Exploration of the relationships between health literacy and experience of care integration in parents of children with special healthcare needs: a cross-sectional study

Abstract

Objective

This study sought to identify health literacy profiles of Australian parents of children with special health care needs (SHCN), and explore the relationships between health literacy, experiences of health care integration, and parent stress.

Methods

A cross-sectional study design was used. Anonymous data were collected via an online survey including the Health Literacy Questionnaire (HLQ), a modified version of the Pediatric Integrated Care survey (PICS), parent perceived stress, and additional child and family demographic questions including child quality of life using the EQ-5D-Y. We performed a cluster analysis using the HLQ to identify health literacy profiles, and used descriptive statistics to examine relationships between health literacy profiles, health care integration and parent perceived stress.

Results

A total of 126 Australian parents of children with varied special health care needs participated, with needs predominantly related to disability (53.7%) and/or respiratory conditions (38.9%). Cluster analysis identified three health literacy profiles, which illustrated different health literacy strengths and gaps. The health literacy profiles were similar in terms of demographic characteristics but the profile with higher health literacy was associated with better experiences of health care integration and lower levels of parent stress.

Conclusion

Parents of children with SHCN have specific health literacy strengths along with skills and knowledge gaps, which are associated with how health care is experienced and perceived stress managing their child’s health. Consideration of parent health literacy can provide valuable information to inform more effective models of care and family supports.

Introduction

Parenting a child with special health care needs (SHCN) requires parents to draw on their health literacy to navigate and manage their child’s care. This responsibility can be challenging, potentially impacting parent wellbeing [1]. Children with special health care needs (SHCN) require complex healthcare due to physical, developmental, behavioural and/or emotional signs and symptoms associated with their condition [2, 3]. Many children with SHCN have a neurodevelopmental disorder, potentially associated with cognitive, communication, behaviour and/or motor impairments [4]. Other SHCN conditions such as asthma, cystic fibrosis and epilepsy are also complex and chronic, and may confer additional vulnerability to social, emotional and physical development [5]. For SHCN conditions, there is high prevalence of multimorbidity [6], thus requiring care from multiple health professionals [7]. Further, the child’s healthcare needs may require support from systems and organisations outside the healthcare sector such as those addressing social, disability and educational needs [8]. The healthcare system is mostly siloed and caters well for the management of single conditions but navigating service provision can be challenging for people with complex needs [9]. Complicated and fragmented health and social service systems have negative consequences for families who must sustain advanced health literacy and navigational skills to achieve appropriate support for their child [10, 11].

Integrated health care is defined as care coordinated across health professionals and health services, focused on continuity of care over time, and tailored to meet the needs of patients and their families [12]. This can include connecting families to resources and services beyond the health system [13]. Care integration has been associated with positive long-term health outcomes for the child and improved family functioning for children with chronic health conditions with identified additional needs [14]. Integrated care requires collaboration and communication between health professionals and provision of reasonable accommodations for the child [15] to deliver the goals of care [16]. Integrated care builds capacity within the family and care team to meet the child’s healthcare needs [17], and reduce risks of adverse outcomes for the child [18]. Care coordination and person-centred care are important components of effective models of care for intellectual disability [19].

Families contribute as partners in their child’s healthcare team in diverse ways [20]. Parents act as a catalyst to their child’s healthcare [8], facilitating information exchange and follow-up between health professionals on the care team [21]. Parents often adopt a coordination role, especially for children with SHCN whose care may include multiple specialites [22]. Parents need to develop specific knowledge and skills to perform these activities effectively, with proficiency likely impacted by their health literacy [23].

Health literacy is defined as the ability to access, understand, appraise and apply health information to manage one’s own health [24] and this extends to managing the health of others, such as one’s children [25]. Health literacy influences how families experience health care [26] and their health behaviours [27]. Health literacy is critically important for parents of children with complex health issues [22]. Health literacy influences parents’ ability to identify when health care is needed for their child, initiate help-seeking and provide ongoing management and support of their child’s health [8], which includes mastery of support or management of medical procedures when required (e.g., subcutaneous injections, enteral feeding) [9]. Increasingly, it also includes the ability to ensure fidelity in the procedural aspects of their child’s care when delivered by support workers or other paid personnel involved in their child’s care [28]. Parent health literacy also includes skills to engage and effectively communicate with health professionals [21, 26, 29], whilst advocating for and supporting their child within the health experience [30]. The support of personal networks (family, friends) is another important domain of health literacy [31], that results in an improved psychological state for parents of children with SHCN [32, 33].

Little is known about how strengths and weaknesses of parent health literacy cluster within parents with a child with SHCN, how this contributes to their experience of healthcare integration or their wellbeing. Whilst there is evidence of relationships between health literacy and healthcare experiences and outcomes [34, 35], typically family-level support is provided based on a child’s diagnosis. Identifying how health literacy interacts with experiences of care and child and family outcomes will provide more nuanced insight on how supports might be better targeted to improve outcomes [1]. Similarly, recognition of health literacy strengths and gaps is important for health services to meet the diverse needs of the community [31], for example, informing how health professionals can adjust what they do to reflect the person’s knowledge and skill [27]. This study sought to identify health literacy profiles of Australian parents of children with SHCN. We then explored the relationships between the identified health literacy profiles, the levels of health care integration experienced, and parental stress.

Methods

The study was designed and reported according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [36].

Design and procedure

A cross-sectional study design was used. The survey was provided in English language only. We recruited particpants by advertising the study through organisations who advocate and/or support children with SHCN and their families. These included chronic-condition focused organisations (e.g., cystic fibrosis, asthma, diabetes), and family support and advocacy organisations related to disability across Australia. Study advertising materials were posted with no cost on organisation social media and in organisation newsletters targeting families. Parents elected to participate if they had a child (aged 0–18 years) who they identified as having a SHCN including those who experience disability, chronic health, or mental health conditions. Participants then gained access to the survey using a QR code. Participants were not provided with an remuneration for participation. Participants were anonymous, although a small fraction elected to provide their contact details for further research opportunities. These details were kept separate to the survey data for analysis and aggregated data were reported. The survey was hosted on the web-based REDCap platform [37], and was open throughout 2024.

Ethics approval for this study (2021/ET000902) was provided by the University of Western Australia Human Research Ethics Committee in accordance to the Australian National Statement on Ethical Conduct in Human Research. This study adhered to the Declaration of Helsinki. Informed consent to participate was obtained from all participants.

Measures

The survey was developed with guidance from our Consumer Reference Group, and piloted with two parents of children with SHCN. The survey comprised four parts. Part 1 was the Health Literacy Questionnaire (HLQ) [38], a multi-dimensional health literacy tool used to better understand specific populations and their needs. The HLQ was developed independently of Sorenson’s model [24] through a grounded, concept mapping approach rather than a direct operationalisation of the model. However, the HLQ domains (e.g., having sufficient information, understanding health information, appraisal of health information, actively managing one’s health) conceptually align with the access, understand, appraise and apply dimensions of the health literacy definition adopted by Sorenson. The HLQ has 44 items across nine conceptually distinct dimensions and has strong composite reliability [38]. Dimensions 1–5 are measured on a 4-point response scale from ‘strongly disagree’ to ‘strongly agree’. Dimensions 6–9 have a 5-point response option: ‘cannot do’, ‘very difficult’, ‘quite difficult’, ‘easy’, and ‘very easy’. Dimension scores are calculated by summing the item scores and dividing by the number of items in the scale to give a mean item score. The nine dimensions of the HLQ are:

1. Feeling understood and supported by healthcare providers (4 items).

2. Having sufficient information to manage my (child’s) health (4 items).

3. Actively managing my (child’s) health (5 items).

4. Social support for health (5 items).

5. Appraisal of health information (5 items).

6. Ability to actively engage with healthcare providers (5 items).

7. Navigating the healthcare system (6 items).

8. Ability to find good health information (5 items).

9. Understand health information well enough to know what to do (5 items).

Part 2 contained an adapted version of the Paediatric Integrated Care Survey (PICS) tool [39] with supplementary questions, to measure parent-reported experience of integrated health care for their child and identify gaps in care delivery. There is evidence that the PICS domains have strong internal consistency and satisfactory test/retest reliability [39]. Questions were adapted to map question content and terms to the Australian context with respect to health and social care systems and common terminology, for example, by using Australian terms for the types of government support for which the family may be eligble. The PICS includes 21 questions about health service utilisation together with 18 items across five domains:

1. Access to care (2 items)

2. Communication with care team (4 items)

3. Family impact (5 items)

4. Care goal creation (2 items)

5. Team functioning and quality (5 items)

Additional to the standard PICS, supplementary modules from the PICS were included, with 13 questions addressing the domains of:

1. Adapting to change (3 items)

2. Connection to resources (3 items)

3. Provision of system information (3 items)

4. Enhancing communication (4 items)

Questions addressing care experience domains use a 6-point Likert scale, ranging from ‘never’ to ‘always’ (scores 0–5), with each domain score calculated as the mean score of the questions within that domain.

Part 3 included demographic questions about the child (age, gender, conditions experienced in the past 12 months) and family (respondent gender, education, household, transport, health insurance status). In addition, the EQ-5D-Y proxy version [40] measured overall health status of the child with questions relating to mobility, looking after onself, doing usual activities, having pain or discomfort, and feeling worried, sad, or unhappy. The EQ-5D-Y has been found as a feasible measure of child health in the general population with fair to moderate test/retest reliability and validity [41], and for a population of children with intellectual disability [42]. Each dimension is scored on three levels: ‘no problems’, ‘some problems’, or ‘a lot of problems’.

Part 4 used the Perceived Stress Scale 4 (PSS-4) [43], a four-item short form derived from the 14-item Perceived Stress Scale [6], to measure the frequency that feelings and thoughts associated with stress were experienced in the past month. Adequate internal consistency, test/retest reliability and validity have been demonstrated [44]. Each item is rated on a 5-point Likert scale ranging from ‘never’ to ‘very often’ (score 1–5) with the mean item score calculated to give the total score.

Analytical approach

Statistical analysis was conducted using Stata 18 [45]. We used all available data and took a complete case analysis approach for identifying health literacy profiles and predictors of profiles. Domain scores for the HLQ were calculated as described above, with 95% confidence intervals calculated using the standard error of the mean. Mean item scores were used to generate the PICS domain scores, in contrast to Ziniel and colleagues who used a composite score calculation for the PICS [39]. Confidence intervals for the PICS domain scores were calculated using the standard error of the mean. Descriptive statistics were used to characterise the study population regarding demographics and health service utilisation.

A cluster analysis was performed on the HLQ data to identify health literacy profiles within our sample, maximising homogeneity of the profiles and heterogeneity between the profiles [46, 47]. The cluster analysis used Ward’s method to combine clusters that have minimum within-cluster variance [48]. We evaluated solutions with 3, 4 and 5 profiles. Selection of the optimum number of profiles was based on visual inspection of domain scores within each profile, considering the homogeneity of scores within each profile and the meaningfulness of differences in domain scores between profiles, and identifying the largest number of profiles before at least two of the profile patterns of domain scores became parallel. The demographic characteristics were described for each profile after profile membership was assigned. Single-factor analysis of variance was conducted to explore differences in the experience of health care integration (PICS) and perceived stress (PSS-4) between profiles, with post-hoc analysis conducted using 2-tailed t-tests. Fisher’s exact test was performed to compare categorical data (EQ-5D-Y, health service use, demographic data) for different health literacy profiles.

Results

While the survey was open, consent to participate was provided by 162 parent caregivers. However, there was attrition across the questionnaire sections with 126 HLQ, 107 PICS, 102 demographic and 98 PSS-4 datasets collected (Fig. 1). Data describing the demographic characteristics of the children with SHCN are presented in Table 1, for the whole sample and for subgroups where the child’s healthcare team comprised a small (0–5), medium (6–10) or large (> 10) number of health and allied health professionals seen within the previous 12-month period.

Fig. 1.

Flowchart of participant attrition through the survey sections. Note: Section A = HLQ; Section B = PICS; Section C = Demographics; Section D = PSS-4

Table 1.

Characteristics of children with SHCN (n = 107)

			All	Small team (0–5)	Moderate team (6–10)	Large team (11+)
			n = 107	n = 28	n = 48	n = 31	p value
	Age (years)	0–4	12	1	5	6	0.12
		5–11	35	12	12	11
		12–18	55	13	30	12
	Gender	Female	40	12	19	9	0.68
		Male	60	15	26	19
		Other	2		1	1
	SHCN Conditions	Behavioural (e.g. ADHD)	66	19	31	16	NA
		Developmental (e.g. autism)	49	14	20	15
		Breathing (e.g. asthma)	42	11	18	13
		Neurological (e.g. seizures)	25	4	12	9
		Disability	58	12	27	19
		Intellectual disability	14	7	5	2
EQ-5D-Y	Mobility	None	69	24	30	14	0.03
		Some	17	1	8	8
		A lot	16	2	7	7
	Self care	None	56	19	25	11	0.20
		Some	25	4	11	10
		A lot	21	4	9	8
	Doing usual activities	None	35	13	17	5	0.09
		Some	43	11	17	14
		A lot	24	3	11	10
	Pain	None	39	14	17	8	0.33
		Some	50	11	24	15
		A lot	13	2	5	6
	Worried/sad/unhappy	None	32	10	13	9	0.76
		Some	56	13	28	14
		A lot	14	3	5	5
Health service use	Has GP	Yes	79	18	37	24	0.41
		No	28	10	11	7
	Seen GP (12 months)	none	15	5	7	3	0.52
		Occasionally (1–5)	60	18	27	15
		Often (6–12)	21	4	10	7
		Very Often (13+)	9	1	3	5
	ED presentation (12 months)	None	44	14	17	13	0.88
		1–2	39	8	20	11
		3–5	15	3	7	5
		6+	9	3	4	2
	Hospital admission (12 months)	None	54	18	23	13	0.5
		1–2	37	7	18	12
		3+	16	3	7	6

Note: Health service use, SHCN conditions and care team size were collected in PICS Section B (n = 107). Seen GP (12 months) had 2 additional missing responses (1 within the moderate team size group, 1 within the large team-ize group, giving n = 105 for this variable). Children could have multiple SCHN conditions reported. Age, gender and EQ-5D-Y were collected in the demographics section of the survey (Section C), completed by 102 of 107 participants. SHCN Special Health Care Needs: GP General Practitioner; ED Emergency department

The mean (SD) child age was 10.8 (4.7) years. Approximately half the children were age 12–18 years, a third were aged 5–11 years, and the smallest proportion was 0–4 years. Disability accounted for 58 (53.7%) of identified SHCN conditions, with respiratory conditions such as cystic fibrosis and asthma accounting for 42 (38.9%). Behavioural conditions such as ADHD were reported in more than 60% of the children, often comorbid with developmental conditions such as autism or intellectual disability. Seizures were reported in approximately one quarter of the children. Measured with the EQ-5D-Y, mobility was not a problem for the majority of children. Self-care was a problem for half the children. In contrast, the largest proportion had some problems with doing usual activities, pain or feeling worried, sad or unhappy (Table 1).

Three quarters of the children had a regular general practitioner (GP). The minority of children (14%) had not visited a GP in the previous year. However, most children had seen a GP in the previous year, including half of those without a regular GP. Frequency of visiting the GP varied, with the majority of children visiting occasionally (1–5 times, 56%), often (6–12 times, 19%) or very often (13 or more times, 8%). The majority of children had presented to the emergency department (ED) in the previous year, with a quarter presenting three or more times. Half the cohort were admitted to hospital in the previous year, with approximately 15% having three or more admissions. There were no associations between the size of the child’s care team and health service utilisation in primary-, hospital- or emergency-based care. Larger care teams were associated with greater child mobility problems; however no associations were observed with the other dimension of the EQ-5D-Y (Table 1).

Parent characteristics are detailed in Table 2. The respondents were predominantly female (89%), university educated (69%), and living in dual parent households (81%). The majority of families had full health insurance (i.e. hospital and ancillary benefits) and 18% had no health insurance. The majority of respondents had adequate transport (78%). The child’s condition had varied impacts on the family; a little (10%), somewhat (22%), quite a bit (26%) or a lot (28%).

Table 2.

Health literacy profile characteristics

		All n = 126	Profile 1 n = 38	Profile 2 n = 68	Profile 3 n = 20	p value
Respondent gender *	Female	96	29	55	12	0.07
	Male	4	2	1	1
	Other	2	2	0	0
Respondent education *	Highschool	3	1	1	1	0.68
	Trade/tech	25	7	15	3
	University	74	25	40	9
Household *	Single parent or shared care	15	5	7	3	0.16
	Dual parent	87	28	49	10
Transport *	Always adequate	82	27	46	9	0.98
	Usually adequate	14	4	7	3
	Inadequate	6	2	3	1
Health insurance *	None	17	7	7	3	0.9
	Extras only	10	3	7
	Hospital only	4	1	3
	Hospital + Extras	71	22	39	10
Care team size	Small	28	8	18	2	0.46
	Medium	48	12	29	7
	Large	31	13	14	4
Child age (years)	Mean (SD)	10.8 (4.6)	10.8 (5.7)	10.5 (4.1)	12.3 (4.0)	0.46
Impact of child’s condition *	Not at all	6	3	3	0	0.98
	A little	7	3	4	0
	Somewhat	18	5	11	2
	Quite a bit	24	8	13	4
	A lot	47	14	25	7
Perceived stress (max possible score /4) ø	Mean (SD)	2.67 (0.80)	2.37 (0.57)	2.58 (0.69)	3.73 (0.86)	< 0.001

Note: Cluster analysis to determine profiles from HLQ (Section A). Missing data for Section B¤ (Profile 1 n = 5, Profile 2 n = 7, Profile 3 n = 7); Missing data for Section C* (Profile 1 n = 5; Profile 2 n = 12; Profile 3 n = 7); Missing data for section Dø (Profile 1 n = 7, Profile 2 n = 13, Profile 3 n = 7)

Health literacy profiles

The cluster analysis identified three health literacy profiles (Fig. 2 for standardised data, Table 3 for actual scale scores), highlighting different strengths and gaps across the health literacy domains. The majority of parents were in the moderate health literacy group (profile 2, n = 68, 54%), about a third with a higher health literacy (profile 1, n = 38, 30%) and the remaining parents in the lower health literacy group (profile 3, n = 20, 16%). The three profiles varied across most health literacy domains. However, the profiles for scale 5 (information appraisal) were similar; as were profiles 1v2, and 2v3 for scale 3 (actively managing health); and profiles 2v3 for scale 9 (having enough information). Differences were evident with respect to the relationships between the caregiver and clinician (e.g., feeling understood and supported by healthcare providers (scale 1), ability to actively engage with the healthcare provider (scale 6), health system navigation (scale 7), social support for health (scale 4), and ability to find enough information to manage the child’s health (scales 2 and 8). Figure 3 illustrates these profiles and how they relate to one another.

Fig. 2.

Health literacy profiles of parents of children with SHCN. Note: Scale 1 = feeling understood and supported by healthcare providers; Scale 2 = having sufficient information to manage my (child’s) health; Scale 3 = actively managing my (child’s) health; Scale 4 = social support for health; Scale 5 = appraisal of health information; Scale 6 = ability to actively engage with healthcare providers; Scale 7 = navigating the healthcare system; Scale 8 = ability to find good health information; and Scale 9 = understand health information well enough to know what to do. Scores for scales 1–5 (out of 4) have been standardised to a common /5 scale to facilitate visual comparisons across scales with different response formats; raw mean scale scores (± 95%CI) shown in Table 3. Scores for scales 6–9 are out of 5. Profile 1 n = 38, profile 2 n = 66, profile 3 n = 20. Error bars denote standardised 95% confidence interval

Table 3.

Health literacy scale scores (raw)

	Scale 1	Scale 2	Scale 3	Scale 4	Scale 5	Scale 6	Scale 7	Scale 8	Scale 9
Profile 1	3.8 (± 0.08)	3.5 (± 0.13)	3.6 (± 0.11)	3.4 (± 0.14)	3.2 (± 0.18)	4.3 (± 0.14)	3.2 (± 0.17)	4.0 (± 0.13)	4.4 (± 0.14)
Profile 2	3.1 (± 0.10)	2.8 (± 0.12)	3.5 (± 0.10)	2.5 (± 0.13)	3.1 (± 0.11)	3.3 (± 0.13)	2.5 (± 0.11)	3.6 (± 0.14)	4.0 (± 0.12)
Profile 3	2.3 (± 0.23)	2.1 (± 0.30)	3.3 (± 0.17)	1.7 (± 0.28)	2.9 (± 0.21)	2.2 (± 0.25)	1.4 (± 0.13)	3.0 (± 0.28)	3.8 (± 0.36)

Fig. 3.

Example scenarios for each health literacy profile

There were no statistically significant differences in profile membership for the level of parent education, type of household, insurance status, adequacy of transportation or age of the child (Table 2).

Relationships between the health literacy profiles, levels of health care integration, and parents’ perceived stress

Membership of the health literacy profiles was associated with different experiences of health care (Fig. 4). The high health literacy group (profile 1) experienced better health care integration across all the measured domains (mean = 3.3, range = 2.2–4.6) when compared to the moderate and low health literacy profiles 2 (mean = 2.1, range = 1.2–3.3), and 3 (mean = 1.4, range = 0.5–2.9) respectively. Profile 2 had higher scores for the care integration domains of system information, access to care, and team functioning than profile 3. This suggests the profile 2 group were more likely to have received information about other services and supports that their child is eligible for, and experienced less difficulty or delay in getting services for their child compared to profile 3. Similarly, their perception of how well the care team was functioning was greater, indicating that their child’s care team could have been working in a cohesive manner.

Fig. 4.

Components of health care integration experienced by different health literacy profiles. Note: PICS health care integration domain scores have a theoretical range of 0–5. Scores above depict the mean domain score for each profile group. Error bars represent 95% confidence intervals. All post-hoc comparisons for Profile 1 vs. Profile 2, and Profile 1 vs. Profile 3 were significant with p < 0.05. * denotes a significant difference between Profile 2 vs. Profile 3 (p < 0.05)

The perceived impact of the child’s condition on the family was similar across all parent health literacy profiles. However, there were significant differences between perceived stress of the group with poorer health literacy (profile 3) compared with those with stronger health literacy (profiles 1 and 2), suggesting that health literacy could be protective against stress (Table 2).

Discussion

This study provides a snapshot of the health literacy strengths and gaps in a population of Australian parents of children with SHCN. Data suggested three health literacy profiles where membership to the stronger health literacy profile group was associated with better experiences of care coordination and less perceived stress. Parents self-identified whether their child met the requirements of having a SHCN and there was a wide range of chronic health conditions and disabilities in our sample. The majority of children were reported to have problems maintaining their usual activities, with pain, and their mental health. The children varied widely with respect to the number of health professionals involved in their care over the previous year.

Approximately a quarter of the children could be considered underinsured (i.e., insurance not likely to cover child’s health needs), which is a lower fraction than has been previously reported elsewhere in a similar SHCN population in the US [49]. This could be problematic because although Australia’s universal health system provides a level of health coverage for children with SHCN, it is a hybrid model [50] funded via government through Medicare, private health insurance, and/or the family. Many of the children in the current study had multiple specialists involved in their care, but we did not identify whether this was via public or private health services. Many programs that manage chronic and/or complex health conditions are only available within the public system. Access to private healthcare, either through adequate insurance or through additional financial cost to the family, could reduce family frustration and anxiety associated with long waiting times [51] as well as burden on the public system.

The majority of participants used primary care services, but it is important to recognise that many families would still experience an out-of-pocket cost to attend the GP. Rather than utilising primary care, ED presentations may be a consequence of this out-of-pocket cost [52]. Despite the heterogeneity of the causes of the SHCN, the majority of children had recent ED visits and hospitalisations. ED utilisation may be driven by low parent health literacy [53], needing doctors out-of-hours [54], or in the absence of appointment availability in primary care, due to the perception of ED having capacity to provide immediate clinical care and parental anxiety [55]. Future research should consider the health care costs to the family, relative urgency of ED presentations for children with SHCN, and the cost-effectiveness and utility of urgent care health services for this cohort. Interventions to build parent health literacy can reduce ED presentations of children with SHCN [52, 55]. These interventions help parents develop and implement health care plans for their child’s respiratory health and give clear instructions on effective management if the child’s health deteriorates. We must consider what other conditions would benefit from explicit instruction about proactive management.

Health literacy profiles

We identified three health literacy profiles within our sample. This is a low number of profiles, in comparison with other studies using a National population across different Australian states (e.g., 11/12 profiles) [47] or diagnosis groups (e.g., 15 profiles for people with Motor Neurone Disease [MND], seven profiles for carers of people with MND) [56] although the latter study included three digital health literacy domains together with the HLQ domains reported here. However, we selected a set of profiles that were clearly distinct in shape and did not simply illustrate parallel between group differences. We acknowledge that our relatively small sample size of 126, in contrast to sample sizes of 1018/923 [47] and 171/56 for the MND studies [56], may yield different profiles if replicated in a larger sample with SHCN. Profile groups within this study were heterogeneous in terms of causal factors of the child’s SHCN, complexity of the child’s needs and child age. Unlike Tschamper and colleagues’ [57] investigation of parents of children with epilepsy, profile membership was unrelated to parent education in this study, likely due to the relatively small sample and recruitment strategy. Future studies should expand the sample size to ensure representation across the population, and consider the influence of factors such as type of condition, time since diagnosis, and complexity of health, disability and social care needs.

The profiles also demonstrated opportunities to develop more tailored supports to specific cohorts who have poorer health literacy [47, 57] particularly support for development of relationships with health care professionals (scale 1), strengthening supports (scale 4), or building better referral pathways to support service navigation (scale 7). Similarly, there is opportunity for health services to recognise how their policies and processes influence their responsiveness to the health literacy needs of their population, termed organisational health literacy [58]. For example, explicit strategies to connect people with new health services, such as primary care [59], or provision of warm handovers from one health professional to another, may improve the uptake of appropriate health services in different sectors [60].

Interestingly, most health literacy domain scores in this study were similar to those identified in the general Australian adult population sample [47], except for the domain of navigating the healthcare system where even the high health literacy group scored substantially lower than the general population. Although paediatric health services are often considered cohesive, fragmentation and silos exist [2, 16], and all our families found system navigation a relative weakness. Navigation supports (e.g. hospital service websites) vary across jurisdictions, and potentially could be improved to build understanding of what different components of the health service do. Parents often need to respond to significant care needs of their child when on a diagnostic odyssey [61] or post diagnosis when needs may be acute [23, 62]. Care coordination for children with complex health needs has been shown to effectively support families [2]. However, other system- and service-level responses to identify opportunities to improve health literacy and pathways to care are required, as outlined by Miller and colleagues [63]. While there is limited understanding of effective programs to build health literacy in parents of children with developmental conditions [64], there are promising approaches using nurse educators [65] and nurse navigation [66, 67].

Health literacy and care integration

Health literacy was associated with how healthcare was experienced by the families, with those with high levels of health literacy generally experiencing more integrated healthcare for their child. Typically, these families were communicating effectively with the care team, and being provided with care that could adapt to changing needs. Despite limited information about the broader system, they were not experiencing significant delays in being able to access the care and support their child needed. Consistent with other literature, these parents appeared to have an active role in facilitating care integration [8], coordination and communication [30], and advocating for their child’s needs and medical safety [18].

In contrast, the group with poor health literacy were less likely to report receiving information about other services or supports that might be helpful, and experienced greater delays in accessing care. They were also less aware of how the care team functioned together, in terms of knowledge of what other members were doing, following through on their responsibilities and thinking about the child’s health beyond their responsibility. This may be due to a lack of responsiveness by the health services and health professionals, not recognising the diverse support needs of their families with lower health literacy. A benefit of integrated care is the feeling that the care team is partnering with the parents to advocate for their child’s needs, thus developing trust between health professionals and the family [68]. However, this was one of the key health literacy domains where the families with poorer health literacy struggled.

Health literacy and perceived stress

Our findings concur with those of Michou and colleagues [69], with lower health literacy associated with higher levels of perceived parental stress. Better health literacy may be associated with lower perceived stress when managing the day-to-day health of a child with SHCN, and dealing with the various services involved in that management. Given that families of children with SHCN are generally experiencing stress and poorer quality of life [70], a key component of support could focus on building parent health literacy. The relationship between perceived stress and each of the domains of health literacy warrants further exploration. Similarly, there is also a need to consider wellbeing and family functioning in more detail, particularly for families of children with developmental disabilities [71].

Future approaches to building capacity within families

Many of the parents in this study were navigating health, mental health, and disability sectors for their child, yet there were differences with respect to how they experienced services. Irrespective of the health conditions being managed, it is common for families of children with SHCN to have unmet information and health care needs [72], and regularly experience inequity within the broader service system [73]. Developing interventions targeting specific domains of health literacy will likely help address any inequities, and build capacity within families to better navigate and engage with the services their child needs, as a critical component of effective models of care [19]. Rather than providing generic supports, our findings suggest that a nuanced approach to service and support provision should consider the different health literacy profiles of the families, and their different support needs. Future research should also consider health literacy within the context of specific population groups such as, parents of children with intellectual disability, life-limiting illness, or rare diseases. This would inform the design of services and supports to meet the specific needs of these families. Co-designed with consumers, development of interventions to build knowledge and skill of proactive management of the child’s health would also be beneficial. This could include detail of when and how to seek support, and assistance with healthcare navigation.

Limitations

While this study provides a snapshot of Australian families with children with SHCN, our participants were not representative of the broader population as they were likely more educated and more often living in a two-parent household. Similarly, we did not ask about ethnicity, nor language proficiency, hence our findings are not specific to any particular group within the general population. Our respondents were predominantly female, so findings cannot be generalised to the health literacy of fathers. This was an initial exploratory study and formal statistical fit indices were not used to determine the optimal number of health literacy profiles; selection was instead based on visual inspection of within-profile homogeneity and between-profile distinctiveness of domain scores, which introduces an element of subjectivity into profile membership. We measured parent’s health literacy, assuming that this influences their ability to manage their child’s health, but did not directly measure the parent’s effectiveness in managing their child’s health. As this study used a cross-sectional design with unadjusted comparisons between health literacy profiles, causal relationships between health literacy and health care experiences or parental stress cannot be inferred; findings reflect associations only. As comparisons between profiles were unadjusted, it is also possible that other characteristics differing between profiles, rather than health literacy itself, contributred to the observed differences in health care integration and perceived stress. Typically, respondents had older children and so likely had more experience in supporting their child’s health. We had a small number of children aged under four and suggest that a larger sample size of parents with younger children may bring more detail regarding the early stages of identification and diagnoses of SHCN. Similarly, our sample included families with children with developmental conditions and/or chronic health conditions and the nuances of health literacy that are relevant to specific SHCN conditions may not have been adequately captured. It is also possible that some perceived experiences of elements of health care integration could be related to a lack of awareness and understanding of how care teams may work behind the scenes. The attrition rate between sections of the survey may have contributed bias to our findings.

Conclusions

Parents of children with SHCN have specific health literacy strengths along with skill and knowledge gaps, as demonstrated by the different health literacy profiles identified within our study. Parents’ health literacy skills appear be associated with how health care is experienced, highlighting that health services may not be sufficiently responsive to the individual needs of families with lower health literacy. Parents with lower health literacy profiles were experiencing greater perceived stress. Importantly, our study showed that health literacy is not immediately apparent by common demographic variables or the child’s SHCN. Therefore, it is critical that we consider parent health literacy when designing models of care [19] and determining eligibility for supports. This may support more euitable care delivery [31], with the potential to improve health outcomes for children with SHCN and their families.

Abbreviations

ADHD

Attention Deficit Hyperactivity Disorder

ED

Emergency Department

EQ-5D-Y

EuroQol 5 Dimensions – Youth

GP

General Practitioner

HLQ

Health Literacy Questionnaire

MND

Motor Neurone Disease

PICS

Pediatric Integrated Care Survey

PSS-4

Perceived Stress Scale 4

SD

Standard Deviation

SHCN

Special health care needs

US

United States

Author contributions

R.S., R.C. and J.D. conceived the study. P.J., J.D. and R.S. performed the analysis. All authors were involved with synthesizing the findings. R.S. and Z.Y. prepared the figures and tables. R.S. drafted the manuscript text. All authors reviewed the manuscript.

Funding

This work was funded by the Australian National Health and Medical Research Council (GTN2015902). J.D. is supported by a Fellowship from the Stan Perron Charitable Foundation.

Data availability

Data are available upon reasonable request from the corresponding author.

Declarations

Ethics approval and consent to participate

Ethics approval for this study (2021/ET000902) was provided by the University of Western Australia Human Research Ethics Committee in accordance to the Australian National Statement on Ethical Conduct in Human Research. This study adhered to the Declaration of Helsinki. Informed consent to participate was obtained from all participants.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.