| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Pathology and Laboratory Medicine, University of Pennsylvania, 3400 Spruce St., Philadelphia, PA 19104-4283.
aAuthor for correspondence. Fax 215-349-5090; e-mail donaldyo{at}mail.med.upenn.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: To study the effect of age on the costs of treating patients, we have evaluated 43 conditions with matching diagnosis-related groups (DRGs) for children and adults. Using a database developed by the University HealthSystems Consortium, we examined the major non-physician components of hospital costs, including accommodation, surgery, pharmacy, radiology, and laboratory for 1 346 028 patient admissions to 60 University hospitals. These costs were derived from the ratio of costs to charges based on the Centers for Medicare and Medicaid Services PPS UB-2 cost reports.
Results: The total non-physician cost of treating adults was generally greater than that for children within paired DRGs. Some of this difference may be attributable to the overall longer stay of adults in hospital. For conditions that were nominally the same, radiology, laboratory, and drug costs, especially tended to be higher for adults than for children. This was most marked when the costs were evaluated on a per diem basis. There tended to be greater variability in the costs of treating children than adults within the paired DRGs, as evidenced by greater differences between the median and mean costs.
Conclusions: Among University hospitals, the costs of managing children are typically less than for adults with the same nominal condition. In these hospitals, there tends to be less use of laboratory, radiology, and pharmacy services for children than for adults.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
database and calculation of costs
The UHC database comprised a total of 1 346 028 patients in 486 DRGs in 1995. The specific criteria for inclusion and exclusion of patients by UHC as well as calculation of costs have been described previously (1)(2). Briefly, the costs used in the calculations were derived by converting the charges for different services within each hospital to the costs of providing these services by applying the ratios of costs to charges (RCC), calculated from detailed Centers for Medicare and Medicaid Services (CMS) PPS UB-2 cost reports. These calculated costs include all pass-through costs (e.g., capital and medical education), but they do not include physician costs. Total costs are derived from the sum of individual departmental charges within a hospital. This method has shown a correlation >0.90 with internal accounting costs (3).
We concentrated our evaluations on median, rather than mean, numbers because these better reflect the typical patient. The following variables were analyzed for comparison of paired adult and pediatric DRGs: overall costs; length of stay; total costs per day; and accommodation, laboratory, and radiology costs, both overall and on a daily basis. The ratios of mean to median total costs were examined to assess heterogeneity among patients in a given DRG. To enable a comparison with Medicare length-of-stay data, which are oriented to the arithmetic mean, we also studied the mean length of stay.
Selected data from the UHC database were entered into a FileMaker Pro file (Ver. 4.0; Claris Corporation), which was also used to derive additional factors for analysis. Data were exported to an Excel (Ver. 5.01; Microsoft Corporation) file to perform the calculations. Correlations were performed using GraphPad Prism 3.0 (GraphPad Software Inc.).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. There were more adults with uncomplicated conditions than children in 27 of the DRGs, an equal number in 1, and in 15 there were more children than adults. The numbers of patients in each of the 43 paired categories varied widely. There were 25 320 adults with psychoses, but only 1763 children with mental disorders, and 14 216 adults with concussion, more than 25 times the number of children. In contrast, there were 12 114 children with bronchitis, the most common childhood DRG, approximately four times the number of adults. Other conditions for which there were considerably more children than adults were myringotomy, viral illness, congenital cardiac valve disease, and otitis media. The category with the lowest number of patients was urethral strictures, which applied to 5 children and 28 adults. All except four paired DRGs—concussion, myringotomy, urethral procedures, and urethral strictures—had more than 100 patients in both of the pairs. With the exception of myringotomy, the smaller number was for children.
|
nature of conditions
The DRG classification is structured primarily around organ systems. These are included in 25 major diagnostic categories (MDCs). Only 16 of the MDCs contain paired adult and pediatric DRGs. In the MDCs for ear, nose, and throat conditions, digestive system conditions, and kidney and urinary tract conditions (MDCs 3, 6, and 11, respectively), there were patients in 10 or more pediatric DRGs. In addition to the absence of diseases related to pregnancy and childbirth (MDCs 13, 14, and 15), there were no pediatric DRGs associated with hepatic or pancreatic diseases, burns, trauma, substance abuse, contact with health services, or HIV infections [Table 1A; available through the on-line version of this article at Clinical Chemistry Online (http://www.clinchem.org/content/vol48/issue1)]. More than one-half of all the DRGs involving kidney and urinary tract disorders (58.8%) are subclassified into adult and pediatric categories. Almost one-half of those associated with diseases of the ear, nose, and throat (46.2%), blood and blood-forming organs (44.4%), and infections and parasitic disease (44.4%) are separately classified into adult and pediatric DRGs.
overall costs
With the exception of acute leukemia, craniotomy, and splenectomy, the total costs of all the paired conditions were relatively low, i.e., less than $5000. Table 2
ranks, by adult costs, the conditions with median total costs greater than $5000. The costs to treat adult leukemia were three times higher than those to treat childhood leukemia. The only other condition for which the costs of the adult disease were more than twice as high as those of the corresponding childhood disease was allergic reactions: $2066 for adults and $1007 for children. For 32 of the pairs, the median total costs were greater for adults than for children, and in 11 they were higher for children [Table 2A; available through the on-line version of this article at Clinical Chemistry Online (http://www.clinchem.org/content/vol48/issue1)]. The mean total costs were higher for adults than for children for 25 of the paired conditions; they were the same for 1, and lower for 17. The median total costs of respiratory infections in children ($6667) were $2377 more than the median total costs for adults without complicated disease. The costs of mental disorders in children were $6610 compared with $5442 for psychoses in adults. After the paired leukemic diseases, the greatest difference in costs between paired DRGs, when the adult costs were higher, was $3258 for septicemia ($6910 for adults and $3652 for children, a ratio of 1.89). For the 43 paired DRGs, the median total costs for adults were positively correlated with those for children (r2 = 0.8780); the mean total costs were only slightly less strongly correlated (r2 = 0.8444).
|
The median total costs varied from a low of $1007 for allergic reactions in children and $2066 in adults to a high of $9969 for acute leukemia without an operative procedure in children (DRG 405) and $30 604 for the corresponding condition in adults (DRG 473). The average ratio of median total costs for adults to those for children was 1.16. The lowest ratio, i.e., costs higher for children compared with adults, was 0.64 for respiratory infections (DRGs 080 and 081 in children and adults, respectively).
When there were paired DRGs and the very different costs of acute leukemia without an operative procedure were excluded, only the median total costs of $11 882 and $9659 for craniotomy and splenectomy in adults, respectively, exceeded the median total costs of all the adult DRGs by more than 2 SD. For children, when the median total costs of acute leukemia without an operative procedure were similarly excluded, the median total costs of $9061 for craniotomy and $6667 for respiratory infections exceeded by 2 SD the mean of the median total costs of all the pediatric diseases. The average of the median costs for the 42 DRGs in adults, when DRG 473 was excluded, was $3584, which was substantially greater than the $3079 for the corresponding DRGs in children. The average of the mean costs, again with DRGs 405 and 473 excluded, were $4599 for adults and $4171 for children.
ratio of mean to median total costs
In the absence of standard deviations, which are not calculated by UHC, comparison of the mean to median costs for a disease provides a simple indication of the heterogeneity of patients within a DRG. When the mean costs exceed the median costs, this is attributable to the greater expenses incurred in caring for a proportion of patients who are outliers compared with the majority. The proportion of patients could be quite small if these patients were very different in their use of resources. The mean total costs were greater than the median total costs for every condition. Table 3
lists the conditions for which the ratio exceeded 1.35, in order of highest ratios of total mean to median costs for children.
|
On average, the mean costs for adults were 1.27 times higher than the median costs, and for children they were 1.36 times higher than the median costs, suggesting a slightly greater variability in the severity of conditions in children. The biggest difference between mean and median costs was for congenital cardiac valvular disease in children, for which the mean costs were 1.90 times the median costs. The mean-to-median cost ratios were typically higher for children than for adults (Table 3
). For adults the highest ratio of mean to median costs was 1.56 for myringotomy. The lowest ratio was 1.12 for fractures, sprains, strains, and dislocations of the forearm, hand, and foot in children and for tonsillectomy and/or adenoidectomy only in adults (data not shown).
comparison of pediatric diseases (DRGs) with comparable adult diseases, with and without complications and comorbidities
Twenty-eight conditions that had pediatric DRGs were also identified in adults as both having and not having complications and/or comorbidities. Table 4
illustrates the mean costs of these conditions together with the ratio of their costs to those for children. We include the mean costs to demonstrate the influence of the more severe conditions within each DRG. The mean total costs were less for only 2 of the 28 adult DRGs with complications and/or comorbidities compared with those for children. These conditions were cardiac congenital and valvular disorders and other disorders of the eye. For cardiac congenital and valvular disorders, the median costs of the adult condition with complications exceeded those for children by $748, median costs of inguinal and femoral hernia procedures exceeded those for children by $1833, and the median costs of other disorders of the eye exceeded those for children by $414.
|
In contrast, for those conditions with separate adult DRGs with or without complications and/or comorbidities, the mean costs of only 14 of the uncomplicated DRGs exceeded those for children. When the pediatric costs were higher than those for adults, the ratio of adult to pediatric costs was <1, as illustrated in Table 4
. The smallest ratios were 0.59 for other disorders of the eye (with a difference of $1811 between pediatric and adult costs) and 0.62 for cardiac congenital and valvular disorders (difference of $2016). The largest ratios, when the costs of the complicated adult condition were higher than those for children, were 2.03 for gastric, esophageal, and duodenal procedures (with a cost difference of $8567) and 1.94 for urinary tract signs and symptoms ($1612). There was much less difference between the mean costs of uncomplicated than of complicated conditions in adults compared with the costs for children. Thus, the largest ratio was 1.49 for kidney and urinary tract signs and symptoms, with an actual difference in costs of $836.
length of stay
The mean lengths of stay for adults with uncomplicated conditions and children are illustrated in Table 5
. Most of the paired DRGs had relatively short mean lengths of stay, ranging from lows of 1.57 and 1.39 days for concussion in adults and children, respectively, to highs of 10.53 days for psychoses in adults and 13.66 days for mental disorders in children. Of the 43 paired conditions, the mean lengths of stay were longer for children with 22 conditions and for adults with 21 uncomplicated conditions. The biggest difference in mean length of stay, 12.67 days, was for acute leukemia without a major operating room procedure. This was the only condition for which the mean length of stay was markedly longer for children. The biggest difference in mean length of stay, when the adult stay was longer than that for children, was 3.62 days.
|
In the UHC hospitals, the overall ratio of the mean lengths of stay for adults to that for children was 1.14. Among these hospitals, the mean length of stay for adults with nonmalignant testes was almost twice that for children in the paired pediatric DRG. The mean lengths of stay for splenectomy, myringotomy with tube insertion, and allergic reactions were 
80% longer for adults than for children.
length of stay in uhc hospitals vs all hospitals
Among all hospitals nationally, Medicare reports that the overall ratio of the lengths of stay for adults to that for children is 1.52. For 29 of the paired DRGs included in the Medicare-reported information, the mean length of stay in all hospitals was longer for children for 10 DRGs and shorter for 19 pediatric DRGs compared with the data from the UHC hospitals. In comparison with the Medicare information for all hospitals, the reported lengths of stay were shorter in the UHC hospitals for all adult DRGs. The greatest differences in mean lengths of stay between UHC hospitals and all reporting hospitals for conditions included in our database were 5.44 days less for adults and 5.43 days less for children in the UHC hospitals, for craniotomy (DRGs 001 and 003, respectively). The mean Medicare length of stay for all 43 paired DRGs was 4.87 days for adults compared with 3.43 days in the UHC hospitals. For children, the mean Medicare length of stay was 3.91 days compared with 3.19 days in the UHC hospitals. In contrast to the generally shorter lengths of stay in the UHC hospitals compared with the Medicare-reported data, the mean length of stay for children with mental disorders (DRG 431) was 6.06 days longer in the UHC hospitals than in all the hospitals, as reported by Medicare.
There was a better correlation of mean length of stay in the UHC hospitals with the Medicare arithmetic length of stay among adults (r2 = 0.8893) than among children (r2 = 0.7636).
total costs per day
Thirteen conditions in adults had median costs greater than $1000 per day [Table 5A; available through the on-line version of this article at Clinical Chemistry Online (http://www.clinchem.org/content/vol48/issue1)]. The majority of the conditions were surgical. Three of the conditions, sinus procedures, procedures for nonmalignant testes, and fractures or strains of arms or legs, cost substantially more for children than for adults. The highest median costs per day for adults were $2971 for craniotomy. The highest median costs per day for children were $4268 for the single day of admission for sinus procedures. The daily cost of sinus procedures in children was 1.9 times higher than that for adults, partly because the median length of stay for children was only 1 day compared with the 2 days for adults. In contrast, the cost of treating poisonings and toxic effects of drugs in adults, $2251 per day, was 2.2 times higher than that for children, but the median length of stay for children, 2 days, was twice that of adults. The median daily costs of $1098 for fractures, strains, and sprains of the arm and leg for adults were just one-half of the $2095 for children.
For adults, the lowest daily costs were $777 for psychoses. For children with mental disorders, the median daily costs were $661 (data not shown).
accommodation costs
Accommodation costs comprise all those costs associated with feeding and supporting a patient, including nursing costs, during hospitalization. For adults, accommodation costs accounted for more than one-half the total costs for 11 DRGs, and for children they accounted for more than one-half of all costs for 15 DRGs [Table 5B; available through the on-line version of this article at Clinical Chemistry Online (http://www.clinchem.org/content/vol48/issue1)]. The highest median accommodation costs as a proportion of total costs were 85.6% for mental disorders in children (DRG 431) and 80.9% for psychoses in adults (DRG 430). For children, accommodation costs were >60% of the total costs for cellulitis (65.3%) and miscellaneous nutrition and metabolic diseases (62.0%). Accommodation costs of $1422 for adults with cellulitis constituted 58.6% of the total costs. The lowest proportions were 12.7% of total costs for acute leukemia without an operative procedure in children (DRG 405) and 19.7% for sinus procedures in adults. The average of the median accommodation costs for the adult DRGs was $1290, not much higher than the $1191 for children.
In 26 of the 43 paired DRGs, accommodation costs represented a smaller proportion of the total costs for adults than for children. For miscellaneous digestive diseases, the accommodation costs, at 58.9% of the total, were 15.2% greater for adults than the costs for children in the paired pediatric DRG (43.7%). For allergic reactions, accommodation costs represented 50.2% of the total costs for adults but were only 37.9% for children.
As illustrated in Table 6
, the median accommodation costs of $11 750, $3700, $3428, and $3265 for adults with acute leukemia without an operative procedure, craniotomy, septicemia, and splenectomy, respectively, exceeded the mean of the median accommodation costs for adults by 2 SD. For children, the median accommodation costs of $3504 for respiratory infection and $3053 for craniotomy exceeded the mean of the median accommodation costs by 2 SD.
|
On average, accommodation costs for adults were 1.12 times higher than those for children, but the accommodation costs for allergic reactions for adults were 2.71 times higher than those for children, whereas the accommodation costs for respiratory infections in adults were only two-thirds as high as those for children. For children, the median accommodation costs for viral illnesses (59.8% of the total costs) were 13.0% higher than those for adults.
The correlation of median accommodation costs with median total costs was r2 = 0.8531 for adult DRGs and r2 = 0.7340 for pediatric DRGs.
laboratory costs
The 13 conditions for which laboratory costs represented >10.0% of the total costs for either the adult or pediatric condition are shown in Table 7
. Laboratory costs were greater for adults than for children in 36 of the 43 paired DRGs. For all the adult DRGs, the average of the median laboratory costs was $249. For children, the average of the median laboratory costs ($142) was substantially less. For adults, the median laboratory costs of $3696 for acute leukemia without an operative procedure, $874 for septicemia, and $836 for splenectomy exceeded the mean of the median laboratory costs for all adult paired conditions by more than 2 SD (Table 6
). For children, the median laboratory costs for acute leukemia without an operative procedure ($1822), kidney and urinary tract disorders ($379), and septicemia ($358) exceeded the mean of the median laboratory costs for all identified pediatric conditions by more than 2 SD (Table 6
). For allergic reactions, laboratory costs were 10.3% of the total costs for adults but only 0.7% for children in the paired DRG. The costs for adults ($213) were 30.4 times greater than those for children ($7). In contrast, laboratory costs represented only 8.2% of the total costs for adults with kidney and urinary tract symptoms, but for children in the paired DRG, they were more than twice as much, 18.3% of the total costs. For kidney and urinary tract symptoms, the laboratory costs ($177) were quite small for adults and substantially less than the average laboratory costs in the adult group, whereas the laboratory costs for children ($278) substantially exceeded the average laboratory costs for pediatric conditions. Laboratory costs for other renal conditions in children also represented a high proportion of the total costs when compared with other DRGs. The biggest difference in laboratory costs between adults and children was $598 for splenectomy. For adults the costs were $836, whereas for children they were $238, representing 8.7% and 4.4%, respectively, of the total costs.
|
Median laboratory costs were, on average, 7.5% of the total costs for adults but were only 5.2% for children. The proportion of total costs attributable to laboratory costs were 2.5, 2.0, and 2.8 times higher for adults than for children for the three diseases of the respiratory system: respiratory infections, simple pneumonia, and bronchitis, respectively (not all data shown). For the four orthopedic conditions, the proportion of total costs attributable to laboratory costs was 2.1–4.5 times higher for adults than for children in the paired DRGs (data not shown). There were no identifiable median laboratory costs for urethral strictures in children, and laboratory costs were <1.0% of total costs for five other pediatric conditions. For adults, the lowest proportion of total costs attributable to the laboratory was 1.7% for lower extremity operative procedures.
Median laboratory costs were better correlated with median total costs (r2 = 0.6033) for adults than for children (r2 = 0.3244).
drug costs
Among the paired DRGs, drug costs exceeded 10.0% of the total costs in six adult DRGs and two pediatric DRGs [Table 7A; available through the on-line version of this article at Clinical Chemistry Online (http://www.clinchem.org/content/vol48/issue1)]. The drug costs of $8502 to treat acute leukemia without an operative procedure for adults (DRG 473) and $3740 for children (DRG 405) were far greater than those for any other condition. These costs were 27.8% and 37.5% of the total costs to manage the disease in adults and children, respectively. When these two extraordinarily high drug costs were excluded, the mean of the median drug costs for the adult DRGs was $233 compared with $155 for the pediatric DRGs. When drug costs exceeded 10.0% in adults, they were higher than those for children except for four conditions: other eye diseases, respiratory infections, congenital cardiac and valvular disease, and other kidney or urinary tract disorders. When the costs of drugs for leukemia were excluded, drug costs in adults generally were 6.11% of the total costs. For children they represented an average of 4.70%. Again, excluding the costs for treating leukemia, drug costs, as a percentage of total costs, ranged from a low of 1.89% in adults with psychoses (DRG 430) to a high of 12.23% for septicemia (DRG 416). In children the comparable proportions for mental disorders (DRG 431) and septicemia (DRG 417) were 0.70% and 9.12%, respectively. Drug costs as a proportion of total costs for respiratory infections in children (10.2%) were higher than those for septicemia. The condition with the biggest difference in drug costs between adults and children was $4762 for acute leukemia without an operative procedure, with the next largest difference being $512 for septicemia. The mean of the mean drug costs for adults was $836, compared with $238 for children. There was a strong positive correlation between the median total costs and drug costs for adults (r2 = 0.8572) and a less positive correlation for children (r2 = 0.7486).
radiology
Median radiology costs were identifiable for both adult and pediatric disorders for only 28 of the 43 paired DRGs. For 10 of the paired conditions, neither adults nor children incurred radiologic expenses. For the other five conditions, the adults incurred median radiologic costs, whereas the children did not. These costs of $172, $59, $42, $32, and $10 for adults were incurred for splenectomy, allergic reactions, cellulitis, poisoning or toxic effects of drugs, and urethral stricture, respectively, whereas no median radiologic costs were incurred for children in the paired DRGs. Radiology costs typically were small. For only three conditions—craniotomy, trauma with stupor and coma, and concussion—did the radiology costs exceed $400. For concussion, the radiology costs accounted for 30.2% of the total costs for adults and 23.2% of the total costs for children. For trauma with stupor and coma, the radiology costs represented 19.8% of the total costs for adults and 17.5% of those for children.
component costs per day
Although the mean of the median lengths of stay was 0.4 days longer for adults than for children in the paired DRGs, the costs per day were not very different, as shown in Table 8
. In particular, accommodation costs did not differ, but Table 8
clearly shows greater use of radiologic and laboratory services for adults than for children. Drug costs per day were also substantially higher for adults than for children.
|
other age classifications
As mentioned earlier, there are 123 DRGs with age in their title. Frequently, three different DRGs are associated with the same condition: age over 17 years with complications and/or comorbidities, age over 17 years without complications and/or comorbidities, and the condition in children age 17 years or less regardless of any intercurrent complication. The influence of complications on costs is assessed in a separate report (3). Unlike other conditions, diabetes is classified with a break point of 35 years, rather than the more typical 17 years. Total costs were 1.25 times higher for those patients >35 years of age than for younger diabetic patients. Accommodation costs were 1.13 times higher, and laboratory costs were almost double. The median length of stay for the younger diabetic patients was 1 day less than for the older group, making the daily cost for the younger patients somewhat higher: $881 vs $828 after age 35 years.
DRG 300 comprises patients with endocrine disorders with complications and/or comorbidities. There is no DRG specifically for endocrine disorders without complications. However, there is a DRG (DRG 301) designated for endocrine diseases in patients >70 years of age. The greater complexity of DRG 300 is reflected in both the total and component service costs. The median total cost for DRG 300 was 1.86 times higher, accommodation costs were 2.08 times higher, laboratory costs were 1.78 times higher, radiology costs were 4.63 times higher, and drug costs were 3.06 times higher than those for DRG 301. The median length of stay for DRG 300 was 4 days, whereas it was only 2 days for DRG 301. Consequently, the median cost per day of the uncomplicated condition was $979, greater than the $913 for endocrine disorders with complications and/or comorbidities.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our study demonstrates that the costs of managing adults are generally greater than those for treating children with conditions that are nominally the same. It must be recognized, however, that the severity of conditions with the same name may be quite different. Thus, surgery to correct a congenital cardiac problem in a child must be done when the patient is acutely ill, whereas a less acute response is often required in an adult. The conditions for which separate adult and pediatric DRGs have been developed are typically low cost. The costs of managing patients in either the adult or pediatric DRG exceeded $5000 for only eight conditions. Some of the higher total costs for children may be attributable to a greater pressure on hospitals to secure early discharge of adult patients because a much larger proportion of these patients have the costs of their treatment paid through Medicare DRG reimbursement. This provides considerable incentive to adult hospitals to discharge patients as early as possible and not to admit patients who could be treated effectively as outpatients. This, for example, might contribute to the much larger number of children than adults hospitalized with bronchitis. Nevertheless, overall the numbers of DRGs with shorter and longer mean lengths of stay were comparable for adults and children.
The UHC database does not include measurements of the standard deviations of the various variables recorded. Thus, the only way we could assess the magnitude of the variability of measurements was to compare their means and medians. Eight pediatric conditions had higher ratios of mean to median total costs than the highest ratio for an adult condition. Overall, the mean-to-median ratio in children (1.36:1) was higher than that in adults (1.27:1), suggesting a greater variability in the complexity of the paired conditions in children than in adults. There are 111 conditions in adults that are separately classified as having preexisting complications and/or comorbidities on the one hand or of not having them on the other. The costs for managing the condition with complications typically exceed those to manage the uncomplicated condition, and the ratio of mean to median costs is typically higher (3). No pediatric DRG is subclassified as having or not having preexisting complications. The influence of complications in some of the patients within a pediatric DRG alone might account for some of the greater variability of costs for children than for adults with the same nominal condition but without complications. This study was also unable to differentiate those patients with more severe conditions from those with milder conditions because the DRG classification scheme includes all patients with the same disease, regardless of its acuity, within the same DRG (4).
Although we included the median costs per day as one of the factors to examine, these are very much influenced by the length of stay. Because median lengths of stay are measured in whole numbers, the costs per day are reduced by one-half if the median length of stay changes from 1 to 2 days, or by one-third if the median length of stay changes from 2 to 3 days. This factor is therefore less robust and its relative importance uncertain. Whereas the overall costs to manage adult conditions were somewhat greater than those for children, the costs per day tended to be comparable. Median total costs were greater for adults than for children and were associated with an overall longer length of stay for adult patients. Median daily accommodation costs differed by only $4 between adults and children. Nevertheless, the daily costs for radiologic and laboratory examinations were 15% and 34% lower for children than for adults. The costs per day of drug treatment were also 23% lower for children than for adults.
Accommodation costs basically comprise nursing costs and the "hotel" type of costs incurred in looking after patients. We have previously shown that median accommodation costs represent 48% of total costs among all medical DRGs and 33% of total costs for surgical DRGs (2). When the accommodation costs of the paired adult and pediatric conditions were broken out, the costs for adults tended to represent a greater proportion of the total costs than those for children, although there is considerable variation among the conditions. On the other hand, Hanson et al. (5) have pointed out that pediatric patients require more time, effort, and skill from hospital staff than do adults. In particular, children under 3 years of age and those with congenital abnormalities or chronic illnesses consume a disproportionate amount of nursing time. Accommodation costs as a proportion of total costs tend to be lowest for simple surgical conditions that are associated with short lengths of stay and greatest for medical conditions with low use of diagnostic procedures. Thus, accommodation costs as a proportion of total costs for psychiatric disorders were 80.9% for adults and 85.6% for children. Although acute leukemia without an operative procedure was associated with a long mean length of stay, 9.50 days for adults and 22.17 days for children, accommodation costs were disproportionately low because of the high cost of drug treatment, which accounted for 27.8% of total costs for adults and 37.5% of the total costs for children (Young DS, Demers R, Sachais BS, Jefferies LC. Drugs as a component of hospital costs, submitted for publication). The high laboratory costs associated with managing acute leukemia, 12.1% of the total costs for adults and 18.3% for children, also contributed to the relatively low proportion of accommodation costs for these conditions.
In conclusion, in this study we compared the costs of treating children with those of treating uncomplicated conditions in adults when the adult conditions were subclassified as having or not having complications and/or comorbidities. In general, costs were greater for adults than for children in paired DRGs. As we have shown previously (3), the cost of treating patients with complications typically is substantially more than the cost of treating the uncomplicated condition. A pediatric DRG includes patients both with and without complications. If only children without complications were compared against the adult condition without complications, the differences in costs would probably be more marked. This study was conducted on children in general hospitals. It would be interesting to compare these results with those obtained in specialized pediatric hospitals to determine whether the exclusively pediatric orientation of such hospitals affects either or both the length of stay and the overall use of resources.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
