996
COEXPOSURE
Klasner et al.
PEDIATRIC MASS CO EXPOSURE
soning: the injury and its treatment. J Am Coll Emerg Med. 1979;8~479-84. 5. Hardy KR, Thom SR. Pathophysiology and treatment of carbon monoxide poisoning. Clin Toxicol. 1994;32:613-29. 6. Binder J W , Roberts W .Carbon monoxide intoxication i n children. Clin l’bxicol. 1980;16:287-95. 7. Crocker P J , Walker JS. Pediatric carbon monoxide toxicity. J Emerg Med. 1985;3:443-8. 8. Baker MD, Henretig FM, Ludwig S . Carboxyhemoglobin levels in children with nonspecific flu-like symptoms. J Pediatr. References 1988;113~501-4. 9. Peterson JE, Stewart RD. Absorption a n d elimination of 1. Cobb N, Etzel RA. Unintentional carbon monoxide-related carbon monoxide by inactive young men. Arch Environ Health. deaths in the United States, 1979 through 1988.JAMA. 1991; 1970;21:165-71. 226:659- 63. 10. Britten J S , Myers RAM.Effects of hyperbaric treatment 2. May M. Formula for carboxyhemoglobin half-life suiting on carbon monoxide elimination i n humans. Undersea Biomed two data point model. Personal communication, 1997. Res. 1985;12:431-8. 3. Burney RE, Wu S , Nemiroff MJ, e t al. Mass carbon mon- 11. Weaver LK, Larson-Lohr V, Howe S, e t al. Carboxyhemooxide poisoning: clinical effects and results in 184 victims. Ann globin (COHb) half-life (t,) in carbon monoxide poisoned paEmerg Med. 1982;11:394-9. tients treated with normobaric oxygen (0,)[abstract]. Under4. Myers RA, Linberg SE, Cowley RA. Carbon monoxide poi- sea Biomed Res. 1994;21:13-4.
ject; and Michael K. May, SJ,PhD, Saint Louis University, Department of Mathematics and Computer Sciences, for his reduction of complex mathematical equations to predict COHb half-life t h a t suited a two-data-point model. They also thank the nurses and staff a t Cardinal Glennon Regional Poison Center for providing treatment advice and for their diligence and efficiency during t h e initial d a t a collection period.
I
Follow-up Compliance in Febrile Children: A Comparison of Two Systems
I
ROBIN R. HEMPHILL,MD, SALLY A. SANTEN, MD, JOHN M. HOWELL,MD, MICHAELF. ALTIERI, MD Abstract. Objectives: Follow-up compliance is critical in febrile children because they may harbor unrecognized life-threatening illnesses. This study compares follow-up rates between 2 systems: Wilford Hall Medical Center (WHMC), with preset appointments after ED release, and free medical care; a n d Fairfax Hospital (FFX), where parents must arrange followu p appointments after ED release, and a r e responsible for payment for their follow-up visits. The study also investigated factors associated with follow-up compliance. Methods: This was a prospective, observational study of febrile children seen in 2 EDs with different systems for patient follow-up. From ED records and parental phone calls, diagnosis, follow-up compliance, and demographics were collected. Data were analyzed using logistic regression and x2. Results: 423 children met entrance criteria, and 330 parents were successfully contacted after the child’s E D release (146 from WHMC; 184 from FFX). The WHMC children were more likely to comply with
From the Joint Military Medical Centers, Emergency Medicine Residency, S a n Antonio, TX (RRH); GeorgetownlGeorge Washington, Emergency Medicine Residency, Washington, DC (JMH); Vanderbilt Department of Emergency Medicine, Nashville, TN @AS); and Department of Emergency Medicine, Inova Fairfax Hospital, Fairfax, VA (MFA). Received September 12, 1997;revisions received February 13,
follow-up t h a n were the children in the FFX system (92% vs 67% follow-up, odds ratio 2.5, 95% CI 1.15.3). Other factors associated with noncompliance with recommended follow-up were: Hispanic ethnicity, non-English-speaking parents, a n d follow-up suggested for >24 hours after ED release. For FFX, self-pay, lack of a follow-up physician, parents’ dissatisfaction with the ED medical care, and diagnosis of otitis media were also significant factors found associated with noncompliance. Conclusion: Febrile children evaluated in a medical system with prearranged follow-up appointments and free medical care are more likely to comply with recommended followu p than a r e those evaluated in a system where payment and appointments a r e t h e responsibility of t h e parents. Efforts should be made to improve follow-up compliance by modeling t h e WHMC system. Key words: follow-up studies; patient compliance; emergency medicine; fever; pediatrics. ACADEMIC EMERGENCY MEDICINE 1995; 5:996- 1001
1998,and March 13,1998;accepted March 25,1998.Presented a t the annual meeting of the American Academy of Pediatrics, Emergency Medicine Section, Dallas, TX, October 1994. Address for correspondence a n d reprints: Sally A. Santen, MD, Vanderbilt Department of Emergency Medicine, 703 Oxford House, Nashville, TN 37232-4700.Fax: 615-936-1316; e-mail:
[email protected]
ACADEMIC EMERGENCY MEDICINE
F
October 1998. Volume 5. Number 10
997
met the following criteria: age >3 months and 24 hours (Table 2). While not statistically significant when combined with other variables in the logistic regression analysis, simple cross-tabulation analysis revealed that children with Medicaid (54%) were less likely to be compliant compared with insured patients (83%) ( p < 0.001). Likewise, there was a trend toward better follow-up if the follow-up physician was contacted at the time of the ED visit (50% vs 81%). For WHMC, poor compliance with follow-up was associated with no diagnostic testing during t h e ED visit, and the diagnosis of seizure or upper respiratory infection (Table 2). Seventy-three children did not follow-up (12 from WHMC, 61 from FFX) (Table 3). In 22 cases, the parents stated t h a t the child was better a n d the parents did not believe t h a t follow-up was necessary. Four parents of children seen at FFX specifically stated they were unable to pay for the cost of the follow-up visit. Seventeen children from FFX were unable to schedule follow-up appointments. I n some of these cases, t h e parents had attempted to get a n appointment but up-front payment was required since the follow-up physician did not take Medicaid or the child had no health insurance. This was particularly true of children referred to the pediatrician on call. The parents of 15 children stated they did not understand t h e written instructions to follow-up with a pediatrician.
TABLE2. Odds Ratios (ORs) and Confidence Intervals (CIS)
FFX Hispanic Self-pay No physician Otitis media Follow-up in 1-3 days
OR
95% CI
2.4 60.2 4.0 0.6
1.1-5.2 4.2-862.3 1.2-13.4 1.1-6.0
46% (26/56) 10% (U10) 72% (5/7) 53% (24/45)
1.3-2.3
48% (23/48)
6.3 28.5
1.4-27.6 1.5-536.9
73% (11115) 50% (l/2)
7.1
1.8-28.0
74%(14119)
1.7
WHMC No diagnostic test Seizure Upper respiratory infection
Follow-up Rate
DISCUSSION To our knowledge, t h e rates of compliance with recommended follow-up specifically for febrile children have not been previously studied. For other populations, without interventions, the rates of folSeveral studlow-up range from 26% to 65%.2.3-6-13 ies have attempted changes to improve compliance by providing patients with specific appointment times and computerized i n s t r u c t i o n ~ . ~ . ~MJag~J~ nusson et al. showed t h a t giving patients a specific appointment time for follow-up improved rates of compliance from 46% to 65%.*Despite low rates of follow-up in some systems, most EDs have not implemented strategies to improve follow-up. This study attempted to show compliance rates in 2 different systems of follow-up a n d identify factors associated with compliance with recommended follow-up. The rate of compliance to follow-up in t h e FFX system, while significantly lower t h a n t h a t at WHMC, was one of t h e best follow-up rates reported for a setting without interventions.8-lo WHMC incorporated several factors t h a t have previously been shown t o yield higher rates of followup. The child was given a time and date of appointment at a specific clinic for follow-up within 2448 hours after E D visit. Additionally, this system provided free health care. This removed much of the responsibility for follow-up from the parents. I n the FFX system, children were 2.5 times less likely to follow-up. Our statistics do not identify the most important factors in ensuring compliance (free care, the set appointment time, or other factors). However, reasons parents gave for noncompliance
TABLE3. Stated Reason for Noncompliance Site
Child Better
Misunderstood Instructions
Couldn’t Afford Visit
Couldn’t Get Appointment
Couldn’t Miss Work
Other Reasons
WHMC FFX
2 20
4 11
1 3
0 17
2 2
3 8
1000
COMPLIANCE
included both payment and scheduling concerns. A study at our institution attempts to clarify these questions.l6 We were able to identify some factors that were associated with decreased compliance to recommended follow-up. For example, Hispanic children were less likely to follow-up. The reason for this is not clear, but since a n inability to speak English was also associated with decreased compliance, the problem may be related to a language barrier in understanding the discharge instructions or making appointments. Other ethnic groups that may have had language problems (i.e., Arab, Asian) were not found to have significantly lower rates of follow-up, but their numbers may have been too small to find significance. Further studies might attempt to elucidate the influence of language and ethnicity on compliance to follow-up. Those children with follow-up recommended in 2-3 days rather than in 24 hours also had a lower rate of follow-up. This finding was true for the entire cohort and for FFX alone. This may suggest that parents are less impressed about the need for follow-up if i t is not instructed for the next day. In the FFX system, children with no insurance had particularly poor rates of follow-up compliance, although the numbers are small. In addition, children with Medicaid had decreased compliance compared with insured patients. These findings imply that payment may play a factor in the ability to comply with recommended follow-up. Factors for compliance for WHMC children were also analyzed. Children were less likely to follow-up if they did not have laboratory work performed as part of the ED evaluation. It is possible that having an ED work-up that includes laboratory studies impresses the need for follow-up on the parents. Parents may believe that laboratory testing might be related to the severity of their children’s illness. These children may be perceived as sicker; therefore, the parents more likely to comply with recommended follow-up The stated reasons for noncompliance may provide suggestions on how to improve follow-up. Many parents attempted to make follow-up appointments as instructed, but were unable to get timely follow-up. During the telephone interview, several parents reported contacting the pediatrician on call, but being informed that Medicaid would not be accepted. Others were not informed of the reason that they were unable to get appointments. Some parents reported that they did not follow-up because they could not pay for the followup visit. Finally, despite written instructions describing the recommended follow-up, some parents did not understand that their children needed to be seen again for a repeat evaluation, even if the child was
Hernphill et al.
FOLLOW-UP COMPLL4NCE IN FEVER
better. The misunderstanding of discharge instructions is a well-documented problem in EDs that holds true for parents of febrile children.8 Emergency physicians evaluate many febrile pediatric patients. Most of these children are healthy, and will improve whether or not the parents comply with recommended follow-up. However, some children will get worse, or develop other problems following their ED visits. It is difficult to predict when this will occur. Close follow-up for febrile children remains critical, but many EDs do not facilitate follow-up arrangements. Therefore, we believe the following efforts by EPs will help improve rates of follow-up for febrile children: 1) Establish specific follow-up dates and times whenever possible. For children without a primary physician, try to arrange protected appointment times in local clinics. 2)Attempt to make special arrangements for circumstances where payment makes the follow-up visit difficult. Based on our findings regarding the parental reasons for noncompliance, we believe i t is important to explain to parents the need for follow-up even if the child is better. A prospective validation of these recommendations is required to verify their impact on follow-up rates.
LIMITATIONSAND FUTURE QUESTIONS There are several limitations to this study. The populations of the 2 sites may not be equivalent and this may introduce bias into the results. It is difficult to determine equivalency of populations because limitless factors could be taken into account. These include the population’s socioeconomic and educational status, the distance each population lives from the site of follow-up, transportation issues, and child care arrangements. Possibly a military population may be more likely to follow “orders” and therefore be more compliant with follow-up than is a civilian population. Other factors not examined may affect compliance with follow-up, such as the degree of fever or the irritability of the child. The compliance with follow-up of children whom we were unable to contact is not known and may change our observed rates for compliance. Children without telephones were excluded for practical reasons, and there was therefore no way to determine compliance in this group. In addition, data about compliance and the reasons for failure to follow-up are limited by parental self-reporting. Finally, the effect of poor compliance on medical outcomes w a s not determined. We recommend t h a t physicians model the WHMC system with specific appointment times and minimal cost of care. Future studies might apply an intervention offering parents appointments and addressing payment concerns, then compare
ACADEMIC EMERGENCY MEDICINE
October 1998, Volume 5, Number 10
the compliance rates before and after the intervention. Such a study in a nonmilitary population would increase the validity and decrease the potential for bias from potentially dissimilar populations.
CONCLUSION Febrile children evaluated in a medical system with preset appointments and free medical care are more likely to comply with recommended follow-up than are those evaluated in a system where payment and appointments are the responsibility of the parents. The authors thank Mary Dietrich, PhD, and Clifford Butzin, PhD, for statistical support.
References 1. Becker MH. Patient adherence to prescribed therapies. Med Care. 1985;23539-55. 2. Magnusson AR, Hedges J R , Vanko M, McCarten K, Moorhead JC. Follow-up compliance after emergency department evaluation. Ann Emerg Med. 1993;22:560-7. 3. Straus J H , Orr ST, Sharney E. Referrals from a n emergency room to primary care practices at a n urban hospital. Am J Public Health. 1983;7357-61. 4. Goldman L, Freidin R, Cook EF, Eigner J , Grich P. A mul-
1001
tivariate approach to t h e prediction of no-show behavior in a primary care center. Arch Intern Med. 1982;142563-7. 6. Rosenweig S . “Emergency rapport.” J Emerg Med. 1993;11: 775-8. 6. Nelson EW, Van Cleve S, Swartz MK, Kessen W, McCarthy PL. Improving the use of early follow-up care after emergency department visits. Am J Dis Child. 1991;145440-4. 7. Rouan GW, Hedges J R , Roltzis R, Goldstein-Wayne B, Brand D, Goldman L. A chest pain clinic to improve the followu p of patients released from a n urban university teaching hospital emergency department. Ann Emerg Med. 1987; 16: 1145- 50. 8. Grover G, Berkowitz CD, Lewis RJ. Parental recall after a visit to t h e emergency department. Clin Pediat. 1994; 33: 194- 201. 9. Chande VT, Exum V. Follow-up phone calls after a n emergency department visit. Pediatrics. 1994;93513-4. 10. Horne A, Ros SP. Telephone follow-up of patients discharged from the emergency department: how reliable? Pediatr Emerg Care. 1995;11:173-5. 11. Vukmir RB, Kremen R, Ellis GL, Dehart GL, Plewa MC, Menegazzi J. Compliance with emergency department referral: t h e effect of computerized discharge instructions. Ann Emerg Med. 1993;222319-23. 12. Macharia WM, Leon G, Rowe BH, Stephenson BJ, Haynes B. An overview of interventions to improve compliance with appointment keeping for medical services. JAMA. 1992;267: 1813-7. 13. Nairn SJ, Sugarman JM. Adequacy to follow-up in children diagnosed with urinary tract infections in a pediatric emergency department. Pediatr Emerg Care. 1995;11:156-7. 14. Jones SL, Jones PK, Katz J. Health belief model intervention to increase compliance with emergency department patients. Med Care. 1988;26:1172-84. 16. Hemphill RR, Santen SA, Morgan A. Compliance to followup appointments in febrile children [abstract]. Ann Emerg Med. 1996;27:126.
