Distances to emergency departments and non-urgent utilization of medical services: a systematic review

Figures & data

Table 1. Ovid Medline® ALL search strategy.

Concept	MeSH (controlled vocabulary)	Natural language keywords
Emergency department	exp Emergency Service, Hospital/or exp Emergency Medical Services/or exp After-Hours Care/	(emergency adj4 (department* or room* or ward* or unit* or service*)) or ((health or medical) adj5 emergenc*) or ER
Non-Urgent use	exp health services misuse/	(non-urgent or nonurgent or non-traumatic or nontraumatic or non-emergency or nonemergency) or (avoidable or inappropriate)
Distance	exp Travel/	(travel or distance or transport*)

Table Key.

*truncation of word.

adj is adjacency search (e.g. adj4 means within four words, adj5 within 5 words)

Figure 1. Proposed conceptual framework adapted from the Andersen and Newman’s Model of health services use.

Table 2. Key findings from the reviewed studies.

Author/Country, Year	Design/Patient or Ecological level	Aim of the study	Study period (Months/Year)	Sample description	Final number of participants	Outcome variable	Definition of non-urgent	Distance Measurement	Key findings	Potential Confounders in the multivariable analysis^^
Valent & Busolin, Italy, 2018 [9]	Cross sectional/Pediatric ecological study	To assess how distance from home to the closest PED affects non-urgent visits among children.	2014	Children from 3 regional pediatric EDs <16 years with a white triage tag and residing within 20 kms^ of the ED	10,327	Non-urgent ED utilization rate	White triage tags	Patient home to ED	1)The proportion of ED non-urgent visits accounted for 24.7% of all visits. 2) The rate of non-urgent visits decreased significantly from the innermost distance band (163.5 ED visits/1,000 person-years) to outermost distance band. (26.6 ED visits per 1000 person years).	None
Hilker, US, 1978 [30]	Cross sectional/Pediatric patient-level study	Examine factors to explain why parents take their children to hospital EDs.	One week each during January and February, 1977.	Randomly selected parents of children visiting a single children’s hospital during two weeks in a 1-year duration	652 parents, and children	Travel time	Patient to whom a delay of 24 hours would make no appreciable difference in the clinical condition.	Travel time	1) Of the non-emergency visits, 61% of the children took less than 15 minutes to get to the ED.	None
Pehlivanturk-Kizilkan et al. Turkey, 2022 [31]	Cross sectional/Pediatric patient-level study	To evaluate the factors that might have an effect on non-urgent PED visits and parental overestimation of emergency severity.	January-June 2015	Children patients who visited a district state hospital during the shifts of 2 physicians (who also assessed the emergency status of all patients over a 6-month duration)	974 patients	Non-urgent ED visits	Pediatric Canadian Triage and Acuity Scale (PCTAS)	ED to patient home	68% of the ED visits were classified as non-urgent. 2) 92.6% of the children lived close to the ED i.e. inside the district border. 3) Distance decay effect was not significant.	Age, Gender, Presence of chronic disease, Time of admission Season of admission, Parents’ age, Parents’ literacy, Father’s working status Admission to ED in past week, Total number of children, Transportation method
Benahmed et al. Belgium, 2012 [32]	Cross sectional/ Pediatric patient-level study	To evaluate the magnitude of non-urgent use of the ED for pediatric patients and identify associated factors.	2 weeks in October and November 2010	All children <15 years of age were randomly selected across 12 Belgian hospitals	3,117	Non-urgent utilization rate	ED attendance was appropriate when at least, one of the following criteria was met: need for a short-stay observation, need for technical examination (X-rays, blood testing, etc.) or orthopedic/surgical treatment, in-patient admission, death, referral by doctor or police, or brought by ambulance.	Travel time from patient home to ED	1) The non-urgent utilization was 39.9%. 2) 87.6% of those visiting the ED for non-urgent reasons took a travel time of less than 30 minutes. 3)The effect of distance on ED use was negative and significant. 4) Those whose travel time was less than 30 minutes, had higher odds (OR = 1.7 [95% CI = 1.3–2.2) of visiting the ED than those with travel times of 30 minutes or more.	Age, Disadvantaged family Hospital setting Out-of-hours visit Registered family doctor
Guckert et al. Germany, 2022 [33]	Cohort study/ Pediatric patient-level study	To determine factors leading to a non-urgent presentation at the ED before and during the first peak of the COVID-19 pandemic.	January-December 2017; March-April 2020	Children 16 years or less visiting a single university hospital and residing within a geographic distance between 0–20 kms.	5038 unscheduled visits	Non-urgent ED visits	Manchester Triage System Algorithm (MTS)	Pediatric ED to the patient’s zip code area’s geographical center	1) 69% of non-urgent visits among pediatric population 2) Non-urgent ED use decreased by 68% from pre-COVID to first peak of the COVID-pandemic 3) 50% of the patients traveled less than 5 kms for non-urgent visits. 4) For children with non-urgent presentations, median distance to the pediatric ED was 5.4 km 5) The distance decay effect was negative and significant.	Age, Sex, Absence of pre-existing conditions, ED visit during PCP office hours.
Wartman et al. US, 1984 [34]	Other: Mixed methods: cross-sectional and interviews/All ages patient level study	To characterize the leavers by examining in detail socioeconomic factors, medical problems, and distance traveled.	One year period	All patients presenting to a university affiliated hospital ED.	335 matched pairs of leavers and stayers/Interviews with 100 leavers	Leaver is identified as one who leaves without treatment.	Non-urgent complaints were graded by level (1–6) according to a classification developed by the ED based on methods and time needed for treatment.	ED to patient home	1) Leavers lived within 2.5 miles of the hospital. 2) 68% percent lived in the first five concentric zones (or within 2 ½ miles of the hospital) compared to 53% of stayers. 3) Half of the leavers living within 2.5 miles of the hospital presented with a mild illness. 4) Multiple discriminant analysis found distance to be one of the distinguishing variables between leavers and stayers.	None
Bianco et al. Italy, 2003 [35]	Cross sectional, Adult patient-level study	Determine the extent of non-urgent visits and effect of different characteristics on such visits.	July 2001-December 2001	Patients (≥15 years) waiting for care in an ED of a 714-bed hospital. Stratified random sample was selected to include 2-h sessions during two weekdays and one weekend for each week of the study period.	500 patients, relatives provided information for 41 patients.	Non-urgent utilization rate	A non-urgent case was defined as one where the patient has no active symptoms or they were recent and minor, without any feeling of emergency and he/she desires a check-up, a prescription refill or a return-to-work release.	ED to patient home	1) 19.6% of the people had non-urgent conditions. 2) The percent of non-urgent visitors declined with increasing distance bands. 3) For non-urgent visits, 23.1% traveled 5 kms or less versus 20% traveled over 35 kms. 4) In the multivariate analysis, the effect of distance on ED use was negative [OR = 0.88, 95% CI 0.65–1.18], but not significant.	Age, Sex, Chronic disease, Day of the week, Education, Duration of presenting problem, Referral to the ED, Number of persons in a household
Naouri et al. France, 2020 [5]	Cross sectional/Adult patient-level study	To explore socioeconomic and geographical determinants of inappropriate ED use in France.	All visits that took place on June 11 (Tuesday), 2013	All patients ≥15 years who visited a French ED during the 24-hour across 734 EDs	29,407 patients	Inappropriate ED use	Three measures of ED use appropriateness: Appropriate Use, Possible GP Use, Resource Utilization.	ED to patient home	1) Based on three methods, 23.6%, 27.4%, 13.5% were considered to be inappropriate ED use, and 6.16% to be inappropriate according to all three measures. 2) 63–66% who visited the ED for inappropriate use traveled 10 kms or less. 3) Distance decay effect was negative and significant. 4) When compared to those living 10 kms or less, those living more than 10 kms away, were less likely to use the ED inappropriately on the three measures of appropriateness: [OR = 0.89***, 95% CI 0.82–0.95]; [OR = 0.85***, 95% CI 0.79–0.91]; [OR = 0.83***, 95% CI 0.76–0.91];	Sex, Age, Insurance type, Having supplemental insurance Having a GP, Employment status, Education, Place of residence, Onset of complaint before the day of the ED visit, Time of visit, Type of hospital, Annual number of ED visits, County medical density.
Oh et al. Korea, 2018 [36]	Cohort/ Adult patient level study	Analyze the demographic, cancer-related and clinical characteristics associated with cancer patients visiting the ED.	January-December, 2016	Adult patients ≥19 years diagnosed of cancer visiting a tertiary care hospital	4,346 patients	Avoidable ED visit	A visit to the ED with a problem that could be resolved at a primary care office, out-patient clinic, or over the telephone.	ED to patient home	1) About 55.7% of the visits were avoidable, and 90.5% of the visits were primary-care treatable. 2) Approximately 38.9 kms were traveled for avoidable visits. 3) The effect of distance on ED use was negative and significant. 4) With a 1-km longer distance to home from ED, the probability of avoidable ED visits dropped by 0.2% (odds ratio [OR] 0.998, 95% CI: 0.997–0.999]	Gender, BMI Type and status of cancer, Chief complaint, ED visit in the last one year, Stay in the ED, Marital status, Final education level, Laboratory and imaging tests in ED.
Chen et al. US, 2015 (South Carolina) [10]	Longitudinal/All ages patient level study	To investigate whether convenience (proxied by travel distance to the hospital ED and to the closest Federally Qualified Health Center (FQHC) is associated with non-urgent ED use.	2006–2010	ED visits in South Carolina between 2005–2010.	6,592,501 ED visits	Probability associated with a NEPCT (Non-urgent or Primary Care Treatable) visit.	New York University – Emergency Department Algorithm (NYU-EDA)	Estimated shortest road distance between patient’s address and ED facility/FQHC visited.	1) The average distance traveled to the hospital ED was 11.685 miles, while that to a FQHC was 10.610 miles. 2) As distance increased, the probability of using the ED for NEPCT visit declined. 3) When compared across different insurance types, the slope coefficients were highest for commercial insurance (β = ‒0.0432, SE = 0.007), followed by self-pay (β = ‒0.0699, SE = 0.007), and those with other payment (β = ‒0.103, SE = 0.02).	Gender, Age, Race/Ethnicity, Insurance type, FQHC location Day of the week^~
Chen et al. US, 2015 (California) [37]	Longitudinal, All ages patient-level study	To examine if travel distances to the ED are associated with avoidable ED utilization and whether Medicaid patients’ avoidable ED utilization patterns differ when a FQHC exists within half a mile of patients’ zip code population centroid.	2006–2010	~7.2% of all ED visits randomly selected in California.	3,912,676 ED visits	Probability associated with non-emergent (NE) visits	New York University – Emergency Department Algorithm (NYU-EDA)	Estimated distance between ED facility or nearest FQHC and the 2008 population centroid of patient’s zip code	1) The average distance traveled to the ED and FQHC was 5.2 miles, and 4.8 miles respectively 2) 23.3% of the ED visits were classified as NE in nature. 4) For Medicare and Medicaid patients, for those who had to travel a short distance to get to the ED, there was a 0.44 [SE = 0.24] and 0.45 [SE = 0.22] percent points change respectively associated with NE care. 5) For Medicaid patients, if a FQHC was located within 0.5 miles of the population centroid of patient’s zip code, there was a reduction of 0.38% points [SE = 0.14] associated with NE care.	Age Gender Race or Ethnicity, Insurance, FQHC location, Year and hospital fixed effects^~
Vaz et al. Portugal, 2014 [38]	Longitudinal, Adult ecological study	To test for the Accident & Emergency (A&E) distance decay effects on the demand for ED in Portugal, and other determinants of ED utilization in Portugal.	January-June 2011, January-June 2012	ED visits of adults ≥18 years from 3 NHS hospitals (basic, medical-surgical, polyvalent EDs)	229,528 visits and 1,490 frequesias	ED utilization rate	Manchester Triage classification	ED to frequesias of patient’s residential address	1) Distance had a negative and significant effect on ED use across the three kinds of EDs examined. 2) For all low severity visits, across the three kinds of EDs, the Incidence Rate Ratios (IRR) were as: Basic ED: IRR= −1.29***; Med-Sur ED:IRR = −2.24***, Polyvalent ED:IRR = −1.27***.	Mean Age, Male population Type of ED Availability of doctors, primary health care, modern health centers, Income, School dropout rate Unemployment rate, Homelessness, Over-crowding of homes Mortality rate, Live births, Psychiatric care Ophthalmological care, Severity or Triage Year
Okunseri et al. US, 2016 [15]	Longitudinal/All ages ecological study	To examine the role of distance in the use of EDs for NTDCs (Non-traumatic dental conditions)	2001–2009	NTDCs visits to ED of the Medicaid population of Wisconsin enrolled in fee-for-service program or managed care organization	4,565,594 person years 80,742 ED visits.	ED visits per person-year	NTDC of first primary diagnosis	Medicaid enrollee’s location to the nearest ED	1) The ED visits per person-year was 17.7 visits per 1000 person years. 2) The median distance to the nearest ED was 2.9 miles. 3) As distance increased, the risk ratio (RR) associated with ED use declined indicative of a distance decay. 4) Compared to enrollees living less than 0.5 miles from the nearest emergency department, those living 3 miles away [RR = 0.88, 95% CI = 0.96–0.84] or further had significantly lower rates of NTDC visits to ED.	Sex^~, Age^~, Race or Ethnicity, Location, Year, Dentist-Population Ratio, Urbanization category
Ingram et al. Canada, 1978 [39]	Cross sectional/All-ages ecological study	Examine the patient’s decision to visit the ED, and study the effect of distance on the pattern of ED utilization	1 May 1975–2 June 1975	A randomly selected sample of ED visits from Humber Memorial Hospital	105	Patient visits per 1,000 population	Patient and medical assessment	Estimated distance between ED to patient’s census tract center	1) The average distance traveled to the ED was 2.37 miles. 2) Over 54% of those who assessed their problem to be non-urgent resided within 3 miles of the hospital 3) The effect of distance on ED use for all visits was negative and significant (β=-0.477 SE = 0.2796].	None
McGarry et al. US, 2023 [40]	Cohort/longitudinal adult ecological study	To examine the relationship between distance to the nearest hospital and Assisted Living (AL) residents’ rate of ED use.	2018–2019	Treat and release ED visits among Medicare beneficiaries aged ≥55 years residing in AL communities spending at least 7 days in 2018 or 2019.	540,944 resident years	ED utilization rate per 1000 AL Resident years	NYU ED Algorithm	Estimated shortest distance between AL and nearest US hospital	The median distance from AL facility to the nearest hospital was 2.5 miles or 4 kms. 2) The distance decay effect was negative and significant. 3) Doubling the distance from nearest hospital was associated with 3% [95% CI: −4.1%, −1.9%] decrease in NE visits.	Beneficiary level controls^~, Assisted living characteristics^~ (e.g. bed size; length of stay) Year fixed effects~

***: Statistically significant at 5% level of significance; ^ kms represents kilometers, where 1 mile = 1.609 kms, ^^ Bold letters signify a known statistical significance of the confounder, ^~ Statistical significance not known, ED: Emergency Department, PED: Pediatric Emergency Department, NEPCT: Non-Emergent or Primary Care Treatable visit.

Table 3. Quality of evidence and risk of bias classification of studies based on GRADE criteria.

Table 4. List of the assessment criteria and studies that were graded as satisfactory based on each assessment criterion.

Criterion	Satisfactory	(% of studies)
Clearly stated aims	[5, 9, 10, 15, 32-42]	100%
Appropriateness of design to meet the aims	[5, 9, 10, 15, 32-42]	100%
Adequate specification of subject group given	[5, 9, 10, 15, 32-42]	100%
Justification of sample size	[5, 9, 10, 15, 32-32]	100%
Likelihood of reliable and valid measurements	Unable to assess	Not applicable
Adequate description of statistical methods	[5, 9, 10, 15 ,32-42]	100%
Adequate description of data	[5, 9, 10, 32, 33, 35, 37-32]	80%
Consistency in number of subjects reported	[5, 9, 10, 32-40]	80%
Assessment of statistical significance	[5, 9, 10, 15, 32-40, 42]	93.3%
Attention to potential biases	None	0%
Meaningful main findings	[5, 9, 10, 15, 32-37, 39, 41, 42]	86%
Interpretations of null findings	[5, 9, 10, 15, 32-37, 39-42]	93.3%
Interpretation of important effects	[5, 9, 10, 15, 32-35, 37, 39-42]	86%
Comparison of results with previous reports	[5, 9, 10, 15, 32-34, 36-42]	93.3%
Implications in real life	[5, 9, 10, 15, 32-34, 36, 42]	93.3%

Figure 2. PRISMA flow chart of included studies.

Valent F, Busolin A. Distance to the pediatric emergency department and nonurgent visits in friuli venezia giulia, Italy. Pediatr Emer Care. 2018;34:193–197. doi: 10.1097/PEC.0000000000000852

 PubMedGoogle Scholar

Hilker TL. Nonemergency visits to a pediatric emergency department. JACEP. 1978;7:3–8. doi: 10.1016/S0361-1124(78)80248-2

 PubMedGoogle Scholar

Pehlivanturk-Kizilkan M, Ozsezen B, Batu ED. Factors affecting nonurgent pediatric emergency department visits and parental emergency overestimation. Pediatr Emer Care. 2022;38:264–268. doi: 10.1097/PEC.0000000000002723

 PubMed Web of Science ®Google Scholar

Benahmed N, Laokri S, Zhang WH, Verhaeghe N, Trybou J, Cohen L, et al. Determinants of nonurgent use of the emergency department for pediatric patients in 12 hospitals in Belgium. Eur J Pediatr. 2012;171:1829–1837. doi: 10.1007/s00431-012-1853-y

 PubMed Web of Science ®Google Scholar

Guckert L, Reutter H, Saleh N, Ganschow R, Muller A, Ebach F. Nonurgent visits to the pediatric emergency department before and during the first peak of the COVID-19 pandemic. Int J Pediatr. 2022;2022:7580546. doi: 10.1155/2022/7580546

 PubMed Web of Science ®Google Scholar

Wartman SA, Taggart MP, Palm E. Emergency room leavers: a demographic and interview profile. J Community Health. 1984;9:261–268. doi: 10.1007/BF01338726

 PubMedGoogle Scholar

Bianco A, Pileggi C, Angelillo IF. Non-urgent visits to a hospital emergency department in Italy. Public Health. 2003;117:250–255. doi: 10.1016/S0033-3506(03)00069-6

 PubMed Web of Science ®Google Scholar

Naouri D, Ranchon G, Vuagnat A, Schmidt J, El Khoury C, Yordanov Y. Factors associated with inappropriate use of emergency departments: findings from a cross-sectional national study in France. BMJ Qual Saf. 2020;29:449–464. doi: 10.1136/bmjqs-2019-009396

 PubMed Web of Science ®Google Scholar

Oh TK, Jo YH, Choi JW. Associated factors and costs of avoidable visits to the emergency department among cancer patients: 1-year experience in a tertiary care hospital in South Korea. Support Cancer Ther. 2018;26:3671–3679. doi: 10.1007/s00520-018-4195-0

 Web of Science ®Google Scholar

Chen BK, Cheng X, Bennett K, Hibbert J. Travel distances, socioeconomic characteristics, and health disparities in nonurgent and frequent use of hospital emergency departments in south carolina: a population-based observational study. BMC Health Serv Res. 2015;15:203. doi: 10.1186/s12913-015-0864-6

 PubMed Web of Science ®Google Scholar

Chen BK, Hibbert J, Cheng X, Bennett K. Travel distance and sociodemographic correlates of potentially avoidable emergency department visits in California, 2006–2010: an observational study. Int J Equity Health. 2015;14:30. doi: 10.1186/s12939-015-0158-y

 PubMed Web of Science ®Google Scholar

Vaz S, Ramos P, Santana P. Distance effects on the accessibility to emergency departments in Portugal. Saude e Sociedade. 2014;23:1154–1161. doi: 10.1590/S0104-12902014000400003

 Web of Science ®Google Scholar

Okunseri C, Vanevenhoven R, Chelius T, Beyer KMM, Okunseri E, Lobb WK, et al. Travel distances by wisconsin medicaid enrollees who visit emergency departments for dental care. J Public Health Dent. 2016;76:213–219. doi: 10.1111/jphd.12138

 PubMed Web of Science ®Google Scholar

Ingram DR, Clarke DR, Murdie RA. Distance and the decision to visit an emergency department. Social Science & Medicine Part D: Medical Geography. 1978;12:55–62. doi: 10.1016/0160-8002(78)90007-2

 Google Scholar

McGarry BE, Mao Y, Nelson DL, Temkin-Greener H. Hospital proximity and emergency department use among assisted living residents. J Am Med Dir Assoc. 2023;2023:1349–1355.e5. doi: 10.1016/j.jamda.2023.05.002

 Google Scholar