ABSTRACT
Aim
The coronavirus disease 2019 (COVID‐19) pandemic has profoundly affected older populations globally. However, limited data are available on the long‐term survival and evolving care needs of older patients with COVID‐19 in Japan. We evaluated 1‐year survival rates and long‐term care needs in this vulnerable population.
Methods
This population‐based cohort study analyzed 2175 hospitalized patients with COVID‐19 aged ≥ 65 years using data from the Nara Kokuho Database between February 2020 and August 2022. The database contains administrative claims data from the National Health Insurance and Late Elders' Health Insurance systems, incorporating information on care needs levels. We examined 1‐year survival and changes in care needs levels.
Results
The overall 1‐year survival rate was 80.9%. Age significantly influenced survival, with rates of 93.6% for patients aged 65–69 years and 56.6% for those ≥ 90 years. Disease severity had a substantial impact: patients not requiring oxygen therapy had a survival rate of 87.5%, compared to 74.0% for those with supplemental oxygen and 49.9% for those receiving invasive mechanical ventilation. Despite a decline in severe cases following the Delta and Omicron‐predominant waves, survival among severely ill patients remained lower. While most patients without prior care needs retained independence, approximately 23% developed new care requirements after COVID‐19. Higher care needs levels were associated with significantly higher mortality, with patients at care needs level ≥ 2 experiencing mortality rates exceeding 35%.
Conclusion
This study highlights the importance of comprehensive management of the long‐term survival and care needs of older patients with COVID‐19.
Keywords: 1‐year survival, coronavirus disease 2019 (COVID‐19), Kokuho database, long‐term care needs, older patients
1. Introduction
The coronavirus disease 2019 (COVID‐19) pandemic, caused by severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2), has profoundly impacted global healthcare and economies since late 2019 [1, 2]. Older people are particularly vulnerable [2, 3, 4], experiencing higher rates of hospitalization, complications, and mortality worldwide [3, 4, 5]. In Japan, where the population is rapidly aging, the pandemic's impact on older adults has been of significant concern [6, 7]. Like many countries, Japan implemented policies recommending hospitalization for patients aged ≥ 65 years with COVID‐19 [6, 7, 8]. Managing care and healthcare resources for this group remains a critical challenge in their long‐term management [9, 10]. Understanding long‐term COVID‐19 outcomes is essential for both Japanese and global healthcare systems to support and prepare for the care needs of the older population. Long‐term survival outcomes, activities of daily living including functional and cognitive impairments, and quality of life are essential indicators for assessing long‐term care needs after COVID‐19. However, limited studies in Japan have evaluated long‐term survival rates and changes in care needs after COVID‐19.
This study evaluated the 1‐year survival rate and changes in care needs levels among hospitalized patients aged ≥ 65 years with COVID‐19 using data from the Nara Kokuho Database (KDB) and identified factors associated with long‐term prognosis. Patients' functional and cognitive impairments in Japan were prospectively assessed using the national standardized system of care requirements in the long‐term care insurance system and classified based on total estimated daily care hours.
2. Methods
2.1. Data Source
We conducted a real‐world population‐based cohort study using the Nara KDB, which contains administrative claims data from Nara Prefecture, Japan. In Japan, three primary public health insurance systems provide nearly universal coverage: Employees' Health Insurance (EHI), National Health Insurance (NHI), and Late Elders' Health Insurance (LEHI). Individuals aged ≥ 75 years are enrolled in LEHI, administered by each prefecture. The Nara KDB includes data from the NHI and LEHI but does not encompass the EHI. Therefore, the data from LEHI capture almost all Nara Prefecture residents aged ≥ 75 years. The Nara KDB covers approximately 500 000 residents; the LEHI contains data on approximately 200 000 older individuals. However, data on individuals aged between 65 and 74 years who are not enrolled in the NHI are not included. We merged the medical and long‐term care claims databases using a unique identifier for each individual. We used a patient‐matching technique to enable longitudinal follow‐up of patient outcomes [11].
2.2. Study Period and Population
We analyzed data from February 1, 2020 to August 31, 2022, incorporating a washout observation period of at least 6 months from March 1 to August 31, 2022 (Figure 1). The study included hospitalized patients aged ≥ 65 years with a first COVID‐19 diagnosis identified in the Nara KDB. For eligible patients, the diagnosis and admission dates were identified, and the index date was the date of admission. The dataset provided registrant information (age, sex, observation period, withdrawal from NHI or LEHI, and death dates), insurance claims (prescribed medications, procedures, and International Classification of Diseases, 10th revision [ICD]‐10 codes), and long‐term care insurance data, including care needs levels and services rendered. Claims data were updated monthly, and ICD‐10 codes were assigned for the duration of treatment for specific diseases.
FIGURE 1.
Schematic diagram of the study design. COVID‐19, coronavirus disease 2019.
2.3. COVID‐19 Diagnosis, Severity Classification, and Predominant SARS‐CoV‐2 Variant Wave
COVID‐19 diagnosis was determined using specific ICD‐10 codes (U07.1 [virologically confirmed cases], U07.2 [clinically diagnosed cases]; Table S1). Disease severity was classified as mild to moderate I (no oxygen therapy), moderate II (respiratory support with oxygen administration, high‐flow therapy, or noninvasive positive pressure ventilation), and severe (invasive mechanical ventilation or extracorporeal membrane oxygenation), following the Japanese Clinical Management of Patients with COVID‐19: A Guide for Front‐line Healthcare Workers, version 2.1 [12]. Information on respiratory support was identified through medical practice codes (Table S2) and obtained for the hospitalization following COVID‐19 diagnosis. For patients who received multiple types of respiratory support, the highest severity category was assigned. This procedure‐based classification approach, in which disease severity was defined by claims‐based identification of oxygen therapy and related interventions, has also been applied in previous studies of COVID‐19 [13], as well as in research on non‐COVID‐19 patients receiving mechanical ventilation [14]. The KDB does not contain information on the SARS‐CoV‐2 variants confirmed in each patient. The study period was divided into predominant virus variant pandemic waves: wild‐type (February 2020–February 2021), Alpha (March–June 2021), Delta (July–November 2021), and Omicron (December 2021–March 2022) [7, 13].
2.4. Comorbidities and Risk Factors
Comorbidities were categorized using ICD‐10 codes (Table S1) and a combination of confirmed disease‐related injury/illness codes and medications (Table S2). We selected general and specific comorbidities of interest based on previously identified risk factors for severe COVID‐19. These comorbidities included hypertension, dyslipidemia, chronic obstructive pulmonary disease, diabetes mellitus with chronic complications, peripheral arterial disease, renal disease, cerebrovascular disease, rheumatologic disease, and liver disease [13, 15, 16].
2.5. Care Needs Level
To receive long‐term care insurance services, the insured candidates or their caregivers must contact the municipal government to obtain a nationally standardized care needs certification [17]. Initially, a trained local government official visits the home or hospital to evaluate nursing care needs using a standardized questionnaire comprising 74 items. These include 20 items on physical function, 12 on daily activity function, 9 on cognitive function, 15 on behavior, 6 on social life adjustment, and 12 on daily use of medical services. The responses are then entered into a computer system, which estimates the time required for nine categories of care: grooming/bathing, eating, toileting, transferring, instrumental activities of daily living, behavioral problems in dementia, rehabilitation, and medical services. Based on the estimated daily care minutes, candidates are initially assigned one of seven care needs levels: Support Level 1 (25–31 min), Support Level 2 (32–49 min), Care Needs Level 1 (32–49 min), Care Needs Level 2 (50–69 min), Care Needs Level 3 (70–89 min), Care Needs Level 4 (90–109 min), and Care Needs Level 5 (≥ 110 min). The Nursing Care Needs Certification Board—comprising physicians, nurses, and other health and social services experts appointed by a mayor—reviews the appropriateness of the initial assessment while considering the assessor's notes and the applicant's primary care physician's statement. The board's decision is final, and the assigned care needs levels determine service benefits covered under long‐term care insurance. Care needs levels are reevaluated at 6‐ to 12‐month intervals. A previous study reported a strong correlation between care needs levels and activities of daily living as calculated using the Barthel index [18]. For instance, Support Levels 1–2 and Care Needs Level 1 corresponded to Barthel index scores of 85–95 (independent with minor assistance); Care Needs Levels 2–3 corresponded to scores of 65–80 (partial dependence); and Care Needs Levels 4–5 corresponded to scores below 40 (complete dependence) [18]. This study categorized preexisting care needs levels into seven groups: no care needs, Support Levels 1–2, and Care Needs Levels 1–5.
2.6. Outcome Measures
The primary outcomes were the 1‐year survival rate and changes in long‐term care needs levels before and after COVID‐19 hospitalization. We analyzed survival rates by age distribution (65–69, 70–79, 80–89, and ≥ 90 years), disease severity, predominant virus variant waves (wild‐type, Alpha, Delta, and Omicron), and preexisting care needs levels. In addition, we evaluated predictive long‐term prognostic factors.
2.7. Statistical Analyses
Patient characteristics were expressed as n (%) for categorical variables and medians (standard deviations) for continuous parameters. We used the Kaplan–Meier analysis to estimate survival rates and log rank tests to compare groups. Univariate and multivariate Cox proportional hazards models were used to identify independent prognostic factors. Variables with p < 0.10 in the univariate analysis were included in the multivariate model, along with clinically significant factors identified in previous studies, such as age, sex, and specific comorbidities [15, 16, 17]. All p values were two‐sided, and statistical significance was set at p < 0.05. Analyses were conducted using SPSS version 27 (IBM, Armonk, NY, USA).
3. Results
3.1. One‐Year Survival
We included 2175 hospitalized patients with a confirmed diagnosis of COVID‐19. Table 1 outlines the baseline characteristics of the study population. The overall 1‐year survival rate was 80.9% (Figure 2A). Stratified by sex, males had a significantly lower survival rate of 77.3%, compared to 83.8% for females (hazard ratio [HR]: 1.47, 95% confidence interval [CI]: 1.20–1.75; p < 0.001) (Figure 2A). Age significantly influenced survival outcomes. The 1‐year survival rate was 56.6% for patients aged ≥ 90 years, compared to 93.6% for those aged 65–69, 91.2% for those aged 70–79, and 76.0% for those aged 80–89 (Figure 2B). Survival outcomes also varied significantly by COVID‐19 severity. The 1‐year survival rate was 87.5% for patients with mild to moderate I, 74.0% for moderate II, and 49.9% for severe disease (Figure 2C). One‐year survival rates based on care needs levels were 90.4% for those with no care needs, 75.4% for Support Levels 1–2, 71.7% for Care Needs Level 1, 63.3% for Levels 2–3, and 58.1% for Levels 4–5 (Figure 2D). Across SARS‐CoV‐2 variant predominant waves, the 1‐year survival rate was 83.2% for wild‐type, 79.3% for Alpha, 87.3% for Delta, and 79.1% for Omicron (180 days) (Figure 2E). Among severe cases, the survival rate was lowest in the Omicron‐predominant wave (25.0%), compared with 58.1% for wild‐type, 48.6% for Alpha, and 40.0% for Delta (Figure 2F). One‐year survival rates among severe cases based on care needs levels were 54.3% for those with no care needs and 31.6% for those with Support Level 1 or higher (Figure 2G). For reference, among non‐COVID‐19 hospitalized patients during the same observation period, the 1‐year survival rates by preexisting care needs level were 94.4% for no care needs, 89.8% for Support Levels 1–2, 88.3% for Care Needs Level 1, 83.3% for Levels 2–3, and 70.7% for Levels 4–5 (Figure 2H).
TABLE 1.
Baseline characteristics.
| All patients, N = 2175 | Mild to moderate I, n = 1293 | Moderate II, n = 784 | Severe, n = 98 | p | |
|---|---|---|---|---|---|
| Sex and age | |||||
| Sex, male | 971 (44.6) | 527 (40.8) | 374 (47.7) | 70 (71.4) | < 0.001 |
| Age, years (SD) | 80.2 (8.3) | 79.9 (8.3) | 81.2 (8.3) | 76.2 (6.2) | < 0.001 |
| 65–69 | 206 (9.5) | 122 (9.4) | 68 (8.7) | 16 (16.3) | 0.05 |
| 70–79 | 873 (40.1) | 550 (42.5) | 268 (34.2) | 55 (56.1) | < 0.001 |
| 80–89 | 753 (34.6) | 418 (32.3) | 311 (39.7) | 24 (24.5) | < 0.001 |
| ≥ 90 | 343 (15.8) | 203 (15.7) | 130–140 | < 10 | 0.001 |
| Comorbidity | |||||
| Hypertension | 1517 (69.7) | 846 (65.4) | 586 (74.7) | 85 (86.7) | < 0.001 |
| Coronary artery disease | 616 (28.3) | 339 (26.2) | 241 (30.7) | 36 (36.7) | 0.01 |
| Chronic heart failure | 730 (33.6) | 387 (29.9) | 284 (36.2) | 59 (60.2) | < 0.001 |
| Atrial fibrillation | 288 (13.2) | 152 (11.8) | 113 (14.4) | 23 (23.5) | 0.002 |
| Chronic pulmonary disease (chronic obstructive pulmonary disease, asthma, and others) | 1072 (49.3) | 604 (46.7) | 409 (52.2) | 59 (60.2) | 0.005 |
| Diabetes mellitus | 664 (30.5) | 308 (23.8) | 285 (36.4) | 71 (72.4) | < 0.001 |
| Dyslipidemia | 1020 (46.9) | 582 (45.0) | 384 (49.0) | 54 (55.1) | 0.05 |
| Peripheral arterial disease | 330 (15.2) | 173 (13.4) | 139 (17.7) | 18 (18.4) | 0.02 |
| Renal disease | 248 (11.4) | 120 (9.3) | 114 (14.5) | 14 (14.3) | 0.001 |
| Hemodialysis | 42 (1.9) | 10–20 | 23 (2.9) | < 10 | < 0.001 |
| Liver disease | 303 (13.9) | 151 (11.7) | 129 (16.5) | 23 (23.5) | < 0.001 |
| Any malignancy including leukemia and lymphoma | 132 (6.1) | 75 (5.8) | 50–60 | < 10 | 0.68 |
| Rheumatologic disease (connective tissue disease) | 107 (4.9) | 63 (4.9) | 40–50 | < 10 | 0.90 |
| Cerebrovascular disease | 637 (29.3) | 349 (27.0) | 260 (33.2) | 28 (28.6) | 0.01 |
| Hemiplegia or paraplegia | 51 (2.3) | 28 (2.2) | 10–20 | < 10 | 0.04 |
| Dementia | 382 (17.6) | 229 (17.7) | 140–150 | < 10 | 0.001 |
| Care needs level | |||||
| No care needs | 1254 (57.7) | 759 (58.7) | 416 (53.1) | 79 (19.4) | < 0.001 |
| Support Level 1–2 | 248 (11.4) | 147 (11.4) | 90–100 | < 10 | 0.55 |
| Care Needs Level 1 | 177 (8.1) | 97 (7.5) | 70–80 | < 10 | 0.03 |
| Care Needs Level 2 | 184 (8.5) | 108 (8.4) | 70–80 | < 10 | 0.41 |
| Care Needs Level 3 | 135 (6.2) | 74 (5.7) | 60–70 | < 10 | 0.02 |
| Care Needs Level 4 | 111 (5.1) | 62 (4.8) | 40–50 | < 10 | 0.07 |
| Care Needs Level 5 | 66 (3.0) | 46 (3.6) | 10–20 | < 10 | 0.17 |
| Respiratory therapy | |||||
| Oxygen supplementation | 855 (39.3) | N/A | 775 (98.9) | 80 (81.6) | N/A |
| Noninvasive positive pressure ventilation | < 10 | N/A | < 10 | < 10 | N/A |
| High‐flow therapy | 114 (5.2) | N/A | 74 (9.4) | 40 (40.8) | N/A |
| Invasive mechanical ventilation | 98 (4.5) | N/A | N/A | 98 (100) | N/A |
| Extracorporeal membrane oxygenation | < 10 | N/A | N/A | < 10 | N/A |
| Predominant SARS‐CoV‐2 waves | |||||
| Wild‐type | 565 (26.0) | 325 (25.1) | 197 (25.1) | 43 (43.9) | < 0.001 |
| Alpha | 678 (31.2) | 296 (22.9) | 345 (44.0) | 37 (37.8) | < 0.001 |
| Delta | 364 (16.7) | 247 (19.1) | 107 (13.6) | 10 (10.2) | 0.001 |
| Omicron | 568 (26.1) | 425 (32.9) | 130–140 | < 10 | < 0.001 |
| Medication and renal replacement therapy | |||||
| Hydrocortisone | 38 (1.7) | < 10 | < 10 | 20–30 | < 0.001 |
| Prednisolone | 102 (4.7) | 34 (2.6) | 51 (6.5) | 17 (17.3) | < 0.001 |
| Methylprednisolone | 101 (4.6) | 10 (0.8) | 65 (8.3) | 26 (26.5) | < 0.001 |
| Dexamethasone | 773 (35.5) | 155 (12.0) | 536 (68.4) | 82 (83.7) | < 0.001 |
| Remdesivir | 51 (2.3) | 10–20 | 36 (4.6) | < 10 | < 0.001 |
| Baricitinib | 43 (2.0) | 10–20 | 22 (2.8) | < 10 | < 0.001 |
| Heparin | 425 (19.5) | 61 (4.7) | 278 (35.5) | 86 (87.8) | < 0.001 |
| Any antibiotics | 588 (27.0) | 186 (14.4) | 321 (40.9) | 81 (82.7) | < 0.001 |
| Renal replacement therapy | 72 (3.3) | 29 (2.2) | 29 (3.7) | 14 (14.3) | < 0.001 |
| Death | 428 (19.7) | 164 (12.7) | 215 (27.4) | 49 (50.0) | < 0.001 |
Note: When the number of patients was fewer than 10 (excluding 0), the data were not disclosed and were adjusted to prevent identification following the National Health Insurance cell size suppression policy.
Abbreviations: N/A, not applicable; SARS‐CoV‐2, severe acute respiratory syndrome coronavirus 2; SD, standard deviation.
FIGURE 2.
One‐year survival rate of hospitalized patients with COVID‐19. (A) Overall survival and by sex. (B) Survival by age group: 65–69, 70–79, 80–89, and ≥ 90 years. (C) Survival by disease severity: Mild to moderate I, moderate II, and severe. (D) Survival by care needs level (CNL): No care needs, support level (SL) 1–2, CNL1, CNL2–3, and CNL4–5. (E) Survival by predominant SARS‐CoV‐2 variant waves: Wild‐type (February 2020–February 2021), Alpha (March–June 2021), Delta (July–November 2021), and Omicron (December 2021–March 2022). (F) Survival in severe cases by predominant SARS‐CoV‐2 variant waves. (G) Survival in severe cases by CNL: No care needs, SL1, and CNL1–5. (H) Survival of non‐COVID‐19 hospitalized patients by preexisting CNL: No care needs, SL1–2, CNL1, CNL2–3, and CNL4–5. Mild to moderate I: no oxygen therapy; moderate II: respiratory support (oxygen administration with/without noninvasive positive pressure ventilation, high‐flow therapy); severe: invasive mechanical ventilation with/without extracorporeal membrane oxygenation. CI, confidence interval; CNL, care needs level; COVID‐19, coronavirus disease 2019; HR, hazard ratio; SARS‐CoV‐2, severe acute respiratory syndrome coronavirus 2; SL, support level.
3.2. Care Needs Levels
We evaluated the changes in long‐term care needs levels before and after COVID‐19 (Figure 3A; Table S3). There was a consistent decrease in care needs across all categories. Among 1254 patients with no preexisting care needs, 971 (77.4%) retained independence after COVID‐19, while 22.6% developed new care needs ranging from Support Levels 1–2 to Level 5 (Figure 3A). A total of 428 deaths were recorded, with the highest number among patients requiring no care needs preinfection (29%, n = 127). Among deceased patients, preexisting care needs levels were distributed as follows: 14% (n = 59) at Support Levels 1–2, 12% (n = 50) at Level 1, 15% (n = 65) at Level 2, 13% (n = 56) at Level 3, 11% (n = 46) at Level 4, and 6% (n = 25) at Level 5 (Figure 3B). One‐year mortality was 35.3% among patients with preexisting Care Needs Level 2 and remained consistently elevated at higher levels, reaching 41.5% at Level 3, 41.4% at Level 4, and 37.9% at Level 5 (Figure 3C). In hospitalized patients with as well as without COVID‐19, 1‐year mortality increased progressively with higher preexisting care needs levels. At every stratum, patients with COVID‐19 showed consistently higher mortality than those without, except at level 5, where mortality was comparably high in both groups (Figure 3C).
FIGURE 3.
Changes in care needs levels and survival outcomes among patients with COVID‐19. (A) Sankey diagram illustrating the changes in care needs levels before and after COVID‐19. (B) Distribution of preexisting care needs levels in patients who died following COVID‐19. (C) One‐year mortality stratified by preexisting care need levels in patients with and without COVID‐19. CNL, care needs level; COVID‐19, coronavirus disease 2019; SL, support level.
3.3. Prognostic Factors
Cox proportional hazards analysis identified several factors significantly associated with poor survival (Table 2). Age, assessed as a continuous variable, was a significant predictor (HR: 1.08, 95% CI: 1.06–1.10; p < 0.001). Male sex was associated with nearly double the risk of poor survival (HR: 2.05, 95% CI: 1.67–2.52; p < 0.001). Comorbid conditions significantly impacting prognosis included diabetes mellitus (HR: 1.33, 95% CI: 1.08–1.64; p = 0.007), hemodialysis (HR: 2.33, 95% CI: 1.36–4.00; p = 0.002), and any malignancies, including leukemia and lymphoma (HR: 2.27, 95% CI: 1.64–3.15; p < 0.001). A significant association was observed between rheumatologic disease (connective tissue disease) and poor outcomes (HR: 1.90, 95% CI: 1.32–2.74; p < 0.001). Care needs levels also served as strong indicators of poor survival: Care Needs Level 1 (HR: 1.59, 95% CI: 1.10–2.31; p = 0.01), Care Needs Levels 2–3 (HR: 2.23, 95% CI: 1.63–3.03; p < 0.001), and Care Needs Levels 4–5 (HR: 2.82, 95% CI: 1.99–4.00; p < 0.001).
TABLE 2.
Cox proportional hazard ratio analysis for 1‐year survival.
| Univariate analysis | Multivariable analysis | |||||
|---|---|---|---|---|---|---|
| HR | 95% CI | p | HR | 95% CI | p | |
| Age | 1.10 | 1.08–1.11 | < 0.001 | 1.08 | 1.06–1.10 | < 0.001 |
| Male | 1.45 | 1.20–1.75 | < 0.001 | 2.05 | 1.67–2.52 | < 0.001 |
| Hypertension | 2.09 | 1.64–2.68 | < 0.001 | 1.11 | 0.84–1.47 | 0.47 |
| Coronary artery disease | 1.70 | 1.40–2.07 | < 0.001 | 0.96 | 0.77–1.20 | 0.73 |
| Chronic heart failure | 2.29 | 1.89–2.77 | < 0.001 | 1.18 | 0.94–1.48 | 0.16 |
| Atrial fibrillation | 1.69 | 1.33–2.15 | < 0.001 | 1.22 | 0.94–1.58 | 0.13 |
| Chronic pulmonary disease (chronic obstructive pulmonary disease, asthma, and others) | 1.26 | 1.04–1.52 | 0.02 | 1.10 | 0.90–1.33 | 0.36 |
| Diabetes mellitus | 1.35 | 1.11–1.65 | 0.003 | 1.33 | 1.08–1.64 | 0.007 |
| Dyslipidemia | 0.86 | 0.71–1.05 | 0.13 | — | — | — |
| Peripheral arterial disease | 1.72 | 1.37–2.15 | < 0.001 | 1.10 | 0.86–1.41 | 0.46 |
| Renal disease (moderate or severe renal disease) | 1.90 | 1.49–2.44 | < 0.001 | 1.04 | 0.78–1.39 | 0.78 |
| Hemodialysis | 2.79 | 1.74–4.47 | < 0.001 | 2.33 | 1.36–4.00 | 0.002 |
| Liver disease | 1.48 | 1.16–1.89 | 0.002 | 1.17 | 0.91–1.50 | 0.23 |
| Any malignancy including leukemia and lymphoma | 1.85 | 1.35–2.53 | < 0.001 | 2.27 | 1.64–3.15 | < 0.001 |
| Rheumatologic disease (connective tissue disease) | 1.72 | 1.21–2.46 | 0.003 | 1.90 | 1.32–2.74 | < 0.001 |
| Cerebrovascular disease | 2.09 | 1.73–2.53 | < 0.001 | 1.11 | 0.89–1.38 | 0.35 |
| Hemiplegia or paraplegia | 2.67 | 1.74–4.11 | < 0.001 | 1.36 | 0.86–2.15 | 0.19 |
| Dementia | 2.09 | 1.69–2.58 | < 0.001 | 1.13 | 0.89–1.43 | 0.33 |
| No care needs | 0.26 | 0.21–0.32 | < 0.001 | Reference | ||
| Support Level 1–2 | 1.29 | 0.98–1.70 | 0.07 | 1.36 | 0.97–1.91 | 0.07 |
| Care Needs Level 1 | 1.63 | 1.21–2.19 | 0.001 | 1.59 | 1.10–2.31 | 0.01 |
| Care Needs Levels 2–3 | 2.78 | 2.10–3.69 | < 0.001 | 2.23 | 1.63–3.03 | < 0.001 |
| Care Needs Levels 4–5 | 2.25 | 1.50–3.38 | < 0.001 | 2.82 | 1.99–4.00 | < 0.001 |
Note: Multivariate Cox proportional hazard models were adjusted for age (continuous), sex (male), hypertension, coronary artery disease, atrial fibrillation, chronic heart failure, chronic pulmonary disease (chronic obstructive pulmonary disease, asthma, and others), diabetes mellitus, peripheral arterial disease, renal disease (moderate or severe renal disease), hemodialysis, liver disease, any malignancy including leukemia and lymphoma, rheumatologic disease (connective tissue disease), cerebrovascular disease, hemiplegia or paraplegia, dementia, level of care needs (including no care needs, Support Levels 1–2, Care Needs Levels 1, 2–3, and 4–5) as covariates.
Abbreviations: CI, confidence interval; HR, hazard ratio.
4. Discussion
4.1. One‐Year Survival
We compared the background characteristics and long‐term survival rates of patients hospitalized with COVID‐19 across four predominant epidemic waves in Japan (wild‐type, Alpha, Delta, and Omicron). The results emphasize that advanced age remains a key determinant of survival in patients with COVID‐19. Moreover, patients with severe COVID‐19 had markedly lower 1‐year survival rates than those with moderate II or mild to moderate I disease. This underscores the critical challenges in managing severe cases, irrespective of the SARS‐CoV‐2 strain. The consistent decline in survival among severely ill patients highlights the importance of not only early intervention and aggressive treatment strategies for COVID‐19 but also long‐term and sustained care. The severity of illness from the Omicron variant is reportedly similar to that of other variants, suggesting the importance of managing severely ill patients [19, 20]. However, Omicron variants have also been associated with the most severe outcomes in critically ill patients [7], especially among older patients [6, 7, 21, 22]. While overall COVID‐19 severity has decreased since the emergence of the Omicron variant, poor prognosis persists for patients who develop severe disease. Despite a reduction in the number of severely ill patients across variant waves, older age remains associated with severe long‐term prognosis once severe illness develops.
4.2. Care Needs Levels
This study presents novel insights into the impact of COVID‐19 on care needs among older hospitalized patients in Nara Prefecture, Japan. A decline in functional status was observed across all care needs levels. Long‐term care needs after COVID‐19 declined significantly among those with no care needs to Care Needs Level 2, while patients requiring Care Needs Level 3 remained stable. In contrast, the number of patients requiring Levels 4 and 5 increased (Figure 3A). These changes illustrate the profound impact of COVID‐19, manifesting as either death or an increased requirement for advanced care needs. Independent associations between preexisting care needs levels and outcomes were observed after adjusting for potential confounders. A previous study in Japan that did not include COVID‐19 found 1‐year mortality rates for patients requiring invasive mechanical ventilation with preexisting Care Needs Levels 2–3 and 4–5 to be 67.8% and 74.1%, respectively, approximately 25%–30% higher than those for patients with no care needs [14]. In this study, 1‐year survival for patients who required invasive mechanical ventilation among those with preexisting Support Level 1 and higher was 31.6%, generally similar to the aforementioned report [14]. However, each preexisting care needs level included < 10 severe cases. The low incidence of invasive mechanical ventilation among patients with higher preexisting care needs suggests limited access to this intervention owing to strain on healthcare resources during the COVID‐19 pandemic or the avoidance of aggressive treatments. One‐year mortality was consistently higher in patients with COVID‐19 than in those without (Figure 3C). However, the non‐COVID‐19 cohort in this study excluded patients with “suspected COVID‐19” diagnoses, which were frequently recorded during the pandemic; therefore, the number of non‐COVID‐19 cases may have been underestimated. Nevertheless, these findings offer important clinical insights for guiding goals of care in older adults with poor functional or cognitive status at baseline.
4.3. Care Needs as a Prognostic Factor
Various risk factors have been identified for COVID‐19 mortality [23, 24, 25, 26, 27, 28, 29], and our findings align with these results. A distinctive feature of our study is the integration of care needs levels into risk assessment. In Japan, with its rapidly aging population, the relationship between COVID‐19 outcomes and care needs levels has rarely been examined. As higher care needs generally predict poorer outcomes, the observed association with increased mortality risk is both reasonable and clinically meaningful. This underscores the significance of considering care needs alongside comorbidities when assessing prognosis in patients with COVID‐19. Further comparative studies are needed to understand the long‐term outcomes of those who newly develop care needs.
4.4. Limitations
Our study has some limitations. First, reliance on the KDB limited the study population to individuals included in the database, potentially introducing selection bias. Although the KDB covers almost all LEHI participants, it omits data on individuals aged 65–74 years who are not enrolled in the NHI. Second, although the outcome was 1‐year survival, we could not separately evaluate in‐hospital and post‐discharge death. Third, COVID‐19 severity was classified based on oxygen therapy, without consideration of clinical findings such as physical examination or imaging results. In some cases, treatment decisions may also have been influenced by advance care planning, leading to management that did not fully reflect the underlying illness severity. Fourth, because patients with non‐COVID‐19 in the KDB were defined as those without a registered COVID‐19 diagnosis, patients with “suspected COVID‐19” listed as their admission diagnosis were excluded. During the pandemic, nearly all hospitalized patients underwent SARS‐CoV‐2 antigen or PCR testing, and “suspected COVID‐19” was commonly coded. Consequently, the non‐COVID‐19 cohort presented here is likely smaller than the actual population of non‐COVID‐19 hospitalized patients. Fifth, we could not assess certain critical factors for COVID‐19—such as obesity, smoking history, vaccination status, physical examination findings, and specific laboratory or imaging findings—limiting the comprehensiveness of our severity analyses. Sixth, we could not determine the severe complications—such as sepsis, multiple organ dysfunction, or shock—and causes of death, which may have provided more profound insights into the long‐term outcomes of COVID‐19. The absence of cause‐of‐death data could impact the interpretation of survival rates, especially concerning distinguishing COVID‐19‐related from non‐COVID‐19‐related mortalities. Seventh, comorbidities were defined using ICD‐10 codes from administrative claims data, which may have resulted in residual confounding due to unmeasured comorbidities. This is an inherent limitation of claims‐based observational studies. Finally, regarding care needs, upward adjustments to care requirements occur relatively early when a patient's condition worsens, while downward adjustments typically occur only during scheduled reassessment periods, even if the patient's condition had previously improved. Consequently, care needs at any given time, as captured in claims data, may not accurately reflect a patient's true care needs. Particularly, while the system effectively captures the deterioration of patients' conditions, it tends to overestimate the care needs of patients who have experienced improvements. Despite these limitations, this study is, to our knowledge, the first to report on long‐term survival outcomes and evolving care needs among hospitalized older patients with COVID‐19 in Japan.
4.5. Conclusion
This study provides critical insights into 1‐year survival rates and care needs among older patients with COVID‐19 in Japan. These results may inform clinical decision‐making, patient management strategies, and resource allocation for older populations affected by COVID‐19.
Author Contributions
K.T. contributed substantially to the study conception and interpretation of data and drafted the manuscript. T.M., Y.N., I.Y., T.N., and T.I. contributed significantly to the paper. All authors reviewed and approved the final version of the manuscript.
Funding
This work was supported by Health Labour Sciences Research Grant (24IA1009).
Ethics Statement
This study was approved by the Ethics Committee of Nara Medical University (Approval number: 1123). The study was conducted in compliance with the Declaration of Helsinki.
Consent
The requirement for informed consent was waived owing to the anonymized nature of the claims datasets.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Table S1: ICD‐10 coding for COVID‐19 and comorbidities.
Table S2: Medical practice and medication code.
Table S3: Changes in the number of patients in each care need level before and after COVID‐19.
Data S1: Supplementary table 1 legend.
Data S2: Supplementary table 2 legend.
Data S3: Supplementary table 3 legend.
Acknowledgments
We would like to thank Editage (www.editage.jp) for English language editing.
Takemoto K., Myojin T., Nishioka Y., Yamazaki I., Noda T., and Imamura T., “One‐Year Survival and Changes in Care Needs Among Hospitalized Older Patients With Coronavirus Disease 2019 in Japan: A Nara Kokuho Database Analysis,” Geriatrics & Gerontology International 26, no. 2 (2026): e70353, 10.1111/ggi.70353.
Data Availability Statement
The data that support the findings of this study are available from Nara Prefectural Government. Restrictions apply to the availability of these data, which were used under license for this study. Data are available from the author(s) with the permission of Nara Prefectural Government.
References
- 1. COVID‐19 Mental Disorders Collaborators , “Global Prevalence and Burden of Depressive and Anxiety Disorders in 204 Countries and Territories in 2020 due to the COVID‐19 Pandemic,” Lancet 398, no. 10312 (2021): 1700–1712. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Izurieta H. S., Graham D. J., Jiao Y., et al., “Natural History of Coronavirus Disease 2019: Risk Factors for Hospitalizations and Deaths Among >26 Million US Medicare Beneficiaries,” Journal of Infectious Diseases 223, no. 6 (2021): 945–956. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Lai C. C., Hsu C. K., Yen M. Y., Lee P. I., Ko W. C., and Hsueh P. R., “Long COVID: An Inevitable Sequela of SARS‐CoV‐2 Infection,” Journal of Microbiology, Immunology, and Infection 56, no. 1 (2023): 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Wang L., Ma H., Yiu K. C., et al., “Heterogeneity in Testing, Diagnosis and Outcome in SARS‐CoV‐2 Infection Across Outbreak Settings in the Greater Toronto Area, Canada: An Observational Study,” CMAJ Open 8, no. 4 (2020): E627–E636. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Lu Y., Jiao Y., Graham D. J., et al., “Risk Factors for COVID‐19 Deaths Among Elderly Nursing Home Medicare Beneficiaries in the Prevaccine Period,” Journal of Infectious Diseases 225, no. 4 (2022): 567–577. [DOI] [PubMed] [Google Scholar]
- 6. Shibata T., Okano S., Onozuka D., Ohta E., and Kutsuna S., “Analysis of Concentrated COVID‐19 Outbreaks in Elderly Facilities in Suita City, Osaka Prefecture, Japan,” International Journal of Environmental Research and Public Health 20, no. 20 (2023): 6926. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Sakuramoto K., Wada D., Maruyama S., et al., “Evaluation of Characteristics and Prognosis of COVID‐19 Patients Requiring Invasive Mechanical Ventilation During Dominance of Nonvariant, Alpha, Delta, and Omicron Variants in Tertiary Hospitals of Japan,” BMC Infectious Diseases 24, no. 1 (2024): 223. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Zha L., Sobue T., Takeuchi T., et al., “Characteristics and Survival of Intensive Care Unit Patients With Coronavirus Disease in Osaka, Japan: A Retrospective Observational Study,” Journal of Clinical Medicine 10, no. 11 (2021): 2477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Nomoto H., Hayakawa K., and Ohmagari N., “Impact of Prioritized Vaccinations for the Elderly on the COVID‐19 Pandemic in Japan,” Global Health & Medicine 4, no. 2 (2022): 129–132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Tsuzuki S., Miyazato Y., Terada M., Morioka S., Ohmagari N., and Beutels P., “Impact of Long‐COVID on Health‐Related Quality of Life in Japanese COVID‐19 Patients,” Health and Quality of Life Outcomes 20, no. 1 (2022): 125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Myojin T., Noda T., Kubo S., Nishioka Y., Higashino T., and Imamura T., “Development of a New Method to Trace Patient Data Using the National Database in Japan,” Advanced Biomedical Engineering 11 (2022): 203–217. [Google Scholar]
- 12. Clinical Management of Patients With COVID‐19: A Guide for Front‐Line Healthcare Workers: Version 2.1, 2020.
- 13. Miyashita K., Hozumi H., Furuhashi K., et al., “Changes in the Characteristics and Outcomes of COVID‐19 Patients From the Early Pandemic to the Delta Variant Epidemic: A Nationwide Population‐Based Study,” Emerging Microbes & Infections 12, no. 1 (2023): 2155250. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Ohbe H., Ouchi K., Miyamoto Y., et al., “One‐Year Functional Outcomes After Invasive Mechanical Ventilation for Older Adults With Preexisting Long‐Term Care‐Needs,” Critical Care Medicine 51, no. 5 (2023): 584–593. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Tanaka C., Tagami T., Nakayama F., et al., “Association Between Mortality and Age Among Mechanically Ventilated COVID‐19 Patients: A Japanese Nationwide COVID‐19 Database Study,” Annals of Intensive Care 11, no. 1 (2021): 171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Vila‐Corcoles A., Satue‐Gracia E., Vila‐Rovira A., et al., “COVID‐19‐Related and All‐Cause Mortality Risk Among Middle‐Aged and Older Adults Across the First Epidemic Wave of SARS‐COV‐2 Infection: A Population‐Based Cohort Study in Southern Catalonia, Spain, March‐June 2020,” BMC Public Health 21, no. 1 (2021): 1795. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Tsutsui T. and Muramatsu N., “Care‐Needs Certification in the Long‐Term Care Insurance System of Japan,” Journal of the American Geriatrics Society 53, no. 3 (2005): 522–527. [DOI] [PubMed] [Google Scholar]
- 18. Matsuda T., Iwagami M., Suzuki T., Jin X., Watanabe T., and Tamiya N., “Correlation Between the Barthel Index and Care Need Levels in the Japanese Long‐Term Care Insurance System,” Geriatrics & Gerontology International 19, no. 11 (2019): 1186–1187. [DOI] [PubMed] [Google Scholar]
- 19. Arabi M., Al‐Najjar Y., Mhaimeed N., et al., “Severity of the Omicron SARS‐CoV‐2 Variant Compared With the Previous Lineages: A Systematic Review,” Journal of Cellular and Molecular Medicine 27, no. 11 (2023): 1443–1464. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Fall A., Eldesouki R. E., Sachithanandham J., et al., “A Quick Displacement of the SARS‐CoV‐2 Variant Delta With Omicron: Unprecedented Spike in COVID‐19 Cases Associated With Fewer Admissions and Comparable Upper Respiratory Viral Loads,” preprint, medRxiv, January 28, 2022, 10.1101/2022.01.26.22269927. [DOI]
- 21. Seyler L., van Nedervelde E., de Cock D., et al., “Surfing the Waves: Differences in Hospitalised COVID‐19 Patients Across 4 Variant Waves in a Belgian University Hospital,” Viruses 15, no. 3 (2023): 618. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Yang X., Shi F., Zhang J., et al., “Disease Severity of COVID‐19 in Different Phases of the Pandemic: Do Healthcare Workers Have Better Outcomes?,” Vaccine: X 15 (2023): 100377. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Matsunaga N., Hayakawa K., Terada M., et al., “Clinical Epidemiology of Hospitalized Patients With Coronavirus Disease 2019 (COVID‐19) in Japan: Report of the COVID‐19 Registry Japan,” Clinical Infectious Diseases 73, no. 11 (2021): e3677–e3689. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Liang W., Guan W., Chen R., et al., “Cancer Patients in SARS‐CoV‐2 Infection: A Nationwide Analysis in China,” Lancet Oncology 21, no. 3 (2020): 335–337. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Lippi G. and Henry B. M., “Chronic Obstructive Pulmonary Disease Is Associated With Severe Coronavirus Disease 2019 (COVID‐19),” Respiratory Medicine 167 (2020): 105941. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Myers L. C., Parodi S. M., Escobar G. J., and Liu V. X., “Characteristics of Hospitalized Adults With COVID‐19 in an Integrated Health Care System in California,” JAMA 323, no. 21 (2020): 2195–2198. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Fadini G. P., Morieri M. L., Longato E., and Avogaro A., “Prevalence and Impact of Diabetes Among People Infected With SARS‐CoV‐2,” Journal of Endocrinological Investigation 43, no. 6 (2020): 867–869. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Zheng Z., Peng F., Xu B., et al., “Risk Factors of Critical & Mortal COVID‐19 Cases: A Systematic Literature Review and Meta‐Analysis,” Journal of Infection 81, no. 2 (2020): e16–e25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Yang J., Zheng Y., Gou X., et al., “Prevalence of Comorbidities and Its Effects in Patients Infected With SARS‐CoV‐2: A Systematic Review and Meta‐Analysis,” International Journal of Infectious Diseases 94 (2020): 91–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1: ICD‐10 coding for COVID‐19 and comorbidities.
Table S2: Medical practice and medication code.
Table S3: Changes in the number of patients in each care need level before and after COVID‐19.
Data S1: Supplementary table 1 legend.
Data S2: Supplementary table 2 legend.
Data S3: Supplementary table 3 legend.
Data Availability Statement
The data that support the findings of this study are available from Nara Prefectural Government. Restrictions apply to the availability of these data, which were used under license for this study. Data are available from the author(s) with the permission of Nara Prefectural Government.