p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 2 5 2 e2 5 7
Available online at www.sciencedirect.com
Public Health journal homepage: www.elsevier.com/puhe
Original Research
Neighborhood socio-economic context and emergency department visits for dental care in a U.S. Midwestern metropolis L. Hong a, Y. Liu b,*, T.L. Hottel c, G.L. Hoff d, J. Cai e a
Department of Pediatric Dentistry and Community Oral Health, College of Dentistry, University of Tennessee Health Science Center, USA b Division of Pharmacy Practice and Administration, University of Missouri-Kansas City School of Pharmacy, USA c Department of Prosthodontics, College of Dentistry, University of Tennessee Health Science Center, USA d Department of Epidemiology, College for Public Health and Social Justice, St. Louis University, USA e Office of Epidemiology and Community Health, Kansas City Health Department, MO, USA
article info
abstract
Article history:
Objective: This study was to examine the association between emergency department (ED)
Received 25 April 2013
visits for dental complaints and neighborhood socio-economic contexts of patients in a
Received in revised form
U.S. Midwestern metropolis.
19 October 2014
Study design: A retrospective study.
Accepted 28 November 2014
Methods: Deidentified data of ED visits for the period 2001e2010 from all facilities serving
Available online 19 February 2015
Kansas City, Missouri and 2007e2011 American Community Survey 5-Year Estimates data were used to determine odds of visit by neighborhood socio-economic characteristics at
Keywords:
the ZIP code level. ED visits with diagnoses of International Classification of Disease 9th
Emergency department
Revision codes related to toothache or tooth injury were included. ZIP code characteristics
Dental care
included percent of non-white population, median family income, percent of population 25
Neighborhood characteristics
years and older with at least a high school degree, and percent of population with a lan-
Health inequalities
guage other than English spoken at home. Each ZIP code characteristic was divided into quartiles. Chi-square tests and two-level hierarchical linear modeling (HLM) were conducted. In the HLM, the outcome variable was whether to have an ED visit for dental complaints (yes/no), the first-level variables were characteristics of individual ED visits, and the second-level variables were ZIP code characteristics. Results: The study population made 1,786,939 ED visits, of which 35,136 (1.9%) were for dental complaints. Among the patients making ED visits for dental complaints, 54.8% were female, 51.9% were younger adults aged 19e35 years, 48.7% were non-Hispanic black, and 35.5% used self-pay as the source of payment. After controlling the first-level variables, the HLM showed that the risk of ED visits for dental complaints significantly increased for individuals residing in ZIP Code Tabulation Areas with lower median family income, or a higher proportion of the population with a language rather than English spoken at home.
* Corresponding author. 4248 Health Sciences Building, University of Missouri-Kansas City School of Pharmacy, 2464 Charlotte Street, Kansas City, MO, 64108, USA. Tel.: þ1 8162356820; fax: þ1 8162356008. E-mail address: [email protected] (Y. Liu). http://dx.doi.org/10.1016/j.puhe.2014.11.014 0033-3506/© 2014 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 2 5 2 e2 5 7
253
Conclusions: Among socio-economic characteristics examined, median family income and percent of population with a language other than English spoken at home are important indicators of ED visits for dental complaints. © 2014 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.
Introduction Increasing use of the hospital emergency department (ED), often accompanied by overcrowding, has become a nationwide challenge to timely and efficient delivery of emergency care in the U.S.1e4 A major contributor to the growth is the increased use by the individuals classified as non-urgent or semiurgent.1,4,5 While it is debatable about what constitutes a non-urgent visit to the ED and whether such visit is inappropriate,4e6 EDs are neither the most appropriate setting for dental care nor are they generally equipped to provide definitive treatment for dental conditions.7e9 ED visits for dental complaints have continued to increase in recent decades with a greater proportion of these visits being for nontraumatic dental problems and largely avoidable by preventive dental visits and early intervention.10e13 In California, the ED visits for preventable dental conditions grew by 12% between 2005 and 2007 e a rate faster than the population growth in the state.14 It has been reported that, during years 2001e2006, Kansas City, Missouri, residents made 19,316 visits to EDs for dental complaints, accounting for 1.7% of all ED visits; there was also a significant increase in dental complaint visits over the six-year period as a proportion of total ED visits.15 Reasons for increased presentation of dental complaints to the ED include lack of insurance, inability to afford the cost or out-of-pocket expense for dental care, not having a usual source of dental care, difficulty in obtaining a dental appointment, unavailability of dental providers during weekend and afterhours, and lack of dental providers willing to accept Medicaid patients.9,11,14,15 Compared to ED visits for other reasons, far fewer visits with dental complaints result in a procedure performed.13,15,16 Most studies that have examined ED visits for dental complaints are based on the experiences at specific institutions or specific sub-groups of the population such as Medicaid beneficiaries, children, and low-income adults.7e13,17,18 Only one study has examined U.S. national data using the 1997 to 2000 National Hospital Ambulatory Medical Care Survey.15 While these studies focused mainly on the demographic and socio-economic characteristics, little is known about the places these patients come from. There has been a growing recognition of the association between the characteristics of the places where people live and their health and health behaviors.19,20 These contextual characteristics have also been shown to affect ED utilization in general.21e23 While the published literature provides important information about the characteristics of individuals who are likely to visit the ED for dental complaints, knowledge of the association with neighborhood
characteristics may allow for more effective policy formulation and interventions. In this study, the association between the socio-economic characteristics of areas defined by ZIP codes within the U.S. and the ED use for dental complaints at all hospitals in the city over a ten-year period were investigated.
Methods This study was based on the ED visits made by the residents of Kansas City, Missouri (KCMO). KCMO, with a population of 480,129 and covering 318 square miles, is the largest city in Missouri and the anchor of the Kansas City bi-state metropolitan area. The final population numerator used, however, was 559,045 as two of the ZIP Code Tabulation Areas (ZCTAs) in this study extended beyond the corporate limits of KCMO.
Study design Annually, the Missouri Department of Health and Senior Services provides the KCMO Health Department a deidentified electronic file of all ED visits within the city limits; these records are from the Missouri Patient Abstract System. Data included in this retrospective analysis were extracted by the KCMO Health Department for the period 2001e2010. Data variables were at the ED visit level, which included the patients' demographic characteristics, ZIP codes and counties of residence (KCMO is located in portions of four Missouri counties), principal and other diagnoses, principal and other procedures, total billed charges, and expected source of payment.
Setting and data selection KCMO residents were selected by the ZIP codes of residence reported in the patient abstract system. All 51 5-digit ZCTAs with the 3-digit prefix 641 were included. A visit was considered to be for dental complaints if the primary diagnosis reported was one or more of the following International Classification of Disease, 9th Revision, Clinical Modification (ICD-9-CM) codes: 520e529, 830, 830.1, 848.1, 873.51, 873.53, 973.54, 873.63e873.69, and 873.73. Socio-economic characteristics of the ZCTAs obtained from 2007 to 2011 American Community Survey 5-Year Estimates were: percent of non-white population, median family income, percent of population 25 years and over with high school education or higher, and percent of population
254
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 2 5 2 e2 5 7
with a language other than English spoken at home. These factors are important neighborhood characteristics and some of them are commonly used in health outcome research, such as race, income, education, and employment rate.9 Details of the area characteristics are displayed in Table 1. The authors hypothesized that these characteristics would impact on dental ED visits.
Data analyses Characteristics of ED visits were expressed as percentages, and quartiles were created within each area socio-economic characteristic variable. Bivariate associations between ZCTA socio-economic characteristics and ED visits for dental complaints were assessed by Chi-square tests. Although they were at the zip code level, area characteristics were extrapolated to the ED visit level. That is, all ED visits in one zip code were assigned the value of an area characteristic for that zip code. Two-level hierarchical linear modeling (HLM) was conducted. In the HLM, the outcome variable was whether to have an ED visit for dental complaints (yes/no). The first-level variables were characteristics of individual ED visits, including sex, age, race, and expected payment source. Each characteristic of ED visits had two or more categories. The secondlevel variables were ZCTA characteristics, including percent of non-white population, the median family income, percent of population 25 years and older with at least a high school degree, and percent of population with a language other than English spoken at home. Because all first-level and secondlevel variables were dichotomous variables with non-normal distribution, the estimation with robust standard error was used. The HLM was carried out using HLM 7.01 (Scientific Software International, Inc., Skokie, Illinois, USA.), and other statistical analyses were carried out using SPSS 18.0 (SPSS, Inc., Chicago, Illinois, USA.).
Results During the period 2001e2010, according to the data of ED visits, the study population made 1,786,939 ED visits, of which 35,136 (1.9%) were for dental complaints. Using the number of ED visits each year and the population number, the average annual rate of ED visits for dental complaints was 6.32 per 1000 persons. Based on data of ED visits, characteristics of the visits for dental complaints along with corresponding characteristics of
Table 1 e Area characteristics of the 51 ZCTAs. ZCTA characteristic
Mean
SD
Min
Max
Non-white population (%) 37.6 28.7 0 100 Median family income ($) 61,637 30,570 4873 144,037 Population 25 years and 85.4 12.3 56.4 100 over: high school graduate or higher (%) Population five years and over: 10.6 11.8 0 56.7 language other than English spoken at home (%) Source: 2007e2011 American Community Survey 5-Year Estimates.
Table 2 e Characteristics of ED visits in KCMO, 2001e2010. Characteristic
ED visits for dental ED visits for other complaints % complaints % (n ¼ 35,136) (n ¼ 1,751,803)
Sex Female Male Age (years) 0e18 19e35 36e50 51þ Race/ethnicity White Black Hispanic Other Expected payment source Private insurance Self-pay Medicaid Medicare Other/unknown
54.8 45.2
55.2 44.8
12.5 51.9 27.6 8.0
26.3 32.5 22.9 18.3
43.7 48.7 4.5 3.1
41.8 47.6 5.2 5.4
15.5 35.5 32.7 5.5 10.8
27.5 21.6 31.8 10.6 8.5
All percentages refer to the category totals within the respective columns.
visits for other reasons are shown in Table 2. Among the patients making ED visits for dental complaints, 54.8% were female, 51.9% were younger adults aged 19e35 years, 43.7% were non-Hispanic white, and 48.7% were non-Hispanic black. Selfpay (35.5%) was the most common expected source of payment. Further details of individual visit characteristics and trends in ED visits were reported elsewhere.15 Using 2007e2011 American Community Survey 5-Year Estimates, the residents of ZCTAs with less than 19.5% nonwhite population (1st quartile) made 1.2% of their ED visits for dental complaints, which was significantly lower than other quartiles (Table 3). ZCTAs with a median family income of less than $36,943 (1st quartile) had the highest proportion (2.2%) of ED visits for dental complaints among all quartiles. Similarly, the residents of ZCTAs with less than 78.7% of the population 25 years or older with high school graduation or higher (1st quartile) had the highest proportion (2.2%). But for populations with language other than English spoken at home, there was not a clear trend. Chi-square tests showed a significant association between these area characteristics and ED visits for dental care (all P-values
Available online at www.sciencedirect.com
Public Health journal homepage: www.elsevier.com/puhe
Original Research
Neighborhood socio-economic context and emergency department visits for dental care in a U.S. Midwestern metropolis L. Hong a, Y. Liu b,*, T.L. Hottel c, G.L. Hoff d, J. Cai e a
Department of Pediatric Dentistry and Community Oral Health, College of Dentistry, University of Tennessee Health Science Center, USA b Division of Pharmacy Practice and Administration, University of Missouri-Kansas City School of Pharmacy, USA c Department of Prosthodontics, College of Dentistry, University of Tennessee Health Science Center, USA d Department of Epidemiology, College for Public Health and Social Justice, St. Louis University, USA e Office of Epidemiology and Community Health, Kansas City Health Department, MO, USA
article info
abstract
Article history:
Objective: This study was to examine the association between emergency department (ED)
Received 25 April 2013
visits for dental complaints and neighborhood socio-economic contexts of patients in a
Received in revised form
U.S. Midwestern metropolis.
19 October 2014
Study design: A retrospective study.
Accepted 28 November 2014
Methods: Deidentified data of ED visits for the period 2001e2010 from all facilities serving
Available online 19 February 2015
Kansas City, Missouri and 2007e2011 American Community Survey 5-Year Estimates data were used to determine odds of visit by neighborhood socio-economic characteristics at
Keywords:
the ZIP code level. ED visits with diagnoses of International Classification of Disease 9th
Emergency department
Revision codes related to toothache or tooth injury were included. ZIP code characteristics
Dental care
included percent of non-white population, median family income, percent of population 25
Neighborhood characteristics
years and older with at least a high school degree, and percent of population with a lan-
Health inequalities
guage other than English spoken at home. Each ZIP code characteristic was divided into quartiles. Chi-square tests and two-level hierarchical linear modeling (HLM) were conducted. In the HLM, the outcome variable was whether to have an ED visit for dental complaints (yes/no), the first-level variables were characteristics of individual ED visits, and the second-level variables were ZIP code characteristics. Results: The study population made 1,786,939 ED visits, of which 35,136 (1.9%) were for dental complaints. Among the patients making ED visits for dental complaints, 54.8% were female, 51.9% were younger adults aged 19e35 years, 48.7% were non-Hispanic black, and 35.5% used self-pay as the source of payment. After controlling the first-level variables, the HLM showed that the risk of ED visits for dental complaints significantly increased for individuals residing in ZIP Code Tabulation Areas with lower median family income, or a higher proportion of the population with a language rather than English spoken at home.
* Corresponding author. 4248 Health Sciences Building, University of Missouri-Kansas City School of Pharmacy, 2464 Charlotte Street, Kansas City, MO, 64108, USA. Tel.: þ1 8162356820; fax: þ1 8162356008. E-mail address: [email protected] (Y. Liu). http://dx.doi.org/10.1016/j.puhe.2014.11.014 0033-3506/© 2014 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 2 5 2 e2 5 7
253
Conclusions: Among socio-economic characteristics examined, median family income and percent of population with a language other than English spoken at home are important indicators of ED visits for dental complaints. © 2014 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.
Introduction Increasing use of the hospital emergency department (ED), often accompanied by overcrowding, has become a nationwide challenge to timely and efficient delivery of emergency care in the U.S.1e4 A major contributor to the growth is the increased use by the individuals classified as non-urgent or semiurgent.1,4,5 While it is debatable about what constitutes a non-urgent visit to the ED and whether such visit is inappropriate,4e6 EDs are neither the most appropriate setting for dental care nor are they generally equipped to provide definitive treatment for dental conditions.7e9 ED visits for dental complaints have continued to increase in recent decades with a greater proportion of these visits being for nontraumatic dental problems and largely avoidable by preventive dental visits and early intervention.10e13 In California, the ED visits for preventable dental conditions grew by 12% between 2005 and 2007 e a rate faster than the population growth in the state.14 It has been reported that, during years 2001e2006, Kansas City, Missouri, residents made 19,316 visits to EDs for dental complaints, accounting for 1.7% of all ED visits; there was also a significant increase in dental complaint visits over the six-year period as a proportion of total ED visits.15 Reasons for increased presentation of dental complaints to the ED include lack of insurance, inability to afford the cost or out-of-pocket expense for dental care, not having a usual source of dental care, difficulty in obtaining a dental appointment, unavailability of dental providers during weekend and afterhours, and lack of dental providers willing to accept Medicaid patients.9,11,14,15 Compared to ED visits for other reasons, far fewer visits with dental complaints result in a procedure performed.13,15,16 Most studies that have examined ED visits for dental complaints are based on the experiences at specific institutions or specific sub-groups of the population such as Medicaid beneficiaries, children, and low-income adults.7e13,17,18 Only one study has examined U.S. national data using the 1997 to 2000 National Hospital Ambulatory Medical Care Survey.15 While these studies focused mainly on the demographic and socio-economic characteristics, little is known about the places these patients come from. There has been a growing recognition of the association between the characteristics of the places where people live and their health and health behaviors.19,20 These contextual characteristics have also been shown to affect ED utilization in general.21e23 While the published literature provides important information about the characteristics of individuals who are likely to visit the ED for dental complaints, knowledge of the association with neighborhood
characteristics may allow for more effective policy formulation and interventions. In this study, the association between the socio-economic characteristics of areas defined by ZIP codes within the U.S. and the ED use for dental complaints at all hospitals in the city over a ten-year period were investigated.
Methods This study was based on the ED visits made by the residents of Kansas City, Missouri (KCMO). KCMO, with a population of 480,129 and covering 318 square miles, is the largest city in Missouri and the anchor of the Kansas City bi-state metropolitan area. The final population numerator used, however, was 559,045 as two of the ZIP Code Tabulation Areas (ZCTAs) in this study extended beyond the corporate limits of KCMO.
Study design Annually, the Missouri Department of Health and Senior Services provides the KCMO Health Department a deidentified electronic file of all ED visits within the city limits; these records are from the Missouri Patient Abstract System. Data included in this retrospective analysis were extracted by the KCMO Health Department for the period 2001e2010. Data variables were at the ED visit level, which included the patients' demographic characteristics, ZIP codes and counties of residence (KCMO is located in portions of four Missouri counties), principal and other diagnoses, principal and other procedures, total billed charges, and expected source of payment.
Setting and data selection KCMO residents were selected by the ZIP codes of residence reported in the patient abstract system. All 51 5-digit ZCTAs with the 3-digit prefix 641 were included. A visit was considered to be for dental complaints if the primary diagnosis reported was one or more of the following International Classification of Disease, 9th Revision, Clinical Modification (ICD-9-CM) codes: 520e529, 830, 830.1, 848.1, 873.51, 873.53, 973.54, 873.63e873.69, and 873.73. Socio-economic characteristics of the ZCTAs obtained from 2007 to 2011 American Community Survey 5-Year Estimates were: percent of non-white population, median family income, percent of population 25 years and over with high school education or higher, and percent of population
254
p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 2 5 2 e2 5 7
with a language other than English spoken at home. These factors are important neighborhood characteristics and some of them are commonly used in health outcome research, such as race, income, education, and employment rate.9 Details of the area characteristics are displayed in Table 1. The authors hypothesized that these characteristics would impact on dental ED visits.
Data analyses Characteristics of ED visits were expressed as percentages, and quartiles were created within each area socio-economic characteristic variable. Bivariate associations between ZCTA socio-economic characteristics and ED visits for dental complaints were assessed by Chi-square tests. Although they were at the zip code level, area characteristics were extrapolated to the ED visit level. That is, all ED visits in one zip code were assigned the value of an area characteristic for that zip code. Two-level hierarchical linear modeling (HLM) was conducted. In the HLM, the outcome variable was whether to have an ED visit for dental complaints (yes/no). The first-level variables were characteristics of individual ED visits, including sex, age, race, and expected payment source. Each characteristic of ED visits had two or more categories. The secondlevel variables were ZCTA characteristics, including percent of non-white population, the median family income, percent of population 25 years and older with at least a high school degree, and percent of population with a language other than English spoken at home. Because all first-level and secondlevel variables were dichotomous variables with non-normal distribution, the estimation with robust standard error was used. The HLM was carried out using HLM 7.01 (Scientific Software International, Inc., Skokie, Illinois, USA.), and other statistical analyses were carried out using SPSS 18.0 (SPSS, Inc., Chicago, Illinois, USA.).
Results During the period 2001e2010, according to the data of ED visits, the study population made 1,786,939 ED visits, of which 35,136 (1.9%) were for dental complaints. Using the number of ED visits each year and the population number, the average annual rate of ED visits for dental complaints was 6.32 per 1000 persons. Based on data of ED visits, characteristics of the visits for dental complaints along with corresponding characteristics of
Table 1 e Area characteristics of the 51 ZCTAs. ZCTA characteristic
Mean
SD
Min
Max
Non-white population (%) 37.6 28.7 0 100 Median family income ($) 61,637 30,570 4873 144,037 Population 25 years and 85.4 12.3 56.4 100 over: high school graduate or higher (%) Population five years and over: 10.6 11.8 0 56.7 language other than English spoken at home (%) Source: 2007e2011 American Community Survey 5-Year Estimates.
Table 2 e Characteristics of ED visits in KCMO, 2001e2010. Characteristic
ED visits for dental ED visits for other complaints % complaints % (n ¼ 35,136) (n ¼ 1,751,803)
Sex Female Male Age (years) 0e18 19e35 36e50 51þ Race/ethnicity White Black Hispanic Other Expected payment source Private insurance Self-pay Medicaid Medicare Other/unknown
54.8 45.2
55.2 44.8
12.5 51.9 27.6 8.0
26.3 32.5 22.9 18.3
43.7 48.7 4.5 3.1
41.8 47.6 5.2 5.4
15.5 35.5 32.7 5.5 10.8
27.5 21.6 31.8 10.6 8.5
All percentages refer to the category totals within the respective columns.
visits for other reasons are shown in Table 2. Among the patients making ED visits for dental complaints, 54.8% were female, 51.9% were younger adults aged 19e35 years, 43.7% were non-Hispanic white, and 48.7% were non-Hispanic black. Selfpay (35.5%) was the most common expected source of payment. Further details of individual visit characteristics and trends in ED visits were reported elsewhere.15 Using 2007e2011 American Community Survey 5-Year Estimates, the residents of ZCTAs with less than 19.5% nonwhite population (1st quartile) made 1.2% of their ED visits for dental complaints, which was significantly lower than other quartiles (Table 3). ZCTAs with a median family income of less than $36,943 (1st quartile) had the highest proportion (2.2%) of ED visits for dental complaints among all quartiles. Similarly, the residents of ZCTAs with less than 78.7% of the population 25 years or older with high school graduation or higher (1st quartile) had the highest proportion (2.2%). But for populations with language other than English spoken at home, there was not a clear trend. Chi-square tests showed a significant association between these area characteristics and ED visits for dental care (all P-values
