The purpose of this study is to determine whether a group intervention for primary health care dyads (two interacting people, in this case: mother-infant or caregiver-infant), which have been screened during pregnancy to be at psychosocial risk, has an impact on parental sensitivity.

The first three years of life of every child is fundamental for their development, these years offer a unique window of opportunity for future health promotion, and different investigations have proved that investing in this period of life not only has great economic benefits, but social ones too.

In Chile the political decision to invest in these valuable years was translated into the creation of a national social program called "Chile Crece Contigo". This Program seeks to intervene in the physical, cognitive and socioemotional areas of development of every Chilean child since they are conceived until they are 4 years old to promote the best development outcome possible. The program identifies families at psychosocial vulnerability and offers them specific interventions according to the risk detected.

In this context different health and social services are provided to pregnant women during their health care visits, such as screening for different psychosocial risks (eg. Low social support, poverty, tobacco or drug use) and according to this screening various interventions are offered by Primary Health Care centers (eg. home visits).

It has been described that in groups at psychosocial risk the prevalence of interactional problems, specifically alterations in the attachment style between the child and the caregiver, is higher than in populations without these risks. This is relevant given that the quality and style of attachment established in infancy predicts subsequent social, emotional and cognitive development. Attachment styles, other than secure, have been associated with behavioural problems in pre-school and school ages, psychopathology during adolescence, later anxiety disorders and higher rates of using emergency health care services.

One of the factors shown to predict the attachment style established between the child and the caregiver is the parental sensitivity (also called maternal sensitivity)- defined as every caregiver's pattern of conduct toward the child that soothes, and increases his comfort and at the same time reduces his stress and disinterest. It also considers the capacity of the adult to pay attention, interpret and respond to the child´s needs. The literature suggests that having an adequate parental sensitivity is associated with a secure attachment style at the end of the first year of life, and having low parental sensitivity detected early in life (even as early as 6-8 weeks) predicts an insecure style of attachment at age 2.

Preliminary Chilean data has shown that 25% of biparental families and up to 52.5% of monoparental families with low income are have low parental sensitivity and are catalogued as having "Interactional risk" (defined as a score of 6 or less points in the CARE-Index scale), indicating the need of offering an intervention specifically aiming to work on the caregiver-child relationship.

Since the creation of the Program "Chile Crece Contigo" there have been great achievements toward the promotion of adequate infant development, especially in the cognitive and physical areas of development, however the initiatives in the socio emotional area are lacking. Issues in caregiver and child relationships are not addressed systematically, which is why counting with interventions that promote parental sensitivity and the development of secure attachment relationships that can be offered to families at psychosocial risk responds to a social necessity and contributes to the promotion of adequate infant mental health and family health.

Considering that up to date an intervention for the means described above and available to be used in Primary Health Care did not exist, during the years 2008-2010 such intervention was designed. The intervention was elaborated according to the method described by the UK Medical Research Council. The intervention was then piloted in two Family Primary Health centers showing an improvement in caregiver-child interactions. Analysis indicate that after 4 months of the intervention: caregivers valued positively having participated, there were no losses to participating, there was an increase of 33% of secure attachment styles (according to the Massie-Campbell Scale) which was not statistically significant, but could result significant in a larger sample size.

The intervention proposed aims to improve caregiver-child relationships, specifically by focusing on parental sensitivity, and is easily applicable and replicable in Primary Health Care considering that it requires little personnel training, the materials used are easily available and are cheap for Primary Health Care. However, the question of "has the intervention proposed any impact on parental sensitivity in dyads at psychosocial risk?" still has to be addressed.

We define the term "dyad" as two interacting people, in this case: mother-infant.

The objective of this study is to evaluate the impact of a group intervention on the parental sensitivity of primary health care dyads that have been screened positive for psychosocial risk during pregnancy.

Specific objectives:

• Evaluate the percentage of dyads catalogued as having "Interactional risk" in the intervention group compared to the control group
• Evaluate the change in the score of parental sensitivity of the dyads in the intervention group compared to the control group
• Evaluate the change in the number of depressive symptoms of the mother of the dyads in the intervention group compared to the control group

Methods: randomized clinical trial. Dyads (mother-infants from 4 to 12 months of age) which have been screened during pregnancy to be at psychosocial risk in 2 Family Health Care Centers in the La Pintana district in Metropolitan Santiago, will be randomized to:

• Receive usual care ("control group") consisting in: home visits, evaluation by social worker and planning interventions according to multidisciplinary team
• Participate in a group intervention besides receiving usual care The randomization will be done with a computer program and sealed in dark envelopes so the sequence remains occult until the dyad enters the study.

Parental sensitivity and interactional risk will be measured with the CARE-Index scale during 3 timings: at enrollment (when the interview is done), immediately after participating in the group workshop (from 1 to 2 weeks after the workshop is finished) and 4 months after (measuring times are also for control group). The scoring of the CARE-Index scale will be done by expert external coders that are blind to the randomization group.

Symptoms for postpartum depression will also be measured with the Edinburgh postnatal depression scale (EPDS).

Statistical analysis: the primary outcome is the change in the category of "Interactional risk" in the CARE-Index scale. Secondary outcomes are the changes in the scores in parental sensitivity and number of depressive symptoms. We will describe the data using averages, percentiles and standard deviations for quantitative variables and proportions for nominal variables. The bivariate analysis includes χ2 test statistics and t tests for independent samples using the SPSS 17.0 Software. The sample size was determined to detect significant differences in the percentage of dyads that change their interactional risk category in the CARE-index scale between the intervention and control groups. Each group should have 90 dyads, assuming a change in 5% due only to the pass of time, a change in 20% of the intervened dyads, a power of at least 80% with a 5% significance level, and estimating a 20% loss of the dyads during the execution of the investigation.

The intervention consists in a group workshop (for up to 7 dyads with children aged between 6 and 12 months and two health care professional monitors) of four sessions of two hours each, held weekly. One of the sessions is characterized as including the fathers or other caregivers relevant to the upbringing of the children. Each session is structured around various activities that specifically deal with the skills associated with parental sensitivity and address relevant issues to child rearing, considering the development of the child.

The intervention was designed through the methodology proposed by the UK Medical Research Council and has been tested in a pilot study of 11 dyads in two primary health care centers and showed an improvement of 33 per cent in the secure attachment style in the dyads (not statistically significant).

Recruiting, enrollment and procedures: mothers that have been previously screened positive during pregnancy to be at psychosocial risk will be contacted for an interview if their child is between 4 and 12 months old. In the interview they will be asked if they would like to participate in the study and sign an informed consent form. Once they have done this an interview will take place to collect socio demographic data regarding the family, record a 3 minute video of the dyads playing (to be coded with the CARE-index scale) and an Edinburgh postnatal depression scale applied. The interviewer will then assign the dyad to the intervention or control group (through the use of closed numbered envelopes to which the interviewer is blind).

If the dyad is recruited for the intervention group they will be invited to participate in the group workshop mentioned above, besides receiving the usual interventions the center does for dyads of psychosocial risk. And asked to come to the centers for the measurements of the following CARE-Index and Edinburgh postnatal depression scales.

If the dyad is recruited for the control group they will receive the usual interventions the center offers dyads at psychosocial risk and asked to come for the following measurements (as in the intervention group).

The recruiting will be done by trained interviewers (different health care professionals that work at the primary health care centers) that will follow a strict protocol to avoid missing entering data and maintain confidentiality at all times. The data will also be managed with strict confidential rules.

If the group intervention should prove to have a positive impact on maternal sensitivity it could be used in other primary health care centers, constituting an effective form for promoting adequate socio emotional development the first years of life.

Inclusion Criteria:

• Dyad with the child aged 4-12 months old
• Have screened positive for 1 or more psychosocial risks during pregnancy

Exclusion Criteria:

• Mother or caregiver with severe mental health disorder, specifically: schizophrenia, mental retardation and severe mood disorders with active suicidal ideas
• Child with important biomedical problems, specifically: genetic syndromes, severe heart disease and extreme preterm birth
• Severe psychosocial problems, specifically: previous history of child abuse or institutionalization of other children from the family
• Not signing the Informed Consent Form