- Email: [email protected]
Diagnostic Imaging Services in Magnet and Non-Magnet Hospitals: Trends in Utilization and Costs
ORIGINAL ARTICLE
Diagnostic Imaging Services in Magnet and Non-Magnet Hospitals: Trends in Utilization and Costs Jayani Jayawardhana, PhD a , John M. Welton, PhD, RN, FAAN b Abstract Purpose: The purpose of this study was to better understand trends in utilization and costs of diagnostic imaging services at Magnet hospitals (MHs) and non-Magnet hospitals (NMHs). Methods: A data set was created by merging hospital-level data from the American Hospital Association’s annual survey and Medicare cost reports, individual-level inpatient data from the Healthcare Cost and Utilization Project, and Magnet recognition status data from the American Nurses Credentialing Center. A descriptive analysis was conducted to evaluate the trends in utilization and costs of CT, MRI, and ultrasound procedures among MHs and NMHs in urban locations between 2000 and 2006 from the following ten states: Arizona, California, Colorado, Florida, Iowa, Maryland, North Carolina, New Jersey, New York, and Washington. Results: When matched by bed size, severity of illness (case mix index), and clinical technological sophistication (Saidin index) quantiles, MHs in higher quantiles indicated higher rates of utilization of imaging services for MRI, CT, and ultrasound in comparison with NMHs in the same quantiles. However, average costs of MRI, CT, and ultrasounds were lower at MHs in comparison with NMHs in the same quantiles. Conclusions: Overall, MHs that are larger in size (number of beds), serve more severely ill patients (case mix index), and are more technologically sophisticated (Saidin index) show higher utilization of diagnostic imaging services, although costs per procedure at MHs are lower in comparison with similar NMHs, indicating possible cost efficiency at MHs. Further research is necessary to understand the relationship between the utilization of diagnostic imaging services among MHs and its impact on patient outcomes. Key Words: Diagnostic imaging services, Magnet hospitals, utilization, costs J Am Coll Radiol 2015;12:1357-1363. Copyright 2015 American College of Radiology
INTRODUCTION Diagnostic imaging services play a key role in the delivery of health care in the United States and have become essential components in identifying, treating, and managing most major medical conditions and diseases. The increased utilization of diagnostic imaging services has contributed to rising health care costs in the United States, accounting for more than 10% of total health care a Department of Health Policy and Management, College of Public Health, University of Georgia, Athens, Georgia. b College of Nursing, University of Colorado-Anschutz Medical Campus, Denver, Colorado. Corresponding author and reprints: Jayani Jayawardhana, PhD, University of Georgia, College of Public Health, Department of Health Policy and Management, 100 Foster Road, Athens, GA 30602; e-mail: jayaward@ uga.edu. This research was funded by a grant from the Harvey L. Neiman Health Policy Institute. The authors have no conflicts of interest related to the material discussed in this article.
ª 2015 American College of Radiology 1546-1440/15/$36.00 n http://dx.doi.org/10.1016/j.jacr.2015.09.011
costs [1-3]. Although overall utilization of diagnostic imaging services grew rapidly in the early 2000s, recent studies indicate a decline in imaging utilization in recent years [4-7]. However, little is known about utilization trends of diagnostic imaging services at Magnet hospitals (MHs) and non-Magnet hospitals (NMHs) and the trends in associated costs of imaging. MHs are known for their high quality of patient care, excellence in nursing practices, higher retention rates for well-qualified nurses, improved outcomes of care, and greater propensity to use evidence-based care [8-12]. As part of obtaining MH designation, MHs make large investments in resources [13], including high-cost resources such as diagnostic imaging services. Given access to greater resources and capabilities, MHs may be more likely to utilize high-cost services such as diagnostic imaging services in comparison with NMHs. Given the availability of resources at MHs, the question of interest
1357
would be whether MHs are more (or less) likely to utilize high-cost resources such as diagnostic imaging services in comparison with NMHs. To date, 418 hospitals in the United States have been designated as MHs by the American Nurses Credentialing Center (ANCC), which was established by the American Nurses Association in the early 1990s [14]. The concept of MHs originates from an early study in the 1980s that found that a unique set of characteristics was responsible for attracting and retaining highly qualified nurses during a nurse shortage [14,15]. Many studies have examined the association between MH status and various outcomes of quality of care, such as mortality rates, length of stay, patient satisfaction, patient safety practices, and net patient revenue gains [8,11,12,16]. However, little is known about utilization trends for high-cost health care services such as diagnostic imaging services within MHs and NMHs and their impact on costs of care. Because MHs invest heavily in the resources needed in the process of obtaining Magnet status, it is likely that MHs have better access to more advanced technological resources and procedures, such as diagnostic imaging services, than NMHs. Similarly, because of their reputation for providing high-quality patient care and excellence in nursing practices, MHs may also attract patients with severe conditions or complications. Generally, MHs tend to be larger in bed size in comparison with NMHs. Although these reasons could lead to higher utilization of diagnostic imaging services at MHs in comparison with NMHs in general, it would be interesting to investigate if MHs tend to have higher utilization and costs of care of diagnostic imaging services in comparison with NMHs when controlling for technological sophistication, severity of patients’ conditions, and hospital size. In this study, we explored trends in the utilization of diagnostic imaging services and associated costs of such services within MHs and NMHs using inpatient data from acute care hospitals across ten states in the United States from 2000 to 2006. Specifically, we examined trends in utilization and costs of care for three diagnostic imaging services at MHs and NMHs: CT, MRI, and ultrasound.
METHODS The sample data set was created by merging hospital-level data from the American Hospital Association’s (AHA) annual survey and Medicare cost reports (Healthcare Cost Report Information System [HCRIS]), Magnet recognition status data from the ANCC, and individual-level inpatient data from the Healthcare Cost and Utilization Project State Inpatient Databases (HCUP-SID). The AHA annual survey 1358
data include information on hospital characteristics such as bed size, location, ownership, personnel information, and types of services provided. HCRIS data include information on costs of care at hospitals. The ANCC lists the hospitals that received Magnet recognition on its website, along with their addresses and the years they received Magnet status, as well as the follow-up years when they were recredentialed. The information from the ANCC website was collected in 2011 and was used to identify the MHs and the years they received Magnet status in the merged data set. The HCUPSID data include information on individual inpatient utilization of health care services, including diagnostic imaging services. The AHA annual survey, HCRIS, and HCUP-SID data are widely used in many research studies. The utilization of individual imaging services in HCUPSID data was identified using International Classification of Diseases, radiologic procedure codes and were aggregated to the hospital-year level. These data were available for individual patient discharges. Cost data identifiable in the HCRIS database included MRI, CT, and ultrasound. We restricted the data sample only to urban hospitals because almost all MHs were located in urban areas. The final data sample included data from Arizona, California, Colorado, Florida, Iowa, Maryland, North Carolina, New Jersey, New York, and Washington from 2000 to 2006 because these are the only states and years for which data were accessible. These states are regionally diverse and include both MHs and NMHs. The total study sample included 178 MH observations and 5,030 NMH observations, a total of 5,208 observations. The unit of analysis was the hospital-year. It should be noted that we did not have a way to identify a hospital that may have had Magnet status during the study period (2000-2006) but lost its Magnet status by 2011 because the ANCC does not disclose this information. Using the study sample, we examined trends in utilization and costs of three diagnostic imaging services, MRI, CT, and ultrasound, within MHs and NMHs from 2000 to 2006. We conducted statistical significance tests, using c2 tests for categorical variables and Student’s t test for continuous variables, to investigate whether there were significant differences between the means of utilization and costs of imaging services at MHs and NMHs within each year. Similar significance tests were also carried out for other key hospital characteristics, such as ownership (1 ¼ nonprofit, 0 ¼ otherwise), percentage of Medicare and Medicaid discharges, total inpatient discharges, number of beds, membership in a system of hospitals (1 ¼ system, 0 ¼ otherwise), the degree of market concentration in the hospital’s health care referral region (Herfindahl-Hirschman index) [17], the hospital’s clinical Journal of the American College of Radiology Volume 12 n Number 12PB n December 2015
technological sophistication (Saidin Index [SI]) [18], and a hospital case mix index (CMI) based on diagnosis-related group weights to adjust for the clinical complexity of patients treated at the hospital. Because MHs could be larger and more technologically sophisticated and thus might attract severe patients in comparison with NMHs, we also performed tests to examine whether there were statistically significant differences between the means of utilization and costs of imaging services at MHs and NMHs when the hospitals were matched on the basis of bed size, severity of patients’ conditions (CMI), and technological sophistication (SI). The analyses were performed using Stata version 13.1 for Windows (StataCorp LP, College Station, Texas).
RESULTS Table 1 shows the descriptive data by Magnet status of hospitals for each year from 2000 to 2006. To decrease the complexity of displaying the findings, statistical significance is summarized over multiple years. MHs are larger (in bed size, numbers of admissions, and discharges; P < .01), have higher nursing and full-time equivalent intensity (P < .05), have a higher percentage of registered nurses (P < .05), are more likely to be teaching hospitals (P < .05), are more technologically sophisticated (SI; P < .01), and treat more severe patients (CMI; P < .01) in comparison with NMHs. MHs are also more likely to be in more competitive markets (Herfindahl-Hirschman index), have longer lengths of stay, have lower shares of Medicaid and Medicare patients, and be nonprofit in comparison with NMHs, although differences between the means of these characteristics were not statistically significant. Interestingly, average costs of MRI examinations, CT scans, and ultrasound procedures were higher at NMHs in comparison with MHs in most years, although the differences in the means were not statistically significant. Although the utilization of MRI examinations, CT scans, and ultrasound procedures was mostly higher at NMHs in early years (2000-2001), utilization at MHs mostly increased between 2003 and 2005 but dropped again in 2006. Figure 1 presents utilization trends in MRI examinations, CT scans, and ultrasound procedures by Magnet status across years. Although the average number of MRI examinations performed per thousand admissions at MHs increased from 2000 to 2003, it decreased between 2003 and 2006. Except for 2000 and 2006, MHs had a higher utilization rate of MRI examinations each year in comparison with NMHs. Although MHs did not report utilization of CT scans during 2000 and 2001, their average utilization of CT scans per thousand admissions increased between 2002 and 2005 but dropped again to almost zero by 2006.
Interestingly, the utilization of CT scans at NMHs was higher in comparison with MHs each year, although NMHs’ CT scan utilization was decreasing over the years. Similarly, the average number of ultrasound procedures performed per thousand admissions at NMHs was highest in 2000 and gradually decreased each year until 2006. Furthermore, the utilization of ultrasound procedures was higher at NMHs in comparison with MHs in 2000, 2001, and 2006. The utilization of ultrasound procedures by MHs increased from 2000 to 2005 but dropped again in 2006. Figure 2 presents trends in average costs of each procedure by Magnet status across years. Average costs of MRI examinations were higher at MHs in comparison with NMHs each year except for 2000 and 2006. Although average costs of MRI examinations at MHs increased between 2000 and 2003, they were on the decline between 2004 and 2006. However, average costs of MRI examinations at NMHs were increasing over the years. Because there was no report of utilization of CT scans by MHs in 2000 and 2001, average costs of CT scans would be zero. Similar to utilization, average costs of CT scans at MHs increased from 2002 to 2005 but dropped again in 2006. Interestingly, average costs of CT scans at NMHs increased over the years and were also higher in comparison with MHs each year except for 2004 and 2005. Although the average costs of ultrasound procedures at NMHs increased over the years, the costs were higher at MHs in comparison with NMHs each year except for 2000 and 2001. Table 2 presents the means of utilization and average costs of diagnostic imaging services across MHs and NMHs when the hospitals were matched on the basis of bed size, severity of patients’ conditions (CMI), and technological sophistication (SI) by quantile across all years. Bed size, CMI, and SI were divided into three quantiles, each with quantile 1 indicating smaller hospitals, less severe patient conditions, or less technologically sophisticated hospitals and quantile 3 indicating larger hospitals, more severe patient conditions, or more technologically sophisticated hospitals. Whereas NMHs in bed quantile 1 used diagnostic imaging services, MHs in bed quantile 1 did not use any of the procedures. In bed quantile 2, NMHs utilized more MRI examinations, whereas MHs utilized more CT scans and ultrasound procedures. Average costs of MRI examinations and ultrasound procedures were higher at MHs in bed quantile 2, whereas average costs of CT scans were higher at NMHs. However, differences between the means were not significant either in utilization or in average costs within bed quantiles 1 and 2. This is likely due to the small sample size of MHs in those respective categories. In bed quantile 3, MHs had higher utilization of
Journal of the American College of Radiology Jayawardhana, Welton n Imaging Services in Magnet and Non-Magnet Hospitals
1359
1360 Table 1. Descriptive statistics of the study sample
Journal of the American College of Radiology Volume 12 n Number 12PB n December 2015
Variable Number of MRI examinations Number of CT scans Number of ultrasound studies AC of MRI AC of CT AC of ultrasound Hospital characteristics Beds Admissions (1,000) Discharges (1,000) Nursing intensity Percentage RNs FTE intensity RN/LPN ratio Nonprofit HHI Medicaid share Medicare share System (1) Teaching (1) SI CMI LOS n
NMH 35.05
2000 2001 2002 MH P NMH MH P NMH MH 18.88 0.77 35.48 65.10 0.55 36.51 126.11
294.68 0.00 0.33 257.90 257.36 112.63 0.62 214.53
27.30 39.92 6.52
0.81 0.61 27.76 0.00 0.50 47.01 0.14 0.50 6.53
2003 P NMH MH P 0.02 34.84 85.41 0.11
NMH 37.72
2004 2005 MH P NMH MH P 57.33 0.48 36.63 74.31 0.14
2006 NMH MH P 32.87 53.93 0.45
0.00 0.33 263.72 206.11 0.78 255.54 169.81 0.63 243.48 226.15 0.91 263.57 288.79 0.87 204.49 2.52 0.29 91.30 0.61 204.20 489.26 0.11 185.62 390.56 0.15 169.21 544.59 0.00 162.89 599.21 0.00 145.04 47.52 0.42
10.69 0.70 0.00 0.45 0.13 0.38
32.04 65.39 9.34
10.09 0.53 38.99 12.00 0.36 79.73 0.90 0.30 10.37
13.66 0.58 22.04 0.27 2.44 0.32
42.27 38.74 0.91 46.96 91.63 22.09 0.22 113.52 11.00 19.47 0.26 13.99
24.38 0.57 65.95 0.48 1.49 0.12
59.03 13.66 0.33 117.75 34.07 0.33 12.14 2.35 0.24
249.42 492.75 0.00 253.46 448.60 0.00 248.27 425.00 0.00 246.60 383.07 0.00 249.00 376.77 0.00 252.25 396.25 0.00 260.81 391.41 0.00 11.40 29.35 0.00 11.61 25.70 0.00 11.92 22.03 0.00 11.99 19.22 0.00 12.22 19.06 0.00 12.36 19.73 0.00 12.53 19.23 0.00 10.73
26.00 0.00
11.07
23.55 0.00
11.42
20.44 0.00
11.49
19.62 0.00
11.69
19.01 0.00
11.88
19.84 0.00
12.06 19.26 0.00
7.52 10.57 0.00 7.37 9.75 0.00 7.49 9.03 0.01 7.60 9.19 0.00 7.73 8.80 0.01 7.87 9.06 0.00 8.29 9.32 0.05 0.88 0.97 0.00 0.88 0.96 0.00 0.89 0.92 0.02 0.89 0.93 0.01 0.90 0.94 0.00 0.90 0.95 0.00 0.91 0.95 0.00 26.24 34.61 0.00 26.38 33.67 0.01 26.56 31.44 0.01 26.09 32.20 0.00 26.21 30.08 0.01 26.42 30.89 0.00 24.66 30.01 0.00 14.99 95.39 0.00 15.17 122.08 0.00 16.09 28.80 0.10 19.08 24.24 0.64 21.19 25.83 0.53 22.92 31.48 0.27 25.01 38.50 0.24 0.66 0.88 0.20 0.67 0.80 0.37 0.67 0.84 0.11 0.66 0.81 0.10 0.66 0.85 0.02 0.66 0.83 0.01 0.59 0.74 0.11 0.12 0.18 0.12 0.12 0.19 0.05 0.12 0.13 0.60 0.12 0.16 0.06 0.12 0.15 0.05 0.12 0.14 0.24 0.12 0.13 0.67 0.15 0.10 0.32 0.15 0.13 0.63 0.15 0.10 0.09 0.16 0.11 0.11 0.16 0.11 0.03 0.15 0.11 0.02 0.17 0.13 0.10 0.32 0.25 0.12 0.33 0.29 0.34 0.34 0.32 0.54 0.34 0.37 0.44 0.35 0.36 0.55 0.35 0.36 0.66 0.32 0.31 0.60 0.61 0.25 0.04 0.61 0.40 0.17 0.62 0.47 0.19 0.64 0.67 0.77 0.63 0.64 0.90 0.63 0.63 0.89 0.69 0.63 0.52 0.09 0.75 0.00 0.09 0.60 0.00 0.09 0.37 0.00 0.09 0.37 0.00 0.09 0.33 0.00 0.09 0.27 0.00 0.08 0.19 0.07 1.77 4.25 0.00 1.87 4.01 0.00 1.97 3.34 0.00 1.96 3.11 0.00 2.02 3.16 0.00 1.99 3.17 0.00 2.04 3.38 0.00 1.41 1.75 0.00 1.41 1.67 0.00 1.40 1.61 0.00 1.38 1.58 0.00 1.39 1.56 0.00 1.39 1.59 0.00 1.43 1.63 0.00 4.70 5.01 0.49 4.77 5.15 0.54 4.78 4.90 0.74 4.73 5.10 0.22 4.72 5.04 0.26 4.65 5.10 0.01 4.48 4.63 0.45 779 8 798 10 749 19 754 27 740 39 700 48 510 27
Note: These represent the total number of observations available for MHs and NMHs for each year in the study sample. AC ¼ average cost; CMI ¼ case mix index; FTE ¼ full-time equivalent; HHI ¼ Herfindahl-Hirschman index; LOS ¼ length of stay; LPN ¼ licensed practical nurse; MH ¼ Magnet hospital; NMH ¼ non-Magnet hospital; RN ¼ registered nurse; SI ¼ Saidin index.
Average number of CT scans per thousand Admissions
by year and Magnet status of Hospitals
by year and Magnet status of Hospitals
0
0
5
1
2
10
3
15
4
20
5
25
Average number of MRI exams per thousand Admissions
2000
2001
2002
2003
Magnet
2004
2005
2006
2000
2001
2002
Non-Magnet
2003
Magnet
2004
2005
2006
Non-Magnet
Average number of Ultrasounds per thousand Admissions
0
10
20
30
by year and Magnet status of Hospitals
2000
2001
2002
2003
2004
Magnet
2005
2006
Non-Magnet
Fig 1. Utilization of MRI examinations, CT scans, and ultrasound procedures by Magnet and non-Magnet hospitals.
MRI examinations and ultrasound procedures, whereas NMHs had higher utilization of CT scans (P < .05). In bed quantile 3, average costs of all three procedures were higher (P < .10) at NMHs in comparison with MHs. In CMI quantile 1, NMHs had higher utilization of each of the procedures and higher average costs for MRI
examinations in comparison with MHs, whereas costs of ultrasound procedures (P < .10) and CT scans were higher at MHs. In CMI quantile 2, NMHs had higher utilization of MRI examinations and CT scans and higher average costs for CT scans and ultrasound procedures, although their utilization of ultrasound procedures was Average Costs of CT scans by year and Magnet status of Hospitals
2000
2001
2002
2003
Magnet
2004
2005
0
0
200
200
400
400
600
600
Average Costs of MRIs by year and Magnet status of Hospitals
2006
2000
2001
Non-Magnet
2002 Magnet
2003
2004
2005
2006
Non-Magnet
Average Costs of Ultrasounds
0
100
200
300
400
500
by year and Magnet status of Hospitals
2000
2001
2002 Magnet
2003
2004
2005
2006
Non-Magnet
Fig 2. Average costs of MRI examinations, CT scans, and ultrasound procedures by Magnet and non-Magnet hospitals.
Journal of the American College of Radiology Jayawardhana, Welton n Imaging Services in Magnet and Non-Magnet Hospitals
1361
Table 2. Mean utilization and average costs of imaging services at MHs and NMHs matched on each hospital characteristic quantile of bed size, severity of patients’ illnesses (CMI), and technological sophistication (SI)
Bed quantile Hospital 1 MHs NMHs P 2 MHs NMHs P 3 MHs NMHs P CMI quantile 1 MHs NMHs P 2 MHs NMHs P 3 MHs NMHs P SI quantile 1 MHs NMHs P 2 MHs NMHs P 3 MHs NMHs P
n 8 1,732 52 1,682 118 1,616
22 1,714 38 1,698 118 1,618
22 1,715 28 1,714 128 1,601
MRI Examinations Mean 0.00 5.89 0.60 4.46 22.95 0.20 106.19 80.90 0.30
Utilization Average Costs CT Scans Ultrasound Studies MRI Examination CT Scan Ultrasound Study Mean Mean Mean Mean Mean 0.00 0.00 0.00 0.00 0.00 87.27 84.65 19.42 49.14 14.88 0.48 0.51 0.63 0.48 0.42 208.71 187.15 54.48 84.76 14.96 204.80 139.95 41.37 88.16 9.60 0.97 0.53 0.62 0.94 0.34 172.86 535.65 7.55 11.55 1.71 494.38 368.67 54.36 93.83 4.52 0.01 0.11 0.06 0.02 0.05
7.05 13.32 0.67 1.37 37.07 0.14 106.41 57.96 0.03
0.77 183.50 0.20 285.05 313.92 0.85 172.90 276.26 0.32
3.59 98.76 0.28 247.50 214.24 0.79 537.75 274.87 0.00
3.56 19.24 0.53 50.99 37.70 0.65 14.47 58.15 0.09
112.92 57.76 0.37 27.67 90.14 0.27 18.94 82.18 0.03
32.35 13.60 0.08 0.86 11.22 0.12 1.99 4.24 0.28
0.00 13.95 0.37 120.89 33.08 0.00 73.27 61.79 0.60
491.23 204.47 0.07 1.50 283.60 0.10 159.39 285.93 0.21
402.23 156.82 0.06 776.25 161.81 0.00 330.90 269.51 0.47
0.00 12.79 0.44 4.34 36.43 0.34 28.14 66.61 0.13
0.03 34.96 0.23 107.58 101.79 0.94 21.54 94.07 0.01
32.04 15.23 0.16 0.50 9.63 0.19 2.04 4.13 0.07
Note: CMI ¼ case mix index; MH ¼ Magnet hospital; NMH ¼ non-Magnet hospital; SI ¼ Saidin index.
lower in comparison with MHs. In CMI quantile 3, MHs had higher utilization of MRI (P < .05) and ultrasound procedures (P < .05), whereas utilization of CT scans and average costs of each procedure were lower in comparison with NMHs, with differences in average costs of CT scans being statistically significant (P < .05). In SI quantile 1, NMHs had higher utilization of MRI examinations, whereas utilization of CT scans and ultrasound procedures was higher at MHs. The average costs of each procedure in SI quantile 1 were lower at MHs in comparison with NMHs, although the differences in the means were not statistically significant. In SI quantile 2, MHs had higher utilization of MRI examinations (P < .05) and ultrasound procedures (P < .05), whereas CT scan utilization was higher at NMHs (P ¼ .10). The average costs of MRI examinations and ultrasound procedures were lower at MHs, whereas average costs of CT scans were higher in 1362
comparison with NMHs. In SI quantile 3, MHs had higher utilization of MRI examinations and ultrasound procedures, whereas utilization of CT scans and average costs of each procedure were lower in comparison with NMHs.
DISCUSSION Our findings show higher rates of utilization of imaging services for MRI, CT, and ultrasound at MHs compared with NMHs over the study period when compared by bed size, severity of illness (CMI), and clinical technological sophistication (SI) quantiles. However, this effect was limited to MHs in the highest quantile of size (number of beds), severity (CMI), and clinical technological sophistication (SI). Structurally, MHs are different from the comparison NMHs, with MHs having more admissions, greater clinical technological sophistication (SI), greater severity of illness among patients treated Journal of the American College of Radiology Volume 12 n Number 12PB n December 2015
(CMI), and higher personnel costs (higher full-time equivalent and nursing care hours). This is consistent with a prior study that found overall costs for hospitals to achieve Magnet status increased [16]. Although the costs of imaging services also rose during the study time period, the average costs per MRI examination, CT scan, and ultrasound procedure were lower at MHs in higher quantiles compared with NMHs in the same quantiles. Overall, the findings show higher utilization of diagnostic imaging services at MHs that are larger in size (number of beds), serve patients with more severe conditions (CMI), and are more technologically sophisticated (SI), although costs per procedure at MHs are lower in comparison with similar NMHs, indicating possible cost efficiency at MHs. Less utilization of imaging services at NMHs seems to result in higher costs per procedure. Could this be because hospitals are trying to recover their investment costs in diagnostic imaging machinery by passing it on to their patients? Or do MHs tend to be more efficient in their practice styles and provision of services? Another interesting question raised in this study is why MHs provide more imaging services compared with similar NMHs. Does the MH practice environment, especially the emphasis on evidence-based care, influence provider practice patterns? If there is a link between excellent nursing care, practice environment and imaging services, could there also be effects on other ancillary hospital services? Further study is needed to better understand the effects of MH designation on the utilization of diagnostic imaging services and other essential hospital services.
Limitations This was an observational study, hence causality cannot be inferred. We focused only on inpatient utilization and costs of diagnostic imaging services and restricted the study sample to hospitals in urban areas within the ten states from 2000 to 2006. The results of this study are not generalizable to the utilization and costs of diagnostic imaging services in general. The small number of MHs in the study sample may have biased the results because of outliers. The link between MHs and diagnostic imaging services has not been previously reported, so there are no prior data for comparison with the findings of this study. TAKE-HOME POINTS -
As of July 2015, there were 418 MHs in the United States, almost all of which were located in urban areas, indicating national interest of urban hospitals in pursuing Magnet credentialing.
-
-
MHs showed higher utilization of imaging services compared with NMHs. This effect was limited to MHs in the highest quantiles of size (number of beds), severity of patients’ illnesses (CMI), and clinical technological sophistication (SI). The average costs of MRI examinations, CT scans, and ultrasound procedures were lower at MHs in comparison with NMHs.
REFERENCES 1. Hendee WR, Becker GJ, Borgstede JP, et al. Addressing overutilization in medical imaging. Radiology 2010;257:240-5. 2. Rothenberg BM, Korn A. The opportunities and challenges posed by the rapid growth of diagnostic imaging. J Am Coll Radiol 2005;2: 407-10. 3. Sistrom C, McKay NL, Weilburg JB, Atlas SJ, Ferris TG. Determinants of diagnostic imaging utilization in primary care. Am J Manage Care 2012;18:e135-44. 4. Lee DW, Duszak R Jr, Hughes DR. Comparative analysis of Medicare spending for medical imaging: sustained dramatic slowdown compared with other services. AJR Am J Roentgenol 2013;201:1277-82. 5. Dodoo M, Duszak R, Hughes DR. Trends in utilization of medical imaging from 2003 to 2011: clinical encounters offer a complementary patient-centered focus. J Am Coll Radiol 2013;10: 507-12. 6. Horny M, Burgess JF Jr, Horwitt J, Cohen AB. Advanced diagnostic imaging in privately insured patients: recent trends in utilization and payments. J Am Coll Radiol 2014;11: 692-7. 7. Kanzaria HK, Probst MA, Ponce NA, Hsia RY. The association between advanced diagnostic imaging and ED length of stay. Am J Emerg Med 2014;32:1253-8. 8. Aiken LH, Smith HL, Lake ET. Lower Medicare mortality among a set of hospitals known for good nursing care. Med Care 1994;32: 771-87. 9. Scott JG, Sochalski J, Aiken LH. Review of Magnet hospital research: findings and implications for professional nursing practice. J Nurs Admin 1999;29:9-19. 10. Havens DS, Johnston MA. Achieving Magnet hospital recognition: chief nurse executives and Magnet coordinators tell their stories. J Nurs Admin 2004;34:579-88. 11. Jayawardhana J, Welton JM, Lindrooth RC. Adoption of National Quality Forum safe practices by Magnet hospitals. J Nurs Admin 2011;41:350-6. 12. Russell J. Journey to magnet: cost vs. benefits. Nurs Econ 2010;28: 340-2. 13. Kramer M, Schmalenberg C. Magnet hospitals: part I. Institutions of excellence. J Nurs Admin 1988;18:13-24. 14. Kramer M, Schmalenberg C. Magnet hospitals: part II. Institutions of excellence. J Nurs Admin 1988;18:11-9. 15. Wagner CM. Is your nursing staff ready for Magnet hospital status? An application of the revised Nursing Work Index. J Nurs Admin 2004;34:463-8. 16. Jayawardhana J, Welton J, Lindrooth RC. Is there a business case for Magnet hospitals? Estimates of the cost and revenue implications of becoming a Magnet. Med Care 2014;52:400-6. 17. Tirole J. The theory of industrial organization. Cambridge, Massachusetts: MIT Press; 2000. 18. Spetz J, Maiuro LS. Measuring levels of technology in hospitals. Q Rev Econ Fin 2004;44:430-47.
Journal of the American College of Radiology Jayawardhana, Welton n Imaging Services in Magnet and Non-Magnet Hospitals
1363
Diagnostic Imaging Services in Magnet and Non-Magnet Hospitals: Trends in Utilization and Costs Jayani Jayawardhana, PhD a , John M. Welton, PhD, RN, FAAN b Abstract Purpose: The purpose of this study was to better understand trends in utilization and costs of diagnostic imaging services at Magnet hospitals (MHs) and non-Magnet hospitals (NMHs). Methods: A data set was created by merging hospital-level data from the American Hospital Association’s annual survey and Medicare cost reports, individual-level inpatient data from the Healthcare Cost and Utilization Project, and Magnet recognition status data from the American Nurses Credentialing Center. A descriptive analysis was conducted to evaluate the trends in utilization and costs of CT, MRI, and ultrasound procedures among MHs and NMHs in urban locations between 2000 and 2006 from the following ten states: Arizona, California, Colorado, Florida, Iowa, Maryland, North Carolina, New Jersey, New York, and Washington. Results: When matched by bed size, severity of illness (case mix index), and clinical technological sophistication (Saidin index) quantiles, MHs in higher quantiles indicated higher rates of utilization of imaging services for MRI, CT, and ultrasound in comparison with NMHs in the same quantiles. However, average costs of MRI, CT, and ultrasounds were lower at MHs in comparison with NMHs in the same quantiles. Conclusions: Overall, MHs that are larger in size (number of beds), serve more severely ill patients (case mix index), and are more technologically sophisticated (Saidin index) show higher utilization of diagnostic imaging services, although costs per procedure at MHs are lower in comparison with similar NMHs, indicating possible cost efficiency at MHs. Further research is necessary to understand the relationship between the utilization of diagnostic imaging services among MHs and its impact on patient outcomes. Key Words: Diagnostic imaging services, Magnet hospitals, utilization, costs J Am Coll Radiol 2015;12:1357-1363. Copyright 2015 American College of Radiology
INTRODUCTION Diagnostic imaging services play a key role in the delivery of health care in the United States and have become essential components in identifying, treating, and managing most major medical conditions and diseases. The increased utilization of diagnostic imaging services has contributed to rising health care costs in the United States, accounting for more than 10% of total health care a Department of Health Policy and Management, College of Public Health, University of Georgia, Athens, Georgia. b College of Nursing, University of Colorado-Anschutz Medical Campus, Denver, Colorado. Corresponding author and reprints: Jayani Jayawardhana, PhD, University of Georgia, College of Public Health, Department of Health Policy and Management, 100 Foster Road, Athens, GA 30602; e-mail: jayaward@ uga.edu. This research was funded by a grant from the Harvey L. Neiman Health Policy Institute. The authors have no conflicts of interest related to the material discussed in this article.
ª 2015 American College of Radiology 1546-1440/15/$36.00 n http://dx.doi.org/10.1016/j.jacr.2015.09.011
costs [1-3]. Although overall utilization of diagnostic imaging services grew rapidly in the early 2000s, recent studies indicate a decline in imaging utilization in recent years [4-7]. However, little is known about utilization trends of diagnostic imaging services at Magnet hospitals (MHs) and non-Magnet hospitals (NMHs) and the trends in associated costs of imaging. MHs are known for their high quality of patient care, excellence in nursing practices, higher retention rates for well-qualified nurses, improved outcomes of care, and greater propensity to use evidence-based care [8-12]. As part of obtaining MH designation, MHs make large investments in resources [13], including high-cost resources such as diagnostic imaging services. Given access to greater resources and capabilities, MHs may be more likely to utilize high-cost services such as diagnostic imaging services in comparison with NMHs. Given the availability of resources at MHs, the question of interest
1357
would be whether MHs are more (or less) likely to utilize high-cost resources such as diagnostic imaging services in comparison with NMHs. To date, 418 hospitals in the United States have been designated as MHs by the American Nurses Credentialing Center (ANCC), which was established by the American Nurses Association in the early 1990s [14]. The concept of MHs originates from an early study in the 1980s that found that a unique set of characteristics was responsible for attracting and retaining highly qualified nurses during a nurse shortage [14,15]. Many studies have examined the association between MH status and various outcomes of quality of care, such as mortality rates, length of stay, patient satisfaction, patient safety practices, and net patient revenue gains [8,11,12,16]. However, little is known about utilization trends for high-cost health care services such as diagnostic imaging services within MHs and NMHs and their impact on costs of care. Because MHs invest heavily in the resources needed in the process of obtaining Magnet status, it is likely that MHs have better access to more advanced technological resources and procedures, such as diagnostic imaging services, than NMHs. Similarly, because of their reputation for providing high-quality patient care and excellence in nursing practices, MHs may also attract patients with severe conditions or complications. Generally, MHs tend to be larger in bed size in comparison with NMHs. Although these reasons could lead to higher utilization of diagnostic imaging services at MHs in comparison with NMHs in general, it would be interesting to investigate if MHs tend to have higher utilization and costs of care of diagnostic imaging services in comparison with NMHs when controlling for technological sophistication, severity of patients’ conditions, and hospital size. In this study, we explored trends in the utilization of diagnostic imaging services and associated costs of such services within MHs and NMHs using inpatient data from acute care hospitals across ten states in the United States from 2000 to 2006. Specifically, we examined trends in utilization and costs of care for three diagnostic imaging services at MHs and NMHs: CT, MRI, and ultrasound.
METHODS The sample data set was created by merging hospital-level data from the American Hospital Association’s (AHA) annual survey and Medicare cost reports (Healthcare Cost Report Information System [HCRIS]), Magnet recognition status data from the ANCC, and individual-level inpatient data from the Healthcare Cost and Utilization Project State Inpatient Databases (HCUP-SID). The AHA annual survey 1358
data include information on hospital characteristics such as bed size, location, ownership, personnel information, and types of services provided. HCRIS data include information on costs of care at hospitals. The ANCC lists the hospitals that received Magnet recognition on its website, along with their addresses and the years they received Magnet status, as well as the follow-up years when they were recredentialed. The information from the ANCC website was collected in 2011 and was used to identify the MHs and the years they received Magnet status in the merged data set. The HCUPSID data include information on individual inpatient utilization of health care services, including diagnostic imaging services. The AHA annual survey, HCRIS, and HCUP-SID data are widely used in many research studies. The utilization of individual imaging services in HCUPSID data was identified using International Classification of Diseases, radiologic procedure codes and were aggregated to the hospital-year level. These data were available for individual patient discharges. Cost data identifiable in the HCRIS database included MRI, CT, and ultrasound. We restricted the data sample only to urban hospitals because almost all MHs were located in urban areas. The final data sample included data from Arizona, California, Colorado, Florida, Iowa, Maryland, North Carolina, New Jersey, New York, and Washington from 2000 to 2006 because these are the only states and years for which data were accessible. These states are regionally diverse and include both MHs and NMHs. The total study sample included 178 MH observations and 5,030 NMH observations, a total of 5,208 observations. The unit of analysis was the hospital-year. It should be noted that we did not have a way to identify a hospital that may have had Magnet status during the study period (2000-2006) but lost its Magnet status by 2011 because the ANCC does not disclose this information. Using the study sample, we examined trends in utilization and costs of three diagnostic imaging services, MRI, CT, and ultrasound, within MHs and NMHs from 2000 to 2006. We conducted statistical significance tests, using c2 tests for categorical variables and Student’s t test for continuous variables, to investigate whether there were significant differences between the means of utilization and costs of imaging services at MHs and NMHs within each year. Similar significance tests were also carried out for other key hospital characteristics, such as ownership (1 ¼ nonprofit, 0 ¼ otherwise), percentage of Medicare and Medicaid discharges, total inpatient discharges, number of beds, membership in a system of hospitals (1 ¼ system, 0 ¼ otherwise), the degree of market concentration in the hospital’s health care referral region (Herfindahl-Hirschman index) [17], the hospital’s clinical Journal of the American College of Radiology Volume 12 n Number 12PB n December 2015
technological sophistication (Saidin Index [SI]) [18], and a hospital case mix index (CMI) based on diagnosis-related group weights to adjust for the clinical complexity of patients treated at the hospital. Because MHs could be larger and more technologically sophisticated and thus might attract severe patients in comparison with NMHs, we also performed tests to examine whether there were statistically significant differences between the means of utilization and costs of imaging services at MHs and NMHs when the hospitals were matched on the basis of bed size, severity of patients’ conditions (CMI), and technological sophistication (SI). The analyses were performed using Stata version 13.1 for Windows (StataCorp LP, College Station, Texas).
RESULTS Table 1 shows the descriptive data by Magnet status of hospitals for each year from 2000 to 2006. To decrease the complexity of displaying the findings, statistical significance is summarized over multiple years. MHs are larger (in bed size, numbers of admissions, and discharges; P < .01), have higher nursing and full-time equivalent intensity (P < .05), have a higher percentage of registered nurses (P < .05), are more likely to be teaching hospitals (P < .05), are more technologically sophisticated (SI; P < .01), and treat more severe patients (CMI; P < .01) in comparison with NMHs. MHs are also more likely to be in more competitive markets (Herfindahl-Hirschman index), have longer lengths of stay, have lower shares of Medicaid and Medicare patients, and be nonprofit in comparison with NMHs, although differences between the means of these characteristics were not statistically significant. Interestingly, average costs of MRI examinations, CT scans, and ultrasound procedures were higher at NMHs in comparison with MHs in most years, although the differences in the means were not statistically significant. Although the utilization of MRI examinations, CT scans, and ultrasound procedures was mostly higher at NMHs in early years (2000-2001), utilization at MHs mostly increased between 2003 and 2005 but dropped again in 2006. Figure 1 presents utilization trends in MRI examinations, CT scans, and ultrasound procedures by Magnet status across years. Although the average number of MRI examinations performed per thousand admissions at MHs increased from 2000 to 2003, it decreased between 2003 and 2006. Except for 2000 and 2006, MHs had a higher utilization rate of MRI examinations each year in comparison with NMHs. Although MHs did not report utilization of CT scans during 2000 and 2001, their average utilization of CT scans per thousand admissions increased between 2002 and 2005 but dropped again to almost zero by 2006.
Interestingly, the utilization of CT scans at NMHs was higher in comparison with MHs each year, although NMHs’ CT scan utilization was decreasing over the years. Similarly, the average number of ultrasound procedures performed per thousand admissions at NMHs was highest in 2000 and gradually decreased each year until 2006. Furthermore, the utilization of ultrasound procedures was higher at NMHs in comparison with MHs in 2000, 2001, and 2006. The utilization of ultrasound procedures by MHs increased from 2000 to 2005 but dropped again in 2006. Figure 2 presents trends in average costs of each procedure by Magnet status across years. Average costs of MRI examinations were higher at MHs in comparison with NMHs each year except for 2000 and 2006. Although average costs of MRI examinations at MHs increased between 2000 and 2003, they were on the decline between 2004 and 2006. However, average costs of MRI examinations at NMHs were increasing over the years. Because there was no report of utilization of CT scans by MHs in 2000 and 2001, average costs of CT scans would be zero. Similar to utilization, average costs of CT scans at MHs increased from 2002 to 2005 but dropped again in 2006. Interestingly, average costs of CT scans at NMHs increased over the years and were also higher in comparison with MHs each year except for 2004 and 2005. Although the average costs of ultrasound procedures at NMHs increased over the years, the costs were higher at MHs in comparison with NMHs each year except for 2000 and 2001. Table 2 presents the means of utilization and average costs of diagnostic imaging services across MHs and NMHs when the hospitals were matched on the basis of bed size, severity of patients’ conditions (CMI), and technological sophistication (SI) by quantile across all years. Bed size, CMI, and SI were divided into three quantiles, each with quantile 1 indicating smaller hospitals, less severe patient conditions, or less technologically sophisticated hospitals and quantile 3 indicating larger hospitals, more severe patient conditions, or more technologically sophisticated hospitals. Whereas NMHs in bed quantile 1 used diagnostic imaging services, MHs in bed quantile 1 did not use any of the procedures. In bed quantile 2, NMHs utilized more MRI examinations, whereas MHs utilized more CT scans and ultrasound procedures. Average costs of MRI examinations and ultrasound procedures were higher at MHs in bed quantile 2, whereas average costs of CT scans were higher at NMHs. However, differences between the means were not significant either in utilization or in average costs within bed quantiles 1 and 2. This is likely due to the small sample size of MHs in those respective categories. In bed quantile 3, MHs had higher utilization of
Journal of the American College of Radiology Jayawardhana, Welton n Imaging Services in Magnet and Non-Magnet Hospitals
1359
1360 Table 1. Descriptive statistics of the study sample
Journal of the American College of Radiology Volume 12 n Number 12PB n December 2015
Variable Number of MRI examinations Number of CT scans Number of ultrasound studies AC of MRI AC of CT AC of ultrasound Hospital characteristics Beds Admissions (1,000) Discharges (1,000) Nursing intensity Percentage RNs FTE intensity RN/LPN ratio Nonprofit HHI Medicaid share Medicare share System (1) Teaching (1) SI CMI LOS n
NMH 35.05
2000 2001 2002 MH P NMH MH P NMH MH 18.88 0.77 35.48 65.10 0.55 36.51 126.11
294.68 0.00 0.33 257.90 257.36 112.63 0.62 214.53
27.30 39.92 6.52
0.81 0.61 27.76 0.00 0.50 47.01 0.14 0.50 6.53
2003 P NMH MH P 0.02 34.84 85.41 0.11
NMH 37.72
2004 2005 MH P NMH MH P 57.33 0.48 36.63 74.31 0.14
2006 NMH MH P 32.87 53.93 0.45
0.00 0.33 263.72 206.11 0.78 255.54 169.81 0.63 243.48 226.15 0.91 263.57 288.79 0.87 204.49 2.52 0.29 91.30 0.61 204.20 489.26 0.11 185.62 390.56 0.15 169.21 544.59 0.00 162.89 599.21 0.00 145.04 47.52 0.42
10.69 0.70 0.00 0.45 0.13 0.38
32.04 65.39 9.34
10.09 0.53 38.99 12.00 0.36 79.73 0.90 0.30 10.37
13.66 0.58 22.04 0.27 2.44 0.32
42.27 38.74 0.91 46.96 91.63 22.09 0.22 113.52 11.00 19.47 0.26 13.99
24.38 0.57 65.95 0.48 1.49 0.12
59.03 13.66 0.33 117.75 34.07 0.33 12.14 2.35 0.24
249.42 492.75 0.00 253.46 448.60 0.00 248.27 425.00 0.00 246.60 383.07 0.00 249.00 376.77 0.00 252.25 396.25 0.00 260.81 391.41 0.00 11.40 29.35 0.00 11.61 25.70 0.00 11.92 22.03 0.00 11.99 19.22 0.00 12.22 19.06 0.00 12.36 19.73 0.00 12.53 19.23 0.00 10.73
26.00 0.00
11.07
23.55 0.00
11.42
20.44 0.00
11.49
19.62 0.00
11.69
19.01 0.00
11.88
19.84 0.00
12.06 19.26 0.00
7.52 10.57 0.00 7.37 9.75 0.00 7.49 9.03 0.01 7.60 9.19 0.00 7.73 8.80 0.01 7.87 9.06 0.00 8.29 9.32 0.05 0.88 0.97 0.00 0.88 0.96 0.00 0.89 0.92 0.02 0.89 0.93 0.01 0.90 0.94 0.00 0.90 0.95 0.00 0.91 0.95 0.00 26.24 34.61 0.00 26.38 33.67 0.01 26.56 31.44 0.01 26.09 32.20 0.00 26.21 30.08 0.01 26.42 30.89 0.00 24.66 30.01 0.00 14.99 95.39 0.00 15.17 122.08 0.00 16.09 28.80 0.10 19.08 24.24 0.64 21.19 25.83 0.53 22.92 31.48 0.27 25.01 38.50 0.24 0.66 0.88 0.20 0.67 0.80 0.37 0.67 0.84 0.11 0.66 0.81 0.10 0.66 0.85 0.02 0.66 0.83 0.01 0.59 0.74 0.11 0.12 0.18 0.12 0.12 0.19 0.05 0.12 0.13 0.60 0.12 0.16 0.06 0.12 0.15 0.05 0.12 0.14 0.24 0.12 0.13 0.67 0.15 0.10 0.32 0.15 0.13 0.63 0.15 0.10 0.09 0.16 0.11 0.11 0.16 0.11 0.03 0.15 0.11 0.02 0.17 0.13 0.10 0.32 0.25 0.12 0.33 0.29 0.34 0.34 0.32 0.54 0.34 0.37 0.44 0.35 0.36 0.55 0.35 0.36 0.66 0.32 0.31 0.60 0.61 0.25 0.04 0.61 0.40 0.17 0.62 0.47 0.19 0.64 0.67 0.77 0.63 0.64 0.90 0.63 0.63 0.89 0.69 0.63 0.52 0.09 0.75 0.00 0.09 0.60 0.00 0.09 0.37 0.00 0.09 0.37 0.00 0.09 0.33 0.00 0.09 0.27 0.00 0.08 0.19 0.07 1.77 4.25 0.00 1.87 4.01 0.00 1.97 3.34 0.00 1.96 3.11 0.00 2.02 3.16 0.00 1.99 3.17 0.00 2.04 3.38 0.00 1.41 1.75 0.00 1.41 1.67 0.00 1.40 1.61 0.00 1.38 1.58 0.00 1.39 1.56 0.00 1.39 1.59 0.00 1.43 1.63 0.00 4.70 5.01 0.49 4.77 5.15 0.54 4.78 4.90 0.74 4.73 5.10 0.22 4.72 5.04 0.26 4.65 5.10 0.01 4.48 4.63 0.45 779 8 798 10 749 19 754 27 740 39 700 48 510 27
Note: These represent the total number of observations available for MHs and NMHs for each year in the study sample. AC ¼ average cost; CMI ¼ case mix index; FTE ¼ full-time equivalent; HHI ¼ Herfindahl-Hirschman index; LOS ¼ length of stay; LPN ¼ licensed practical nurse; MH ¼ Magnet hospital; NMH ¼ non-Magnet hospital; RN ¼ registered nurse; SI ¼ Saidin index.
Average number of CT scans per thousand Admissions
by year and Magnet status of Hospitals
by year and Magnet status of Hospitals
0
0
5
1
2
10
3
15
4
20
5
25
Average number of MRI exams per thousand Admissions
2000
2001
2002
2003
Magnet
2004
2005
2006
2000
2001
2002
Non-Magnet
2003
Magnet
2004
2005
2006
Non-Magnet
Average number of Ultrasounds per thousand Admissions
0
10
20
30
by year and Magnet status of Hospitals
2000
2001
2002
2003
2004
Magnet
2005
2006
Non-Magnet
Fig 1. Utilization of MRI examinations, CT scans, and ultrasound procedures by Magnet and non-Magnet hospitals.
MRI examinations and ultrasound procedures, whereas NMHs had higher utilization of CT scans (P < .05). In bed quantile 3, average costs of all three procedures were higher (P < .10) at NMHs in comparison with MHs. In CMI quantile 1, NMHs had higher utilization of each of the procedures and higher average costs for MRI
examinations in comparison with MHs, whereas costs of ultrasound procedures (P < .10) and CT scans were higher at MHs. In CMI quantile 2, NMHs had higher utilization of MRI examinations and CT scans and higher average costs for CT scans and ultrasound procedures, although their utilization of ultrasound procedures was Average Costs of CT scans by year and Magnet status of Hospitals
2000
2001
2002
2003
Magnet
2004
2005
0
0
200
200
400
400
600
600
Average Costs of MRIs by year and Magnet status of Hospitals
2006
2000
2001
Non-Magnet
2002 Magnet
2003
2004
2005
2006
Non-Magnet
Average Costs of Ultrasounds
0
100
200
300
400
500
by year and Magnet status of Hospitals
2000
2001
2002 Magnet
2003
2004
2005
2006
Non-Magnet
Fig 2. Average costs of MRI examinations, CT scans, and ultrasound procedures by Magnet and non-Magnet hospitals.
Journal of the American College of Radiology Jayawardhana, Welton n Imaging Services in Magnet and Non-Magnet Hospitals
1361
Table 2. Mean utilization and average costs of imaging services at MHs and NMHs matched on each hospital characteristic quantile of bed size, severity of patients’ illnesses (CMI), and technological sophistication (SI)
Bed quantile Hospital 1 MHs NMHs P 2 MHs NMHs P 3 MHs NMHs P CMI quantile 1 MHs NMHs P 2 MHs NMHs P 3 MHs NMHs P SI quantile 1 MHs NMHs P 2 MHs NMHs P 3 MHs NMHs P
n 8 1,732 52 1,682 118 1,616
22 1,714 38 1,698 118 1,618
22 1,715 28 1,714 128 1,601
MRI Examinations Mean 0.00 5.89 0.60 4.46 22.95 0.20 106.19 80.90 0.30
Utilization Average Costs CT Scans Ultrasound Studies MRI Examination CT Scan Ultrasound Study Mean Mean Mean Mean Mean 0.00 0.00 0.00 0.00 0.00 87.27 84.65 19.42 49.14 14.88 0.48 0.51 0.63 0.48 0.42 208.71 187.15 54.48 84.76 14.96 204.80 139.95 41.37 88.16 9.60 0.97 0.53 0.62 0.94 0.34 172.86 535.65 7.55 11.55 1.71 494.38 368.67 54.36 93.83 4.52 0.01 0.11 0.06 0.02 0.05
7.05 13.32 0.67 1.37 37.07 0.14 106.41 57.96 0.03
0.77 183.50 0.20 285.05 313.92 0.85 172.90 276.26 0.32
3.59 98.76 0.28 247.50 214.24 0.79 537.75 274.87 0.00
3.56 19.24 0.53 50.99 37.70 0.65 14.47 58.15 0.09
112.92 57.76 0.37 27.67 90.14 0.27 18.94 82.18 0.03
32.35 13.60 0.08 0.86 11.22 0.12 1.99 4.24 0.28
0.00 13.95 0.37 120.89 33.08 0.00 73.27 61.79 0.60
491.23 204.47 0.07 1.50 283.60 0.10 159.39 285.93 0.21
402.23 156.82 0.06 776.25 161.81 0.00 330.90 269.51 0.47
0.00 12.79 0.44 4.34 36.43 0.34 28.14 66.61 0.13
0.03 34.96 0.23 107.58 101.79 0.94 21.54 94.07 0.01
32.04 15.23 0.16 0.50 9.63 0.19 2.04 4.13 0.07
Note: CMI ¼ case mix index; MH ¼ Magnet hospital; NMH ¼ non-Magnet hospital; SI ¼ Saidin index.
lower in comparison with MHs. In CMI quantile 3, MHs had higher utilization of MRI (P < .05) and ultrasound procedures (P < .05), whereas utilization of CT scans and average costs of each procedure were lower in comparison with NMHs, with differences in average costs of CT scans being statistically significant (P < .05). In SI quantile 1, NMHs had higher utilization of MRI examinations, whereas utilization of CT scans and ultrasound procedures was higher at MHs. The average costs of each procedure in SI quantile 1 were lower at MHs in comparison with NMHs, although the differences in the means were not statistically significant. In SI quantile 2, MHs had higher utilization of MRI examinations (P < .05) and ultrasound procedures (P < .05), whereas CT scan utilization was higher at NMHs (P ¼ .10). The average costs of MRI examinations and ultrasound procedures were lower at MHs, whereas average costs of CT scans were higher in 1362
comparison with NMHs. In SI quantile 3, MHs had higher utilization of MRI examinations and ultrasound procedures, whereas utilization of CT scans and average costs of each procedure were lower in comparison with NMHs.
DISCUSSION Our findings show higher rates of utilization of imaging services for MRI, CT, and ultrasound at MHs compared with NMHs over the study period when compared by bed size, severity of illness (CMI), and clinical technological sophistication (SI) quantiles. However, this effect was limited to MHs in the highest quantile of size (number of beds), severity (CMI), and clinical technological sophistication (SI). Structurally, MHs are different from the comparison NMHs, with MHs having more admissions, greater clinical technological sophistication (SI), greater severity of illness among patients treated Journal of the American College of Radiology Volume 12 n Number 12PB n December 2015
(CMI), and higher personnel costs (higher full-time equivalent and nursing care hours). This is consistent with a prior study that found overall costs for hospitals to achieve Magnet status increased [16]. Although the costs of imaging services also rose during the study time period, the average costs per MRI examination, CT scan, and ultrasound procedure were lower at MHs in higher quantiles compared with NMHs in the same quantiles. Overall, the findings show higher utilization of diagnostic imaging services at MHs that are larger in size (number of beds), serve patients with more severe conditions (CMI), and are more technologically sophisticated (SI), although costs per procedure at MHs are lower in comparison with similar NMHs, indicating possible cost efficiency at MHs. Less utilization of imaging services at NMHs seems to result in higher costs per procedure. Could this be because hospitals are trying to recover their investment costs in diagnostic imaging machinery by passing it on to their patients? Or do MHs tend to be more efficient in their practice styles and provision of services? Another interesting question raised in this study is why MHs provide more imaging services compared with similar NMHs. Does the MH practice environment, especially the emphasis on evidence-based care, influence provider practice patterns? If there is a link between excellent nursing care, practice environment and imaging services, could there also be effects on other ancillary hospital services? Further study is needed to better understand the effects of MH designation on the utilization of diagnostic imaging services and other essential hospital services.
Limitations This was an observational study, hence causality cannot be inferred. We focused only on inpatient utilization and costs of diagnostic imaging services and restricted the study sample to hospitals in urban areas within the ten states from 2000 to 2006. The results of this study are not generalizable to the utilization and costs of diagnostic imaging services in general. The small number of MHs in the study sample may have biased the results because of outliers. The link between MHs and diagnostic imaging services has not been previously reported, so there are no prior data for comparison with the findings of this study. TAKE-HOME POINTS -
As of July 2015, there were 418 MHs in the United States, almost all of which were located in urban areas, indicating national interest of urban hospitals in pursuing Magnet credentialing.
-
-
MHs showed higher utilization of imaging services compared with NMHs. This effect was limited to MHs in the highest quantiles of size (number of beds), severity of patients’ illnesses (CMI), and clinical technological sophistication (SI). The average costs of MRI examinations, CT scans, and ultrasound procedures were lower at MHs in comparison with NMHs.
REFERENCES 1. Hendee WR, Becker GJ, Borgstede JP, et al. Addressing overutilization in medical imaging. Radiology 2010;257:240-5. 2. Rothenberg BM, Korn A. The opportunities and challenges posed by the rapid growth of diagnostic imaging. J Am Coll Radiol 2005;2: 407-10. 3. Sistrom C, McKay NL, Weilburg JB, Atlas SJ, Ferris TG. Determinants of diagnostic imaging utilization in primary care. Am J Manage Care 2012;18:e135-44. 4. Lee DW, Duszak R Jr, Hughes DR. Comparative analysis of Medicare spending for medical imaging: sustained dramatic slowdown compared with other services. AJR Am J Roentgenol 2013;201:1277-82. 5. Dodoo M, Duszak R, Hughes DR. Trends in utilization of medical imaging from 2003 to 2011: clinical encounters offer a complementary patient-centered focus. J Am Coll Radiol 2013;10: 507-12. 6. Horny M, Burgess JF Jr, Horwitt J, Cohen AB. Advanced diagnostic imaging in privately insured patients: recent trends in utilization and payments. J Am Coll Radiol 2014;11: 692-7. 7. Kanzaria HK, Probst MA, Ponce NA, Hsia RY. The association between advanced diagnostic imaging and ED length of stay. Am J Emerg Med 2014;32:1253-8. 8. Aiken LH, Smith HL, Lake ET. Lower Medicare mortality among a set of hospitals known for good nursing care. Med Care 1994;32: 771-87. 9. Scott JG, Sochalski J, Aiken LH. Review of Magnet hospital research: findings and implications for professional nursing practice. J Nurs Admin 1999;29:9-19. 10. Havens DS, Johnston MA. Achieving Magnet hospital recognition: chief nurse executives and Magnet coordinators tell their stories. J Nurs Admin 2004;34:579-88. 11. Jayawardhana J, Welton JM, Lindrooth RC. Adoption of National Quality Forum safe practices by Magnet hospitals. J Nurs Admin 2011;41:350-6. 12. Russell J. Journey to magnet: cost vs. benefits. Nurs Econ 2010;28: 340-2. 13. Kramer M, Schmalenberg C. Magnet hospitals: part I. Institutions of excellence. J Nurs Admin 1988;18:13-24. 14. Kramer M, Schmalenberg C. Magnet hospitals: part II. Institutions of excellence. J Nurs Admin 1988;18:11-9. 15. Wagner CM. Is your nursing staff ready for Magnet hospital status? An application of the revised Nursing Work Index. J Nurs Admin 2004;34:463-8. 16. Jayawardhana J, Welton J, Lindrooth RC. Is there a business case for Magnet hospitals? Estimates of the cost and revenue implications of becoming a Magnet. Med Care 2014;52:400-6. 17. Tirole J. The theory of industrial organization. Cambridge, Massachusetts: MIT Press; 2000. 18. Spetz J, Maiuro LS. Measuring levels of technology in hospitals. Q Rev Econ Fin 2004;44:430-47.
Journal of the American College of Radiology Jayawardhana, Welton n Imaging Services in Magnet and Non-Magnet Hospitals
1363