Impact of hospital volume on mortality and cost in sepsis based on a Japanese nationwide medical claims database study—the Japan Sepsis Alliance study group

doi:10.21203/rs.3.rs-1964450/v1

Download PDF

Research Article

Impact of hospital volume on mortality and cost in sepsis based on a Japanese nationwide medical claims database study—the Japan Sepsis Alliance study group

https://doi.org/10.21203/rs.3.rs-1964450/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background:

The impact of hospital volume on the mortality rate and medical costs in sepsis has not been fully elucidated. We hypothesized that high-volume hospitals have low mortality rates and high medical costs in sepsis cases.

Methods:

This retrospective study used the administrative data from 2010 to 2017. The hospital volume categorization into quartiles was performed according to the annual number of sepsis cases. The primary outcome was in-hospital mortality. The secondary outcomes were medical costs and length of hospital stay. A mixed-effects logistic model with a two-level hierarchical structure was used to adjust for baseline imbalances. Fractional polynomials and restricted cubic splines were investigated to determine the significance of the association between the hospital volume and clinical outcomes. Subgroup and sensitivity analyses were performed for the primary outcome.

Results:

Among 1,781,014 sepsis patients from 1,622 hospitals, the crude in-hospital mortality rate was 22.1% and 17.0% in the lowest and highest quartile of sepsis volume, respectively (p < 0.0001). After adjustment for confounding factors, the in-hospital mortality in the highest quartile was significantly lower than that of the lowest quartile (odds ratio, 0.789; 95% confidence interval, 0.774–0.804; p < 0.0001). Investigations with fractional polynomials and restricted cubic splines revealed that the hospital volume was significantly associated with in-hospital mortality (p < 0.0001). The highest quartile had higher daily medical costs per person and shorter length of hospital stay than those of the lowest quartile. Subgroup analyses showed that the high-volume hospitals with patients subjected to intensive care unit admission, mechanical ventilation, vasopressor therapy, and renal replacement therapy had a significantly low in-hospital mortality (p < 0.0001). The sensitivity analysis, excluding patients who were transferred to other hospitals, demonstrated a result consistent with that of the primary test.

Conclusions:

This nationwide sepsis study using the medical claims database suggested that a higher hospital volume leads to lower in-hospital mortality and higher daily medical costs per person. Further research is needed to investigate the mechanism of the case volume–outcome relationship in sepsis cases.

hospital volume

sepsis

diagnosis procedure combination

medical cost

critical care

Sepsis affects millions of patients due to the associated lethality, persistent organ dysfunction, and impaired quality of life, making it a public concern worldwide [1, 2]. Although over 100 clinical trials could not prove the efficacy of drug therapies in sepsis [3], an integrated approach, such as the implementation of sepsis management bundles and performance improvement programs, could be a promising solution [4, 5]. As high-volume hospitals have several advantages, including abundant medical resources and high capability in implementing protocol-based care, elucidating the impact of the hospital volume could help formulate efficient strategies to improve patient outcomes [6–9].

In the field of surgery and trauma, high hospital volume is associated with decreased mortality rate, low frequency of surgical complications, and reduced length of hospital stay [10–14]. In accordance with these benefits, centralization of patients has been regarded as an essential policy for achieving positive outcomes in specific situations. With respect to sepsis cases, previous reports showed a significant association between high-volume hospitals and an increased survival rate [15, 16]. However, two studies on severe sepsis patients contradicted these results [17, 18]. Therefore, the impact of hospital volume on clinical outcomes in sepsis cases remains unclear.

Besides the debatable impact of hospital volume on mortality in sepsis, previous studies investigating this association have rarely focused on medical costs. While the case volume–medical costs relationship has been presented in the fields of surgery, trauma, and burn injury [19–22], only few studies have clarified the association between the sepsis case volumes and economic outcomes [16]. Since the treatment intensity and amount of healthcare resources [23, 24] substantially affect the medical expenses, the sepsis case volume could affect both the medical costs and mortality rate. Therefore, using a Japanese nationwide medical claims database, we tested the hypothesis that high-volume hospitals have a low mortality rate but high medical costs in sepsis patients.

Study setting and patient data

This retrospective study used the Japanese nationwide medical claims database known as the Diagnosis Procedure Combination (DPC) [25]. With the introduction of this system, clinical outcomes, medical costs, and patient information could be obtained efficiently. DPC comprises the diagnostic classification and medical procedures, which enable us to collect patient demographics and economic outcomes from the database. As described in a previous report [26, 27], sepsis patients were screened based on a combined diagnosis of presumed serious infection and organ dysfunction. Presumed serious infection was defined by four consecutive days of antibiotic administration with blood culture drawn 48 h before or after the first dose of antibiotics. Since laboratory data were not recorded retrospectively in this database, organ dysfunction was determined based on other integrated conditions such as vasopressor use, oxygen support, ventilation support, renal replacement therapy (RRT), or diagnostic codes related to kidney dysfunction, hepatic disorder, thrombocytopenia/coagulopathy, or metabolic acidosis.

This study was approved by the Institutional Review Board of Chiba University Graduate School of Medicine (approval number: 3429). The need for obtaining written consent from the patients was waived due to the retrospective nature of this study in accordance with the Ethical Guidelines for Medical and Health Research Involving Human Subjects in Japan.

Data collection

The following patient demographic data were collected from the database: age, sex, chronic diseases (malignant tumor, hypertension, diabetes mellitus, heart failure, cerebrovascular disease, ischemic heart disease, chronic respiratory disease, or chronic renal failure), medical procedures, medical costs, intensive care unit (ICU) admission, length of hospital stay, and site of infection (respiratory, abdominal, urogenital, skin and soft tissue, neurological, blood, heart, or unknown). Admission diagnosis, comorbidities, or complications during the hospital stay were recorded as codes based on the International Statistical Classification of Diseases and Related Health Problems 10th revision (Additional File 1: Table S1). Patients with multiple codes for the “site of infection” parameter were categorized as “multiple,” whereas those with missing data (n = 765,460) were categorized as “unknown.” Repeat admissions were excluded from the analysis. Patients whose cultures or antibiotics were initiated within 48 h of hospital admission were defined as having community-acquired sepsis.

Categories of hospital volume

The hospital volume was estimated by the annual number of sepsis cases. Previous reports regarded the hospital volume as continuous or categorical variables according to the number of sepsis patients [15, 16]. In this study, we used both continuous and categorical variables to analyze the hospital volume. Hospital volume categorization into quartiles was performed according to the annual number of sepsis cases.

Medical costs

The representative value of medical costs comprises the total medical fees for drugs, examinations, laboratory tests, radiological tests, and surgical procedures. Reimbursement of medical fees was mainly determined based on the most common payment diagnosis and medical procedures performed during the hospital stay on the bundled payment basis; however, we calculated the total fees according to the reference prices in the Japanese fee schedule, as described in a previous report [28]. Considering the different admission years of the patients included in the study, the value of medical costs was standardized by the consumer index price in each year. After normalization, the price was converted from Japanese yen to U.S. dollars as of February 3, 2022 (115.25 yen = $1 USD).

Statistical analysis

In this study, the primary outcome was the in-hospital mortality according to the hospital volume or the annual number of sepsis patients. The secondary outcomes were medical costs and length of hospital stay. We used a mixed-effects logistic model with a two-level hierarchical structure, including hospital-level clustering, to adjust for the baseline imbalances. Potential confounders included age, sex, site of infection, community-acquired sepsis, chronic diseases, number of organ dysfunctions, ICU admission, admission year, number of hospital beds, and academic affiliation. As the severity scores using laboratory tests were not available in our study, we used the number of organ dysfunctions to estimate the extent of the severity [29]. The average number of hospitalized patients per institution was used as a substitute value representing the number of hospital beds.

In the analysis using the number of admitted patients as a continuous variable, fractional polynomials were used to determine the significance of the association between the hospital volume and clinical outcomes. Furthermore, restricted cubic splines (4 knots) were used to examine the correlation in the case of the non-linear association between the number of admitted patients and the mortality, length of hospital stay, and medical costs. For sensitivity analysis, we analyzed the cohort after excluding patients who were transferred to other hospitals. Because critically ill patients are likely to be transferred from low-volume hospitals to high-volume hospitals, the transferred patients could potentially cause a bias in the clinical outcomes. A robust method is needed to track the detailed information regarding patient transfer; however, in this study, we were unable to determine whether the patients were transferred to high- or low-volume hospitals. We also conducted subgroup analyses in terms of patients subjected to ICU admission, mechanical ventilation, vasopressor therapy, and RRT to validate the results in a specific population.

Categorical variables are expressed as numbers and percentages. These values were analyzed using Pearson’s chi-square test. Continuous variables are expressed as means (standard deviation) or medians (interquartile ranges [IQR]), as appropriate. Student’s t-test or Mann–Whitney U test was used to analyze the continuous variables according to the normality of distribution. For comparisons between more than two groups, one-way analysis of variance was performed to estimate the difference. Data were manipulated and analyzed using SQL (mariadb v10.4.17), R version 4.1.2 (R Foundation for Statistical Computing, Vienna, Austria, http://www.R-rojec t.org/), Stata version 17.0 (StataCorp LLC), and pandas (v1.0.5), scipy (v1.7.3), numpy (v1.21.4), seaborn (v0.11.2), matplotlib (v3.5.1), and statsmodels (v0.13.2) in Python (v3.9.0).

Baseline patient characteristics

During our study period, among the 50,490,128 registered patients on the DPC system from 1,622 hospitals, the data of 1,781,014 sepsis patients were included in this study (Additional file 2: Fig. S1). According to the annual number of admitted patients per institution, the hospital volumes were categorized into four quartiles in the ascending order as follows: Q1, Q2, Q3, and Q4 (Fig. 1). There were significant differences in age, sex, chronic diseases, percentage of community-acquired sepsis, and infection sites among the four quartiles. The most frequent infection site was the respiratory tract in all quartiles. The proportion of ICU admissions was 10.2% in Q1, 18.0% in Q2, 21.6% in Q3, and 23.6% in Q4 (p < 0.0001). Q4 showed a high proportion of organ dysfunctions and organ support, i.e., patients receiving vasopressor therapy, oxygen therapy, mechanical ventilation, kidney dysfunction, or RRT or those with hepatic disorder, thrombocytopenia/coagulopathy, or acidosis (p < 0.0001) (Table 1, Additional file 1: Table S2).

Table 1

Baseline characteristics of the patients in the cohort^a
	Total (n = 1,781,014)	Hospital volume Q1 (n = 446,279)	Hospital volume Q2 (n = 445,962)	Hospital volume Q3 (n = 444,921)	Hospital volume Q4 (n = 443,852)	p-value
Sepsis patients (n/year)	306 (163–474)	94 (58–129)	233 (203–272)	392 (344–425)	653 (541–795)	< 0.0001
Age, years	76 (66–84)	79 (69–86)	76 (67–84)	75 (65–83)	74 (64–82)	< 0.0001
Male, n (%)	1,037,354 (58.2)	248,621 (55.9)	258,112 (58.2)	264,381 (59.3)	266,240 (59.7)	< 0.0001
Chronic diseases
Cancer, n (%)	597,513 (33.5)	128,314 (28.8)	154,219 (34.7)	158,599 (35.6)	156,381 (35.0)	< 0.0001
Hypertension, n (%)	475,158 (26.7)	121560 (27.3)	115,903 (26.1)	116,295 (26.1)	121,400 (27.2)	< 0.0001
Diabetes mellitus, n (%)	385,810 (21.7)	97,050 (21.8)	97,431 (22.0)	95,636 (21.4)	95,693 (21.4)	< 0.0001
Heart failure, n (%)	327,401 (18.4)	86,912 (19.5)	82,820 (18.7)	77,777 (17.4)	79,892 (17.9)	< 0.0001
Stroke, n (%)	262,220 (14.7)	75,904 (17.1)	63,577 (14.3)	61,228 (13.7)	61,511 (13.8)	< 0.0001
Ischemic heart disease, n (%)	193,081 (10.8)	44,484 (10.0)	47,203 (10.6)	49,962 (11.2)	51,432 (11.5)	< 0.0001
Chronic respiratory disease, n (%)	148,480 (8.3)	38,882 (8.7)	37,915 (8.5)	35,606 (8.0)	36,077 (8.1)	< 0.0001
Chronic renal failure, n (%)	67,690 (3.8)	16,417 (3.7)	16,570 (3.7)	15,589 (3.5)	19,114 (4.3)	< 0.0001
Community-acquired sepsis, n (%)	964,893 (54.2)	256,523 (57.7)	242,648 (54.7)	235,981 (52.9)	229,741 (51.5)	< 0.0001
Infection site						< 0.0001
Respiratory, n (%)	387,995 (21.8)	101,366 (22.8)	95,725 (21.6)	93,776 (21.0)	91,487 (20.5)
Abdominal, n (%)	152,424 (8.6)	36,688 (8.2)	39,733 (9.0)	39,902 (8.9)	39,146 (8.8)
Urogenital, n (%)	75,313 (4.2)	20,257 (4.6)	17,746 (4.0)	17,576 (3.9)	17,223 (3.9)
Bone and soft tissue, n (%)	41,030 (2.3)	10,531 (2.4)	10,118 (2.3)	9,992 (2.2)	9,976 (2.2)
Meninges/brain/ spinal cord, n (%)	13,368 (0.8)	2,677 (0.6)	3,051 (0.7)	3,720 (0.8)	4,294 (1.0)
Heart, n (%)	7,580 (0.4)	1,602 (0.4)	1,993 (0.4)	2,169 (0.5)	2,207 (0.5)
Blood, n (%)	684 (0.04)	117 (0.03)	158 (0.04)	265 (0.06)	185 (0.04)
Multiple, n (%)	318,832 (17.9)	80,946 (18.2)	80,050 (18.0)	77,657 (17.4)	79,283 (17.8)
Unknown, n (%)	784,857 (44.0)	190,737 (42.9)	195,278 (44.0)	200,905 (45.0)	202,478 (45.4)
ICU admission, n (%)	327,389 (18.4)	45,499 (10.2)	80,076 (18.0)	96,319 (21.6)	105,495 (23.6)	< 0.0001
Vasopressor therapy, n (%)	232,163 (13.0)	53,719 (12.1)	59,696 (13.4)	56,712 (12.7)	62,036 (13.9)	< 0.0001
Oxygen therapy, n (%)	1,450,421 (81.4)	363,528 (81.7)	360,452 (81.2)	361,395 (81.0)	365,046 (81.8)	< 0.0001
Mechanical ventilator, n (%)	332,619 (18.7)	67,980 (15.3)	81,834 (18.4)	88,315 (19.8)	94,490 (21.2)	< 0.0001
Kidney dysfunction, n (%)	832,784 (46.8)	207,781 (46.7)	209,648 (47.2)	210,641 (47.2)	204,714 (45.9)	< 0.0001
RRT, n (%)	124,602 (7.0)	25,264 (5.7)	30,389 (6.8)	33,316 (7.5)	35,633 (8.0)	< 0.0001
Hepatic disorder, n (%)	67,456 (3.8)	17,531 (3.9)	16,575 (3.7)	16,595 (3.7)	16,755 (3.8)	< 0.0001
Thrombocytopenia/ coagulopathy, n (%)	198,927 (11.2)	48,160 (10.8)	49,698 (11.2)	50,543 (11.3)	50,526 (11.3)	< 0.0001
Acidosis, n (%)	15,592 (0.8)	2,846 (0.6)	4,103 (0.9)	3,489 (0.8)	3,561 (0.8)	< 0.0001
Number of organ dysfunctions						< 0.0001
Mean (standard deviation)	1.60 (0.77)	1.58 (0.76)	1.60 (0.77)	1.60 (0.77)	1.60 (0.78)
Median (interquartile range)	1 (1–2)	1 (1–2)	1 (1–2)	1 (1–2)	1 (1–2)
IQR, interquartile range; ICU, intensive care unit; RRT, renal replacement therapy
^aQ1: 0–162 patients per year; Q2: 163–305 patients per year; Q3: 306–474 patients per year; and Q4: 475–1,644 patients per year.

Clinical outcomes and medical costs

In-hospital mortality was 18.9% in the cohort, with the highest mortality recorded in Q1 (22.1%) and lowest in Q4 (17.0%) (p < 0.0001). The length of hospital stay was shorter in Q4 (28 [15–52]) than in Q1 (32 [16–62]). Gross medical costs were higher in Q4 ($1.03×10¹⁰) than in the other quartiles ($9.14×10⁹ in Q1, $9.45×10⁹ in Q2, and $9.92×10⁹ in Q3). The median cost per hospitalization (p < 0.0001) and median daily costs per person (p < 0.0001) increased proportionally with the hospital volume (Table 2).

Table 2

Clinical outcomes and medical costs in the cohort^a
	Total (n = 1,781,014)	Hospital volume Q1 (n = 446,279)	Hospital volume Q2 (n = 445,962)	Hospital volume Q3 (n = 444,921)	Hospital volume Q4 (n = 443,852)	p-value
In-hospital mortality, n (%)	363,363 (18.9)	98,248 (22.1)	84,919 (19.1)	77,724 (17.4)	75,706 (17.0)	< 0.0001
Ventilator-free days						< 0.0001
Mean (SD)	18.2 (10.2)	17.3 (10.8)	18.1 (10.2)	18.4 (10.0)	18.6 (9.8)
Median (IQR)	23 (11–27)	22 (6–27)	22 (11–27)	23 (12–26)	23 (13–26)
Vasopressor-free days						< 0.0001
Mean (SD)	23.6 (7.9)	22.7 (8.9)	23.7 (7.9)	23.7 (7.7)	24.2 (7.2)
Median (IQR)	27 (25–27)	27 (24–27)	27 (25–27)	27 (25–27)	27 (26–27)
Length of hospital stay (days)						< 0.0001
Mean (SD)	46.6 (56.0)	54.4 (130.8)	46.4 (133.9)	43.1 (75.6)	42.6 (92.0)
Median (IQR)	30 (16–56)	32 (16–62)	30 (16–56)	29 (16–53)	28 (15–52)
Length of ICU stay (days)						< 0.0001
Mean (SD)	6.9 (6.2)	6.7 (5.7)	6.7 (5.6)	6.8 (6.1)	7.4 (6.9)
Median (IQR)	5 (2–12)	5 (2–11)	5 (2–11)	5 (2–11)	5 (2–12)
Length of antibiotic treatment (days)						< 0.0001
Mean (SD)	18.6 (21.0)	18.8 (21.4)	18.2 (20.4)	18.5 (20.9)	18.9 (21.5)
Median (IQR)	12 (8–21)	12 (8–22)	12 (8–21)	12 (8–21)	12 (8–21)
Inter-hospital transfer, % (n)	20.7 (369,182)	18.2 (80,994)	19.8 (87,895)	22.2 (99,065)	22.7 (101,228)	< 0.0001
Gross medical costs ($)	3.88×10¹⁰	9.14×10⁹	9.45×10⁹	9.92×10⁹	1.03×10¹⁰	NA
Medical costs per hospitalization ($)						< 0.0001
Mean (SD)	21,796 (31,215)	20,547 (27,128)	21,285 (38,448)	22,236 (29,135)	23,110 (28,857)
Median (IQR)	13,472 (6,940 – 26,099)	12,686 (6,484–24,443)	13,194 (6,783–25,525)	13,703 (7,090 − 26,702)	14,323 (7,459–27,848)
Daily medical costs per person ($)						< 0.0001
Mean (SD)	531 (457)	448 (343)	517 (485)	561 (504)	598 (466)
Median (IQR)	416 (328–580)	366 (296–482)	408 (325–563)	436 (344–618)	465 (362–662)
SD, standard deviation; IQR, interquartile range; ICU, intensive care unit; NA, not applicable
^aQ1: 0–162 patients per year; Q2: 163–305 patients per year; Q3: 306–474 patients per year; and Q4: 475–1,644 patients per year.

Regression analysis

After adjustment for baseline imbalances, the highest hospital volume was significantly associated with reduced in-hospital mortality as compared to the lowest hospital volume (odds ratio [OR] 0.862, 95% confidence interval [CI] 0.850–0.875; p < 0.0001). Additionally, in-hospital mortality in Q2 and Q3 was lower than that in Q1 (OR 0.816, 95% CI 0.802–0.830; p < 0.0001 and OR 0.789, 95% CI 0.774–0.804; p < 0.0001, respectively) (Table 3).

Table 3

Multivariable regression analysis for in-hospital mortality^a,b
Variable	Odds ratio	95% CI	p-value
Hospital volume categories
Q1	1.000	Reference
Q2	0.862	0.850–0.875	< 0.0001
Q3	0.816	0.802–0.830	< 0.0001
Q4	0.789	0.774–0.804	< 0.0001
CI, confidence interval; IQR, interquartile range
^aQ1: 0–162 patients per year; Q2: 163–305 patients per year; Q3: 306–474 patients per year; and Q4: 475–1,644 patients per year.
^bThe independent variable was adjusted for age, sex, site of infection, community-acquired sepsis, chronic diseases, number of organ dysfunctions, intensive care unit admission, admission year, number of hospital beds, and academic affiliation.

Fractional polynomials revealed that the annual number of patients per institution was significantly associated with reduced in-hospital mortality (p < 0.0001). While the length of hospital stay decreased in proportion to the sepsis case volume (p < 0.0001), the daily medical costs per person significantly increased with increase in the case volume (p < 0.0001) (Fig. 2). Restricted cubic splines showed similar correlation patterns as that of fractional polynomials between the annual number of patients per institution and the daily medical costs per person, length of hospital stay, or in-hospital mortality (Additional file 3: Fig. S2).

Subgroup analysis

In the subpopulation subjected to ICU admission, mechanical ventilation, vasopressor therapy, and RRT, the annual number of sepsis patients was significantly associated with reduced in-hospital mortality (p < 0.0001) (Fig. 3).

Sensitivity analysis

The proportion of inter-hospital transfers was the highest in Q4 (22.7%; p < 0.0001). After excluding patients who were transferred to other hospitals, Q4 still showed lower in-hospital mortality than that of Q1 (OR 0.828, 95% CI 0.811–0.845; p < 0.0001), which was consistent with the primary test results. Moreover, the fractional polynomials showed that the annual sepsis case volume was significantly associated with reduced in-hospital mortality (Table 2, Additional file 1: Table S3, Additional file 4: Fig. S3).

Using the Japanese nationwide medical claims database, this study demonstrated that a high hospital volume significantly contributed to decreasing the in-hospital mortality in sepsis cases. Moreover, higher volume hospitals were associated with higher daily medical costs per person in sepsis.

Several studies have investigated the relationship between the hospital volume and mortality in sepsis patients. While some studies revealed a beneficial effect of the hospital volume on mortality in sepsis [15, 16], other investigations showed contradictory results [17, 18]. The disparities among the studies could be attributed to the differences in the study design, number of enrolled patients, adjusted variables, and medical systems in each country. An insufficient number of study participants and an arbitrary threshold of high-volume hospitals would lead to underpowered studies to detect valid results. Our study clearly demonstrated that high-volume hospitals and the number of hospitalized patients significantly correlated with reduced in-hospital mortality in sepsis patients. The strength of our study is its large cohort data of 1,781,014 patients, which outnumbered that of the previous studies. Additionally, we used various statistical methods to enhance the robustness of our analysis. While the investigations using quartile groups of hospital volume presented plausible results, application of fractional polynomials and restricted cubic splines with the hospital volume as a continuous variable compensated for the limitation of the arbitrary categorization.

The mechanism of the beneficial effect of high-volume hospitals on mortality is potentially explained by several components, including abundant medical resources, more experienced medical staff, and higher capability of implementing evidence-based practices [30–32]. In general, high-volume hospitals have more ICU staff, beds, and medical equipment, such as mechanical ventilators and blood purification devices. It seems reasonable that the availability of abundant medical resources with more experienced staff contributes to improving the clinical outcomes of sepsis patients. On the other hand, considering that patients subjected to mechanical ventilation, RRT, or ICU admission comprised less than 20% of the cohort, these components may not substantially affect the benefits of high-volume hospitals. The consistent benefit of high-volume hospitals despite the confounder adjustment for ICU admission and organ dysfunctions implies that the other elements contributed to the advantages. As the intensity of physician staffing and compliance with implementing guidelines-based treatment were not investigated in our study, detailed mechanisms of the effect of hospital volume on mortality should be clarified in future investigations.

Although we demonstrated higher daily medical costs per person and shorter length of stay in association with higher sepsis case volume, no associations were reported in a previous publication [16]. Besides sepsis, overall cancer surgery costs were inversely associated with the hospital volume as well as surgeon volume through a shorter length of stay [19, 20]. In these reports, abundant experience of the staff was suggested as the possible reason for the shorter length of hospitalization in high-volume hospitals than in the low-volume ones. Similarly, Endo et al. showed better survival and lower total cost per admission in high-volume hospitals for severe trauma patients using an administrative database [22]. In contrast, a previous study enrolling burn injury patients demonstrated a positive correlation between the hospital volume and medical costs during hospitalization [21]. The plausible reason for the increased medical costs was the high proportion of intensive treatments performed in high-volume hospitals, such as mechanical ventilation and cultured skin grafts. This study demonstrated that not only daily medical costs per person but also medical costs per hospitalization increased according to the number of annual sepsis patients, despite a shorter length of stay. As the proportion of patients on artificial organ support is greater in high-volume hospitals than in low-volume hospitals, the proportional association between the hospital volume and medical costs would be conceivable.

Hospital volume was shown to be a crucial factor in improving in-hospital mortality in this study; however, centralization of sepsis patients requires deep consideration and meticulous debate prior to policy implementation. While the acceleration of the centralization improved the survival rate after pancreatic cancer surgery [33], it might not be appropriate to apply such a policy to sepsis management without careful deliberation. Since sepsis patients require a time-sensitive approach with early recognition and immediate initiation of treatment, transportation of sepsis patients could reduce the treatment effect due to delayed antibiotic administration and the introduction of sepsis management bundles [34, 35]. Therefore, investigation of the detailed mechanisms responsible for better clinical outcomes in high-volume hospitals should be conducted rather than encouraging the transportation of sepsis patients to high-volume hospitals.

This study has several limitations that must be addressed prior to drawing conclusions. First, the administrative database did not include laboratory data to calculate the severity scores. Second, treatment policy such as withholding or withdrawal of life-sustaining treatment was not collected from the database, which could be a confounding factor affecting the results. Third, categorization into quartiles might lead to a bias in determining the effect of hospital volume on outcomes. However, in our study, the complementary analysis using hospital volume as a continuous variable potentially increased the robustness. Fourth, the detailed proportions of the medical costs were not described in this study. Fifth, long-term mortality rate and quality of life were not investigated in the cohort. Sixth, plausible reasons for the association between the hospital volume and the mortality or medical costs were not sufficiently explained. Further investigations are warranted to clarify the mechanisms underlying these intimate relationships.

This study, using a Japanese nationwide medical claims database, suggests that higher volume hospitals are likely to have better survival and higher daily medical costs per person in sepsis cases as compared to the lower volume hospitals. Further research is warranted to validate these results and clarify the mechanisms underlying these associations.

DPC, diagnosis procedure combination; RRT, renal replacement therapy; ICU, intensive care unit; IQR, interquartile range; OR, odds ratio; CI, confidence interval.

Ethics approval and consent to participate

This study was approved by the Institutional Review Board of Chiba University Graduate School of Medicine. The Review Board waived the need for written informed consent from the enrolled patients or their surrogates.

Consent for publication

Not applicable

Availability of data and materials

The datasets used and analyzed in this study will be available from the corresponding author upon reasonable request.

Competing interests

The authors have no conflicts of interest to declare.

Funding

TN is the CEO of Smart119 Inc. and owns stock. YY owns stock of Smart119 Inc. Smart119 Inc. had no role in the study design, data analysis, or preparation of the manuscript. Other authors received no specific funding for this work.

Authors’ contribution

TO, TI, TN, and NT are responsible for the study concept and design, statistical analysis and interpretation of data, drafting of the manuscript, and critical revision of the manuscript for important intellectual content. TA performed statistical analysis and data interpretation. YY performed the computing process to extract the necessary data, and KF performed the data acquisition. All the other authors interpreted the data and critically revised the article for important intellectual content. All the authors have read and approved the final manuscript.

Acknowledgements

We would like to thank all the contributors of the Japan Sepsis Alliance (JaSA) study group. The study group is a joint research organization, which consisted of the Japanese Society of Intensive Care Medicine, the Japanese Association of Emergency Medicine, and the Japanese Society of Infectious Diseases, holds sepsis seminars and public lectures calling for the prevention, early detection and early treatment of sepsis in Japan.

Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, Bellomo R, Bernard GR, Chiche JD, Coopersmith CM et al: The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016, 315(8):801-810.
Rudd KE, Johnson SC, Agesa KM, Shackelford KA, Tsoi D, Kievlan DR, Colombara DV, Ikuta KS, Kissoon N, Finfer S et al: Global, regional, and national sepsis incidence and mortality, 1990-2017: analysis for the Global Burden of Disease Study. Lancet 2020, 395(10219):200-211.
Marshall JC: Why have clinical trials in sepsis failed? Trends Mol Med 2014, 20(4):195-203.
Evans L, Rhodes A, Alhazzani W, Antonelli M, Coopersmith CM, French C, Machado FR, McIntyre L, Ostermann M, Prescott HC et al: Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock 2021. Crit Care Med 2021, 49(11):e1063-e1143.
Seymour CW, Gesten F, Prescott HC, Friedrich ME, Iwashyna TJ, Phillips GS, Lemeshow S, Osborn T, Terry KM, Levy MM: Time to Treatment and Mortality during Mandated Emergency Care for Sepsis. N Engl J Med 2017, 376(23):2235-2244.
Luft HS, Bunker JP, Enthoven AC: Should operations be regionalized? The empirical relation between surgical volume and mortality. N Engl J Med 1979, 301(25):1364-1369.
Halm EA, Lee C, Chassin MR: Is volume related to outcome in health care? A systematic review and methodologic critique of the literature. Ann Intern Med 2002, 137(6):511-520.
Lee KC, Sethuraman K, Yong J: On the Hospital Volume and Outcome Relationship: Does Specialization Matter More Than Volume? Health Serv Res 2015, 50(6):2019-2036.
Sewalt CA, Wiegers EJA, Venema E, Lecky FE, Schuit SCE, Den Hartog D, Lingsma HF: The volume-outcome relationship in severely injured patients: A systematic review and meta-analysis. J Trauma Acute Care Surg 2018, 85(4):810-819.
Birkmeyer JD, Siewers AE, Finlayson EV, Stukel TA, Lucas FL, Batista I, Welch HG, Wennberg DE: Hospital volume and surgical mortality in the United States. N Engl J Med 2002, 346(15):1128-1137.
Ikoma N, Kim B, Elting LS, Shih YT, Badgwell BD, Mansfield P: Trends in Volume-Outcome Relationship in Gastrectomies in Texas. Ann Surg Oncol 2019, 26(9):2694-2702.
O'Donnell TFX, Boitano LT, Deery SE, Lancaster RT, Siracuse JJ, Schermerhorn ML, Scali ST, Patel VI: Hospital Volume Matters: The Volume-Outcome Relationship in Open Juxtarenal AAA Repair. Ann Surg 2020, 271(1):184-190.
Alsfasser G, Leicht H, Gunster C, Rau BM, Schillinger G, Klar E: Volume-outcome relationship in pancreatic surgery. Br J Surg 2016, 103(1):136-143.
Wright JD: The Volume-Outcome Paradigm for Gynecologic Surgery: Clinical and Policy Implications. Clin Obstet Gynecol 2020, 63(2):252-265.
Gaieski DF, Edwards JM, Kallan MJ, Mikkelsen ME, Goyal M, Carr BG: The relationship between hospital volume and mortality in severe sepsis. Am J Respir Crit Care Med 2014, 190(6):665-674.
Walkey AJ, Wiener RS: Hospital case volume and outcomes among patients hospitalized with severe sepsis. Am J Respir Crit Care Med 2014, 189(5):548-555.
Shahin J, Harrison DA, Rowan KM: Relation between volume and outcome for patients with severe sepsis in United Kingdom: retrospective cohort study. BMJ 2012, 344:e3394.
Goodwin AJ, Simpson KN, Ford DW: Volume-Mortality Relationships during Hospitalization with Severe Sepsis Exist Only at Low Case Volumes. Ann Am Thorac Soc 2015, 12(8):1177-1184.
Ho V, Aloia T: Hospital volume, surgeon volume, and patient costs for cancer surgery. Med Care 2008, 46(7):718-725.
Shi HY, Chang HT, Culbertson R, Chen YJ, Liao YC, Hou MF: Breast cancer surgery volume-cost associations: hierarchical linear regression and propensity score matching analysis in a nationwide Taiwan population. Surg Oncol 2013, 22(3):178-183.
Endo A, Shiraishi A, Otomo Y, Fushimi K, Murata K: Volume-outcome relationship on survival and cost benefits in severe burn injury: a retrospective analysis of a Japanese nationwide administrative database. J Intensive Care 2019, 7:7.
Endo A, Shiraishi A, Fushimi K, Murata K, Otomo Y: Increased Severe Trauma Patient Volume is Associated With Survival Benefit and Reduced Total Health Care Costs: A Retrospective Observational Study Using a Japanese Nationwide Administrative Database. Ann Surg 2018, 268(6):1091-1096.
Chalupka AN, Talmor D: The economics of sepsis. Crit Care Clin 2012, 28(1):57-76, vi.
Pittoni GM, Scatto A: Economics and outcome in the intensive care unit. Curr Opin Anaesthesiol 2009, 22(2):232-236.
Hayashida K, Murakami G, Matsuda S, Fushimi K: History and Profile of Diagnosis Procedure Combination (DPC): Development of a Real Data Collection System for Acute Inpatient Care in Japan. J Epidemiol 2021, 31(1):1-11.
Imaeda T, Nakada TA, Takahashi N, Yamao Y, Nakagawa S, Ogura H, Shime N, Umemura Y, Matsushima A, Fushimi K: Trends in the incidence and outcome of sepsis using data from a Japanese nationwide medical claims database-the Japan Sepsis Alliance (JaSA) study group. Crit Care 2021, 25(1):338.
Rhee C, Zhang Z, Kadri SS, Murphy DJ, Martin GS, Overton E, Seymour CW, Angus DC, Dantes R, Epstein L et al: Sepsis Surveillance Using Adult Sepsis Events Simplified eSOFA Criteria Versus Sepsis-3 Sequential Organ Failure Assessment Criteria. Crit Care Med 2019, 47(3):307-314.
Oami T, Imaeda T, Nakada TA, Abe T, Takahashi N, Yamao Y, Nakagawa S, Ogura H, Shime N, Umemura Y et al: Temporal trends of medical cost and cost-effectiveness in sepsis patients: a Japanese nationwide medical claims database. J Intensive Care 2022, 10(1):33.
Sasabuchi Y, Matsui H, Lefor AK, Fushimi K, Yasunaga H: Risks and Benefits of Stress Ulcer Prophylaxis for Patients With Severe Sepsis. Crit Care Med 2016, 44(7):e464-469.
Keeler EB, Rubenstein LV, Kahn KL, Draper D, Harrison ER, McGinty MJ, Rogers WH, Brook RH: Hospital characteristics and quality of care. JAMA 1992, 268(13):1709-1714.
Joynt KE, Harris Y, Orav EJ, Jha AK: Quality of care and patient outcomes in critical access rural hospitals. JAMA 2011, 306(1):45-52.
Fawzy A, Walkey AJ: Association Between Hospital Case Volume of Sepsis, Adherence to Evidence-Based Processes of Care and Patient Outcomes. Crit Care Med 2017, 45(6):980-988.
Lemmens VE, Bosscha K, van der Schelling G, Brenninkmeijer S, Coebergh JW, de Hingh IH: Improving outcome for patients with pancreatic cancer through centralization. Br J Surg 2011, 98(10):1455-1462.
Mohr NM, Harland KK, Shane DM, Ahmed A, Fuller BM, Torner JC: Inter-hospital transfer is associated with increased mortality and costs in severe sepsis and septic shock: An instrumental variables approach. J Crit Care 2016, 36:187-194.
Faine BA, Noack JM, Wong T, Messerly JT, Ahmed A, Fuller BM, Mohr NM: Interhospital Transfer Delays Appropriate Treatment for Patients With Severe Sepsis and Septic Shock: A Retrospective Cohort Study. Crit Care Med 2015, 43(12):2589-2596.

No competing interests reported.

Additionalfile1.docx
Additional file 1:Table S1. Diagnostic categories with corresponding International Statistical Classification of Diseases and Related Health Problems 10th revision codesTable S2. Characteristics of hospitals across quartiles of annual sepsis volumeTable S3. Multivariable regression analysis for in-hospital mortality after exclusion of transferred patients
AdditionalFile2.pdf
Additional file 2: Figure S1. Flow chart of the study population
AdditionalFile3.pdf
Additional file 3: Figure S2. Association between the annual number of sepsis patients and in-hospital mortality using restricted cubic splines
AdditionalFile4.pdf
Additional file 4: Figure S3. Subgroup analyses of the impact of hospital volume on the in-hospital mortality in the cohort after excluding transferred patients

Download PDF

Version 1

posted

You are reading this latest preprint version

Impact of hospital volume on mortality and cost in sepsis based on a Japanese nationwide medical claims database study—the Japan Sepsis Alliance study group

Status:

Version 1

Abstract

Background:

Methods:

Results:

Conclusions:

Figures

Background

Methods

Study setting and patient data

Data collection

Categories of hospital volume

Medical costs

Statistical analysis

Results

Baseline patient characteristics

Clinical outcomes and medical costs

Regression analysis

Subgroup analysis

Sensitivity analysis

Discussion

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1