Journal of the Neurological Sciences 353 (2015) 137–142
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Positive correlation between care given by specialists and registered nurses and improved outcomes for stroke patients Kyu-Tae Han a,b, Sun Jung Kim a,c, Sung-In Jang a,b,d, Seung Ju Kim a,b, Seo Yoon Lee b,e, Hyo Jung Lee b,e, Eun-Cheol Park b,d,⁎ a
Department of Public Health, Graduate School, Yonsei University, Seoul, Republic of Korea Institute of Health Services Research, Yonsei University College of Medicine, Seoul, Republic of Korea Department of Health Administration, Namseoul University, Cheonan, Republic of Korea d Department of Preventive Medicine, Yonsei University College of Medicine, Seoul, Republic of Korea e Department of Health Policy and Management, Graduate School of Public Health, Yonsei University, Seoul, Republic of Korea b c
a r t i c l e
i n f o
Article history: Received 2 December 2014 Received in revised form 12 March 2015 Accepted 17 April 2015 Available online 29 April 2015 Keywords: Stroke Quality of care Readmission Death Specialist Registered nurse
a b s t r a c t Background: Cerebrovascular diseases are the second-highest cause of death in South Korea (9.6% of all causes of mortality in 2013). South Korea has a shortage of trained medical personnel compared with other countries and the demands for health care are continuously increasing. Our study sought to determine the relationship between hospital human resources and the outcomes of stroke patients. Methods: We used data from NHI claims (n = 99,464) at 120 hospitals to analyze readmission or death within 30 days after discharge or hospitalization for stroke patients during 2010–2013. We used multilevel models that included both patient-level and hospital-level variables to examine factors associated with readmission or death within 30 days. Results: A total of 1782 (1.8%) patients were readmitted within 30 days, and death occurred within 30 days for 6926 (7.0%) patients. Patients cared for by a higher percentages of specialists or registered nurses had a lower risk of readmission or death within 30 days (readmission per 10% increase in registered nurses, OR = 0.89 and SD = 0.85–0.94; death per 10% increase in specialists, OR = 0.93 and SD = 0.89–0.98). Conclusions: The percentages of specialist and registered nurses caring for stroke patients were positively correlated with better patient outcomes, particularly for patients with cerebral infarction. © 2015 Elsevier B.V. All rights reserved.
1. Introduction The health status and longevity of people in South Korea improved rapidly during the latter 20th century due to the introduction of National Health Insurance (NHI) and the development of health technologies [1]. This led to increases in the elderly population, the emergence of new health problems associated with geriatrics (individuals aged greater than 65 years, who represented 12.2% of the total population in 2013), and increases in morbidity and mortality among the aged [2]. These health problems include chronic diseases and deterioration in mental health such as suicide among the elderly. Many health-care specialists are gaining expertise in working with and treating geriatric diseases.
Abbreviations: OECD, Organization for Economic Cooperation and Development; NHI, National Health Insurance; RNs, registered nurses; LPNs, licensed practical nurses; ICD, International Classification of Diseases; CCI, Charlson Comorbidity Index; ANOVA, analysis of variance; OR, odds ratio. ⁎ Corresponding author at: Department of Preventive Medicine & Institute of Health Services Research, College of Medicine, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-752, Republic of Korea. Tel.: +82 2 2228 1862; fax: +82 2 392 7734. E-mail address: [email protected] (E.-C. Park).
http://dx.doi.org/10.1016/j.jns.2015.04.034 0022-510X/© 2015 Elsevier B.V. All rights reserved.
Chronic diseases remain as the primary cause of mortality among elderly populations in South Korea [3]. Cerebrovascular diseases are one of the primary causes of mortality in South Korea. National statistics in Korea indicate that cerebrovascular diseases are the second-highest cause of death (9.6% of all causes of mortality in 2013) [4]. The burden of cerebrovascular disease is expected to continuously increase in the immediate future because the major risk factor for stroke is advanced age [5,6]. Therefore, it is important to identify factors that can reduce or prevent mortality due to stroke. Here, we focus on the role of human resources and medical personnel during stroke patient hospitalization. South Korea is recently experiencing a shortage in trained medical personnel, including specialist doctors and registered nurses. However, the demand for health care has increased during this time. The numbers of trained medical personnel in South Korea is lower than those in other countries within the Organization for Economic Cooperation and Development (OECD). In South Korea, there are 2.0 practicing doctors and 4.8 practicing nurses per 1000 people, whereas in the OECD there are an average of 3.2 practicing doctors and 8.8 practicing nurses per 1000 people [7]. Many studies have investigated the role of professional human resources in hospitals [8–10]. These studies indicate that the
138
K.-T. Han et al. / Journal of the Neurological Sciences 353 (2015) 137–142
quality of care and patient outcomes were significantly improved when the ratio of medical personnel to patients was larger, and when the medical personnel had specialized training [11–15]. The shortage of health-care human resources in South Korea is caused by practical constraints and limited resources, and it will be challenging to solve these problems in the immediate future. Therefore, our study investigated the relationship between the numbers of specialist doctors and registered nurses (RNs) in hospitals and stroke patient outcomes. We performed sub-group analysis to identify differences based on patient diagnoses. 2. Methods 2.1. Study population There are about 1730 hospitals including 39 public hospitals during 2010–2013 year in South Korea, but, the data we used in this study was only included 156 hospitals (117 private and 39 public) after extracting through propensity score matching-methods (1:3) adjusting some variables as follows: region of hospitals, nursing staffing level, number of total beds, number of intensive care unit beds, number of emergency room beds, and number of doctors. And then, this study only included hospitalization due to stroke. Stroke was classified according to the International Classification of Diseases (ICD-10; including I61, I62, and I63). We excluded the hospitals without stroke inpatient cases (N = 36). Finally, 120 hospitals (public = 32 vs private = 88, 99,464 hospitalization cases) were included for analysis. The unit of analysis was one hospitalization case. 2.2. Variables To reflect the quality of care and outcomes in stroke inpatient, we used readmission within 30 days after discharge for stroke and death within 30 days after hospitalization for stroke as outcome variables in this study. We identified the patient's first discharge or hospitalization in the calendar year as the first index discharge or hospitalization. Readmission or death within 30 days was defined as readmission or death within 30 calendar days based on the first index discharge or hospitalization. The primary variables of interest with respect to readmission or death within 30 days after discharge or hospitalization were the percentages of specialist doctors and RNs. These were defined as the percentage of specialist doctors to the total number of doctors, and the percentage of RNs to the total number of nurses [including RNs and licensed practical nurses (LPNs)] in each hospital. These indicators were used to reflect the richer skill mix in human resource pool, and were calculated as shown in the following two equations: Percentage of specialist doctors ¼ ðNumber of specialist doctors=Total number of doctorsÞ 100 Percentage of RNs ¼ðNumber of RNs=Total number of nursesÞ 100: We adjusted these correlation analyses for patient-level and hospital-level variables. Patient-level variables included major diagnosis, age, gender, Charlson Comorbidity Index (CCI) score, health insurance type, hospitalization year, and length of hospitalization. Major diagnoses were classified according to the following ICD groupings: I61–I62, intracerebral hemorrhage; and I63, cerebral infarction. These classifications reflected specific pathological mechanisms. The CCI was used to account for the effect of comorbid disorders or diseases. Hospital-level variables included structural characteristics, human resources, and the number of stroke patients in each hospital. The structural characteristic variables included ownership status, teaching status, and number of beds. The human resource variables included the number
of doctors and nurses per bed, and the number of neurosurgeons and neurologists. 2.3. Statistical analysis We analyzed the distribution of each categorical variable by examining the frequencies and percentages of the variables and performing χ2 tests to identify correlations with patient readmission or death within 30 days. These analyses were performed for both patient-level and hospital-level variables. We also performed analysis of variance (ANOVA) to compare the average values and standard deviations for continuous variables. Then, we used multilevel models including both patient-level and hospital-level variables to identify correlations with patient readmission or death within 30 days after discharge or hospitalization, respectively. We also performed sub-group analyses with respect to major diagnoses for stroke patients. All statistical analyses were performed using SAS statistical software version 9.2. All calculated P-values were two-sided and considered significant at P b 0.05. 3. Results The data used in this study included 99,464 hospitalization cases. The number of patients readmitted within 30 days after discharge was 1782 (1.8%), and the number of mortalities within 30 days after hospitalization was 6926 (7.0%). Table 1 presents the results of univariate associations between each variable and readmission or death within 30 days after discharge or hospitalization, respectively, for stroke. The numbers of readmissions and deaths within 30 days after discharge or hospitalization were greater for patients with major diagnoses of intracerebral hemorrhage (2.3% and 17.1%, respectively) than cerebral infarction (1.7% and 4.5%, respectively). Analysis of CCI scores showed that the higher the score, the higher the percentages of readmission within 30 days. Mortality due to stroke was inversely correlated to the time since hospitalization (decreasing over the years of analysis), whereas readmission due to stroke was directly correlated with time (increasing during the years of analysis). These correlations also were observed when analyzing the length of hospitalization. The analyses of hospital characteristics revealed significant differences in readmissions or deaths within 30 days for teaching hospitals (1.7% and 6.6%, respectively) and non-teaching hospitals (2.0% and 7.6%, respectively). The average percentage of RNs was lower for cases that were readmitted or perished within 30 days after discharge or hospitalization. The average numbers of neurosurgeons, neurologists, beds, and stroke patient admittance were lower in patients that were readmitted or perished within 30 days after discharge or hospitalization for stroke. Table 2 presents the results of multilevel analyses considering both patient-level and hospital-level variables for readmission or death within 30 days after discharge or hospitalization for stroke. Patients hospitalized for intracerebral hemorrhage had a higher risk of readmission or death within 30 days [for readmission, odds ratio (OR) = 1.20 and SD = 1.04–1.39; for death, OR = 8.68 and SD = 8.07–9.34] than patients hospitalized for cerebral infarction (this group was used as the reference). Multilevel analyses of age groups indicated that older patients had a higher risk of death within 30 days after hospitalization, and a lower risk of readmission within 30 days. Analyses of CCI scores showed that higher scores were associated with greater risk for readmission or death within 30 days after discharge or hospitalization. Multilevel analyses of hospital-level variables show that patients at non-teaching hospitals have lower trend for readmission or death within 30 days after discharge or hospitalization, but this was not statistically significant. Patients at private hospitals had lower risk for death within 30 days (OR = 0.82 and SD = 0.75–0.91; teaching hospitals were used as the reference). Patients at hospitals with a higher percentage of specialist doctors (for each 10% more specialists, OR = 0.93 and SD = 0.89–0.98) or
K.-T. Han et al. / Journal of the Neurological Sciences 353 (2015) 137–142
139
Table 1 Univariate correlations between readmission and death within 30 days after discharge or hospitalization. Variables
Readmission within 30 days (n = 99,464) Yes
Patient-level Major diagnoses Intracerebral hemorrhage Cerebral infarction Age (years) b64 65+ Gender Male Female Charlson Comorbidity Index 0 1 2 3+ Type of health insurance NHI Medical-Aid Hospitalization year 2010 2011 2012 2013 Length of stay Hospital-level Teaching status Teaching hospital (n = 43) Non-teaching hospital (n = 77) Ownership Public (n = 32) Private (n = 88) Percentage of specialists Percentage of RNs Number of total doctors per bed Number of total nurses per bed Number of neurosurgeons Number of neurologists Number of beds Stroke patient admittance Total
Death within 30 days (n = 99,464)
No
P-value
N (mean)
% (SD)
N (mean)
% (SD)
445 1337
2.3 1.7
19,159 78,523
97.7 98.3
705 1077
2.0 1.7
35,266 62,416
976 806
1.8 1.8
53 147 246 1336
Dead
Alive
P-value
N (mean)
% (SD)
N (mean)
% (SD)
b.0001
3358 3568
17.1 4.5
16,246 76,292
82.9 95.5
b.0001
98.0 98.3
0.0026
1734 5192
4.8 8.2
34,237 58,301
95.2 91.8
b.0001
53,610 44,072
98.2 98.2
0.9248
3439 3487
6.3 7.8
51,147 41,391
93.7 92.2
b.0001
1.4 1.7 1.8 1.8
3864 8489 13,249 72,080
98.6 98.3 98.2 98.2
0.1661
186 448 636 5656
4.7 5.2 4.7 7.7
3731 8188 12,859 67,760
95.3 94.8 95.3 92.3
b.0001
1498 284
10.3 0.3
13,093 84,589
89.7 99.7
0.0019
5913 1013
6.9 7.6
80,174 12,364
93.1 92.4
0.0029
229 441 535 577 (34.4)
1.4 1.3 1.6 3.7 (±50.8)
16,082 33,674 33,038 14,888 (18.8)
98.6 98.7 98.4 96.3 (±27.1)
b.0001
1196 2429 2296 1005 (8.3)
7.3 7.1 6.8 6.5 (±6.7)
15,115 31,686 31,277 14,460 (19.9)
92.7 92.9 93.2 93.5 (±28.6)
0.0136
1036 746
1.7 2.0
61,384 36,298
98.3 98.0
b.0001
4118 2808
6.6 7.6
58,302 34,236
93.4 92.4
b.0001
204 1578 (77.6) (84.1) (20.6) (50.7) (3.4) (2.5) (435.8) (383.2) 1782
1.7 1.8 (±19.2) (±13.9) (±12.3) (±19.4) (±2.5) (±2.2) (±208.6) (±344.1) 1.8
11,954 85,728 (75.9) (86.6) (22.2) (54.1) (3.9) (2.8) (470.5) (507.5) 97,682
98.3 98.2 (±19.7) (±10.7) (±12.5) (±19.6) (±2.8) (±2.3) (±224.1) (±428.6) 98.2
0.3131
895 6031 (77.2) (85.6) (21.0) (52.3) (3.6) (2.5) (446.3) (483.0) 6926
7.4 6.9 (±19.6) (±11.1) (±12.4) (±19.1) (±2.6) (±2.2) (±214.7) (±428.4) 7.0
11,263 81,275 (75.8) (86.6) (22.3) (54.2) (3.9) (2.8) (471.7) (507.0) 92,538
92.6 93.1 (±19.7) (±10.7) (±12.5) (±19.6) (±2.8) (±2.3) (±224.4) (±427.4) 93.0
0.0657
higher numbers of doctors per bed (for one doctor per 10 beds, OR = 0.96 and SD = 0.92–0.99) had lower risks for death within 30 days. Patients at hospitals with a higher percentage of RNs with respect to the total number of nurses had lower risks for readmission within 30 days (for each 10% more RNs, OR = 0.89 and SD = 0.85–0.94). Patients at hospitals that have higher intake of stroke patients had lower risk for readmission within 30 days. We also performed a sub-group analysis to investigate the correlation between hospital human resources (percentages of specialist doctors, number of total doctor per bed, number of neurologists, number of neurosurgeons, and percentages of RNs) and readmission or death within 30 days after discharge or hospitalization. First, the data were stratified with respect to each major diagnosis. Higher percentages of specialists among total doctors and higher number of neurosurgeon have lower trend for readmission within 30 days after discharge for both intracerebral hemorrhage and cerebral infarction, but these were not statistically significant. And, higher percentages of RNs with respect to the total number of nurses were correlated with lower readmission rates within 30 days for intracerebral hemorrhage (OR = 0.88, SD = 0.77–0.99) and cerebral infarction (OR = 0.90, SD = 0.84–0.95); these values were for each 10% increase in RNs (Fig. 1). In the sub-group analysis for 30 day mortality, higher percentages of specialists with respect to the total number of doctors were correlated with lower risks of death within 30 days for intracerebral hemorrhage
b.0001
0.0002 b.0001 b.0001 b.0001 b.0001 b.0001 b.0001 b.0001
b.0001
b.0001 b.0001 b.0001 b.0001 b.0001 b.0001 b.0001 b.0001
(OR = 0.93, SD = 0.86–0.99) and cerebral infarction (OR = 0.93, SD = 0.88–0.99); these values were for each 10% increase in specialists. Also, higher number of total doctors per bed was correlated with lower risk of 30 day mortality for cerebral infarction (OR = 0.88, SD = 0.80– 0.97). In addition, higher number of neurologists and neurosurgeons has lower trends for 30 day mortality in both intracerebral hemorrhage and cerebral infarction, although these results were not statistically significant. Higher percentages of RNs with respect to the total number of nurses were correlated with lower risks of death within 30 days for cerebral infarction (OR = 0.93, SD = 0.89–0.98); these values were for each 10% increase in RNs (Fig. 2). 4. Discussion The overall health status of South Korea has improved in recent years due to the introduction of NHI and health technology advances. However, the aging and elderly population in South Korea is now expanding rapidly, and the country is faced with new health-care challenges related to geriatric medicine. Most of these new health-care challenges involve chronic diseases [3]. Cerebrovascular diseases such as stroke are reported as the leading cause of morbidity and mortality in South Korea [16–18]. Many studies and governmental committees have explored ways to address these emerging health-care challenges, but efficient and long-term solutions have not been identified.
140
K.-T. Han et al. / Journal of the Neurological Sciences 353 (2015) 137–142
Table 2 Odds ratios (OR) illustrating factors associated with readmission or death within 30 days after discharge or hospitalization for stroke; derived from a multilevel model. Variable
Patient-level Major diagnoses Intracerebral hemorrhage Cerebral infarction Age (years) b64 65+ Gender Male Female Charlson Comorbidity Index 0 1 2 3+ Type of health insurance NHI Medical-Aid Hospitalization year 2010 2011 2012 2013 Length of stay Hospital-level Teaching status Teaching hospital Non-teaching hospital Ownership Public Private Percentage of specialists (per 10% increase) Percentage of RNs (per 10% increase) Number of total doctors per bed (per 10 beds) Number of total nurses per bed (per 10 beds) Number of neurosurgeons Number of neurologists Number of beds (per 100 beds increase) Stroke patient admittance (per 100-fold increase)
Readmission within 30 days
Death within 30 days
OR
OR
95% CI
95% CI
1.20 1.04 1.39 8.68 8.07 9.34 1.00 – – 1.00 – – 1.00 – – 1.00 – – 0.76 0.65 0.89 1.78 1.62 1.95 1.00 – – 1.00 – – 1.03 0.91 1.17 1.14 1.06 1.22 1.00 1.23 1.40 1.53
– 0.80 0.94 1.03
– 1.88 2.11 2.27
1.00 1.43 1.23 2.10
– 1.13 0.97 1.69
– 1.81 1.55 2.61
1.00 – – 1.00 – – 1.14 0.96 1.35 1.14 1.03 1.26 1.00 1.06 1.36 2.91 1.01
– 0.85 1.10 2.37 1.01
– 1.32 1.68 3.58 1.01
1.00 0.94 0.89 0.85 0.91
– 0.85 0.80 0.75 0.90
– 1.05 0.99 0.96 0.91
1.00 – – 1.00 – – 0.89 0.68 1.16 0.89 0.77 1.04 1.00 1.18 0.98 0.89 1.00 0.98 0.86 1.15 0.98 0.95
– 0.99 0.90 0.85 0.88 0.93 0.56 0.70 0.90 0.92
– 1.41 1.06 0.94 1.13 1.02 1.32 1.90 1.05 0.98
1.00 0.82 0.93 0.98 0.91 0.98 1.07 0.83 0.96 1.01
– 0.75 0.89 0.94 0.85 0.95 0.86 0.63 0.92 0.99
– 0.91 0.98 1.02 0.98 1.00 1.34 1.11 0.99 1.02
If government directed more resources into these issues, it is expected that these issues could be solved [19] using effective management strategies. In 2013, there were conflicting opinions in South Korea on policy for human resource supply and demand in the health-care services. These conflicting opinions arose during proposals to use alternative human resources to meet the overwhelming need for qualified RNs [20]. Many health-care policy advisors voiced concerns about possible reductions in the quality of care if less qualified individuals were recruited into health care. These debates have continued and many reviews have been conducted. However, there is a lack of substantive data on this subject, which is necessary for establishing effective policies. We performed a study focusing on morbidity and mortality of individuals with chronic cerebrovascular disease, and assessed whether patient-level and hospital-level factors affected patient outcomes. We analyzed correlations between stroke patient outcomes (measured as readmission or death within 30 days after discharge or hospitalization, respectively, for stroke) and hospital human resources, including the percentages of specialist doctors and RNs with respect to the total numbers of doctors and nurses, respectively. Our results indicate that higher percentages of specialists correlate with lower risk for death within 30 days and higher percentages of RNs correlate with lower risk for
readmission within 30 days. Also, the numbers of neurologists and neurosurgeons have positive trend with outcomes of stroke inpatients, although these results were not statistically significant. Our findings suggest that hospitals with richer staffing such as specialist and RNs have positive relation with quality of care and outcomes in stroke inpatients. Therefore, effective health-care policy should be based on efforts to increase the available human resource pool in South Korea by recruiting trained health-care professionals and/or providing incentives for students to enter health-care studies. However, this strategy would require significantly more financial resources than are currently available. A realistic alternative would be to increase the pool of health-care assistants and aides. Our data suggest that this alternative strategy will not produce the best possible outcomes for stroke patients. Many studies have shown that care from specialists and RNs results in better patient outcomes and patient management [21–25]. Other studies suggest that the quality of care is reduced when care is performed by LPNs instead of RNs [26–28]. Our results are consistent with all of these studies. Therefore, it will be necessary for healthcare policy makers to re-evaluate the current health-care system to ensure the best outcomes for stroke patients. New strategies should be developed to effectively utilize health-care resources and improve the quality of care for management of chronic diseases. This study has several strengths compared with previous studies. First, we conducted multilevel analysis using National Health Insurance claim data that reflected both patient-level and hospital-level characteristics. These data are especially useful for establishing evidence-based health policies. Second, to our knowledge, this was the first study in South Korea that analyzed correlations between the percentages of specialists and RNs and stroke patient outcomes. Although there are many factors that influence the management of chronic diseases such as stroke, studies investigating those factors are lacking in South Korea. Our study examined these factors, and the results can be used to improve stroke patient outcomes. Next, we considered both readmission and death within 30 days as health care outcomes in stroke inpatient. Thus, the results of this study could reflect the quality of care and outcomes in stroke inpatients [29,30]. Finally, we used percentages of specialists and RNs to reflect the skill mix of hospital staffing [9]. Thus, our study suggests strategies for improvement of hospital human resource pools, which can be implemented by health-care policy makers and hospital managers. Our study also has some limitations. First, the NHI claim data we analyzed included hospitalization cases from 120 hospitals. It may be difficult to generalize our results from this relatively small sample to the entire country of South Korea. We could not consider details related with admitted or transferred to other hospitals, because data used in our study not included cases. Our results also do not include cost comparisons. Next, although surgical procedures are an important factor in caring stroke patients, they were not considered in our analyses because the relevant details were not included in the NHI data. We did not analyze patient care details such as medications. Also, we could not consider details about whether each stroke inpatient received better care by general physician or specialists/RNs or LPNs due to limitation of NHI data. Third, we did not assessed severity of stroke inpatient through a specific scale, because information related with that was not included in data. But, we considered severity as Charlson Comorbidity Index to more detailed study. Such index was used to adjust severity of stroke inpatients in many previous studies [31]. Fourth, there were positive trends of 30 day mortality in stroke inpatients by hospital with more neurologist and neurosurgeons, but it was not statistically significant. It seems to occur by limitations of data. Given that specialist including neurologist or neurosurgeon would substantially affect in caring stroke inpatient, these issues were need to treat through using more detailed data. Finally, our analysis considered the causes of death as all-cause mortality. Therefore, death within 30 days could be caused by accident or another disease; mortality did not have to be caused by stroke.
K.-T. Han et al. / Journal of the Neurological Sciences 353 (2015) 137–142
141
Fig. 1. Odds ratios showing factors correlated with readmission within 30 days of discharge, stratified by diagnosis. *The OR as marked to circle point was calculated by multilevel analysis adjusted for inpatient-level characteristics and hospital-level characteristics, and results were statistically significant if each bar as marked to SD is not reached the cutoff line in 1.00.
Fig. 2. Odds ratios showing factors correlated with death within 30 days of hospitalization, stratified by diagnosis. *The OR as marked to circle point was calculated by multilevel analysis adjusted for inpatient-level characteristics and hospital-level characteristics, and results were statistically significant if each bar as marked to SD is not reached the cutoff line in 1.00.
142
K.-T. Han et al. / Journal of the Neurological Sciences 353 (2015) 137–142
In conclusion, our results indicate that care given by specialist doctors and RNs has a substantial positive effect to reduce the risk of readmission or death within 30 days after discharge or hospitalization for stroke, respectively. Based on these results a solution for managing the health-care system could be to recruit more trained health-care workers. Additional studies will be required to establish effective and sustainable strategies that can be implemented by health-care policy makers. 5. Conclusions The percentages of specialists and RNs delivering care were positively correlated with better outcomes of stroke patients. In particular, better outcomes as readmission and death within 30 days of patients with cerebral infarction were correlated with care given by staff with a higher percentage of specialists and RNs. Therefore, health care professionals have to effort in training richer hospital staffing such as specialists or RNs, and rewarding for them to provide better treatment in managing for stroke inpatients. Authors' contributions K.T.H. designed the study, researched data, performed statistical analyses and wrote the manuscript. S.J.K., S.I.J., C.O.K., S.Y.L., H.J.L., and E.C.P. contributed to the discussion and reviewed and edited the manuscript. E.C.P. is the guarantor of this work and as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Conflict of interest The authors report no conflicts of interest in this work. References [1] Kwon S. Thirty years of national health insurance in South Korea: lessons for achieving universal health care coverage. Health Policy Plan 2009;24(1):63–71. [2] Statistics Korea. Population trends survey; 2013. [3] Yoon S-J. Future directions of chronic disease management in South Korea. J Korean Med Assoc 2012;55(5):414–6. [4] Statistics Korea. All causes of mortality; 2013. [5] Awad I, Spetzler R, Hodak J, Awad C, Carey R. Incidental subcortical lesions identified on magnetic resonance imaging in the elderly. I. Correlation with age and cerebrovascular risk factors. Stroke 1986;17(6):1084–9. [6] Hajat C, Dundas R, Stewart JA, Lawrence E, Rudd AG, Howard R, et al. Cerebrovascular risk factors and stroke subtypes differences between ethnic groups. Stroke 2001; 32(1):37–42. [7] Organization for Economic Cooperation and Development. Health at a glance 2013. OECD Indicators; 2013.
[8] WORKSHEET PPS. Panel size: how many patients can one doctor manage? Fam Pract Manag 2007;14(4):44–51. [9] Lankshear AJ, Sheldon TA, Maynard A. Nurse staffing and healthcare outcomes: a systematic review of the international research evidence. Adv Nurs Sci 2005; 28(2):163–74. [10] Rodgers H, Dennis M, Cohen D, Rudd A. British Association of Stroke Physicians: benchmarking survey of stroke services. Age Ageing 2003;32(2):211–7. [11] Grilli R, Minozzi S, Tinazzi A, Labianca R, Sheldon T, Liberati A. Do specialists do it better? The impact of specialization on the processes and outcomes of care for cancer patients. Ann Oncol 1998;9(4):365–74. [12] Silvestri GA, Handy J, Lackland D, Corley E, Reed CE. Specialists achieve better outcomes than generalists for lung cancer surgery. Chest J 1998;114(3):675–80. [13] Buerhaus PI, Donelan K, Ulrich BT, Norman L, Dittus R. Is the shortage of hospital registered nurses getting better or worse? Findings from two recent national surveys of RNs. Nurs Econ 2004;23(2):61–71 96, 55. [14] Clark PA, Leddy K, Drain M, Kaldenberg D. State nursing shortages and patient satisfaction: more RNs—better patient experiences. J Nurs Care Qual 2007;22(2): 119–27. [15] Starfield B, Shi L, Grover A, Macinko J. The effects of specialist supply on populations' health: assessing the evidence. Cancer 2005;103(20.93):23.18. [16] Choi-Kwon S, Kim H-S, Kwon SU, Kim JS. Factors affecting the burden on caregivers of stroke survivors in South Korea. Arch Phys Med Rehabil 2005;86(5):1043–8. [17] Hong K-S, Kim J, Cho Y-J, Seo S-Y, Hwang S-I, Kim S-C, et al. Burden of ischemic stroke in Korea: analysis of disability-adjusted life years lost. J Clin Neurol 2011; 7(2):77–84. [18] Kim H-J, Kim Y, Seo H-Y, Kim E-J, Yoon S-J, Oh I-H. The economic burden of stroke in 2010 in Korea. J Korean Med Assoc 2012;55(12):1226–36. [19] Evans JR, Hall KL, Warford J. Health care in the developing world: problems of scarcity and choice (Shattuck lecture). 1981;61:74 (population). [20] Kim BYC Boon Han, Kim Jin Kyung, Lee Ae-young, Hwang Seon Young, Cho Jonn Ah, Kim Jung A. Current situation and the forecast of the supply and demand of the nursing workforce in Korea. Korean J Adult Nurs 2013;25(6):701–11. [21] Mark BA, Harless DW, McCue M, Xu Y. A longitudinal examination of hospital registered nurse staffing and quality of care. Health Serv Res 2004;39(2):279–300. [22] Kane RL, Shamliyan TA, Mueller C, Duval S, Wilt TJ. The association of registered nurse staffing levels and patient outcomes: systematic review and meta-analysis. Med Care 2007;45(12):1195–204. [23] McGahan M, Kucharski G, Coyer F. Nurse staffing levels and the incidence of mortality and morbidity in the adult intensive care unit: a literature review. Aust Crit Care 2012; 25(2):64–77. [24] Manawadu D, Choyi J, Kalra L. The impact of early specialist management on outcomes of patients with in-hospital stroke. PLoS One 2014;9(8):e104758. [25] Saposnik G, Baibergenova A, O'Donnell M, Hill M, Kapral M, Hachinski V. Hospital volume and stroke outcome: does it matter? Neurology 2007;69(11):1142–51. [26] Kim H, Harrington C, Greene WH. Registered nurse staffing mix and quality of care in nursing homes: a longitudinal analysis. Gerontologist 2009;49(1):81–90. [27] Bostick JE, Rantz MJ, Flesner MK, Riggs CJ. Systematic review of studies of staffing and quality in nursing homes. J Am Med Dir Assoc 2006;7(6):366–76. [28] Needleman J, Buerhaus P, Mattke S, Stewart M, Zelevinsky K. Nurse-staffing levels and the quality of care in hospitals. N Engl J Med 2002;346(22):1715–22. [29] Thomas JW, Holloway JJ. Investigating early readmission as an indicator for quality of care studies. Med Care 1991;29(4):377–94. [30] Ashton CM, Del Junco DJ, Souchek J, Wray NP, Mansyur CL. The association between the quality of inpatient care and early readmission: a meta-analysis of the evidence. Med Care 1997;1044–59. [31] Goldstein LB, Samsa GP, Matchar DB, Horner RD. Charlson Index comorbidity adjustment for ischemic stroke outcome studies. Stroke 2004;35(8):1941–5.
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Positive correlation between care given by specialists and registered nurses and improved outcomes for stroke patients Kyu-Tae Han a,b, Sun Jung Kim a,c, Sung-In Jang a,b,d, Seung Ju Kim a,b, Seo Yoon Lee b,e, Hyo Jung Lee b,e, Eun-Cheol Park b,d,⁎ a
Department of Public Health, Graduate School, Yonsei University, Seoul, Republic of Korea Institute of Health Services Research, Yonsei University College of Medicine, Seoul, Republic of Korea Department of Health Administration, Namseoul University, Cheonan, Republic of Korea d Department of Preventive Medicine, Yonsei University College of Medicine, Seoul, Republic of Korea e Department of Health Policy and Management, Graduate School of Public Health, Yonsei University, Seoul, Republic of Korea b c
a r t i c l e
i n f o
Article history: Received 2 December 2014 Received in revised form 12 March 2015 Accepted 17 April 2015 Available online 29 April 2015 Keywords: Stroke Quality of care Readmission Death Specialist Registered nurse
a b s t r a c t Background: Cerebrovascular diseases are the second-highest cause of death in South Korea (9.6% of all causes of mortality in 2013). South Korea has a shortage of trained medical personnel compared with other countries and the demands for health care are continuously increasing. Our study sought to determine the relationship between hospital human resources and the outcomes of stroke patients. Methods: We used data from NHI claims (n = 99,464) at 120 hospitals to analyze readmission or death within 30 days after discharge or hospitalization for stroke patients during 2010–2013. We used multilevel models that included both patient-level and hospital-level variables to examine factors associated with readmission or death within 30 days. Results: A total of 1782 (1.8%) patients were readmitted within 30 days, and death occurred within 30 days for 6926 (7.0%) patients. Patients cared for by a higher percentages of specialists or registered nurses had a lower risk of readmission or death within 30 days (readmission per 10% increase in registered nurses, OR = 0.89 and SD = 0.85–0.94; death per 10% increase in specialists, OR = 0.93 and SD = 0.89–0.98). Conclusions: The percentages of specialist and registered nurses caring for stroke patients were positively correlated with better patient outcomes, particularly for patients with cerebral infarction. © 2015 Elsevier B.V. All rights reserved.
1. Introduction The health status and longevity of people in South Korea improved rapidly during the latter 20th century due to the introduction of National Health Insurance (NHI) and the development of health technologies [1]. This led to increases in the elderly population, the emergence of new health problems associated with geriatrics (individuals aged greater than 65 years, who represented 12.2% of the total population in 2013), and increases in morbidity and mortality among the aged [2]. These health problems include chronic diseases and deterioration in mental health such as suicide among the elderly. Many health-care specialists are gaining expertise in working with and treating geriatric diseases.
Abbreviations: OECD, Organization for Economic Cooperation and Development; NHI, National Health Insurance; RNs, registered nurses; LPNs, licensed practical nurses; ICD, International Classification of Diseases; CCI, Charlson Comorbidity Index; ANOVA, analysis of variance; OR, odds ratio. ⁎ Corresponding author at: Department of Preventive Medicine & Institute of Health Services Research, College of Medicine, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-752, Republic of Korea. Tel.: +82 2 2228 1862; fax: +82 2 392 7734. E-mail address: [email protected] (E.-C. Park).
http://dx.doi.org/10.1016/j.jns.2015.04.034 0022-510X/© 2015 Elsevier B.V. All rights reserved.
Chronic diseases remain as the primary cause of mortality among elderly populations in South Korea [3]. Cerebrovascular diseases are one of the primary causes of mortality in South Korea. National statistics in Korea indicate that cerebrovascular diseases are the second-highest cause of death (9.6% of all causes of mortality in 2013) [4]. The burden of cerebrovascular disease is expected to continuously increase in the immediate future because the major risk factor for stroke is advanced age [5,6]. Therefore, it is important to identify factors that can reduce or prevent mortality due to stroke. Here, we focus on the role of human resources and medical personnel during stroke patient hospitalization. South Korea is recently experiencing a shortage in trained medical personnel, including specialist doctors and registered nurses. However, the demand for health care has increased during this time. The numbers of trained medical personnel in South Korea is lower than those in other countries within the Organization for Economic Cooperation and Development (OECD). In South Korea, there are 2.0 practicing doctors and 4.8 practicing nurses per 1000 people, whereas in the OECD there are an average of 3.2 practicing doctors and 8.8 practicing nurses per 1000 people [7]. Many studies have investigated the role of professional human resources in hospitals [8–10]. These studies indicate that the
138
K.-T. Han et al. / Journal of the Neurological Sciences 353 (2015) 137–142
quality of care and patient outcomes were significantly improved when the ratio of medical personnel to patients was larger, and when the medical personnel had specialized training [11–15]. The shortage of health-care human resources in South Korea is caused by practical constraints and limited resources, and it will be challenging to solve these problems in the immediate future. Therefore, our study investigated the relationship between the numbers of specialist doctors and registered nurses (RNs) in hospitals and stroke patient outcomes. We performed sub-group analysis to identify differences based on patient diagnoses. 2. Methods 2.1. Study population There are about 1730 hospitals including 39 public hospitals during 2010–2013 year in South Korea, but, the data we used in this study was only included 156 hospitals (117 private and 39 public) after extracting through propensity score matching-methods (1:3) adjusting some variables as follows: region of hospitals, nursing staffing level, number of total beds, number of intensive care unit beds, number of emergency room beds, and number of doctors. And then, this study only included hospitalization due to stroke. Stroke was classified according to the International Classification of Diseases (ICD-10; including I61, I62, and I63). We excluded the hospitals without stroke inpatient cases (N = 36). Finally, 120 hospitals (public = 32 vs private = 88, 99,464 hospitalization cases) were included for analysis. The unit of analysis was one hospitalization case. 2.2. Variables To reflect the quality of care and outcomes in stroke inpatient, we used readmission within 30 days after discharge for stroke and death within 30 days after hospitalization for stroke as outcome variables in this study. We identified the patient's first discharge or hospitalization in the calendar year as the first index discharge or hospitalization. Readmission or death within 30 days was defined as readmission or death within 30 calendar days based on the first index discharge or hospitalization. The primary variables of interest with respect to readmission or death within 30 days after discharge or hospitalization were the percentages of specialist doctors and RNs. These were defined as the percentage of specialist doctors to the total number of doctors, and the percentage of RNs to the total number of nurses [including RNs and licensed practical nurses (LPNs)] in each hospital. These indicators were used to reflect the richer skill mix in human resource pool, and were calculated as shown in the following two equations: Percentage of specialist doctors ¼ ðNumber of specialist doctors=Total number of doctorsÞ 100 Percentage of RNs ¼ðNumber of RNs=Total number of nursesÞ 100: We adjusted these correlation analyses for patient-level and hospital-level variables. Patient-level variables included major diagnosis, age, gender, Charlson Comorbidity Index (CCI) score, health insurance type, hospitalization year, and length of hospitalization. Major diagnoses were classified according to the following ICD groupings: I61–I62, intracerebral hemorrhage; and I63, cerebral infarction. These classifications reflected specific pathological mechanisms. The CCI was used to account for the effect of comorbid disorders or diseases. Hospital-level variables included structural characteristics, human resources, and the number of stroke patients in each hospital. The structural characteristic variables included ownership status, teaching status, and number of beds. The human resource variables included the number
of doctors and nurses per bed, and the number of neurosurgeons and neurologists. 2.3. Statistical analysis We analyzed the distribution of each categorical variable by examining the frequencies and percentages of the variables and performing χ2 tests to identify correlations with patient readmission or death within 30 days. These analyses were performed for both patient-level and hospital-level variables. We also performed analysis of variance (ANOVA) to compare the average values and standard deviations for continuous variables. Then, we used multilevel models including both patient-level and hospital-level variables to identify correlations with patient readmission or death within 30 days after discharge or hospitalization, respectively. We also performed sub-group analyses with respect to major diagnoses for stroke patients. All statistical analyses were performed using SAS statistical software version 9.2. All calculated P-values were two-sided and considered significant at P b 0.05. 3. Results The data used in this study included 99,464 hospitalization cases. The number of patients readmitted within 30 days after discharge was 1782 (1.8%), and the number of mortalities within 30 days after hospitalization was 6926 (7.0%). Table 1 presents the results of univariate associations between each variable and readmission or death within 30 days after discharge or hospitalization, respectively, for stroke. The numbers of readmissions and deaths within 30 days after discharge or hospitalization were greater for patients with major diagnoses of intracerebral hemorrhage (2.3% and 17.1%, respectively) than cerebral infarction (1.7% and 4.5%, respectively). Analysis of CCI scores showed that the higher the score, the higher the percentages of readmission within 30 days. Mortality due to stroke was inversely correlated to the time since hospitalization (decreasing over the years of analysis), whereas readmission due to stroke was directly correlated with time (increasing during the years of analysis). These correlations also were observed when analyzing the length of hospitalization. The analyses of hospital characteristics revealed significant differences in readmissions or deaths within 30 days for teaching hospitals (1.7% and 6.6%, respectively) and non-teaching hospitals (2.0% and 7.6%, respectively). The average percentage of RNs was lower for cases that were readmitted or perished within 30 days after discharge or hospitalization. The average numbers of neurosurgeons, neurologists, beds, and stroke patient admittance were lower in patients that were readmitted or perished within 30 days after discharge or hospitalization for stroke. Table 2 presents the results of multilevel analyses considering both patient-level and hospital-level variables for readmission or death within 30 days after discharge or hospitalization for stroke. Patients hospitalized for intracerebral hemorrhage had a higher risk of readmission or death within 30 days [for readmission, odds ratio (OR) = 1.20 and SD = 1.04–1.39; for death, OR = 8.68 and SD = 8.07–9.34] than patients hospitalized for cerebral infarction (this group was used as the reference). Multilevel analyses of age groups indicated that older patients had a higher risk of death within 30 days after hospitalization, and a lower risk of readmission within 30 days. Analyses of CCI scores showed that higher scores were associated with greater risk for readmission or death within 30 days after discharge or hospitalization. Multilevel analyses of hospital-level variables show that patients at non-teaching hospitals have lower trend for readmission or death within 30 days after discharge or hospitalization, but this was not statistically significant. Patients at private hospitals had lower risk for death within 30 days (OR = 0.82 and SD = 0.75–0.91; teaching hospitals were used as the reference). Patients at hospitals with a higher percentage of specialist doctors (for each 10% more specialists, OR = 0.93 and SD = 0.89–0.98) or
K.-T. Han et al. / Journal of the Neurological Sciences 353 (2015) 137–142
139
Table 1 Univariate correlations between readmission and death within 30 days after discharge or hospitalization. Variables
Readmission within 30 days (n = 99,464) Yes
Patient-level Major diagnoses Intracerebral hemorrhage Cerebral infarction Age (years) b64 65+ Gender Male Female Charlson Comorbidity Index 0 1 2 3+ Type of health insurance NHI Medical-Aid Hospitalization year 2010 2011 2012 2013 Length of stay Hospital-level Teaching status Teaching hospital (n = 43) Non-teaching hospital (n = 77) Ownership Public (n = 32) Private (n = 88) Percentage of specialists Percentage of RNs Number of total doctors per bed Number of total nurses per bed Number of neurosurgeons Number of neurologists Number of beds Stroke patient admittance Total
Death within 30 days (n = 99,464)
No
P-value
N (mean)
% (SD)
N (mean)
% (SD)
445 1337
2.3 1.7
19,159 78,523
97.7 98.3
705 1077
2.0 1.7
35,266 62,416
976 806
1.8 1.8
53 147 246 1336
Dead
Alive
P-value
N (mean)
% (SD)
N (mean)
% (SD)
b.0001
3358 3568
17.1 4.5
16,246 76,292
82.9 95.5
b.0001
98.0 98.3
0.0026
1734 5192
4.8 8.2
34,237 58,301
95.2 91.8
b.0001
53,610 44,072
98.2 98.2
0.9248
3439 3487
6.3 7.8
51,147 41,391
93.7 92.2
b.0001
1.4 1.7 1.8 1.8
3864 8489 13,249 72,080
98.6 98.3 98.2 98.2
0.1661
186 448 636 5656
4.7 5.2 4.7 7.7
3731 8188 12,859 67,760
95.3 94.8 95.3 92.3
b.0001
1498 284
10.3 0.3
13,093 84,589
89.7 99.7
0.0019
5913 1013
6.9 7.6
80,174 12,364
93.1 92.4
0.0029
229 441 535 577 (34.4)
1.4 1.3 1.6 3.7 (±50.8)
16,082 33,674 33,038 14,888 (18.8)
98.6 98.7 98.4 96.3 (±27.1)
b.0001
1196 2429 2296 1005 (8.3)
7.3 7.1 6.8 6.5 (±6.7)
15,115 31,686 31,277 14,460 (19.9)
92.7 92.9 93.2 93.5 (±28.6)
0.0136
1036 746
1.7 2.0
61,384 36,298
98.3 98.0
b.0001
4118 2808
6.6 7.6
58,302 34,236
93.4 92.4
b.0001
204 1578 (77.6) (84.1) (20.6) (50.7) (3.4) (2.5) (435.8) (383.2) 1782
1.7 1.8 (±19.2) (±13.9) (±12.3) (±19.4) (±2.5) (±2.2) (±208.6) (±344.1) 1.8
11,954 85,728 (75.9) (86.6) (22.2) (54.1) (3.9) (2.8) (470.5) (507.5) 97,682
98.3 98.2 (±19.7) (±10.7) (±12.5) (±19.6) (±2.8) (±2.3) (±224.1) (±428.6) 98.2
0.3131
895 6031 (77.2) (85.6) (21.0) (52.3) (3.6) (2.5) (446.3) (483.0) 6926
7.4 6.9 (±19.6) (±11.1) (±12.4) (±19.1) (±2.6) (±2.2) (±214.7) (±428.4) 7.0
11,263 81,275 (75.8) (86.6) (22.3) (54.2) (3.9) (2.8) (471.7) (507.0) 92,538
92.6 93.1 (±19.7) (±10.7) (±12.5) (±19.6) (±2.8) (±2.3) (±224.4) (±427.4) 93.0
0.0657
higher numbers of doctors per bed (for one doctor per 10 beds, OR = 0.96 and SD = 0.92–0.99) had lower risks for death within 30 days. Patients at hospitals with a higher percentage of RNs with respect to the total number of nurses had lower risks for readmission within 30 days (for each 10% more RNs, OR = 0.89 and SD = 0.85–0.94). Patients at hospitals that have higher intake of stroke patients had lower risk for readmission within 30 days. We also performed a sub-group analysis to investigate the correlation between hospital human resources (percentages of specialist doctors, number of total doctor per bed, number of neurologists, number of neurosurgeons, and percentages of RNs) and readmission or death within 30 days after discharge or hospitalization. First, the data were stratified with respect to each major diagnosis. Higher percentages of specialists among total doctors and higher number of neurosurgeon have lower trend for readmission within 30 days after discharge for both intracerebral hemorrhage and cerebral infarction, but these were not statistically significant. And, higher percentages of RNs with respect to the total number of nurses were correlated with lower readmission rates within 30 days for intracerebral hemorrhage (OR = 0.88, SD = 0.77–0.99) and cerebral infarction (OR = 0.90, SD = 0.84–0.95); these values were for each 10% increase in RNs (Fig. 1). In the sub-group analysis for 30 day mortality, higher percentages of specialists with respect to the total number of doctors were correlated with lower risks of death within 30 days for intracerebral hemorrhage
b.0001
0.0002 b.0001 b.0001 b.0001 b.0001 b.0001 b.0001 b.0001
b.0001
b.0001 b.0001 b.0001 b.0001 b.0001 b.0001 b.0001 b.0001
(OR = 0.93, SD = 0.86–0.99) and cerebral infarction (OR = 0.93, SD = 0.88–0.99); these values were for each 10% increase in specialists. Also, higher number of total doctors per bed was correlated with lower risk of 30 day mortality for cerebral infarction (OR = 0.88, SD = 0.80– 0.97). In addition, higher number of neurologists and neurosurgeons has lower trends for 30 day mortality in both intracerebral hemorrhage and cerebral infarction, although these results were not statistically significant. Higher percentages of RNs with respect to the total number of nurses were correlated with lower risks of death within 30 days for cerebral infarction (OR = 0.93, SD = 0.89–0.98); these values were for each 10% increase in RNs (Fig. 2). 4. Discussion The overall health status of South Korea has improved in recent years due to the introduction of NHI and health technology advances. However, the aging and elderly population in South Korea is now expanding rapidly, and the country is faced with new health-care challenges related to geriatric medicine. Most of these new health-care challenges involve chronic diseases [3]. Cerebrovascular diseases such as stroke are reported as the leading cause of morbidity and mortality in South Korea [16–18]. Many studies and governmental committees have explored ways to address these emerging health-care challenges, but efficient and long-term solutions have not been identified.
140
K.-T. Han et al. / Journal of the Neurological Sciences 353 (2015) 137–142
Table 2 Odds ratios (OR) illustrating factors associated with readmission or death within 30 days after discharge or hospitalization for stroke; derived from a multilevel model. Variable
Patient-level Major diagnoses Intracerebral hemorrhage Cerebral infarction Age (years) b64 65+ Gender Male Female Charlson Comorbidity Index 0 1 2 3+ Type of health insurance NHI Medical-Aid Hospitalization year 2010 2011 2012 2013 Length of stay Hospital-level Teaching status Teaching hospital Non-teaching hospital Ownership Public Private Percentage of specialists (per 10% increase) Percentage of RNs (per 10% increase) Number of total doctors per bed (per 10 beds) Number of total nurses per bed (per 10 beds) Number of neurosurgeons Number of neurologists Number of beds (per 100 beds increase) Stroke patient admittance (per 100-fold increase)
Readmission within 30 days
Death within 30 days
OR
OR
95% CI
95% CI
1.20 1.04 1.39 8.68 8.07 9.34 1.00 – – 1.00 – – 1.00 – – 1.00 – – 0.76 0.65 0.89 1.78 1.62 1.95 1.00 – – 1.00 – – 1.03 0.91 1.17 1.14 1.06 1.22 1.00 1.23 1.40 1.53
– 0.80 0.94 1.03
– 1.88 2.11 2.27
1.00 1.43 1.23 2.10
– 1.13 0.97 1.69
– 1.81 1.55 2.61
1.00 – – 1.00 – – 1.14 0.96 1.35 1.14 1.03 1.26 1.00 1.06 1.36 2.91 1.01
– 0.85 1.10 2.37 1.01
– 1.32 1.68 3.58 1.01
1.00 0.94 0.89 0.85 0.91
– 0.85 0.80 0.75 0.90
– 1.05 0.99 0.96 0.91
1.00 – – 1.00 – – 0.89 0.68 1.16 0.89 0.77 1.04 1.00 1.18 0.98 0.89 1.00 0.98 0.86 1.15 0.98 0.95
– 0.99 0.90 0.85 0.88 0.93 0.56 0.70 0.90 0.92
– 1.41 1.06 0.94 1.13 1.02 1.32 1.90 1.05 0.98
1.00 0.82 0.93 0.98 0.91 0.98 1.07 0.83 0.96 1.01
– 0.75 0.89 0.94 0.85 0.95 0.86 0.63 0.92 0.99
– 0.91 0.98 1.02 0.98 1.00 1.34 1.11 0.99 1.02
If government directed more resources into these issues, it is expected that these issues could be solved [19] using effective management strategies. In 2013, there were conflicting opinions in South Korea on policy for human resource supply and demand in the health-care services. These conflicting opinions arose during proposals to use alternative human resources to meet the overwhelming need for qualified RNs [20]. Many health-care policy advisors voiced concerns about possible reductions in the quality of care if less qualified individuals were recruited into health care. These debates have continued and many reviews have been conducted. However, there is a lack of substantive data on this subject, which is necessary for establishing effective policies. We performed a study focusing on morbidity and mortality of individuals with chronic cerebrovascular disease, and assessed whether patient-level and hospital-level factors affected patient outcomes. We analyzed correlations between stroke patient outcomes (measured as readmission or death within 30 days after discharge or hospitalization, respectively, for stroke) and hospital human resources, including the percentages of specialist doctors and RNs with respect to the total numbers of doctors and nurses, respectively. Our results indicate that higher percentages of specialists correlate with lower risk for death within 30 days and higher percentages of RNs correlate with lower risk for
readmission within 30 days. Also, the numbers of neurologists and neurosurgeons have positive trend with outcomes of stroke inpatients, although these results were not statistically significant. Our findings suggest that hospitals with richer staffing such as specialist and RNs have positive relation with quality of care and outcomes in stroke inpatients. Therefore, effective health-care policy should be based on efforts to increase the available human resource pool in South Korea by recruiting trained health-care professionals and/or providing incentives for students to enter health-care studies. However, this strategy would require significantly more financial resources than are currently available. A realistic alternative would be to increase the pool of health-care assistants and aides. Our data suggest that this alternative strategy will not produce the best possible outcomes for stroke patients. Many studies have shown that care from specialists and RNs results in better patient outcomes and patient management [21–25]. Other studies suggest that the quality of care is reduced when care is performed by LPNs instead of RNs [26–28]. Our results are consistent with all of these studies. Therefore, it will be necessary for healthcare policy makers to re-evaluate the current health-care system to ensure the best outcomes for stroke patients. New strategies should be developed to effectively utilize health-care resources and improve the quality of care for management of chronic diseases. This study has several strengths compared with previous studies. First, we conducted multilevel analysis using National Health Insurance claim data that reflected both patient-level and hospital-level characteristics. These data are especially useful for establishing evidence-based health policies. Second, to our knowledge, this was the first study in South Korea that analyzed correlations between the percentages of specialists and RNs and stroke patient outcomes. Although there are many factors that influence the management of chronic diseases such as stroke, studies investigating those factors are lacking in South Korea. Our study examined these factors, and the results can be used to improve stroke patient outcomes. Next, we considered both readmission and death within 30 days as health care outcomes in stroke inpatient. Thus, the results of this study could reflect the quality of care and outcomes in stroke inpatients [29,30]. Finally, we used percentages of specialists and RNs to reflect the skill mix of hospital staffing [9]. Thus, our study suggests strategies for improvement of hospital human resource pools, which can be implemented by health-care policy makers and hospital managers. Our study also has some limitations. First, the NHI claim data we analyzed included hospitalization cases from 120 hospitals. It may be difficult to generalize our results from this relatively small sample to the entire country of South Korea. We could not consider details related with admitted or transferred to other hospitals, because data used in our study not included cases. Our results also do not include cost comparisons. Next, although surgical procedures are an important factor in caring stroke patients, they were not considered in our analyses because the relevant details were not included in the NHI data. We did not analyze patient care details such as medications. Also, we could not consider details about whether each stroke inpatient received better care by general physician or specialists/RNs or LPNs due to limitation of NHI data. Third, we did not assessed severity of stroke inpatient through a specific scale, because information related with that was not included in data. But, we considered severity as Charlson Comorbidity Index to more detailed study. Such index was used to adjust severity of stroke inpatients in many previous studies [31]. Fourth, there were positive trends of 30 day mortality in stroke inpatients by hospital with more neurologist and neurosurgeons, but it was not statistically significant. It seems to occur by limitations of data. Given that specialist including neurologist or neurosurgeon would substantially affect in caring stroke inpatient, these issues were need to treat through using more detailed data. Finally, our analysis considered the causes of death as all-cause mortality. Therefore, death within 30 days could be caused by accident or another disease; mortality did not have to be caused by stroke.
K.-T. Han et al. / Journal of the Neurological Sciences 353 (2015) 137–142
141
Fig. 1. Odds ratios showing factors correlated with readmission within 30 days of discharge, stratified by diagnosis. *The OR as marked to circle point was calculated by multilevel analysis adjusted for inpatient-level characteristics and hospital-level characteristics, and results were statistically significant if each bar as marked to SD is not reached the cutoff line in 1.00.
Fig. 2. Odds ratios showing factors correlated with death within 30 days of hospitalization, stratified by diagnosis. *The OR as marked to circle point was calculated by multilevel analysis adjusted for inpatient-level characteristics and hospital-level characteristics, and results were statistically significant if each bar as marked to SD is not reached the cutoff line in 1.00.
142
K.-T. Han et al. / Journal of the Neurological Sciences 353 (2015) 137–142
In conclusion, our results indicate that care given by specialist doctors and RNs has a substantial positive effect to reduce the risk of readmission or death within 30 days after discharge or hospitalization for stroke, respectively. Based on these results a solution for managing the health-care system could be to recruit more trained health-care workers. Additional studies will be required to establish effective and sustainable strategies that can be implemented by health-care policy makers. 5. Conclusions The percentages of specialists and RNs delivering care were positively correlated with better outcomes of stroke patients. In particular, better outcomes as readmission and death within 30 days of patients with cerebral infarction were correlated with care given by staff with a higher percentage of specialists and RNs. Therefore, health care professionals have to effort in training richer hospital staffing such as specialists or RNs, and rewarding for them to provide better treatment in managing for stroke inpatients. Authors' contributions K.T.H. designed the study, researched data, performed statistical analyses and wrote the manuscript. S.J.K., S.I.J., C.O.K., S.Y.L., H.J.L., and E.C.P. contributed to the discussion and reviewed and edited the manuscript. E.C.P. is the guarantor of this work and as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Conflict of interest The authors report no conflicts of interest in this work. References [1] Kwon S. Thirty years of national health insurance in South Korea: lessons for achieving universal health care coverage. Health Policy Plan 2009;24(1):63–71. [2] Statistics Korea. Population trends survey; 2013. [3] Yoon S-J. Future directions of chronic disease management in South Korea. J Korean Med Assoc 2012;55(5):414–6. [4] Statistics Korea. All causes of mortality; 2013. [5] Awad I, Spetzler R, Hodak J, Awad C, Carey R. Incidental subcortical lesions identified on magnetic resonance imaging in the elderly. I. Correlation with age and cerebrovascular risk factors. Stroke 1986;17(6):1084–9. [6] Hajat C, Dundas R, Stewart JA, Lawrence E, Rudd AG, Howard R, et al. Cerebrovascular risk factors and stroke subtypes differences between ethnic groups. Stroke 2001; 32(1):37–42. [7] Organization for Economic Cooperation and Development. Health at a glance 2013. OECD Indicators; 2013.
[8] WORKSHEET PPS. Panel size: how many patients can one doctor manage? Fam Pract Manag 2007;14(4):44–51. [9] Lankshear AJ, Sheldon TA, Maynard A. Nurse staffing and healthcare outcomes: a systematic review of the international research evidence. Adv Nurs Sci 2005; 28(2):163–74. [10] Rodgers H, Dennis M, Cohen D, Rudd A. British Association of Stroke Physicians: benchmarking survey of stroke services. Age Ageing 2003;32(2):211–7. [11] Grilli R, Minozzi S, Tinazzi A, Labianca R, Sheldon T, Liberati A. Do specialists do it better? The impact of specialization on the processes and outcomes of care for cancer patients. Ann Oncol 1998;9(4):365–74. [12] Silvestri GA, Handy J, Lackland D, Corley E, Reed CE. Specialists achieve better outcomes than generalists for lung cancer surgery. Chest J 1998;114(3):675–80. [13] Buerhaus PI, Donelan K, Ulrich BT, Norman L, Dittus R. Is the shortage of hospital registered nurses getting better or worse? Findings from two recent national surveys of RNs. Nurs Econ 2004;23(2):61–71 96, 55. [14] Clark PA, Leddy K, Drain M, Kaldenberg D. State nursing shortages and patient satisfaction: more RNs—better patient experiences. J Nurs Care Qual 2007;22(2): 119–27. [15] Starfield B, Shi L, Grover A, Macinko J. The effects of specialist supply on populations' health: assessing the evidence. Cancer 2005;103(20.93):23.18. [16] Choi-Kwon S, Kim H-S, Kwon SU, Kim JS. Factors affecting the burden on caregivers of stroke survivors in South Korea. Arch Phys Med Rehabil 2005;86(5):1043–8. [17] Hong K-S, Kim J, Cho Y-J, Seo S-Y, Hwang S-I, Kim S-C, et al. Burden of ischemic stroke in Korea: analysis of disability-adjusted life years lost. J Clin Neurol 2011; 7(2):77–84. [18] Kim H-J, Kim Y, Seo H-Y, Kim E-J, Yoon S-J, Oh I-H. The economic burden of stroke in 2010 in Korea. J Korean Med Assoc 2012;55(12):1226–36. [19] Evans JR, Hall KL, Warford J. Health care in the developing world: problems of scarcity and choice (Shattuck lecture). 1981;61:74 (population). [20] Kim BYC Boon Han, Kim Jin Kyung, Lee Ae-young, Hwang Seon Young, Cho Jonn Ah, Kim Jung A. Current situation and the forecast of the supply and demand of the nursing workforce in Korea. Korean J Adult Nurs 2013;25(6):701–11. [21] Mark BA, Harless DW, McCue M, Xu Y. A longitudinal examination of hospital registered nurse staffing and quality of care. Health Serv Res 2004;39(2):279–300. [22] Kane RL, Shamliyan TA, Mueller C, Duval S, Wilt TJ. The association of registered nurse staffing levels and patient outcomes: systematic review and meta-analysis. Med Care 2007;45(12):1195–204. [23] McGahan M, Kucharski G, Coyer F. Nurse staffing levels and the incidence of mortality and morbidity in the adult intensive care unit: a literature review. Aust Crit Care 2012; 25(2):64–77. [24] Manawadu D, Choyi J, Kalra L. The impact of early specialist management on outcomes of patients with in-hospital stroke. PLoS One 2014;9(8):e104758. [25] Saposnik G, Baibergenova A, O'Donnell M, Hill M, Kapral M, Hachinski V. Hospital volume and stroke outcome: does it matter? Neurology 2007;69(11):1142–51. [26] Kim H, Harrington C, Greene WH. Registered nurse staffing mix and quality of care in nursing homes: a longitudinal analysis. Gerontologist 2009;49(1):81–90. [27] Bostick JE, Rantz MJ, Flesner MK, Riggs CJ. Systematic review of studies of staffing and quality in nursing homes. J Am Med Dir Assoc 2006;7(6):366–76. [28] Needleman J, Buerhaus P, Mattke S, Stewart M, Zelevinsky K. Nurse-staffing levels and the quality of care in hospitals. N Engl J Med 2002;346(22):1715–22. [29] Thomas JW, Holloway JJ. Investigating early readmission as an indicator for quality of care studies. Med Care 1991;29(4):377–94. [30] Ashton CM, Del Junco DJ, Souchek J, Wray NP, Mansyur CL. The association between the quality of inpatient care and early readmission: a meta-analysis of the evidence. Med Care 1997;1044–59. [31] Goldstein LB, Samsa GP, Matchar DB, Horner RD. Charlson Index comorbidity adjustment for ischemic stroke outcome studies. Stroke 2004;35(8):1941–5.
