1278
Wang et al.
d
PEDIATRIC ED AFTERCARE COMPLIANCE
Socioeconomic Disparities Are Negatively Associated with Pediatric Emergency Department Aftercare Compliance N. Ewen Wang, MD, Michael A. Gisondi, MD, Mana Golzari, BA, Theresa M. van der Vlugt, MD, Methodius Tuuli, MBChB, MPH Abstract Objectives: This study sought to identify demographic, socioeconomic, and clinical predictors of aftercare noncompliance by pediatric emergency department (ED) patients. Methods: The authors conducted a prospective, observational study of pediatric patients presenting to a university teaching hospital ED from July 1, 2002, through August 31, 2002. Demographic and clinical information was obtained from guardians during the ED visit. Guardians were contacted after discharge to determine compliance with ED aftercare instructions. Subjects were excluded if they were admitted or if guardians were unavailable or unwilling to consent. Data were analyzed using multivariable logistic regression to identify predictors of noncompliance from a list of predeter-
mined variables. Results: Of the 409 patients enrolled in the study, 111 were prescribed medications and 364 were given specific follow-up instructions. Subtypes of the variable ‘‘insurance status’’ were significantly associated with medication noncompliance in multivariable regression analysis. ‘‘Insurance status’’ and ‘‘low-acuity discharge diagnoses’’ were significantly associated with follow-up noncompliance. Conclusions: Disparity in health insurance has been shown to be a predictor of poor aftercare compliance for pediatric ED patients within the patient population. Key words: pediatrics; emergency medicine; social class; aftercare; logistic models; follow-up studies. ACADEMIC EMERGENCY MEDICINE 2003; 10:1278–1284.
Successful outcomes for pediatric patients discharged from the emergency department (ED) are dependent on guardian compliance with aftercare instructions.1–3 Many pediatric illnesses can be managed on an outpatient basis if guardians are agreeable to timely follow-up and medication compliance.4,5 However, physicians may choose to admit those patients whose families are unable or unwilling to comply with critical aftercare instructions, rather than risk a poor health outcome. Hospital admission has been shown to be a less desirable alternative because it is associated with increased costs and potential complications.6 As the number of families using the ED for routine pediatric sick care continues to increase,7,8 emergency physicians will need assurance that
optimal outpatient management can be arranged for all children. The term ‘‘aftercare’’ refers to all facets of ED discharge planning, including the prescription of medications and instructions for physician followup. Aftercare compliance has not been thoroughly studied in pediatric ED populations, with only a few papers describing either medication9 or follow-up10–13 compliance rates. Some of the predictors of noncompliance that have been identified in these studies suggest that socioeconomic disparities influence patient compliance. This study examines demographic, socioeconomic, and clinical predictors of pediatric ED aftercare noncompliance. The observation of disparities within our study population may provide a pattern for recognizing social and clinical conditions that increase the risk of a poor health outcome.
From the Division of Emergency Medicine, Stanford University School of Medicine, Palo Alto, CA (NEW, TMvdV, MG); Division of Emergency Medicine, Feinberg School of Medicine, Northwestern Memorial Hospital, Chicago, IL (MAG); Department of Emergency Medicine, Kaiser Permanente Medical Center, Santa Clara, CA(TMvdV); and University of California, Berkeley, Berkeley, CA (MT). Received March 5, 2003; revisions received April 28, May 25, and June 29, 2003; accepted July 2, 2003. Mana Golzari is a recipient of funding from the Stanford University Medical Scholars Program. Address for correspondence and reprints: N. Ewen Wang, MD, Stanford University, Division of Emergency Medicine, 701 Welch Road–Bldg. C, Palo Alto, CA 94304. E-mail:
[email protected]. doi:10.1197/S1069-6563(03)00499-8
METHODS Study Design. This was a prospective, observational study. The protocol was approved by our Human Subjects Committee. Study Setting and Population. This study was conducted at a university teaching hospital ED that sees approximately 38,000 patients per year. Children represent one fourth of the overall census. We enrolled pediatric patients (0–18 years) presenting to the ED
ACAD EMERG MED
d
November 2003, Vol. 10, No. 11
d
1279
www.aemj.org
from July 1, 2002, to August 31, 2002, between the weekday hours of 4:00 PM and 12:00 AM, and between 12:00 PM and 12:00 AM on Saturdays and Sundays. These enrollment periods were predetermined by review of our registration database, identifying highvolume pediatric visit times over the previous year. If English was not the guardian’s primary language, professional interpretive services were used for all interviews. Detailed consent forms were available in both English and Spanish. Subjects were excluded if they were admitted to the hospital or if their guardians were unavailable or unwilling to consent during the ED visit. Study Protocol. Twenty potential predictors of noncompliance were identified from the literature or from clinical experience.14,15 These variables included: 1) age,10,16–19 2) gender,10,17–20 3) race,11,13 4) birthplace, 5) low household income,21–23 6) government aid recipient, 7) guardian educational level,22,23 8) guardian primary language,11,22–24 9) transportation availability,13,20,21 10) insurance status,10,11,16,21–23 11) primary provider at a county clinic,8 12) hospital interpreter at discharge,11,22–24 13) immunization status,19,22,23,25,26 14) specialty care provider,20 15) low-acuity discharge diagnoses,13 16) traumatic injury,10,18,19 17) x-rays performed,10,19 18) laboratory tests performed,13 19) chronic illness,10,20,26 and 20) medications prescribed.20 ‘‘Low household income’’ was defined as a total income of less than $20,000 per year. $20,000 represents the Federal Poverty Level (FPL) cutoff for a family of four individuals, and 200% of the FPL for a family of two. Families with pregnant women or infants that have an income at or below 200% FPL qualify for some Medi-Cal assistance programs.27 Whereas specific qualifications for Medicaid may vary across the country, this definition of poverty was considered applicable for most areas of the United States. ‘‘Government aid recipient’’ was defined by the use of public assistance such as welfare payments, Supplemental Security Income (SSI), Medicaid/Medi-Cal, food stamps, or public housing assistance. ‘‘Specialty care provider’’ was defined as ongoing care by a specialist who was not identified as the patient’s primary care provider. ‘‘Low-acuity discharge diagnoses’’ represents any of the following: nasal congestion, afebrile viral upper respiratory infection, sprain, strain, contusion with no laceration, minor closed head injury with no laceration or imaging, diaper rash, or viral exanthem. These represent our most common pediatric ED diagnoses that do not necessarily mandate a follow-up appointment. ‘‘Chronic illness’’ was identified by frequent or continued use of medications for the ongoing treatment of a disease such as asthma, diabetes, or seizure disorder.
Demographic data and telephone numbers were obtained through guardian interview during the ED visit. Clinical data were abstracted from the medical chart after discharge. Guardians were then contacted by phone one week after the patient’s recommended follow-up appointment, or one week after the ED visit if medications were prescribed but no follow-up instructions were given. Guardians were asked to recall the specific aftercare that their child received, including the timeframe for obtaining prescription medications and/or physician follow-up. Responses were compared with the aftercare instructions written in their child’s ED record. Aftercare compliance was determined through guardian self-report. Outcome Measures. Outcome measures included medication and follow-up noncompliance. Guardians were considered noncompliant with medication instructions if prescribed medications were not obtained within two days of the ED visit. Guardians were considered noncompliant with follow-up instructions if they failed to bring their child to a physician within 1.5 times the number of days recommended. Data Analysis. Data were uncoupled from patients’ names and entered into the study database. Potential predictors were analyzed as dichotomous variables except ‘‘age,’’ ‘‘race,’’ ‘‘guardian educational level,’’ and ‘‘insurance status.’’ ‘‘Age’’ was analyzed as continuous, ‘‘race’’ as four subtypes (white, African American, Hispanic, and other), ‘‘guardian educational level’’ as four categories of years of schooling (0–8, 9–12, 13–16, and $17) and ‘‘insurance status’’ as three subtypes (private, Medi-Cal, and uninsured). Univariate analysis using chi-square and logistic regression was performed to assess the association between predictor and outcome variables. Those with an association at a significance level of 5% were included in the multivariable regression model. A stepwise backward multiple regression was performed, with a p-value of 0.05 to remove or reenter. Coefficients from the final model were used to derive odds ratios and 95% confidence intervals. This procedure was performed separately for medication and follow-up noncompliance data. All analyses were performed using Stata 7.0 (Stata Statistical Software [7.0], 1984–2000, Stata Corporation, College Station, TX). Subjects who were not prescribed medications or were not given specific follow-up instructions (i.e., ‘‘PRN’’ or ‘‘as needed’’) were not included in the data analyses. Subjects who were lost to study follow-up also were excluded from these analyses.
RESULTS Approximately 1,400 children were seen in the ED during the study period. Of the 790 pediatric patients
1280
Wang et al.
who presented during enrollment hours, 509 were recruited for the study. Total subjects numbered 409, because 100 patients met exclusion criteria. Twentyfive subjects were lost to study follow-up. Demographic and clinical characteristics are reported in Table 1. The median patient age was 4.35 TABLE 1. Demographics, Clinical Characteristics, and Outcomes Characteristic Demographics* Patient age Median (yr) Interquartile range Patient gender Male Female Patient race White African American Hispanic Asian Other Birthplace U.S. born Non-U.S. born Household income [$20,000 (high) \$20,000 (low) Government aid recipient Yes No Guardian educational level Mean years schooling 6 SD 0–8 years 9–12 years 13–16 years [17 years Guardian primary language English Non-English Transportation Available Unavailable Insurance status Private Medi-Cal Uninsured Primary provider county clinic Yes No Specialty care provider Yes No
Result
Percent
4.35 7.52
NA NA
244 165
59.7 40.3
131 35 144 37 62
32.0 8.6 35.2 9.0 15.2
378 31
92.4 7.6
275 133
67.4 32.6
71 338
17.4 82.6
13.7 6 4.5 49 127 133 98
12.1 31.2 32.6 24.1
246 163
60.1 39.9
355 54
86.8 13.2
240 155 14
58.7 37.9 3.4
93 313
22.9 77.1
76 336
17.9 82.2
Clinical characteristics Low-acuity discharge diagnosis Traumatic injury X-rays performed Laboratory tests performed Chronic illness Medication prescribed
218 133 96 116 81 111
53.2 32.4 23.5 28.4 19.8 27.1
Outcomes Medication noncompliancey Follow-up noncompliancez
25 123
23.0 36.6
*N ¼ 409; yN ¼ 109; zN ¼ 336.
d
PEDIATRIC ED AFTERCARE COMPLIANCE
years. Age distribution was skewed. The interquartile range was 7.52 years, with 25% of patients under 1.5 years of age. Males constituted 244 of 409 (60%) of the study population. The largest racial subgroups were white 131 (32%) and Hispanic 144 (35.2%). Most of the subjects were U.S.-born (378 [92.4%]). A slight majority of our population had a household income greater than $20,000/year (275 [67%]). The mean guardian educational level was 13.7 years, with a bell-shaped distribution. Of the 409 patients enrolled, 111 (27%) of the subjects were prescribed medications at discharge; two subjects had incomplete data. Guardians who reported medication compliance numbered 84 (77%), while 25 subjects (23%) were found to be noncompliant. Results of the univariate and stepwise regressions are shown in Table 2. Whereas univariate analysis identified a number of potential predictors, only subgroups of the variable ‘‘insurance status’’ were found to be significantly associated to medication noncompliance in the multivariable model. Specific follow-up instructions were given to 364 (89%) of the 409 subjects. Data were complete for 336 subjects. Of these, 213 (63.4%) reported compliance with follow-up instructions, while 123 (36.6%) were noncompliant. Table 3 shows the results of the regression analyses. The multivariable model demonstrated ‘‘insurance status’’ and ‘‘low-acuity discharge diagnoses’’ to be significantly associated with followup noncompliance.
DISCUSSION Clinical researchers often attempt to link demographic differences among study subjects to disease patterns within populations. If the researchers demonstrate a difference with substantial social implications, then a disparity exists within their study population. Disparities are not simply benign differences among subjects, but instead represent a meaningful inequality of social class, wealth, power, or privilege. Disparities are differences defined by their social context, which increase the risk of harm or neglect. Research studies concerning health care access have demonstrated that significant socioeconomic disparities are present within pediatric ED patient populations. Jones et al. found that a majority of children seen in an urban ED lacked a regular source of care.28 Patients who identified the ED as their regular health care provider were more likely to be uninsured than those who did not.29 ED use for routine pediatric sick visits also has been shown to be associated with economic and social disparities.7,8 The few studies that have examined medication or follow-up compliance by pediatric ED patients have found some socioeconomic differences within their populations.10,11,13 These findings suggested that social disparities that impact health care access may
ACAD EMERG MED
d
November 2003, Vol. 10, No. 11
d
1281
www.aemj.org
TABLE 2. Predictors of Medication Noncompliance Predictor Univariate analysis Patient race Hispanic Household income Low Guardian educational level Guardian primary language Non-English Insurance status Medi-Cal Uninsured Traumatic injury Yes
95% Confidence Interval
Ratio
% Noncompliance
Odds Ratio
12/35
34.29%
3.65
1.04, 12.86
0.044
14/37 NA
37.84% NA
3.17 0.58
1.23, 8.17 0.35, 0.96
0.015 0.034
14/38
36.84%
3.03
1.18, 7.80
0.019
17/41 3/5
41.46% 60.00%
8.85 \0.062
2.69, 29.19 2.39, 146.93
2/23
8.70%
4.39
0.95, 20.28
Ref* 8.85 18.75
Ref 2.69, 29.19 2.39, 146.93
Multivariate analysis Insurance Status Private Medi-Cal Uninsured
p-value
\0.001 0.005 0.042
Ref \0.001 0.005
*Reference value. N ¼ 109.
also affect the outpatient management of those discharged from the ED. Using multivariable regression analyses, our study demonstrates a significant association between health insurance type (‘‘insurance status, uninsured’’ and ‘‘insurance status, Medi-Cal’’) and pediatric ED aftercare noncompliance. If the significant insurance
subtypes are grouped together, one can argue that a lack of private health insurance is predictive of potential aftercare noncompliance for similar populations. Poor aftercare compliance may be the result of inadequate health care access that arises from such insurance disparity. Uninsured and Medi-Cal patients may have more difficulty obtaining clinic
TABLE 3. Predictors of Follow-up Noncompliance Predictors Univariate analysis Patient race African American Hispanic Asian Other Household income Low Guardian educational level Guardian primary language Non-English Insurance status Medi-Cal Uninsured Low-acuity diagnosis Yes Traumatic injury Yes Multivariate analysis Insurance status Private Medi-Cal Uninsured Low-acuity diagnoses Yes N ¼ 336.
95% Confidence Interval
Rates
% Noncompliance
Odds Ratio
p-value
14/27 50/114 12/30 21/54
51.85% 43.86% 40.00% 38.89%
4.03 2.92 2.49 2.38
1.66, 1.62, 1.05, 1.16,
9.75 5.27 5.91 4.86
0.002 \0.001 0.038 0.017
54/104 NA
51.92% NA
2.67 0.67
1.65, 4.31 0.53, 0.86
\0.001 0.001
57/128
44.53%
1.82
1.15, 2.88
0.010
59/118 7/12
50.00% 58.33%
2.78 3.89
1.73, 4.47 1.18, 12.77
\0.001 0.025
75/164
45.73
2.29
1.45, 3.62
\0.001
28/107
26.17
1.94
1.17, 3.22
0.010
Ref 2.01 2.54
Ref 1.03, 3.94 0.65, 9.85
Ref 0.042 0.179
2.23
1.36, 3.65
0.001
1282 appointments or affording prescription medications, thus contributing to poor ED aftercare compliance. Medication Noncompliance. The overall rate of medication noncompliance in our study was 25 of 109 (23%). This rate is much higher than that observed in a Canadian study of children with similar clinical presentations, which reported a noncompliance rate as low as 8%.9 Differences between American and Canadian health insurance systems may account for these dissimilar results. The rate of prescription noncompliance for our pediatric population more closely resembles those observed in studies of adult ED patients.20–22,30 Subtypes of ‘‘insurance status,’’ the only significant predictor identified through multivariable analysis in our study, have been previously shown to be associated to medication noncompliance in adult ED populations.22 Guardians were considered noncompliant if prescribed medications were not obtained within two days of the ED visit. This endpoint, known as ‘‘primary’’ medication noncompliance, is used in two other studies in the literature.1,21 A third study uses a similar definition of primary medication noncompliance, but allows for a much longer period in which to obtain the medication.22 We did not attempt to confirm ‘‘secondary’’ medication noncompliance, which refers to the completion of the exact dosing regimen prescribed.31,32 Because our hospital does not have an outpatient pharmacy on site, we could not easily confirm a guardian’s report with pharmacy records.32 Oral antibiotics (50 subjects [46%]) and analgesics (18 subjects [16%]) were the two largest categories of medications prescribed. Although antibiotics may be overused, compliance with such prescriptions is generally viewed as necessary for a good clinical outcome. Of the 25 subjects who were noncompliant with their prescriptions, 11 (44%) were prescribed antibiotics and one (4%) was prescribed analgesics. Eleven (22%) of the 50 patients prescribed antibiotics and one (6%) of the 18 prescribed analgesics were noncompliant. We did not compare the prescribed medications with the Medi-Cal formulary; it is unclear if Medi-Cal patients who were noncompliant were prescribed off-formulary medications. Follow-up Noncompliance. We observed that 123 (36.6%) of the 336 pediatric ED patients given followup instructions did not comply. Previous studies in pediatric populations have demonstrated noncompliance rates of 8% to 50%.10,12,13 Multiple studies of both children10–12 and adults16,22,33,34 have shown that if follow-up appointments are made from the ED, or if a referral to a specific clinic is made while the patient is still in the ED, rates of follow-up compliance increase. Our study reports observational data for
Wang et al.
d
PEDIATRIC ED AFTERCARE COMPLIANCE
a system in which guardians are given the name and phone number for a clinic in their area, but not a specific appointment. ‘‘Insurance status’’ and ‘‘low-acuity discharge diagnoses’’ were shown through multivariable regression to be significantly associated with follow-up noncompliance. These results suggest that in addition to observed disparities in health insurance, a guardian’s perception of his or her child’s disease severity influences follow-up compliance. Guardians were considered noncompliant with follow-up instructions if they failed to bring their child to a physician within 1.5 times the number of days recommended. This definition was used to emphasize the importance of promptness of follow-up. Other definitions of physician follow-up compliance in the literature include timeframes of seven to ten days,16,20 one month,18,23 or greater.11,35 Seven subjects sought physician follow-up outside our defined time limit. Four patients were seen within seven days when directed to follow-up in two days; three subjects were seen within two weeks when instructed to follow-up within one week. It is unclear why these delays occurred, or if these delays in physician followup resulted in any poor outcomes. Of the 409 subjects enrolled, 45 had ‘‘as needed’’ (or ‘‘PRN’’) follow-up written in their discharge instructions. These patients were not included in the statistical analyses. We therefore assume that subjects with ‘‘low-acuity discharge diagnoses’’ had some clinical indication for directed follow-up.
LIMITATIONS Potential sampling biases exist for this study. We enrolled only 64% of the children who presented to the ED during the enrollment hours. Because we had inconsistent numbers of recruitment staff present during these hours, some children were missed when patient volume was highest. Patients who presented to the ED during daytime hours also were missed, because recruitment staff were scheduled to work only during high pediatric volume times (evenings and weekends). Patients who presented to the ED during daytime hours may have done so for reasons of poor health care access. Also, we enrolled just ten uninsured patients; according to registration records, we know that the number of uninsured ED patients is an order of magnitude higher. Those patients not enrolled during daytime hours may have been largely uninsured. The average number of years of school for guardians in our population is quite high, at 13.7 years. Our results may not be applicable to populations with lower guardian educational levels. Systems issues may have influenced the results. Our physicians record the names and phone numbers of patients for whom follow-up is essential. Each of
ACAD EMERG MED
d
November 2003, Vol. 10, No. 11
d
www.aemj.org
these families receives a call from a pediatrics resident within 24 hours of the ED visit. A number of these families also are told to follow-up in the ED, as opposed to an outside clinic (an intervention noted to improve return rates in previous studies).10– 12,16,19,22,33,35 It is unclear how such phone calls influenced our follow-up data. In addition, we use computerized discharge instructions, which have been shown to improve ED patient follow-up compliance.18 Because our data were dependent on guardian selfreport some time after their ED visit, recall bias may exist. Some guardians may have forgotten the details of their aftercare instructions, or may not admit to aftercare noncompliance. In addition, guardians consented to the research protocol before leaving the ED. Because they were aware that they were research subjects, they may have intentionally been more compliant with aftercare instructions (the Hawthorne effect). There may be a higher number of insured children in our community as a result of new county-sponsored programs that began in January 2003. Because our study demonstrated the importance of insurance status in aftercare noncompliance, these programs may improve health outcomes for some disadvantaged children. The impact of these programs should be assessed.
CONCLUSIONS This study demonstrates a significant association between health insurance type and pediatric ED aftercare noncompliance. Uninsured or Medi-Cal–insured patients were less compliant with aftercare instructions than patients with private insurance. A lack of private insurance may be a predictor of poor aftercare compliance. This finding may influence the ED aftercare of uninsured or Medi-Cal–insured pediatric patients. For those in which prescription compliance is essential, medications might best be dispensed from the ED at discharge. Hospital admission may be warranted for those who would require close follow-up. The authors thank Sasha Costanza-Chock; Elizabeth Levey; Fredysha McDaniels; Sonal Patil, MD; Ambili Ramachandran; Matthew Rollee; Shaival Shah; Grace Sun; Jimmy Wu; and Janet Young, MD, for their assistance in data collection; Dr. David A. Bergman for his comments on study design; and Dr. Helena C. Kraemer, Dr. Jerry Halpern, and Dr. Ram Duriseti for their thoughtful statistical consultation.
References 1. Nelson EW, Van Cleve S, Swartz MK, Kessen W, McCarthy PL. Improving the use of early follow-up care after emergency department visits. A randomized trial. Am J Dis Child. 1991; 145:440–4. 2. Becker MH. Patient adherence to prescribed therapies. Med Care. 1985; 23:539–55.
1283
3. Eraker SA, Kirscht JP, Becker MH. Understanding and improving patient compliance. Ann Intern Med. 1984; 100:258–68. 4. Baker MD, Bell LM, Avner JR. Outpatient management without antibiotics of fever in selected infants. N Engl J Med. 1993; 329:1437–41. 5. Jaskiewicz JA, McCarthy CA, Richardson AC, et al. Febrile infants at low risk for serious bacterial infection—an appraisal of the Rochester criteria and implications for management. Febrile Infant Collaborative Study Group. Pediatrics. 1994; 94:390–6. 6. DeAngelis C, Joffe A, Wilson M, Willis E. Iatrogenic risks and financial costs of hospitalizing febrile infants. Am J Dis Child. 1983; 137:1146–9. 7. Liu T, Sayre MR, Carleton SC. Emergency medical care: types, trends, and factors related to nonurgent visits. Acad Emerg Med. 1999; 6:1147–52. 8. Halfon N, Newacheck PW, Wood DL, St Peter RF. Routine emergency department use for sick care by children in the United States. Pediatrics. 1996; 98(1):28–34. 9. Matsui D, Joubert GI, Dykxhoorn S, Rieder MJ. Compliance with prescription filling in the pediatric emergency department. Arch Pediatr Adolesc Med. 2000; 154:195–8. 10. Fosarelli P, DeAngelis C, Kaszuba A. Compliance with follow-up appointments generated in a pediatric emergency room. Am J Prev Med. 1985; 1(3):23–9. 11. Hemphill RR, Santen SA, Howell JM, Altieri MF. Follow-up compliance in febrile children: a comparison of two systems. Acad Emerg Med. 1998; 5:996–1001. 12. Komoroski EM, Graham CJ, Kirby RS. A comparison of interventions to improve clinic follow-up compliance after a pediatric emergency department visit. Pediatr Emerg Care. 1996; 12(2):87–90. 13. Scarfone RJ, Joffe MD, Wiley JF 2nd, Loiselle JM, Cook RT. Noncompliance with scheduled revisits to a pediatric emergency department. Arch Pediatr Adolesc Med. 1996; 150:948–53. 14. Sackett DL, Haynes RB. Compliance with Therapeutic Regimens. Baltimore: Johns Hopkins University Press, 1976. 15. Haynes RB, Taylor DW, Sackett DL. Compliance in Health Care. Baltimore: Johns Hopkins Press, 1979. 16. Magnusson AR, Hedges JR, Vanko M, McCarten K, Moorhead JC. Follow-up compliance after emergency department evaluation. Ann Emerg Med. 1993; 22:560–7. 17. Murray MJ, LeBlanc CH. Clinic follow-up from the emergency department: do patients show up? Ann Emerg Med. 1996; 27:56–8. 18. Vukmir RB, Kremen R, Ellis GL, DeHart DA, Plewa MC, Menegazzi J. Compliance with emergency department referral: the effect of computerized discharge instructions. Ann Emerg Med. 1993; 22:819–23. 19. Vukmir RB, Kremen R, Dehart DA, Menegazzi J. Compliance with emergency department patient referral. Am J Emerg Med. 1992; 10:413–7. 20. Jones PK, Jones SL, Katz J. A randomized trial to improve compliance in urinary tract infection patients in the emergency department. Ann Emerg Med. 1990; 19:16–20. 21. Saunders CE. Patient compliance in filling prescriptions after discharge from the emergency department. Am J Emerg Med. 1987; 5:283–6. 22. Thomas EJ, Burstin HR, O’Neil AC, Orav EJ, Brennan TA. Patient noncompliance with medical advice after the emergency department visit. Ann Emerg Med. 1996; 27: 49–55. 23. Enguidanos ER, Rosen P. Language as a factor affecting follow-up compliance from the emergency department. J Emerg Med. 1997; 15(1):9–12.
1284 24. Hampers LC, Cha S, Gutglass DJ, Binns HJ, Krug SE. Language barriers and resource utilization in a pediatric emergency department. Pediatrics. 1999; 103(6 pt 1):1253–6. 25. Christakis DA, Mell L, Koepsell TD, Zimmerman FJ, Connell FA. Association of lower continuity of care with greater risk of emergency department use and hospitalization in children. Pediatrics. 2001; 107:524–9. 26. Yamamoto LG, Zimmerman KR, Butts RJ, et al. Characteristics of frequent pediatric emergency department users. Pediatr Emerg Care. 1995; 11:340–6. 27. Available at: http://www.smchsa.org/health/medi-elig.html. Accessed Apr 2003. 28. Jones DS, McNagny SE, Williams MV, Parker RM, Sawyer MF, Rask KJ. Lack of a regular source of care among children using a public hospital emergency department. Pediatr Emerg Care. 1999; 15(1):13–6. 29. O’Brien GM, Stein MD, Zierler S, Shapiro M, O’Sullivan P, Woolard R. Use of the ED as a regular source of care: associated factors beyond lack of health insurance. Ann Emerg Med. 1997; 30:286–91.
Wang et al.
d
PEDIATRIC ED AFTERCARE COMPLIANCE
30. Freeman CP, Guly HR. Do accident and emergency patients collect their prescribed medication? Arch Emerg Med. 1985; 2(1):41–43. 31. Pontali E, Feasi M, Toscanini F, et al. Adherence to combination antiretroviral treatment in children. HIV Clin Trials. 2001; 2:466–73. 32. Bartlett SJ, Lukk P, Butz A, Lampros-Klein F, Rand CS. Enhancing medication adherence among inner-city children with asthma: results from pilot studies. J Asthma. 2002; 39(1):47–54. 33. Murray MD, Stang P, Tierney WM. Health care use by inner-city patients with asthma. J Clin Epidemiol. 1997; 50:167–74. 34. Fletcher SW, Appel FA, Bourgois M. Improving emergency-room patient follow-up in a metropolitan teaching hospital. Effect of a follow-up check. N Engl J Med. 1974; 291:385–8. 35. Silverman GK, Silverman HM. Efficacy of the follow-up system in the community hospital emergency department. Am J Emerg Med. 1984; 2:119–22.
