549544
research-article2014
CPJXXX10.1177/0009922814549544Clinical PediatricsBrixey et al
Article
The Impact of Behavioral Risk Assessments and Tailored Health Information on Pediatric Injury
Clinical Pediatrics 2014, Vol. 53(14) 1383–1389 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/0009922814549544 cpj.sagepub.com
Suzanne N. Brixey, MD1, Nancy L. Weaver, PhD, MPH2, Clare E. Guse, MS1, Haydee Zimmermann, BA1, Janice Williams, MSED3, Timothy E. Corden, MD1, and Marc H. Gorelick, MD, MSCE1
Abstract Background. Integrating age appropriate injury prevention messages during a well-child visit is challenging in the face of competing demands. Purpose. To describe a 7-month pilot using technology to facilitate injury prevention risk assessment and education integration. Methods. We prospectively tracked responses to the computer-based injury prevention self-assessment tool, safety product distribution, and any subsequent contact with the local hospital system for related unintentional injuries. Results. A total of 2091 eligible visits by 1368 unique patients were assessed. Eight hundred forty-three unique patients completed the Safe N’ Sound assessment and 7 were subsequently injured, with an injury related to a Safe N’ Sound target area. Conclusions. A kiosk-based tailored injury assessment tool can be successfully integrated into a busy pediatric practice. Unintentional injury outcomes can be linked to the tailored anticipatory guidance and can identify the effectiveness of this electronic integration of injury prevention messaging into well-child examinations. Keywords unintentional injury, children, anticipatory guidance, prevention
Introduction Unintentional injuries remain the leading cause of childhood morbidity and mortality.1 For example, among 1to 4-year-olds in the United States, the rate of nonfatal unintentional injuries is 13 795.9 per 100 000, while the rate of unintentional injury deaths is 8.6 per 100 000, or nearly 1400 fatalities in 2010.2 Owens et al3 found that nearly one-third of pediatric emergency department (ED) visits were because of injuries, and children from low income communities were more likely to have an injury-related ED visit than their high income community peers. The American Academy of Pediatrics4 has highlighted the importance of injury prevention anticipatory guidance when children arrive for well-child care. However, studies have identified that families may not receive ideal injury prevention counseling.5-7 There are many barriers to integrating injury prevention into every well-child check, including competing demands and the preparation of the clinic staff and physicians to provide such counseling. In order to support the inclusion of injury prevention in well-child visits, it is important to evaluate the efficacy of approaches to injury
prevention counseling, not only on short-term outcomes, but also on long-term injury rates. There are few evidence-based strategies to improve home safety, especially for certain subgroups of the population that are at highest risk for injury.8,9 Appropriate injury prevention strategies to combat these high rates of injury and death should be tailored to a child’s developmental stage and their risk of various mechanisms of injury.10,11 One such program, Safe N’ Sound (SNS), delivers such tailored information to parents of young children to address the highest impact injuries.12-14 SNS has been evaluated and found to result in the greater adoption of safety devices and practices.13 Notably, 1
Medical College of Wisconsin, Milwaukee, WI, USA Saint Louis University, St. Louis, MO, USA 3 Carolinas Medical Center, Charlotte, NC, USA 2
Corresponding Author: Suzanne N. Brixey, General Pediatrics, Medical College of Wisconsin, Injury Research Center, Children’s Research Institute, CMG-Downtown Health Center Pediatric Clinic, 1020 North 12th Street, Milwaukee, WI, 53233, USA. Email: [email protected]
Downloaded from cpj.sagepub.com at SUNY STONY BROOK on June 30, 2015
1384
Clinical Pediatrics 53(14)
these program effects were greatest in populations with low education levels. While SNS, and other programs, show promise in reducing risk factors for injuries among children, there is little data linking injury prevention counseling with a decrease in actual injury rates.3 The objective of this article is to describe a 7-month pilot of a novel implementation of SNS using technology to facilitate its integration into an urban academic clinic, the Downtown Health Center Pediatric Clinic (DHC) and measure concurrence of risk identification with actual injury rates.
Methods This was a prospective cohort study. This study was approved by the Institutional Review Board of the Children’s Hospital of Wisconsin.
kiosk generates a tailored 4- page safety assessment for the family. SNS topics include child passenger safety, burn prevention, fall prevention, unintentional poisoning, and choking prevention. The first 3 pages of the assessment are for the caregiver and are designed in a manner to facilitate social learning and influence the perceived social norms and support self-efficacy, all nested within the context of effective parenting. The fourth page is a summary of the injury prevention health risk assessment and it is reviewed by the child’s health care provider and remains in the child’s chart. This process allowed tracking of the individual responses and services over time. If the patient was called back to the examination room, prior to completion of the survey, the educational material would not print and the survey would not be counted.
Data Sources
Study Setting The Downtown Health Center, part of the Children’s Hospital of Wisconsin system, was established to improve the health and advance health equity for the most vulnerable children in Milwaukee through an academic medical home. It serves more than 7200 children and adolescents, of which 80% are African American and 95% live in poverty with 42% of the families reporting an annual household income of less than $10 000. The Children’s Hospital of Wisconsin is the major children’s hospital in the metropolitan area and has more than 80 000 ED or urgent care visits each year.
Study Participants The cohort consisted of patients 0 to 4 years old who sought preventive primary care services at the DHC between July 1, 2011 and January 31, 2012. All caregivers of eligible patients were invited to participate in the SNS program, which was offered in English only; the registration staff provided caregivers with a paper “Prescription for Safety” stickered with a unique bar-coded visit number. While in the waiting room, independent of the staff, the caregiver inserted the prescription slip into the kiosk where a scanner read the bar code to identify the patient and then initiated survey questions electronically linked to the unique patient. The SNS program accounts for developmental changes by targeting questions in age intervals as follows: 0 to 6 months; 7 to 11 months; 12 to 17 months; 18 to 23 months; 2 years; 3 years; and 4 years. After SNS identifies the 2 highest priority injury prevention measures for the child’s age and developmental stage, based on questions regarding home and vehicle safety measures, the
Kiosk. The SNS program served both as the risk data collection tool and the program that generated the tailored reports. All data collected from the kiosk including the individual visit number (which enabled linkage to the child’s medical record number, age grouping, and date of service) as well as the 2 identified risk areas were electronically captured and housed in a password-protected database that was downloaded monthly. Data were reviewed monthly to generate clinic level reports and to assess data quality. Injury and Administrative Data. Billing and diagnosis data on pediatric injury visits between July 1, 2011 and January 31, 2012 were acquired for all children that presented to Children’s Hospital of Wisconsin (CHW children) either in the ED, urgent care clinics, specialty clinics, or direct inpatient admissions whose diagnosis was a targeted unintentional injury. Linkages were made by medical record number and date of service to identify patients of the DHC who were subsequently seen within the CHW system for an unintentional injury visit, defined as a visit with an associated International Classification of Diseases, 9th Edition (ICD-9) external cause of injury (E-code). Registration and administrative data provided information on the date of birth, race/ ethnicity, insurance status, ICD-9 diagnoses and external cause of injury codes. Visit data from hospitals outside the CHW system were not available.
Measures Primary Outcome Measures. The primary outcome measures were (1) the completion rate of SNS over the implementation period (completed kiosk/given prescription),
Downloaded from cpj.sagepub.com at SUNY STONY BROOK on June 30, 2015
1385
Brixey et al Table 1. E-Codes for Downtown Health Center Pediatric Clinic Safety Corner Targeted Injuries. E-Codes E015; E924 E811-E816; E818-E819 E826 E850-858; E860-E864 E880 E884 E890, E893, E895, E898-899 E910; E013 E913
Target Injury Burn prevention Child passenger safety Bike helmet Poison prevention Fall—stairs Fall—bed Fire prevention Drowning prevention Safe sleep
(2) the injury types identified as the top 2 priorities by age group, and (3) the occurrence of an injury-related visit during the 7 months of the study subsequent to the eligible DHC visit for all DHC patients whether exposed to SNS information or not. Unintentional injuries were assessed using the following E-codes: E001.0-E029.9; E800-E844.5; E846-E870.0; E880-E903; E907-E929.9; E980-E989. The E-codes were then categorized to reflect the injury topics addressed by the DHC Safety Corner kiosk (see Table 1). Unique children were the unit of measure. If a single visit resulted in multiple E-codes identifying different target injuries, we identified the E-code that captured the primary cause of injury. For example, if a child fell down a flight of stairs on a tricycle, and the injury was coded as both a fall down stairs and a fall off of a bike, the E-code for fall down stairs was used, since a recommended stair gate may have prevented this injury. Only 3 such multiple E-code events occurred. Primary Exposure Measure. The primary exposure measure was completion of the SNS risk assessment at wellchild care visits.
Analysis Descriptive statistics were computed for the number of referrals to and completions of the SNS program. Fisher’s exact test was used to conduct comparisons of the distribution of injury types between the DHC and the CHW children, by age group. Injury rates per 1000 visits within age group for the DHC were computed as the number of injuries divided by the number of unique child visits to the DHC multiplied by 1000. The relative risk and its 95% confidence interval were calculated for the comparison of injuries in unscreened and screened children. Lastly, for all screened patients, injuries identified within the CHW system were assessed and compared with injury areas addressed in their SNS report.
Results Demographics Over the 7-month time period of the study there were 2091 eligible preventive health visits of children aged 0 to 4 years. There were 1084 kiosk completions by the 1406 patients provided a Prescription for Safety for a 77% kiosk completion rate. The first kiosk assessment for a child was used for the remainder of the analyses to prevent duplicate counting, resulting in 843 SNS assessments for the 1368 unique patients (62% assessed at least once). As shown in Figure 1, rates of referral and completion increased during the first 5 months of the program, and then stabilized. The kiosk survey was usually completed by mothers (84%) and those who described themselves as African American (77%). Forty-two percent reported that their highest level of education completed was a high school diploma or GED equivalent, and 68% indicated an annual household income of less than $10 000.
Screenings and Risks The highest number of screenings was for children ages 0 to 6 months (n = 419; 38% of the total), while the fewest occurred among children 4 years of age (n = 59; 5% of the total). Risks for injuries from falls were most commonly identified as 1 of the 2 highest risks for the first 2 years of life. For children in the first 6 months of life, 14% were identified as at risk for a fall from bed and 11% for a fall down stairs. This risk increased for infants 7 to 11 months to 28% for fall from bed and 14% at risk for a fall down stairs. Being at risk of a fall from a bed remained an elevated risk at 18% for 12- to 17-month-old toddlers, and 15% for 18-to 23-month-old children. After 6 months, child passenger safety risks were highly prioritized for all age groups (14% for 7- to 11-month-olds, 16% for 12- to 17-month-olds, 17% for 18- to 23-month-olds, 29% for 2-year-olds, 42% for 3-year-olds, and 41% for 4-year-olds). Choking risks were most often identified from 7 months until 2 years of age (15% for 7-11 months, 26% for 12-17 months, 33% for 18- to 23-month-olds, and 38% for 2-year-old children).
Injuries During the time period of the screenings, a total of 11 DHC children were injured and seen within the CHW system for an unintentional injury targeted for review by the DHC SNS implementation. Seven of the 843 screened children (0.8%) had a subsequent injury visit related to a SNS injury area (Table 2), while 4 of the 525 unscreened
Downloaded from cpj.sagepub.com at SUNY STONY BROOK on June 30, 2015
1386
Clinical Pediatrics 53(14)
Figure 1. Eligible visits, kiosk referrals, and kiosk completions by month. Table 2. Kiosk-Screened Children (DHC patients) With Safe N’ Sound Identified Risks and Subsequent Injury, by Age Group. Age Group in Months (Total Number Injured)
Screening Age Screened Date (Months)
0-6 (1) 7-11 (2)
8/17/11 11/3/11
1 7
12-17 (1)
8/22/11 9/13/11
3 12
18-23 (3)
7/28/11
17
9/23/11
19
10/6/11
20
Risks Identified
Age Injured (Months)
No risks Falls—walker, poisons Crib, hot water Falls—Stairs, Falls—walker Drowning, choking—food Car safety, drowning Falls—stairs, choking
5 7 8 16 19 21 22
Date of Injury
Injury (E-Code)
12/17/11 Fall from bed (E884.4) 11/14/11 Fall from bed (E884.4) 1/19/12 1/11/12
Fall from bed (E884.4) Motor vehicle collision NOSPassenger (E812.1) 9/22/11 Hot liquids & vapors, including. steam (E924.0) 11/26/11 Hot substance accident NEC (E924.8) 12/16/11 Accidental poisoning by corrosives and caustics NEC (E864.3)
Abbreviations: NOS, not otherwise specified; NEC, not elsewhere classified.
eligible children had a subsequent injury visit related to a SNS targeted injury (0.8%). Of the 7 children screened and injured there was no case of injury in a screened child that was related to the specific SNS identified priority risks for that child (Figure 2).
Discussion To our knowledge, this is the first effectiveness evaluation of the injury outcomes associated with the use of a tailored injury prevention program in a clinic setting. This study demonstrated that a kiosk-based prevention tool can be successfully incorporated as best practice
standard of care for a clinic. We were able to screen 77% of all patients receiving the Prescription for Safety. The percentage of referrals and completions by those eligible increased over time and appeared to become more institutionalized after 5 months. In addition, we used a standardized previsit form with a unique patient bar code to activate the kiosk and to link the Safe N’ Sound responses to the identified injury outcomes. Previous work implementing SNS in a clinical setting showed much lower reach (Weaver,14 2011). In a field trial without a research or practice infrastructure to facilitate integration, SNS completion averaged 10%. The greater reach in this application of the program speaks to several
Downloaded from cpj.sagepub.com at SUNY STONY BROOK on June 30, 2015
1387
Brixey et al
2091 Eligible Visits
1368 Unique Pa ents
843
525
SNS Assessments
No SNS Assessment
7
836
Injured
No Injury
4
521
Injuries
No Injuries
5 mo – Fall From Bed 7 mo – Fall From Bed
16 mo – Fall From Bed
8 mo – Fall From Bed
18 mo - Motor Vehicle Collision
16 mo - Motor Vehicle Collision 2 yr – Fall From Bed 19 mo - Burn 4 yr – Poisoning
21 mo - Burn 22 mo - Poisoning
Figure 2. Eligible Visits, Safe N’ Sound screenings and subsequent injuries (July 1, 2011 to January 31, 2013).
important factors that should be considered when implementing evidence-based programs. First, the use of the “Prescription for Safety” provided a credible cue for engaging parents with the program which helped make the use of the kiosk a commonplace and expected behavior. Second, the physicians and staff at the clinic were highly committed to the program. Several champions
helped foster this commitment by providing routine reports and encouraging dialogue about the organizational changes that were needed to accommodate the use of the program. Globally, we found a significant overlap between the SNS identified risks collectively and the ultimate injuries seen during the same time period. For children 0 to
Downloaded from cpj.sagepub.com at SUNY STONY BROOK on June 30, 2015
1388
Clinical Pediatrics 53(14)
6 months of age, burns and falls were both identified as a self-reported risk and as actual injuries. For 7- to 11-month-old infants falls from both bed and stairs showed congruency. For children aged 12 to 23 months, falls and child passenger safety were seen in both selfreported risks and actual injuries. There was limited overlap for children at age 2 years but by 3 years of age overlap was seen for car crash risks, and for 4-year-olds car crashes and burns were both identified in selfreported risks and actual injuries. Importantly, while the study design does not control for all factors that could account for these differences, the 7 children screened by SNS were not subsequently injured by a risk covered in their customized report. This may speak to the benefit of the tailored education provided but more research is needed. Injury risks also vary by developmental age; risks identified at 2 months (such as unsafe sleep in a crib) may no longer be relevant as a child matures. While another study used focused counseling with educational handouts and attempted to measure injuries by parent report and medical record data, there was limited enrollment and no significant difference in the number of injuries between groups was found.15 There are several limitations to this study. Although all children 0 to 4 years old who presented for their wellchild checkup were eligible, due to the emphasis of maintaining clinic flow, not all children were referred to the kiosk by the front desk staff or had the time to complete the survey prior to being called back into an examination room to be seen by their provider. Caregivers who were not fluent in English may not have been able to use the program. The kiosk identifies self-reported risks. Although there is concern that safety behaviors may be overestimated, there is data to support that selfreports of ownership of a car seat are reliable.16,17 There is an inexact match between the SNS questions and E-codes. As a pilot study there was not enough power in the current number of patients screened and injured to establish statistical significance. The SNS survey is designed to stop collecting data after identifying 2 “atrisk” responses; so data were not collected about each question at each visit. This limitation of SNS to only identify 2 risks is supported by clinical evidence of the amount of information an individual caregiver can absorb at any individual time and supports clinical efficiencies in a busy practice. This limitation of SNS to only identify 2 risks is supported by health education, but clearly limits the analysis. Although there was no child screened who was injured by the targeted risk identified it is not clear that there is causality as we did not always know the response for the targeted injury risk question due to the SNS program ceasing once 2 risks are identified.
Conclusion Our results, joined with the limited research on the use of electronic devices in other pediatric settings, should encourage the adoption of theory-driven electronic communication strategies to support families in making changes to improve child safety. We were able to successfully screen families and provide needed safety information that may impact injury outcomes with minimal impact on patient flow. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The author(s) received no financial support for the research, authorship and/or publication of this article.
References 1. Centers for Disease Control and Prevention. Vital signs: Unintentional injury deaths among persons aged 0-19 years—United States, 2000-2009. MMWR Morb Mortal Wkly Rep. 2012;61:270-276. 2. Web-based injury statistics query and reporting system. Centers for Disease Control and Prevention Web site. http://www.cdc.gov/ncipc/wisqars. Updated 2012. Accessed February 19, 2013. 3. Owens PL, Zodet MW, Berdahl T, Dougherty D, McCormick MC, Simpson LA. Annual report on health care for children and youth in the United States: focus on injury-related emergency department utilization and expenditures. Ambul Pediatr. 2008;8:219 -240. 4. Gardner HG, American Academy of Pediatrics Committee on Injury, Violence, and Poison Prevention. Office-based counseling for unintentional injury prevention. Pediatrics. 2007;119:202-206. 5. Gielen AC, McDonald EM, Forrest CB, Harvilchuck JD, Wissow L. Injury prevention counseling in an urban pediatric clinic. Analysis of audiotaped visits. Arch Pediatr Adolesc Med. 1997;151:146-151. 6. Bass JL, Christoffel KK, Widome M, et al. Childhood injury prevention counseling in primary care settings: a critical review of the literature. Pediatrics. 1993;92:544-550. 7. Barrios LC, Runyan CW, Downs SM, Bowling JM. Pediatric injury prevention counseling: an observational study of process and content. Patient Educ Couns. 2001;44:141-149. 8. Kendrick D, Coupland C, Mulvaney C, et al. Home safety education and provision of safety equipment for injury prevention. Cochrane Database Syst Rev. 2007;(1):CD005014. 9. Kendrick D, Groom L, Stewart J, Watson M, Mulvaney C, Casterton R. “Risk watch”: cluster randomised controlled
Downloaded from cpj.sagepub.com at SUNY STONY BROOK on June 30, 2015
1389
Brixey et al trial evaluating an injury prevention program. Inj Prev. 2007;13(2):93-98. 10. Agran PF, Winn D, Anderson C, Trent R, Walton-Haynes L. Rates of pediatric and adolescent injuries by year of age. Pediatrics. 2001;108:E45. 11. Agran PF, Anderson C, Winn D, Trent R, Walton-Haynes L, Thayer S. Rates of pediatric injuries by 3-month intervals for children 0 to 3 years of age. Pediatrics. 2003;111(6 pt 1):e683-e692. 12. Weaver NL, Williams J, Jacobsen HA, et al. Translation of an evidence-based tailored childhood injury prevention program. J Public Health Manag Pract. 2008;14:177-184. 13. Nansel TR, Weaver N, Donlin M, Jacobsen H, Kreuter MW, Simons-Morton B. Baby, Be safe: The effect of tailored communications for pediatric injury prevention
provided in a primary care setting. Patient Educ Couns. 2002;46:175-190. 14. Weaver NL, Nansel TR, Williams J, Tse J, Botello Harbaum M, Willson K. Reach of a kiosk-based pediatric injury prevention program. Transl Behav Med. 2011;1:515-522. 15. Sangvai S, Cipriani D, Colborn DK, Wald ER. Studying injury prevention: practices, problems and pitfalls in implementation. Clin Pediatr (Phila). 2007;46:228. 16. Uherick L, Gorelick MH, Biechler R, Brixey SN, MelzerLange M. Validation of two child passenger safety questionnaires. Inj Prev. 2010;16:343-347. 17. Robertson A, Rivara F, Ebel BE, Lymp JF, Christakis DA. Validation of parent self reported home safety practices. Inj Prev. 2005;11:209-212.
Downloaded from cpj.sagepub.com at SUNY STONY BROOK on June 30, 2015
research-article2014
CPJXXX10.1177/0009922814549544Clinical PediatricsBrixey et al
Article
The Impact of Behavioral Risk Assessments and Tailored Health Information on Pediatric Injury
Clinical Pediatrics 2014, Vol. 53(14) 1383–1389 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/0009922814549544 cpj.sagepub.com
Suzanne N. Brixey, MD1, Nancy L. Weaver, PhD, MPH2, Clare E. Guse, MS1, Haydee Zimmermann, BA1, Janice Williams, MSED3, Timothy E. Corden, MD1, and Marc H. Gorelick, MD, MSCE1
Abstract Background. Integrating age appropriate injury prevention messages during a well-child visit is challenging in the face of competing demands. Purpose. To describe a 7-month pilot using technology to facilitate injury prevention risk assessment and education integration. Methods. We prospectively tracked responses to the computer-based injury prevention self-assessment tool, safety product distribution, and any subsequent contact with the local hospital system for related unintentional injuries. Results. A total of 2091 eligible visits by 1368 unique patients were assessed. Eight hundred forty-three unique patients completed the Safe N’ Sound assessment and 7 were subsequently injured, with an injury related to a Safe N’ Sound target area. Conclusions. A kiosk-based tailored injury assessment tool can be successfully integrated into a busy pediatric practice. Unintentional injury outcomes can be linked to the tailored anticipatory guidance and can identify the effectiveness of this electronic integration of injury prevention messaging into well-child examinations. Keywords unintentional injury, children, anticipatory guidance, prevention
Introduction Unintentional injuries remain the leading cause of childhood morbidity and mortality.1 For example, among 1to 4-year-olds in the United States, the rate of nonfatal unintentional injuries is 13 795.9 per 100 000, while the rate of unintentional injury deaths is 8.6 per 100 000, or nearly 1400 fatalities in 2010.2 Owens et al3 found that nearly one-third of pediatric emergency department (ED) visits were because of injuries, and children from low income communities were more likely to have an injury-related ED visit than their high income community peers. The American Academy of Pediatrics4 has highlighted the importance of injury prevention anticipatory guidance when children arrive for well-child care. However, studies have identified that families may not receive ideal injury prevention counseling.5-7 There are many barriers to integrating injury prevention into every well-child check, including competing demands and the preparation of the clinic staff and physicians to provide such counseling. In order to support the inclusion of injury prevention in well-child visits, it is important to evaluate the efficacy of approaches to injury
prevention counseling, not only on short-term outcomes, but also on long-term injury rates. There are few evidence-based strategies to improve home safety, especially for certain subgroups of the population that are at highest risk for injury.8,9 Appropriate injury prevention strategies to combat these high rates of injury and death should be tailored to a child’s developmental stage and their risk of various mechanisms of injury.10,11 One such program, Safe N’ Sound (SNS), delivers such tailored information to parents of young children to address the highest impact injuries.12-14 SNS has been evaluated and found to result in the greater adoption of safety devices and practices.13 Notably, 1
Medical College of Wisconsin, Milwaukee, WI, USA Saint Louis University, St. Louis, MO, USA 3 Carolinas Medical Center, Charlotte, NC, USA 2
Corresponding Author: Suzanne N. Brixey, General Pediatrics, Medical College of Wisconsin, Injury Research Center, Children’s Research Institute, CMG-Downtown Health Center Pediatric Clinic, 1020 North 12th Street, Milwaukee, WI, 53233, USA. Email: [email protected]
Downloaded from cpj.sagepub.com at SUNY STONY BROOK on June 30, 2015
1384
Clinical Pediatrics 53(14)
these program effects were greatest in populations with low education levels. While SNS, and other programs, show promise in reducing risk factors for injuries among children, there is little data linking injury prevention counseling with a decrease in actual injury rates.3 The objective of this article is to describe a 7-month pilot of a novel implementation of SNS using technology to facilitate its integration into an urban academic clinic, the Downtown Health Center Pediatric Clinic (DHC) and measure concurrence of risk identification with actual injury rates.
Methods This was a prospective cohort study. This study was approved by the Institutional Review Board of the Children’s Hospital of Wisconsin.
kiosk generates a tailored 4- page safety assessment for the family. SNS topics include child passenger safety, burn prevention, fall prevention, unintentional poisoning, and choking prevention. The first 3 pages of the assessment are for the caregiver and are designed in a manner to facilitate social learning and influence the perceived social norms and support self-efficacy, all nested within the context of effective parenting. The fourth page is a summary of the injury prevention health risk assessment and it is reviewed by the child’s health care provider and remains in the child’s chart. This process allowed tracking of the individual responses and services over time. If the patient was called back to the examination room, prior to completion of the survey, the educational material would not print and the survey would not be counted.
Data Sources
Study Setting The Downtown Health Center, part of the Children’s Hospital of Wisconsin system, was established to improve the health and advance health equity for the most vulnerable children in Milwaukee through an academic medical home. It serves more than 7200 children and adolescents, of which 80% are African American and 95% live in poverty with 42% of the families reporting an annual household income of less than $10 000. The Children’s Hospital of Wisconsin is the major children’s hospital in the metropolitan area and has more than 80 000 ED or urgent care visits each year.
Study Participants The cohort consisted of patients 0 to 4 years old who sought preventive primary care services at the DHC between July 1, 2011 and January 31, 2012. All caregivers of eligible patients were invited to participate in the SNS program, which was offered in English only; the registration staff provided caregivers with a paper “Prescription for Safety” stickered with a unique bar-coded visit number. While in the waiting room, independent of the staff, the caregiver inserted the prescription slip into the kiosk where a scanner read the bar code to identify the patient and then initiated survey questions electronically linked to the unique patient. The SNS program accounts for developmental changes by targeting questions in age intervals as follows: 0 to 6 months; 7 to 11 months; 12 to 17 months; 18 to 23 months; 2 years; 3 years; and 4 years. After SNS identifies the 2 highest priority injury prevention measures for the child’s age and developmental stage, based on questions regarding home and vehicle safety measures, the
Kiosk. The SNS program served both as the risk data collection tool and the program that generated the tailored reports. All data collected from the kiosk including the individual visit number (which enabled linkage to the child’s medical record number, age grouping, and date of service) as well as the 2 identified risk areas were electronically captured and housed in a password-protected database that was downloaded monthly. Data were reviewed monthly to generate clinic level reports and to assess data quality. Injury and Administrative Data. Billing and diagnosis data on pediatric injury visits between July 1, 2011 and January 31, 2012 were acquired for all children that presented to Children’s Hospital of Wisconsin (CHW children) either in the ED, urgent care clinics, specialty clinics, or direct inpatient admissions whose diagnosis was a targeted unintentional injury. Linkages were made by medical record number and date of service to identify patients of the DHC who were subsequently seen within the CHW system for an unintentional injury visit, defined as a visit with an associated International Classification of Diseases, 9th Edition (ICD-9) external cause of injury (E-code). Registration and administrative data provided information on the date of birth, race/ ethnicity, insurance status, ICD-9 diagnoses and external cause of injury codes. Visit data from hospitals outside the CHW system were not available.
Measures Primary Outcome Measures. The primary outcome measures were (1) the completion rate of SNS over the implementation period (completed kiosk/given prescription),
Downloaded from cpj.sagepub.com at SUNY STONY BROOK on June 30, 2015
1385
Brixey et al Table 1. E-Codes for Downtown Health Center Pediatric Clinic Safety Corner Targeted Injuries. E-Codes E015; E924 E811-E816; E818-E819 E826 E850-858; E860-E864 E880 E884 E890, E893, E895, E898-899 E910; E013 E913
Target Injury Burn prevention Child passenger safety Bike helmet Poison prevention Fall—stairs Fall—bed Fire prevention Drowning prevention Safe sleep
(2) the injury types identified as the top 2 priorities by age group, and (3) the occurrence of an injury-related visit during the 7 months of the study subsequent to the eligible DHC visit for all DHC patients whether exposed to SNS information or not. Unintentional injuries were assessed using the following E-codes: E001.0-E029.9; E800-E844.5; E846-E870.0; E880-E903; E907-E929.9; E980-E989. The E-codes were then categorized to reflect the injury topics addressed by the DHC Safety Corner kiosk (see Table 1). Unique children were the unit of measure. If a single visit resulted in multiple E-codes identifying different target injuries, we identified the E-code that captured the primary cause of injury. For example, if a child fell down a flight of stairs on a tricycle, and the injury was coded as both a fall down stairs and a fall off of a bike, the E-code for fall down stairs was used, since a recommended stair gate may have prevented this injury. Only 3 such multiple E-code events occurred. Primary Exposure Measure. The primary exposure measure was completion of the SNS risk assessment at wellchild care visits.
Analysis Descriptive statistics were computed for the number of referrals to and completions of the SNS program. Fisher’s exact test was used to conduct comparisons of the distribution of injury types between the DHC and the CHW children, by age group. Injury rates per 1000 visits within age group for the DHC were computed as the number of injuries divided by the number of unique child visits to the DHC multiplied by 1000. The relative risk and its 95% confidence interval were calculated for the comparison of injuries in unscreened and screened children. Lastly, for all screened patients, injuries identified within the CHW system were assessed and compared with injury areas addressed in their SNS report.
Results Demographics Over the 7-month time period of the study there were 2091 eligible preventive health visits of children aged 0 to 4 years. There were 1084 kiosk completions by the 1406 patients provided a Prescription for Safety for a 77% kiosk completion rate. The first kiosk assessment for a child was used for the remainder of the analyses to prevent duplicate counting, resulting in 843 SNS assessments for the 1368 unique patients (62% assessed at least once). As shown in Figure 1, rates of referral and completion increased during the first 5 months of the program, and then stabilized. The kiosk survey was usually completed by mothers (84%) and those who described themselves as African American (77%). Forty-two percent reported that their highest level of education completed was a high school diploma or GED equivalent, and 68% indicated an annual household income of less than $10 000.
Screenings and Risks The highest number of screenings was for children ages 0 to 6 months (n = 419; 38% of the total), while the fewest occurred among children 4 years of age (n = 59; 5% of the total). Risks for injuries from falls were most commonly identified as 1 of the 2 highest risks for the first 2 years of life. For children in the first 6 months of life, 14% were identified as at risk for a fall from bed and 11% for a fall down stairs. This risk increased for infants 7 to 11 months to 28% for fall from bed and 14% at risk for a fall down stairs. Being at risk of a fall from a bed remained an elevated risk at 18% for 12- to 17-month-old toddlers, and 15% for 18-to 23-month-old children. After 6 months, child passenger safety risks were highly prioritized for all age groups (14% for 7- to 11-month-olds, 16% for 12- to 17-month-olds, 17% for 18- to 23-month-olds, 29% for 2-year-olds, 42% for 3-year-olds, and 41% for 4-year-olds). Choking risks were most often identified from 7 months until 2 years of age (15% for 7-11 months, 26% for 12-17 months, 33% for 18- to 23-month-olds, and 38% for 2-year-old children).
Injuries During the time period of the screenings, a total of 11 DHC children were injured and seen within the CHW system for an unintentional injury targeted for review by the DHC SNS implementation. Seven of the 843 screened children (0.8%) had a subsequent injury visit related to a SNS injury area (Table 2), while 4 of the 525 unscreened
Downloaded from cpj.sagepub.com at SUNY STONY BROOK on June 30, 2015
1386
Clinical Pediatrics 53(14)
Figure 1. Eligible visits, kiosk referrals, and kiosk completions by month. Table 2. Kiosk-Screened Children (DHC patients) With Safe N’ Sound Identified Risks and Subsequent Injury, by Age Group. Age Group in Months (Total Number Injured)
Screening Age Screened Date (Months)
0-6 (1) 7-11 (2)
8/17/11 11/3/11
1 7
12-17 (1)
8/22/11 9/13/11
3 12
18-23 (3)
7/28/11
17
9/23/11
19
10/6/11
20
Risks Identified
Age Injured (Months)
No risks Falls—walker, poisons Crib, hot water Falls—Stairs, Falls—walker Drowning, choking—food Car safety, drowning Falls—stairs, choking
5 7 8 16 19 21 22
Date of Injury
Injury (E-Code)
12/17/11 Fall from bed (E884.4) 11/14/11 Fall from bed (E884.4) 1/19/12 1/11/12
Fall from bed (E884.4) Motor vehicle collision NOSPassenger (E812.1) 9/22/11 Hot liquids & vapors, including. steam (E924.0) 11/26/11 Hot substance accident NEC (E924.8) 12/16/11 Accidental poisoning by corrosives and caustics NEC (E864.3)
Abbreviations: NOS, not otherwise specified; NEC, not elsewhere classified.
eligible children had a subsequent injury visit related to a SNS targeted injury (0.8%). Of the 7 children screened and injured there was no case of injury in a screened child that was related to the specific SNS identified priority risks for that child (Figure 2).
Discussion To our knowledge, this is the first effectiveness evaluation of the injury outcomes associated with the use of a tailored injury prevention program in a clinic setting. This study demonstrated that a kiosk-based prevention tool can be successfully incorporated as best practice
standard of care for a clinic. We were able to screen 77% of all patients receiving the Prescription for Safety. The percentage of referrals and completions by those eligible increased over time and appeared to become more institutionalized after 5 months. In addition, we used a standardized previsit form with a unique patient bar code to activate the kiosk and to link the Safe N’ Sound responses to the identified injury outcomes. Previous work implementing SNS in a clinical setting showed much lower reach (Weaver,14 2011). In a field trial without a research or practice infrastructure to facilitate integration, SNS completion averaged 10%. The greater reach in this application of the program speaks to several
Downloaded from cpj.sagepub.com at SUNY STONY BROOK on June 30, 2015
1387
Brixey et al
2091 Eligible Visits
1368 Unique Pa ents
843
525
SNS Assessments
No SNS Assessment
7
836
Injured
No Injury
4
521
Injuries
No Injuries
5 mo – Fall From Bed 7 mo – Fall From Bed
16 mo – Fall From Bed
8 mo – Fall From Bed
18 mo - Motor Vehicle Collision
16 mo - Motor Vehicle Collision 2 yr – Fall From Bed 19 mo - Burn 4 yr – Poisoning
21 mo - Burn 22 mo - Poisoning
Figure 2. Eligible Visits, Safe N’ Sound screenings and subsequent injuries (July 1, 2011 to January 31, 2013).
important factors that should be considered when implementing evidence-based programs. First, the use of the “Prescription for Safety” provided a credible cue for engaging parents with the program which helped make the use of the kiosk a commonplace and expected behavior. Second, the physicians and staff at the clinic were highly committed to the program. Several champions
helped foster this commitment by providing routine reports and encouraging dialogue about the organizational changes that were needed to accommodate the use of the program. Globally, we found a significant overlap between the SNS identified risks collectively and the ultimate injuries seen during the same time period. For children 0 to
Downloaded from cpj.sagepub.com at SUNY STONY BROOK on June 30, 2015
1388
Clinical Pediatrics 53(14)
6 months of age, burns and falls were both identified as a self-reported risk and as actual injuries. For 7- to 11-month-old infants falls from both bed and stairs showed congruency. For children aged 12 to 23 months, falls and child passenger safety were seen in both selfreported risks and actual injuries. There was limited overlap for children at age 2 years but by 3 years of age overlap was seen for car crash risks, and for 4-year-olds car crashes and burns were both identified in selfreported risks and actual injuries. Importantly, while the study design does not control for all factors that could account for these differences, the 7 children screened by SNS were not subsequently injured by a risk covered in their customized report. This may speak to the benefit of the tailored education provided but more research is needed. Injury risks also vary by developmental age; risks identified at 2 months (such as unsafe sleep in a crib) may no longer be relevant as a child matures. While another study used focused counseling with educational handouts and attempted to measure injuries by parent report and medical record data, there was limited enrollment and no significant difference in the number of injuries between groups was found.15 There are several limitations to this study. Although all children 0 to 4 years old who presented for their wellchild checkup were eligible, due to the emphasis of maintaining clinic flow, not all children were referred to the kiosk by the front desk staff or had the time to complete the survey prior to being called back into an examination room to be seen by their provider. Caregivers who were not fluent in English may not have been able to use the program. The kiosk identifies self-reported risks. Although there is concern that safety behaviors may be overestimated, there is data to support that selfreports of ownership of a car seat are reliable.16,17 There is an inexact match between the SNS questions and E-codes. As a pilot study there was not enough power in the current number of patients screened and injured to establish statistical significance. The SNS survey is designed to stop collecting data after identifying 2 “atrisk” responses; so data were not collected about each question at each visit. This limitation of SNS to only identify 2 risks is supported by clinical evidence of the amount of information an individual caregiver can absorb at any individual time and supports clinical efficiencies in a busy practice. This limitation of SNS to only identify 2 risks is supported by health education, but clearly limits the analysis. Although there was no child screened who was injured by the targeted risk identified it is not clear that there is causality as we did not always know the response for the targeted injury risk question due to the SNS program ceasing once 2 risks are identified.
Conclusion Our results, joined with the limited research on the use of electronic devices in other pediatric settings, should encourage the adoption of theory-driven electronic communication strategies to support families in making changes to improve child safety. We were able to successfully screen families and provide needed safety information that may impact injury outcomes with minimal impact on patient flow. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The author(s) received no financial support for the research, authorship and/or publication of this article.
References 1. Centers for Disease Control and Prevention. Vital signs: Unintentional injury deaths among persons aged 0-19 years—United States, 2000-2009. MMWR Morb Mortal Wkly Rep. 2012;61:270-276. 2. Web-based injury statistics query and reporting system. Centers for Disease Control and Prevention Web site. http://www.cdc.gov/ncipc/wisqars. Updated 2012. Accessed February 19, 2013. 3. Owens PL, Zodet MW, Berdahl T, Dougherty D, McCormick MC, Simpson LA. Annual report on health care for children and youth in the United States: focus on injury-related emergency department utilization and expenditures. Ambul Pediatr. 2008;8:219 -240. 4. Gardner HG, American Academy of Pediatrics Committee on Injury, Violence, and Poison Prevention. Office-based counseling for unintentional injury prevention. Pediatrics. 2007;119:202-206. 5. Gielen AC, McDonald EM, Forrest CB, Harvilchuck JD, Wissow L. Injury prevention counseling in an urban pediatric clinic. Analysis of audiotaped visits. Arch Pediatr Adolesc Med. 1997;151:146-151. 6. Bass JL, Christoffel KK, Widome M, et al. Childhood injury prevention counseling in primary care settings: a critical review of the literature. Pediatrics. 1993;92:544-550. 7. Barrios LC, Runyan CW, Downs SM, Bowling JM. Pediatric injury prevention counseling: an observational study of process and content. Patient Educ Couns. 2001;44:141-149. 8. Kendrick D, Coupland C, Mulvaney C, et al. Home safety education and provision of safety equipment for injury prevention. Cochrane Database Syst Rev. 2007;(1):CD005014. 9. Kendrick D, Groom L, Stewart J, Watson M, Mulvaney C, Casterton R. “Risk watch”: cluster randomised controlled
Downloaded from cpj.sagepub.com at SUNY STONY BROOK on June 30, 2015
1389
Brixey et al trial evaluating an injury prevention program. Inj Prev. 2007;13(2):93-98. 10. Agran PF, Winn D, Anderson C, Trent R, Walton-Haynes L. Rates of pediatric and adolescent injuries by year of age. Pediatrics. 2001;108:E45. 11. Agran PF, Anderson C, Winn D, Trent R, Walton-Haynes L, Thayer S. Rates of pediatric injuries by 3-month intervals for children 0 to 3 years of age. Pediatrics. 2003;111(6 pt 1):e683-e692. 12. Weaver NL, Williams J, Jacobsen HA, et al. Translation of an evidence-based tailored childhood injury prevention program. J Public Health Manag Pract. 2008;14:177-184. 13. Nansel TR, Weaver N, Donlin M, Jacobsen H, Kreuter MW, Simons-Morton B. Baby, Be safe: The effect of tailored communications for pediatric injury prevention
provided in a primary care setting. Patient Educ Couns. 2002;46:175-190. 14. Weaver NL, Nansel TR, Williams J, Tse J, Botello Harbaum M, Willson K. Reach of a kiosk-based pediatric injury prevention program. Transl Behav Med. 2011;1:515-522. 15. Sangvai S, Cipriani D, Colborn DK, Wald ER. Studying injury prevention: practices, problems and pitfalls in implementation. Clin Pediatr (Phila). 2007;46:228. 16. Uherick L, Gorelick MH, Biechler R, Brixey SN, MelzerLange M. Validation of two child passenger safety questionnaires. Inj Prev. 2010;16:343-347. 17. Robertson A, Rivara F, Ebel BE, Lymp JF, Christakis DA. Validation of parent self reported home safety practices. Inj Prev. 2005;11:209-212.
Downloaded from cpj.sagepub.com at SUNY STONY BROOK on June 30, 2015
