ABSTRACT
Background and Aims
In the ambulatory setting, mildly low hemoglobin (Hb), < 13 g/dL in men or < 12 g/dL in women, is associated with reduced cognition. However, in hospitalized patients, clinicians tolerate much lower Hb levels due to restrictive transfusion policies. Older adults who are at higher risk for cognitive impairment may also be particularly susceptible to impaired cognition from lower Hb levels. The purpose of this study was to test for an association between restrictive Hb levels during hospitalization and patient's cognitive function, particularly in older adults.
Methods
The data for this study came from an ongoing prospective observational study of hospitalized general medicine patients at the University of Chicago Medical Center (UCMC). Any adult with a Hb < 10 g/dL between March 2017 and February 2020 was eligible. Cognition was measured using the Short Portable Mental Status Questionnaire (SPMSQ), which ranges from 0 to 10. Higher scores indicate worse cognition. Zero‐inflated Poisson (ZIP) and logistic regression were used to test the association between cognition as the dependent variable, measured by the SPMSQ, and patients' nadir Hb and age as predictor variables, controlling for race, sex, Charlson Comorbidity Index, transfusion, and a diagnosis of dementia.
Results
The SPMSQ questionnaire was completed by 5112 patients who had Hb < 10 g/dL. Lower nadir Hb levels were associated with decreased cognition (higher SPMSQ). Across all ages, a 1 g/dL decrease in Hb was associated with a 7.2% increase in SPMSQ score (95% confidence interval (CI) = 2.9%–11.7%). The effect of decreasing Hb on cognition was greater in older populations, where a 1 g/dL decrease in Hb was associated with an 8.8% (3.1%–14.8%) and 20.8% (9.7%–33.0%) increase in SPMSQ score for patients above 50 and 65, respectively.
Conclusions
Lower Hb levels during hospitalization are associated with decreased cognition, especially in older adults.
Keywords: anemia, cognitive impairment, older adults
Summary
Lower nadir hemoglobin level during hospitalization is associated with decreased cognition.
The effect is greater in older patients, especially those above 65 years old.
1. Introduction
The association between anemia (Hb < 13 g/dL in men, < 12 g/dL in women) and cognitive impairment in the ambulatory setting, particularly in older adults, has been well described [1, 2, 3, 4, 5, 6]. This effect results from decreased cerebral oxygen delivery due to lower Hb levels [7, 8]. In the hospital, clinicians tolerate far lower Hb levels in patients than are typical in the ambulatory setting because of restrictive transfusion policies [9]. This raises the question of whether these lower Hb levels (7–8 g/dL) impair cognition in older adults during hospitalization.
Transfusion trials have demonstrated the safety of maintaining Hb levels between 7 and 8 g/dL for most patient populations during hospitalization with respect to stroke and mortality [10]. However, trial data has not been reported on these outcomes specifically in subgroups of older adults. Additionally, the transfusion trials have not reported on whether lower Hb levels and reduced oxygen delivery to the brain have milder and/or transient effects on cognition during hospitalization. Older adults are at risk for cognitive impairment and/or delirium during hospitalization, so they may also be at higher risk of impaired cognition from lower Hb levels compared to younger adults [11]. If lower Hb levels during hospitalization exacerbate cognitive function it may also increase the risk of delirium, impaired physical functioning, and the odds of hospital readmission [12, 13, 14]. Moreover, since some etiologies of anemia in hospitalized patients are easily treatable (i.e., nutritional anemias), determining whether lower Hb during hospitalization is associated with decreased cognition could result in clinically actionable treatments that can directly improve the care and outcomes of patients during and after hospitalization.
Therefore, the purpose of this study was to test for the association between Hb levels supported by restrictive transfusion practices and cognitive function in hospitalized older adults.
2. Methods
2.1. Study Design and Eligibility
Data for this study came from an ongoing prospective observational study of hospitalized adult patients with anemia at the University of Chicago Medical Center (UCMC) [15]. Between March 2017 and February 2020, any adult patient (≥ 18 years old) admitted to a general medicine service and with a Hb < 10 g/dL was eligible for participation in this study. General medicine services at UCMC include any adult with a non‐cardiac or non‐oncologic (i.e., cardiology and oncology are separate services) primary medical indication for hospitalization (i.e., excludes surgical admissions). The overwhelming majority of general medicine patients are admitted via the emergency room. A small fraction of patients are transferred to general medicine from another service (i.e., ICU), and a very small minority of patients are directly admitted from an ambulatory clinic. The Institutional Review Board approved the study procedures, and all study subjects or their healthcare power of attorney provided informed written consent to participate.
2.2. Measuring Cognitive Impairment
Cognitive impairment was measured with the Short Portable Mental Status Questionnaire (SPMSQ), a 10‐item survey designed for detecting delirium and dementia [16]. The SPMSQ can detect moderate‐severe cognitive impairment, and changes in SPMSQ scores represent clinically significant changes in cognitive function [17, 18]. We analyzed SPMSQ both as a continuous and a binary variable. As a binary outcome, we used a threshold of ≥ 3 mistakes to represent mild cognitive impairment, compared to < 3 mistakes representing no impairment [16].
2.3. Demographic and Clinical Data
Hospital administrative data were used to determine patients' age, sex, race/ethnicity, and comorbidities. Race was categorized as white, black, or other due to the low percentage of non‐black/African‐American patients at UCMC. A Charlson Comorbidity Index Score was calculated for each patient using the International Classification of Diseases, 10th Revision [19]. The patients' first Hb level < 10 g/dL, making them eligible for study participation, was obtained through manual review of the EHR. All additional Hb levels, including nadir Hb, were obtained from the hospital's clinical data repository. For analytic purposes, we used patients' nadir Hb to test for an association with their cognition, because nadir Hb is most closely associated with symptoms of anemia during hospitalization [20].
2.4. Statistical Analysis
Descriptive statistics were used to characterize the demographic and clinical characteristics of all participants. Patients who were eligible (i.e., admitted to a general medicine service and with a Hb < 10 g/dL) and provided informed consent for the study but were discharged prior to data collection (n = 510, Figure 1) were not included in the analysis. For our primary analysis, we used zero‐inflated Poisson (ZIP) regression to test the association between cognitive impairment, measured by the SPMSQ, and nadir Hb and an age‐Hb interaction term as the primary predictor variables. We chose ZIP regression because SPMSQ scores are discrete, and 62% of patients in our sample scored 0, which would jeopardize the normality of errors assumption of ordinary least squares regression [21]. We also compared the goodness of fit of the ZIP model to a Poisson model using the Bayesian Information Criterion (BIC) which significantly favored the ZIP model (z = 3.33, p < 0.001) [22]. Additionally, we performed a pre‐specified age‐stratified analysis in patients ≥ 50 years old and those ≥ 65 years old. In these models the BIC favored the use of a ZIP model in patients older than 50 (z = 2.46, p = 0.007) but not in patients above 65 (z = 0.65, p = 0.26), so Poisson regression was used in the above 65 subgroup. We also performed a sensitivity analysis by excluding patients with a known diagnosis of dementia. Last, we used a logistic regression model to test the association between Hb levels and mild cognitive impairment (SPMSQ ≥ 3). All models controlled for age, sex, race, Charlson Comorbidity Index, transfusion of red blood cells, and dementia. Prior to analysis we chose a two‐tailed analysis and a 95% level of confidence as the threshold for statistical significance. All statistical analysis was performed in R.Results.
Figure 1.
Patient enrollment flow chart. The Short Portable Mental Status Questionnaire (SPMSQ) ranges from 0 to 10, where higher scores indicate greater cognitive impairment.
2.5. Patient Characteristics
Of 5662 patients who consented to participate, 5112 had complete data (Figure 1). The sample was 59% female, and the mean age was 57 years old (Table 1). The sample's racial composition was 74% black/African‐American, 21% white, and 5% other (Table 1). The mean nadir Hb of the sample was 7.9 g/dL, the median Charlson Comorbidity Index was 3, and 96% of patients had an SPMSQ score below 3 (Table 1).
Table 1.
Patient characteristics.
| Overall (N = 5112) | |
|---|---|
| Age (years) | |
| Mean (SD) | 57 (18) |
| Sex | |
| Female | 2989 (58%) |
| Race | |
| Black/African‐American | 3790 (74%) |
| White | 1092 (21%) |
| Other | 230 (5%) |
| Nadir lab Hb (g/dL) | |
| Median (IQR) | 8.0 (6.8,9.0) |
| Hb < 8 | 2537 (50%) |
| Charlson comorbidity score | |
| 0 | 762 (15%) |
| 1–2 | 1519 (30%) |
| 3–4 | 1022 (20%) |
| >=5 | 1810 (35%) |
| Length of stay (days) | |
| Median (IQR) | 6.1 (3.63,10.21) |
| Dementia | |
| Dementia | 123 (2%) |
| Transfusion status | |
| Received transfusion | 1401 (27%) |
| SPMSQ score (0–10)a | |
| <=2 | 4915 (96%) |
| 3–4 | 165 (3%) |
| 5–7 | 18 (0.003%) |
| >=8 | 14 (0.002%) |
The Short Portable Mental Status Questionnaire (SPMSQ) ranges from 0–10, where higher scores indicate greater cognitive impairment. A score of 0–2 indicates intact intellectual functioning, 3–4 mild intellectual impairment, 5–7 moderate intellectual impairment, and >=8 severe intellectual impairment. Allow one additional point for black patients.
2.6. Anemia and the SPMSQ
Lower nadir Hb and increasing age were significantly associated with higher SPMSQ scores in all regression models (Table 2). In our primary model, a 1 g/dL lower Hb was associated with a 7% higher in SPMSQ score (1/rate ratio = 1/0.933 = 1.072, 1.029–1.117). There was also a statistically significant effect of the age/Hb interaction term, such that decreases in cognition associated with lower Hb become greater as patients get older (rate ratio = 1.002, 1.000–1.004) (Table 2). The sensitivity analysis excluding patients with dementia observed a Hb term of 0.938 (0.899–0.978), which was not clinically significantly different than the association between anemia and SPMSQ scores observed in the full model, 0.933 (0.895–0.972).
Table 2.
The association between hemoglobin level and SPMSQa scores.
| All patients (ZIP)b | Age >=50 (ZIP)b | Age >=65 (Poisson) | SPMSQ >=3 (logit) | |||||
|---|---|---|---|---|---|---|---|---|
| Coefficient | Rate ratio | p value | Rate ratio | p value | Rate ratio | p value | Odds ratio | p value |
|
Nadir Hb (95% CI) |
0.933 (0.895–0.972) |
0.001 |
0.919 (0.871–0.970) |
0.002 |
0.828 (0.752–0.911) |
< 0.001 |
0.840 (0.718–0.982) |
0.03 |
|
Age (95% CI) |
1.020 (1.015–1.025) |
< 0.001 |
1.020 (1.012–1.028) |
< 0.001 |
1.035 (1.023–1.047) |
< 0.001 |
1.050 (1.033–1.067) |
< 0.001 |
|
Sex: female (95% CI) |
0.996 (0.888–1.118) |
0.95 |
1.032 0.915–1.163) |
0.61 |
0.951 (0.827–1.185) |
0.33 |
0.970 (0.722–1.303) |
0.85 |
|
Race: black (95% CI) |
1.709 (1.467–1.990) |
< 0.001 |
1.654 (1.423–1.923) |
< 0.001 |
1.427 (1.227–1.660) |
< 0.001 |
4.036 (2.850–5.715) |
< 0.001 |
|
Race: other (95% CI) |
1.927 (1.320–2.815) |
< 0.001 |
1.834 (1.270–2.648) |
0.001 |
1.140 (0.833–1.560) |
0.42 |
3.403 (1.404–8.249) |
0.007 |
|
Charlson comorbidity index (95% CI) |
0.994 (0.975–1.013) |
0.55 |
0.992 (0.974–1.011) |
0.43 |
0.990 (0.972–1.008) |
0.25 |
1.017 (0.967–1.069) |
0.51 |
|
Dementia (95% CI) |
2.284 (1.883–2.771) |
< 0.001 |
2.270 (1.876–2.758) |
< 0.001 |
2.344 (1.939–2.833) |
< 0.001 |
4.535 (3.459–5.949) |
< 0.001 |
|
Transfused (95% CI) |
0.917 (0.687–1.222) |
0.16 |
0.887 (0.767–1.025) |
0.09 |
0.724 (0.619–0.846) |
< 0.001 |
0.511 (0.316–0.826) |
0.006 |
|
Age: Hb interaction (95% CI) |
1.002 (1.000–1.004) |
0.03 |
1.002 (0.999–1.005) |
0.14 |
1.006 (1.001–1.011) |
0.02 |
1.003 (0.995–1.011) |
0.48 |
The Short Mental Status Questionnaire (SPMSQ) ranges from 0–10, where higher scores indicate greater cognitive impairment.
Zero‐inflated Poisson (ZIP).
The logistic regression model demonstrated similar effects, whereby lower nadir Hb was associated with an increased probability of mild cognitive impairment (1/odds ratio = 1/0.840 = 1.19, 1.02–1.39) (Table 2). This represents a 19% change in the probability of mild cognitive impairment per 1 g/dL decrease in Hb. However, in this model, the age–Hb interaction term was not statistically significant.
Last, the age‐stratified analysis also showed a greater effect of lower nadir Hb on increased cognitive impairment in older adults. In the ≥ 50 and ≥ 65 years old subgroups, a 1 g/dL decrease in Hb was associated with 9% and 21% increases in SPMSQ scores, respectively.
3. Discussion
Our study observed that lower nadir Hb during hospitalization was associated with statistically significant decreases in cognitive function in hospitalized older adults. Moreover, the effect of lower nadir Hb on decreased cognition was greater as patients' age increased. We saw this effect in each of our regression models, and the effect size of nadir Hb on cognitive function was greatest in patients 65 and older. For patients in the ≥ 65 group, a 21% increase in predicted SPMSQ per 1 g/dL decline in Hb represents a clinically meaningful decline in cognition, putting patients at greater risk for passing the threshold from mild to medium or medium to severe cognitive impairment [17]. For example, our model predicts that a 2 g/dL change in Hb in a patient older than 65, could change their SPMSQ score from a 4 to a 5 or greater (4 * 1.208 * 1.208 = 5.8). This change would be sufficient to change the SPMSQ's cognitive assessment of the patient from mild to moderate intellectual impairment.
Our finding that lower nadir Hb is associated with decreased cognition is consistent with past data demonstrating an association between Hb levels and cognitive impairment seen in the ambulatory setting [2, 3]. However, this is the first study to investigate the relationship between Hb level and cognitive impairment in the context of the much lower Hb levels (7–8 g/dL) tolerated during hospitalization as a result of restrictive transfusion practices. Our findings are important because they demonstrate that there can be large changes in SPMSQ resulting from lower Hb levels, and even small changes in SPMSQ are associated with increased risk of post‐hospitalization disability in older patients [18]. Poorer SPMSQ and mild cognitive impairment in the hospital are also associated with higher rehospitalization rates and could result in long‐term or permanent declines in cognition [17, 23, 24]. Moreover, reduced cognition can impact a patient's autonomy and their ability to interact with their providers and direct their care. Although it is unlikely that transfusion can reverse such declines, our study suggests that robust anemia testing and treatment protocols should be more widely implemented to ensure that treatable etiologies, like iron deficiency, are not neglected during hospitalization and result in adverse effects. The strengths of this study include a large sample size, a diverse range of patient ages and cognitive states, and the robustness of our results to changes in model specifications. Future research should consider comparing changes in cognition with changes in Hb during hospitalization, comparing cognition between anemic and healthy adults, or comparing cognitive status prior and post transfusion.
This study has limitations. As a single‐center observational study, the results may not be generalizable to other settings. Additionally, the SPMSQ is not the only tool for measuring cognition in the hospital, and other validated instruments might have also been appropriate. While the SPMSQ may be limited in its ability discriminate between patients with low/mild cognitive impairment form those with intact cognition, in our study it would have therefore underestimated the impact of anemia on cognition, especially in patients who scored a 0, 1, or 2 [25]. We also did not capture patient's primary indication for hospitalization, and as a result our study could not evaluate heterogeneity in the association between Hb and cognition that may exist in patients admitted to the hospital for different reasons. Also, we lacked data on when patients' SPMSQ scores and nadir Hb were collected, which may confound the association between anemia and cognition. Without data on measurement times, we chose nadir hemoglobin as our predictor because it is the best predictor of symptoms of anemia during hospitalization [20]. Further, the 510 patients not included in our analysis may confound our results, as these patients may have not been able to complete the SPMSQ or lost interest in participation for reasons that could depend on cognitive impairment. Last, while the results from our models were consistent, we may not have accounted for all measured or unmeasured confounders that influence the relationship between Hb and cognition, particularly in older adults.
4. Conclusion
In summary, this study observed that lower Hb was significantly associated with worse cognition (SPMSQ scores) in hospitalized patients with Hb < 10 g/dL and that this effect is greater in patients older than 65. Worse cognition, measured by the SPMSQ, is associated with greater risk for functional decline and worse outcomes in old patients, and anemia during hospitalization may exacerbate these risk factors in older adults.
Author Contributions
Micah Prochaska: conceptualization, investigation, funding acquisition, writing – review and editing, data curation, supervision, methodology, formal analysis. Ethan Harper: writing – original draft, software, formal analysis.
Conflicts of Interest
The authors declare no conflicts of interest.
Transparency Statement
Ethan Harper affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.
Acknowledgments
All individuals who made significant contributions to the conception, design, execution, or interpretation of this project have been listed as authors. All authors have read and approved the final version of the manuscript. Ethan Harper, as corresponding author, had full access to all of the data in this study and takes complete responsibility for the integrity of the data and the accuracy of the data analysis. Dr. Prochaska is supported by grants from the National Institutes of Health (NIA R43).
Data Availability Statement
The data for this study includes identifiable protected patient health information. As a result, we are not allowed to share the data with anyone outside of the Institutional Review Board (IRB) approved investigative team. However, individuals interested in the data are encouraged to reach out to the authors to determine if there is a collaborative capacity where the data could be potentially de‐identified and/or shared under proper IRB and/or institutional Data Use Agreements (DUA).
References
- 1. Dlugaj M., Winkler A., Weimar C., et al., “Anemia and Mild Cognitive Impairment in the German General Population,” Journal of Alzheimer's Disease 49, no. 4 (2016): 1031–1042, 10.3233/JAD-150434. [DOI] [PubMed] [Google Scholar]
- 2. Yavuz B. B., Cankurtaran M., Haznedaroglu I. C., et al., “Iron Deficiency Can Cause Cognitive Impairment in Geriatric Patients,” Journal of Nutrition, Health and Aging 16, no. 3 (2012): 220–224, 10.1007/s12603-011-0351-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Andro M., Le Squere P., Estivin S., and Gentric A., “Anaemia and Cognitive Performances in the Elderly: A Systematic Review,” European Journal of Neurology 20, no. 9 (2013): 1234–1240, 10.1111/ene.12175. [DOI] [PubMed] [Google Scholar]
- 4. Jáuregui‐Lobera I., “Iron Deficiency and Cognitive Functions,” Neuropsychiatric Disease and Treatment 10 (2014): 2087–2095, 10.2147/NDT.S72491. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Son S. J., Lee K. S., Na D. L., et al., “The Effect of Anemia and White Matter Hyperintensities (WMH) on Cognitive Impairment in Patients With Amnestic Mild Cognitive Impairment (MCI),” Archives of Gerontology and Geriatrics 55, no. 2 (2012): 251–256, 10.1016/j.archger.2011.10.015. [DOI] [PubMed] [Google Scholar]
- 6. Zamboni V., Cesari M., Zuccalà G., et al., “Anemia and Cognitive Performance in Hospitalized Older Patients: Results From the GIFA Study,” International Journal of Geriatric Psychiatry 21, no. 6 (2006): 529–534, 10.1002/gps.1520. [DOI] [PubMed] [Google Scholar]
- 7. Wang X., Cui L., and Ji X., “Cognitive Impairment Caused by Hypoxia: From Clinical Evidences to Molecular Mechanisms,” Metabolic Brain Disease 37, no. 1 (2022): 51–66, 10.1007/s11011-021-00796-3. [DOI] [PubMed] [Google Scholar]
- 8. Zhang X. and Le W., “Pathological Role of Hypoxia in Alzheimer's Disease,” Experimental Neurology 223, no. 2 (2010): 299–303, 10.1016/j.expneurol.2009.07.033. [DOI] [PubMed] [Google Scholar]
- 9. Tobian A. A. R., Heddle N. M., Wiegmann T. L., and Carson J. L., “Red Blood Cell Transfusion: 2016 Clinical Practice Guidelines From AABB,” Transfusion 56, no. 10 (2016): 2627–2630, 10.1111/trf.13735. [DOI] [PubMed] [Google Scholar]
- 10. Carson J. L., Stanworth S. J., Roubinian N., et al., “Transfusion Thresholds and Other Strategies for Guiding Allogeneic Red Blood Cell Transfusion,” Cochrane Database of Systematic Reviews 10 (2016): CD002042, 10.1002/14651858.CD002042.pub4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Chinnappa‐Quinn L., Makkar S. R., Bennett M., et al., “Is Hospitalization a Risk Factor for Cognitive Decline in Older Age Adults?,” International Psychogeriatrics 34, no. 11 (2022): 963–980, 10.1017/S1041610220001763. [DOI] [PubMed] [Google Scholar]
- 12. Fogg C., Griffiths P., Meredith P., and Bridges J., “Hospital Outcomes of Older People With Cognitive Impairment: An Integrative Review,” International Journal of Geriatric Psychiatry 33, no. 9 (2018): 1177–1197, 10.1002/gps.4919. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Hartley P., Gibbins N., Saunders A., et al., “The Association Between Cognitive Impairment and Functional Outcome in Hospitalised Older Patients: A Systematic Review and Meta‐Analysis,” Age and Ageing 46, no. 4 (2017): 559–567, 10.1093/ageing/afx007. [DOI] [PubMed] [Google Scholar]
- 14. Callahan K. E., Lovato J. F., Miller M. E., Easterling D., Snitz B., and Williamson J. D., “Associations Between Mild Cognitive Impairment and Hospitalization and Readmission,” Journal of the American Geriatrics Society 63, no. 9 (2015): 1880–1885, 10.1111/jgs.13593. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Meltzer D., Manning W. G., Morrison J., et al., “Effects of Physician Experience on Costs and Outcomes on an Academic General Medicine Service: Results of a Trial of Hospitalists,” Annals of Internal Medicine 137, no. 11 (2002): 866–874, 10.7326/0003-4819-137-11-200212030-00007. [DOI] [PubMed] [Google Scholar]
- 16. Pfeiffer E., “A Short Portable Mental Status Questionnaire for the Assessment of Organic Brain Deficit in Elderly Patients,” Journal of the American Geriatrics Society 23, no. 10 (1975): 433–441, 10.1111/j.1532-5415.1975.tb00927.x. [DOI] [PubMed] [Google Scholar]
- 17. Schönstein A., Wahl H. W., Katus H. A., and Bahrmann A., “SPMSQ for Risk Stratification of Older Patients in the Emergency Department: An Exploratory Prospective Cohort Study,” Zeitschrift für Gerontologie und Geriatrie 52, no. S4 (2019): S222–S228, 10.1007/s00391-019-01626-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Zisberg A., Sinoff G., Agmon M., Tonkikh O., Gur‐Yaish N., and Shadmi E., “Even a Small Change Can Make a Big Difference: The Case of In‐Hospital Cognitive Decline and New Iadl Dependency,” Age and Ageing 45, no. 4 (2016): 500–504, 10.1093/ageing/afw063. [DOI] [PubMed] [Google Scholar]
- 19. Roffman C. E., Buchanan J., and Allison G. T., “Charlson Comorbidities Index,” Journal of Physiotherapy 62, no. 3 (2016): 171, 10.1016/j.jphys.2016.05.008. [DOI] [PubMed] [Google Scholar]
- 20. Prochaska M. T., Newcomb R., Block G., Park B., and Md D. O. M., “Association Between Anemia and Fatigue in Hospitalized Patients: Does the Measure of Anemia Matter?,” Journal of Hospital Medicine 12, no. 11 (2017): 898–904, 10.12788/jhm.2832. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Loeys T., Moerkerke B., De Smet O., and Buysse A., “The Analysis of Zero‐Inflated Count Data: Beyond Zero‐Inflated Poisson Regression,” British Journal of Mathematical and Statistical Psychology 65, no. 1 (2012): 163–180, 10.1111/j.2044-8317.2011.02031.x. [DOI] [PubMed] [Google Scholar]
- 22. Desmarais B. A. and Harden J. J., “Testing for Zero Inflation in Count Models: Bias Correction for the Vuong Test,” Stata Journal: Promoting communications on Statistics and Stata 13, no. 4 (2013): 810–835, 10.1177/1536867X1301300408. [DOI] [Google Scholar]
- 23. Ishihara K., Izawa K. P., Kitamura M., et al., “Impact of Mild Cognitive Impairment on Unplanned Readmission in Patients With Coronary Artery Disease,” European Journal of Cardiovascular Nursing 21, no. 4 (2022): 348–355, 10.1093/eurjcn/zvab091. [DOI] [PubMed] [Google Scholar]
- 24. Goldberg T. E., Chen C., Wang Y., et al., “Association of Delirium With Long‐Term Cognitive Decline: A Meta‐Analysis,” JAMA Neurology 77, no. 11 (2020): 1373–1381, 10.1001/jamaneurol.2020.2273. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Smyer M. A., Hofland B. F., and Jonas E. A., “Validity Study of the Short Portable Mental Status Questionnaire for the Elderly,” Journal of the American Geriatrics Society 27, no. 6 (1979): 263–269, 10.1111/j.1532-5415.1979.tb06128.x. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data for this study includes identifiable protected patient health information. As a result, we are not allowed to share the data with anyone outside of the Institutional Review Board (IRB) approved investigative team. However, individuals interested in the data are encouraged to reach out to the authors to determine if there is a collaborative capacity where the data could be potentially de‐identified and/or shared under proper IRB and/or institutional Data Use Agreements (DUA).