ORIGINAL ARTICLE
Additional Medical Costs Due to Hospital-Acquired Falls Masahiro Hirose, MD, PhD, Dr PH,* Narue Nakabayashi, HIM, BA, M Med Sci.,† Seiji Fukuda, MD, PhD,* Shuhei Yamaguchi, MD, PhD,* Mikio Igawa, MD, PhD,‡ Koichi Egami, RT,§ Junichi Honda, MD, PhD,§ and Hiroji Shima, MD, PhD||
Objectives: To explore the additional medical costs (AMCs) due to hospital-acquired falls (falls), as well as their impact on clinical services within hospitals under the nationally uniform universal health insurance system in Japan. Methods: With the use of administrative profiling data based on accounting systems linked with the Japanese social insurance medical fee schedule, we analyzed data from 2 teaching hospitals: Shimane University Hospital (SUH) and St. Mary's Hospital (SMH). We extracted 588 fall cases from 4669 incident reports in SUH and 1168 fall cases from 7717 incident reports in SMH that potentially incurred AMCs. Results: Additional medical costs were 364 ± 2129 USD for minor injuries and 4336 ± 3645 USD for major injuries at SUH (P < 0.001) and 114 ± 124 USD for minor injuries and 2267 ± 2811 USD for major injuries at SMH (P < 0.001). Among the clinical services provided, imaging services were the most frequently used, with 89.9% (n = 205) of 228 minor injuries at SUH and 86.7% (n = 339) of 391 minor injuries at SMH; imaging services were used in all major injury cases at both hospitals. Although the number of cases using additional procedure/surgery services was lower than those using imaging services at both hospitals, AMCs for procedure/ surgery services accounted for the highest proportions of total AMCs in both hospitals. Conclusions: Although falls with minor injuries outnumbered falls with major injuries, fall-related AMCs for the latter were higher at both teaching hospitals because procedure/surgery services were required for cases with major injuries such as femoral neck and trochanteric fractures. The findings suggest that hospital administrators and policy makers have to take appropriate measures to prevent major injuries inpatients due to hospital-acquired falls. Key Words: hospital-acquired falls, additional medical costs, major and minor injuries, clinical services (J Patient Saf 2015;00: 00–00)
ospital-acquired falls (hereafter referred to as “falls”) is a very common issue all over the world, with the proportion of fall-related incident reports ranging from 9.8% to 41%.1–3 A previous study reported that 23% of 922 patient-reported adverse events were medication-related incidents, followed by fall-related adverse events at 21%.4 The incidence of falls increases with
H
From the *Patient Safety Division, and †Division of Health Care Services, Shimane University Hospital, Shimane, Japan; ‡President, Shimane University Hospital, Shimane, Japan; §Medical Care Quality Headquarters, St. Mary's Hospital; and ||President, St. Mary's Hospital, Fukuoka, Japan. Correspondence: Masahiro Hirose MD, PhD, Department of Community-Based Health Policy and Quality Management, Shimane University Faculty of Medicine, 89-1 Enya-Chou, Izumo-Shi, Shimane 693-8501, Japan (e‐mail:
[email protected]). The authors disclose no conflict of interest. This work was funded in part by a Grant-in-Aid for Scientific Research B (No. 22390106) from the Japan Society for the Promotion of Science. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially.
J Patient Saf • Volume 00, Number 00, Month 2015
patient age, and several studies have reported rates of falls per 1000 patient-days that range from 0.8 to 6.12 (overall, 3.38) at a US university hospital with 1300 beds,5 2.7 at a Swiss university hospital with 800 beds,6 and 8.6 to 9.1 at a 300-bed urban public hospital.7 These rates vary according to bed type, bed number, and ownership in health care institutions. Furthermore, serious fallrelated injuries can differ by hospital type, where these injuries accounted for 3.7% and 2.2% of all fall-related injuries in an academic hospital and nonacademic hospital, respectively.1 Falls can increase health resource expenditure due to extended length of hospital stay (LOS) and additional diagnostic procedures and/or surgeries.8 A study found that when compared with nonfallers, inpatient LOS durations were longer for fallers in the diagnosisrelated groups (DRGs) “delirium”, “stroke”, and “respiratory conditions”; in addition, the total associated costs for fallers for the DRG “stroke with severe/complicating diagnosis/procedure” were significantly higher than those for nonfallers.9 Medical costs after incidents and accidents within hospitals are predominantly unrelated with the primary diseases, and these additional costs may be borne by hospitals in cases where patients become injured. Therefore, these unnecessary costs must be addressed by hospital administrators. Furthermore, the government should acknowledge the additional financial burden for such incidents from the viewpoint of hospital management and public health in a “superaging society” like Japan. Although some studies in other countries have focused on the estimated additional medical costs (AMCs) associated with falls,10–12 their results might not be applicable to hospitals in Japan because of the differences in health care systems. In addition, the LOS in Japan is the longest (18.7 days) among the OECD countries (mean LOS, 7.2 days) according to OECD 2011 health data,13 and fees for clinical services are essentially uniform throughout Japan under the social insurance medical fee schedule, regardless of hospital ownership, hospital type, and bed numbers. With regard to fall-related AMCs, Bates et al14 found that fallers were charged 4233 USD higher than their nonfaller counterparts who had been matched using age, sex, and LOS up to the time of the fall incidents. Moreover, the Quality in Australian Health Care Study estimated that the total cost of falls in hospitals was 2.1 million AUD per annum.15 Another study in the UK demonstrated that the cost of treating fall-related injuries amounted to approximately 70,000 GBP, with an additional 56 weeks of hospital stay in total required.16 However, because these findings may not be directly applicable to Japan, it is necessary to conduct an analysis of the AMCs associated with falls in the Japanese health care setting. Several studies have analyzed falls in Japanese hospitals but have generally only focused on small samples.17–19 In addition, few studies have quantified the AMCs or analyzed the breakdown of costs according to clinical service types. As a result, Japanese researchers and policy makers do not have access to basic precise data regarding the AMCs due to hospital-acquired falls. This study aimed to explore the AMCs arising from falls and their impact on various clinical services within hospitals by analyzing administrative profiling data linked with claims data. www.journalpatientsafety.com
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
1
J Patient Saf • Volume 00, Number 00, Month 2015
Hirose et al
METHODS Setting and Methods This study was approved by the respective institutional review board of each participating hospital. 1. Hospitals and incident reporting systems There were 2 participating hospitals in this analysis: Shimane University Hospital (SUH) and St. Mary's Hospital (SMH). Shimane University Hospital is a 616-bed (including 40 psychiatric care type beds) tertiary referral center located in Izumo, Shimane prefecture situated in the southwest of Honshu, which is Japan's largest island; SUH serves approximately 170,000 inpatients and 230,000 outpatients per year. St. Mary's Hospital is also a tertiary referral center and is located in Kurume, Fukuoka prefecture situated in the north of Kyushu Island; it has 1354 beds including 100 long-term care beds and 60 psychiatric care beds with an emergency medical care center. St. Mary's Hospital provides care for approximately 320,000 inpatients and 370,000 outpatients, and its emergency medical care center receives approximately 10,000 ambulances and 60,000 emergency patients annually. The average LOS durations were 18.0 days and 21.0 days at SUH and SMH, respectively. Both hospitals have implemented the same online electronic incident reporting system, which has been in use at SUH since 2006 and at SMH since 2003. The taxonomies, categorization methods, data forms, goals, and principles of the system are similar in both hospitals. 2. Injury level (influence level of harm) When submitting an incident report, the reporting party has to input information relevant to the incidents, including the injury level the patient experienced. Injury level is categorized based on an 8-level classification scheme according to National University Hospitals. These levels are level zero, where an incident did not occur; level 1, where an incident occurred but the patient was not harmed; level 2, where an incident occurred and the patient required examination and/or medical observation; level 3a, where an incident occurred and the patient required minor procedures; level 3b, where an incident occurred and the patient required major procedures (ie, surgical procedures or extended hospital stay) regardless of disability; level 4a, where an incident occurred and the patient was disabled without significant dysfunction and
cosmetic problems; level 4b, where an incident occurred and the patient was disabled with significant dysfunction and cosmetic problems; and finally, level 5, in which the patient died due to the incident. Therefore, incidents at level 2 or higher are associated with possible AMCs. We dichotomized severity into 2 classes: “major” injuries were defined as cases at level 3b and above and “minor” injuries as cases at levels 2 or 3a (Table 1). 3. Data and definition of variables We collected all incident reports filed in each institution between FY 2007 and FY 2009 (ie, April 2007 to March 2010). Injury level was categorized at both SUH and SMH according to the 8-level classification scheme described earlier. We identified 1190 cases with falls from 4669 incident reports at SUH and 1790 cases of falls from 7717 reports at SMH based on their incident reporting data. Additional medical costs were calculated using both participant hospitals' administrative profiling data linked with the Japanese social insurance medical fee schedule. 1) Medical costs in Japan In Japan, medical costs are computed using medical fees, which are calculated by summing charges billed during hospitalization (1 USD = 100 Japanese Yen) on a detailed fee-for-service basis according to the social insurance medical fee schedule under the nationally uniform universal health insurance system. Medical fees include those for major clinical service items, such as medical administrative charges, medication charges, injection charges, procedure/surgery charges, laboratory examination charges, diagnostic imaging charges, and hotel charges. As a result, medical fees are considered and frequently used as a good estimator of the costs of health care in Japan. The social insurance medical fee schedule undergoes minor revisions every 2 years. 2) Definition of variables ➢ FR: Fall rate (FR) was calculated using the following formula:
FR per 1000 patient-days = Number of reports for falls / [number of inpatients length of hospital stay] 1000. ➢ AMCs: As described in the “Injury level” of the “Setting and Methods” subsection, cases at level 2 or higher require examination and/or treatment and can therefore incur additional charges arising from these medical services that are
TABLE 1. Injury Level (Influence Level of Harm) Level 0 1
Continuity of Disability
–
2 Temporary 3a 3b
4a Permanent 4b 5
2
Severity
Description of Harm
No occurrence (Preventing an incidence)
Death
An error occurred and/or there was something wrong with drugs/medical devices, but an incident due to the error did not reach the patient. An incident occurred, but it did not do real harm to the patient (the incident may have had some influence on a patient.) Mild No procedure and treatment (e.g. the patient was under more stringent medical observation, his/her vital signs were slightly changed, and an examination for securing his/her safety was required.) Moderate Minor procedures and treatments were performed (eg, disinfection, fomentation, suturing of a skin, administration of analgesic, and so on, were performed.) Severe Major procedures and/or treatments were done.(eg, the patient’s vital signs was highly changed, a respirator was equipped, a surgery was done, extension of length of hospital stay, and admission of out-patient) Mild to moderate A patient remains permanently disabled and sequelae without significant dysfunction and cosmetic problem Moderate to severe A patient remains permanently disabled and sequelae with significant dysfunction and cosmetic problem A patient has died from a reason unrelated to the natural course of the primary disease.
www.journalpatientsafety.com
© 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Patient Saf • Volume 00, Number 00, Month 2015
not linked to their primary diseases. These costs are designated AMCs and calculated according to the Japanese social insurance medical fee schedule. In this study, medical administrative charges and hotel charges were not taken into account because the extensions in hospital stay as a direct result of falls could not be discerned. ➢ N: number of incident reports ➢ Nf: number of incident reports with falls ➢ Ne: number of potential cases with AMCs (cases at level 2 and higher based on incident reporting) ➢ iNe: number of reports in which AMCs were identified by researchers using administrative profiling data ➢ iCe: identified AMCs in the reports with AMCs ➢ ∑iCe: total identified AMCs ➢ ∑iCe / iNe: identified AMCs per case ➢ ∑eCe: total estimated AMCs; ∑eCe = Ne ∑iCe/iNe
4. Data collection and calculation of additional medical costs using administrative profiling data This study was conducted according to the following process, as illustrated in Figure 1. 1) Reports with injuries at level 2 and higher were extracted from incident reports filed during the target period. 2) Fall-related medical fees unrelated to the primary disease upon admission were extracted from all medical fees based on clinical services data and analyzed as medical costs. 3) With regard to clinical services actually provided, medical costs for each clinical service were collected from the administrative profiling data (claims data). 4) As stated previously, medical costs were calculated by the combination of clinical service items that were actually provided.
In the Japanese health care setting, lists of medical costs on a day-to day basis during hospitalization categorized by clinical
Additional Medical Costs Due to Falls
services are compiled at each hospital every month. In our investigation, we checked these lists and identified fall-related medical costs unrelated to the primary diseases.
RESULTS Overview of the 2 Participant Hospitals and Fall Rates Shimane University Hospital and SMH are both teaching hospitals, and the former is also an affiliate with its medical school. The hospitals are similar in their functions as tertiary referral hospitals. Table 2 provides an overview of both hospitals. There were a total of 4669 incident reports at SUH and 7717 reports at SMH, and these reports included 1190 cases and 1790 cases for falls at SUH and SMH, respectively. Therefore, the FR in SUH was 2.17, calculated from 1190 falls (number of incident reports for falls), 30,352 inpatients, and 18.0 days for LOS; similarly, the FR in SMH was 1.87, calculated from 1790 falls, 45,787 inpatients, and 21.0 days for LOS. There were no significant differences among fiscal years in the number of incident reports for falls and FR in each hospital.
Additional Medical Costs by Injury Level Table 3 shows AMCs by fiscal year and injury level at each hospital. In SUH, ∑iCe/iNe was 502 ± 3303 USD for minor injuries in FY 2007, 187 ± 237 USD for minor injuries and 5387 ± 6934 USD for major injuries in FY 2008 (P < 0.001), and 405 ± 1726 USD for minor injuries and 3636 ± 836 USD for major injuries in FY 2009 (P < 0.001). For the total study period, these values were 364 ± 2129 USD for minor injuries and 4336 ± 3645 USD for major injuries (P < 0.001). As there was no case with major injuries in FY 2007, ∑iCe / iNe for major injuries was not calculated. Similarly, ΣiCe / iNe in SMH was 114 ± 153 USD for minor injuries and 2505 ± 3538 USD for major injuries in FY 2007 (P < 0.001), 114 ± 114 USD for minor injuries and 1431 ± 1845 USD for major injuries in FY 2008 (P < 0.001),
FIGURE 1. Data collection and AMC calculation strategy. © 2015 Wolters Kluwer Health, Inc. All rights reserved.
www.journalpatientsafety.com
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
3
J Patient Saf • Volume 00, Number 00, Month 2015
Hirose et al
TABLE 2. An Overview of 2 Hospitals Shimane University Hospital (SUH) 2007 FY
2008 FY
St. Mary's Hospital (SMH)
2009 FY
total
2007FY
2008FY
2009FY
Total
Length of hospital 19.3 18.5 17.1 18.0 20.6 20.8 21.4 21.0 stay, days No. of in-patients 9738 10,051 10,563 30,352 15,549 15,076 15,112 45,737 No. of person-days 184,694 185,432 179,870 549,996 321,286 312,717 324,372 958,375 No. of incident reports 1442 1558 1669 4669 2760 2559 2398 7717 No. of reports for falls 382 (26.5%) 399 (25.6%) 409 (24.5%) 1190 (25.5%) 610 (22.1%) 565 (22.1%) 615 (25.6%) 1790 (23.2%) (proportion of incident reports) 2.07 2.15 2.27 2.17 1.90 1.81 1.90 1.87 Falls rate* (1000 patient-days) Falls rate 1000 patient-days = Number of reports for falls / (number of in-patients length of hospital stay) 1000. = Number of reports for falls/number of person-days 1000. *No statistical difference among fiscal years at each hospital.
and 114 ± 87 USD for minor injuries and 2470 ± 2695 USD for major injuries in FY 2009 (P < 0.001). For the total study period, these values were 114 ± 124 USD for minor injuries and 2267 ± 2811 USD for major injuries (P < 0.001). There were statistically significant differences in ΣiCe / iNe between injury levels but none by injury level among the fiscal years at each hospital.
Use of Clinical Services Table 4 shows AMCs by clinical services and injury level. When a patient is injured owing to a fall, he or she is generally required to be examined and treated. Specifically, a patient frequently requires x-ray photography of the injured region to evaluate whether there is a need for further examination and treatment.
TABLE 3. Additional Medical Costs by Injury Level SUH 2007
2008
2009
All years
SMH
Injury Level
Minor Injury
Major Injury
Minor Injury
Major Injury
iNe (Ne) ΣiCe ΣiCe/iNe ΣeCe iNe (Ne) ΣiCe ΣiCe/iNe ΣeCe iNe (Ne) ΣiCe ΣiCe/iNe ΣeCe iNe (Ne) ΣiCe ΣiCe/iNe ΣeCe
73 (192) 36,635 502 ± 3303 96,384 75 (207) 14,051 187 ± 237*1 38,709 80 (184) 32,388 405 ± 1726*2 74,520 228 (583) 83,074 364 ± 2,129*3 212,212
0 (0) 0 – – 2 (2) 10,774 5387 ± 6934*1 10,774 3 (3) 10,877 3636 ± 836*2 10,877 5 (5) 21,681 4336 ± 3645*3 21,681
166 (388) 19,002 114 ± 153*4 44,232 87 (353) 9932 114 ± 114*5 40,242 138 (381) 15,690 114 ± 87*6 43,434 391 (1122) 43,310 114 ± 124*7 127,908
11 (19) 27,551 2505 ± 3538*4 47,586 7 (8) 10,020 1431 ± 1845*5 11,448 16 (19) 39,520 2470 ± 2695*6 46,930 34 (46) 77,090 2267 ± 2811*7 104,282
Total
233,893
232,190
Additional medical cost: Fall-related medical charges arise from their examinations and/or procedures are not linked with primary diseases in case of injury level 2 and higher. Their costs are defined as additional medical costs. Ne: no. of potential cases with additional medical costs. There was no significance in ΣiCe / iNe between fiscal years at each injury level in both hospitals. iNe: no. of identified cases with additional medical costs. There were statistical significances in ΣiCe / iNe between levels (P < 0.001 except *2 and P value in *2 = 0.002). iCe: identified additional medical costs. ΣiCe: total identified additional medical costs. ΣeCe: total estimated additional costs; eCe = Ne iCe / iNe.
4
www.journalpatientsafety.com
© 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Patient Saf • Volume 00, Number 00, Month 2015
Additional Medical Costs Due to Falls
TABLE 4. Additional Medical Costs by Clinical Services SUH Injury Level (n) Medication Injection Procedure/Surgery Examination Imaging Others Total
SMH
Minor Injury (228)
Major Injury (5)
Minor Injury (391)
Major Injury (34)
1041 (43) 24 ± 59 4755 (12) 396 ± 1308 34,780 (26) 1338 ± 4078*1 1341 (12) 112 ± 129 33,741 (205) 165 ± 220*2 7417 (6) 1236 ± 1921 83,074 (228) 364 ± 2129*3
0 – 0 – 18,449 (5) 3690 ± 3073*1 0 – 1867 (5) 373 ± 416*2 1365 (4) 341 ± 281 21,681 (5) 4336 ± 3645*3
495 (110) 5 ± 7#1 2304 (28) 82 ± 156 2229 (52) 43 ± 82#2 392 (22) 13 ± 10#3 37,981 (339) 112 ± 72#4 7 (2) 4 ± 2#5 43,310 (391) 114 ± 124#6
1310 (23) 57 ± 86#1 1616 (17) 95 ± 109 61,922 (15) 4128 ± 2494#2 2301 (21) 110 ± 102#3 9726 (34) 286 ± 173#4 211 (9) 23 ± 20#5 77,090 (34) 2267 ± 2811#6
Upper: subtotal additional medical cost (identified case number); lower; mean additional cost. *1: no significant difference between injury levels. *2: A significant difference between injury levels (P = 0.042). *3: A significant difference between levels (P < 0.001). #1, 2, 3, 4, and 6: significant differences between injury levels (P < 0.001). #5: a significant difference between injury levels (P = 0.015).
Among the clinical services provided to fall patients at SUH, the most commonly used were imaging services at 90.1% (n = 210) of 233 cases, with 89.9% (n = 205) of 228 cases with minor injuries and 100% (n = 5) of cases with major injuries. Medication services were used in 18.5% (n = 43) of 233 cases, with 18.9% (n = 43) of 228 cases with minor injuries. Similarly, procedure/surgery services were used among 13.3% (n = 31) of 233 cases, with 11.4% (n = 26) of 228 cases with minor injuries and 100% (n = 5) of cases with major injuries. In SMH, imaging services were also the most commonly used clinical services, followed by medication services and procedure/ surgery services. Imaging services were used in 87.8% (n = 373) of 425 cases, with 86.7% (n = 339) of 391 cases with minor injuries and 100% (n = 34) of cases with major injuries. Medication services were used in 31.3% (n = 133) of 425 cases, with 28.1% (n = 110) of 391 cases with minor injuries and 67.6% (n = 23) of 34 cases with major injuries. Procedure/surgery services were used in 15.8% (n = 67) of 425 cases, with 13.3% (n = 52) of 391 cases with minor injuries and 44.1% (n = 15) of 34 cases with major injuries.
Additional Medical Costs by Clinical Services In both hospitals, AMCs for procedure/surgery services comprised the highest proportion of total AMCs, followed by AMCs for imaging services. In SUH, the proportions of AMCs for procedure/surgery services (53,229 USD: 34,780 for minor injuries and 18,449 USD for major injuries) to total AMCs (104,755 USD: 83,074 for minor injuries and 21,681 USD for major injuries) were 50.8% in total, 41.9% for minor injuries, and 85.1% for major injuries. Additional medical costs per case for procedure/surgery by injury level were 1338 ± 4078 USD for minor injuries and 3690 ± 3073 USD for major injuries, and there was a statistically significant difference between injury levels (P < 0.001). Although imaging services were the most commonly used services, the proportions of AMCs for this service to total AMCs were 34.0% in total, © 2015 Wolters Kluwer Health, Inc. All rights reserved.
40.6% for minor injuries, and 8.6% for major injuries. Additional medical costs per case for imaging services were 165 ± 220 USD for minor injuries and 373 ± 416 USD for major injuries, and there was a statistically significant difference between injury levels (P = 0.042). In SMH, the proportions of AMCs for procedure/surgery services (64,151 USD: 2229 for minor injuries and 61,922 USD for major injuries) to total AMCs (120,400 USD: 43,310 for minor injuries and 77,090 USD for major injuries) were 53.3% in total, 5.1% for minor injuries, and 80.3% for major injuries. Additional medical costs per case for procedure/surgery services by injury level were 43 ± 82 USD for minor injuries and 4128 ± 2494 USD for major injuries, and there was a statistically significant difference between injury levels (P < 0.001). Despite the frequent use of imaging services, the impact of this service on overall costs was relatively small from a monetary perspective. In contrast, although procedure/surgery services were used less frequently, these services had a comparatively large impact on the additional financial burden due to falls. In SUH, for example, the proportion of imaging services to clinical services was 90.1% in all 233 cases with AMCs, and the proportion of this service to total AMCs was only 34.0%. In contrast, the proportion of procedure/surgery services to all clinical services was just 13.3% of the 233 cases, but the proportion of this service to total AMC was 50.8%.
DISCUSSION Although hospital-acquired falls is a critical and common issue in the inpatient health care setting, it has not been sufficiently addressed from the viewpoint of financial burden and types of clinical services used, particularly in Japan.
Basic Epidemiology of Hospital-Acquired Falls The proportions of incident reports for falls were 25.5% (1190/4669) at SUH and 23.2% (1790/7717) at SMH, and these www.journalpatientsafety.com
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
5
J Patient Saf • Volume 00, Number 00, Month 2015
Hirose et al
proportions fall into the ranges previously reported.1–3 Incident reports for falls with minor injuries comprised 99.1% (583/588) of cases at SUH and 96.1% (1122/1168) of cases at SMH. Fall rates in this study were 2.17 at SUH and 1.87 at SMH, which were lower than those reported in Hitcho et al5 (overall FR, 3.38; ranging from 0.80 [orthopedics)] to 6.12 [Medicine]). This difference in FRs may be in part due to the fact that the study by Hitcho had been performed in 2004 as a prospective descriptive study. In contrast, our study had been done between 2007 and 2009 as a retrospective observational case study. Our target hospitals might have been taking actions for falls compared with Hictho's hospital. In addition, the FR of SUH was higher than that of SMH. This may be partly because SUH is an academic center, whereas SMH is not. Furthermore, the mean age of patients who experienced falls in SUH was higher than in SMH. This may be due to differences in population aging rate: the population aging rates in 2009 were 22.7% in Japan, 22.0% in Fukuoka prefecture, and 29.0% in Shimane prefecture.20
Additional Medical Costs In this study, the numbers of cases with identified AMCs (iNe) were 233 (39.6%) among 588 potential cases at SUH and 425 (36.4%) among 1168 potential cases at SMH. At SUH, the total amount of identified AMCs (ΣiCe) was 104,755 USD and the identified AMC per case (ΣiCe / iNe) was 450 ± 2235 USD, regardless of injury level; at SMH, these values were 120,400 USD and 286 ± 985 USD, respectively. AMCs for minor injuries at SUH (364 ± 2129 USD) were statistically larger (P = 0.021) than at SMH (114 ± 124 USD), and AMCs for major injuries at SUH (4336 ± 3645 USD) were numerically larger (P = 0.1246) than at SMH (2267 ± 2811 USD). When examining cases with major injuries according to fallrelated conditions and their associated AMCs, the 5 cases in SUH with major injuries included the following conditions: acute subdural hematoma (men: n = 1; 72 years; AMC,10,290 USD), femoral neck fracture (women: n = 1; 75 years; AMC,4502 USD), trochanteric fracture (women: n = 2; 87 years and 79 years, AMC, 2834 USD and 3572 USD), and avulsion fracture of the tibial attachment of the posterior cruciate ligament (men: n = 1; 54 years; AMC,484 USD). In SMH, the 34 cases with major injuries included 9 femoral neck fractures (men: n = 1; 81 years; AMC, 3903 USD; women: n = 8; 80.3 ± 8.4 years, AMC,2438 ± 2468 USD), 6 trochanteric fractures (men: n = 3; 74.0 ± 5.6 years; AMC,2038 ± 2456 USD; women: n = 3; 81.3 ± 7.5 years; AMC, 7562 ± 3231 USD), one femoral fracture (men: n = 1, 61 years, AMC = 4263 USD), and one acute subdural hematoma (men: n = 1; 79 years; AMC,7870 USD). Of the 34 cases, there were 12 cases with AMCs greater than 2000 USD; further examination showed that there were 5 femoral neck fractures (AMC, 4571 ± 466 USD and 3903–4951 USD), 4 trochanteric fractures (AMC,6889 ± 2962 USD and 4871–11,284 USD), one femoral fracture (AMC,4263 USD), one patellar fracture (AMC, 5060 USD), and one acute subdural hematoma (AMC,7870 USD). Therefore, the AMCs for femoral neck fracture, trochanteric fracture, and acute subdural hematoma were higher at both hospitals among cases with major injuries. In 1995, Bates et al reported 6.6 falls per 1000 admissions, and patients who had falls had hospital charges that were 4233 USD higher than those of nonfallers. The researchers found that when compared with 37 patients with less serious injuries, 25 patients who had experienced fractures, dislocations, or intracranial injuries had median charges that were 10,421 USD higher, but this result was not statistically significant.14 The 37 cases in that report may correspond to the cases with minor injuries in our study, whereas
6
www.journalpatientsafety.com
the 25 patients may correspond to the cases with major injuries. Diseases with major injuries were common between the 2 studies. In SUH, the 5 patients with major injuries had median charges of 3572 USD (mean: 4330 USD, 484–10,290 USD), and the highest hospitalization charge was in a patient with acute subdural hematoma. In SMH, the 34 cases with major injuries had median charges of 559 USD (mean: 2267 USD, 58–11,284 USD), and the highest hospitalization charge was in a patient with trochanteric fracture. The AMCs for patients with major injuries in the study by Bates et al were larger than in our study. This may be partly affected by our noninclusion of administrative charges and hotel charges in our AMC calculations. Furthermore, the observation period of the study by Bates et al was from January 1, 1987, to March 31, 1991; in contrast, our study duration was between April 1, 2007, and March 31, 2010. In addition, the prices and exchange rates of both countries should be taken into account. Finally, these differences in the results between the studies may be influenced by methodological differences. As Bates et al is a case-control study and our study is an observational case study, the evidence level of our study may be lower than that of Bates et al. Nevertheless, our study aimed to explore how AMCs due to falls were used in clinical services, and we could not detail the AMCs without implementing our study method. Therefore, we should avoid crude comparisons of AMCs between the 2 studies. In Japan, several studies have analyzed AMCs associated with falls using an approach similar to that of our study. Kamioka et al17 surveyed 6 cases of home-dwelling elderly persons with falls that required hospitalization and explored their associated medical costs. The 6 cases included 3 subjects with surgery and 3 subjects without. In the 3 cases with surgery (similar to the cases with major injuries in our study), hospitalization charges per case were 2174 USD; our results were 4336 ± 3645 USD at SUH and 2267 ± 2811 USD at SMH. Although our study subjects were inpatients at teaching hospitals and Kamioka et al analyzed homedwelling elderly patients, these 2 studies were not fundamentally different, as both used similar analytical approaches. Our study period was from April 2007 to March 2009, whereas Kamioka et al was conducted between May 2000 and July 2001. The differences in health care charges may be due in part to revisions to Japan's social insurance medical fee schedule. According to Kobayashi et al,18 medical costs per incident were 94.35 ± 69.10 USD (28 cases; median, 83.24 USD; minimum, 20.40 USD; maximum, 429.70 USD), which was similar to that of our cases with minor injuries (364 ± 2129 USD at SUH and 114 ± 124 USD at SMH). In their study, there was only one case with an accident with an additional cost of 87.45 USD, and the AMC per incident in their study was lower than in ours. Kobayashi et al performed their study in 2002 with a methodology similar to ours in a regional medical center with 400 beds. The differences in AMCs between the 2 studies may be influenced by differences in the number of cases and revisions to the social insurance medical fee schedule. With regard to accident cases, the results were not comparable. In an analysis conducted over the period of 1 year from April 2008 to March 2009, Shibuya et al19 reported that falls occurred in 19 cases, including 4 cases with level 3a, 12 cases with level 3b, 2 cases with level 4, and one case with level 5. Although they had calculated the average charges per patient without consideration of injury level, the calculated cost was 2917.53 USD, which was similar to our results, partly perhaps because both studies were conducted around the same time period. Among the 228 cases with minor injuries at SUH, imaging services were used in 205 cases (89.9%), and their AMCs accounted for 40.6% (33,741/83,074 USD) of total AMCs, as © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Patient Saf • Volume 00, Number 00, Month 2015
shown in Table 4. Medication services were the second most frequently used, with 43 (18.9%) of 228 cases incurring AMCs for these services. However, the amount of AMCs was only 1.25% (1041/83,074 USD) for medication services. Similarly, imaging services were used in 339 (86.7%) of the 391 cases with minor injuries at SMH, and their AMCs accounted for 87.7% (37,981/ 43,310 USD) of total AMCs. Medication services were again the second most frequently used, with 110 (28.1%) of 391 cases. The amount of AMCs was also low at SMH, at only 1.14% (495/ 43,310 USD). In contrast, procedure/surgery services at SUH were less frequently used in cases with minor injuries (26/228 cases [11.4%]) but used in all cases with major injuries (5/5 cases). Similarly, although these services were the third most frequently used in the minor injury cases (52/391 cases [13.3%]) at SMH, the proportions of AMCs for these services were 5.1% (2229/43,310 USD) at the minor injury level and 80.3% (61,922/77,090 USD) at the major injury level. These results revealed that imaging services were the most frequently used services in cases with falls, but procedure/surgery services incurred the most AMCs, thereby making procedure/ surgery services the most expensive from the viewpoint of medical costs. These findings are novel, and there are no published studies to our knowledge that have calculated the AMCs using administrative profiling data based on incident reporting data. In addition, few studies have addressed the various costs and therapeutic items by clinical services due to falls as conducted in this study.
CONCLUSIONS It is essential to quantify the AMCs due to various adverse events. In particular, hospital-acquired falls account for most adverse events, and understanding the actual situation is an important step in reducing these events. This is the first study to approach this research problem by using administrative profiling data linked with claims data under the Japanese universal social insurance medical fee schedule, and the first to calculate AMCs according to clinical services. Our study showed that fall-related AMCs of cases with major injuries were higher than those with minor injuries at 2 teaching hospitals, mainly because the former generally required procedure/ surgery services for conditions such as hip fractures and acute subdural hematoma. The results in this study may be useful for medical institutions seeking to ensure patient safety and to improve overall health care quality. REFERENCES 1. Krauss MJ, Nguyen SL, Dunagan WC, et al. Circumstances of patient falls and injuries in 9 hospitals in a Midwestern Healthcare System. Infect Cont Hosp Ep. 2007;28:544–550. 2. Shaw R, Drever F, Hughes H, Osborn S, et al. Adverse events and near miss reporting in the NHS. Qual Saf Health Care. 2005;14:279–283.
© 2015 Wolters Kluwer Health, Inc. All rights reserved.
Additional Medical Costs Due to Falls
3. Healey F, Scobie S, Oliver D, et al. Falls in English and Welsh hospitals: a national observational study based on retrospective analysis of 12 months of patient safety incident reports. Qual Saf Health Care. 2008;17:424–430. 4. Naessens JM, Campbell CR, Huddleston JM, et al. A comparison of hospital adverse events identified by three widely used detection methods. Int J Qual Health C. 2009;21:301–307. 5. Hitcho EB, Krauss MJ, Birge S, et al. Characteristics and circumstances of falls in a hospitalized setting. J Gen Intern Med. 2004;19:732–739. 6. Halfon P, Eggli P, Van Melle G, et al. Risk of falls for hospitalized patients: a predictive model based on routinely available data. J Clini Epidemiol. 2001;54:1258–1266. 7. Schwendimann R, Buhler H, De Geest A, et al. Falls and consequent injuries in hospitalized patients: effects of an interdisciplinary falls prevention program. BMC Health Serv Res. 2006;6:69. 8. Hendrich A, Nyhuis A, Kippenbrock T, et al. Hospital falls: development of a predictive model for clinical practice. Appl Nurs Res. 1995;8:129–139. 9. Hill DK, Vu M, Walsh W. Falls in the acute hospital setting—impact on resource utilization. Aust Health Rev. 2007;31:471–477. 10. Johnson WG, Brennan TA, Newhouse JP, et al. The economic consequences of medical injuries. Implications for a no-fault insurance plan. J Am Med Assoc. 1992;267:2487–2492. 11. Thomas EJ, Studdert DM, Newhouse JP, et al. Costs of medical injuries in Utah and Colorado. Inquiry. 1999;36:255–264. 12. Brown P, McArthur C, Newby L, et al. Cost of Medical Injury in New Zealand: a retrospective cohort study. J Health Serv Res Pol. 2002;17: S29–S34. 13. OECD. Health at a glance 2011: OECD indicators. Available at: http://www.oecd.org/els/health-systems/49105858.pdf. 14. Bates DW, Pruess K, Sourney P, et al. Serious falls in hospitalized patients: correlates and resource utilization. Am J Med. 1995;99:137–143. 15. Rigby K, Clark RB, Runciman WB. Adverse events in health care: setting priorities based on economic evaluation. J Qual Clin Pract. 1999;19:7–12. 16. Nadkarni JB, Iyengar KP, Dussa C, et al. Orthopaedic injuries following falls by hospital in-patients. Gerontology. 2005;51:329–333. 17. Kamioka H, Kuroyanagi R, Mutoh Y, et al. Influence of accidental falls by home-dwelling elderly on the household economy. J Phys Educ Med. 2002;3:35–46. [in Japanese]. 18. Kobayashi M, Ikeda S, Muto M. The estimation of medical cost incurred by incidents and adverse events. Health Care and Society (Iryou to Shakai). 2006;16:85–95. [in Japanese]. 19. Shibuya A, Hanai K, Arai Y, et al. The cost of adverse events in hospitalized patients. Kitasato Medical Journal. 2009;39:165–170. 20. Statistics Bureau, Ministry of Internal Affairs and Communications. Statistics Japan: Available at: http://www.stat.go.jp/data/jinsui/2009np/pdf/ gaiyou.pdf#page=12.
www.journalpatientsafety.com
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
7