Matching parenting support needs to service ... - Wiley Online Library

care, health and development Child: Original Article bs_bs_banner

doi:10.1111/j.1365-2214.2011.01315.x

Matching parenting support needs to service provision in a universal 13-month child health surveillance visit cch_1315

665..674

C. Wilson,* L. Thompson,† A. McConnachie‡ and P. Wilson§ *University of Glasgow †Public Health Resource Unit, NHS Greater Glasgow and Clyde ‡Robertson Centre for Biostatistics, University of Glasgow, and §Centre for Population Sciences, University of Glasgow, Glasgow, UK Accepted for publication 12 August 2011

Abstract

Keywords attachment, infant mental health, maternal depression, screening Correspondence: Philip Wilson, DPhil MRCPCH FRCGP, Centre for Population Sciences, University of Glasgow, Caledonia House, Royal Hospital for Sick Children, Yorkhill, Glasgow G3 8SJ, UK E-mail: philip.wilson@ glasgow.ac.uk

Background The Glasgow Parenting Support Framework is an intervention to support families with preschool children. It provides structured support through universal and targeted interventions. Two routine family visits by health visitors have been piloted, one involving a home assessment when the child is 13 months old. Aims To establish the need for parenting support in the population at 13 months and whether or not the home assessment improved the match of service provision to need. Methods Health visitors were asked to collect data on existing problems and service provision to families, the mental state of the child’s principal carer, an observation of the parent–child relationship and details of management plans. Data from the Child Health Surveillance System were also used. Results Data were obtained for 549 families. Nine families were noted to have problems in the parent–child relationship. Carers in these families had poorer mental health. In all, 20% of families had a revisit scheduled: they were also more likely to have poor mental health. All families with possible problems in the relationship had a revisit scheduled. Depression scores measured during the pilot were a significant predictor of revisiting and referral, with an odds ratio of 1.37 for every one point increase in score on the Adult Wellbeing Scale depression subscale. Conclusions Current service provision matches need to some extent but routine visits focused on parenting difficulties at 13 months, particularly parental depression, may help to identify families needing support who would not otherwise have received it.

Introduction There is growing evidence that a child’s early environment affects cognitive and emotional development. There are critical or sensitive periods for brain development in the early years which shape future social behaviour (eds Bailey et al. 2001; Chugani et al. 2001). Attachments formed between child and

© 2011 Blackwell Publishing Ltd

primary caregiver shape the child’s emotional development and influence the child’s relationships in the future (Bowlby 1969). Parental emotional well-being is a major determinant of a child’s social and emotional development (Eisenberg et al. 1999; Tough et al. 2008). As primary healthcare professionals in routine contact with families, health visitors in the UK are ideally placed to offer

665

666 C. Wilson et al.

support to parents of preschool children. Systematic reviews of health visiting services in the UK suggest they can deliver improvements in parenting skills, rates of breastfeeding and child intellectual and behavioural development (Elkan et al. 2000). Early support provided by health visitors can play a valuable role in managing the stresses encountered during early parenthood (Hogg & Worth 2000). There is however some evidence that parents may feel excluded from service provision targeted on the basis of socioeconomic determinants (Roche et al. 2005). The Scottish guidance (Scottish Executive 2005) on implementation of the Health for All Children report (Hall & Elliman 2003) proposed that universal face-to-face contacts between health visitors and families after the first few weeks of life need only be offered to families with evident risk factors for the child. More recent evidence suggests that, even in the context of regular home visits, stratification of vulnerability in the first year of life is difficult (Wright et al. 2009). The Glasgow Parenting Support Framework is a needsbased approach to service provision based on the judgement of both professionals and families. The Framework (Wilson et al. 2007) aims to provide structured support through universal and targeted interventions such as the Solihull Approach to infant mental health (Blackwell 2004) and the triple-P parenting programme (Sanders et al. 2003). One pilot component of the Framework involved a home visit by a health visitor to all families when their child is 13 months old. In Glasgow there is a routine contact at 13 months for immunizations within the universal child health programme (Hall & Elliman 2003) but in general the contact is used for immunization alone. The home visit reported here was planned to be a separate contact from the immunization consultation but the latter could be used as an opportunity for raising the issue of the home assessment. The timing of the home visit was also selected because assessment of parent–child attachment is difficult before the age of 1 year (Ainsworth et al. 1978). The visit involved observation of the parent–child interaction and assessment of the psychological well-being of the primary caregiver. The aim of the evaluation of the 13-month visit was to assess whether or not the visit could improve the match between need for parenting support and provision of services to families.

Methods The 13-month visits by health visitors were piloted for 6 months in West Glasgow. All families in the West Glasgow Community Health and Care Partnership with children aged 13

© 2011 Blackwell Publishing Ltd, Child: care, health and development, 38, 5, 665–674

months during the 6-month pilot period were eligible for a visit. Health visitors were asked to visit all of these families in their caseload, but there were a number of ‘vacant’ caseloads caused by sickness absence and non-recruitment to posts during the pilot period. All health visitors in the area were consulted 6 months before the pilot and given a training session 2 weeks before it started. During the visits, health visitors recorded any existing problems and service provision to families then were asked to observe the parent–child interaction and record whether or not they had any concerns. They were asked to observe the parent and child engaged in an activity such as nappy changing or playing for a few minutes: just long enough to gain an impression of their relationship. A checklist was provided for health visitors to expand on what areas were causing concern. Further details of the validity of this assessment are reported elsewhere (Wilson et al. 2010). Health visitors were also asked to administer the Adult Wellbeing Scale (AWBS – http://www. dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/@dh/@ en/documents/digitalasset/dh_4079490.pdf) to the child’s primary caregiver, to provide information on psychological well-being. The AWBS is part of the Department of Health Framework for the Assessment of Children in Need and their Families and has been validated but has not been widely used clinically. (Snaith et al. 1978) The anxiety and depression subscales correspond to those used in the popular Hospital Anxiety and Depression Scale (Zigmond & Snaith 1983) and there are additional items addressing irritability, generally considered to be useful in the assessment of parenting stresses. The AWBS was sent to families to complete in advance of the visit, although only 26% of families completed it in advance of the visit. In order to assess need for parenting support in West Glasgow, information recorded during the 13-month visit was combined with population-level data from the Child Health Surveillance System. This provided demographic information such as the Scottish Index of Multiple Deprivation (SIMD) (http://www.scotland.gov.uk/Topics/Statistics/SIMD/) based on the domains of income, employment, health, education, skills and training, housing, geographic access and crime. During visits, health visitors were asked to record the action they had taken as a result of the visit. This may have been a revisit, referral and/or change in the child’s Health Plan Indicator (HPI) status. The HPI is a means of stratifying need within community health care, in line with the Scottish implementation guidance for the Health for All Children recommendations (Scottish Executive 2005). Families are classified as ‘Core’, ‘Additional’ or ‘Intensive’ by health visitors, usually by the time the

A universal 13-month health surveillance contact 667

child has reached 8 weeks. A revisit, referral and/or change in the child’s HPI status were used as indicators of intention to provide further services, in order to assess whether or not this matched need for support.

Statistical analysis Data were analysed using R for Windows, version 2.11.0. Analysis of families with possible problems in the parent–child relationship was primarily qualitative because of the small number recorded as having possible problems. AWBS results were analysed as scores for each of four subscales: depression, anxiety, inwardly directed irritability and outwardly directed irritability. Scores were summarized by mean and standard deviation according to HPI status (prior to the 13-month visit), quintile of SIMD, whether the family was know to other services, whether there were possible problems with the relationship and revisit/referral status after the visit. Subgroups were compared with Kruskal–Wallis and Mann–Whitney tests. The number and percentage of families with a revisit scheduled or who were referred was summarized by HPI at the start of the visit, SIMD quintile, the family being known to other services and change in HPI status (if a post-visit HPI was not recorded, it was assumed to be unchanged. Revisit/referral rates were compared between subgroups with Fisher’s exact test. Qualitative analysis of those families who changed their HPI status as a result of the visit was chosen because of the small numbers of families who changed status. Logistic regression was used to establish if AWBS scores could predict a revisit or referral in addition to current indicators of need: SIMD quintile, HPI at the start of the visit and the family being known to other services. Each predictor was assessed univariately, and a multivariate model was fitted with all predictors simultaneously. Factors associated with revisit or referral in the multivariate model were used to create a predictive model, which was illustrated graphically. Ethical approval was not required for the evaluation but Caldicott Guardian approval was obtained.

Results Health visitors gathered information for 549 families: 61.3% of 896 families in West Glasgow eligible for a visit during the pilot (Fig. 1). SIMD and HPI distribution of visited families closely matched those eligible for a contact in West Glasgow. The only information available for those families for whom

data were not gathered are SIMD and HPI. Of the 549 families, 83% were successfully visited; the distribution of HPI, SIMD and whether or not the family was known to other services were similar between those families visited and not visited. Analysis by caseload number (where attributable to individual health visitor) showed great variation in individual practice. The number of successful visits per health visitor ranged from 0 to 36 (median 15), with the proportion of caseload successfully visited ranging from 0% to 200% per health visitor. Health visitors noted possible problems with the parent– child relationship in nine families. Seven of them were in SIMD 1 (the most deprived quintile) and six were known to other services. These other services included social work services, community paediatrics, parents and children together teams, psychiatric services, addiction workers and child protection services. All of these families had significantly higher AWBS scores across all subscales. AWBS depression, anxiety and inwardly directed irritability scores were also higher (P < 0.001) in the greater need HPIs. Depression scores were lower (P = 0.004) in the more affluent SIMD quintiles. Depression (P = 0.001) and anxiety (P = 0.023) scores were higher in those families known to other services (Table 1). There was no association between an incomplete AWBS questionnaire and either SIMD or HPI (P > 0.05). Of all those completing the AWBS, 9.5% (n = 36) received a score which may indicate a problem with depression. Ten of these families were in the Core HPI and only 10 were known to other services. In all, 6.8% (n = 26) may have had a problem with anxiety, seven of these families being in the Core category and only six being known to other services. In all, 2.6% (n = 10) may have had a problem with outwardly directed irritability, four being in the Core category and two being known to other services. Similarly, 2.6% (n = 10) may have had a problem with inwardly directed irritability but unlike outwardly directed irritability, all of these families were in the Additional or Intensive HPI, and five were known to other services. In total, 20% of families had a revisit scheduled. Fifteen referrals were made: five were to a general practitioner and four to a parenting class. All families with possible parent–child relationship problems were revisited but only two were referred. All AWBS scores were increased in those families who were revisited and all except outwardly directed irritability were higher in those referred (Table 1). Socio-economic deprivation was associated with revisiting (P = 0.003) but not referral (P = 0.484); revisiting was significantly associated (P < 0.001) with the family being known to other services but referral was not associated (Table 2).


668 C. Wilson et al.

Data returned n = 549

Family visited n = 458 (83.4%)

Core 323

Core 276 (85.4)

HPI at end of visit n = 370 Missing = 88

Intended revisit or referral n = 96 (21%) Yes 21 (7.6)

11

No 255 (92.4)

59

2

1

Not visited 47 (14.6) Yes 40 (27.6)

Intensive 45

35 55

No 105 (72.4)

17

1

Intensive 37 (82.2)

Additional 101

5

Additional 145 (81.5)

Not visited 33 (18.5)

HPI not known 61

8

33

Additional 178

Core 239

195

HPI not known 22

1 30

Yes 35 (94.6)

Intensive 30

4 No 2 (5.4)

1

HPI not known 5

Not visited 8 (17.8)

HPI not known 3

Not visited 3 (100) Figure 1. Flowchart. HPI, Health Plan Indicator.

Thirty-five families moved to a lower HPI status, with 33 moving from Additional to Core and two moving from Intensive to Additional. Nine families moved up an HPI status from Core to Additional. The changes in HPI status were evenly distributed throughout the SIMD quintiles. Of those increasing in HPI status, 8/9 (89%) had a revisit scheduled. Only one of the families with a reduced HPI status had a revisit scheduled. The 15 families who were referred did not have a change in HPI


status, although only five were in the Core group (Table 2). In the nine families with an observable problem in the parent– child relationship, four families stayed at the Additional HPI status and three at the Intensive. Two families, one initially classed as Additional and the other as Intensive, did not have their HPI status recorded at the end of the visit. In logistic regression analyses (Table 3), all AWBS scores, as well as deprivation, contact with other services and HPI

A universal 13-month health surveillance contact 669

Table 1. AWBS subscale scores by HPI status at start of visit, by SIMD quintile, by whether the family was known to other services and by revisit/referral status

AWBS subscale score Total

Depression

Anxiety

Inward irritability

What was the HPI at the start of the visit? Core 229 2.6 (2.0) 2.8 (2.4) 0.8 (1.3) Additional 123 3.8 (2.7) 4.4 (3.0) 1.6 (2.1) Intensive 28 4.6 (3.0) 4.4 (3.6) 2.0 (3.3) P-value