Parenting-Related Exhaustion During the Italian COVID-19 Lockdown

Daniela Marchetti, PHD; Lilybeth Fontanesi, PHD; Cristina Mazza, PSYD; Serena Di Giandomenico, PSYD; Paolo Roma, PHD; Maria Cristina Verrocchio, PSYD


J Pediatr Psychol. 2020;45(10):1114-1123. 

In This Article


Participants and Procedures

The survey was completed by a sample of 2,173 Italian mothers and fathers. Of these, 219 (10.1%) were excluded from the analysis because they did not meet the inclusion criteria, while 728 (33.5%) were eliminated due to incomplete or missing information. The final sample was comprised of 1,226 caregivers (89% mothers; M age 39.13 ± 6.77). The online survey was administered on the Qualtrics platform, between April 3 and 14. Participants completed the survey only after indicating their consent on a form that described the study aims, participant rights, and data treatment procedure. Participants were recruited through social media (i.e., a survey link posted on Facebook and Instagram, with instructions), including snowball sampling via WhatsApp. The survey took approximately 20 min to complete, and participation was voluntary, anonymous, and free. The study was approved by the local ethics committee (Board of the Department of Human Neuroscience, Faculty of Medicine and Dentistry, Sapienza University of Rome, n. 6.2020). The inclusion criteria were as follows: (a) at least 18 years old and (b) parent to at least one child, aged 0–13 years, who was living with them during the lockdown. With respect to the latter criterion, we selected this age range for the children because we expected that the parents of these children would be experiencing: (a) a higher education-related burden during the outbreak, since younger children often require more parental assistance in their lessons and homework than do adolescents; and (b) a higher emotional burden as a result of having to take care of toddlers round the clock, during a highly stressful situation. Descriptive information for the sample is reported in Table I.


An ad hoc questionnaire was used to gather demographical information and information relating to particular COVID-19 experiences (Table I). The questionnaire encompasses personal information as parental role (mother/father); age; nationality; marital status (married/cohabitant or unmarried); educational level; occupation (employed or unemployed); number of children and age; presence of children with special needs (cognitive or physical disabilities). COVID-19 lockdown experiences were assessed by the following: residence during the lockdown (North, South, or Central Italy); living in the most infected areas (Lombardy, Emilia-Romagna, Piedmont, Veneto, and Marche, which were the regions with the higher infection rates); with whom they were spending the lockdown (only with children or with other family members); working (from home or at the workplace); and finally a set of yes/no questions about the COVID-19 infection(if the participant or significant others were infected, if participants suffered the loss of significant others due to COVID-19).

Parenting-related exhaustion was assessed using the Emotional Exhaustion subscale (EE) (eight items evaluated on a 7-point Likert scale) of a translated version (Fontanesi et al., 2020) of the Parental Burnout Inventory (Roskam et al., 2017). The Parental Burnout Inventory is a reliable instrument for measuring parenting-related burnout (Mikolajczak et al., 2019; Sánchez-Rodríguez et al., 2020; Van Bakel et al., 2018). Example items are "I feel tired when I get up in the morning and have to face another day with my children" and "I feel my parental role is breaking me down." In the present sample, Cronbach's α was .87. Cut-off scores were calculated following Maslach et al.'s (2010) indications, as follows: low ≤16, moderate 17–24, and high ≥25.

To assess parents' protective factors, we used the Parental Resilience (PR) and Social Connections (SC) subscales of a translated version (Fontanesi et al., 2020) of the Parents' Assessment of Protective Factors (Kiplinger & Browne, 2014). Each of these subscales is comprised of nine items, which are evaluated on a 5-point Likert scale. Example items are "I take good care of my child even when I am sad" and "I manage the daily responsibilities of being a parent/caregiver" for the PR subscale; and "I have someone who will help me get through tough times" and "I am willing to ask for help from my family" for the SC subscale. Internal consistencies in our sample were .85 and .93, respectively.

The General Health Questionnaire-12 (GHQ-12)—a shorter version of Goldberg's General Health Questionnaire (Goldberg, 1978)—was employed to assess parents' psychological distress during the lockdown. Cronbach's α was .84. Cut-off scores were calculated on the basis of previous research on general and clinical populations in Italy, as follows: scores ≤13 indicated low or no significance presence of psychological distress, and scores ≥14 indicated the (clinical) presence of psychological distress (Giorgi et al., 2014; Piccinelli & Politi, 1993; Piccinelli et al., 1993).

Statistical Analyses

Statistical analyses were run using IBM SPSS 26. Post-hoc power was calculated using G*Power Descriptive statistics of all study variables and prevalence rates for emotional exhaustion and distress among the surveyed parents were computed. Differences between mother and father subgroups on emotional exhaustion and stress were assessed using t-test analyses. A multivariate linear regression model was used to evaluate the influence of risk and protective factors on parental emotional exhaustion scores. Predictors comprised pre-existing conditions (i.e., parental role, employment, education, age, marital status, children age, having a child with special needs, number of children), COVID-19-related experiences (i.e., working during lockdown, living with others during lockdown, living in a highly infected area, number of relevant others infected, losses due to COVID-19), and psychological variables (i.e., PR, SC, GHQ-12). These factors were sequentially included as predictors of total emotional exhaustion in three steps. A post-hoc power analysis (1—β) suggested that the sample size was sufficient to provide robust power (1—β = 1.0) for multiple linear regression with an effect size of 0.41, a critical α of .05, and 16 predictors.