If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address correspondence to Jean L. Raphael, MD, MPH, Clinical Care Center Suite D.1540.00, Texas Children’s Hospital, 6701 Fannin St, Houston, TX 77030.
Department of Pediatrics, University of Washington, Seattle, WashCenter for Child Health, Behavior and Development, Seattle Children’s Research Institute, Seattle, Wash
Children with chronic diseases represent a high-cost and resource-intensive population of children. With continued gaps in chronic disease management and persistent fragmentation in the health care system, stakeholders are seeking new strategies to address the needs of these children.
Objective
We sought to systematically assess the effectiveness of lay health worker interventions in improving health care utilization, symptom management, and family psychosocial outcomes for children with chronic conditions.
Data Source
PubMed, PsycINFO, and Web of Science (January 1961 to February 2013).
Study Eligibility Criteria, Participants, and Interventions
We developed a strategy to search citations to identify relevant articles. Search terms included randomized controlled trial (RCT), lay worker, parent mentor, peer mentor, peer educator, community health workers, community health aids, patient advocate, patient facilitator, patient liaison, promotoras(es), care ambassadors, patient navigator, and nonprofessional. Additional studies were identified by searching the reference lists of retrieved articles and contacting clinical experts. RCTs of lay health worker interventions for children with chronic conditions were included. Studies were restricted to those concentrated on children 0–18 years of age with chronic illnesses.
Study Appraisal and Synthesis Methods
Abstracts were independently screened by 2 reviewers. Articles with relevant abstracts underwent full text review and were evaluated for inclusion criteria. A structured tool was used to abstract data from selected articles. Because of the heterogeneous interventions and outcomes, we did not conduct a meta-analysis.
Results
The search yielded 736 unique articles, of which 17 met inclusion criteria. All interventions focused on specific conditions: asthma, type I diabetes, obesity, and failure to thrive. Interventions were heterogeneous in frequency, mode, and duration of interactions between lay health workers and subjects. Several interventions were multifaceted, including both one-on-one and group interactions. Improved outcomes most commonly reported were reduced urgent care use, decreases in symptoms, fewer missed work and school days, and increased parental quality of life. One study demonstrated that lay health worker interventions were cost-effective.
Conclusions
Lay health workers interventions in children with chronic conditions may lead to modest improvements in urgent care use, symptoms, and parental psychosocial outcomes. Such interventions may also be cost-effective. Future research should focus on interventions targeted toward other chronic conditions such as sickle cell disease or cystic fibrosis and medically complex children whose conditions are noncategorical.
Lay health worker interventions lead to modest improvements in urgent care use, symptoms, and parental psychosocial outcomes.
•
Lay health worker interventions may lead to cost savings.
•
Lay health worker intervention effects may persist after the intervention ends.
How to Use This Systematic Review
•
Compare and contrast different models of lay health worker interventions.
•
Identify outcome measures most relevant to lay health worker interventions.
•
Findings may be applied to designing interventions or incorporating aspects of lay health worker interventions into practice.
Children with chronic diseases represent a high-cost and resource-intensive group of children. The prevalence of chronic conditions in children has quadrupled in the past 4 decades, now comprising 12–16% of the pediatric population.
What is the prevalence of children with special health care needs? Toward an understanding of variations in findings and methods across three national surveys.
The impact of pediatric chronic diseases on child health-related cost is enormous with respect to health care utilization, school and work absenteeism, and family functioning.
Minority and low-income children with chronic diseases are particularly vulnerable as they experience inequities in access to care, utilization, unmet medical needs, and patient satisfaction.
Satisfaction with care and ease of using health care services among parents of children with special health care needs: the roles of race/ethnicity, insurance, language, and adequacy of family-centered care.
There is mounting concern that the current model of pediatric primary care is not equipped to support the provision of high-quality care for chronically ill children.
With continued gaps in chronic disease management and persistent fragmentation in the health care system, policy makers, clinicians, and other stakeholders are seeking new strategies to promote partnership models of care in which patients take a more active role in their own management.
Increasing evidence suggests that patients with chronic diseases may potentially benefit from individualized assistance to navigate the health care system.
Lay health worker interventions (eg, community health workers, patient navigators) for chronic disease have emerged as innovative models of individualized care. Lay health workers are individuals who perform functions related to health care delivery.
Because they have no formal or paraprofessional training, they are typically provided with informal job-related training. They may work in paid positions or as volunteers. The term lay health workers is broad in scope and could include community health workers, village health workers, and cancer supporters. Although heterogeneous in specific content and delivery, lay health workers programs provide flexible and tailored one-on-one guidance to individuals. Previous studies among adults indicate that such interventions improve diabetes knowledge, hypertension management, timeliness of cancer care, health care utilization, and patient satisfaction.
Although numerous individual studies have assessed lay health worker interventions in pediatrics, there is little consensus about the effectiveness of such programs for children with chronic conditions.
For example, a prior review of lay health worker interventions among children narrowly focused on asthma with heavy emphasis on environmental controls. Evidence for the effectiveness of these interventions among a broad spectrum of children with chronic conditions is needed, particularly with respect to health care use, symptom management, and family psychosocial outcomes. The objectives of this study were to summarize the available studies on lay health worker interventions, determine the efficacy of such programs for children with chronic conditions, and identify gaps in current knowledge regarding efficacy.
Methods
Eligibility Criteria
We searched for randomized controlled trials (RCT) of interventions that were delivered by lay health workers to families of children 18 years of age or younger with chronic health conditions. Lay health workers were identified as individuals who were specifically trained to deliver a health-related intervention but who had no formal professional or paraprofessional training in health care. A lay health worker could not be a family member trained to provide care or support exclusively for members of his or her family. Because by definition lay health workers deliver one-on-one guidance, interventions had to have one-on-one interactions between the lay health worker and family. All RCTs were included, regardless of type. We excluded studies that focused on adults, studies evaluating pediatric to adult transition, nonempirical work such as case studies or commentaries, and unpublished literature. From included studies, we extracted data on characteristics, theoretical frameworks, training, intervention types, outcomes, and cost-effectiveness.
Information Sources
We conducted a systematic search of 3 major electronic databases comprising medical and social science studies (PubMed, PsycINFO, and Web of Knowledge) for titles and abstracts relevant to our research question (January 1961 to February 2013). We additionally hand searched bibliographies from retrieved articles and from published reviews. Local experts in behavioral interventions were consulted for additional studies. We did not contact study authors for further details with respect to methodology or reporting of results.
Search Terms
On the basis of our inclusion/exclusion criteria, we started with Medical Subject Headings (MeSH) and the following keyword search terms: lay worker, parent mentor, peer mentor, community health worker, and promotoras(es). We added more terms by identifying keywords associated with these searches or within articles found with these searches. The final compiled list of search terms consisted of the following keywords or keyword combinations: lay worker, parent mentor, peer mentor, peer educator, community health workers, community health aids, patient advocate, patient facilitator, patient liaison, promotoras(es), care ambassadors, patient navigator, and nonprofessional.
Study Selection
Publications were screened and selected in a 2-step process in order to minimize bias. In the first phase, publications were screened for inclusion independently by 2 investigators (JR and AR) using titles of articles and, if available, the abstracts derived from the search. The inclusion criteria were RCTs of lay health worker interventions among children with chronic conditions. In the second phase, full text of articles that met inclusion criteria were retrieved and independently reviewed by JR and AR. The investigators met regularly to discuss the design and validity of the studies to determine whether they met inclusion criteria. Any disagreements were resolved by consensus between investigators.
Data Collection Process
We developed a data extraction document based on the Cochrane Consumers and Communication Review Group’s data extraction template,
pilot tested it on 10 included studies, and refined it accordingly. JR and AR independently collected the following data elements from the included studies: design, trial inclusion and exclusion criteria, condition type, population characteristics, setting, theoretical framework, lay health worker training, and intervention type (including duration and frequency).
Assessment of Risk of Bias of Included Studies
The methodological risk of bias of included studies was assessed in accordance with the Cochrane Handbook and the guidelines of the Cochrane Consumers and Communication Review Group,
which recommends the explicit reporting of the following individual elements for RCTs: random sequence generation; allocation sequence concealment; blinding (participants, personnel); blinding (outcome assessment); completeness of outcome data, selective outcome reporting; and other sources of bias. JR and AR independently assessed the risk of bias of all included studies, with any disagreements resolved by discussion and consensus.
Summary Measures
Given that there are no standard outcome measures for lay health worker interventions, we did not prespecify or place limitations on the outcomes examined. We used all outcomes for each study. To minimize bias, we only report results from prespecified analyses and do not include findings of post hoc analyses.
Results
Results of the Search
The literature review yielded 736 unique articles, of which 27 articles met criteria for initial review (Fig.). After abstract review, 10 of the 27 studies were excluded because they did not meet inclusion criteria: 8 articles were eliminated as a result of nonrandomized design and 2 were eliminated because the study design did not include one-on-one interaction with the lay health workers. The final group of 17 studies comprised 4 specific pediatric chronic conditions: asthma, type I diabetes, obesity, and failure to thrive (FTT). The heterogeneity of interventions and outcomes precluded us from conducting any meta-analysis.
FigureFlow of titles, abstracts, and articles included in review.
None of the studies scored low risk of bias in all of the 7 items assessed (Table 1). Six studies were rated high risk in at least one category of risk of bias. These articles were rated low risk or unclear risk in all other categories and were therefore deemed appropriate to include in the review. However, the results of included studies are stratified according to whether or not they rate as high risk in any categories. Certain quality limitations were common in most articles. Three-quarters of the articles did not adequately describe the allocation concealment process. Because of the design of the interventions, which required the lay health workers to have face-to-face contact with the subjects, it was impossible to blind the study participants and personnel. Of all the articles, 8 stated that the outcome assessment was blinded; 2 stated that the outcome assessment was not blinded; and the rest of the articles made no statement regarding blinding of the outcome assessment.
A randomized controlled trial of asthma self-management support comparing clinic-based nurses and in-home community health workers: the Seattle–King County Healthy Homes II Project.
Evaluation of Community Action Against Asthma: a community health worker intervention to improve children’s asthma-related health by reducing household environmental triggers for asthma.
The Seattle–King County Healthy Homes Project: a randomized, controlled trial of a community health worker intervention to decrease exposure to indoor asthma triggers.
included social cognitive theory, self-efficacy theory, and social support theory. All studies were 2- or 3-arm RCTs conducted in urban areas (Table 2), and 3 of the 17 studies were multicenter trials.
The included studies involved 3,806 children 0–18 years old, with the majority of the study populations having a mean age of 5–8 years, and the total study size ranging from 46 to 937 children. The subjects were followed between 3 and 24 months after intervention. Most studies targeted minority populations of low socioeconomic status. Subjects were recruited from schools, outpatient clinics, and emergency departments (ED).
Table 2Study Characteristics
Primary Author (Year), Study Design
N
Target Population
Eligibility Requirements
Recruitment
Asthma
Bryant-Stephens (2009), 2-arm RCT
264
Urban minority, low SES
2–16 y old; physician diagnosed; on controller medication; ≥1 hospitalization or ≥2 ED visits within 12 mo
Self-referral or from physician
Flores (2009), 2-arm RCT
220
Urban minority
2–18 y old; AA/LA; asthma as primary ED or hospital diagnosis;
ED or inpatient service at 4 local hospitals
Krieger (2009), 2-arm RCT
271
Low SES;
3–13 y old; persistent or poorly controlled asthma; income <200% FPL or on Medicaid
Community and public health clinics
Bryant-Stephens (2008), 3-arm RCT
281
Urban minority
2–16 y old; receive primary care at Children’s Hospital of Philadelphia; ≥1 hospitalization or ≥2 ED visits within 12 mo
Age criteria NR; BMI ≥85 percentile; enrolled at 1 of 2 study schools; K–5th grade
School
Failure to Thrive
Black (1995), 2-arm RCT
130
Low-income children
<25 mo of age; weight for age <5 percentile; gestational age ≥36 wk; weight appropriate for gestational age
Pediatric clinics; community health maintenance organizations
AA/LA = African American/Latino American; BMI = body mass index; ED = emergency department; FPL = federal poverty limit; NR = not reported; SES = socioeconomic status.
∗ High risk of bias in 1 or more quality category.
In most studies, lay health workers were chosen on the basis of social congruence with the study population according to race/ethnicity, socioeconomic status, or having a child with the same chronic condition. All of the lay health workers received training before beginning the study intervention, although 7 studies
A randomized controlled trial of asthma self-management support comparing clinic-based nurses and in-home community health workers: the Seattle–King County Healthy Homes II Project.
The Seattle–King County Healthy Homes Project: a randomized, controlled trial of a community health worker intervention to decrease exposure to indoor asthma triggers.
Evaluation of Community Action Against Asthma: a community health worker intervention to improve children’s asthma-related health by reducing household environmental triggers for asthma.
Impact of a household environmental intervention delivered by lay health workers on asthma symptom control in urban, disadvantaged children with asthma.
reported oversight by trained professionals or use of protocols to maintain the quality, accuracy, and consistency of the lay health worker interventions.
Table 3Intervention Characteristics
Primary Author (Year)
Lay Health Worker Training
Content
Mode
Dose/Frequency
Intervention Length
Assessment Frequency
Control Group Study Protocol
Asthma
Bryant-Stephens (2009)
Didactic training with practice assessments
Home education and environmental intervention
HV
Active phase: biweekly visits for 24 wk; inactive phase: monthly visits for 6 mo
12 mo
Monthly for 12 mo
Observation (crossover)
Flores (2009)
Training for 2.5 d; training manual
Asthma education; peer support
HV, phone calls; group meetings
2 HV/y; monthly phone calls
12 mo
Monthly for 12 mo
Observation only
Krieger (2009)
NR
Action plan development; review of educational topics; social support
HV, phone calls
Initial visit plus 0–5 follow up visits (mean 3.1)
12 mo
At 12 mo
Clinic visits with asthma nurses
Bryant-Stephens (2008)
NR
Asthma education; review of medications and asthma action plan; pest control
HV
Assessment HV + 5 weekly follow-up visits, then 1 HV/mo to collect symptom diaries, review medications and action plans
12 mo
At 12 mo
Given information about asthma self-management classes
CA group received information on scheduling clinic visits with monitoring of clinic attendance; CA+ group received 8 written teaching modules over 2 y about type 1 DM care
Sessions in the clinic by CA
CA group, mean of 7.3 visits in 24 mo; CA+ group, mean of 7.5 visits in 24 mo
24 mo
4, 8, 12, 24 mo
Delayed intervention
Laffel (1998)
NR
CA assisted with appt scheduling, monitored clinic attendance and contacted families who missed their appointments
Interventions consisted of lay health worker-family interactions through home visits, phone calls, or e-mails (Table 3). The services provided by lay health workers varied across studies. The lay health workers in the 11 studies of asthma primarily focused on education about environmental trigger reduction, provision of engineering controls (eg, air filters, mattress encasements), asthma action plans, medication management, and increasing parental recognition of symptoms. Two asthma interventions involved monthly group meetings in the community.
The frequency of interactions between the lay health workers and the subjects varied among studies and was inconsistently reported. Most of the studies did not report the duration of individual encounters.
Study Findings
Results were stratified according to the following categories: studies that rated low risk or unclear risk in all risk of bias elements and studies that had any element rated as high risk. Given the heterogeneity of outcomes assessed and analyses performed, standardized measures of intervention effect could not be reported across the different studies. We present the most informative measures of effect and statistical confidence available from the published studies (Table 4).
Table 4Findings
Study (Year)
Psychosocial/Behavior Outcomes
Symptoms/Medication Use
Urgent Care Use
Asthma
Bryant-Stephens (2009)
No difference in knowledge
No difference in nighttime cough, wheeze, or albuterol use
No difference in ED visits or hospitalizations
Flores (2009)
Increased know when serious breathing problem controllable at home; 0.7 unit increase, P = .02; no differences in QOL, asthma satisfaction
Fewer rapid breathing episodes/mo; 57.3 to 41.5, P = .04; fewer asthma exacerbations/mo; 2.9 to 1.8, P = .01; fewer wheezing episodes/mo; 92 to 60, P = .01; fewer missed parental work days/mo; 2.9 to 0.3, P = .01; fewer missed school days/mo; 3.7 to 0.8, P = .03
Fewer ED visits/y; 0.5 to 0.1, P = .03; no differences for hospitalizations, urgent care visits
Krieger (2009)
Increase in mean caregiver QOL score; 5.6 to 6.2, P < .001; no differences in behavior change
Increase in no. of symptom-free days/2 wk; IE 0.94 (0.02–1.86) or 24.4 more days/y; no differences in missed work or school days
Fewer caregiver depressive symptoms; 1.62 to 1.5, P value not reported
Increased nadir PF; IE 8.2 (1.1–15.2); increased nadir FEV1; IE 10.0 (0.9–19.1); fewer bouts of persistent cough; 3.81 to 3.36; decrease in cough with exercise; 4.27 to 3.69; decrease in any symptoms for more than 2 d/wk and no controller medication; IE 0.39 (0.20–0.73)
Less unscheduled urgent care; IE for last 12 mo 0.40 (0.22–0.74); IE for last 3 mo 0.43 (0.23–0.80)
Williams (2006)
Not assessed
Lower asthma functional severity score; 33% to 20%, P < .01
Fewer no. of days with symptoms/2 wk; −0.82, P < .0001 at 12 mo; −0.60, P < .0001 at 24 mo; fewer caretaker nights wakening/2 wk; −0.61, P < .0001 at 12 mo; −0.37, P = .006 at 24 mo; days caretaker plans changed/2 wk; −0.31, P < .0001 at 12 mo; fewer missed school days/2 wk; −0.17, P = .003 at 12 mo; −0.17, P = .009 at 24 mo
Fewer no. of visits/y to ED or clinic; −0.35, P = .04 at 12 mo
Bonner (2002)
Increased asthma knowledge, P < .001; increased self-efficacy for managing asthma, P < .001
Less symptom persistence, P < .01; fewer activity restrictions, P < .01; increased family adherence, P < .001; increased prophylactic use of bronchodilator, P < .05; increased physician pharmacotherapy, P < .001
Increase in father confidence related to managing DM, P = .02; no differences in paternal concern, worry, DM impact, perceived amount and helpfulness of daily management
Not assessed
Not assessed
Sullivan-Bolyai (2010)
No differences in parental concern, worry, confidence, DM impact on family
Not assessed
Not assessed
Svoren (2003)
Not assessed
Lower rate of severe hypoglycemic events 45.4/100 patient-y in CA+ vs 60.6/100 patient-y in SC plus CA, P = .02; fewer severe hypoglycemic events requiring parenteral therapy; 4.2/100 patient-y in CA+ vs 10.5/100 patient-y in SC plus CA, P = .01; increase in risk of worsening glycemic control; SC plus CA 3.4 times higher compared to CA+, P = .002
Increased mean no. of clinic follow-up visits at 24 mo in both intervention groups; CA 7.3, CA+ 7.5, SC 5.4, P = .0001; less hospitalization; 8.9/100 patient-y in CA+ vs 15.3/100 patient-y in SC plus CA, P = .04; fewer ED visits; 21/100 patient-y in CA+ vs 34.9/100 patient-y in SC plus CA, P = .004
Laffel (1998)
Not assessed
Lower rate of severe hypoglycemic events; 10.1/100 patient-y vs 22.5/100 patient-y, P = .009
Fewer ED visits or hospitalizations; 10.6/100 patient-y vs 20.5/100 patient-y, P = .034; increased mean no. of clinic visits/24 mo; 7.1 visits for intervention vs 5.2 for control, P = .0001
No differences in parent confidence; no differences in behavior
No differences in BMI
Not assessed
Failure to Thrive
Black (2005)
Not assessed
No differences in weight
Not assessed
CA = care ambassador; ED = emergency department; FEV1 = forced expiratory volume at 1 second; IE = intervention effect; PF = peak expiratory flow; QOL = quality of life; SC = standard care; BMI = body mass index.
Confidence intervals are provided in parentheses. The term “difference” refers to statistically significant differences for intervention group relative to control.
∗ High risk of bias in 1 or more quality category.
A randomized controlled trial of asthma self-management support comparing clinic-based nurses and in-home community health workers: the Seattle–King County Healthy Homes II Project.
The Seattle–King County Healthy Homes Project: a randomized, controlled trial of a community health worker intervention to decrease exposure to indoor asthma triggers.
Impact of a household environmental intervention delivered by lay health workers on asthma symptom control in urban, disadvantaged children with asthma.
A randomized controlled trial of asthma self-management support comparing clinic-based nurses and in-home community health workers: the Seattle–King County Healthy Homes II Project.
The Seattle–King County Healthy Homes Project: a randomized, controlled trial of a community health worker intervention to decrease exposure to indoor asthma triggers.
Impact of a household environmental intervention delivered by lay health workers on asthma symptom control in urban, disadvantaged children with asthma.
This study also demonstrated significant reductions in asthma exacerbations per year (2.9 to 1.8) in the intervention but not control group. The study by Krieger et al reported an increase in symptom-free days by 24.4 days/year in the intervention group.
A randomized controlled trial of asthma self-management support comparing clinic-based nurses and in-home community health workers: the Seattle–King County Healthy Homes II Project.
The Seattle–King County Healthy Homes Project: a randomized, controlled trial of a community health worker intervention to decrease exposure to indoor asthma triggers.
The Seattle–King County Healthy Homes Project: a randomized, controlled trial of a community health worker intervention to decrease exposure to indoor asthma triggers.
A randomized controlled trial of asthma self-management support comparing clinic-based nurses and in-home community health workers: the Seattle–King County Healthy Homes II Project.
Impact of a household environmental intervention delivered by lay health workers on asthma symptom control in urban, disadvantaged children with asthma.
A randomized controlled trial of asthma self-management support comparing clinic-based nurses and in-home community health workers: the Seattle–King County Healthy Homes II Project.
The Seattle–King County Healthy Homes Project: a randomized, controlled trial of a community health worker intervention to decrease exposure to indoor asthma triggers.
The mean number of work days missed by the parent decreased by 2.6 days. The remaining 2 studies assessing school and work days did not show significant differences between intervention and control groups.
A randomized controlled trial of asthma self-management support comparing clinic-based nurses and in-home community health workers: the Seattle–King County Healthy Homes II Project.
The Seattle–King County Healthy Homes Project: a randomized, controlled trial of a community health worker intervention to decrease exposure to indoor asthma triggers.
A randomized controlled trial of asthma self-management support comparing clinic-based nurses and in-home community health workers: the Seattle–King County Healthy Homes II Project.
The Seattle–King County Healthy Homes Project: a randomized, controlled trial of a community health worker intervention to decrease exposure to indoor asthma triggers.
Impact of a household environmental intervention delivered by lay health workers on asthma symptom control in urban, disadvantaged children with asthma.
A randomized controlled trial of asthma self-management support comparing clinic-based nurses and in-home community health workers: the Seattle–King County Healthy Homes II Project.
The Seattle–King County Healthy Homes Project: a randomized, controlled trial of a community health worker intervention to decrease exposure to indoor asthma triggers.
Improved outcomes included caregiver quality of life and caregiver depressive symptoms. The study by Bonner et al reported a 41% increase in participant self-efficacy for managing asthma versus only a 9% increase in the control group.
Evaluation of Community Action Against Asthma: a community health worker intervention to improve children’s asthma-related health by reducing household environmental triggers for asthma.
Evaluation of Community Action Against Asthma: a community health worker intervention to improve children’s asthma-related health by reducing household environmental triggers for asthma.
Parker et al demonstrated that, over 12 months, the proportion of the intervention group that needed unscheduled clinic visits decreased by 6%, whereas the control group’s unscheduled visits increased by 15%.
Evaluation of Community Action Against Asthma: a community health worker intervention to improve children’s asthma-related health by reducing household environmental triggers for asthma.
In the study by Morgan et al, there were significant reductions in the disruption of caretakers’ plans and school days missed by the children in the intervention group.
Evaluation of Community Action Against Asthma: a community health worker intervention to improve children’s asthma-related health by reducing household environmental triggers for asthma.
Evaluation of Community Action Against Asthma: a community health worker intervention to improve children’s asthma-related health by reducing household environmental triggers for asthma.
the study with low or unclear risk of bias in all categories showed no differences in parental concern, worry, confidence, or diabetes impact on family.
The studies by Laffel et al and Svoren et al measured severe hypoglycemic events, glycemic control, outpatient clinic visits, and emergency care visits and hospitalizations.
Both of these studies reported improved glycemic control in the intervention groups. In addition, both reported increased outpatient clinic attendance and a decrease in hospitalization and ED visits for the intervention group. In the study by Laffel et al, appointment attendance in the diabetes clinic improved with a mean of 7.1 visits/24 months for the intervention group compared to 5.2 visits/24 months for the control group.
Obesity
The study by Resnick et al (high risk in one category) reported a modest reduction of children’s BMI percentile. The mean BMI percentile decreased from 94.1 to 90.6 at the end of the 12-month intervention for all children (intervention and control).
There were no differences in BMI reductions between groups.
Failure to Thrive
The lay health worker intervention in the study by Black et al (low or unclear risk in all categories) resulted in improved growth parameters for children with FTT regardless of intervention status.
Two studies related to asthma assessed the sustainability of intervention outcomes. In the study by Krieger et al (low or unclear risk of bias in all categories), the sustainability of the intervention outcome was assessed 6 months after exit from the intervention group.
The Seattle–King County Healthy Homes Project: a randomized, controlled trial of a community health worker intervention to decrease exposure to indoor asthma triggers.
Improvements in caregiver quality of life, urgent care use, and days with activity limitation were all sustained. The control group was not followed because members of this group crossed over and received the intervention. In the study by Morgan et al (high risk of bias in 1 category), sustained intervention effects (relative to control) were observed 12 months after exit for caretaker nighttime wakening, missed school days, and number of days with symptoms.
Two studies (both low or unclear risk of bias in all categories) examined cost-effectiveness or cost savings of lay health worker interventions. In the study by Flores et al,
the average monthly cost of the intervention per patient was $60.42. The intervention group experienced savings of $361.48 for hospitalizations and $50.33 for ED encounters. The mean reduction in asthma exacerbation days was 1.26 days for the intervention group and 0.78 for the control group. The incremental cost-effectiveness ratio for the intervention group was −$597.10 per asthma exacerbation-free day gained, indicating a total cost savings for the intervention group. In the 2005 study by Krieger et al, the estimated marginal cost of the high-intensity intervention relative to the low-intensity intervention was $1124 per child.
The Seattle–King County Healthy Homes Project: a randomized, controlled trial of a community health worker intervention to decrease exposure to indoor asthma triggers.
The savings in urgent care costs (ED visits, hospitalization, and unscheduled clinic visits) during a 2-month period was calculated to range from $57 to $80 per child.
Discussion
The evidence for lay health worker interventions improving the care of children with chronic conditions is generally positive according to this review. Benefits of interventions included reduced urgent health care use, decreased symptoms, improved child health status, fewer missed parental work days, fewer missed child school days, improved parental quality of life, and increased self-efficacy. However, these findings were not consistent across all studies. Two studies demonstrated that intervention effects may be sustained long after the intervention ends. Additional studies indicated that lay health worker interventions may offer cost savings. Current data are limited to outcomes among a small number of specific diseases. In our review, the majority of interventions were conducted in asthma with far fewer in diabetes, obesity, and FTT. Overall, the current evidence suggests that lay health worker interventions may provide an important strategy for improving care and therefore warrant further study.
This work adds to the findings of previous reviews of lay health worker interventions.
This review included 17 trials involving 7442 participants. It concluded that lay health worker interventions led to modest, short-term improvements in patients’ confidence to manage their condition and perceptions of their own health without altering health care utilization or quality of life. A review by Postma et al specifically assessed the effectiveness of community health worker interventions for children with asthma.
Seven studies with 2,316 participants were examined. The review found positive outcomes associated with lay health worker interventions, including decreased asthma symptoms, day-time activity limitations, and urgent care use. Six of the studies in the review by Postma were among the 11 included in our review. A most recent review conducted by the Agency for Healthcare Research and Quality on adults and children concluded that evidence of the effectiveness of lay health worker interventions is mixed.
The development of innovative and effective care delivery mechanisms for children with chronic conditions has evolved as a central priority in redefining pediatric practice. Various models of care (eg, medical homes, comprehensive hospital-based care programs) have been implemented with limited evidence and variable results.
In adult care, newer lay health worker interventions have provided insights on how to facilitate timely, high-quality care for those with chronic conditions. Most recently, patient navigation has emerged as an innovative care model with the principal function of eliminating barriers to timely delivery of services for individual patients across the health care continuum.
Although patient navigation includes components grounded in self-efficacy and social support theories similar to lay health worker interventions, it also incorporates practical assistance to achieve desired outcomes. It has its conceptual and theoretical roots in cancer care where it was developed in order to support and guide adults with abnormal cancer screening or a new cancer diagnosis.
Patient navigators are non–health workers who target at-risk populations such as racial/ethnic minorities and those from low-income populations for delays in care.
Positive outcomes associated with patient navigation have included timeliness of diagnosis, time to initiation of primary therapy, patient satisfaction, quality of life, and cost-effectiveness.
Metrics for evaluating patient navigation during cancer diagnosis and treatment: crafting a policy-relevant research agenda for patient navigation in cancer care.
Patient navigation may represent a new model for lay health worker intervention for children in the future.
This review had several strengths. These included a focus on children with chronic conditions, inclusion of conditions other than asthma, and incorporation of more current studies. The inclusive search strategy and rigorous evaluation of risk of bias have resulted in a comprehensive and critical assessment of the current state of the science for lay health worker interventions. However, limitations to this review also merit discussion. Although the individual studies assessed a diverse set of outcomes, they did little in providing a theoretical basis for improvements. Only 10 of the 17 studies cited specific conceptual models that informed the interventions. Of the studies that did use conceptual models, few operationalized theoretical concepts in such a way that the mediators of behavioral change could be identified. For example, if an intervention resulted in improved quality of life, it could not be determined which element of the intervention led to the change. Therefore it was not possible to summarize the individual elements of the causal pathways that led to improvements. As more lay health worker or patient navigation interventions are developed, more attention must be directed toward the operationalization of the theoretical models underpinning these interventions and analysis of which elements lead to improvements. Only 2 studies in this review examined costs. Though both studies documented cost-savings, they did not rigorously assess cost-effectiveness as recommended for such studies.
Most studies were limited to a small set of chronic conditions in urban populations. Future studies should focus on other chronic conditions and different settings (eg, rural).
There were also limitations inherent to the review methodology. The gray literature (evaluation of theses, dissertations, or unpublished work) was not included in the search as a result of limitations in resources. The studies that met the inclusion criteria were heterogeneous with varied medical conditions, designs, interventions, and outcome measures. The reporting of results by individual studies also varied widely with authors reporting intervention effects, percentage changes, or simply P values to demonstrate outcomes. These differences limited comparisons among studies. Given the heterogeneity of outcomes, we could not construct a funnel plot to assess publication bias or perform a meta-analysis. Future studies should report intervention effects with confidence intervals and all elements of study design that could lead to systematic bias.
Policy Implications
Although statistically significant positive outcomes of lay health worker interventions have been demonstrated, the clinical benefits are relatively small in scale. Though any improvement is noteworthy, a critical question for policy makers will be whether this scale of improvement will make any appreciable impact on costs associated with management of pediatric chronic disease, particularly when there is increasing emphasis on identifying interventions to produce cost savings. Even if lay health worker interventions do result in cost savings, the returns might not go to those who fund the programs but might be realized by other parties. Therefore, cost savings must be considered from the perspective of all stakeholders, both within and outside the health care system. This determination may better elucidate which entities should be most invested in funding these programs (eg, hospitals, insurers, or employers). Aside from cost, another challenge to implementation of lay health worker interventions will be the current dynamic context of health care. Increasing focus on health literacy and technology-based interventions may change the role of lay health workers, as well as how they communicate with patients.
Conclusions
Lay health worker interventions are emerging as innovative models of care for individuals with chronic conditions, especially those from minority and low socioeconomic backgrounds. Many of the pediatric studies of lay health worker interventions report modest positive results including reduced health care use, decreased symptoms, and improved parental quality of life. Gaps in knowledge remain regarding the long-term sustainability of these benefits or cost-effectiveness of such programs. Core metrics must be identified to elucidate the impact of lay health worker interventions on children with chronic conditions. Lay health worker interventions should be further assessed in other specific conditions such as sickle cell disease and cystic fibrosis as well as among medically complex children who have multiple diagnoses and medical needs. Because most lay health worker studies have been conducted in urban areas, more trials in rural areas are warranted.
Conflict of Interest
The authors declare that they have no conflict of interest.
Acknowledgments
Supported in part by a grant to JLR ( NIH grant 1K23 HL105568-01A1 ).
References
Newacheck P.W.
Halfon N.
Prevalence and impact of disabling chronic conditions in childhood.
What is the prevalence of children with special health care needs? Toward an understanding of variations in findings and methods across three national surveys.
Satisfaction with care and ease of using health care services among parents of children with special health care needs: the roles of race/ethnicity, insurance, language, and adequacy of family-centered care.
A randomized controlled trial of asthma self-management support comparing clinic-based nurses and in-home community health workers: the Seattle–King County Healthy Homes II Project.
Evaluation of Community Action Against Asthma: a community health worker intervention to improve children’s asthma-related health by reducing household environmental triggers for asthma.
The Seattle–King County Healthy Homes Project: a randomized, controlled trial of a community health worker intervention to decrease exposure to indoor asthma triggers.
Impact of a household environmental intervention delivered by lay health workers on asthma symptom control in urban, disadvantaged children with asthma.
Metrics for evaluating patient navigation during cancer diagnosis and treatment: crafting a policy-relevant research agenda for patient navigation in cancer care.
00125-3&id=gr1.jpg){kind=link}