Objective To identify the effectiveness of information resources to help parents decide when to seek medical care for an acutely sick child under 5 years of age, including the identification of factors influencing effectiveness, by systematically reviewing the literature.
Methods 5 databases and 5 websites were systematically searched using a combination of terms on children, parents, education, acute childhood illness. A narrative approach, assessing quality via the Mixed Methods Appraisal Tool, was used due to non-comparable research designs.
Results 22 studies met the inclusion criteria: 9 randomised control trials, 8 non-randomised intervention studies, 2 qualitative descriptive studies, 2 qualitative studies and 1 mixed method study. Consultation frequency (15 studies), knowledge (9 studies), anxiety/reassurance (7 studies), confidence (4 studies) satisfaction (4 studies) and antibiotic prescription (4 studies) were used as measures of effectiveness. Quality of the studies was variable but themes supported information needing to be relevant and comprehensive to enable parents to manage an episode of minor illness Interventions addressing a range of symptoms along with assessment and management of childhood illness, appeared to have the greatest impact on the reported measures. The majority of interventions had limited impact on consultation frequencies, No conclusive evidence can be drawn from studies measuring other outcomes.
Conclusions Findings confirm that information needs to be relevant and comprehensive to enable parents to manage an episode of minor illness. Incomplete information leaves parents still needing to seek help and irrelevant information appears to reduce parents’ trust in the intervention. Interventions are more likely to be effective if they are also delivered in non-stressful environments such as the home and are coproduced with parents.
- PUBLIC HEALTH
This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: http://creativecommons.org/licenses/by/4.0/
Statistics from Altmetric.com
Strengths and limitations of this study
This is the first review of the outcome of information resources which aid parental decision-making utilising systematic search and quality assessment criteria.
The strengths of this review lies in its inclusiveness. Using an integrative narrative approach enabled us to identify influences on effectiveness across a wider range of studies and topics than would have been possible with a single study type or topic focused review.
The findings are limited by the quality of the studies and not being able to control for the impact of different healthcare delivery systems.
Acute illness is a universal experience for children and families and represents the most common type of illness in childhood, particularly in 0–5-year-olds. Acute illness includes short-term illnesses, predominantly infections such as coughs, colds, diarrhoea, vomiting and ear infections. Home management is often supported by consultations in primary care, where children under 5 years old constitute 40% of general practitioner (GP) workload,1 with most consultations for acute illness.2 ,3 Under 1-year olds are seen more often than all other age groups other than the over 75s2 and urgent care and emergency department service use by young children appears to be rising.4–6
Parents’ anxiety about acute childhood illness leads them to seek information to help them decide whether or not to seek help from a healthcare professional.7–11 A wide range of information is available for families, such as written leaflets or via websites much of which is either unknown to parents5 ,7 or does not seem to be making any impact on service use when children are acutely sick at home.11–14 The increase in consultation rates for non-urgent care4–6 suggests more effective information sources are needed.
We aimed to systematically review the literature to identify the effectiveness of information resources to help parents decide when to seek medical care for an acutely sick child under 5 years of age, including the identification of factors influencing effectiveness.
Our research questions were:
What measures of effectiveness have been used to evaluate such interventions?
How effective are existing interventions in helping parents know when to seek help for an acutely sick child at home?
What factors influence effectiveness of information provision to help parents know when to seek help for an acutely sick child at home?
We systematically searched five electronic databases (MEDLINE, CINAHL, PsycNET, ASSIA Web of Knowledge) and five websites (Centre for Review and Dissemination York, National Institute for Health and Care Excellence, Health Technology Assessment programme, NHS Evidence and the Cochrane Library) using a combination of terms on children, parents/carers, education, acute childhood illness (see online supplementary appendix 1). We scanned reference lists of key articles, and attempted to contact authors when further information was required to determine eligibility and inform quality assessment.
Studies which met all the following criteria were included:
Studies which included children from 0 to 14 years with research participants being their parents or caregivers. Initial pilot searches aimed solely at children under 5 years yielded minimal results.
An educational intervention on acute childhood illness was provided to parents/caregivers in any form (written, visual, verbal or electronic) designed to help with decision-making about whether or not to seek medical help.
The study was conducted in primary care, emergency departments, ambulatory settings or in the home, in high income countries as defined by Organisation for Economic Co-operation and Development (OECD). We included all study types.
Studies were excluded if they focused on chronically ill children, hospital inpatient settings or educational interventions designed for health professionals. We limited our search to papers published in the English language, between January 1990 and June 2014 (inclusive). The decision to search from 1990 was taken pragmatically as health services have evolved considerably since the latter half of the 20th century. We did not exclude studies on the basis of quality alone but have noted the quality of studies when discussing their impact. To have excluded low quality studies would have reduced the comprehensiveness of the review, especially given the likely heterogeneity of study design.
The titles and abstracts of studies identified in the search were retrieved and assessed by one reviewer who excluded those that were clearly not relevant. The full text of remaining studies was assessed for inclusion by two reviewers; discrepancies were resolved by discussion between all authors. Reasons for exclusion were recorded (see online supplementary appendix 2).
Data extraction and quality assessment
Data from included studies were extracted by one reviewer and checked by a second reviewer. All studies which met the inclusion criteria were included regardless of quality, which was assessed independently by two other reviewers using the Mixed Methods Appraisal Tool (MMAT).15 This gives a rating between zero stars (lowest quality) and 4 stars (****, highest quality).
The search identified 7863 studies, of which 22 were included (figure 1). Table 1 shows the characteristics of included studies of which there were nine randomised controlled trials, eight non-randomised intervention studies, two qualitative descriptive studies, two qualitative studies and one mixed method study. Thirteen were conducted in the USA, six in the UK, two in Canada and one in Denmark. Parents/caregivers of children aged 0–14 years were included across all studies, with 12 studies limiting inclusion to parents of children under the age of 6 years. Studies were conducted in primary care (9), emergency department/hospital (7), child health clinics (3) and children's health centres (3).
Interventions involved written information in all but one study, which used video alone.18 Written information was augmented by video/slide presentations,19–23 home visits,12 ,24 reinforcement within consultations19 ,23 ,25–28 or was part of a structured educational programme.29–31 Three separate studies reported on the same ‘Baby Check’ intervention in different settings/populations.24 ,32 ,33
Quality of included studies is summarised in table 1, and detailed in online supplementary appendix 3. Only two studies were given the highest quality score, with many being given low scores, often due to insufficient reporting of methods.
Measures of effectiveness
The most frequently used measures of effectiveness were: consultation frequency (15 studies), parent knowledge (9 studies), parent anxiety/reassurance (7 studies), parent satisfaction (4 studies), parent confidence and clinician antibiotic prescribing (both 4 studies).
Six of the 15 studies which measured this outcome showed a significant reduction in either actual consultation rates or intention to consult in the future (see table 2). Three of these studies evaluated effects on consultation rates over a longer (1–3-year) period postintervention and found persistence of effect.29 ,34 ,35 (2 low and 1 high quality). One study (low quality) showed a reduction in home visits but with an increase in out-of-hours visits.35 The eight remaining studies on consultant frequency showed no difference on consultation rates with the specified intervention.
Nine studies assessed the effect of interventions on parental knowledge of childhood illnesses including fever, upper respiratory infections, febrile convulsion and otitis media (see table 3). Most interventions used multiple methods to provide information, such as written materials supported by verbal explanations (one high-quality study).12 ,19 ,22 ,23 ,27 ,28 ,36 Timing of outcome measurement ranged from immediately to 32 months later. Eight studies (one high quality) found a significant increase in parental knowledge after interventions18 ,19 ,22 ,23 ,27 ,28 ,31 ,36 with a spread of 24 h to 12 months for postintervention reassessment. One (high quality) study showed reduction in knowledge at 7 months.12
Of the seven randomised controlled studies that reported this outcome (table 4), only one reported significantly reduced concern compared with control group following intervention26 (2* quality rating). Using Baby Check to score their baby's illness reassured 41% (14/34)32 and 46%24 of parents, respectively. In Herman and Jackson's29 (high-quality) study the percentage of parents reporting that they were ‘very worried’ when their child was sick reduced by one-third.
Four studies assessed the effects of interventions on parent's satisfaction with their communication with health professionals,19 ,25 and with the educational information received.27 ,37 Two studies reported non-significantly increased satisfaction in control and interventions groups19 ,25 (one high quality), while another reported significantly increased satisfaction for both intervention groups compared to controls27 (2* quality). The fourth study suggested a web-based self-triage tool would be well received by parents37 (low quality).
Two of four studies12 ,19 (one high quality) measuring the effect of interventions on parents’ confidence in managing childhood illness at home did not show an increase in levels of confidence. However, Thornton et al's24 (high quality) field trials of ‘Baby Check’ found parents’ confidence in the tool itself increased over time, while Kai's32 (2* quality) qualitative exploration found that parents felt ‘Baby Check’ had increased their confidence to monitor their child and given them ‘moral support’ for their decision to consult a doctor.
Four studies assessed the effect of interventions on antibiotic prescription. Francis et al25 (high quality) found a significant reduction in In antibiotic prescriptions given by clinicians in the intervention group (19.5% intervention vs 40.8% control (95% CI 13.7 to 28.9, p<0.001)); and Stockwell et al31 showed a reduction in the number of parents who sought antibiotics without a prescription or used over the counter medication inappropriately; however this small study (11 parents) failed to report effects on antibiotics sought by parents from health professionals. Two other studies (both high quality)12 ,33 found no significant differences in antibiotic prescribing.
Factors influencing the effectiveness of an intervention
Factors which may have influenced the effectiveness of interventions were identified from a comparison of study populations and/or the setting of the study and the content, format and delivery of the educational interventions.
Content of interventions: range of topics addressed by the interventions
Eleven studies assessed interventions which focused on a single symptom or type of childhood illness alone (such as fever, febrile convulsions, respiratory tract infection, otitis media), while 10 provided information on a range of different childhood illnesses.
Three single-topic studies measured consultation behaviour, of which Francis et al25 found reduced intention to consult in the intervention compared to control group while two did not.18 ,22 Two single-topic studies assessed anxiety/reassurance, one found no effect25 and the other a reduction in intervention and control groups.30 Confidence was assessed in one single-topic study19 which found no effect. Antibiotic prescribing was assessed in two respiratory focused studies,25 ,31 one of which showed a significant reduction in prescribing in the intervention group in the first 2 weeks postintervention25 and the other a non-significant reduction in seeking antibiotics without prescription after the intervention31 (only Francis et al studied rated as high quality).
Four of the 10 studies evaluating the effects of providing information on multiple childhood illnesses or symptoms showed trends towards reduction in consultation rates or intention to consult26 ,29 ,34 ,35 (one high-quality). Four multitopic intervention studies reported a reduction in anxiety or increased reassurance24 ,26 ,29 ,32 (one high quality). Confidence improved in two of the ‘Baby Check’ studies24 ,32 (one high quality) but in another (high-quality) study, there was no effect on confidence.12 Neither of two high-quality multitopic studies demonstrated a significant reduction in antibiotic prescribing.12 ,33
In summary, reduction in consultation rates, reduction in anxiety and increases in confidence appeared more common in multitopic compared to single-topic interventions, while reduction in antibiotic prescribing was more effective with single illness-focused interventions.
Content of interventions: information on assessment and/or management of childhood illness
Four interventions specifically intended to enable parents to assess the severity of their baby's illness and know when to seek medical attention for their child23 ,24 ,32 ,33 (two high quality). One of these interventions (a low quality study) informed parents about fever and home management of fever and found that 90% of parents rated the information helpful in decision-making and as a communication tool.19 In contrast, nearly one-third of parents did not think the ‘Baby Check’ educational tool was useful,24 and a qualitative study of the same tool32 revealed that even when parents scored their child's illness as minor they still consulted for the illness within 24 h after the assessment, because they wanted practical advice on management.
Content of the interventions: accessibility of the information
Many of the papers provided brief descriptions of the strategies used to make interventions easy to understand for parents. Three (one high quality) designed their interventions specifically for parents with low levels of health literacy.29 ,31 ,38 The language used in the ‘Baby Check’ score card was simplified to accommodate low health literacy through the translation of professional terms such as ‘reduced tone’ as ‘floppiness’24 and a further three studies reported that their interventions were designed for age 11–12-year-old reading level.30 ,34 ,39 One study specifically mentioned using cartoons and humour to increase the accessibility of information.34 There was no identifiable relationship on outcomes between studies which did or did not design interventions for easy reading. However, Krantz's38 qualitative study evaluating parents’ views of a fever guide found that parents liked the one page, easy-to-read style, the use of simple diagrams such as a thermometer showing both Fahrenheit and Celsius and pictures of how to measure a child's temperature. Parents felt that these pictures were likely to enhance recall of the information.
Delivery method for interventions: interactive or one-way flow
Six studies provided educational interventions to parents in an interactive manner, that is, the parent could engage with the intervention rather than just receiving information:19 ,23 ,25 ,29–31 ,36 two (high-quality studies) showed significant reductions in consultation rates or intention to consult25 ,29 and four significantly improved parental knowledge19 ,23 ,31 ,36 (low to 2* quality).
Two additional but low to 2* quality studies19 ,26 used a relatively simple non-discursive method to provide information to parents, showing significant reductions in consultations of up to 88% in a comparison of attendances to an emergency department per month 1 year following the intervention. These shared a common feature: when health professionals gave their booklets to parents, they emphasised that the content was important and would help them to look after their acutely sick child. These findings intimate that educational interventions can be successful even when they are provided using a simple method, but clearly further studies are needed to demonstrate this.
None of the four interventions which were delivered in the waiting room of an emergency department18 ,20 ,21 ,30 (one high quality) had significant effects on consultation rates, anxiety or parental knowledge. These studies involved both single topic and multitopic interventions with varying delivery mechanisms and suggest that it is the environment in which the intervention was delivered which is associated with effectiveness, rather than the content of the intervention itself.
Two US studies29 ,31 took place in children's health centres: one high-quality study reduced consultation rates in local emergency departments and primary care29 and the other improved parental knowledge.31 Peer support and a trustworthy environment were two important factors suggested by the authors as related to this success.
Parent involvement in intervention development or evaluation
One high-quality study involved parents in the development25 and four in the evaluation of the educational intervention.19 ,26 ,29 ,35 Four showed reduction in consultation rates, intention to consult, or improved parental knowledge.19 ,25 ,26 ,29 In comparison, studies using existing educational materials as their intervention, without modification and evaluation by its target population, were less successful12 ,33 (both high quality).
This systematic review and synthesis of information resources intending to help parents decide when to seek medical help for an acutely sick child identified measures of effectiveness used to evaluate interventions, as well as factors which appear to influence the effectiveness of interventions. Unlike previous reviews which focused on interventions specifically for respiratory tract infections40 or acute paediatric hospital admissions,41 our review was broader as we identified factors influencing effectiveness of interventions on parents’ help-seeking behaviour for all common acute illnesses at home.
Measures of effectiveness
Consultation frequency, knowledge, reassurance/anxiety, satisfaction, confidence and antibiotic prescribing were used as measures of effectiveness. Studies which found reductions in consultation rates27 ,29 ,34 were all conducted in the USA, which may reflect differences in health service delivery systems and possible financial costs associated with unscheduled consultations. These differences in parental motivations may limit applicability in other countries such as the UK where direct parent-incurred health service costs are less relevant.
Results from studies measuring parents’ knowledge of acute childhood illness indicate that when both verbal and written information were provided, parents were more likely to retain knowledge in the long term than when only given written information.19 ,22 ,23 ,28 ,31 ,33 ,37 Verbal reinforcement may signal to parents that health professionals endorse the information.
Providing information did not seem to be directly linked to increased satisfaction, although it is not clear whether the studies we found used a valid measurement tool. Limited information was available about the methods used to measure parent satisfaction, which included a question over the phone,27 or using one or two items within a rating scale administered by phone.19 ,25 Satisfaction is a complex phenomenon and it is therefore unlikely that such simple measures will elucidate factors which influence it. No conclusions can be drawn regarding the impact of interventions on parents’ confidence to care for their child.
The effectiveness of interventions at reducing antibiotic prescriptions mirror those of Andrews et al's40 review of interventions specifically focused on reducing consultation and antibiotic use in respiratory tract infection, which found that educational materials reduced consultation rates by up to 40%. The two respiratory focused studies which we identified, one from the UK and one from the USA, both indicated a reduction in antibiotic use, while neither of the less focused interventions demonstrated any effect on antibiotic use.
We were unable to easily identify an intervention which works consistently to reduce consultation rates, to improve parents’ knowledge, confidence or satisfaction.
Factors influencing the effectiveness of an intervention
Interventions providing information on multiple childhood illnesses or symptoms appeared to be more effective (eg, reduction in consultation rates or intention to consult, reduction in anxiety or increased reassurance), compared to interventions addressing single symptoms. This may be because common childhood symptoms, such as fever, cough, sore throat, vomiting and diarrhoea, often occur simultaneously. Therefore, although parents receiving fever education may feel more competent in managing fever, they may continue to seek a medical consultation for other symptoms about which they have less knowledge or confidence. Moreover, educational material which addressed the assessment of illness severity as well as management of minor illness appear to be more effective in supporting parents to care for their children and seek help when necessary: if information is only provided on assessment this may still leave parents needing advice about how to manage, even minor, illness.
Parents’ involvement in the development of educational interventions may improve effectiveness. These findings support the general trend towards involving patients and the public in research,4 emphasising the importance of working collaboratively with the end users of interventions.
O'Neill-Murphy et al30 argued that information provided in an interactive method is more effective in improving knowledge than non-interactive methods. However, our findings do not clearly support this position as we noted significant effects for interventions delivered with, and without, interaction. Involving health professionals in the distribution of booklets, with or without an interactive discussion, may increase the perceived value and reliability of the information and motivate parents to read the booklets, trust the home management strategies suggested and, finally, impact on their behaviour. Parents have previously been found to trust information from doctors more than that from other sources.9
Studies in the review were conducted in a range of settings; those conducted in emergency departments were the least effective.18 ,20 ,21 ,30 Having an acutely sick child is a stressful time for parents, generating considerable anxiety and uncertainty about when to seek medical help.5 ,9 ,11 Stress can impair learning,42 ,43 therefore it is not surprising that in Chande et al's study only 65% of participants in the intervention group remembered the video in the emergency department. However, two US studies29 ,31 conducted in children's health centres showed reduction in consultation rates in local emergency departments and in primary care29 and improved parental knowledge.31 We do not know whether interventions delivered in children's centres would similarly work in the UK, although community education on childhood illness has been suggested in a recent UK survey of parents’ first contact choices.43
Strengths and limitations
The strengths of our review lie in its inclusiveness. Given the non-comparable research designs, we used an integrative narrative approach, recognised as an effective method for summarising and synthesising findings across multiple study designs.16 ,17 This approach enabled us to identify influences on effectiveness across a wider range of studies and topics than would have been possible with a single study type or topic focused review. This comprehensive strategy does result in the inclusion of low quality studies whose impact may be questioned and means our recommendations need to confirmed in further studies.
It is possible some studies were missed as the screening of titles and abstracts for inclusion was performed by only one person. The highly heterogeneous nature of the included studies in terms of design, as well as interventions, outcomes measured, populations and settings limited our ability to perform more quantitative syntheses. The literature search was of papers published in English since January 1990. However, it was evident that some of the earlier included studies are already of limited direct relevance to contemporary health services. For example, the ‘Baby Check’ tool used in three studies included a requirement for parents to measure rectal temperature, which is no longer recommended practice. Also no studies compared differing healthcare delivery systems; health systems are likely to have implications on the impact of different interventions.
Recommendations for clinical practice: how best to provide information to help parents decide when to seek help for an acutely sick child
Our findings indicate that interventions with the following characteristics are more likely to be effective:
Comprehensive information on childhood illness;
Information on assessment of children's need for a medical consultation and on how to manage minor illness at home;
Reinforcement or support by local healthcare professionals;
Delivery away from the stressful environment of the emergency department. This could be in primary care, in the home or in social care settings;
Coproduction with parents.
Even without the development of new materials for parents of acutely ill children, there are messages here for clinicians using existing materials. Clinicians need to select resources which provide information on multiple common symptoms of childhood illness. Evidence from focus groups parents indicates development with parents is good practice. Interventions in this area can have unexpected consequences which need to be considered prior to implementation, as, for example, one primary care-based intervention which resulted in shifting consultation from day time home visits to the out of hours service.35
Information is best provided in primary care or social care settings. Community centres such as SureStart Children's Centres in the UK provide a potential route for the delivery of health information by health professionals, such as health visitors.
Directions for future research
Most of the studies included in the review were quantitative, providing valuable information on the effects of educational interventions. More qualitative studies are needed, which are able to provide in-depth understanding about what, how and why interventions affect parents’ abilities to assess and manage acute childhood illnesses. This information should be underpinned by research which identifies both parents’ and health professionals’ current use of information resources, and their views on how these resources need to be developed. Finally it is important that any future interventions for parents should be co-developed with parents themselves.44 ,45 Given the rising rates of consultations and the considerable impact this is having on the health service in the UK, as well as on parents, there is a pressing need for larger scale implementation studies taking into account the findings of this review.
Overall, the majority of reviewed interventions had limited effects on consultation rates. Although many studies showed an improvement in parental knowledge of childhood illness, this did not necessarily lead to more confidence and less anxiety in parents when looking after their child at home. Interventions providing comprehensive information on childhood illness which can be used for both assessing children's need for a medical consultation and for managing minor illness at home were more effective in reducing consultation rates than those focused on a single symptom/illness or only on assessing the child's level of acuity. Interventions also appeared more effective if parents were involved in their development or evaluation.
The authors thank Dr Chenyu Shang, for the early work searching and reviewing the literature.
Contributors SN, ML, CHDJ and MT conceived the original idea. Initial data searching was performed by SN and quality analysis undertaken by all authors. SN prepared an initial manuscript which DR revised. All authors contributed to the final version.
Funding This publication presents independent research funded by the University of Leicester. The views expressed in this publication are those of the author(s) and not necessarily those of the University of Leicester. MT and CHDJ: This report is independent research arising from MT’s Career Development Fellowship supported by the National Institute for Health Research. The views expressed in this publication are those of the author(s) and not necessarily those of the NHS, the National Institute for Health Research or the Department of Health.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement No additional data are available.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.