational definitions was not clearly described. This is particularly important for addressing high-quality research questions.

RE: Regarding the selection of the predictor variables considered for the multivariable model we preferred to test each potential predictor at a time, before to include it into the multivariable model. In the first step we aimed to select the predictors with a stronger association with our outcome using a log-rank test with a p-value <0.05 as inclusion criteria, since we are in a Cox model setting. In the second step we tested for independence among predictors, resulted significant in the first step, using a Chi-Square test since all predictors were qualitative. In the third step we run the explicative multivariable Cox proportional hazards model including all predictors significantly associated to the outcome and independent among each other.

According to the request of the reviewer, the selection of predictor variables included in this study have been described in more details (lines 144-157).

4. In addition, it is hard to understand the "raised surface falls reported" among children within 6 months. Because we know such young children cannot independently walk. Please explain the reason for it.

RE: Falls are the most common mechanism of unintentional injury but of course injury’s mechanisms change with age. In this study falls from raised surface that affected children in the first and second year of life were described. In Table 2 (line 170) it is reported the place (eg. bed, chair, table...) from where the child fell in the first 12 months of life (period in which we can assume that the child is not yet able to independently walk). Results show that falls from a raised surface occurred more frequently from the bed (57%), followed by table/chair/high chair (11.6%), changing table (6%) and etc.

Reviewer #2: This paper describes unintentional injuries and risk factors of falls in young children from a Piccolipiù Italian birth cohort. This paper addresses an important topic in a niche population. However, it would benefit from restructuring, strengthening the introduction, distinguishing the results and the discussion, and carefully reading through for any grammatical and clarity concerns.

RE: We thank the reviewer for pointing this out. We have accepted your suggestions and the final manuscript was revised by a native English speaker.

Abstract

• Include any hypothesized findings

RE: We are not sure to have understand this comment; anyway, we added a paragraph in the Abstract to clarify why unintentional injuries are important to be investigated in birth cohorts such as Piccolipiù (lines 22-25).

Introduction

• The introduction starts off strong but weakens near the end. The structure somewhat falls apart and some of the implications of the study are lost.

RE: We thank the reviewer for this comment. We modified the text according to the suggestions (lines 60-68).

• Lines 58-60 on page 3 are a little confusing and could be reworded.

RE: We thank the reviewer for these comments. We modified the text according to the suggestions (lines 60-68).

• Why are we specifically describing injuries in this cohort and in Italy? I think the introduction would benefit for more context on why this study is relevant and how it can be helpful to this population.

RE: We thank the reviewer for this comment. The Introduction has been enriched with more information from the literature available in this field (lines 60-68).

Methods

• I’m a little confused about the meaning of “Piccolipiù.” It’s described as a “birth cohort” but what does that mean in this context? At first I though it was a city in Italy but that does not seem accurate either. Further clarity on this is essential to really understand the population of this paper.

RE: We thank you for this comment. We clarified in the text that Piccolipiù is the name of our birth cohort (line 75).

• Do you have any examples that can be provided regarding the details about the raised surface falls?

RE: In this study, falls from raised surface that affected children in the first and second year of life were described. In Table 2 were reported some details about the age of child, the place where the fall happened (bedroom, living room, nursery etc),the place from where the child fell (bed, table/chair/high chair, changing table, arms, etc), the person who was with the child (mother, father, grandparents, siblings, baby sitter, etc), the type of injury reported (bruise, bleeding, blow, etc) and the treatment required (medical examination, emergency hospital department, hospitalization, etc).

• Do you have any psychometrics to report on the EHII?

RE: The Equivalised Household Income Indicator (EHII), was used as indicator of the total disposable monthly household income at birth standardized by household size and composition. It was derived using Piccolipiù cohort data (maternal age, cohabitation status, country of birth, educational level, occupational status and occupational code; paternal/partner age, country of birth, educational level, occupational status and occupational code; and household size and tenure status) and external data from the Italian 2011 “European Union Statistics on Income and Living Conditions” (EU- SILC) survey. This is definitely an interesting aspect but this is not the focus of this article.

• On page 5, the line about children’s sleeping behavior (lines 121-123) is confusing. And why did it get dichotomized?

RE: Children’s sleeping behavior was assessed by considering several items of the questionnaire. One of them is the parental perception of child’s sleeping behavior which was a categorical variable in the questionnaire with 3 levels. We decided to recode this variable in a dichotomous way (no/yes), collapsing in “yes” the categories 2=”somewhat a problem” and 3=”quite a problem”, since they had low frequencies.

• Lines 128-134 feel out of place in the methods section. I recommend moving them to the Statistical Analysis section.

RE: We thank the reviewer for pointing this out. We deleted this paragraph and reworded it with a detailed explanation of the entire process of variables selection in the Statistical Analysis section (lines 144-149).

Statistical Analysis

• Lines 137-142 should be placed in the Methods section.

RE: These lines are already in the Method Section (sub-section “Statistical Analysis”). We think they are important to understand the type of analysis performed and how the model was built. We prefer to maintain this part in the same place.

• I don’t understand what is meant by the following: “Children without documented falls at the end of the study were censored at the date of the last available questionnaire.” What does “censored” mean in this context?

RE: Censoring is referred to subjects who have not achieved any fall from raised surface. For example, if a child didn’t have any event of fall and had a questionnaire filled in at 24 months (the end of study period or follow-up), the time to event considered was 24.

Results

• Page 7, lines 152-153 need to be revised as they are currently confusing: “Unintentional injuries occurred to children of the Piccolipiù cohort in the first and in the second year of life are presented in Table 1.”

RE: We slightly modified this sentence to make it clearer and more understandable (lines 160-161).

• For Table 2, why were some categories multiple choice and not others?

RE: The questionnaires at 12 and 24 months of children’s age included some specific questions regarding falls from raised surface which occurred in the first and in the second year. For all questions only one answer was admitted, except for the question related to the type of injury (since many children had more than one type of injury), as reported in Table 2.

• This section overall needs improved set up and structure. It is difficult to follow the separate ideas because of the current presentation.

RE: We agree with the reviewer and we added some sentences in order to improve this section (lines 160-161, 164, 166, 181, 193-195).

Discussion

• Why might younger mothers be at higher risk? There is no interpretation of this finding.

RE: We hypothesize that younger mothers could have a lower risk perception and therefore they could pay lesser attention to prevent children’s unintentional injuries. We explained better this concept in lines 212-214.

• Any thoughts on the contrast in findings listed on page 12 lines 118-221? Also, is that supposed to be a separate paragraph? It seems randomly placed.

RE: We thank the reviewer for pointing this out. We tried to highlight that these different findings, obtained in a study conducted in UK, could be due to a different study population, since Mulvaney and Kendrick [ref 32], differently from our study which have been performed in general population, enrolled only mothers living in socio-economically disadvantage areas (line 226).

• I am not sure what is meant for lines 231-232 (page 13), can you elaborate further?

RE: We found that children living in northern cities of Italy (Turin and Trieste) are at lower risk of FERF compared to those living in central cities (Rome, Florence and Viareggio). This result is somewhat difficult to explain with the data at hand but we could hypothesize that this could be attributable to different safety behavior of parents who live in different cultural contests (lines 239-242)

• The limitations section can benefit from adding what can be done differently next time and/or why the limitation existed to begin with.

RE: We think the self-collected data used in the study could be one of the major limitations due to high risk of information bias (eg. underreporting of injuries) according to child’s health, family characteristics and environmental variables (eg. educational level, parity, household density). Moreover, a questionnaire administrated by a properly trained interviewer could definitely reduce the amount of missing data. We added a paragraph in lines 258-260.

• Consider adding some of the reported findings currently in the discussion to the results section. Right now the discussion feels too much like the results.

RE: We agree with the reviewer and have shortened the discussion section.

Conclusion

• Could benefit from a comment on any intervention being done to address these concerns.

RE: We agree with the reviewer and we better clarified the kind of interventions could be done to reduce the unintentional injuries in early life. Since the role of parents in maintaining toddler’s safety is crucial, we think these interventions should especially promote environmental safety and supervision of the child, therefore supporting parents from the birth or during birth classes (lines 274-276).

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

21 Jul 2022

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,

Angela Lupattelli, PhD

Academic Editor

PLOS ONE

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: (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: Partly

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: No

Reviewer #2: No

**********

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: Thanks for the replies to my comments. However, I don't think the replies and revisions towards comments 1 and 3 that I raised are adequately addressed. In this case, the innovations of this paper look to be weak.

Reviewer #2: This paper remains to be critical and meaningful to the unintentional injury literature, but there are still some issues with the quality of writing and structure of the paper when considering clarity, reproducibility, and transparency about the data and statistical analyses.

Introduction

• The introduction could still benefit from added context about the significance of the research questions and research population. Please reevaluate the current literature included in the introduction and consider a section that focuses on the regions of interest.

• There remains some issues with clarity. Although I appreciate the effort put into improving lines 58-60 on page 3, it continues to be difficult to follow. It is clear that this is an important point so please take the time to revise and improve clarity.

Methods

• I appreciate the authors’ considerations of revisions for this section and especially am grateful for the additional information provided within the statical analysis section. However, there are some big issues that continue to remain.

o The sleeping behavior paragraph continues to be difficult to follow. The response to the reviewers was helpful and should potentially be integrated into the text.

o Although I understand the EHII is not the focus of the paper, given that it is not a standardized measure, it is relevant and critical to include psychometric properties to help the readers understand the validity and reliability of the measure.

o I appreciate the clarification of “censored” but do not agree it is the best way to describe what happens to those participants. Please review what you wrote in the response to the reviewer and consider if it can be better integrated in the text.

Results

• I appreciate the attention provided by the reviewers toward revisions in this section but believe it would benefit from additional revisions. Regarding the differences in questionnaires and items (e.g, some were multiple choice and some were not), the authors might consider explaining this more clearly earlier on in the methods so that it can be more easily understood in the results.

• Overall, the structure of this section is much more easy to follow.

Discussion

• Although I appreciate the interpretation about young mothers, I want to know why this may be the case. Do you have any citations to demonstrate young mothers might have lower risk perceptions? Or any theory as to why? I know this is not the main purpose of the paper but when considering interpretations provided in the discussion, think about how to best explaining the findings in your results because that information is most helpful to the readers. You do a great job of this with the maternal psychological distress comments.

• Lines 243-247 are very difficult to follow, please revise for clarity.

• The authors did a good job of addressing revisions to the limitations and conclusion sections.

**********

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: No

**********

[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.

Submitted filename: PLOS-ONE R.1.docx

Click here for additional data file.

8 Sep 2022

Comments to the Author

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: No

Reviewer #2: No

RE: As suggested the entire text of the article was revised by a native English speaker.

________________________________________

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: Thanks for the replies to my comments. However, I don't think the replies and revisions towards comments 1 and 3 that I raised are adequately addressed. In this case, the innovations of this paper look to be weak.

RE: We thank the reviewer for this observation. We tried to improve the Introduction section in order to clarify that our cohort could contribute to increase the knowledge in this field of research (lines 63-69). We have therefore also reviewed the literature, adding several new references.

Regarding the third comment previously mentioned, to define the set of potential predictors to be included in the multivariable model, we considered some a priori risk factors based on literature (such as: sex, maternal age, maternal distress) and some risk factors based on biological plausibility, for example child sleeping behavior. All risk factors included in this study were assessed trough questionnaires.

Reviewer #2: This paper remains to be critical and meaningful to the unintentional injury literature, but there are still some issues with the quality of writing and structure of the paper when considering clarity, reproducibility, and transparency about the data and statistical analyses.

Introduction

• The introduction could still benefit from added context about the significance of the research questions and research population. Please reevaluate the current literature included in the introduction and consider a section that focuses on the regions of interest.

RE: We thank the reviewer for this observation. We tried to improve the Introduction section in order to clarify why our cohort could contribute to increase the knowledge in this field of research (lines 63-69). We have therefore also reviewed the literature, as suggested, adding several new references.

• There remains some issues with clarity. Although I appreciate the effort put into improving lines 58-60 on page 3, it continues to be difficult to follow. It is clear that this is an important point so please take the time to revise and improve clarity.

RE: We thank the reviewer for this observation. We better clarify this paragraph in lines 63-69.

Methods

• I appreciate the authors’ considerations of revisions for this section and especially am grateful for the additional information provided within the statical analysis section. However, there are some big issues that continue to remain.

o The sleeping behavior paragraph continues to be difficult to follow. The response to the reviewers was helpful and should potentially be integrated into the text.

We thank the reviewer for this comment and we added an explanation in the text as suggested (lines 141-143).

o Although I understand the EHII is not the focus of the paper, given that it is not a standardized measure, it is relevant and critical to include psychometric properties to help the readers understand the validity and reliability of the measure.

RE: We thank the reviewer for pointing this out. We have not understood the meaning of psychometric properties. However, the EHII is a measure of the total disposable household monthly income equivalised for household size and composition (currency: log-euro).

This income measure is obtained with a prediction model applied to the EUSILC data (Italian EUSILC survey for Piccolipiù) including as predictors the following variables (maternal age, maternal cohabitation status, maternal country of birth, maternal educational level, maternal occupational status and maternal occupational code; paternal age, paternal country of birth, paternal educational level, paternal occupational status and occupational code, household size and tenure status). This income measure has been derived for approximately 20 European birth cohort studies within the framework of the H2020 LifeCycle project. As described in the paper by Pizzi et al, the prediction models for each study have been already validated (temporal validation) and the derived measures compared with other SEP-related questionnaire-based information, including self-reported income, showed a strong correlation.

o I appreciate the clarification of “censored” but do not agree it is the best way to describe what happens to those participants. Please review what you wrote in the response to the reviewer and consider if it can be better integrated in the text.

We thank the reviewer for this comment and we added an explanation in the text as suggested (lines 151-152).

Results

• I appreciate the attention provided by the reviewers toward revisions in this section but believe it would benefit from additional revisions. Regarding the differences in questionnaires and items (e.g, some were multiple choice and some were not), the authors might consider explaining this more clearly earlier on in the methods so that it can be more easily understood in the results.

RE:

We thank the reviewer for this comment and we added an explanation in the text as suggested (lines 100-101)

Discussion

• Although I appreciate the interpretation about young mothers, I want to know why this may be the case. Do you have any citations to demonstrate young mothers might have lower risk perceptions? Or any theory as to why? I know this is not the main purpose of the paper but when considering interpretations provided in the discussion, think about how to best explaining the findings in your results because that information is most helpful to the readers. You do a great job of this with the maternal psychological distress comments.

RE: We thank the reviewer for this observation. Some demographic risk factors for childhood injury were already identified in different studies and of course the maternal age is one of them. Our result about the association between maternal age and FERF is very consistent with results from other studies. In the systematic review of Mitton et al., which identified cohorts of school-aged children and adolescents, is hypothesized that younger mothers, compared with older, may be less aware of the risks that children encounter as they develop and grow (Mitton et al. 2008). We added a sentence about this in order to better explain this concept (lines 221-225)

• Lines 243-247 are very difficult to follow, please revise for clarity.

RE: We thank the reviewer for this comment and we better explain the concept (lines 247-249).

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

19 Sep 2022

Unintentional injuries and potential determinants of falls in young children: results from the Piccolipiù Italian birth cohort

PONE-D-21-37522R2

Dear Dr. culasso,

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,

Angela Lupattelli, PhD

Academic Editor

PLOS ONE

26 Sep 2022

PONE-D-21-37522R2

Unintentional injuries and potential determinants of falls in young children: results from the Piccolipiù Italian birth cohort.

Dear Dr. culasso:

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. Angela Lupattelli

Academic Editor

PLOS ONE